2020 Annual Conference Conférence Annuelle

doi:10.3138/jammi.6.s1.abst

. 2021 Jul 21;6(Suppl):1–85. doi: 10.3138/jammi.6.s1.abst

2020 Annual Conference Conférence Annuelle

PMCID: PMC9632078 PMID: 36339318

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):1.

A01. Incidence and timing of prosthetic joint infections after primary and revision arthroplasty: A population-based retrospective cohort study

Christopher Kandel ^1,^✉, Nick Daneman ², Jessica Widdifield ^3,^4,⁵, Bettina E Hansen ⁶, Richard Jenkinson ^7,⁸, Allison McGeer ⁹

Objectives

Hip and knee joint replacements are among the most common operations performed in Ontario, but rates of prosthetic hip or knee joint infections (PJIs) are unknown. The objectives of this study were to describe the incidence of PJIs after primary and revision arthroplasty in Ontario from 2003 until 2017.

Methods

A population-based retrospective cohort of all primary hip and knee arthroplasties performed in Ontario over the 15-year study period was created using linked administrative databases. Time-to-event models were used to determine the incidence of a PJI with death as a competing event, and a Poisson model was used to model trends in annual rates over time with annual surgical volume as the offset.

Results

Over the study period, 504,332 primary arthroplasties were performed, and 7,331 were complicated by a PJI. The cumulative incidence of a PJI after primary arthroplasty was 0.59% (95% CI 0.57% to 0.61%) at 90 days, 0.85% (95% CI 0.82% to 0.87%) at 1 year, and 1.29% (95% CI 1.26% to 1.33%) at 5 years. PJIs complicated a lower proportion of primary knee arthroplasties than hip arthroplasties (hazard ratio [HR] 0.89; 95% CI 0.85% to 0.93%; p <0.001). Among the 13,327 revision operations performed for a non-infectious indication, there were 697 PJIs with a 90-day incidence of 1.93% (95% CI 1.70% to 2.16%), 1-year incidence of 3.09 (95% CI 2.80% to 3.39%), and 5-year incidence of 4.89% (95% CI 4.52% to 5.26%). The proportion of all hip and knee arthroplasties complicated by a PJI at 1 year increased over time (range 0.70% in 2003 to 0.97% in 2016; p <0.001).

Conclusion

In a large population-based cohort, the continued increase in the number of primary arthroplasties is being mirrored by a rise in PJIs, particularly after hip operations. Further research into the mechanisms driving these infections and identifying individuals at highest risk is needed to target interventions capable of reducing the pronounced morbidity caused by these infections.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):1–2.

A02. Clinical characteristics, management, and outcomes of patients with indeterminate Clostridioides difficile tests using a modified two-step algorithm

Laura E Burnes ^1,^✉, Stephanie W Smith ^1,², Kinga Kowalewska-Grochowska ^1,³, Justin Z Chen ^1,²

Objectives

Clostridioides difficile infection (CDI) is associated with significant morbidity and mortality. Guidelines recommend multi-step testing algorithms. It is not clear how patients with an indeterminate combined glutamate dehydrogenase antigen–toxin enzyme immunoassay (EIA) clinically present or are managed. We aim to gain a better understanding of the clinical presentation, pharmacologic management, and outcomes of this group of patients.

Methods

A retrospective chart review of adult patients admitted to a tertiary academic centre with C. difficile testing performed at a provincial laboratory between November 1, 2018, and October 31, 2019, was undertaken. The laboratory employs a modified two-step algorithm using an enzyme immunoassay arbitrated by nucleic acid amplification testing (polymerase chain reaction [PCR]) for indeterminate EIA results. Specimens were excluded if they were referred in, from outpatients (including the emergency department), from inpatients discharged within 24 hours, and represented repeat episodes.

Results

After inclusion and exclusion criteria were applied, 190 patients tested indeterminate. Of those, 77 (41%) tested PCR- (non-toxigenic) and 113 (59%) tested PCR+. Of the 113 who tested PCR+, the antimicrobial stewardship program retrospectively assessed that 98 had possible current CDI (three or more loose stools/24 h), 1 had recent CDI, and 14 were colonized (fewer than three loose stools/24 h). Of 98 possible current CDI cases, 40 had mild, 30 had severe uncomplicated, and 28 had severe complicated disease, and 24% did not receive guideline-concordant treatment. Ten of 14 (71%) patients with colonization and 18 of 77 (23%) patients with non-toxigenic C. difficile were unnecessarily treated with antibiotics.

Conclusion

Of 91 patients with either colonization or non-toxigenic C. difficile, 31% were unnecessarily treated with antibiotics. Of 98 patients assessed as having possible current C. difficile infection, 24% did not receive guideline-concordant antibiotic therapy. These findings stress an opportunity for antimicrobial stewardship intervention in CDI management.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):2.

A03. Urinary tract infection investigation and treatment in older adults presenting to the emergency department with confusion: A health record review of local practice patterns

Rhiannan AM Pinnell ^1,^✉, Tim Ramsay ², Han X Wong ³, Pil Joo ^2,³

Objectives

Confusion is a common emergency department (ED) presentation among elderly patients. Urinary tract infection (UTI) investigation and treatment appear to be commonly initiated in confused patients with or without urinary symptoms, despite a paucity of supporting evidence. We therefore aimed to describe the prevalence of UTI investigation, diagnosis, and treatment in older patients presenting to the ED with confusion.

Methods

A retrospective health record review of older adults aged 65 years and older presenting to one of two academic EDs with confusion was performed. Patients were excluded if there was documentation of transfer from another hospital, Glasgow Coma Scale score <13, current treatment for UTI, indwelling catheters, nephrostomy tubes, or anuria. Secondary outcomes were associations among patient characteristics, rates of UTI investigation and treatment, and patient outcomes.

Results

Of the 499 patients included in the analysis, 324 (64.9%) received urine tests, 57 (11.4%) were diagnosed with UTI, and 176 (35.2%) were prescribed antibiotics. Within a subgroup of 342 patients who had no urinary symptoms, fever, or other obvious indication for antibiotics, these numbers were 199 (58.2%), 26 (7.6%), and 62 (18.1%) respectively. Patients who received urine tests were older (mean 82.4 versus 80.1 years) but did not differ in sex distribution. Patients who received antibiotics were more likely to be admitted (odds ratio [OR] = 2.9) and had higher 30-day (OR = 4.0) and 6-month (OR = 2.8) mortality.

Conclusion

Older patients presenting to the ED with confusion were frequently investigated and treated for UTI, even in the absence of urinary symptoms, despite a lack of evidence regarding whether this practice is beneficial. Antibiotic treatment was associated with higher hospitalization and mortality.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):2–3.

A04. Impact of Cobas® PCR media freezing on SARS-CoV-2 viral RNA integrity and whole genome sequencing analyses

Patrick Benoit ^1,^✉, Floriane Point ², Simon Gagnon ³, Daniel Kaufmann ^1,^2,³, Cécile Tremblay ^1,^2,³, Richard Paul Harrigan ⁴, Isabelle Hardy ^1,^2,³, François Coutlée ^1,^2,³, Simon Grandjean-Lapierre ^1,^2,³

Objectives

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genomic sequencing supports many aspects of our response to the pandemic. Freezing of primary samples and shipment to reference laboratories is frequently required because sequencing rarely occurs in clinical laboratories where diagnoses and outbreak investigations happen. Cobas® PCR Media transport medium facilitates high throughput SARS-CoV-2 reverse transcription polymerase chain reaction (PCR) analyses on Cobas multianalyzer platforms. The manufacturer does not recommend freezing Cobas PCR Media because of risks of degrading molecular templates and impairing test results. Our objective was to compare the quality and results of SARS-CoV-2 genomic sequencing when performed on fresh or frozen samples in Cobas PCR Media.

Methods

We selected 10 oral and naso-pharyngeal swab samples received in Cobas PCR Media and positive on Cobas 8800. We included cycle threshold from 14.39 to 34.74 for the ORF1 a/b target. Samples were split, and sequencing was performed after 4°C storage or −80°C freezing for 7 days. Sequencing was performed on the Oxford Nanopore Technologies MinION platform using the ARTIC Network protocol.

Results

Genomic coverage at multiple depth thresholds did not significantly differ between fresh and frozen samples. More important, 20× sequencing depth, allowing for wild-type or variant allelic identification within our protocol, was achieved for an average of 83.9% and 83.7% of the viral genome, respectively, before and after freezing (p = 0.99). Sequencing accuracy and single nucleotide polymorphism detection did not differ for samples with a PCR cycle threshold <30. In simulated outbreak investigation, using a catalog of viral genomes, and in simulated public health surveillance, using Nextstrain SARS-CoV-2 global reference sequences repository, phylogenetic placement was identical for pre- and post-freezing sequencing.

Conclusion

Freezing of Cobas PCR Media does not affect SARS-CoV-2 genomic sequencing quality and results. It is a suitable medium for outsourcing sequencing to reference laboratories and will simplify the collection and storage of samples in laboratories in which it is used.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):3.

B01. Verification of the new bioMérieux piperacillin–tazobactam ETEST® against reference broth microdilution

Philippe Lagacé-Wiens ^1,^2,^✉, Heather Adam ^1,², Christine Turenne ^1,², Markus Stein ^1,², Andrew Walkty ^1,², James Karlowsky ^1,²

Objectives

The piperacillin–tazobactam ETEST® by bioMérieux has recently been reformulated to address limitations of the strip and was re-introduced to clinical laboratories in 2019. We undertook a verification of the piperacillin–tazobactam ETEST strip for Enterobacteriaceae, Pseudomonas aeruginosa, and Acinetobacter baumannii complex.

Methods

We selected 100 gram-negative clinical isolates—40 Enterobacteriaceae, 30 A. baumannii, and 30 Pseudomonas aeruginosa—from frozen stocks for the study. Isolates were selected to include a representative mixture of piperacillin–tazobactam susceptible, intermediate, and resistant strains and included 5 carbapenemase-producing and 8 ESBL-producing Enterobacteriaceae, 2 carbapenemase-producing P. aeruginosa, and 8 piperacillin-tazobactam-resistant A. baumannii. Susceptibility testing to piperacillin–tazobactam was determined by broth microdilution following Clinical and Laboratory Standards Institute (CLSI) recommendations and ETEST following manufacturer’s recommendations and using the same inoculum. Reading of panels and strips was by four independent microbiologists, and all isolates with discrepant minimum inhibitory concentration (MIC) values were repeated to resolve the discrepancy. We determined essential agreement (EA), categorical agreement (CA), minor error rate (mE), major error (ME) rate, and very major error rate (VME) in accordance with CLSI criteria and CLSI breakpoints. Successful verification required CA >90%, mE <10%, ME <3%, and VME <3%. Given that >20% of isolates had MIC values within one dilution of the intermediate breakpoint (ie, between 16 and 128 µg/mL, inclusive), we also used the CLSI error-bound rate method to assess the performance of the strip.

Results

EA met verification criteria for P. aeruginosa (90%) but not for A. baumannii (83.3%) or Enterobacteriaceae (77.5%). CA did not meet acceptance for P. aeruginosa (83.3%), A. baumannii (86.7%), or Enterobacteriaceae (80%). mEs exceeded the threshold for A. baumannii (13.3%), P. aeruginosa (16.7%), and Enterobacteriaceae (20.0%). No MEs or VMEs occurred. Using the error-bound method did not resolve the excess mEs.

Conclusion

When compared with broth microdilution, the piperacillin–tazobactam ETEST did not meet validation criteria established by CLSI, primarily because of an excess number of mEs. Although this may be due to a high proportion of isolates with MIC values near the CLSI breakpoint (commonly seen in extended-spectrum beta-lactamase–producing and inducible AmpC-positive Enterobacteriaceae and other Enterobacterales), the error-bound method did not resolve the excess mEs.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):3–4.

B02. Performance of matrix-assisted laser desorption/ionization time of flight mass spectrometry for identification of filamentous moulds using several different extraction methods

Lisa Li ^1,^2,^✉, Elizabeth Bryce ^1,², Billie Velapatino ², Ramin Najafi ³, Leane Kishi ³, Bing Wang ⁴, Claire Swanston ⁴, Linda Hoang ^2,⁵, Vincent Tang ³, Marthe Charles ^1,²

Objectives

Identification of filamentous moulds using matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) mass spectrometry has been explored as a tool to supplement traditional microscopic identification methods. We aimed to evaluate the performance of MALDI-TOF for filamentous mould identification using several different isolate extraction methods.

Methods

A total of 45 saved clinical and proficiency testing fungal isolates (14 Aspergillus species, 11 other hyaline moulds, 14 dematiaceous fungi, 5 dermatophytes, and 1 Zygomycetes isolates) were tested on the Bruker Biotyper instrument (MBT Filamentous Fungal Library 2.0) at 2–3 days of growth, using the direct extended transfer, modified tube extraction, and liquid culture extraction methods. Mould identification to species was considered acceptable if a MALDI-TOF score of ≥1.7 was obtained. Definitive morphological identification using microscopy (and internal transcribed spacer sequencing when this was insufficient) was used as the reference method. Rates of acceptable identification using different extraction methods were compared.

Results

The rates of acceptable identification to species were 47%, 31% and 71% with the extended direct transfer, tube extraction, and liquid culture extraction methods, respectively. The rates of successful identification to genus only was 4% for the direct transfer method (the other two were able to provide a species for all isolates with an acceptable result). Lactophenol cotton blue staining at day 2–3 of growth was also undertaken and was able to provide identification to species in 13% of cases and to genus in 49% of cases.

Conclusion

For MALDI-TOF identification of moulds using the Bruker MBT Filamentous Fungal Library 2.0 database, liquid culture extraction had the best success rates in this study. The direct transfer method performed better than the tube extraction method.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):4.

B03. Evaluation of isavuconazole gradient strips for susceptibility testing of Mucorales isolates

Lillian Feng ¹, LeeAnn Turnbull ², Brad Jansen ², Tanis C Dingle ^1,^2,^✉

Objectives

Isavuconazole is an azole antifungal agent that is approved for the treatment of invasive aspergillosis and mucormycosis in adult patients. The reference method for determining susceptibility of isavuconazole to fungal pathogens is broth microdilution (BMD), which is not widely used by clinical laboratories. In this study, we evaluated the performance of gradient strips as an alternative method for isavuconazole susceptibility testing of Mucorales isolates.

Methods

Frozen stock clinical isolates of members of the Mucorales (including Rhizopus, Lichtheimia, Rhizomucor, Mucor, and Chaetomium species) (N = 100) were tested by isavuconazole BMD per the Clinical and Laboratory Standards Institute (CLSI) M38 method and gradient strips (Liofilchem) per manufacturer’s instructions. Results were assessed for essential agreement (defined as a minimum inhibitory concentration [MIC] by the BMD method within ±2 doubling dilutions of the gradient strip method), MIC₅₀, and MIC₉₀. Discrepancy resolution was performed by repeating BMD and gradient strip testing in duplicate.

Results

Before discrepancy resolution, essential agreement for the 100 Mucorales isolates between BMD and gradient strips was 89% and 96% when gradient strips were read at 80% (manufacturer recommended) and 100% inhibition (CLSI recommended), respectively. At 100% inhibition, all essential disagreements occurred with Rhizopus species isolates (4/24 = 16.7%). Upon discrepancy resolution, 3 of these discrepancies resolved, giving an overall essential agreement for Rhizopus species of 95.8% (23/24). Essential agreement for all of the other Mucorales isolates tested was 100% between BMD and gradient strip testing. MIC₅₀ and MIC₉₀ values for all Mucorales species tested by the isavuconazole gradient strip aligned with previously published BMD values from the literature.

Conclusion

Isavuconazole gradient strip testing is an accurate method for susceptibility testing of Mucorales isolates. Given that most laboratories have experience in the gradient strip method for bacterial isolates, this may provide an opportunity to bring isavuconazole susceptibility testing in house when isavuconazole susceptibility test requests are frequent with minimal additional expertise required.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):4–5.

B04. Evaluation of whole-genome sequencing–based subtyping methods for the surveillance of Shigella flexneri and Shigella sonnei and the confounding effect of mobile genetic elements in propagated outbreaks

Isabelle Bernaquez ^1,^✉, Christiane Gaudreau ^2,³, Pierre A Pilon ^4,⁵, Florence Doualla-Bell ¹, David Roy ¹, Sadjia Bekal ^1,³

Objectives

Canadian public health laboratories have begun implementing whole-genome sequencing (WGS) for the surveillance and outbreak detection of food-borne pathogens and have determined that most single-strain outbreaks are contained within 0–10 multi-locus sequence typing (MLST)–based allele differences or core genome single nucleotide variants (SNVs). However, many Shigella spp cases occur through continuous person-to-person transmission, predominantly involving men who have sex with men (MSM). WGS-based subtyping methods must now be validated for large sexual transmission clusters in addition to food- and travel-associated Shigella outbreaks.

Methods

An evaluation of four WGS-based subtyping methods (SNVPhyl, coreMLST, core genome [cg] MLST, and whole-genome [wg] MLST) was performed on nine retrospective outbreaks from a collection of 91 S. flexneri and 232 S. sonnei isolates to determine their discriminatory power, epidemiological concordance, correlations, robustness, extent of variation detected in mobile genetic elements (MGEs), and ability to generate stable interpretation criteria.

Results

The discriminatory powers were ranked as follows: coreMLST < cgSNV < cgMLST < wgMLST (range 0.970–1.000). The genetic differences observed for non-MSM-related Shigella spp outbreaks respect the standard 0–10 allele/SNV guideline; however, MGE-encoded loci caused inflated genetic variation and discrepant phylogenies for propagated MSM-related S. sonnei outbreaks via wgMLST. The S. sonnei correlation coefficients of wgMLST were also the lowest at 0.68, 0.70, and 0.71 for SNVPhyl, coreMLST, and cgMLST, respectively. The coreMLST approach was shown to be the most robust, followed by SNVPhyl and wgMLST.

Conclusion

Our results highlight the need to validate species-specific subtyping methods based on microbial genome plasticity and outbreak dynamics in addition to the importance of filtering confounding MGEs for cluster detection. We also demonstrate the usefulness of core genome–based approaches for inter-laboratory comparability and the creation of stable WGS interpretation criteria that can be applied to outbreaks of different transmission natures.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):5–6.

C01. Children hospitalized with community-acquired pneumonia complicated by effusion: A single-centre, retrospective cohort study

Gelila Alemayheu ^1,^✉, Claire SJ Lee ¹, Jeffrey M Pernica ¹

Objectives

To describe the workup and treatment of children hospitalized with community-acquired pneumonia complicated by effusion (cCAP) at a Canadian children’s hospital.

Methods

Records of all children hospitalized between September 2014 and September 2019 whose discharge diagnoses contained any pneumonia code were reviewed using a structured questionnaire. Those who had cCAP confirmed by ultrasound or drainage were eligible. Neonates and children with lung abscess, congenital heart disease, prior pulmonary disease, repeated aspiration, malignancy, immunodeficiency, renal or hepatic disease, or tuberculosis were excluded. Associations between categorical variables were assessed using Fisher’s exact test, and those involving dependent continuous variables were assessed using linear regression.

Results

Of 2,428 hospitalized children labeled with any pneumonia diagnostic code, 106 (4%) were eligible for inclusion. The median age was 4.8 years (IQR 3.2–6.8 y). The median length of stay (LOS) was 9 days (IQR 7–11 d). Forty-one participants (39%) were admitted to the intensive care unit (ICU); the median ICU stay was 5 days (IQR 2–8 days). The causative pathogen was isolated in only 61 participants (56%) via blood culture, pleural fluid culture, or pleural fluid polymerase chain reaction; there were 38 S. pneumoniae cases, 15 S. pyogenes cases, 7 methicillin-susceptible S. aureus cases, and 1 S. anginosus case. Most children (n = 80; 76%) had their effusion drained, at a median of 1 d (IQR 1–3 d) after admission. There was no relationship between LOS and effusion size, but LOS was statistically associated with time to drainage; each day’s delay in drainage increased LOS by 0.52 day (95% CI 0.11 d to 0.94 d; p = 0.014).

Conclusion

cCAP resulted in long hospitalizations and often caused critical illness. Despite the use of molecular diagnostics, causative pathogens were not identified in almost half the cases. S. pneumoniae and S. pyogenes were the most frequently identified pathogens, and no cases of MRSA cCAP were observed. Expediting drainage procedures may reduce overall LOS.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):6.

C02. Highest prevalence of Mycoplasma genitalium infection among other sexually transmitted diseases and macrolide and fluoroquinolone resistance-mediating mutations: A Saskatchewan perspective

Nidhi R Parmar ^1,^2,^✉, Linda Mushanski ³, Tasker Wanlin ², Aurora Lepe ^1,², Amanda Lang ³, Jessica Minion ³, Jo-Anne R Dillon ^1,²

Objectives

Mycoplasma genitalium is an emerging sexually transmitted infection (STI) that is more prevalent than Chlamydia trachomatis in some regions. The prevalence of resistance to antibiotics used for treating M. genitalium infections (ie, azithromycin and moxifloxacin) is also rising. This is the first detailed description of the prevalence of M. genitalium and its associated antimicrobial resistance (AMR) in Saskatchewan, Canada.

Methods

Aptima urine specimens (N = 1,977), collected (January and March–April 2019) for the diagnosis of C. trachomatis/Neisseria gonorrhoeae, were tested for M. genitalium using the Aptima Mycoplasma genitalium Assay (MG-TMA). AMR was ascertained using polymerase chain reaction and DNA sequencing of 23S rRNA (azithromycin) and parC (moxifloxacin) from MG-TMA–positive specimens; mutations predictive of resistance were noted.

Results

The prevalence of M. genitalium in Saskatchewan was 9.6% (189/1,977), higher than the prevalence of N. gonorrhoeae (3.09%) and C. trachomatis (6.85%) during the same time period. Predicted resistance to azithromycin (substitutions at positions 2058/2059 in 23S rRNA) was observed in 63.6% (70/110) of the specimens tested, and resistance to moxifloxacin (S83I in ParC) was observed in 10.6% (9/85) of the specimens. Mutations in both 23S rRNA and ParC were observed in 2.1% (4/189) of the specimens. Women aged 20–24 years had the highest prevalence (18.3%; p <0.001) of M. genitalium infection. Among women, M. genitalium was significantly associated with C. trachomatis or N. gonorrhoeae/C. trachomatis co-infection (p <0.001).

Conclusion

The prevalence of M. genitalium was highest among individuals with concurrent C. trachomatis or N. gonorrhoeae infection in the province. A high prevalence (63.6%) of macrolide resistance-mediating mutations suggests that empiric azithromycin therapy may not be adequate for treating M. genitalium infections. A potential risk of azithromycin treatment failure could be avoided by performing resistance-guided therapy because azithromycin is also a recommended drug for the treatment of N. gonorrhoeae (combined with ceftriaxone) and C. trachomatis infections in Canada.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):6–7.

C03. Population-level compliance with provincial treatment guidelines for the management of gonorrhea in Alberta, Canada, 2000–2019

Alejandra Ugarte Torres ^1,^✉, Carolina Diaz Pallarez ², John Niruban ³, Petra Smyczek ³, David Strong ⁴, Jennifer Gratrix ³, Ameeta E Singh ⁵

Objectives

Appropriate antimicrobial therapy is one of the essential strategies in the global control of gonorrhea. We sought to determine the compliance with Alberta treatment guidelines for gonococcal infections in Alberta and to identify factors associated with non-compliance.

Methods

Retrospective population-based analysis of compliance with provincial treatment guidelines for gonorrhea from January 1, 2000, to December 31, 2019, using data extracted from communicable disease outbreak database. Bivariate analyses were used. Characteristics with statistical significance (p <0.05) associated with non-compliance were included in a logistic regression analysis to identify predictors of compliance.

Results

We identified a total of 42,875 cases of gonorrhea; 58% were male and 42% were female; 40% were Caucasian, 28% were First Nations, and 15% were “unknown race.” The majority of cases were treated by a family physician (43%), followed by STI clinics (32%) and emergency departments (20%). Overall, 80.3% of individuals were treated using the preferred or alternative therapies according to guidelines. The lowest rate of compliance (32%) was registered in 2012 after the introduction of multiple changes in the guidelines: ceftriaxone for infections in men having sex with men (MSM) and throat and increased dose of cefixime, among others. Overall, treatment of gonorrhea among MSM had the highest rate of compliance (87%), as did those treated at STI clinics (92%) and correctional facilities (94%). Female gender (odds ratio [OR] 1.5; 95% CI 1.32% to 1.61%), individuals reported as “unknown ethnicity” (OR 1.4; 95% CI 1.26% to 1.62%), and those treated in the emergency department (OR 2.8; 95% CI 2.45% to 3.19%) were found to be predictors of non-compliance to treatment guidelines.

Conclusion

Our study informs areas of need for targeted interventions to improve uptake of treatment guidelines. Delivery of STI care by nurse-led care models had the highest level of compliance to treatment guidelines. This study also highlights gender and race disparities in the management of gonorrhea in Alberta.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):7.

C04. Cerebrospinal fluid shunt infections in children: A descriptive cohort study of Canadian and US patients 2013–2019

Alastair McAlpine ^1,^✉, Michelle Barton ², Archana Balamohan ³, H Dele Davies ⁴, Gwenn Skar ⁴, Marie-Astrid Lefebvre ⁵, Dolores Freire ⁶, Nicole Le Saux ⁷, Jennifer Bowes ⁸, Jocelyn A Srigley ¹, Patrick Passarelli ⁹, John Bradley ⁹, Sarah Khan ¹⁰, Rupeena Purewal ¹¹, Isabelle Viel-Thériault ¹², Adriana Ranger ², Joan L Robinson ¹³

Objectives

Current guidelines for cerebrospinal fluid (CSF) shunt infections are based on expert opinion with little data to inform them. We aimed to describe contemporary pathogens and management of shunt infections among children in Canada and the United States.

Methods

A descriptive cohort study conducted by the Paediatric Investigators Collaborative Network on Infections in Canada (PICNIC) at tertiary care hospitals in Canada (n = 8) and the United States (n = 3) of children aged up to 18 years with CSF shunt infections from July 1, 2013, through June 30, 2019.

Results

There were 183 infections in 151 children, with 9 having 2 infections and 9 having 3 or more. The most common reason for shunt placement was intraventricular haemorrhage (78 infections; 43%), and 64 infections occurred in children with congenital hydrocephalus (35%). Median time between last shunt placement or manipulation and infection was 14 days (>60 days in 65 cases; 36%). Common pathogens were coagulase-negative staphylococcus (n = 53; 28%), methicillin-susceptible Staphylococcus aureus (n = 36; 20%), methicillin-resistant S. aureus (MRSA) (n = 18; 10%), and Pseudomonas aeruginosa (n = 11; 6%); 21 infections were polymicrobial (12%), and 5 were fungal (3%). Shunt was retained in 4 cases (2%). Intrathecal antibiotics were used in 7 (4%). Duration of antibiotics was a median 17 days. Poor outcomes (death, ventriculitis, and pneumatosis intestinalis) occurred in 3 cases in which empiric antibiotics did not cover the pathogen (P. aeruginosa, Enterobacter cloacae, and Klebsiella aerogenes). Ten children died during the admission; 8 deaths were attributed to shunt infection (CFR = 4.4%). Two had possible recurrences (both with MRSA at 30–60 d), 40 had one reinfection with a different pathogen, and 25 had two or more. Presence of a gastrostomy tube was a risk factor for reinfection (p <0.001).

Conclusion

CSF shunt infections are almost always managed with replacement of the shunt. Recurrences are rare with current management.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):7–8.

D01. Impact of antibiotic prescribing feedback to high-volume primary care physicians: A randomized controlled trial

Kevin L Schwartz ^1,^2,^3,^✉, Noah Ivers ⁴, Bradley J Langford ¹, Monica Taljaard ⁵, Drew Neish ⁶, Kevin A Brown ¹, Valerie Leung ¹, Nick Daneman ⁷, Javed Alloo ⁸, Michael Silverman ⁹, Emily Shing ¹, Jeremy Grimshaw ⁵, Jerome A Leis ⁷, Julie HC Wu ¹, Gary Garber ¹

Objectives

Antibiotic overuse contributes to antimicrobial resistance. We evaluated peer-comparison audit and feedback to high-prescribing primary care physicians, assessing the impact of targeted messaging on avoiding unnecessary antibiotic initiation and avoiding long-duration prescribing.

Methods

We conducted a three-arm randomized controlled trial. The intervention targeted high-volume antibiotic-prescribing primary care physicians in Ontario, Canada. We recruited 3,500 primary care physicians who prescribed the highest volume of antibiotics in the region. Physicians were randomized 3:3:1 to a letter addressing antibiotic initiation, prescribing duration, or control (no letter). The primary outcome was total number of antibiotic prescriptions over 12 months post-intervention; secondary outcomes were number of prolonged duration prescriptions (>7 d) and antibiotic drug costs (CAD$). Robust Poisson regression controlling for baseline prescriptions was used for analysis.

Results

Physicians receiving the antibiotic initiation letter or duration letter prescribed 3.6% (risk ratio [RR] 0.964; 97.5% CI 0.923 to 1.007; p = 0.057) and 4.8% (RR 0.952; 97.5% CI 0.911 to 0.996; p = 0.014) fewer antibiotic prescriptions, respectively, than controls, with no significant difference between letters. The antibiotic duration letter led to an 8.1% reduction (RR 0.919; 97.5% CI 0.871 to 0.971; p < 0.001) in prolonged-duration prescriptions and a 6.1% reduction in antibiotic drug costs (RR 0.939; 97.5% CI 0.892 to 0.989; p = 0.006). On average, physicians receiving the duration letter prescribed 42 fewer antibiotic prescriptions and 24 fewer prolonged-duration prescriptions, resulting in $771 in drug cost savings per physician over 12 months.

Conclusion

A single mailed letter to the highest-prescribing physicians in the jurisdiction that provided peer comparison feedback including messaging on limiting antibiotic prescribing durations led to significant reductions in total and prolonged-duration antibiotics, as well as drug costs.

Trial registration

Clinicaltrials.gov NCT03776383

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):8–9.

D02. Molecular and epidemiological characterization of adult health care–associated and community-associated Clostridioides difficile infections, 2015–2019, Canada

Tim Du ^1,^✉, Kelly Baekyung Choi ², Romeo Hizon ¹, Anada Silva ², Linda Pelude ², James Brooks ^2,³, Ghada N Al-Rawahi ⁴, Jun C Collet ⁴, Blanda Chow ⁵, Jeannette L Comeau ⁶, Ian Davis ⁷, Gerald Evans ⁸, Charles Frenette ⁹, Guanghong Han ¹⁰, Susy S Hota ¹¹, Jennie Johnstone ¹², Kevin C Katz ¹³, Pamela Kibsey ¹⁴, Joanne Langley ⁶, Bonita E Lee ¹⁵, Yves Longtin ¹⁶, Dominik Mertz ¹⁷, Jessica Minion ¹⁸, Michelle Science ¹⁹, Jocelyn Srigley ⁴, Paula Stagg ²⁰, Kathryn N Suh ³, Nisha Thampi ²¹, Alice HM Wong ²², on behalf of the Canadian Nosocomial Infection Surveillance Program (CNISP) ²

Objectives

We describe the epidemiological and molecular characteristics of health care–associated (HA) and community-associated (CA) Clostridioides difficile infection (CDI) in Canadian Nosocomial Infection Surveillance Program hospitals from 2015 to 2019.

Methods

Epidemiologic data were collected using standardized case definitions from hospitalized patients identified with C. difficile infections, confirmed by polymerase chain reaction (PCR) and further characterized using PCR ribotyping and antimicrobial susceptibility using ETEST.

Results

This study included 18,455 adult cases, of which 13,735 (74.4%) were HA-CDI and 4,720 (25.6%) were CA-CDI. The overall HA-CDI rate was 4.2/10,000 patient-days, and the CA-CDI rate was 1.4/1,000 admissions, with decreasing rates seen over time (27% decrease since 2015 for HA-CDI, 23% for CA-CDI). Female patients made up 49% of those with HA-CDI and 56% of those with CA-CDI. The mean ages of HA-CDI and CA-CDI patients were 68.3 (SD 16.9) and 64.4 (SD 18.4) years. Comparing HA-CDI and CA-CDI, there was no significant difference in intensive care unit admissions, colectomy, or attributable mortality; however, 30-day all-cause mortality was significantly higher among HA-CDI cases (11.4% versus 3.0%; p <0.0001). Ribotyping and susceptibility data were available for 2,506 samples, of which 1,887 (75.3%) were HA and 619 (24.7%) were CA. The most common HA-CDI and CA-CDI ribotypes (RTs) seen were 027 (16.0% and 7.9%, respectively), 106 (11.5% and 12.3%), 014 (8.6% and 8.1%), and 020 (6.4% and 8.4%). The overall resistance rates of all HA-CDI and CA-CDI isolates were similar for all drugs tested, except moxifloxacin, for which rates were 21.7% versus 12.4%, with the frequency of moxifloxacin resistance in HA-CDI specimens decreasing from 34.3% to 13.5% during the study period. No resistance to metronidazole, vancomycin, or tigecycline was observed.

Conclusion

HA-CDI and CA-CDI rates, and fluoroquinolone resistance among HA-CDI, are decreasing in Canada. Although prevalence of RT 027 has declined, there has been an increase in prevalence of RTs 014, 020, and 106. Thirty-day all-cause mortality was more strongly associated with HA-CDI than with CA-CDI; however, no significant differences in other severe outcomes were found.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):9.

D03. Optimizing the treatment of Staphylococcus aureus bloodstream infection with the implementation of a molecular assay and antimicrobial stewardship intervention

Hilal Al Sidairi ^1,², Emma K Reid ^3,^✉, Navjot Sandila ⁴, Jason LeBlanc ^2,³, Ian Davis ^2,³, Paul E Bonnar ^2,³

Objectives

Vancomycin is often used for Staphyococcus aureus bloodstream infections, pending sensitivities. Earlier sensitivity reporting should decrease vancomycin use and optimize early beta-lactam antibiotic therapy among patients with methicillin-susceptible S. aureus (MSSA) infections. We assessed the impact of implementing Xpert® MRSA/SA BC (Cepheid, Sunnyvale, CA) molecular testing coupled with an antimicrobial stewardship (AMS) intervention on the time to optimal antimicrobial therapy for S. aureus bloodstream infections.

Methods

Xpert molecular testing was performed for patients with S. aureus bloodstream infections between January and July 2020. With this sensitivity information, the AMS team contacted the treating physician with the results via phone and provided advice. The primary outcome of interest was time to optimal therapy, defined as the time from first blood culture draw to time of optimal therapy. Optimal therapy was defined as the first appropriately dosed cefazolin or cloxacillin monotherapy, or the time at which vancomycin concomitant to cefazolin or cloxacillin was discontinued. The outcome was compared with a historical cohort from 2017–2018.

Results

Fifty-six patients (29 intervention, 27 control) were included. The median time to optimal therapy was shorter in the intervention group than in the historical cohort (38.0 h, 95% CI 32.8 to 47.0, versus 50.1 h, 95% CI 29.8 to 69.5). Time to optimal therapy was characterized using Kaplan–Meier plots, and the log-rank test indicated that the curves differed significantly by group (p = 0.0405). Participants in the intervention group at any time point during the study period were 77% more likely to start optimal therapy (hazard ratio 1.77; 95% CI 1.02 to 3.09; p = 0.0432).

Conclusion

Implementing Xpert molecular testing with AMS notification significantly reduced the time to optimal therapy and likely decreased vancomycin use in patients with MSSA bloodstream infections. Implementing molecular testing can improve therapy optimization, particularly in remote hospitals where there may be delays in sensitivity testing.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):9–10.

D04. Assessment of empirical vancomycin use and appropriateness at a tertiary care pediatric hospital

Khaled Alsager ^1,^✉, Joseph Vayalumkal ², Cora Constantinescu ³

Objectives

Vancomycin has multiple empiric indications in pediatrics, such as coverage of methicillin-resistant Staphylococcus aureus (MRSA) and central line–associated infections. However, vancomycin use has been an antimicrobial stewardship (AMS) target because of difficulties in reaching and maintaining therapeutic levels as well as toxicities. We sought to describe empirical vancomycin use in a tertiary care pediatric centre over an 8-week period to inform local guideline development.

Methods

A pediatric AMS fellow in collaboration with AMS staff and pharmacist performed prospective audit and feedback (PAF) for all vancomycin starts, dosage adjustments in relation to levels, and 48-hour reviews. Data were collected prospectively over an 8-week period between November 16, 2020, and January 8, 2021, and analyzed using frequencies and percentages.

Results

Twenty-two patients received vancomycin during the PAF period. Stratified by location of utilization, 55% of cases were on general pediatrics unit, 22% on the oncology unit, and 13% in the neonatal intensive care unit. Indications for use are shown in Figure D04-1. Starting dose was 60 mg/kg/day in 47% and 45 mg/kg/day in 40%. Initial fifth-dose trough level was therapeutic in 29% of cases. Duration of vancomycin use was less than 48 hours in 67%. The indications to continue beyond 48 hours in the remaining 33% were deemed appropriate upon review, as all were for confirmed gram-positive bacteremia (MRSA, coagulase-negative Staphylococcus species, Streptococcus mitis) except one case of pending MRSA screening swabs. A total of four antimicrobial stewardship program interventions were recommended and were accepted.

Figure D04-1: — Indication for vancomycin

Conclusion

Our PAF identified that the AMS concerns associated with inpatient vancomycin use were not around initiation, de-escalation, or duration. This needs assessment showed that the AMS concerns were around inconsistency in the starting dose, frequency, and subsequently achieving and monitoring for therapeutic levels. Future vancomycin local guideline and education needs to be tailored to vancomycin dosage and monitoring.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):10–11.

E01. Waning anti-nucleocapsid IgG signal among SARS-CoV-2 seropositive blood donors: May–November 2020

Sahar Saeed ^1,^✉, Steven J Drews ^2,³, Chantale Pambrun ⁴, Sheila F O’Brien ¹

Objectives

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) seroprevalence survey estimates may be affected by declining anti-nucleocapsid immunoglobulin G (IgG) antibodies. We assessed waning signal-to-cut-off ratios (S/Co) among seropositive Canadian blood donors.

Methods

This longitudinal study was nested within a national Canadian blood donor seroprevalence survey. The Abbott Architect SARS-CoV-2 IgG assay (nucleocapsid antigen based) was used to determine reactivity at a S/Co of ≥1.4. A mixed-effects linear regression model was used to evaluate the rate of S/Co decline among repeat donors (≥2 measurements). We then compared the rate of S/Co decline between consistently positive donors and those who became negative.

Results

Between May 9 and November 30, 2020, 108,015 EDTA plasma retention samples were screened for SARS-CoV-2 IgG. Seroprevalence increased from 0.75% (552/74,642; May–July) to 0.98% (327/33,373; October–November). Ninety seropositive donors had two or more measurements over 27 weeks. At baseline, the median age was 49 (IQR 34 to 62), and 66 (73%) were men; S/Co ratios were 2.8 (IQR 1.75, 5.2), decreasing to 1.95 (IQR 1.41 to 2.92) at last screening. Adjusting for variance between donors, the rate of decline was −0.06 units/week (95% CI −0.08 to −0.05). Of donors, 74% (67/90) remained positive (S/Co ≥1.4), and 26% (23/90) of donors dropped to negative S/Co levels (<1.4). Among subsequent negative donors, the rate of decline was significantly faster: −0.11 unit/week (95% CI −0.14 to −0.08) versus donors that remained positive (−0.04 units/wk; 95% CI −0.06 to −0.02). Age, sex, region, ABO blood group, and date of baseline donation were not associated with baseline S/Co ratios or the rate of decline.

Conclusion

We describe waning anti-nucleocapsid IgG S/Co ratios among SARS-CoV-2 seropositive repeat blood donors. Although the sample size was limited, more data from later time points will be accrued. As the coronavirus disease 2019 pandemic progresses, developing methods to adjust for waning antibody signals in seroprevalence studies should be a key research priority.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):11.

E02. SARS-CoV-2 serology: Validation of high-throughput chemoluminescent immunoassay (CLIA) platforms and a field study in British Columbia

Annie S Mak ^1,^✉, Inna Sekirov ¹, Darrel Cook ¹, Paul Levett ¹, Agatha Jassem ¹, Muhammad Morshed ¹, Andrea Olmstead ¹, Vilte Barakauskas ², Janet Simons ³, Shazia Masood ⁴, Marthe Charles ⁵, Althea Hayden ⁶, Mel Krajden ¹

Objectives

Multi-site laboratory validation of six high-throughput severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) chemoluminescent immunoassays (CLIA) and subsequently evaluation of the six different SARS-CoV-2 assays in a serosurvey at two health care facilities affected by coronavirus disease 2019 (COVID-19) outbreaks.

Methods

Six high-throughput CLIA platforms were evaluated and validated using characterized pre-COVID samples and performance characteristics claimed by each manufacturer. The platform evaluated are as follows: (1) LIAISON^® SARS-CoV-2 S1/S2 IgG (DiaSorin IgG; DiaSorin, Saluggia, Italy); (2) ARCHITECT SARS-CoV-2 IgG (Abbott IgG; Abbott, Abbott Park, IL, USA); (3) VITROS® Anti-SARS-CoV-2 Total (Ortho T; Ortho Clinical Diagnostics, Raritan, NJ, USA); (4) VITROS Anti-SARS-CoV-2 IgG (Ortho IgG; Ortho Clinical Diagnostics); (5) SARS-CoV-2 Total Assay (Siemens T; Siemens, Chicago, IL, USA); and (6) Elecsys® Anti-SARS-CoV-2 (Roche T; Roche, Indianapolis, IN, USA). For the serosurvey of facilities affected by COVID-19 outbreaks, serum samples were collected from consenting residents and staff as part of a public health investigation.

Results

Overall sensitivity was lowest for DiaSorin IgG (78.6%) and highest for Ortho T (100%); only DiaSorin IgG showed significantly lower sensitivity than the other assays for samples collected 0–14 days post-onset. Specificities were high for all assays (range 95.4% to 100%) and were not significantly different. An excellent specificity correlation among samples was found for cross-reactivity across platforms.

Conclusion

In the field study, there was consensus agreement for the majority of known positive (95.7%) and negative (98.4%) patient samples, but no assay had 100% accuracy. Discordant results may indicate variability in serological responses among patients or in antibody detection by the assays. When using a given assay for clinical diagnostic purposes, especially where positive predictive value is low, orthogonal testing with a second assay, perhaps with a different antigen target, would increase confidence that dual reactive results indicate true antibody positivity.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):11–12.

E03. Analysis of the performance of water gargle samples for SARS-CoV-2 detection using heat extraction followed by laboratory-developed nucleic acid amplification testing

Sarah Gobeille Paré ¹, Julie Bestman-Smith ¹, Judith Fafard ², Florence Doualla-Bell ², Mariève Jacob-Wagner ¹, Manon St-Hilaire ³, Stéphanie Beauchemin ³, Valérie Boucher ⁴, Fabiola Vancol-Fable ³, Christian Lavallée ³, Annie-Claude Labbé ^3,^5,^6,^✉

Objectives

With the increasing demand for coronavirus disease 2019 (COVID-19) testing, self-collected non-invasive specimens should be assessed on large-scale testing platforms. We validated the use of water gargle (WG), in comparison with oral and naso-pharyngeal swab (ONPS), for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) detection using heat extraction followed by an in-house laboratory-developed nucleic acid amplification test (HE-LD-NAAT).

Methods

Health care workers and individuals from the general population were prospectively recruited at two screening clinics in Quebec City (n = 816) and Montreal (n = 509). After the ONPS sampling, participants gargled with 5 mL of Eska or Naya natural spring water for 20 seconds. The paired specimens were analyzed using HE-LD-NAAT assay for detection of SARS-CoV-2, and a comparative performance analysis was performed. An individual was considered infected if a positive result was obtained on either sample.

Results

Among the 1,325 participants, the mean age was 36 (range 6–78), and 1,035 had symptoms (78%). Invalid results were obtained for 12 ONPS and 6 WG (these 18 pairs were excluded from analysis). SARS-CoV-2 was detected in 148/1,307 (11.3%) participants—125 from both samples, 20 from ONPS, and 3 from WG (κ= 0.91). Sensitivity was estimated at 98.0% (95% CI 93.9 to 99.4) for ONPS and 86.5% (95% CI 79.9 to 91.1) for WG. WG sensitivity was slightly higher when restricted to symptomatic individuals (87.9%; 95% CI 81.3 to 92.3). Mean ONPS cycle threshold (Ct) value was lower for concordant paired samples than for discordant ones (Ct 23.1 versus 32.3; p <0.001). The median ΔCt between WG and ONPS among concordant positive samples was 7.6 (range −7.5 to 15.3), which might explain the loss of sensitivity among samples with a lower amount of RNA.

Conclusion

Being non-invasive, the higher acceptability of WG might compensate for the loss of sensitivity and enable a higher rate of detection on a population-based perspective. Nonetheless, among patients with a high clinical suspicion of COVID-19, a repeated analysis with ONPS should be considered.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):12.

E04. Saliva assay for viral examination of SARS-CoV-2 (SAVE-CoV): A pilot study

Susan M Poutanen ^1,^2,^✉, Camille Lemieux ^2,³, Bryn Hazlett ¹, Megan Landes ^2,³, Allison McGeer ^1,², Tony Mazzulli ^1,², Steven Friedman ^2,³

Objectives

Saliva is as an alternate to the gold standard naso-pharyngeal (NP) swabs for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) detection by reverse transcription polymerase chain reaction (RT-PCR) that saves health care worker time, personal protective equipment, and NP swab supplies. This pilot study sought to assess the sensitivity of saliva compared with NP swabs collected in viral transport media for SARS-CoV-2 detection by RT-PCR using the Seegene’s Allplex™ 2019-nCoV Assay.

Methods

Seven hundred patients attending a coronavirus disease 2019 (COVID-19) assessment centre in May 2020 who met screening criteria for symptomatic COVID-19 self-collected neat saliva samples in sterile containers (Starplex, Etobicoke, Ontario, Canada); NP swabs were also collected following routine practice. Saliva was tested neat and mixed 1:1 with sterile normal saline alongside the NP swab on the same Seegene Allplex 2019-nCoV Assay run. Results were compared for yield and cycle threshold (Ct).

Results

A total of 689 NP–saliva pairs were received with sufficient quantity of saliva for testing. Of these, 47 (6.82%) were positive for SARS-CoV-2 in either one or both of the specimens. Of these, 74.5% had positive NP swabs, 57.5% had positive diluted saliva, and 59.6% had positive neat saliva. Compared with NP swabs, average Ct values were 3.39 and 3.72 higher for diluted saliva and neat saliva results. Discrepancies between NP and saliva occurred with higher Ct (average Ct >33.5). Invalid results were noted in 0.1%, 2.5%, and 13.6% of NP, diluted saliva, and neat saliva specimens, respectively.

Conclusion

Using the Seegene Allplex 2019-nCoV Assay, COVID-19 cases will be missed with either NP swabs (75% sensitive) or saliva (58% sensitive) as the sole specimen type. For patients with high pre-test probability of COVID-19, combination testing or repeat testing should be considered. Comparing novel tests with NP swab RT-PCR as the gold standard should take into consideration the limitations of the yield of NP swabs.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):12–13.

P01. OPAT quality indicators at a tertiary care centre reveal vast room for improvement

Jorge L Martinez-Cajas ¹, Susan McKenna ², Beatriz Alvarado-Llano ¹, Evan Wilson ^1,^✉, Kirk Leifso ¹, Santiago Perez-Patrigeon ¹, Gerald Evans ¹

Objectives

Outpatient parenteral antimicrobial therapy (OPAT) allows patients to receive intravenous antibiotic treatment outside the hospital. Like many other health centres in Ontario, the current practice of OPAT at Kingston Health Sciences Centre (KHSC) has no formal structured OPAT program. We describe key OPAT quality indicators of the current KHSC OPAT practice.

Methods

We identified patients who were discharged on OPAT from KHSC between July 2017 and June 2018. We manually retrieved data from medical records, including sociodemographic characteristics, diagnoses, comorbidities, antimicrobial type and duration, clinical outcomes, complications, and emergency department (ED) visits and readmissions. We focused on the quality indicators suggested by Berrevoets et al. (2019).

Results

During the study period, 313 patients were discharged from KHSC on parenteral antimicrobial treatment. The most common diagnoses were bacteremia (35.6%), pneumonia (13.2%), osteomyelitis or septic arthritis (11.6%), and endocarditis (9.6%). An infectious disease specialist consultation was obtained in 48%; no OPAT-related instructions were given to primary care physicians in 87%; and instructions to patients were insufficient in 85% and non-existent in 12.2%. Monitoring labs were not ordered in 74% of the cases. Pharmacist input on OPAT before discharge occurred in 54% and was non-existent post-discharge. Ninety patients (28.9%) were readmitted within 60 days of discharge, 69% for causes related to OPAT. ED visits occurred among 21.1%, and 86% were related to OPAT, with most cases representing uncomplicated intravenous catheter issues.

Conclusion

Our results indicate major patient safety issues and poor patient outcomes in the absence of a formal structured OPAT program. The results likely underestimate actual rates because we had no access to records outside KHSC. In contrast, a recent publication reported ED visit and readmission rates as low as 5% and 14%, respectively, in a tertiary-care hospital with a formal OPAT program. Formal OPAT programs at tertiary care centres would likely improve these outcomes.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):13–14.

P02. Outcomes of a risk-based decision-making process for reducing the risk of hepatitis E virus transfusion transmission via the Canadian blood supply

Sheila Ward ¹, Gilles Delage ², Yves Gregoire ³, Gordon Hawes ⁴, Christopher Brennan ¹, Lisa Potter ¹, Margaret Fearon ⁵, Christy Simpson ^6,^7,⁸, Michael McDonald ^7,⁹, Bartha Knoppers ^7,¹⁰, Sheila O’Brien ¹, Steven J Drews ^11,^12,^✉

Objectives

Hepatitis E virus (HEV) can asymptomatically infect healthy blood donors, usually after donors eat raw or undercooked pork. Transfusion transmission (TT) and infection can occur, with severe liver disease impacting stem cell–solid organ transplant patients and patients with hematological malignancies. No prior HEV TT has been described in Canada, and Canadian blood operators do not screen donors for HEV. We describe the outcomes of a risk-based decision-making (RBDM) process aimed at reducing the risk of HEV TT in Canada.

Methods

Pertinent literature was assembled, and risk information was generated or made available to the risk assessment team. The team included experts in medical microbiology, epidemiology, health economics, ethics, stakeholder engagement, donor testing, and risk management. The Alliance of Blood Operators RBDM framework was used to guide information review and analytical stages involving preparation, problem formulation, stakeholder consultation, assessment, evaluation, and decision.

Results

The risk of a significant HEV infection via TT (1 in 1 million to 1 in 10 million) is roughly equivalent to residual risks for infection with HIV, hepatitis B virus, or hepatitis C virus after donor laboratory testing. RBDM options identified were (1) status quo, (2) test all blood donations for HEV, (3) test selected blood donations for HEV, (4) increase physician awareness of HEV and improve adverse event and enhanced donor surveillance, (5) introduce pathogen reduction technology, (6) identify high-risk donors through an additional question on the Record of Donation regarding food habits, (7) vaccinate a panel of donors against HEV, and (8) test mini pools rather than single donors. Options varied in feasibility; options 5–7 are not possible, options 2, 3, and 8 are not immediately feasible, and option 1 had low ethical acceptability.

Conclusion

HEV TT risks are extremely rare in Canada. Option 4 (increase physician awareness of HEV and improve adverse event and enhanced surveillance) is the recommended option.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):14.

P03. Evaluation of the NG-Test Carba 5 multiplex immunochromatographic assay and Cepheid Xpert Carba-R assay for the detection of carbapenemase genes in gram-negative organisms

Ghada N Al-Rawahi ^1,^2,^✉, Ryan Penny ¹, Audrey Cuarte ¹, Suk Dhaliwal ¹, Tammy Chen ¹, Dale Purych ^2,³, David Boyd ⁴, Linda Hoang ^2,⁵, Peter Tilley ^1,²

Objectives

Timely and accurate detection of carbapenemase-producing organisms (CPOs) is crucial for patient management and infection prevention and control. The aim of this study was to evaluate the performance of the NG-Test Carba 5 multiplex immunochromatographic assay (NG Biotech, Guipry, France) and the Cepheid Xpert Carba-R assay (Cepheid, Sunnyvale, CA, USA) for the detection of carbapenemase genes from pure bacterial culture.

Methods

A total of 61 molecularly characterized isolates were included in this evaluation; 51 positive for at least one of the five carbapenemase types (KPC, NDM, IMP, VIM, and OXA-48) and 10 negative isolates resistant to at least one carbapenem. The isolates included members of the order Enterobacterales (n = 43), Pseudomonas aeruginosa (n = 16), and Acinetobacter baumannii (n = 2).

Results

The concordance with the reference method was 88.5% (54/61) and 83.6% (51/61) for the NG-Test Carba 5 and Xpert Carba-R, respectively. Two false-positive NDMs with faint bands were seen with the NG-Test Carba 5 and tested negative upon repeat. Almost all of the false-negative results in both assays were seen with the IMP type. The overall sensitivity and specificity were, respectively, 89.8% and 83.3% for the NG-Test Carba 5 and 82% and 100% for the Xpert Carba-R.

Conclusion

The NG-Test Carba 5 and Xpert Carba-R assays provided rapid results and are easy to implement in the clinical laboratory. The NG-Test Carba 5 assay performed slightly better for IMP type detection, and false positives could be reduced by repeating when faint bands are produced.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):14–15.

P04. SARS-CoV-2 transmission through cells, tissues, and organs transplantation

Amaury Gaussen ^1,^✉, Laura Hornby ^2,³, Gary Rockl ¹, Sheila O’Brien ³, Gilles Delage ⁴, Ruth Sapir-Pichhadze ^5,^6,⁷, Steven J Drews ^8,⁹, Matthew J Weiss ^10,^11,¹², Antoine Lewin ^4,¹³

Objectives

The emergence of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus responsible for the coronavirus disease 2019 (COVID-19) pandemic has raised concerns for programs overseeing donation and transplantation of cells, tissues, and organs (CTO) that this virus might be transmissible by transfusion or transplantation. This review summarizes the published cases of CTO transplantation from donors infected with SARS-CoV-2 and assesses the current state of knowledge on the detection of this virus in different biological specimens.

Methods

Two independent search strategies were used by two independent teams to identify eligible publications between December 2019 and October 2020 on the transmission of SARS-CoV-2 to recipients after CTO transplantation. The review was extended to include 39 references related to the detection of viral genomic material, proteins, and infectious SARS-CoV-2 virus in clinical samples, cells, and tissue specimens collected from different organs in COVID-19-positive patients.

Results

Autopsy studies of COVID-19-positive patients reveal that different organs may be affected, but to date whether these injuries are directly attributable to SARS-CoV-2 infection is still uncertain. SARS-CoV-2 RNA has been detected in kidney, liver, heart, bowel, and intestines. Although several studies support the possibility of the virus replicating in tissues of several organ types, absence of experimental controls or validation of these observations by independent teams make it difficult to inform on transmissibility of SARS-CoV-2 by solid organ transplants. We identified four case reports involving liver or hematopoietic stem cells from COVID-19-positive donors. Ultimately, no transmission of SARS-CoV-2 by CTO transplantation was documented.

Conclusion

Evidence collected to date raises the possibility of SARS-CoV-2 infection and replication in some CTO, which makes it impossible to exclude transmission through transplantation. Improved standardization of donor and CTO screening practices, along with a systematic follow-up of the respective transplant recipients, could facilitate the assessment of SARS-CoV-2 transmission risk through transplantation.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):15.

P05. One health genomic study of extended-spectrum beta-lactamase–producing Salmonella enterica in Canada, 2012–2016

Amrita Bharat ^1,^2,^✉, Laura Mataseje ¹, E Jane Parmley ^3,⁴, Brent P Avery ⁴, Graham Cox ^1,², Carolee Carson ⁴, Rebecca J Irwin ⁴, Anne E Deckert ⁴, Danielle Daignault ⁵, David C Alexander ⁶, Vanessa G Allen ⁷, Sameh El Bailey ⁸, Sadjia Bekal ⁹, Linda Chui ¹⁰, Greg J German ¹¹, David Haldane ¹², Linda Hoang ¹³, Jessica Minion ¹⁴, George Zahariadis ¹⁵, Richard J Reid-Smith ⁴, Michael R Mulvey ¹

Objectives

Extended-spectrum beta-lactamases (ESBLs) confer resistance to extended-spectrum cephalosporins, which are an important class of antimicrobials. Here, we used whole-genome sequencing to investigate whether food and animal sources may be a reservoir of ESBL-producing Salmonella for human infection.

Methods

The Canadian Integrated Program for Antimicrobial Resistance Surveillance collects Salmonella from human sources from provincial public health laboratories and from food and animal samples from farms, abattoirs, retail stores, and animal health laboratories. We carried out short-read whole-genome sequencing on ESBL Salmonella isolates identified from 2012 to 2016 and long-read sequencing on a subset of isolates.

Results

From 2012 to 2016, the prevalence of ESBL enzymes in Salmonella isolated from humans in Canada was 0.35% (54 ESBL positive of 15,401 screened); it was 0.27% in animals and meat (41/14,923). Salmonella from animals and meat included isolates from turkeys (n = 22), pigs or pork (n = 11), cattle or beef (n = 4), chickens (n = 3), and domestic cat (n = 1). The most common ESBLs were bla_CTX-M-65 (n = 18) and bla_SHV-2 (n = 12) in humans, and bla_CTX-M-1 (n = 19) and bla_SHV-2 (n = 15) in animals and meat. We frequently observed S. infantis carrying bla_CTX-M-65 in humans and S. Albany carrying bla_CTX-M-1 in the food chain. Most ESBL serotypes and alleles differed in humans versus animals. Two exceptions were bla_SHV-2 IncI1 and bla_CTX-M-1 IncI1 plasmids, which were found in both sources. A subclade of S. Heidelberg from humans and chicken meat carrying the same IncI1-bla_SHV-2 plasmid differed by only one to seven single nucleotide variants. Many ESBL Salmonella were multiclass resistant, with resistance to six classes observed in 46.2% of human isolates and 29.3% of animal and meat isolates.

Conclusion

In general, there were few instances of similar ESBL-producing Salmonella and plasmids in the two sources, suggesting that the sampled animals and meat may only be a minor reservoir of ESBL Salmonella with respect to human infections.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):15–16.

P06. Canadian cost-utility analysis of isavuconazole for the treatment of patients with invasive aspergillosis and mucormycosis

Catherine Beauchemin ^1,², Kimberly Guinan ¹, Karine Mathurin ^1,², Simon Frédéric Dufresne ³, Coleman Rotstein ⁴, David Claveau ^5,^✉, Liette Landry ⁵, Jean Lachaine ^1,²

Objectives

Invasive mould infections (IMIs), including invasive aspergillosis (IA) and mucormycosis (IM), occur in immunocompromised patients, directly affecting their life expectancy. Because IA is more common than IM, voriconazole is often initiated with patients with suspected pulmonary IMI, despite its ineffectiveness against IM. Isavuconazole, which targets both IA and IM, was recently approved in Canada and has become a key empirical treatment for suspected pulmonary IMI. The primary objective was to assess the cost-effectiveness of isavuconazole compared with voriconazole for the treatment of suspected pulmonary IMI in Canada. A secondary objective was to assess the impact of varying the study time horizon to address the wide spectrum of life expectancies of IMI patients according to their underlying diseases.

Methods

In the base-case analysis, a decision tree was developed over a 10-year time horizon from the Canadian Ministry of Health (MoH) perspective. The target population was patients with suspected pulmonary IMI (including IA [95%] and IM [5%]). Efficacy parameters were extracted from the SECURE and VITAL trials. Costs included were those associated with treatment acquisition, hospitalization, adverse event management, and productivity loss. Results were based on incremental cost-utility ratios (ICUR), analyzing the cost per quality-adjusted life-year (QALY). A scenario analysis using a 3- and a 5-year time horizon was conducted to assess the impact of different life expectancies that might be expected with hematological malignancies. Extensive sensitivity analyses were also conducted.

Results

From a MoH perspective, 3-, 5-, and 10-year time horizons resulted in an estimated deterministic ICUR of $45,424/QALY, $30,160/QALY, and $16,778/QALY, respectively. Results of the probabilistic sensitivity analyses showed that isavuconazole remained cost-effective for all time horizons, relative to a willingness-to-pay threshold of $50,000/QALY.

Conclusion

This economic evaluation demonstrates that, in comparison to voriconazole, isavuconazole is a cost-effective strategy for all patients with IMI regardless of their life expectancy.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):16.

P07. A virtual care model using patient-directed oximetry monitoring for outpatients with COVID-19: A quality improvement study

Megan K Devlin ^1,^2,^✉, Michael J Nicholson ^1,², Jaclyn Ernst ^1,³, Inderdeep Dhaliwal ^1,³, Marko Mrkobrada ^1,³, Erin Spicer ^1,³

Objectives

A care gap exists for outpatients with coronavirus disease 2019 (COVID-19), with many lacking access to standardized medical care. We combined virtual clinical assessment with patient-directed oximetry to enhance clinical care of these patients. The aim of this study was to assess the role of oximetry and clinical outcomes of those enrolled in this novel clinical initiative.

Methods

A team of general internal medicine, infectious diseases, and respirology physicians partnered with the local public health unit to enroll outpatients diagnosed with COVID-19. We assessed patients virtually and arranged for the same-day delivery of an oximetry device to at-risk patients’ homes to assess for hypoxemia. In this quality improvement study, we performed a review of the use of oximeters in our virtual care clinic and present 30-day patient outcomes using this novel clinical model.

Results

Between April 23 and May 19, 2020, we assessed and monitored 51 patients with COVID-19 in the community. Of these, 47% had an oximeter delivered to their residence. A majority of patients (91%) who experienced severe dyspnea had normal oxygen saturations. Our clinical intervention resulted in three direct admissions to a designated COVID-19 unit at a local hospital for decompensating patients. No deaths were noted. We have characterized a number of significant outcomes that warrant further medical and allied health follow-up.

Conclusion

We present a clinical model that supports the care and symptomatic management of patients in the community with COVID-19. Oximetry was found to primarily exclude the presence of hypoxemia in dyspneic patients while identifying few patients with true hypoxemia.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):16–17.

P08. Case report: Actinotignum schaalii identified in a neck abscess of a pediatric patient

Ruwandi M Kariyawasam ^1,^2,^✉, Estelle M Morin ³, Eduard Eksteen ⁴, Kinga Kowalewska-Grochowska ^1,^2,⁵

Objectives

Actinotignum schaalii is a gram-positive facultative anaerobic non-motile coccoid rod commonly identified along the human genitourinary tract and implicated in urinary tract infections (UTIs). Few reports have documented A. schaalii infections among pediatric patients, specifically those of a non-UTI origin. We present a case of A. schaalii identified in a neck abscess of a boy aged 22 months with no underlying immunodeficiency, prior medical history, or urogenital infections.

Methods

The patient presented with a 3-day history of an erythematous and swollen nodule below the left clavicle. The nodule had been progressively enlarging with increasing tenderness and was 2–3 cm at time of presentation. He was systemically well; however, upon ultrasound, an abscess was identified. This prompted incision and drainage, at which point 5 mL of purulent material was sent for culture. He was started on intravenous clindamycin and, after the removal of his Penrose drain, was discharged on amoxicillin–clavulanic acid for 1 week.

Results

Culture was performed using blood agar plates (BAPs) under anaerobic and CO₂ conditions, MacConkey plate, brain heart infusion (BHI) plate, phenylethyl alcohol agar (PEA) plate, and a cooked meat broth. After 48 hours of growth on the BAP and the PEA, A. schaalii appeared as small, transparent colonies similar to coryneform bacteria and was identified using the Vitek matrix-assisted laser desorption ionization–time of flight mass spectrometry 5 days after patient discharge. Coryneform bacilli and mixed anaerobic flora were also identified. In our case, antimicrobial susceptibility testing was not performed.

Conclusion

Our patient showed near complete resolution of subcutaneous collection on repeat ultrasound 1 month later and was advised to be monitored for progression or evolution in the future. From a clinical standpoint, A. schaalii is increasingly encountered in non-UTI infections and should be monitored particularly in the pediatric population to determine whether it may be considered an emerging pathogen.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):17–18.

P09. Synergistic activity of combinations of colistin with other antibiotics against multidrug- and extensively drug-resistant clinical isolates of Klebsiella pneumoniae, Acinetobacter baumannii, and Pseudomonas aeruginosa

Tatiana A Petrovskaya ¹, Dmitry V Tapalski ¹

Objectives

To reveal antibiotics capable of potentiating the antimicrobial activity of colistin against multidrug- and extensively drug-resistant strains of Klebsiella pneumoniae, Acinetobacter baumannii, and Pseudomonas aeruginosa.

Methods

The minimum inhibitory concentrations (MICs) of colistin alone and in combination with fixed concentrations of antibiotics of different groups were determined for 272 multidrug- and extensively drug-resistant strains of K. pneumoniae, A. baumannii, and P. aeruginosa. Bactericidal activity was determined at fixed pharmacokinetic/pharmacodynamic breakpoint concentrations of antibiotics with the use of a modified method for testing bactericidal activity of various combinations (multiple combination bactericidal antibiotic testing).

Results

Potentiation of colistin antibacterial activity in the presence of fixed concentration of rifampicin (0.5 mg/L) was observed as a 4- to 16-fold MIC decrease for K. pneumoniae and A. baumannii. In the presence of fixed concentrations of azithromycin (2 mg/L) or clarithromycin (1 mg/L), the colistin MICs decreased 64–512 times for K. pneumoniae, 4–32 times for A. baumannii, and 16–64 times for P. aeruginosa. Two- fold or more reduction of MIC of colistin in the presence of 1 mg/L clarithromycin was observed for 85.2% of K. pneumoniae, 86.3% of A. baumannii, and 60.2% of P. aeruginosa strains. In the presence of 1 mg/L clarithromycin and 8 mg/L meropenem, the potentiation effect was enhanced and was observed for an even larger percent of isolates: K. pneumoniae, 96.1%; A. baumannii, 98.0%; and P. aeruginosa, 61.3%. Colistin-based combinations with clarithromycin–meropenem and clarithromycin–doripenem were bactericidal against most isolates of A. baumannii and P. aeruginosa (91.4–100%), and against colistin-sensitive K. pneumoniae (95.3%) and colistin-resistant K. pneumoniae (79.1%).

Conclusion

The ability of macrolides to significantly potentiate the colistin antimicrobial activity against both colistin-sensitive and colistin-resistant strains of K. pneumoniae, A. baumannii, and P. aeruginosa was shown. This potentiation effect was enhanced in the presence of carbapenems.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):18.

P10. Vancomycin minimum inhibitory concentration of Staphylococcus species isolated from children’s blood culture

Eugene YH Yeung ^1,^2,^✉, Nicole MA Le Saux ³

Objectives

Since 2020, the Infectious Diseases Society of America (IDSA) has recommended use of an area-under-the-curve-to-minimum-inhibitory-concentration (AUC:MIC) ratio of 400–600, in place of trough level, for therapeutic monitoring of vancomycin in treatment of serious methicillin-resistant Staphylococcus aureus (MRSA) infections. The IDSA assumes most that S. aureus isolates have a vancomycin MIC of 1 μg/mL. Limited data suggest that coagulase-negative Staphylococcus (CoNS) endocarditis with vancomycin MIC ≥2 μg/mL had the highest mortality rate. The current study aimed to determine the vancomycin MIC of the Staphylococcus isolates among the patients in our pediatric teaching hospital.

Methods

An audit was conducted of all patients with Staphylococcus bacteremia identified in our 167-bed children’s hospital from November 2019 to October 2020 (inclusive). The vancomycin MICs of these Staphylococcus isolates were obtained from broth microdilution performed at the microbiology laboratory.

Results

Twenty patients (mean age 8.0 y) had S. aureus bacteremia; 15 had vancomycin MIC of 1 μg/mL, and the rest had an MIC of 0.5 μg/mL. Seventeen patients’ S. aureus bacteremia became positive within 36 hours of collection. Two patients had MRSA bacteremia. In comparison, 44 patients (mean age 6.3 y) had CoNS bacteremia, 24 of whom had vancomycin MIC of 1 μg/mL, 17 of whom had a MIC of 2 μg/mL, and 3 of whom had a MIC of 0.5 μg/mL. Thirty-nine patients’ CoNS bacteremia became positive within 36 hours of collection. The vancomycin MIC of S. aureus was significantly different from that of CONS (p <0.01).

Conclusion

Our children’s hospital may apply the IDSA guidance on the use of AUC:MIC because most of the S. aureus isolates had a vancomycin MIC of 1 μg/mL. However, the use of AUC:MIC may be limited because our hospital had only two cases of MRSA bacteremia. What the target AUC:MIC should be for CoNS bacteremia is unknown.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):18–19.

P11. Microbiological efficiency of the combinations of two carbapenems against Klebsiella pneumoniae carbapenemase-producing strains

Elena V Timoskova ^1,^✉, Dmitry V Tapalski ¹

Objectives

We assessed synergistic activity of combinations of two carbapenems against extensively drug-resistant K. pneumoniae strains producing different types of carbapenemases.

Methods

For 60 antibiotic-resistant K. pneumoniae strains, the minimum inhibitory concentrations (MICs) of colistin and carbapenems were determined by broth microdilution method. Detection of NDM, KPC, and OXA-48 carbapenemase genes was performed. The MICs of meropenem and doripenem were determined in the presence of a fixed pharmacokinetic–pharmacodynamic (PK–PD) concentration of 0.5 mg/L ertapenem. The sensitivity to combination of doripenem and ertapenem was assessed by checkerboard method.

Results

The bla_KPC genes were detected in 15 K. pneumoniae strains, and bla_OXA-48 in 31 strains, bla_NDM in 14 strains. Only 33 strains (55.0%) were sensitive to colistin (MIC ≤2 mg/L).

The OXA-48–producing strains had less pronounced resistance to carbapenems (meropenem MIC₉₀ 64 mg/L, doripenem MIC₉₀ 128 mg/L). The levels of resistance to doripenem were maximal for NDM-producing strains (MIC₉₀ 256 mg/L). The effect of potentiating the activity of doripenem with ertapenem at PK–PD concentration was observed for 20.0% of KPC-producing strains and 29.0% of OXA-48–producing strains and was absent for NDM-producing strains. The potentiating effect did not depend on the sensitivity to colistin. The potentiating effect of meropenem by ertapenem in PK–PD concentration was observed only for 8.3% of strains.

For five of six strains—producers of KPC and OXA-48 carbapenemases in which the effect of potentiation for the combination of doripenem–ertapenem was revealed in a preliminary experiment—the synergistic effect of this combination was confirmed by the checkerboard method (summation fractional inhibitory concentration [ΣFIC] 0.31 to 0.50). For one more strain, the interaction effect was additive (ΣFIC 1.00). The effect of the doripenem–ertapenem combination was neutral for all NDM producers (ΣFIC 1.50 to 2.00).

Conclusion

The ability of ertapenem to potentiate the antimicrobial activity of doripenem and meropenem against some of the strains producing carbapenemases KPC and OXA-48 was confirmed.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):19.

P12. Waning anti–SARS-CoV2-2 plaque reduction neutralization test titres in repeat Canadian convalescent plasma donors: April–December 2020

Steven J Drews ^1,^2,^✉, Dana V Devine ^3,⁴, Janet McManus ³, Emelissa Mendoza ⁵, Kathy Manguiat ⁵, Roxie Girardin ⁶, Alan Dupuis ⁶, Kathleen McDonough ^6,⁷, Y-C James Lin ^8,⁹, David H Evans ^8,⁹, Chantale Pambrun ¹⁰, Michael Drebot ^5,¹¹, Heidi Wood ⁵

Objectives

Since April 2020, we have used plaque reduction neutralization test 50 (PRNT₅₀)–generated values to qualify convalescent plasma donations supporting clinical trials in Canada. We describe waning PRNT₅₀ titres of repeat plasma donors.

Methods

All donations met standard criteria for plasma donations. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)–specific neutralizing antibody titres were determined using PRNT₅₀ in Vero E6 cell cultures. Plasma units demonstrating titres of ≥1:160 were issued to trial sites. Donors with titres of ≥1:160 were encouraged to return for repeat donation as frequently as weekly. Donors with titres of 1:40 or 1:80 were asked to donate one additional time in hopes that their titres would reach the 1:160 level. Data were stored in an Excel file, and statistical analysis was undertaken using GraphPad Prism.

Results

From April 29, 2020, to December 5, 2020, there were 408 donations from 128 repeat donors making two or more donations. Of the repeat donors, 36.7% (47/128) showed a eight-fold or more decrease in PRNT₅₀ titres from peak to trough. The median time from onset of symptoms to a eight-fold or more decrease in PRNT₅₀ titres was 129 days (range 63–233 days). Of repeat donors, 46.4% (64/128) stopped donating as a result of dropping titres or donations that did not meet the PRNT₅₀ titre of ≥1:160. At “stop donation,” the median PRNT₅₀ was 1:40 (range 0–1:80) with ranges of 1:80 (18.8% [12/64]), 1:40 (32.8% [21/64], and <1:40 (48.4%, 31/64). Regression analysis of titre versus time past resolution of symptoms indicated a significant relationship (p = 0.0005; Spearman r = –0.1726 with Gaussian approximation; confidence interval −0.2679 to −0.07389.

Conclusion

As a result of physiologic waning, blood operators cannot infer that SARS-CoV-2 PRNT₅₀ titres will remain high in repeat plasma donors 4 months after onset of COVID-19 symptoms. With an increasing number of COVID-19 cases in Canada, we have increased recruiting activity to ensure ongoing inventory for clinical trials.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):19–20.

P13. Impact of the COVID-19 pandemic on the incidence and serotype distribution of invasive pneumococcal disease in south central Ontario

Allison McGeer ^1,^✉, Kevin R Barker ², Aaron Campigotto ³, Larissa Matukas ⁴, David Richardson ⁵, Christie Vermeiren ⁶, Reena Lovinsky ⁷, Nataly Farshait ⁸, Agron Plevneshi ¹, Kazi Hassan ¹, Irene Martin ⁹

Objectives

To evaluate the impact of the coronavirus disease 2019 (COVID-19) pandemic on the incidence and serotype distribution of invasive pneumococcal disease (IPD) in south central Ontario

Methods

The Toronto Invasive Bacterial Diseases Network has performed population-based surveillance for IPD in the Metropolitan Toronto and Peel Region since 1995. All laboratories serving residents of the population area report cases in which S. pneumoniae is isolated from a sterile site to the central study office, with annual audits conducted to ensure complete reporting. Clinical and demographic information are collected by chart review and patient and physician interview. Isolates are serotyped at Canada’s National Microbiology Laboratory. Population data are obtained from Statistics Canada.

Results

The annual incidence of IPD increased from 5.95 per 100,000 population in 2015–2016 to 7.9 per 100,000 in 2018–2019, then declined to 3.70 per 100,000 population in 2020 (a 54% decrease compared with 2018–2019). The decreased rate became significant in April and persisted through December (Figure P13-1). The decline in incidence was greatest in children (81%; 5.33 in 2018–2019 to 0.99 per 100,000 in 2020) and lesser in adults aged 65 years and older (56%; 20.2 to 8.80 per 100,000) and those aged 16–64 years (53% decrease: 5.87 to 2.76 per 100,000). The decrease occurred uniformly in different vaccine serotype groups. Overall, in 2020, 15% of isolates causing IPD were of serotypes included in PCV17, 36% were included in pneumococcal conjugate vaccine (PCV) 13, 54% were included in PCV20, and 9% were included in pneumococcal polysaccharide vaccine 23 but not in conjugate vaccines. There was no difference in the case fatality rate in cases in the data available in 2020 compared with 2018–2019 (15/81 [18.5%] versus 88/589 [14.9%], p = 0.40).

Figure P13-1: — Number of cases of IPD per month, Toronto/Peel

IPD = Invasive pneumococcal disease

Conclusion

The COVID-19 pandemic is associated with a significant decline in IPD in all serotype and age groups in our population, suggesting that pandemic public health measures have reduced IPD risk.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):20–21.

P14. Impact of an antimicrobial stewardship initiative to reduce asymptomatic bacteriuria in a tertiary hospital

Maggie O Wong ^1,^✉, Natalie Makhoulian ¹, Sandra Hosseini ¹, Kevin Afra ¹

Objectives

Reducing asymptomatic bacteriuria (ASB) is an important antimicrobial stewardship (AMS) target. Up to 57% of antibiotics for urinary tract infection (UTI) are prescribed in cases with unclear symptoms. This offers no medical benefit, but it has significant risk of harm, such as antimicrobial resistance or Clostridium difficile infection. This quality-improvement project describes a multipronged approach to reduce antibiotic usage for ASB in a 700-bed tertiary hospital that consisted of education plus prospective audit and feedback and, more uniquely, empowered two hospitalist champions to provide peer feedback.

Methods

Patients aged >18 years with a positive urine culture more than 48 hours after hospital admission and started on antibiotics due to suspected or confirmed UTI were included. Patients in intensive care, psychiatric, and maternity units were excluded. The pre-intervention and intervention periods were from October 2018 to April 2019 and from October 2019 to April 2020, respectively. The interventions included education sessions between periods, ongoing AMS pharmacist audit and feedback, and longitudinal peer feedback from two hospitalist champions. The main objective was to decrease inappropriate antibiotic use for ASB. Definition of UTI was based on the Association of Medical Microbiology and Infectious Disease Canada “symptom-free pee: let it be” algorithm. We compared the percentage of patients prescribed antibiotics for ASB during pre-intervention and intervention periods.

Results

Of the 57 cases treated pre-intervention, 16 (28%) were identified as ASB. This decreased to 16% (12/74 cases) in the intervention period. In the intervention period, documentation of UTI symptoms improved drastically from 50% to 88% of the cases, likely due to peer feedback from the hospitalist champions. As a result of sample size constraints, results did not achieve statistical significance.

Conclusion

Collaborating with hospitalist champions successfully decreased the percentage of patients who received antibiotic(s) for ASB. Improved chart documentation of UTI symptoms may also improve continuity of care and reinforce not treating ASB.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):21.

P15. Development of quality indicators to evaluate appropriate empiric antimicrobial use in pediatric patients

Emily Black ^1,^2,^✉, Holly MacKinnon ¹, Jeannette Comeau ^1,², Kathryn Timberlake ³, Michelle Science ³, Kathryn Slayter ^1,²

Objectives

The objective of this study was to obtain consensus on quality indicators that characterize appropriate empiric antimicrobial use for the management of infectious syndromes in hospitalized pediatric patients.

Methods

This study was completed using the Delphi technique, an interactive forecasting method that relies on a panel of experts. The research team developed an initial list of quality indicators, informed by a literature search. The list was electronically disseminated to expert panelists (infectious disease physicians and pharmacists) using Opinio survey software. Panelists were asked to rate indicators on a 9-point Likert scale considering the following criterion: the importance of each item in determining appropriateness considering benefit or harm at the individual or population level. In the first round, panelists also had the option to recommend additional quality indicators or suggest changes in wording. Consensus was defined as ≥75% agreement and a median score of ≥7 after three rounds.

Results

Of 31 invited experts, 12 completed at least one round of the survey, and 10 completed all rounds. Consensus was achieved in 28 of 31 indicators proposed after three rounds. Indicators that met consensus were categorized under empiric choice (n = 12), dose (n = 5), administration (n = 4), duration (n = 2), diagnosis (n = 2), and documentation (n = 3). Six indicators that achieved consensus were re-phrased by experts.

Conclusion

Consensus was achieved on quality indicators to assess appropriate empiric antimicrobial use in pediatric patients. Clinicians and researchers can use these consensus-based indicators to assess adherence to best practice.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):21–22.

P16. Penicillin susceptibility in Staphylococcus aureus: An opportunity for antimicrobial stewardship?

Philippe Lagacé-Wiens ^1,^2,^✉, Markus Stein ^1,², Peter Pieroni ¹, Heather Adam ^1,², Christine Y Turenne ^1,², Andrew Walkty ^1,², Terry Wuerz ^2,³, James Karlowsky ^1,²

Objectives

Staphylococcus aureus is widely assumed to be resistant to penicillin because of the presence of the blaZ penicillinase. Reports from the 1970s indicate that approximately 90% of hospital-associated methicillin-susceptible S. aureus (MSSA) were resistant to penicillin by virtue of penicillinase production. However, more recent studies have suggested the re-emergence of susceptible isolates. Because penicillin may offer both pharmacokinetic and stewardship advantages for the treatment of S. aureus infections, we sought to establish a current estimate of the proportion of MSSA isolates that are both susceptible to penicillin and lack penicillinase activity in three urban hospital centres in Manitoba, Canada.

Methods

One hundred sixty-three consecutive MSSA isolates from three urban microbiology laboratories were tested for penicillin susceptibility using the Clinical and Laboratory Standards Institute (CLSI) disk diffusion method and penicillin zone-edge test. As per CLSI guidelines, isolates tested with a penicillin 10U disk that produced a zone size >28 mm and a diffuse zone edge (“beach”) were deemed susceptible to penicillin. Isolates demonstrating a sharp zone edge (“cliff”) were considered penicillin resistant regardless of the zone diameter. Isolates that tested negative for beta-lactamase were confirmed negative by blaZ polymerase chain reaction (PCR).

Results

Thirty-two of 163 (19.6%) MSSA were penicillin susceptible. Among susceptible isolates, zone sizes varied from 34 to 57 mm in diameter. No isolates displaying a diffuse zone edge (ie, penicillinase negative) were resistant to penicillin by disk diffusion testing. All isolates that tested negative for beta-lactamase by the zone edge test were blaZ PCR negative.

Conclusion

Of MSSA isolates, 19.6% were susceptible to penicillin. Although the proportion of susceptible isolates is relatively low, it appears higher than those reported in older studies and suggests that penicillin may represent an acceptable treatment option for some MSSA infections in which the pharmacokinetic profile and narrower spectrum of penicillin are advantageous compared with those of penicillinase-stable beta-lactams. Negative zone-edge tests reliably predict the absence of blaZ.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):22.

P17. Nosocomial SARS-CoV-2 infection in health care workers: descriptive analysis of a large acute care hospital COVID-19 outbreak on cardiac units

Devika Dixit ^1,^2,^✉, John Conly ^1,³, Kate Snedeker ^1,³, Nicholas Etches ^1,³, Amanda Weiss ¹, Michael Suddes ¹, Peter Jameison ^1,³, Stephen Tsekrekos ^1,⁴

Objectives

To provide a detailed assessment of hospital-acquired severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections in health care workers (HCWs) during a cardiac unit outbreak.

Methods

Epidemiological data on HCWs infected with SARS-CoV-2 during this outbreak were gathered retrospectively using a detailed standardized questionnaire and analyzed as part of the outbreak response to identify infection source. Serial asymptomatic prevalence testing was conducted on HCWs every 5 days for three cycles to identify pre-symptomatic and asymptomatic HCWs. Outbreak control measures were implemented on the basis of investigation results.

Results

Forty-two nosocomial HCW infections were identified in 17 registered nurses, 9 licensed practical nurses, 8 physicians, 5 allied health workers, and 3 health care aides. HCWs aged 20–39 years made up 65% (28/42) of infections, and 70% (30/42) of total HCWs were female. Serial asymptomatic prevalence testing initially identified 4 HCWs; 2 developed symptoms within 48 hours of their prevalence test, and 2 remained asymptomatic. In total, 40/42 HCWs (95%) ultimately had symptomatic coronavirus disease 2019 infection. Although the majority of HCW infections are thought to be due to patient exposures secondary to missed point-of-care risk assessment and poor hand hygiene practices, HCW-to-HCW transmission risk also occurred during carpooling, breaks, and a teaching session without continuous masking. Visitor-to-HCW exposures may have also played a role in some acquisitions.

Conclusion

Almost all the infected HCWs developed symptoms during this outbreak as identified by means of a detailed questionnaire and HCW follow-up, in contrast to other reports in the literature that have noted high proportions of asymptomatic cases. Outbreak control measures targeted sources of infection among HCWs including other HCWs, patients, and visitors. These measures included contact tracing and quarantine of exposed HCWs, serial HCW testing, isolation of infected HCWs, evaluation of break rooms and other common areas, and HCW cohorting to a single work location.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):22–23.

P18. Verification of the Vitek® 2 N391 gram-negative AST card: Results are method dependent

Philippe Lagacé-Wiens ^1,^2,^✉, Markus Stein ^1,², Peter Pieroni ¹, Heather Adam ^1,², Christine Turenne ^1,², Andrew Walkty ^1,², James A Karlowsky ^1,²

Objectives

Verification of commercial antimicrobial susceptibility testing (AST) panels is typically required before use for patient isolates. The M52 Clinical and Laboratory Standards Institute document provides guidance on the verification of these systems. Verification is required when panels are modified to accommodate new breakpoints, to add new drugs, or to reformulate existing drugs to address testing limitations. As a comparator, a currently verified commercial AST system with arbitration by a reference method or direct comparison with a reference method such as disk diffusion or broth microdilution (BMD) is recommended. The purpose of this study was to compare verification results of the newly released Vitek® 2 N391 gram-negative AST card with an in-use commercial system comparator and a reference broth dilution comparator.

Methods

Thirty gram-negative isolates, including six challenge strains (eg, ESBL production, AmpC production, and carbapenemases) were used in this study. For all isolates, AST was performed on the new N391 card, on the previously verified N208 card, and by broth microdilution. Essential agreement (EA), categorical agreement (CA), minor error rate (mE), major error (ME) rate, and very major error rate (VME) were determined using both comparator methods. Successful verification required EA >90%, CA >90%, mE <10%, ME <3%, and VME <3%.

Results

Using BMD as the comparator, essential agreement was unacceptable for ceftazidime (75%) and nitrofurantoin (83%), and categorical agreement was unacceptable for ertapenem (89%), nitrofurantoin (83%), and tobramycin (89%). Minor errors exceeded the threshold for ertapenem (11%) and nitrofurantoin (17%), and major errors exceeded the threshold for meropenem (4.5%) and tobramycin (4.5%). Using the GN208 card with arbitration by BMD as the comparator, only tobramycin failed verification with categorical agreement of 89% and ME of 4.8%.

Conclusion

Using previously validated commercial AST panels for verification of new panels may underestimate mE and ME while overestimating CA and EA. Laboratories should consider using a reference method for verification of new commercial panels.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):23.

P19. Evaluation of the Solana® Influenza A and B nucleic acid amplification assay

Philippe Lagacé-Wiens ^1,^2,^✉, Markus Stein ^1,², Peter Pieroni ¹, Heather Adam ^1,², Christine Turenne ^1,², Andrew Walkty ^1,², James Karlowsky ^1,²

Objectives

The Solana® Influenza A + B assay uses helicase-dependent amplification for the rapid detection of influenza A and B from naso-pharyngeal swabs without an extraction step. We compared the performance of the Solana system with the GeneXpert® influenza A/B assay (Xpert® Flu) in use in our laboratory.

Methods

One hundred fifty-three naso-pharyngeal swabs from three urban microbiology laboratories were selected from −70°C frozen stocks that had been previously tested for influenza A and B by GeneXpert. The samples were thawed and tested using the Solana instrument. Discrepant results were repeated by both GeneXpert and Solana in parallel to account for possible loss of target RNA during the freeze–thaw cycles. Parallel testing results were considered correct if they differed from the original testing results. Where possible, specimens that remained discrepant after parallel testing were tested by third party testing using an alternate assay for influenza A and B (Seegene® RV16). Results of the third-party assay were considered correct for analysis. Sensitivity and specificity of the assays were compared using McNemar’s χ².

Results

The Solana Influenza A + B assay detected 61 of 74 influenza A–positive samples (sensitivity 82.4%; 95% CI 71.5% to 90.0%) and 57 of 72 influenza B–positive samples (79.2% sensitivity, 95% CI 67.7% to 87.5%). Specificity was 100% for both analytes. Samples that were falsely negative with the Solana instrument tended to have high cycle threshold values by GeneXpert PCR (>28 cycles). Sensitivity of the Solana system was significantly lower than that of the GeneXpert system (p <0.001 for both targets).

Conclusion

The Solana instrument has lower sensitivity for the detection of influenza A and B than the GeneXpert influenza A/B assay (Xpert Flu).

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):23–24.

P20. Stepwise implementation of an antimicrobial stewardship prospective audit and feedback intervention at a large academic centre in Canada

Justin Z Chen ^1,^2,^✉, Carlos Cervera ¹, Stephanie W Smith ^1,², Dima Kabbani ^1,², Cecilia Lau ³, Serena Bains ³, Karen Fong ³, Jackson Stewart ³, Karen E Doucette ^1,²

Objectives

Effective implementation strategies for establishing antimicrobial stewardship program (ASP) inpatient interventions are not well described. We describe a methodology for prospective audit and feedback (PAF) implementation at a large Canadian hospital and report its outcomes.

Methods

A PAF intervention targeting restricted antibiotics (carbapenems, daptomycin, linezolid, and tigecycline) was launched in March 2018. Stepwise implementation by inpatient programs was prioritized by collaboration readiness because immediate sitewide implementation was deemed unfeasible. New (past 1–3 d) prescriptions were assessed by the antimicrobial stewardship team. Exclusions included single dose, discontinued before assessment, or discharge before assessment. Prescription, prescriber, intervention type, and acceptance data were collected through December 2020.

Results

Sixteen steps over 2 years were required for sitewide implementation. Over the study period, 1,629 prescriptions were evaluated; 1,064 (65%) were empiric, and the remainder were culture directed. Of empiric prescriptions, 192 (18%) were not guideline concordant by indication, 240 (23%) did not follow guidelines but were reasonable by expert opinion, and 632 (59%) were guideline concordant. Only 977 prescriptions (60%) were optimally prescribed (evaluated by agent, regimen, and duration).

ASP recommendations included the following: no change, 972 (60%); change agent, 291 (18%); change regimen, 194 (12%); discontinue antibiotic, 119 (7%); and other, 53 (3%). The recommendation acceptance rate was 86% (656 cases evaluable). Infectious disease (ID) was involved in 523 (32%) prescriptions. With ID involvement, higher guideline-concordant empiric prescribing (67% versus 57%; p = 0.003) and regimen appropriateness (75% versus 53%; p <0.001) was observed. ASP suggested no changes more often with ID involvement (74% versus 53%; p <0.001).

Conclusion

Stepwise implementation was acceptable to programs and allowed evolution and implementation of ASP processes and workflow. Of prescriptions audited, 40% resulted in intervention, suggesting a possible role for further education, guidelines, or restriction. ID involvement was associated with more optimal prescribing. A stepwise implementation of an antimicrobial stewardship PAF intervention is viable.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):24.

P21. SARS-CoV-2 outbreak among physicians at a Canadian curling bonspiel: A descriptive observational study

Bonnie Meatherall ^1,^✉, Sampson Law ², Chris Rice ², Jia Hu ³, Christopher Fung ⁴, Allan Woo ⁵, Kevin Fonseca ^6,⁷, Amanda Lang ⁸, Jamil Kanji ^9,¹⁰, Kelly Burak ^2,¹¹

Objectives

Transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is known to occur among individuals who congregate in large groups, especially during indoor activities. Our objective was to provide a detailed clinical description of an outbreak of coronavirus disease 2019 (COVID-19) that occurred after a sporting and social event during the early days of the pandemic in Canada.

Methods

We conducted a descriptive observational study of a SARS-CoV-2 outbreak that occurred during a curling bonspiel in Edmonton, Alberta, March 11–14, 2020. We developed a standardized questionnaire to conduct interviews of all bonspiel participants between April 17 and May 5, 2020. Information regarding demographics, travel history, symptoms (type, onset, and duration), self-reported testing results, and clinical outcomes was collected.

Results

Seventy-three curlers (56 health care workers) from four Canadian provinces participated. Convalescent SARS-CoV-2 immunoglobulin G serology was completed for 85% of participants. Of the 73 participants (75.3% male; median age 51 [range 26 to 79]; 72.6% physicians), 40 curlers (54.8%) tested positive for SARS-CoV-2 by reverse transcription polymerase chain reaction. An additional 16 participants developed symptoms but had negative swabs or were not tested (14 were probable cases). Anosmia with ageusia or dysgeusia occurred among 72.2%. The clinical course was mild in the majority (one emergency visit, no hospitalizations). Transmission likely occurred from multiple individuals with minor non-specific symptoms during the event, possibly during shared meals.

Conclusion

Our study confirms the wide spectrum of symptoms patients may experience with SARS-CoV-2 infection, and the high frequency of anosmia may help distinguish it from other respiratory infections. The 74% attack rate (confirmed or probable cases) in our study highlights the infectivity of SARS-CoV-2 during sporting and social events. This reinforces the need for public health measures (masking, physical distancing, and limiting the size of social gatherings) during the subsequent waves of the pandemic.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):25.

P22. Verification of 2019 CLSI M100 revised ciprofloxacin breakpoints for Pseudomonas aeruginosa using the Vitek® 2 N391 gram-negative AST card

Philippe Lagacé-Wiens ^1,^2,^✉, Markus Stein ^1,², Peter Pieroni ¹, Heather Adam ^1,², Christine Turenne ^1,², Andrew Walkty ^1,², James Karlowsky ^1,²

Objectives

In 2019, the 29th edition of the M100 Clinical and Laboratory Standards Institute (CLSI) document changed the Pseudomonas aeruginosa ciprofloxacin susceptibility breakpoint from 1 mg/mL to 0.5 mg/mL, justified by pharmacokinetic–pharmacodynamic data supporting a lower breakpoint. We undertook a verification study of the new breakpoint applied to the Vitek® 2 N391 card in use in our laboratory.

Methods

One hundred sixty-five consecutive non-repeat P. aeruginosa isolates from three urban microbiology laboratories were subjected to ciprofloxacin susceptibility testing using the Vitek 2 GN391 card. Minimum inhibitory concentrations (MICs) were interpreted using 2019 CLSI M100 breakpoints (S, ≤0.5 mg/mL; I, 1 mg/mL; R, ≥2 mg/mL). Susceptibility results were compared with Kirby-Bauer as the reference gold standard. We determined categorical agreement (CA), minor error rate (mE), major error (ME) rate, and very major error rate (VME). Successful verification required CA >90%, mE <10%, ME <3%, and VME <3%. If verification failed, the error-bound rate method was also used to assess the performance of the card with the new breakpoint.

Results

Categorical agreement was observed for 144/165 (87.3%) isolates with minor errors occurring in 19 of 165 (11.5%) and major errors identified in 2 of 134 (1.5%) susceptible isolates. No VMEs were detected. Using the error-bound rate calculation, the minor error rate remained excessive at 28.6% in the “intermediate plus 2 dilutions” (MIC ≥4 mg/mL) category, confirming a failed verification.

Conclusion

Applying the new P. aeruginosa ciprofloxacin breakpoints to the N391 Vitek 2 AST card resulted in an unacceptably low categorical agreement and unacceptably high minor error rate with both the conventional and the error-bound methods. This is likely the result of the new ciprofloxacin breakpoint bisecting the wild-type MIC distribution of P. aeruginosa.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):25–26.

P23. Congenital SARS-CoV-2 infection in a pre-term neonate with viable virus from the placenta

Joseph V Vayalumkal ^1,^✉, Amuchou Soraisham ¹, Ayman Abou-Mehrem ¹, Jessica Dunn ¹, Kevin Fonseca ², Byron Berenger ^2,³, Marie-Anne Brundler ³, Elaine Chan ³, James Lin ⁴, David Evans ⁴, John M Conly ⁵

Objectives

Congenital infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is rare, and existing reports vary in their strength of evidence. We sought to (1) confirm congenital SARS-CoV-2 infection in a neonate born to a mother with active coronavirus disease 2019 and (2) describe the clinical, virologic, and pathologic data to support transplacental transmission.

Methods

The clinical courses of the neonate and mother were obtained from clinical records. Naso-pharyngeal (NP) specimens were collected from the neonate at 24 and 48 hours of life and tested for the envelope (E) gene by real-time polymerase chain reaction (RT-PCR). Placental tissue, umbilical cord tissue, and cord blood were cultured for SARS-CoV-2 on Vero cells, and quantitative RT-PCR targeting the E and nucleocapsid (N) genes was performed. Pathological examination was completed on the placenta.

Results

The mother developed fever 2 days before delivery, and pregnancy was complicated by pre-term premature rupture of membranes for 11 days. The baby was born at 30 weeks, 2 days gestation by vaginal delivery. The neonate initially required non-invasive ventilation but was weaned to room air by day 4. E gene PCRs from the infant were positive from NP swabs (cycle threshold [Ct] 17.9 and 22.8), as well as the cord blood (Ct 36.95) and the placenta (E gene Ct 24.9, N gene Ct 16.03). Culture of the placental tissue yielded SARS-CoV-2 at 2.80 × 10² pfu/mL, showing viral plaque with typical morphology for SARS-CoV-2. Viral culture was negative from umbilical tissue and cord blood. Placental examination showed histiocytic intervillositis with perivillous fibrin deposition in a subchorionic distribution.

Conclusion

Virologic, PCR, pathologic, and clinical data support a congenital infection. This may be the first described congenital case in a pre-term neonate to be confirmed with viable virus cultivated from placental tissue.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):26.

P24. Comparative analysis of capillary versus venous blood for serologic detection of SARS-CoV-2 antibodies by rPOC lateral flow tests

Tamara Pidduck ^1,^✉, Inna Sekirov ¹, Meghan McLennan ¹, Paul Levett ¹, Navdeep Chahil ¹, Annie Mak ¹, Erin Carruthers ¹, Jesse Kustra ¹, Ken Chu ¹, Fred Burgess ¹, Lori Willis ¹, Ray Wada ¹, Rosemarie Blancaflor ¹, Mel Krajden ¹, Muhammad Morshed ¹

Objectives

To evaluate the performance of rapid point of care (rPOC) in the field, the British Columbia Centre for Disease Control Public Health Laboratory (BCCDC PHL) conducted a comparative assessment of the performance of rPOC lateral flow assays in a laboratory (using venous blood samples) versus a field (using fingerpick capillary blood) setting. Field testing was conducted in two long-term-care facilities (LTCFs) affected by coronavirus disease 2019 (COVID-19) outbreaks

Methods

Initially, seven rPOC products were screened in the laboratory using venous samples obtained from known COVID-19 patients at 0–7, 8–14, and >14 days post–illness onset (N = 79), as well as pre-pandemic negative samples stored at BCCDC PHL tested for other serology before 2019 (n = 63). The Artron Diagnostics Inc. product was later selected for a dual laboratory–field trial.

Results

The Artron rPOC showed excellent performance in the laboratory setting; however, in the field, we found that finger-prick–based sensitivity of the Artron rPOC test was inferior to that done with venous blood in a laboratory setting, with immunoglobulin G sensitivity dropping to 82% (specificity 99%) and immunoglobulin M sensitivity dropping to 66% (specificity 92%).

Conclusion

Our results demonstrate poorer performance of rPOC assays in field settings relative to the laboratory, possibly because of reduced standardization in blood inoculum in the field. Capillary blood inoculum may vary in volume with possible effects on sensitivity. The nature of capillary blood collection also predisposes the sample to hemolysis, which might interfere with test specificity. Variability of lighting in the field and operator training may further compound these effects.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):26–27.

P25. Cefiderocol therapy in a Canadian urban academic centre: A new option to treat carbapenem-resistant gram-negative infections

Christine Ondro ^1,^✉, Irina Rajakumar ¹, Samuel Bourassa-Blanchette ^1,², Alejandra Ugarte-Torres ^1,², Joanne Salmon ^1,²

Objectives

Cefiderocol is a novel siderophore cephalosporin stable against many classes of beta-lactamases, including metallo-beta-lactamase. It is approved in the United States for complicated urinary tract infections (cUTIs), including pyelonephritis, and hospital-acquired pneumonia and ventilator-associated pneumonia caused by susceptible gram-negative organisms.

Methods

We describe two cases of compassionate cefiderocol use through Health Canada’s Special Access Programme (SAP), including treating OXA-23–producing Acinetobacter baumannii complex acute epididymitis and VIM-producing Pseudomonas aeruginosa prostatitis.

Results

A 79-year-old man with hypertension was hospitalized in India for polytrauma after a motor vehicle accident. He was then transferred to Canada, where he was found to have rectal colonization with NDM carbapenemase-producing Escherichia coli, OXA-23 carbapenemase-producing Acinetobacter baumannii complex, and OXA-48–like carbapenemase-producing Klebsiella pneumoniae complex. Five months later, he was readmitted with acute epididymitis secondary to carbapenemase OXA-23 A. baumannii, developing acute renal failure on day 3 of colistin. After a 14-day course of cefiderocol, his symptoms resolved, and his urine and rectal cultures became negative. The second patient, a 65-year-old man with a history of recurrent UTIs, was admitted with prostatitis complicated by an abscess. Urine culture grew VIM carbapenemase-producing Pseudomonas aeruginosa. After initial therapy with colistin and a transurethral resection of the prostate, he was switched to intravenous fosfomycin and piperacillin–tazobactam while awaiting cefiderocol approval. He completed 20 days of cefiderocol therapy along with piperacillin–tazobactam. A repeat ultrasound showed resolved abscess, and his urine cultures were negative. He was discharged 8 weeks later after a complicated hospital stay.

Conclusion

Cefiderocol is a new antibiotic indicated for the treatment of multi-drug-resistant gram-negative infections with limited treatment options. In our two cases, cefiderocol was well tolerated and effective to treat infections secondary to metallo-beta-lactamase–producing organisms.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):27.

P26. Survey results from a pilot antimicrobial stewardship prospective audit and feedback in general surgery

April J Chan ^1,^✉, Melanie Tsang ^1,², Bradley J Langford ^3,⁴, Mark Downing ^1,²

Objectives

Prospective audit and feedback (PAF) is established in critical care settings but not in the surgical population. We piloted once-weekly face-to-face PAF for the acute care surgery (ACS) service in a community hospital and conducted a survey to determine feasibility and barriers.

Methods

All surgical residents, fellows, and staff surgeons who were on the ACS service from November 23, 2015, to April 29, 2019, were invited to participate in a web-based anonymous survey by email. The survey was open for 3 months (from November 2019 to February 2020). Respondent demographics and questions using a 5-point Likert scale were measured using counts. Answers to free-text questions were analyzed and grouped into common categories.

Results

The overall response rate was 25% (n = 10). Of the respondents, 60% were staff surgeons, and the rest were residents. Fifty percent of respondents agreed that PAF provided them with skills to use antimicrobials more judiciously, and 80% of respondents agreed that PAF improved the quality of antimicrobial treatment for their patients. The most useful aspects of PAF were that it served as a reminder to reassess duration and was a scheduled review of patients on antimicrobials. In contrast, the least useful aspects of PAF were its lack of a more formal structure and the inconvenience of in-person rounds. Respondents found it difficult to use antimicrobials judiciously because of a lack of high-quality evidence, unique cases with lack of source control, and the high number of prescribers. Conversely, improved dissemination of evidence-based guidelines based on local resistance rates was thought to improve antimicrobial use. Overall, 70% of respondents agreed that this pilot should continue, with no respondent against it.

Conclusion

Once-weekly scheduled PAF appears to be well received and is perceived as beneficial by surgical staff. There are opportunities to refine this stewardship intervention in surgery.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):27–28.

P27. Associated antibiotic use for community infections in British Columbia: A review of prescribing from 2000 to 2018

Ariana Saatchi ^1,^✉, Andrew M Morris ², David M Patrick ^3,⁴, James McCormack ¹, Fawziah Marra ^1,³

Objectives

With 92% of all antibiotics in Canada being used in the community setting, delineating for which indications these antibiotics are used is important. In British Columbia, efforts to curb the use of these essential medications include stewardship campaigns and updated practitioner guidelines. This study examines prescribing for common infections in order to quantify antibiotic use for specific indications in the community and identify new targets for stewardship.

Methods

Prescription data from 2000 to 2018 were extracted from PharmaNet, a system that connects all pharmacies in the community setting and is linked to the Medical Service Plan (MSP) that records claims submitted by family physicians for services provided, using the International Classification of Diseases, Ninth Revision, diagnostic codes. Rates were calculated for upper respiratory tract infection (URTI) and lower respiratory tract infection (LRTI), urinary tract (UTI), skin–soft tissue infections (SSTI), and acute otitis media (AOM). Unlinked antibiotics were defined as prescriptions dispensed without a corresponding MSP record.

Results

In any study year, URTI was the category of infection most prescribed for, with an average of 179 prescriptions dispensed per 1,000 population, followed by UTI, SSTI, AOM, and LRTI. Although prescribing for UTI and SSTI infections remained relatively consistent over the study period, prescribing for cystitis increased by 26% over the study period. By 2018, URTI prescribing decreased across multiple indications: bronchitis (48%), laryngitis (76%), nasopharyngitis (45%), and unspecified URTI (50%). AOM also decreased significantly over the study period (68%) with a rate of 10 prescriptions per 1,000 by 2018.

Conclusion

Outpatient prescribing for URTI has decreased steadily since 2000. Many of these indications do not warrant antibiotic use, and initiatives to improve the use of these medications have been in place since 2005. Further analyses are necessary to evaluate prescribing quality to fully delineate the state of antibiotic use in British Columbia.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):28.

P28. Impact of a reported beta-lactam allergy on cefazolin administration in surgical prophylaxis: Cefazolin is still best, but is it given?

Holly L Hoang ^1,^2,^✉, Katelynn C Crick ³, Justin Z Chen ^1,⁴, Susan R Fryters ⁵, A Uma Chandran ⁴, Alena W Tse-Chang ^4,⁶, David C Williams ^4,⁷, Tyler W Myroniuk ³, Roseanne O Yeung ³, Denise Campbell-Scherer ³, Lynora M Saxinger ^1,⁴

Objectives

Pre-operative surgical antibiotic prophylaxis is an established best practice to minimize surgical site infections (SSIs), with cefazolin the most recommended agent. Patients with a reported beta-lactam (BL) allergy often receive non-BL antibiotics with an associated 50% increased odds of a SSI. Provincial recommendations list cefazolin for surgical prophylaxis unless there is cefazolin allergy or severe non–immunoglobulin E–mediated reaction to a BL, because there is no expected cross-reactivity. We evaluated cefazolin surgical prophylaxis in patients with listed BL allergy to determine current trends and areas for improvement.

Methods

A cross-sectional prospective chart review was conducted from June 4 to August 23, 2019, on surgical procedures performed at five teaching hospitals. Data captured from post-operative charts included patient demographics (age, gender, allergy history), surgical procedure, and details of antibiotic prophylaxis (antimicrobial selection, timing from incision, and duration). Cefazolin indication was assessed by provincial recommendations and absence of contraindications.

Results

There were 3,218 surgical procedures documented, of which 370 (11.5%) involved patients with a BL allergy, 296 (80%) to penicillin. Assessment of adherence to provincial recommendations (by matching procedure codes) was possible in 253 cases; cefazolin was indicated in 204 cases (80.6%) and administered in 152 (74.5%). Multivariable analysis, controlling for gender, site, and surgical specialty, confirmed that BL allergy decreased the likelihood of receiving cefazolin (adjusted odds ratio 0.43; 95% CI 0.30 to 0.61; p <0.001). Appropriate administration of cefazolin to BL-allergic patients differed by site but not by surgical specialty.

Conclusion

The prevalence of reported BL allergy among surgical patients is 11.5%. Although cefazolin is still being administered as recommended to almost 75% of this population, a reported BL allergy is associated with 57% decreased odds of receiving cefazolin. This represents an area in which antimicrobial stewardship interventions can improve surgical prophylaxis practices.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):28–29.

P29. Review of surgical antibiotic prophylaxis practice in adult urology procedures: Opportunities for antimicrobial stewardship collaboration

Justin Z Chen ^1,^2,^✉, Holly L Hoang ^1,³, Katelynn C Crick ⁴, Susan R Fryters ⁵, A Uma Chandran ², David C Williams ^6,², Tyler Myroniuk ⁴, Roseanne O Yeung ⁴, Denise Campbell-Scherer ⁴, Lynora M Saxinger ¹

Objectives

Surgical antibiotic prophylaxis (SAP) is an established best practice in certain urology procedures; however, there is a paucity of data on practice and adherence. The purpose of this study was to assess the quality of SAP selection and duration according to provincial recommendations.

Methods

A prospective cross-sectional chart review of adult urology procedures performed in operating theatres at four major hospitals in a large Canadian city was conducted from June 4 to August 23, 2019. Data captured from chart reviews included patient demographics (age, gender), surgical procedure (incision time and duration), and details of SAP (agent, timing from incision, and duration). Trans-rectal prostate biopsies performed by interventional radiology were excluded.

Results

There were 358 urology procedures reviewed; 318 (93%) cases received SAP. According to provincial SAP recommendations, cefazolin was recommended in 75 procedures, but 14 (20%) did not receive it. This was not influenced by beta-lactam allergy status. In 170 procedures for which cefazolin SAP was not recommended, 48 (28%) received cefazolin. Of 183 cases who received cefazolin SAP, late administration occurred in 13 (7.1%), 3 at incision and 10 after incision. Four cases were eligible for repeat intra-operative dosing (duration of surgery >4 h); 1 of 4 received a repeat dose. Nearly half (135/318) received non–cefazolin-based SAP consisting primarily of ciprofloxacin. This was used predominantly in cystoscopy day procedures. Eighty-nine (66%) of these received one or more post-operative doses.

Conclusion

Nearly half of urology procedures received non–cefazolin-based SAP, with the majority receiving ciprofloxacin in a day procedure setting. Although not recommended, post-operative prophylaxis with ciprofloxacin occurred in two-thirds of procedures. In addition, nearly one-third received cefazolin SAP when it was not recommended. These findings highlight opportunities for antimicrobial stewardship collaboration to optimize SAP practices.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):29.

P30. Real-life experience with ceftobiprole in Canada: Results from the CLEAR (Canadian LEadership on Antimicrobial Real-life usage) Registry

George G Zhanel ^1,^✉, Justin Kosar ², Melanie Baxter ¹, Rita Dhami ³, Sergio Borgia ⁴, Neal Irfan ⁵, Kelly S MacDonald ¹, Gordon Dow ⁶, Philippe Lagacé-Wiens ¹, Maxime Dube ⁷, Marco Bergevin ⁸, Yoav Keynan ¹, Anna Lee ⁹, Andrew Walkty ¹, James Karlowsky ¹

Objectives

Ceftobiprole is an intravenous cephalosporin with broad-spectrum activity and a favourable safety profile. Published data on the clinical use of ceftobiprole are limited. We report on the use of ceftobiprole with Canadian patients using data captured by the CLEAR (Canadian LEadership on Antimicrobial Real-life usage) registry.

Methods

A ceftobiprole usage questionnaire was developed with the input of infectious disease–medical microbiology specialists (physicians and pharmacists) across Canada. The CLEAR registry protocol–questionnaire was submitted and received approval by the University of Manitoba Ethics Committee (April 2019). The CLEAR registry uses the web-based research data management program, REDCap™ (online survey, https://is.gd/CLEAR_ceftobiprole) to facilitate clinicians voluntarily entering details associated with their experiences using ceftobiprole.

Results

Data were available for 42 patients (compiled January 5, 2021). The most common infections treated with ceftobiprole were endocarditis (42.8%), bone and joint infection (26.2%), and hospital-acquired bacterial pneumonia (14.3%). Of the patients, 92.9% had bacteremia, and 23.8% were in the intensive care unit. Ceftobiprole was primarily used as directed therapy for methicillin-resistant Staphylococcus aureus (MRSA) infections (95.2% of patients). Ceftobiprole susceptibility testing was performed in isolates from 50.0% of patients. It was used concomitantly with daptomycin in 57.1% of patients and with vancomycin in 21.4% of patients. Ceftobiprole was used because of failure of (66.6%), resistance to (16.7%), or adverse effects from (11.9%) previously prescribed antimicrobial agents. The dosage regimen was customized in all patients on the basis of their creatinine clearance. Treatment duration was primarily >10 days (64.3% of patients) with microbiological success in 94.7% of patients and clinical success in 83.8%. Overall, 30-day mortality was 9.5%. Of the patients, 2.4% had gastrointestinal adverse effects.

Conclusion

In Canada to date, ceftobiprole is used as directed therapy to treat a variety of severe infections caused by MRSA. It is primarily used in combination with daptomycin or vancomycin, has high microbiological and clinical cure rates, and has an excellent safety profile.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):29–30.

P31. Examination of selection, timing, and duration of surgical prophylaxis for vascular procedures at a major Canadian vascular surgery centre

Holly L Hoang ^1,^2,^✉, Gerrit B Winkelaar ^3,⁴, Katelynn C Crick ⁵, Justin Z Chen ^1,⁶, Susan R Fryters ⁷, A Uma Chandran ⁶, David C Williams ^3,⁶, Tyler W Myroniuk ⁵, Rose O Yeung ⁵, Denise Campbell-Scherer ⁵, Lynora M Saxinger ^1,⁶

Objectives

Preoperative surgical antibiotic prophylaxis (SAP) is an established best practice to minimize surgical site infections, yet data for vascular procedures are limited, particularly with respect to duration of antibiotics. The aim of this study was to investigate SAP in vascular procedures at a large Canadian centre.

Methods

A cross-sectional prospective chart review was conducted from June 4 to August 23, 2019, for surgical procedures performed at a teaching hospital that includes a large vascular surgery centre. Data captured from chart reviews included patient demographics, surgical procedure, and details of antibiotic prophylaxis (antimicrobial selection, timing from incision, and duration).

Results

There were 103 adult vascular procedures reviewed, of which assessment for recommended surgical prophylactic agent, by procedure codes, was possible in 64 cases. According to provincial recommendations, all 64 procedures warranted cefazolin prophylaxis, and 60 of 64 (94%) received cefazolin appropriately. Five of the 64 patients had a non-cefazolin beta-lactam allergy, and 4 of these were appropriately prescribed cefazolin. Timing of cefazolin administration was appropriate in 94% (87/92) of procedures for which timing data were available, with a median administration time of 24 (IQR 17 to 35) minutes before incision. Twenty-one procedures exceeded 4 hours in duration, qualifying for repeat intraoperative antibiotic dosing; 15/21 (71.4%) received a second dose of cefazolin. There were no intraoperative concerns for infection in 89 of the procedures, yet 57 (65.2%) received extended surgical prophylaxis for 24 hours or more.

Conclusion

Surgical prophylaxis in vascular procedures adheres to provincial recommendations with respect to selection and timing. One opportunity for improvement is in the re-dosing of cefazolin for procedures longer than 4 hours. Administration of cefazolin beyond 24 hours occurred in more than 50% of vascular procedures, representing an area for further study given current equipoise on duration of surgical prophylaxis in vascular surgery.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):30.

P32. Audit of surgical antibiotic prophylaxis in orthopedic surgery with particular focus on duration of prophylaxis

Susan R Fryters ^1,^✉, Justin Z Chen ², A Uma Chandran ³, Holly L Hoang ², Lynora M Saxinger ², Katelynn C Crick ⁴, Tyler W Myroniuk ⁴, David C Williams ⁵, Roseanne O Yeung ⁴, Denise Campbell-Scherer ⁴, Alena W Tse-Chang ⁶

Objectives

Surgical antibiotic prophylaxis (SAP) is an established best practice to minimize surgical site infections. There is increasing evidence for single-dose SAP in orthopedic procedures because of its equivalent efficacy and the decreased risk of antibiotic-related complications. SAP data were collected to assess adherence to provincial recommendations and trends in duration of surgical prophylaxis.

Methods

A prospective chart review of a convenience sample of orthopedic procedures was conducted from June 4 to August 23, 2019, at four adult teaching hospitals. Data collected included patient demographics, surgical procedure (including incision time and duration), and details of SAP (antimicrobial allergy, selection, timing from incision, and duration).

Results

There were 699 adult orthopedic procedures reviewed, of which 530 were assessable. Cefazolin was the recommended agent in 475 procedures; 453 (95%) received cefazolin. There were 55 procedures for which cefazolin was not indicated but 47 (85.4%) received cefazolin. Of the 475 patients, 55 (11.6%) had a reported beta-lactam allergy, and 44 (80%) of them were appropriately prescribed cefazolin. Timing of cefazolin administration was appropriate in 92% of procedures. Fourteen procedures exceeded 4 hours in duration, qualifying for repeat intra-operative antibiotic dosing; 6 (42.9%) of these received an intra-operative dose. Post-operative SAP was ordered in 478 of 699 procedures (68.4%); 5.5% received one additional dose, 15.1% received two doses, and 79.5% received ≥24 hours of cefazolin. Data on duration of prophylaxis by specific surgical procedure will be described.

Conclusion

SAP for orthopedic procedures adheres to provincial recommendations with respect to antibiotic selection and timing. Opportunities for improvement in orthopedic SAP include antibiotic re-dosing for procedures lasting longer than 4 hours and eliminating antibiotic prophylaxis for those procedures for which it is not recommended. Evolving orthopedic literature indicates that postoperative SAP is unnecessary; this has yet to become standard practice locally, thus representing an area for antimicrobial stewardship collaboration.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):30–31.

P33. Assessment of surgical antibiotic prophylaxis practices at a Canadian tertiary pediatric centre

Alena W Tse-Chang ^1,^2,^✉, Katelynn C Crick ³, Susan R Fryters ⁴, Uma Chandran ^2,⁵, David C Williams ^2,⁶, Tyler W Myroniuk ³, Roseanne O Yeung ³, Denise Campbell-Scherer ³, Lynora M Saxinger ^2,⁷

Objectives

Surgical antibiotic prophylaxis (SAP) decreases the incidence of surgical site infections. However, SAP recommendations in pediatrics are generally based on expert opinion or inferred from those for adults. We aimed to describe SAP practices at our institution and assess antibiotic selection, timing, and duration according to provincial recommendations.

Methods

A prospective chart review was conducted from June 4 to August 23, 2019, on surgical procedures performed at Canadian tertiary pediatric centre. Data collected included patient demographics (age, gender, allergies), surgical procedure (incision time and duration), and SAP details (antimicrobial selection, timing, and duration).

Results

Of the 166 procedures included, the majority were otolaryngology procedures (27.1%; 45/166), followed by oral surgery (17.5%), general surgery (17%), cardiac surgery (12%), plastic surgery (10%), urology (9%), orthopaedic surgery (6%), and neurosurgery 91%). Overall, 34.3% received SAP.

Sixty-eight procedures (41%) were assessable for whether cefazolin SAP was required. Cefazolin was indicated in 38 procedures and was given appropriately 60.5% (23/38) of the time. Cefazolin was not indicated in 30 procedures but was given in 5 (16.7%). Beta-lactam allergy was reported in 16 of 166 procedures (9.6%), with 3 of 4 receiving cefazolin as recommended.

Pre-operative timing was appropriate (within 1 h before incision) in 84.3% (43/51) of procedures. Six procedures warranted intra-operative re-dosing; 4 (66.7%) appropriately received a second dose, but 8 cardiac surgery patients were reposed within 2 hours after incision. Thirty-two patients (19.3%) received post-procedural SAP, which was extended for 24 hours or more in 11 (34.4%) of the cases.

Conclusion

The timing of SAP in pediatric cases can be optimized, and post-procedural SAP is frequent, representing opportunities for antimicrobial stewardship collaboration. A large proportion of the cases were not assessable against our provincial adult recommendations because there were no specific recommendations listed, demonstrating a need for pediatric-specific recommendations.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):31–32.

P34. Comparison of contact precautions with routine practices for the prevention of vancomycin-resistant enterococci infections in hospitals: The secret is in the routine

Heather Glassman ^1,^✉, Laura E Burnes ¹, Alexandra C McFarlane ^1,², Aruna Uma Chandran ^1,²

Objectives

Contact precautions are often implemented in hospital settings to prevent the transmission of vancomycin-resistant enterococci (VRE). Drawbacks of contact precautions include financial cost and impact on quality of patient care. We reviewed the literature to determine whether contact precautions, in comparison with infection control routine practices, have been shown to reduce VRE infection incidence rates in hospitals in non-outbreak settings.

Methods

Published literature from sources including MEDLINE, PubMed, and reference lists were searched for articles evaluating rates of VRE infection with use of contact precautions compared with routine practices. The search was limited to English language publications with human subjects of all study designs, apart from mathematical models. Studies that stopped contact precautions in colonized (but not infected) patients, or that used different comparators, were excluded.

Results

Thirteen studies, including two systematic reviews, were included in the final analysis. Eleven studies did not demonstrate a statistically significant difference in VRE infection rates between routine practices and contact precautions. One study showed an increase in VRE infection rates with cessation of contact precautions in the malignant hematology patient subgroup only. Another study showed a population-wide increased rate of VRE bacteremia after stopping isolation procedures for VRE.

Conclusion

Of the reviewed studies, most showed no difference in VRE infection rates after discontinuing standard use of contact precautions for VRE. Limitations include differences in study design, degree of surveillance, and duration of follow-up. We propose that discontinuation of contact precautions for VRE in non-outbreak settings is a reasonable approach for institutions to consider on the basis of local epidemiology and infection control practices. This change should be accompanied by organizational support of, and investment in, more robust horizontal approaches that incorporate routine practices and antimicrobial stewardship, as well as a structured surveillance plan to monitor for adverse clinical outcomes.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):32.

P35. What’s in a number? The value of titres as routine proof of immune protection

Natalie C Marshall ^1,^2,^✉, Ashley-Nicole M Bailey ¹, L Alexa Thompson ¹, Jamil N Kanji ^1,^2,³, Carmen L Charlton ^1,²

Objectives

Vaccination is an essential public health effort against preventable infectious diseases, such as measles, mumps, and rubella (MMR). Those at higher risk of exposure (eg, health care workers) require proof of immune protection, for which the gold standard is documentation of vaccine doses received, not lab tests. However, a recent increase in test requests suggests that some medical schools require quantitative serological tests (ie, titres) to demonstrate immunity. In Canada, titres are not performed for this indication, thereby compromising Canadian students’ applications. Here, we investigated the prevalence of medical schools requiring titres for admission and the availability of such tests.

Methods

We reviewed and compared MMR immunization guidelines, literature, and admission requirements for the 17 Canadian and 63 American medical schools admitting Canadian students according to the Association of American Medical Colleges. Finally, we surveyed all provincial public health laboratories to assess how titre requests for MMR immunity are handled across Canada.

Results

To demonstrate immunity to MMR viruses, 100% of Canadian medical schools considered documentation of vaccine doses sufficient, in concordance with Canada’s National Advisory Committee on Immunization guidelines. In contrast, 15% of eligible US medical schools required immunoglobulin G (IgG) titres, conflicting with both US and Canadian national guidelines. Moreover, no responding Canadian public health laboratory would approve quantitative IgG testing for this indication.

Conclusion

To document MMR immunity, the admission requirements of at least nine US medical schools conflict with national guidelines. Moreover, the required test is neither recommended nor available for this purpose, according to national guidelines and laboratories. Accordingly, qualified Canadian students are unable to meet the admission criteria of these medical schools. Until these requirements are updated, Canadian applicants must use out-of-country, pseudo-quantitative tests that do not align with current guidelines.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):32–33.

P36. A case of multisystem inflammatory syndrome in adults: A rare manifestation of COVID-19

Teagan King ^1,^✉, Jordan Mah ², Andrew Johnson ², Marvin Fritzler ^1,³, Bryan Yipp ⁴, Mark Gillrie ^2,⁵

Objectives

Multisystem inflammatory syndrome in adults (MIS-A) is a rare and severe complication of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). MIS-A is defined by multi-system organ failure with shock, cardiac dysfunction, or abdominal pain in adults aged older than 21 years, with elevated inflammatory markers, positive coronavirus disease 2019 (COVID-19) serology or polymerase chain reaction (PCR) within 12 weeks, and the absence of severe respiratory disease. We describe a case of MIS-A related to SARS-CoV-2 in a 38-year-old man of Bangladeshi origin presenting to a tertiary care hospital in Canada.

Methods

We summarized the clinical case and epidemiology of MIS-A cases related to SARS-CoV-2 through a targeted literature review.

Results

On day 1, the patient tested positive for COVID-19 on PCR. He presented to hospital on day 25 with 4 days of fever, epigastric pain, diarrhea, macular rash, chest pressure, and bilateral conjunctivitis. Initial labs showed marked neutrophilic leukocytosis, elevated creatinine, c-reactive protein, international normalized ratio, B-natriuretic peptide, and troponin. Transthoracic echo revealed severe biventricular dysfunction, initially requiring vasoactive support. Serology was strongly positive to multiple SAR-CoV-2 antigens with elevations in serum CCL2, interleukin (IL)-6, IL-10, and IP-10. Extensive infectious, cardiac, and inflammatory work-up was conducted and was negative for alternative diagnoses; ultimately, he was clinically diagnosed with MIS-A. He received 5 days of oral dexamethasone, no other therapeutic agents, and was discharged home on day 40.

According to a US Centers for Disease Control and Prevention case series, MIS-A is being described in patients aged 21–50 years, with limited past medical history, 2–5 weeks after initial COVID-19 infection. Most cases occur among ethnic minorities, with half never experiencing respiratory symptoms. Many cases required intensive care; however, most patients (24/27) survived with supportive care or empiric immunomodulatory therapies, including steroids and intravenous immunoglobulin.

Conclusion

MIS-A is a rare and severe complication of SARS-CoV-2, with no proven therapies. Our patient had all the cardinal features of MIS-A plus marked leukocytosis. A high degree of clinical suspicion is required to recognize MIS-A.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):33.

P37. Absence of transmission of NDM and OXA-48–like carbapenemase genes in a long-term-care facility

Carl Boodman ^1,^✉, Natalie Gibson ², Davenna Conrod ², Christine Turenne ¹, David Alexander ¹, David Boyd ³, Laura Mataseje ³, Michael Mulvey ³, James Karlowsky ⁴, Molly Blake ², John Embil ¹

Objectives

To evaluate the transmission of carbapenemase genes NDM and OXA-48 in a long-term-care facility (LTCF) after 14 months of routine practices. The status of the index case, colonized with a carbapenemase-producing Enterobacteriaceae (CPE), was originally unknown to the LTCF and thus additional infection prevention and control (IPC) precautions were not implemented.

Methods

Contacts of the index case underwent screening for CPE colonization using rectal swabs collected on days 0, 7, and 21 after the exposure was identified. Swabs were cultured on CHROMagar mSuperCARBA agar (CHROMagar, Paris, France) and gram-negative, carbapenem-resistant microorganisms were identified to species level using matrix-assisted laser desorption/ionization-time of flight mass spectrometry (Bruker Daltonics; Billerica, MA, USA). Phenotypic confirmation of carbapenemase production by Enterobacteriaceae isolates was performed using the Neo-Rapid CARB kit (Rosco Diagnostica; Taastrup, Denmark). Analysis of the index CPE isolate as well as other CPE isolates was performed using whole-genome sequencing (WGS) to establish whether the CPE isolates were identical.

Results

Of 19 screened contacts, 1 was colonized with a CPE. The presence of OXA-48 was discovered by polymerase chain reaction. WGS of the index isolate and the contact isolate revealed that the contact harbored an OXA-181, an OXA-48–like element differing by four amino acids but causing a positive PCR result, whereas the index case harbored a true OXA-48 by WGS. This established a lack of relation to OXA-48 or NDM of the index case. The discovery of the OXA-181 isolate led to another 11 contacts screened for CPE. Among these, CPE was not recovered.

Conclusion

No transmission of NDM and OXA-48 genes occurred among contacts of a CPE-positive index case after 14 months of routine IPC practices in a LTCF.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):33–34.

P38. Sensitivity of Cobas® 68/8800 SARS-CoV-2 test using gargle specimens with two different processing methods

Patrick Benoit ^1,^✉, Annie-Claude Labbé ^1,^2,³, Linda Lalancette ⁴, Simon Gagnon ², Eric Bonneau ⁴, Judith Fafard ⁵, Stéphanie Beauchemin ², Angie Jetté ², Sonia Tremblay ⁴, Karine Filiatrault ⁴, Fabiola Vancol-Fable ², François Coutlée ^1,²

Objectives

In the context of high-volume severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing, autonomous, convenient, and non-invasive specimen collection, such as gargling, could replace oral and naso-pharyngeal swabs (ONPS) in some settings but requires further evaluation. Our objective was to compare the performance of gargle specimens in Cobas® PCR Media or in lysis buffer with that of an ONPS in Cobas PCR Media on the Cobas 6800 and 8800 platforms.

Methods

Participants were recruited prospectively in two designated coronavirus disease 2019 (COVID-19) screening clinics. They were asked to gargle with 5 mL of Amaro (Boisbriand; n = 300) or Eska (Montreal; n = 347) water in addition to the standard ONPS. The gargle specimen was split as follows: 1 mL was added to 4.3 mL of Cobas PCR Media and 400 mL was added to 200 mL of lysis buffer. Testing was performed on the Cobas 6800 or 8800 platform according to the package insert. An individual was considered infected if a positive result was obtained for at least one of the three samples.

Results

Among 647 participants, 3 were excluded because an invalid result was obtained for at least one sample. Overall, 134 (20.7%) were infected. Sensitivity was 96.3% (95% CI 91.3 to 98.5) for ONPS, 89.6% (95% CI 83.1 to 93.7) for gargle directly in Cobas PCR Media, and 91.0% (95% CI 84.7 to 94.8) for gargle in lysis buffer. For positive samples, mean cycle threshold (Ct) for E-gene target was lower for ONPS (23.4) than gargle in Cobas PCR Media (29.9) and in lysis buffer (28.9).

Conclusion

Sensitivity for gargle specimens was slightly, but not significantly, lower than for an ONPS on Cobas 68/8800 SARS-CoV-2 test. Discordant results were obtained from samples with higher Ct values. Both methods for processing the gargle sample before testing on the Cobas 6800 or 8800 platforms were equivalent.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):34.

P39. Evaluation of the Q score for assessing wound swab quality in a tertiary care centre

Jessica D Forbes ¹, Shawn T Clark ^1,^✉, Larissa M Matukas ^1,²

Objectives

Wound swabs submitted for routine culture may be contaminated with commensal microbiota from improper collection technique. This may complicate downstream microbiological analyses. The Q score is a numerical rating of wound swab quality that can be used as a tool to guide swab processing, workup, and reporting. This scoring system uses the relative abundance of polymorphonuclear cells to epithelial cells seen in a Gram smear to deem a specimen acceptable for culture. The objective of our study was to apply the Q score to retrospectively collected wound swab cultures and predict its impact on the workup and reporting of potential pathogens in a tertiary care centre.

Methods

We performed a retrospective survey of wound swabs processed for routine culture by the microbiology laboratory at a large tertiary care centre between November 29, 2018, and December 23, 2019.

Results

We analyzed 968 wound swab cultures. Adopting modified Q score criteria, 67 swabs (6.9%) would be considered low quality (Q0). Variability in swab quality was observed across different hospital units, with the highest frequency of Q0 swabs found among family medicine clinics (8.0%) and inpatient units (7.4%). Lower extremity swabs were the most commonly submitted swab (n = 476; 7.1% Q0), with perineal–anal swabs (n = 8; 24.4% Q0) often being low quality. A total of 1,419 isolates were reported from these swabs, with 1,020 having a complete ID and AST workup. With Q score, only 944 isolates would have had ID and AST workup on the basis of swab quality, a reduction of 76 isolates.

Conclusion

Our Q score-based approach identified few low-quality wound swabs over a 390-day period. Given the low frequency of poor-quality swabs at this centre, stakeholders should be engaged to determine whether this rate of poor-quality swabs is sufficient to make it a target for change.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):34–35.

P40. Susceptibility patterns for Enterococcus spp clinical isolates collected over a 14-year period

Victoria Bugaj ^1,^✉, Sandra AN Walker ^1,², Christine Peragine ²

Objectives

Enterococci are commensal gram-positive cocci that are a common cause of urinary tract infections, endocarditis, and intra-abdominal, skin, soft tissue, and wound infections. Although E. faecalis remains the most common pathogenic enterococci, the frequency of E. faecium infections have increased globally. This research investigated the antimicrobial resistance patterns of enterococcus spp clinical isolates collected over a 14-year study period.

Methods

A retrospective review of susceptibility data for Enterococcus spp clinical isolates collected from patients admitted between October 2002 and September 2016 was completed through data extraction from the microbiology database. Trends in susceptibility to ampicillin, ciprofloxacin, gentamicin, nitrofurantoin, tetracycline, and vancomycin were analyzed using linear regression models with a significance level of <0.05.

Results

Among the 2,617 Enterococcus isolates, 57% (n = 1,486) were E. faecalis, 30% (n = 775) were E. faecium, and the remaining 14% (n = 356) were classified as other Enterococcus species. The majority of isolates came from blood cultures (n = 1,176; 45%). Susceptibility trends were assessed from 2008 forward. Of the E. faecalis and E. faecium, <1% and 10%, respectively, were vancomycin resistant. For the aggregate of all Enterococcus isolates, the sensitivity to ampicillin and nitrofurantoin decreased each year by 3.2% (p = 0.0055) and 3.4% (p = 0.006), respectively. For E. faecalis, susceptibilities for ciprofloxacin (+3.9% susceptible/y; p = 0.0117), gentamicin (+3.9% susceptible/y; p = 0.0144), and tetracyclines (+7% susceptible/y; p = 0.42) increased over time. Conversely, nitrofurantoin susceptibility increased by 24% each year (p = 0.0013) for E. faecium. Although E. faecium susceptibility to tetracyclines (−4.5% susceptible/y) demonstrated a noteworthy decline, it was not statistically significant (p = 0.562).

Conclusion

Enterococcal antibiotic susceptibility improved during the study period. However, E. faecium infections constituted a substantial proportion (30%) of enterococcal clinical isolates during the study period. E. faecalis and E. faecium demonstrated heterogeneous susceptibility patterns, supporting species-targeted antimicrobial stewardship interventions for Enterococcus.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):35.

P41. Maternal and fetal infection of non-typhoidal Salmonella during pregnancy

Xena X Li ^1,^2,^✉, Carson K Lo ¹, Jeffrey Pernica ³, Philippe El-Helou ¹

Objectives

Salmonella species are known to cause invasive metastatic disease after gastroenteritis. Although non-invasive Salmonella infections often do not require antibiotics, there is controversy regarding the need to treat pregnant women. We aim to review the literature and describe our case, highlighting the need to identify and treat invasive Salmonella infections in pregnancy to improve maternal and fetal outcomes.

Methods

A healthy 18-year-old woman whose baby was at 36 weeks gestational age presented with 4 weeks of intermittent nausea and diarrhea without fever or systemic symptoms. She had no history of travel, animal exposure, or sick family contacts. She did attend a wedding in the preceding month. Her stool culture resulted in Salmonella spp, and she was initially given oral azithromycin. Later, blood cultures resulted in Salmonella enterica subsp. enterica serovar Schwarzengrund, and she was treated with 14 days of intravenous ceftriaxone.

Results

Subsequently, the patient became febrile while on ceftriaxone, and her baby was delivered by emergency caesarean section for fetal tachycardia. The baby initially showed no signs of infection and was discharged without antibiotics; however, the baby was re-admitted on day of life 23 with fevers, tachycardia, and diarrhea. Stool culture was positive for the same Salmonella spp A blood culture was negative, the baby’s cerebrospinal fluid was normal, and a full-body MRI showed no osteomyelitis. The baby was treated with ampicillin–ceftriaxone for 14 days. There are few reported cases of intra-amniotic Salmonella infections. Although the mothers generally have good outcomes with treatment, fetal prognosis is worse, including fatal outcomes.

Conclusion

There are insufficient data regarding the treatment of Salmonella infections during pregnancy. Our case demonstrates the heightened need to identify invasive Salmonella in pregnant patients who may not present with systemic illness. Moreover, what specific measures to prevent neonatal infection should be taken in mothers with invasive Salmonella infections is not clear.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):35–36.

P42. Real-life experience with ceftolozane–tazobactam in Canada: Results from the CLEAR Registry

George G Zhanel ^1,^✉, Rita Dhami ², Melanie Baxter ¹, Justin Kosar ³, Carlos Cervera ⁴, Neal Irfan ⁵, Rosemary Zvonar ⁶, Sergio Borgia ⁷, Jean-Francois Tessier ⁸, Gordon Dow ⁹, Rob Ariano ¹⁰, Maxime Dube ¹¹, Michel Savoie ⁸, Sarah Lutes ¹², Andrew Walkty ¹, James Karlowsky ¹

Objectives

Ceftolozane–tazobactam is an intravenous cephalosporin–beta-lactamase inhibitor combination with activity against gram-negative bacilli. We report the use of ceftolozane–tazobactam in Canada using a national registry.

Methods

A ceftolozane–tazobactam usage questionnaire was developed with input from infectious disease and medical microbiology specialists (physicians and pharmacists) across Canada. The questionnaire was submitted to and received approval by the University of Manitoba Ethics Committee (April 2019). The CLEAR (Canadian LEadership on Antimicrobial Real-life usage) registry uses the web-based research data management program, REDCap™ (online survey, https://is.gd/CLEAR_ceftolozanetazobactam) to facilitate clinicians voluntarily entering details of their clinical experiences using ceftolozane–tazobactam.

Results

Data were available for 53 patients (January 5, 2021) treated with ceftolozane–tazobactam. The most common infections treated were hospital-acquired bacterial pneumonia (35.8% of patients), ventilator-associated bacterial pneumonia (17.0%), bone and joint infection (11.3%), complicated intra-abdominal infection (7.5%), and complicated skin or skin structure infection (7.5%). Of the patients, 17.0% had bacteremia; 48.1% were in an intensive care unit. Ceftolozane–tazobactam was used primarily as directed therapy for Pseudomonas aeruginosa infections (94.3% of patients). Ceftolozane–tazobactam susceptibility testing was performed on isolates from 88.7% of patients. Ceftolozane–tazobactam was used in combination with another antimicrobial active versus gram-negative bacilli in 39.6% of patients (aminoglycosides, 17.0%; fluoroquinolones, 7.5%; and colistin–polymyxin B, 7.5%). Ceftolozane–tazobactam was used because of resistance to (86.8%), failure of (11.3%), or adverse effects from (1.9%) previously prescribed antimicrobial agents. Treatment duration was primarily >10 days (60.4% of patients), with 62.5% microbiological success and 66.0% clinical success. Overall 30-day mortality was 13.2%. Of patients, 7.5% had adverse effects not requiring drug discontinuation.

Conclusion

In Canada, ceftolozane–tazobactam is used as directed therapy to treat various severe infections caused by multi-drug-resistant P. aeruginosa. It is commonly used in combination with other antimicrobials, is associated with relatively high microbiological–clinical cure rates, and has an excellent safety profile.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):36–37.

P43. Risk factors for severe outcomes in patients hospitalized with COVID-19 in a network of Canadian acute-care hospitals

Robyn Mitchell ^1,^✉, Kelly Baekyung Choi ¹, Linda Pelude ¹, Wallis Rudnick ¹, James Brooks ¹, Jeannette Comeau ², John Conly ³, Chelsey Ellis ⁴, Jennifer Ellison ⁵, John Embil ⁶, Gerald A Evans ⁷, Gregory J German ⁸, Lynn Johnston ⁹, Jennie Johnstone ¹⁰, Kevin C Katz ¹¹, Pamela Kibsey ¹², Bonita E Lee ¹³, Marie-Astrid Lefebvre ^14,¹⁵, Yves Longtin ¹⁶, Allison McGeer ¹⁰, Dominik Mertz ¹⁷, Jessica Minion ¹⁸, Caroline Quach ¹⁹, Stephanie Smith ²⁰, Jocelyn Srigley ²¹, Paula Stagg ²², Kathryn N Suh ²³, Alice Wong ²⁴, Nisha Thampi ²⁵, Charles Frenette ¹⁵

Objectives

To describe risk factors associated with severe outcomes among patients hospitalized with coronavirus disease 2019 (COVID-19).

Methods

We analyzed data from 2,583 adult and pediatric laboratory-confirmed COVID-19 cases between March 1 and November 30, 2020, from 55 Canadian acute-care hospitals in 10 provinces and one territory. Multivariable logistic regression was used to assess associations among demographics, medical conditions, symptoms, source of acquisition with intensive care unit (ICU) admission, and 30-day all-cause mortality.

Results

Patients hospitalized with COVID-19 were older (median age 70 years, interquartile range 53–82 y), and 86.2% (2,205/2,557) had medical conditions. Few pediatric COVID-19 hospitalizations were reported (3.2%; 82/2,582). Twenty-three percent of hospitalized cases (585/2,582) were admitted to the ICU, and of those, 60.4% received mechanical ventilation (348/576). Among all hospitalized cases, 30-day all-cause mortality was 18.3% (472/2,578). Independent factors associated with ICU admission were ages 40–59 years (adjusted odds ratio [aOR] 1.61), 60–79 years (aOR 1.80), and ≥80 years (aOR 0.55) versus 0–39 years, male sex (aOR 1.48), community acquisition of COVID-19 (aOR 2.91), diabetes (aOR 1.46), fever (aOR 1.56), and shortness of breath (aOR 2.24). Fewer older patients (≥80 y) were admitted to ICU, likely as a result of advance planning directives. Independent factors associated with 30-day all-cause mortality were ages 40–59 years (aOR 1.83), 60–79 years (aOR 9.05), and ≥80 years (aOR 31.29) versus 0–39 years, male sex (aOR 1.55), health care acquisition of COVID-19 (aOR 2.93), kidney disease (aOR 1.85), shortness of breath (aOR 2.64), receipt of mechanical ventilation (aOR 2.80), and ICU admission (aOR 1.83).

Conclusion

A substantial proportion of patients hospitalized with COVID-19 received critical interventions and had poor outcomes. The strongest predictor of 30-day all-cause mortality was increasing age. Prevention of infections among vulnerable populations, especially the elderly, through public health measures including vaccination is a priority to reduce deaths due to COVID-19.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):37–38.

P44. Reliability of E-gene cycle threshold values in interpreting duration of COVID-19 infection

William Stokes ^1,^2,^3,^✉, Jamil Kanji ^1,^2,³, Jia Hu ^4,⁵, Nathan Zelyas ^2,⁶, Byron Berenger ^7,⁸

Objectives

There is growing interest in using cycle threshold (Ct) values generated by real-time polymerase chain reaction (PCR) assays to determine infectiousness and timing of an individual’s coronavirus disease 2019 (COVID-19) infection. We sought to determine the correlation between E-gene Ct values from the Alberta Public Health Laboratory’s lab-developed test (LDT) for COVID-19 and duration of symptoms.

Methods

Provincial data recorded from March to May 2020 from the Alberta Health Services Public Health and Alberta Precision Laboratory databases were linked and analyzed. Symptom duration at the time of collection was determined during case investigation by Public Health. Only E-gene Ct values from the LDT were included. Specimens included nasal swabs, throat swabs, naso-pharyngeal swabs, and endotracheal tube aspirates.

Results

There were 7,974 positive COVID-19 cases observed in Alberta during this time period. Of those, 5,756 had symptom onset provided and were performed using the LDT. At time of collection, 787 (13.7%) were classified as asymptomatic; 92 (1.6%), pre-symptomatic (defined as symptom onset within 48 hours after collection); 3,107 (54.0%), with symptom onset ≤7 days; and 1,770 (30.7%), with symptom onset >7 days. When excluding day 1 of symptom onset, there was a linear trend toward higher median Ct values with increasing days of symptom onset (R² = 0.970; p <0.001; Figure P44-1). However, Ct values ranged widely, regardless of symptom onset. For example, 25% and 10% of individuals with symptoms ≤7 days had Ct values >29.1 and >32.8, respectively, whereas 25% and 10% of individuals with symptom onset >7 days had Ct values <24.3 and <20.3.

Figure P44-1: — Scatter plot of E gene Ct value per symptom onset day *Note: Red dots represent median E gene Ct values. From day 2–14, R2 = 0.970, p<0.001*

Ct = Cycle threshold

Conclusion

Although there was a linear trend toward increasing Ct values with duration of symptom onset at time of specimen collection, there is a wide enough variation such that a Ct value alone cannot reliably predict an individual’s date of symptom onset or time of infection.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):38.

P45. Susceptibility of Pseudomonas aeruginosa to empiric therapy and alternative beta-lactam antimicrobials in intensive care unit patients with respiratory tract infections in Canadian centres: SMART surveillance (2016–2019)

Radwan El Ali ^1,^✉, Marcela Gonzalez ², Christiane Ghakis ¹, Emma O’Callaghan ¹

Objectives

To describe trends in susceptibility of Pseudomonas aeruginosa in Canadian tertiary centres and to evaluate co-resistance to both empiric and early switch beta-lactams in intensive care unit patients, aged 20 years and older, with respiratory tract infections (RTI).

Methods

Data were collected from seven Canadian health centres between 2016 and 2019 as part of the SMART surveillance program. Susceptibility testing (mg/L) was performed by broth microdilution methods. Applied interpretive criteria to determine susceptibility were Clinical and Laboratory Standards Institute breakpoints for meropenem (MEM), cefepime (CEP), ceftazidime (CAZ), piperacillin–tazobactam (TZP), and ceftolozane–tazobactam (CT).

Results

A total of 1,176 P. aeruginosa isolates were collected in 2016–2019, of which 223 were of respiratory source. Although susceptibility for CEP, CAZ, MEM, and TZP was below 70% throughout the years (except for MEM in 2016), CT maintained a high susceptibility of more than 90% with an average susceptibility of 93% between 2016 and 2019. Further analysis of 2019 data showed P. aeruginosa co-resistance among empiric therapy and first-alternative antibiotics. If P. aeruginosa was non-susceptible to TZP, there was 21% or less susceptibility for CEP, CAZ, and MEM. However, CT offered more than 80% in vitro coverage (Table P45-1).

Table P45-1:

2019 respiratory isolates

	TZP NS	MEM NS	CAZ NS
Non-susceptible, % of total	41	43	39
CEP, % susceptibility	21	30	15
CAZ, % susceptibility	10	30	0
CT, % susceptibility	83	83	81
MEM, % susceptibility	17	0	22
TZP, % susceptibility	0	20	4

Open in a new tab

TZP = Piperacillin–tazobactam; MEM = Meropenem; CAZ = Ceftazidime; CEP = Cefepime; CT = Ceftolozane–tazobactam

Conclusion

CT sustained high coverage for P. aeruginosa across the years of this study for patients with respiratory tract infections. Moreover, it offered more than 80% in vitro coverage when the isolate is resistant to empiric therapy and early alternative beta-lactams. The clinical benefits of such findings need to be further explored.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):38–39.

P46. Naso-pharyngeal expression of angiotensin-converting enzyme 2 and transmembrane serine protease 2 in children compared with adults in family clusters exposed to COVID-19

Mohammad Rubayet Hasan ^1,^2,^✉, Muneera Naseer Ahmad ¹, Alaa Al Hashemi ¹, Patrick Tang ^1,²

Objectives

Evidence is accumulating that children have lower levels of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) host entry factors, angiotensin-converting enzyme 2 (ACE2) and transmembrane serine protease 2 (TMPRSS2), in their nasal epithelium compared with adults, which may be related to a lower incidence of COVID-19 compared with adults. However, no direct evidence is available to support this hypothesis. In this study, we compared the transcript levels of ACE2 and TMPRSS2 in naso-pharyngeal (NP) swabs from children and adult members in COVID-19–exposed families in relation to their infection status.

Methods

Residual NP swab specimens (N = 207) from pediatric patients and their adult family members submitted for COVID-19 testing at Sidra Medicine, Qatar, were assessed for ACE2, TMPRSS2, and b-actin gene expression using predesigned Taqman assays (Thermofisher). Only families that had one or more members positive for SARS-CoV-2 by quantitative reverse transcription polymerase chain reaction of NP swabs were included.

Results

ACE2 and TMPRSS2 expression normalized to the expression of endogenous control b-actin was higher in adults than children (n = 115 adults and 92 children; p <0.05). The expression of these genes was not significantly different between COVID-19–positive and –negative patients of all ages or within the same age groups. Using paired data from families with a history of exposure to SARS-CoV-2, both ACE2 and TMPRSS2 expression were significantly higher among COVID-19–positive adults than among COVID-19–negative children (n = 47 pairs; p = 0.0004 and 0.0049 by two-tailed, paired T-test for ACE2 and TMPRSS2, respectively). However, ACE2 and TMPRSS2 expression between COVID-19–positive adults and COVID-19–positive children (n = 17 pairs) or COVID-19–negative adults and COVID-19–positive children (n = 28 pairs) were not significantly different.

Conclusion

Our results provide evidence in favour of the hypothesis that children with lower expression of ACE2 and TMPRSS2 remain COVID-19 negative despite being exposed to a COVID-19–positive adult family member.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):39.

P47. Association of frailty and outcomes among hospitalized COVID-19 patients in Toronto

Altynay Shigayeva ¹, Melissa K Andrew ², Kevin Katz ^3,⁴, Andrew Simor ^3,⁵, Don Melady ¹, Eric Coomes ^3,⁶, Brenda Coleman ^1,³, Jeffrey Kwong ^3,⁷, Allison McGeer ^1,³, Christopher Kandel ^1,^3,^✉

Objectives

We investigated the association between frailty and case fatality, intensive care unit (ICU) admission, and hospital length of stay (LOS) among hospitalized coronavirus disease 2019 (COVID-19) patients.

Methods

This population-based cohort study included 950 hospitalized adults (aged ≥16 y) with COVID-19. Demographic and medical data were abstracted by chart review. Multivariable Cox regression was performed to identify factors associated with mortality and ICU admission. Frailty was assessed using the Clinical Frailty Scale (CFS), Frailty Index (FI), and Laboratory FI (Lab-FI).

Results

Of 950 patients, 514 (54.1%) were male, median age was 68 years (IQR 21 to 81), median Charlson Comorbidity Index (CCI) was 1 (IQR 0 to 3). One hundred eighty-five (19.5%) were nursing home residents (NHRs). Overall, 30.5% (290/950) had an advance directive precluding ICU admission, including 69% (192/277) of those aged >80 years and 73.5% (136/185) of NHRs. Median CFS score was 4 (IQR 3 to 6), median FI was 0.12 (IQR 0.06–0.32), median Lab-FI was 0.26 (IQR 0.18–0.36). Frailty and CCI increased with age. Overall, 26.8% (255/950) died after a median of 7 days (IQR 4–16). Mortality was associated with age (aHR = 1.04; 95% CI 1.03 to 1.05/1 y increase), male sex (aHR = 1.45; 95% CI 1.12 to 1.87), CFS (aHR = 1.17; 95% CI 1.07 to 1.29/1-point increase), and Lab-FI (aHR = 1.22; 95% CI 1.11 to 1.34/0.1-point increase). Similar associations were found for FI. Two hundred fifty-nine cases (27.3%) were admitted to ICU, which was associated with higher Lab-FI (aHR = 1.32; 95% CI 1.21 to 1.44/0.1 increase), being male and aged 40–79 years (aHR = 1.40; 95% CI 1.05 to 1.86). Younger (aged <39 y ; aHR = 0.44) and older (aged >80 y; aHR = 0.56) patients of both sexes were less often admitted. CFS and FI were not associated with ICU admission. For patients discharged alive, median LOS was 11 days (IQR 4–28). Frailty was associated with longer LOS, median 5 days (IQR 3–13) for CFS = 1–2 versus 30 days (IQR 12.5–57) for CFS ≥6.

Conclusion

Frailty among hospitalized COVID-19 patients is a significantly associated with death and prolonged hospital stay. ICU admissions were frequent among young non-frail adults.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):39–40.

P48. Real-life experience with IV fosfomycin in Canada: Results from the CLEAR Registry

George G Zhanel ^1,^✉, Anna Lee ², Melanie Baxter ¹, Neal Irfan ³, Coleman Rotstein ⁴, Gabriel Girouard ⁵, Maxime Dube ⁶, Andrew Walkty ¹, James Karlowsky ¹

Objectives

Intravenous (IV) fosfomycin recently received Health Canada approval and became available for use. We report the use of IV fosfomycin in Canada using a national registry.

Methods

An IV fosfomycin usage questionnaire was developed. The CLEAR (Canadian LEadership on Antimicrobial Real-life usage) registry protocol–questionnaire was submitted and received approval by the University Ethics Committee (April 2019). The CLEAR registry uses the web-based research data management program, REDCap™ (online survey, https://is.gd/CLEAR_IVfosfomycin), to facilitate clinicians voluntarily entering details associated with their clinical experiences using IV fosfomycin.

Results

As of January 5, 2021, data were available for six patients. Infections treated were bacteremia–sepsis (n = 2), complicated urinary tract infections (cUTI) (n = 2), and central nervous system infection and community-acquired pneumonia (n = 1 each). IV fosfomycin was used as directed therapy to treat Escherichia coli (1 ESBL and 1 CRE), Klebsiella spp (2 CRE), Pseudomonas aeruginosa (n = 1), and Enterococcus faecium (n = 1). IV fosfomycin susceptibility testing was performed on all isolates. IV fosfomycin was used in combination with other antimicrobials in all infections except the two patients with cUTI. IV fosfomycin was used because of resistance to (n = 5) and failure of (n = 1) previously prescribed antimicrobial agents. The dosage regimen was customized in all patients on the basis of their creatinine clearance. Treatment duration was >10 days (n = 3), 7–10 days (n = 2), and <7 days (n = 1). Microbiological success was 100% (50% eradicated, 50% presumed eradicated); clinical success was 100% (50% cured, 50% improvement). Thirty-day mortality was 0%. Two patients had adverse effects (hypokalemia, elevated alanine aminotransferase).

Conclusion

To date, IV fosfomycin is used as directed therapy to treat a variety of severe infections caused by pathogens resistant to other antimicrobials. Other than for cUTI it is used in combination with other antimicrobials with high microbiological–clinical cure rates. More data are needed to fully elucidate the efficacy and safety of IV fosfomycin in Canada.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):40.

P49. Antimicrobial resistance trends of Staphylococcus aureus isolates collected from patients over 14 years

Dylan Bedi ^1,^✉, Sandra AN Walker ¹, Christine Peragine ²

Objectives

Staphylococcus aureus are designated as methicillin-susceptible (MSSA) or methicillin-resistant (MRSA) on the basis of beta-lactam antibiotic resistance mediated by the mecA gene. Published Canadian data are heavily focused on MRSA resistance trends. The present analysis discusses antimicrobial resistance trends for both MRSA and MSSA isolates, offering a more complete picture of S. aureus resistance patterns.

Methods

Susceptibility data were extracted from the Sunnybrook Health Sciences Centre (SHSC) Microbiology Database for S. aureus clinical isolates collected from patients from October 2002 until September 2016. Using univariate linear regressions with a significance level of <0.05, resistance trends for cefazolin, ciprofloxacin, clindamycin, cloxacillin, erythromycin, gentamicin, moxifloxacin, nitrofurantoin, penicillin, rifampin, tetracycline, sulfamethoxazole–trimethoprim, and vancomycin were generated for MSSA and MRSA isolates.

Results

The prevalence of antimicrobial-resistant and multi-drug-resistant MSSA increased over time (+1.0%/y and +0.5%/y, respectively). MSSA resistance increased to ciprofloxacin (+0.5%/y), penicillin (+1.7%/y), and sulfamethoxazole–trimethoprim (+0.1%/y). Conversely, a significant decrease in resistance was found for MRSA isolates to ciprofloxacin (−1.7%/y), clindamycin (−4.0%/y), erythromycin (−1.4%/y), moxifloxacin (−4.0%/y), and rifampin (−0.5%/y). For all other antimicrobials analyzed, there were no significant trends in MSSA or MRSA resistance. One hundred percent (7,398/7,398) of MSSA isolates were susceptible to cloxacillin, and more than 99% (3,556/3,559) were susceptible to cefazolin. In addition, 100% (1,819/1,819) of MRSA isolates identified across the 14-year study period were susceptible to vancomycin.

Conclusion

MSSA resistance rates to individual antibiotics increased or remained stable, whereas MRSA resistance rates decreased or remained stable. The modest reduction in MRSA resistance at SHSC may be a commentary on the reversibility of institution-level antimicrobial resistance as a result of more prudent antimicrobial use and infection control policies. However, multivariate models would be required to confirm this optimism.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):40–41.

P50. Emergency department visits and readmission rates of patients discharged on OPAT: A retrospective observational study to identify predictors

Jorge Martinez-Cajas ¹, Susan McKenna ², Beatriz Alvarado-Llano ¹, Evan Wilson ^1,^✉, Kirk Leifso ¹, Santiago Perez-Patrigeon ¹, Gerald Evans ¹

Objectives

Kingston Health Science Centre (KHSC) is an academic tertiary care hospital that lacks a structured outpatient parenteral antimicrobial treatment (OPAT) team and program. We identified high rates of emergency department (ED) visits and subsequent readmissions in patients discharged from KHSC who were prescribed OPAT. Thus, we sought to identify predictors associated with ED visits and readmissions.

Methods

All OPAT courses initiated between July 2017 and June 2018 were identified using an electronic retrieval protocol. The first OPAT course per patient was included. Reasons for subsequent ED visits were obtained from patient records. Poisson multivariate regressions were performed to identify predictors, which included patient characteristics (demographic characteristics, type of infection, microorganism, antibiotic, Charlson Comorbidity Index [CCI], and treatment duration) and process indicators (infectious disease [ID] specialist involvement before discharge, person responsible for care after discharge, instructions to patients, instructions to physicians), among others.

Results

Among 313 patients discharged with parenteral antibiotics, 21.1% had at least one ED visit within 30 days, and 28.9% were re-admitted. ED visits were more likely to occur among those who had higher social vulnerability (people who use drugs or are homeless), resided in special homes (eg, Indigenous lodge), or were incarcerated. Higher readmission rates were seen in those who had methicillin-resistant Staphylococcus aureus or Pseudomonas spp infections and those with higher CCI. Other factors, including ID specialist and pharmacy involvement before discharge, were not associated with ED visits or readmission.

Conclusion

We identified predictors for ED visits and readmission to hospital among patients undergoing OPAT in an unstructured setting. Plans for a future structured OPAT program at KHSC will need to ensure that all discharged patients on OPAT have an appropriate monitoring plan and adequate follow-up with appropriate specialty input. Socially vulnerable patients will need special attention.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):41.

P51. Prescribing pattern of restricted antimicrobials at a quaternary care academic hospital

Ryan J LeBlanc ¹, Dima Kabbani ², Stephanie W Smith ², Karen E Doucette ², Cecilia Lau ³, Serena Bains ³, Karen Fong ³, Jackson Stewart ³, Justin Z Chen ^2,^✉

Objectives

There is a paucity of data regarding prescribing appropriateness comparing learners versus staff and the timing of prescriptions. We aimed to describe prescribing appropriateness of restricted antibiotics among hospitalized patients on the basis of prescriber level of training, time, and date of prescription.

Methods

Restricted antimicrobial (carbapenems, linezolid, daptomycin, and tigecycline) prescriptions in adult inpatients at a large academic Canadian centre were prospectively audited for guideline concordance and regimen appropriateness by the antimicrobial stewardship program (ASP). A retrospective review of an ASP database for prescriptions between March 2018 and December 2020 was undertaken. Appropriateness was analyzed by prescriber level of training, and subgroup analyses were performed on the basis of prescribing service, time, and date of prescription. Time between 8:00 a.m. and 5:59 p.m. was defined as working hours. Time between 6:00 p.m. and 7:59 a.m. was defined as after hours.

Results

A total of 1,629 restricted antimicrobial prescriptions were reviewed; 54% were written by medicine, 27% by surgery, and 18% by critical care. Of the prescriptions, 60% (977/1,629) were optimally prescribed by agent, regimen, and duration as assessed by the ASP at the time of audit. No statistical difference was found between staff and learners (62% versus 59%; p = 0.31). Prescriptions written by infectious disease (ID) prescriber groups were more often optimally written than prescriptions written by non-ID prescriber groups (82% versus 55%; p <0.001). Most prescriptions were written during working hours (69%; 1,101/1,607) and during weekdays (69%; 1,119/1,629). Appropriateness was significantly worse after hours (62% versus 54%; p <0.001) and significantly worse during weekends (67% versus 53%; p <0.001).

Conclusion

Restricted antimicrobial prescriptions assessed by agent, dose, route, frequency, and duration during ASP audit were frequently suboptimally written by both staff and learners at a similar rate. Prescriptions written by ID prescriber groups were optimally written more often than prescriptions written by non-ID prescriber groups. There was a significant decline in optimal prescribing after hours and on weekends.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):41–42.

P52. Quality audit of surgical antibiotic prophylaxis timing, redosing, and post-operative dosing practices at five hospital sites: The devil is in the details

Lynora M Saxinger ^1,^2,^✉, David C Williams ^2,³, Katelynn C Crick ⁴, Justin Z Chen ^1,², Holly L Hoang ^1,⁵, Susan R Fryters ⁶, A Uma Chandran ², Tyler W Myroniuk ⁴, Roseanne O Yeung ⁴, Denise Campbell-Scherer ⁴

Objectives

Stewardship in peri-operative surgical antimicrobial prophylaxis (SAP) has a focus on optimizing pre-operative timing, re-dosing in cases >4 hours, and elimination of post-operative SAP, as recommended in major guidelines. This audit was done to assess the timing and duration of SAP to inform a multi-site, transdisciplinary quality intervention. In addition, a national cefazolin shortage occurred during the audit period, and the effect of cefazolin stewardship and conservation efforts is described.

Methods

A prospective chart review (N = 3,262) occurred at five teaching hospitals from June 4 to August 23, 2019. Data collected included patient demographics (age, gender), type of surgical procedure (incision time, duration), and details of antibiotic prophylaxis (documented allergy, antimicrobials, time, duration.)

Results

There were 3,218 analyzable cases, of which 84.6% received SAP (89.8% cefazolin). Prophylaxis was outside of target timing (>60 min before, at time of, or after incision) in 11% of cases. Surgical specialty influenced pre-operative SAP timing, with early (>60 min) administration more common in cardiovascular surgery and neurosurgery and late administration more common in obstetrics and gynecology (O&G), otolaryngology, and general surgery. Re-dosing was received in 59.1% of the 164 indicated cases (ongoing procedure >4 h from pre-operative SAP). Post-procedure SAP was documented in 1,128 (35.1%) cases, with 74.8% extended for >24 hours, with the lowest prescribers of post-operative SAP O&G (11%) and general surgery (14.8%) and the highest prescribers orthopedic surgery (67.8%) and cardiovascular surgery (76.2%). Data comparing practices before and during the cefazolin shortage are presented.

Conclusion

Pre-operative timing was appropriate in 89% of cases, but re-dosing practices were suboptimal. Non-recommended extension of SAP into the post-operative period occurred in 35% of cases (range 11% to 76% across subspecialties). Both site-based and specialty-related factors appear to influence SAP, highlighting the need for a context-dependent and transdisciplinary approach to quality interventions in this area.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):42.

P53. Targeted plasma metabolomics method for the diagnosis of acute SARS-CoV-2

Anthony T Le ¹, Manhong Wu ¹, Nicholas A Phillips ¹, Pranav S Rajpurkar ¹, Mamdouh Sibai ¹, ChunHong Huang ¹, Malaya K Sahoo ¹, Justin Mak ¹, Raffick Bowen ¹, Tina M Cowan ¹, Benjamin A Pinsky ¹, Catherine A Hogan ^1,^✉

Objectives

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) may induce host metabolite alterations by infecting respiratory epithelial cells. The current reference standard for acute SARS-CoV-2 diagnosis is nucleic acid testing, which is subject to frequent reagent shortages and may lack sensitivity. The objective of this study was to investigate the test performance of a distinct approach, targeted plasma metabolomics, for the diagnosis of acute SARS-CoV-2 infection.

Methods

Liquid chromatography triple quadrupole time-of-flight mass spectrometry testing was combined with machine learning to identify metabolic signatures that differentiate SARS-CoV-2–infected individuals from uninfected individuals. Plasma aliquots of 20 mL were selected from individuals with confirmed SARS-CoV-2 infection within 1 week of diagnosis and uninfected inpatient and outpatient controls with and without a history of coronavirus disease 2019 (COVID-19). Test performance was characterized by area under the receiver operating characteristic curve (AUC) analysis and median metabolite quantitation comparison by Mann–Whitney tests.

Results

A total of 262 individuals were included, and samples were collected a median of 1 day after the SARS-CoV-2 diagnosis (interquartile range 0 to 5 days). Analysis showed an AUC of 0.91 overall (95% CI 0.85 to 0.98) and an AUC of 0.96 for individuals without a history of COVID-19 (95% CI 0.92 to 1.0). Of the top differentiating metabolites identified, arginine was shown to be lowered in the plasma of infected individuals (33.5 versus 80.4 µM; p <0.0001), whereas the opposite was seen with pyroglutamic acid (214,127 versus 152,641 peak area; p <0.0001).

Conclusion

This plasma metabolomic approach demonstrated high accuracy for the diagnosis of acute SARS-CoV-2 infection. This technique may be feasible to implement in other settings given the availability of mass spectrometry instruments in clinical laboratories. Further work will assess naso-pharyngeal specimens and longitudinal measurements to extend these findings.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):43.

P54. Lineage distributions of SARS-CoV-2 during the second wave in British Columbia

Kimia Kamelian ^1,^2,^✉, Shannon Russell ¹, Hind Sbihi ^1,³, Kimberley A Macdonald ^1,⁴, Dan Fornika ¹, Tracy Lee ¹, Rebecca Hickman ¹, John Tyson ¹, Robert Azana ¹, Corrinne Ng ¹, Loretta Janz ¹, Linda Hoang ^1,², Natalie Prystajecky ^1,²

Objectives

Real-time phylogenetic classification of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) through a nomenclature system can provide insight into circulating strains and identify genetic clusters. In this study, the spatiotemporal distribution of SARS-CoV-2 lineages during the second wave of infections in British Columbia is examined.

Methods

Unique cases collected between August 1 and December 31, 2020, underwent whole-genome sequencing on an Illumina platform. Phylogenetic Assignment of Named Global Outbreak LINeages (pangolin; version 2.1.6) with PangoLEARN date of January 1, 2021, was used to assign SARS-CoV-2 lineages. Sequences were assigned a lineage if they contained more than 85% genome coverage. Subsequent phylogenetic analysis occurred under a maximum likelihood framework.

Results

In total, 4,613 cases (4,605 unique individuals; 8 mink) were assigned lineages and represented 9% of all reported cases (n = 49,122) in British Columbia during the study period. Individuals were predominantly female (n = 2,367; 51%) with a median age of 30 years (25th–75th percentile 17 to 50). Cases were primarily from the Fraser Health Authority (n = 2,938; 64%) followed by Vancouver Coastal Health (n = 1,034; 22%). The B parent lineage dominated cases in British Columbia (n = 4,608; 99.9%), with sub-lineages B.1.128 (n = 883; 19%), B.1.36.1 (n = 812; 18%), B.1.2 (n = 606; 13%), and B.1.1.306 (n = 580; 13%) accounting for the majority of the assigned lineages. Temporally, observations of B.1.128 and B.1.1.306 decreased over the study period, and observations of B.1.2 and B.1.36.1 increased. The majority of mink cases were associated with lineage B.1.1.91 and clustered in close proximity apart from other lineages.

Conclusion

Currently, major SARS-CoV-2 lineages present in British Columbia include B.1.128 and B.1.36.1. Although lineage assignments present a method to account for and classify the global and local changing characteristics of SARS-CoV-2, the evolving nature of lineage designations as a response to novel circulating strains presents challenges to data reproducibility.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):43–44.

P55. Assessment of fungal ITS PCR performance in formalin-fixed paraffin-embedded tissues

Shawn T Clark ^1,^✉, Yvonne CW Yau ^1,², Aaron Campigotto ^1,², Susan E Richardson ^1,², Farhad Gharabaghi ², Manal Tadros ^1,²

Objectives

The molecular detection and identification of fungal DNA in clinical specimens can be used to confirm infection when other laboratory analyses are inconclusive. The objective of this study was to assess the performance of an in-house end-point polymerase chain reaction (PCR) assay targeting the ITS2 region of the fungal genome to detect fungal DNA in formalin-fixed paraffin-embedded (FFPE) biopsy specimens.

Methods

Our workflow includes reviewing the histopathology of samples for fungal elements, deparaffinization and deproteination of FFPE tissues, followed by PCR amplification of an approximately 400 bp portion of the fungal genome using universal ITS3–ITS4 primers, amplicon purification, Sanger sequencing, and analysis. All requests for fungal PCR on FFPE received between January and December 2019 were reviewed. Reports were examined for specimen type, smear characteristics, and amplification parameters (number of amplicons, inhibition). Positive results were categorized into “positive–contaminant” and “positive–clinically significant” on the basis of sequence identification of the fungus and correlation between smear and PCR results.

Results

A total of 88 FFPE blocks were submitted for fungal PCR during the study period. Lung biopsy was the most common specimen received (64.8%). Fifty-one specimens were smear positive for fungal elements. Of 88 specimens, 87 had at least one band detected post-amplification; 53 (60.2%) were categorized as positive–contaminants and 34 (38.6%) as positive–clinically significant. Fungal PCR had a positivity rate (positive–clinically significant) of 62.7% and 2.9% for smear-positive and smear-negative specimens, respectively. Of the organisms reported Blastomyces dermatitidis (17.1%), Cryptococcus neoformans (17.1%), Histoplasma capsulatum (14.3%), and Mucormycetes (11.4%) were the most common fungi identified.

Conclusion

Pan-fungal PCR is a useful diagnostic tool, especially in smear-positive FFPE specimens. Its utility may improve by implementing additional measures to minimize contamination and overcome PCR inhibition in some specimens.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):44.

P56. Comparative genomics analysis of an IncR plasmid encoding blaNDM-1 from Enterobacter hormaechei involved in an outbreak in Quebec

Florence Doualla-Bell ^1,^✉, David A Boyd ², David Roy ³, Khadidja Yousfi ^1,⁴, Isabelle Bernaquez ¹, Simon Wong ¹, Patrice Savard ⁵, Sadjia Bekal ^1,⁴

Objectives

The New Delhi metallo-beta-lactamase (NDM) genes confer resistance to almost all classes of beta-lactams and are increasingly reported in clinical isolates. In 2019–2020, the Quebec surveillance program on carbapenemase-producing Enterobacteriaceae (CPE) characterized an unprecedented increase in blaNDM-1 gene, representing 19% of all CPE. Hence, we performed a comparative analysis of an IncR plasmid encoding blaNDM-1 unit in Enterobacter hormaechei.

Methods

Three clinical isolates of E. hormaechei, recovered from hospitalized patients involved in a putative outbreak, were found positive for the blaNDM-1 gene. Whole-genome sequence (WGS) analysis was performed to characterize the blaNDM-1 carbapenem resistance-encoding determinants and their genetic environment.

Results

Genome analysis, supported by PFGE data, revealed that two of the strains were highly similar, and both harbored a non-conjugative but mobilizable IncR plasmid (~45 kb) carrying a multi-drug-resistant (MDR) region consisting of blaNDM-1, blaOXA-1, blaOXA-10, sul1, ant(3’’)-la, arr-3, catB4, and aac(6ʹ)-Ib-cr. The IS26-mphA-mrx-mphR(A)-IS6100 unit, previously identified in Shigella spp and involved in the acquisition and spread of azithromycin (AZT) resistance, was identified. Both strains also harbor a multiresistance-encoding IncFII conjugative plasmid (~130 kb). We also showed that a complex class 1 integron located within the IncR plasmid may be responsible for a site-specific acquisition of the blaNDM-1, most probably by homologous recombination from a previously coexisting conjugative IncF plasmid.

Conclusion

This study highlights the putative role of a non-conjugative IncR plasmid in the acquisition and persistence of blaNDM-1 genes in Enterobacteriaceae throughout the interspecies mobilization of multiresistant conjugative plasmids. The presence of the IS26-mphA-mrx-mphR(A)-IS6100 unit suggests that CPE may act as reservoir for such an AZT resistance. MDR variable regions related to a complex class I integron suggest an important role of DNA recombination in IncR plasmids and should therefore be monitored in the resistance surveillance program.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):44–45.

P57. Yokenella regensburgei infection in an immunocompetent host

Fatimah Hussain Al Mutawa ^1,^✉

Objectives

Yokenella regensburgei is a member of the Enterobacteriaceae family. It is found in the environment and as part of the insect intestinal flora. It resembles Hafnia alvei, with some different biochemical reactions. It rarely causes human disease.

Methods

This is a case presentation.

Results

A 38-year-old man who was otherwise healthy presented on November 2020 with chronic draining wound of his left arm after a remote injury in September 2020 when he fell down and incurred a puncture of his left elbow by a tree branch. Since then, he had been having recurrent swelling and draining requiring seven emergency room visits and multiple courses of oral and intravenous antibiotics, including cephazolin and amoxicillin–clavulanate, with no improvement. He was stable at the time of assessment. X-rays ruled out fracture; a white blood cell scan was consistent with subcutaneous inflammation, and ultrasound showed a foreign body as well as fluid collection. The patient underwent incision and drainage. The fluid grew pure culture of Y. regensburgei on blood agar and MacConkey’s agar plates that was identified by matrix-assisted laser desorption ionization–time of flight mass spectrometry. The organism was susceptible to ciprofloxacin, gentamicin, imipenem, tobramycin, and trimethoprim–sulfamethoxazole. Treatment with trimethoprim–sulfamethoxazole started; the patient improved clinically, and the wound healed.

Conclusion

To our knowledge, infections caused by Y. regensburgei are rarely reported in the literature. They range from skin–soft tissue infections to bacteremia and arthritis. Predisposing factors include diabetes, alcoholism, chronic kidney disease, and immunosuppression. Our patient did not have any risk factors. His likely source of infection was the penetrating injury that happened back in September. This organism is commonly known to have multidrug resistance, but fortunately in our case it was susceptible to multiple classes of antimicrobials. Given that our patient is immunocompetent and had no risk factors, this emphasizes the importance of investigating and identifying unusual pathogens and involvement of infectious disease experts for better patient management.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):45.

P58. Performance comparison of micro-neutralization assays based on surrogate SARS-CoV-2 and WT SARS-CoV-2 in assessing virus-neutralizing capacity of anti-SARS-CoV-2 antibodies

Inna Sekirov ^1,², Martin Petric ^1,², Erin Carruthers ^1,^✉, David Lawrence ¹, Navdeep Chahil ¹, Paul Levett ^1,², Heidi Wood ³, Robbin Lindsay ³, Mike Drebot ³, Mel Krajden ^1,², Muhammad Morshed ^1,²

Objectives

We describe the results of testing using a cell culture-independent commercial enzyme-linked immunosorbent assay, designated as the surrogate virus neutralization test (sVNT), which measures the serum-induced inhibition of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein receptor binding domain (RBD) binding to the host cell receptor, angiotensin converting enzyme-2.

Methods

A serum panel from the Canadian National Microbiology Laboratory (NML) (21 polymerase chain reaction–confirmed SARS-CoV-2 patients’ sera and 19 SARS-CoV-2–negative sera) was evaluated by the conventional SARS-CoV-2 virus micro-neutralization (cVNT) and sVNT (GenScript USA, Inc., Piscataway, NJ, USA) assays at the British Columbia Centre for Disease Control Public Health Laboratory (BCCDC PHL).

Results

For the NML panel, PRNT₅₀/PRNT₉₀, cVNT, and sVNT assays yielded very comparable results for all 40 samples. All BCCDC PHL samples negative for SARS-CoV-2 antibodies by high-volume platforms (Siemens, Ortho, and Abbott) were also negative on the sVNT assay and had titres of <1:8 on cVNT. All negative samples had no more than 11% inhibition of binding. All serology-positive samples (n = 42) were positive for inhibition by sVNT, but cVNT results were variable. Overall, sVNT–cVNT positive agreement on coronavirus disease 2019 positive sera from combined panels was 75%, or 81% with equivocal cVNT results counted as positive.

Conclusion

Laboratory-developed and commercially available sVNT assays allow for accessible high-volume assessments of antibody neutralizing capacity. Surrogate SARS-CoV-2 assays offer the potential to assess anti-SARS-CoV-2 antibodies’ neutralizing capacity outside level 3 containment.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):45–46.

P59. Necrotic myositis in an immunocompetent patient with persistent Bacillus cereus bacteremia and acute liver failure: A case report

Anna Cvetkovic ^1,^✉, Vaibhav Mokashi ¹, Kevin Woodward ¹

Objectives

Bacillus cereus bacteremia is commonly considered to be a contaminant in immunocompetent patients given its environmental distribution and usual role as a foodborne pathogen. However, it is known to cause disease in patients who use drugs, patients with indwelling venous catheters, patients who are immunocompromised, and, rarely, patients with liver cirrhosis. We describe an immunocompetent patient without a history of substance use, indwelling catheter, preceding trauma, or prosthetic material presenting with persistent B. cereus bacteremia with associated necrotizing myositis.

Results

A 62-year-old woman was transferred from a community-based hospital with uncontrolled atrial flutter requiring cardioversion. She developed acute shock liver secondary to severe hypotension. During a second episode of atrial flutter, she developed severe and sudden right thigh pain. A MRI of her right leg revealed quadriceps myositis with focal fascial edema and fasciitis without evidence of abscess. Multiple blood cultures grew B. cereus. Intravenous vancomycin was initiated, and despite therapeutic drug levels, she remained persistently bacteremic for 12 days until her cultures cleared. Her symptoms resolved without surgical management. A transthoracic echocardiogram was negative for endocarditis and a computed tomography scan of her abdomen and pelvis showed no mycotic aneurysms. She was successfully treated with 4 weeks of vancomycin with complete recovery of her liver injury.

Conclusion

This case presents an opportunity to review the varied presentations of B. cereus infection, from common foodborne illness to rare gas gangrene-like infections. To our knowledge, this represents the first case of primary cutaneous necrotizing infection in a patient with no history of preceding trauma or underlying immunocompromise, outside of acute liver failure and well-controlled diabetes.

In presenting this case alongside those previously reported, we add to the building literature suggesting that B. cereus can cause severe, life-threatening illness in both immunocompetent and immunocompromised patients.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):46.

P60. How has antimicrobial use changed during the COVID-19 pandemic? A look at community antimicrobial use in Canada

Glenys Smith ¹, Braden Knight ^1,^✉, Drew Greydanus ¹, Jayson Shurgold ¹, James Brooks ¹, Denise Gravel Tropper ¹

Objectives

Approximately 15.7 antimicrobial doses/1,000 inhabitants are dispensed daily by Canadian pharmaceutical retail outlets. As a consequence of the coronavirus disease 2019 (COVID-19) pandemic, medical practice in Canada changed, with many in-person visits being replaced by remote care. The potential effect on antimicrobial prescribing was unknown, but there was the possibility that it would increase as a result of diagnostic uncertainty associated with the absence of physical examinations. The objective of this study was to assess the impact of the COVID-19 pandemic on the rate of community antibiotic prescribing in Canada.

Methods

Data extracted from IQVIA’s Canadian CompuScript data set, covering the period from November 2014 to October 2020, were analyzed using interrupted time series analysis to evaluate changes in antimicrobial dispensing. The number of defined daily doses (DDDs) dispensed nationally, by age and sex and by physician specialty, were compared for each month, comparing 2020 with 2019.

Results

After accounting for seasonality, there was a large, statistically significant, and durable reduction in antimicrobial prescribing during the first 8 months of the COVID-19 pandemic (p <0.0001), with a peak reduction of 38% in May 2020. Children aged 0–18 years saw the greatest drop in the number of antimicrobial prescriptions in April 2020, compared with April 2019 (females, 68%; males, 72%). The decrease in antimicrobial prescribing was higher among primary care practitioners than specialists (non-general practice–family practice specialties), with decreases of 41% and 35% in April 2020 compared with April 2019, respectively.

Conclusion

Contrary to what was predicted, the COVID-19 pandemic was associated with a significant reduction in the amount of antimicrobials that were prescribed in the community in Canada.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):46–47.

P61. Isolating Candida auris at Humber River Hospital: A comparative evaluation of media to develop a screening protocol for an emerging pathogen

Chantal Morris ^1,^✉, Joanna Widla ¹, Jajn McLay ¹, William Dubinski ¹

Objectives

Humber River Hospital became the first health care institution in Ontario to isolate the emerging pathogen Candida auris after a preliminary screening program was implemented in early 2019. Finding no existing provincial guidelines, we performed this study to select the best selective or differential media for the recovery of C. auris from patient screens and environmental samples by comparing two direct plate inoculation methods and an enrichment broth method. We also determined the optimal temperature and length of time for sample incubation.

Methods

We performed parallel testing of patient samples using known isolates of C. auris and other Candida species. Growth on two different yeast-specific chromogenic solid agar media was compared after both direct inoculation and broth enrichment. Cultures were incubated at varying temperatures and checked over a period of 5 days to determine optimal conditions for growth.

Results

C. auris growth was detectable after 48 hours of incubation. No differences were identified among direct inoculation, 0.5 McFarland suspension, or a 1:10 dilution of a 0.5 McFarland suspension. When challenged against a second yeast species, C. auris was consistently isolated. C. auris grew at 37°C and 42°C, but fewer competing yeast species of a similar appearance grew at the higher temperature.

Conclusion

Oxoid’s Brilliance Candida agar was chosen as the superior agar in this study. It supported the growth of C. auris while showing more variety in colour for other yeast, thereby reducing the time spent identifying non-significant organisms. Direct inoculation and incubation at 42°C is preferred for patient swabs because it provides excellent sensitivity while inhibiting some other yeasts and provides adequate turnaround time because cultures may be positive in as little as 48 hours. Enrichment broth is preferred for environmental swabs, assuming a low burden of organism on surfaces.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):47.

P62. WITHDRAWN

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):47.

P63. Community antibiotic use at the population level during the SARS-CoV-2 pandemic in British Columbia, Canada

Abdullah A Mamun ^1,^✉, Ariana Saatchi ², Max Xie ¹, Hannah Lishman ³, Edith Blondel-Hill ⁴, Fawziah Marra ², David M Patrick ^3,¹

Objectives

To examine the aggregate rates of antibiotic use at the population level and to compare these rates over time against historical averages in order to identify the effect of severe acute respiratory syndrome coronavirus 2, and the resulting control measures, on community prescribing.

Methods

We collected antibiotic prescriptions and physician office visits from January 1, 2016, to July 30, 2020. We calculated monthly prescription rates stratified by sex, age group, profession, diagnosis type, and antibiotic class. We also looked at the monthly prescription rate per 1,000 population as a moving average over time; centred and calculated using a 3-month time window. Using interrupted time series analysis (ITSA) method, we estimated the changes in prescription rates after March 2020.

Results

The moving average of overall monthly prescription rates from January to June 2020 was below the minimum of the historical years’ moving averages (2016–2019). We observed >30% reduction in overall monthly prescription rates in April, May, and July 2020 compared with the same months in 2019. The results of ITSA showed that, on average, overall monthly provincial prescription rates experienced a significant level change of −12.79 (p <0.001) after coronavirus disease 2019 restrictions were put into place in March 2020, with the greatest level change of −18.02 observed among those aged 1–4 years (p <0.001). We estimated an average –5.94 (p <0.001) change in respiratory tract infection (RTI)–associated monthly prescription rates after March 2020. Overall monthly prescription rates comparing January–July 2019 and their 2020 counterparts showed a decrease in monthly prescribing with a range of −1 to −5 prescriptions per 1,000 population for amoxicillin, amoxicillin and enzyme inhibitors, azithromycin, clarithromycin, and sulfamethoxazole.

Conclusion

In British Columbia, Canada, overall and RTI-specific monthly antibiotic prescription rates declined significantly during April to July 2020 compared with the same months in pre-pandemic years.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):47–48.

P64. Prospective audit and feedback in level II neonatal intensive care units: Is it worth it?

Khaled Alsager ^1,^✉, Kwadwo Mponponsuo ¹, Genevieve Kerkerian ¹, Anish Krishnan ², Deonne Dersch-Mills ², Joseph Vayalumkal ¹, Cora Constantinescu ¹

Objectives

There are limited antimicrobial stewardship data in the level II neonatal intensive care unit (NICU) population. In 2020, prospective audit and feedback (PAF) was performed for all the level II NICUs in Calgary, using electronic chart review. Subsequently, the antimicrobial stewardship program (ASP) interventions were delivered via telephone communication and documentation in patients’ electronic medical records. Our aim was to describe antibiotic usage in level II NICUs over a 1-month period and assess the success rate of ASP interventions.

Methods

PAF was implemented in three level II NICUs in Calgary during a 4-week period from November 16, 2020, to December 11, 2020, by infectious disease fellows supervised by an antimicrobial stewardship physician and pharmacist. Data were collected prospectively regarding indications, antibiotics prescribed, duration of therapy, and uptake of recommendations.

Results

In total, 29 patients started antibiotics during the PAF period. Early-onset sepsis was the indication in 86% of cases, followed by late-onset sepsis (10%) and others (meningitis and cellulitis). The most common antibiotic regimen was a combination of ampicillin and gentamicin (90%), followed by ampicillin and cefotaxime (7%), then cloxacillin and gentamicin (3%). Duration of antibiotics was ≤48 hours in 86% of cases. Only 4 patients received antibiotics for longer than 48 hours. One patient was on antibiotics for 7 days (cellulitis), 1 for 14 days, and 2 for 21 days (meningitis). Only three ASP interventions were required during the PAF (one for stop date addition, and the others for stopping antibiotics) with a 33% acceptance rate.

Conclusion

There was consistency in level II NICU practice regarding indication of therapy, antibiotics used, and duration of therapy. In cases in which duration of therapy exceeded the initial 48 hours, there was a clear and appropriate indication (meningitis, cellulitis). Further study is required to determine whether PAF in level II NICUs is required or whether alternative strategies could be used to optimize antimicrobial utilization.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):48.

P65. Validation of the triplex real-time RT-PCR assay for detection of chikungunya, dengue, and Zika viruses

Benjamin Hon ¹, Ella Grandbois ¹, Min-Kuang Lee ^1,^✉, Navdeep Chahil ¹, Muhammad G Morshed ^1,²

Objectives

To respond to the need for accurate and affordable molecular diagnosis of clinically indistinguishable arbovirus infections, we aim to expand our existing Zika real-time reverse transcription polymerase chain reaction (RT-PCR) by integrating dengue 1–4 and chikungunya detection into one triplex test.

Methods

Multiple primers, probes, and reporters were tested for their sensitivity and compatibility. The final selection of the triplex test is to target the non-structure protein 1 gene of chikungunya, 3' UTR region of dengue, and membrane glycoprotein E gene of Zika. RNA was extracted using QIAGEN Viral RNA extraction kit, and real-time RT-PCR was performed on the ABI Taqman 7500 using ABI TaqMan 2X Fast Virus PCR Master Mix. Synthesized DNA gBlocks were used for analytical validation, and real specimens were used for clinical validation (including our retro specimens, isolates from ATCC and specimens from the National Microbiology Lab in Winnipeg).

Results

The reportable range of the assay is 1E3 to 1E7, and the R² values for all three targets are >0.99, indicating a great linearity throughout the reportable range. The detection limit for all tests is 1E2/1E3 copies per reaction. Precision was calculated from triplicates over 10 runs. All targets have a coefficient of variation ≤2.5% (within the acceptable 15% range). The clinical sensitivity is 87.5% for chikungunya, 94.7% for dengue, and 100% for Zika. The clinical specificity for all targets is 100%.

Conclusion

The real-time RT-PCR assay described here is a reliable method for screening of chikungunya, dengue, and Zika viruses. Implementation of this triplex assay, along our pan-flavivirus RT-PCR assay, will enhance detection of arbovirus infection in British Columbia.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):48–49.

P66. Viral co-circulation during the COVID-19 pandemic: Broad respiratory testing to inform diagnostic stewardship

Natalie C Marshall ^1,², Ruwandi M Kariyawasam ^1,^2,^✉, Nathan Zelyas ^1,², Mathew A Diggle ^1,²

Objectives

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection can present with a broad clinical differential, so accurate tests are crucial to distinguish true COVID-19 cases from pathogens that do not require urgent public health interventions. Co-circulation of other respiratory viruses is largely unknown during the coronavirus disease 2019 (COVID-19) pandemic but would inform strategies to rapidly and accurately test patients with respiratory symptoms.

Methods

This study retrospectively examined 298,415 COVID-19 tests on respiratory specimens collected in the 3 months after COVID-19 was first documented in Alberta, all from symptomatic patients. Of these, 52,285 were also tested for 17 other pathogens using the Luminex Respiratory Pathogen Panel to understand the prevalence of co-circulating pathogens and their relative rates compared with previous years.

Results

SARS-CoV-2 was identified in 2.2% of specimens. Parallel broad multiplex testing detected additional pathogens in only 3.4% of these specimens—significantly less than in SARS-CoV-2–negative specimens (p <0.0001), suggesting very low rates of SARS-CoV-2 co-infection. Moreover, the overall co-infection rate was significantly lower among specimens with SARS-CoV-2 detected (p <0.0001). Finally, <0.005% of all specimens tested positive for both SARS-CoV-2 and any of the four endemic coronaviruses tested, strongly suggesting neither co-infection nor cross-reactivity between these coronaviruses.

Conclusion

Broad respiratory pathogen testing rarely detected additional pathogens in SARS-CoV-2–positive specimens. Although helpful to understand co-circulation of respiratory viruses causing symptoms similar to COVID-19, these broad tests were ultimately resource intensive and inflexible at a time when clinical laboratories face unprecedented demand for respiratory virus testing, with further increases expected during influenza season. A transition from broad multiplex tests toward streamlined diagnostic algorithms targeting respiratory pathogens of public health concern could simultaneously reduce the overall burden on clinical laboratories while prioritizing testing of pathogens of public health importance.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):49.

P67. Impact of ID consultation on patients with Enterococcus species bloodstream infection

Brittany E Kula ^1,^✉, Tanis C Dingle ², Justin Z Chen ¹

Objectives

Enterococcal blood stream infection (E-BSI) is challenging to treat. Emerging literature suggests mortality benefit with infectious disease consultation (IDC). We aimed to compare 30-day mortality between patients with E-BSI who received IDC and those who did not.

Methods

We included all first episodes of E-BSI in adult patients admitted to a large academic centre from December 1, 2019, to November 30, 2020. We performed a retrospective chart review to evaluate for patient characteristics, source, appropriateness of antimicrobial therapy, whether IDC was performed and its timing, and 30-day mortality.

Results

Sixty-eight patients met inclusion criteria. The average age was 67 years, and 47 (69%) were male. Enterococcus faecalis was isolated in 44 patients (65%), E. faecium in 23 (34%), and E. gallinarum in 1 (1%). Polymicrobial bacteremia occurred in 19 (28%) cases. E-BSI was community acquired occurred in 21 (31%). Overall, 18 (26%) died within 30 days of bacteremia. The average length of hospitalization was 31 days. IDC was obtained in 49 (72%) of cases, and this was early (<24 h from positive Gram stain) in 20 (41%) cases. The mortality was 7 (35%) in the early IDC group compared with 9 (31%) in the late IDC group. Of those with IDC, 16 patients (33%) died at 30 days compared with 2 (11%) in the group without IDC. The average Charlson Comorbidity Index was higher in the group with IDC compared with the group without IDC (3.18 versus 2.47). The 30-day mortality was lower in patients who had IDC, follow-up blood culture(s), and echocardiography than among those who did not (19% versus 30%).

Conclusion

In our retrospective cohort, IDC was not associated with a mortality benefit; IDC was more likely to occur in the more medically complex patients with a higher risk of mortality.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):49–50.

P68. Comparison of the BD Phoenix CPO Detect Test and the NG-Test® CARBA-5 lateral flow assay for the detection of carbapenemase-producing organisms

Lisa Li ^1,^2,^✉, Munjoat Sidhu ³, Leane Kishi ³, Tasnim Vegdani ³, Gnanalaksmi Arumugam ³, Mellichie Hantid ³, Jane Speakman ³, Marthe Charles ^1,²

Objectives

Carbapenemase-producing organisms (CPOs) have emerged as a global health threat in recent years, and the development of new microbiological tests to facilitate their detection is of increasing importance. Our objective was to compare the performance of the BD Phoenix™ CPO Detect test, a component of the newly released BD Phoenix NMIC-450 gram-negative panel (BD, Franklin Lakes, New Jersey, USA), with the NG-Test® CARBA-5 lateral flow assay (LFA) (NG Biotech, Guipry, France) for the detection of CPOs.

Methods

A total of 90 CPO-positive (62 clinical and 28 from the US Centers for Disease Control and Prevention Enterobacteriales carbapenemase diversity panel) and 30 CPO-negative clinical isolates were tested on both the CPO Detect test and the LFA in parallel. Sensitivity, specificity, and reproducibility were compared for both tests. Discrepant results were tested on an in-house CPO polymerase chain reaction to resolve CPO status.

Results

Overall, 30 KPC, 31 NDM, and 29 OXA-type CPOs, as well as 30 CPO-negative isolates (10 ESBLs, 10 wild-type gram negatives, 5 AMP-Cs, and 5 others, including Pseudomonas aeruginosa, Acinetobacter spp, and Stenotrophomonas maltophilia) were tested. Post-discrepant analysis, the sensitivities of the LFA and the CPO Detect were 97.8% and 91.1%, respectively. The LFA yielded 2 false negatives, both Proteus mirabilis isolates, 1 with NDM and 1 with both NDM and OXA-48 genes. The CPO Detect missed 4 KPC, 1 NDM, and 2 OXA-48 positive isolates. The LFA was 100% specific, whereas the CPO Detect test was 96.7% specific (n = 30). The 1 false positive on the CPO Detect test was a P. aeruginosa isolate. The reproducibility of both tests was excellent, with the LFA and CPO Detect test achieving 100% and 93.3% reproducibility, respectively (n = 15).

Conclusion

Both the LFA and the BD Phoenix CPO Detect test performed very well for CPO detection, with the LFA demonstrating a superior sensitivity (97.8% versus 91.1%). The specificity of both was more than 95%.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):50.

P69. Performance of three different screening media for the detection of carbapenemase-producing organisms

Lisa Li ^1,^2,^✉, Munjoat Sidhu ³, Leane Kishi ³, Tasnim Vegdani ³, Gnanalaksmi Arumugam ³, Mellichie Hantid ³, Jane Speakman ³, Marthe Charles ^1,²

Objectives

Infections with carbapenemase-producing organisms (CPOs) have emerged as a pressing global health concern. There is a paucity of published literature comparing the most recently available commercial CPO screening agars that can detect KPC, NDM, and OXA-type CPOs. Our objective was to compare the performance of three different CPO screening media.

Methods

A total of 90 previously characterized CPO-positive isolates (62 clinical and 28 from the US Centers for Disease Control and Prevention Enterobacteriales carbapenemase diversity panel) and 30 known CPO-negative clinical isolates were tested on three different media, the chromogenic mSupercarba^TM plate (Micronostyx, Ottawa, Ontario), the modified MacConkey agar with carbapenem (ThermoFisher Scientific, Burnaby, British Columbia) plate, and the chromogenic chromID® CarbaSmart Carb/OXA bi-plate (bioMérieux, Montreal, Quebec), in parallel. Sensitivity, specificity, and reproducibility were compared. Isolates yielding discrepant results were subsequently tested on an in-house CPO PCR to resolve CPO status.

Results

Overall, 30 KPC, 31 NDM, and 29 OXA-type CPOs, as well as 30 CPO-negative isolates (10 ESBLs, 10 wild-type gram-negatives, 5 AMP-Cs, and 5 others including Pseudomonas aeruginosa, Acinetobacter spp, and Stenotrophomonnas maltophilia) were tested. Post-discrepant analysis, the sensitivities of the different media were 97.8%, 93.3%, and 91.1% (n = 90 for each medium) for the mSupercarba plate, MacConkey with carbapenem plate, and CarbaSmart Carb/OXA bi-plate, respectively. The mSupercarba medium missed 2 KPC-positive organisms (1 Proteus mirabilis and 1 Escherichia coli). The MacConkey medium missed 4 KPCs, 1 NDM, and 1 isolate with both NDM and OXA-48 genes. The CarbaSmart medium missed 4 KPCs and 4 NDMs. The most specific medium was the mSupercarba (93.3%), followed by the CarbaSmart (90%), and then the MacConkey with carbapenem (80%) (n = 30 for each medium). Reproducibility of all media was 100% (n = 15 for each medium).

Conclusion

All three CPO screening media had sensitivities of >90%. The mSupercarba medium performed the best in terms of sensitivity and specificity.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):50–51.

P70. Comparison of two commercial broth microdilution methods for testing susceptibility to colistin in multi-drug-resistant gram-negative organisms

Lisa Li ^1,^2,^✉, Jennifer Grant ^1,², Munjoat Sidhu ³, Joanne Hsu ³, Josie Hughes ³, Marthe Charles ^1,²

Objectives

With the increasing prevalence of multi-drug-resistant (MDR) gram-negative organisms, routine front-line antimicrobials and broader spectrum drugs such as carbapenems are ineffective, leaving clinicians with limited treatment options. Colistin is a last-line drug for which susceptibility testing has proved challenging for clinical microbiology laboratories. Currently, the only Clinical and Laboratory Standards Institute–approved minimum inhibitory concentration (MIC) testing method is by broth microdilution (BMD). Our objective was to compare the performance of two commercially available BMD testing systems for colistin for MDR organisms in an acute care laboratory.

Methods

A total of 50 MDR gram-negative isolates (47 from the US Centers for Disease Control and Prevention [CDC] Enterobacteriales carbapenemase diversity panel and 3 clinical isolates), were tested in parallel on the SensiTest^TM Colistin panel (Liofilchem, Roseto degli Abruzzi, Italy), which tests colistin only, and the Sensititre^TM CAN1MSTF panel (ThermoFisher, Burnaby, British Columbia), which tests colistin and other antibiotics. Categorical agreement of each method with published reference colistin MICs from the CDC (for CDC isolates) or with MICs obtained using a previously validated BMD test (for clinical isolates) was evaluated and compared. In addition, 5 isolates were tested in triplicate for each method to assess reproducibility.

Results

A variety of resistance genes (42% NDM, 32% KPC, 8% OXA-48 or OXA-181, 12% SME, 2% VIM, 2% IMP, and 2% ESBL) and a range of established colistin MICs (72% sensitive, 28% resistant) were represented among the 50 MDR isolates. The categorical agreement was 98% for both the SensiTest and Sensititre panels. Both panels had one major error; both were with the same clinical NDM-positive Pseudomonas aeruginosa, which had an established MIC of 2 (right at the clinical breakpoint). The reproducibility of both methods (tested using 2 KPCs, 2 NDMs, and 1 OXA-181) was 100%.

Conclusion

Both the SensiTest and the Sensititre panels demonstrated excellent performance (>95% categorical agreement with reference MICs) in this study involving well-characterized MDR gram-negative organisms.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):51.

P71. Audit of surgical antibiotic prophylaxis in obstetrical and gynecological surgery

Susan R Fryters ^1,^✉, Justin Z Chen ², A Uma Chandran ³, Holly L Hoang ², Lynora M Saxinger ², Katelynn C Crick ⁴, Tyler W Myroniuk ⁴, David C Williams ⁵, Roseanne O Yeung ⁴, Denise Campbell-Scherer ⁴

Objectives

Pre-operative surgical antibiotic prophylaxis (SAP) is an established best practice to minimize surgical site infections. There are robust evidence-based guidelines in obstetrical and gynecological (O&G) surgeries, but studies about local practice patterns and adherence to recommendations are limited. Data were collected to assess the quality of SAP with respect to selection, timing, and duration of antibiotic prophylaxis according to provincial recommendations.

Methods

A prospective cross-sectional chart review was conducted from June 4 to August 23, 2019, of a convenience sample of O&G procedures performed at three teaching hospitals. Data captured included patient demographics (age, gender), surgical procedure (including incision time, duration), and antibiotic prophylaxis (antimicrobial allergy, selection, timing from incision, duration).

Results

There were 118 O&G procedures included in the study. Of the 107 patients with assessable procedures with adequate data and surgical codes encompassed in provincial recommendations, 95 required cefazolin according to provincial recommendations, of whom 86 (90.5%) received cefazolin appropriately (plus another agent in 30% of cases). SAP regimens used by procedure are described. There were 12 procedures for which cefazolin prophylaxis was not indicated; only 2 (16.7%) of these received cefazolin. Of the 107 patients, 8 had a beta-lactam allergy and 6 (75%) of these were appropriately prescribed cefazolin in accordance with recommendations. The timing of cefazolin administration was appropriate (within 60 min before incision) in 88.8% (79/89) of procedures, with a median administration time of 10 minutes (IQR 4–16 minutes) before incision. Although not recommended, post-operative surgical prophylaxis was ordered in 11% (13/118) of procedures. Information about duration of prophylaxis by specific surgical procedure is described.

Conclusion

Surgical prophylaxis for O&G procedures at the hospitals audited adheres to provincial recommendations with respect to antibiotic selection and timing. Opportunities for antimicrobial stewardship include improving pre-operative SAP timing and eliminating post-operative SAP.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):51–52.

P72. Using propidium monoazide in real-time quantitative PCR to distinguish viable versus non-viable enteric viruses in water samples

Melissa B Glier ^1,^✉, Ray Han ², Stephanie Man ³, Christine Tchao ¹, Frankie Tsang ¹, Natalie Prystajecky ^1,²

Objectives

More than 150 types of enteric viruses have been detected in wastewater. Enteric viruses are labour intensive and, in some cases, impossible to grow in the laboratory setting, requiring molecular methods (real-time polymerase chain reaction [PCR]) for detection. However, these methods detect both viable and non-viable viruses, affecting the conclusions that can be made regarding public health risks. As such, data are limited on the viral occurrence and removal in Canadian wastewater treatment plants. The membrane impermeable dye propidium monoazide (PMA) has been used to distinguish between viable and non-viable viruses and works by preventing PCR amplification of nucleic acids from damaged viruses. In this study, PMA in combination with real-time quantitative PCR (PMA-qPCR) was used to distinguish between viable and non-viable enteric viruses in water samples.

Methods

Coxsackievirus (RNA genome) and adenovirus (DNA genome) were rendered non-viable by heat inactivation (95°C for 30 min) or by chlorine inactivation (0.5 mg/L for 5 and 15 min). Viable and non-viable viruses were treated with PMA (200 µM) and without, and qPCR was then used to detect and quantify viruses (viral copies/mL). Each inactivation method was tested in triplicate, PMA-qPCR was tested in duplicate, and the limit of detection was 500 copies/mL.

Results

Treatment with PMA reduced the signal of heat-inactivated coxsackievirus and adenovirus by 3.04 logs and 3.03 logs compared with viable PMA-treated viruses. Likewise, treatment with PMA reduced the signal of chlorine-inactivated coxsackievirus (at 5 and 15 min) and adenovirus (at 15 min) by 1.63 logs and 2.53 logs, compared with viable PMA-treated viruses.

Conclusion

This proof-of-concept study suggests that the use of PMA treatment of adenovirus and coxsackievirus before real-time qPCR allowed differentiation of heat- and chlorine-inactivated viruses from viable control samples. These methods are being adapted for use with environmental samples, including wastewater samples.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):52.

P73. Validation of gargle samples on the point-of-care Cobas® Liat® for the detection of SARS-CoV-2

Judith Fafard ^1,^✉, Francine Tourangeau ², Jeannot Dumaresq ^3,⁴, Florence Doualla-Bell ¹, Stéphanie Fournier ², Louise Josée Caron ², Stéphanie Beauchemin ^5,⁶, Annie-Claude Labbé ^5,^6,⁷

Objectives

The performance of the Cobas® SARS-CoV-2 & Influenza A/B nucleic acid test on the Cobas Liat® system on naso-pharyngeal swab specimens has been shown equivalent to that of the Cobas 6800/8800. Our objective was to validate the Liat on gargle specimens, compared with oral and naso-pharyngeal swabs (ONPS).

Methods

Participants were recruited in a designated coronavirus disease 2019 screening clinic in Rimouski, Quebec. In addition to the standard ONPS collected in 3 mL of molecular-grade water, they were asked to gargle with 5 mL of spring water. Both specimens were tested in parallel on the Liat and on the Alinity m SARS-CoV-2 assay.

Results

Among the 111 participants, the mean age was 37 years, and 60 (54%) had symptoms. With the Liat, concordant positive results were obtained on 22 pairs (ONPS and gargle), and concordant negative results were obtained on 85 pairs; 1 pair had an invalid result on the ONPS sample (the gargle sample was negative). Three pairs were discordant: one was positive on the gargle sample only, and 2 were positive on the ONPS sample only. Those two gargle samples were also negative on the Alinity m; the cycle threshold (Ct) values of their paired ONPS sample were 36.8 and 37.4 (compared with a mean Ct value of 24.1 on the 22 concordantly positive pairs). The positive agreement, negative agreement, and overall agreement between the Liat and Alinity m for gargle samples were, respectively: 100% (95% CI 85.2% to 100%), 98.9% (95% CI 93.8% to 100%), and 99.1% (95% CI 95.1% to 100%). Corresponding figures on ONPS were, respectively, 100% (95% CI 85.8% to 100%), 97.7% (95% CI 91.9% to 99.7%), and 98.2% (95% CI 93.6% to 99.8%).

Conclusion

Agreement between the Liat and Alinity m was excellent for both sample types. Agreement between ONPS and gargle sample results on the Liat was also excellent. Although the number of participants was limited, our results suggest that gargle samples could be used on this platform.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):52–53.

P74. Cohort study of children hospitalized with COVID-19 infection

Michelle Barton ¹, Adriana Trajtman ², Rolando Ulloa-Gutierrez ³, Helena Brenes-Chacon ³, Adriana Yock-Corrales ³, Gabriela Ivankovich-Escoto ³, Alejandra Soriano-Fallas ³, Marcela Hernandez-de Mezerville ³, Ari Bitnun ⁴, Shaun Morris ⁴, Tala El Tal ⁴, E Ann Yeh ⁴, Peter Gill ⁴, Ron Laxer ⁴, Alireza Nateghian ⁵, Behzad Haghighi Aski ⁵, Ali Manafif ⁵, Jesse Papenburg ⁶, Marie-Astrid Lefebvre ⁶, Suzette Cooke ⁷, Tammie Dewan ⁷, Lea Restivo ⁷, Isabelle Viel-Thériault ⁸, Rachel Dwillow ², Jared Bullard ², Ashley Roberts ⁹, Manish Sadarangani ⁹, Nicole Le Saux ¹⁰, Jennifer Bowes ¹⁰, Jacqueline Wong ¹¹, Rupeena Purewal ¹², Janell Lautermilch ¹², Kirk Leifso ¹³, Cheryl Foo ¹⁴, Leigh Anne Newhook ¹⁴, Ann Bayliss ¹⁵, Joan L Robinson ¹⁶

Objectives

The purpose of this study was to describe the burden and characteristics of coronavirus disease 2019 (COVID-19) in children hospitalized in three countries.

Methods

Consecutive children admitted to 15 hospitals in Canada (n = 13), Iran, and Costa Rica up to November 16, 2020, with COVID-19 confirmed by molecular detection or serology were included. Clinical and demographic data were collected using standardized case report forms.

Results

There were 211 cases (Canada, n = 95; Costa Rica, n = 84; Iran, n = 32), of which 26 met World Health Organization multi-system inflammatory syndrome in children criteria. Only 104 (49%) were suspected to have COVID-19 at admission. Of the other 107, 26 had their stay prolonged by COVID-19, and 81 did not (incidental COVID-19 cases). Fever was documented in 143 (74%) of 194 cases with data recorded. Children admitted in Canada were older than those admitted in Costa Rica or Iran (median age 4.4 y versus 2.2 y; p <0.05) and were less likely to require supplemental oxygen (19/95 [20%] versus 46/116 [40%]; p <0.05) or mechanical ventilation (3/95 [3%] versus 14/116 [12%]; p <0.05) but not intensive care unit (ICU) admission (12/95 [13%] versus 22/116 [19%]; p = 0.2). Excluding the 81 incidental COVID-19 cases, 70 of 130 (54%) had comorbidities. Thirty (23%) required ICU admission because of COVID-19, of whom 17 (57%) required mechanical ventilation and 3 died (2 had malignancies, and 1 had mild renal disease). Another child with malignancy died without being ventilated. Severe disease (defined as need for oxygen) was more likely in boys, neonates and infants, and those with shortness of breath or a thrombotic or bleeding complication (p <0.05).

Conclusion

The majority of children positive for COVID-19 during hospitalization were admitted with other diagnoses. Approximately half had no comorbidities. This signals a need for vigilance for COVID-19 in all admitted children to reduce the risk of nosocomial transmission. Death was rare, but risk appears to be increased in children with end-stage malignancy.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):53–54.

P75. Evaluation of the Allplex^TM 2019-nCoV assay for the detection of SARS-CoV-2 in a variety of clinical matrices

Shawn T Clark ^1,^✉, Bryn Hazlett ², Sandra Isabel ¹, Ana Aquino ², Qin Liu ², Tony Mazzulli ^1,², Susan M Poutanen ^1,^2,³

Objectives

Laboratories in Ontario began polymerase chain reaction (PCR) testing for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in naso-pharyngeal (NP) and throat specimens in March 2020. We implemented the Allplex^TM 2019-nCoV Assay (Seegene Inc, Seoul, Republic of Korea) to detect SARS-CoV-2 RNA after verifying 25 positive and 50 negative NP specimens with a reference laboratory. Our objective was to assess the performance of this assay on various transport media and non-NP specimens required because of supply chain issues and evolving test requests.

Methods

Assays were prepared using the Microlab STARlet IVD platform (Hamilton Company, Franklin, Massachusetts, USA). Bronchoalveolar lavage (BAL) fluid, neat or sputolysin-treated sputum, oral saline rinse, oral sterile water rinse, saliva diluted 1:1 in 0.85% saline, and nine transport media (ESwab^TM medium [COPAN, Brescia, Italy], Cobas^® PCR Media [Roche, Basel, Switzerland], Aptima^® Specimen Transport Medium [Hologic, Marlborough, Massachusetts, USA], PrimeStore^® Molecular Transport Medium [Longhorn LLC, Bethesda, Maryland, USA], EZPro Transport Medium [ESBE, Markham, Ontario], Norgen media [Norgen, Thorold, Ontario], Bartels [Trinity Biotech-Nova Century Scientific, Inc, Burlington, Ontario], Virus Sampling Kit [Deaou Biotechnology, Guangzhou Science City, China], and Virus Sampling Kit [Yocon, Beijing, China]) (n = 5–10 specimens each) were inoculated with aliquots of SARS-CoV-2 positive NP specimens. Matrix effects relative to paired Universal Transport Medium (UTM) (COPAN) were determined by cycle threshold (Ct) comparison, and specimens deemed acceptable if within 3.3 Ct. Average Ct differences were documented.

Results

SARS-CoV-2 detection was comparable to COPAN UTM for BAL fluid (100%), oral saline rinse (100%), oral sterile water rinse (100%), saliva diluted 1:1 in 0.85% saline (100%), and all nine transport media. Assay performance decreased with sputum regardless of mucolytic agent (20% [95% CI 5 to 52] neat and 60% [95% CI 31 to 83] with sputolysin) with an average increase in Ct of 6.31 in neat sputum and 3.45 in sputum treated with sputolysin.

Conclusion

The Allplex^TM 2019-nCoV Assay could detect SARS-CoV-2 RNA in multiple transport media and clinical specimens with sputum being a notable exception.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):54.

P76. Analysis of the performance of water gargle samples for SARS-CoV-2 detection by RT-PCR in a pediatric population

Emilie Vallières ^1,^2,^✉, Judith Fafard ³, Annie-Claude Labbé ^2,^4,⁵, Marco Bergevin ^2,⁶, Zineb Laghdir ¹, Stéphanie Beauchemin ^4,⁵, Marc Desforges ^1,²

Objectives

Naso-pharyngeal swab (NPS) collection for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing is an uncomfortable experience. Consequently, some parents are reluctant to bring their children for coronavirus disease 2019 (COVID-19) testing, even more so when it is for school outbreak investigation purposes. Therefore, the performance of self-collected non-invasive specimens such as water gargle (WG) warrants evaluation among children. In this study, we validated the use of WG for SARS-CoV-2 detection in comparison to NPS in a pediatric population.

Methods

Children (aged 5–12 y; n = 75) and teenagers (aged 13–17 y; n = 29) were prospectively recruited at CHU Sainte-Justine (Montreal, Quebec). Immediately after NPS sampling, participants gargled with 5 mL of Naya natural spring water. Paired specimens were simultaneously analyzed for the detection of SARS-CoV-2 using chemical extraction followed by a laboratory developed reverse transcription polymerase chain reaction. A comparative performance statistical analysis was performed. An individual was considered infected if a positive result was obtained on either sample.

Results

Among the 104 participants, mean age was 10.9 years. SARS-CoV-2 was detected in 14/75 (18.7%) children and in 1/29 (3.4%) teenagers, for a total of 15/104 (14.4%). No invalid result was obtained but three paired samples had discordant results: two were SARS-CoV-2 SNP positive and WG negative, and one was SARS-CoV-2 positive in WG only. Sensitivity was estimated at 93.3% (95% CI 66% to 99.7%) for NPS and 86.7% (95% CI 58.4% to 97.7%) for WG. In concordant positive samples, the cycle threshold was higher in WG than in NPS; mean difference was 10.0 (SD 3.2) with a range of 6.0 to 15.7.

Conclusion

In a pediatric population, WG appears slightly less sensitive than NPS for the detection of SARS-CoV-2. Nonetheless, because WG is non-invasive, non-painful, and self-collected, more children might accept being tested on repeated occasions using this method. Consequently, availability of WG testing might facilitate the investigation of outbreaks in schools by public health services.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):54–55.

P77. Persistent Pandoraea sputorum bacteremia in a patient with extensive burns: A case report

Victoria Weaver ^1,^✉, Matthew Clifford-Rashotte ¹, Valéry Lavergne ², Titus Wong ²

Background

Identified as a distinct genus in 2005, Pandoraea sputorum is an emerging, multi-drug-resistant pathogen known to cause chronic infection in patients with cystic fibrosis (CF). There has been one previously reported case of P. sputorum bacteremia in an immunocompromised patient following orthotopic liver transplantation. Here, we present a case of P. sputorum bacteremia in a previously immunocompetent patient with 50% total body surface area (TBSA) burns.

Case Presentation

A previously healthy 42-year-old man of no fixed address was transferred to a tertiary care centre with 50% TBSA burns. He underwent multiple surgeries for debridement and skin grafting and received meropenem, daptomycin, doxycycline, and fluconazole for polymicrobial wound infections. On post-admission day (PAD) 25, blood cultures were positive for a gram-negative bacillus, identified by matrix-assisted laser desorption ionization–time of flight as P. sputorum. The broth-based microdilution system Phoenix (BD, Franklin Lakes, New Jersey, USA) revealed intermediate susceptibility to imipenem–cilastatin, ciprofloxacin, tigecycline, and doxycycline. All other antimicrobials tested—including colistin—were characterized as resistant according to the Clinical and Laboratory Standards Institute breakpoints (M100-Ed29) for non-Enterobacteriaceae. All lines were changed on PAD 29, with imipenem–cilastatin extended infusion started on PAD 31 and addition of intravenous ciprofloxacin on PAD 32. Blood cultures drawn on PAD 33 were negative. Neither transthoracic nor transesophageal echocardiogram could be completed to assess for endocarditis or suppurative thrombophlebitis because of the degree of the patient’s burns.

Conclusion

In this patient, acquired immunodeficiency from significant burns and exposure to broad-spectrum antimicrobials likely contributed to development of bacteremia with P. sputorum. Despite its extensive drug resistance, blood cultures cleared with use of two antibiotics to which the isolate tested intermediate. This is the second reported case of P. sputorum bacteremia, although previous cases may have been misidentified as Burkholderia or Ralstonia spp.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):55.

P78. Validation of the EuroImmun anti-SARS-CoV-2 ELISA assay for detection of IgG and IgA antibodies in human breast milk

Shaista Anwer ^1,^✉, Samantha Ismail ², Yvonne Yau ^2,³, Sharon Unger ^2,⁴, Deborah O’Connor ^2,^3,⁴, Susan M Poutanen ^1,²

Objectives

The potential protective effect of breast milk through severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibodies is not known. Our objectives were to validate the use of the EuroImmun Anti-SARS-CoV-2 enzyme-linked immunosorbent assay (ELISA) to detect anti-SARS-CoV-2 antibodies (immunoglobulin A [IgA] and immunoglobulin G [IgG]) in human milk.

Methods

SARS-CoV-2 IgG and IgA positive serum was used to spike serum and defatted acellular human milk from pre-2019 in triplicate, which were then diluted 1:2 to 1:32 (IgG) and 1:2 to 1:128 (IgA). The limit of detection (LOD) and paired t-test comparing the average optical density was determined. One hundred human milk samples collected pre-2019 and 50 spiked samples at a dilution of 1:2 and at their defined LOD were tested to verify the impact of different human milk matrices.

Results

Both SARS-CoV-2 IgA and IgG antibodies were detected in spiked milk. Although the LOD for spiked milk and serum samples was equal (1:16 for IgG and 1:32 for IgA), the mean optimal density at each dilution was significantly lower in breast milk than in serum (paired t-test, p = 0.0005 IgG; paired t-test, p = 0.0099 IgA). Of the 100 pre-2019 breast milk samples, all samples tested negative for SARS-CoV-2 IgG (100%), whereas 2% (2/100) tested borderline and 1% (1/100) tested positive for IgA. Of the 50 spiked samples, 100% (50/50) tested positive for IgG at 1:2, but 30% (15/50) tested borderline at 1:16; IgA was detected in all samples (100%) at 1:2 and 1:32 dilutions.

Conclusion

The EuroImmun Anti-SARS-CoV-2 ELISA kits effectively detected SARS-CoV-2 IgA and IgG antibodies in human breast milk. There is a small matrix impact of human milk affecting the optical density that did not have a significant impact on the limit of detection. Of the samples, 3% had potential IgA cross-reactivity or non-specificity, which is being further explored by neutralization assay.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):55–56.

P79. Ct as an indicator of SARS-CoV-2 WGS success

Kimberley A Macdonald ^1,^2,^✉, Chris Fjell ², Hind Sbihi ^2,³, Dan Fornika ², Linda Hoang ^2,⁴, Natalie Prystajecky ^2,⁴

Objectives

The BC Centre for Disease Control Public Health Lab (BCCDC PHL) performs whole-genome sequencing (WGS) of coronavirus disease (COVID) positives to help investigate clusters. Testing is done via quantitative polymerase chain reaction (qPCR), in which cycle threshold (Ct) is considered an indicator of viral load in samples. Ct values and WGS success were assessed to determine whether Ct values could help predict success of samples submitted for genomics.

Methods

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) qPCR testing was done in house and by British Columbia hospital and private laboratories. Ct values were obtained for five different SARS-CoV-2 gene targets (RdRp, E, N, N2, ORF1ab) for 43% of the total samples (7,448) sequenced in British Columbia. Data analysis was carried out using Microsoft PowerBI Desktop (version 2.68.5432.841). Sequencing was performed at the BCCDC PHL using an Illumina MiSeq (1,200 bp Freed amplicons). WGS quality control (QC) metrics were used to determine success or failure (85% or higher genome completeness) in phylogenomic analysis. QC metrics were produced using the BCCDC fork (artic v1.1, ncov-tools v1.3) of the Connor Lab Pipeline.

Results

Of the 3,220 samples with Ct values, 87.5% passed WGS QC; of the 4,228 without Ct values, 83.1% passed WGS QC. Samples that passed QC had Ct values that ranged from 9 to 38. Those of failed samples ranged from 16 to ≥40. At Ct 34, 38.5% of samples passed QC. After this, the WGS success rate dropped more steadily (25.3% at Ct 35, 14.7% at Ct 36, and 3.9% at Ct 37), and failures increased.

Conclusion

If capacity was exceeded for the SARS-CoV-2 WGS response, a Ct cut-off of 36 or 37 could be established to focus priorities on samples more likely to succeed in phylogenomic analysis for outbreak investigations. Urgent investigations may still require some flexibility, however, because the need to sequence a critical sample may surpass the risk and costs of WGS failure.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):56.

P80. Sex-based differences in asexual organisms? Differences in the prevalence of antibiotic resistance and multi-drug resistance in a retrospective sample of bacterial isolates collected from acute care male and female inpatients

Diana Peragine ¹, Christine Peragine ^2,^✉

Objectives

Despite well-known differences in immune and infection response, gender-based analyses exploring differences in antibiotic resistance are absent from the literature. To address this gap, this research explored differences in the prevalence of antibiotic-resistant organisms (ARO) and multi-drug-resistant organisms (MDRO) in a retrospective sample of bacterial isolates collected from male and female acute care inpatients at Sunnybrook Health Sciences Centre (SHSC).

Methods

Susceptibility data for aerobic gram-negative clinical isolates collected from inpatients from October 2002 to September 2016 were extracted from the SHSC Microbiology Database. Isolates were classified by host sex (male, female), infection type (hospital acquired, community acquired), and resistance status (ARO, MDRO). Chi-square tests were used to explore relationships among host sex, infection type, and resistance status.

Results

Of the 41,678 unique isolates identified, 46% (19,152) were collected from males, 52% (21,686) were hospital acquired, 70% (29,043) were classified as ARO, and 30% (12,376) were classified as MDRO. Isolates collected from males exhibited higher ARO rates (74% of male isolates were ARO versus 66% of female isolates, p <0.001; 79% of hospital-acquired male isolates were ARO versus 71% of hospital-acquired female isolates, p <0.001; 67% of community-acquired male isolates were ARO versus 62% of community-acquired female isolates, p <0.001). Isolates collected from males also exhibited higher MDRO rates (32% of male isolates were MDRO versus 27% of female isolates, p <0.001; 36% of hospital-acquired male isolates were MDRO versus 30% of hospital-acquired female isolates, p <0.001; 27% of community-acquired male isolates were MDRO versus 24% of community-acquired female isolates, p <0.001).

Conclusion

This is the first research to explore sex-based differences in ARO and MDRO prevalence. Isolates collected from male inpatients exhibited higher rates of antibiotic resistance and multi-drug resistance, but the underlying cause and implication of these sex differences are unclear.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):56–57.

P81. Expediting QC for COVID WGS analysis via dashboards

Kimberley A Macdonald ^1,^2,^✉, Dan Fornika ², Chris Fjell ², Hind Sbihi ^2,³, Linda Hoang ^2,⁴, Natalie Prystajecky ^2,⁴

Objectives

The BC Centre for Disease Control (BCCDC) Public Health Laboratory’s severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genomics (WGS) workflow involves several pipelines and steps (including metadata linkage) to generate quality control (QC) metrics before phylogenomic cluster analysis is done for coronavirus disease (COVID) cases. Dashboards were added to the workflow to expedite and automate genomic metadata linkage and interpretation and communication of QC results.

Methods

QC results were produced using the BCCDC fork (artic v1.1, ncov-tools v1.3) of the Connor Lab Pipeline for Illumina data. Microsoft Excel 2010, Visual Basic for Applications (VBA), and Microsoft PowerBI Desktop (version 2.68.5432.841) were used to combine, clean, and summarize SARS-CoV-2 WGS QC results and compared them with manual processing. Turnaround time and feasibility were considered when comparing the methods.

Results

Dashboards pulled data from multiple sources (SQL server database/extracts and spreadsheets) to generate a metadata file for WGS analysis and display visual summaries of QC metrics, flag questionable results, and monitor lineage distributions over time. Generating the ncov-tools metadata file using the dashboard took 10–22 minutes for 96–576 samples, respectively. Manually, the process took 1 minute/sample or about 1.5 hours for 96 samples (versus ~25 min using Excel with VBA macros). However, Excel can only process approximately 1.05 million rows of data, which BCCDC has already surpassed for COVID, making it unfeasible to rely on Excel alone. The Power BI dashboard, however, does not have this limitation and allows users with no programming skill to quickly create metadata files. The process for summarizing and interpreting COVID WGS QC results takes about 20 minutes for up to 576 samples, using Excel, VBA, and a dashboard (compared with about 2 hours using Excel only).

Conclusion

The use of dashboards has saved several hours of manual labour per week compared with manual metadata linkage. It has also improved the reproducibility of data processing through a standardized process and expedited the interpretation and communication of QC data for SARS-CoV-2 WGS.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):57.

P82. Frequency of carbapenemase-producing Enterobacterales during COVID-19

Shaista Anwer ^1,^✉, Bryn Hazlett ¹, Melissa Kissoon ¹, Allison McGeer ^1,², Tony Mazzulli ^1,², Susan M Poutanen ^1,²

Objectives

Carbapenemase-producing gram-negative Enterobacterales (CPE) are an increasing threat. Introduction of CPE is facilitated by international travel from areas of high prevalence to regions of low prevalence. The goal of this study was to review the frequency of CPE isolated in a large tertiary-care clinical microbiology laboratory serving a metropolitan urban region in the context of coronavirus disease 2019 (COVID-19), in which international travel has been dramatically reduced.

Methods

The total number of CPE per year (all CPE/y excluding duplicates) and the total number of individual carbapenemases per year (all carbapenemase/y excluding duplicates) were graphed. The trend over 2020 in which international travel was dramatically reduced was reviewed.

Results

See Figure P82-1.

Conclusion

A drop in the frequency of CPE was noted in 2020, correlating with a drop in all carbapenemases with the exception of OXA-48/OXA-48-like and IMP enzymes. A silver lining of the disruption COVID-19 has caused the world may be a decrease in the rise of CPE related to a drop in international travel.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):57–58.

P83. Realization of theoretical savings after implementation of bioMérieux’s VITEK® MS for rapid microbial identification in patients with bloodstream infections

Georgia L Carr ^1,^✉, Wendy Gouliquer ¹, Aaron Skillen ¹, David Welbourne ¹, Gregory Gamble ¹

Objectives

Rapid pathogen identification can improve patient flow and optimize patient care in patients admitted with bacteremia or candidemia. We implemented matrix-assisted laser desorption ionization–time of flight VITEK® MS to provide rapid microbial identification results and investigated the change in inpatient bed occupancy and theoretical cost savings.

Methods

We performed a pre–post observational study analyzing data of patients admitted with bloodstream infections (BSI) in a remote Canadian secondary care hospital. Data from April 2016 to March 2017 (pre) was compared with data from June 2019 to May 2020 (post). We performed subgroup analysis stratified by intensive care unit (ICU) and general ward admission. Data were extracted from the hospital information system (Meditech version 5.67). We adjusted for the expense (in dollars) and volume variances, thereby correcting for the increase in number of patients with BSI in the post-period.

Results

Two hundred fifty patients with confirmed BSI were admitted to the ICU (pre, n = 116; post, n = 134) and 680 were admitted to the inpatient wards (pre, n = 294; post, n = 386) over the total study period. The difference in average length of stay was 5.8 days for the ICU-admitted patients and 3.11 days for patients on the wards. Theoretical savings for ICU patients (cost/day $1,385) was discounted by 3.24 days at general ward costs (cost/day $287). Total theoretical savings for all included patients counted up to $944,145. Accounting for expense and volume variance, average cost per patient stay savings were 23.9% and 17.9%, respectively. Investment payback time was calculated to be 9.19 months.

Conclusion

The implementation of VITEK MS was associated with a reduction in inpatient bed occupancy, translated into total theoretical savings. Laboratory innovation and investment results in downstream opportunities for economic efficiency and is financially viable when laboratory and clinical budget silos are removed.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):58–59.

SP01. Examination of the horizontal gene transfer dynamics of an integrative and conjugative element in Histophilus somni

Mai Mohamed Farghaly ^1,^✉, Mohammad Mostafa Nazari ¹, Michael Francis Hynes ², Sylvia Checkley ¹, Neil Rawlyk ³, Karen Liljebjelke ¹

Objectives

Histophilus somni is a gram-negative bacterium that causes histophilosis in cattle and contributes to bovine respiratory disease (BRD). These diseases are significant causes of morbidity and mortality in feedlot cattle. Integrative and conjugative elements (ICEs) are autonomous-transferred mobile genetic elements that play a significant role in disseminating antimicrobials between bacteria via horizontal gene transfer (HGT). One recently sequenced ICE from H. somni isolated from feedlot cattle in Alberta and named ICEHs02 is 72,914 base pairs in length and comprises 79 genes, including the tetracycline, aminoglycosides, florfenicol, sulfonamide, and multicopper oxidase resistance genes. This study aimed to investigate the host range of ICEHs02 and to assess the effect of antimicrobial stressors on the frequency of the transfer of the ICE.

Methods

In vitro conjugation assays were conducted to examine the frequency of transfer of ICEHs02 into other bacteria. Polymerase chain reaction (PCR) and sequence analysis were used to confirm the presence of ICEHs02 and its circular intermediate in the recipient strains. Susceptibility testing of the ICEHs02-carrying recipients were conducted by broth microdilution. The effect of the antimicrobials on the excision rates and transfer frequency was investigated by mating assays and quantitative PCR.

Results

ICEHs02 was shown to transfer into H. somni and Pasteurella multocida strains. PCR assays confirmed the ICE-associated core genes, accessory genes, and the excised circular form in the transconjugants. Susceptibility testing confirmed the functional activity of the ICEHs02-associated resistance in the recipient strains. The copy numbers of ICEHs02 per chromosome exhibited a significant increase of approximately 37- fold after tetracycline exposure and an approximately 4-fold increase after ciprofloxacin treatment. The transfer rates were increased significantly upon tetracycline and ciprofloxacin induction.

Conclusion

This study emphasized the importance of ICEs in disseminating antimicrobial resistance between bacteria and demonstrated the effect of antimicrobial stressors on the transfer rates of ICEs.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):59–60.

SP02. Antiviral use among Canadian children hospitalized for influenza, 2010–2019

Kayur Mehta ^1,^✉, Shaun Morris ², Julie A Bettinger ³, Wendy Vaudry ⁴, Taj Jadavji ⁵, Scott A Halperin ⁶, Christina Bancej ⁷, Manish Sadarangani ^3,⁸, Nandini Dendukuri ^1,⁹, Jesse Papenburg ^1,^9,^10,¹¹

Objectives

Antivirals are recommended for children hospitalized with influenza but are underused. Data on their use are scarce and have focused on the 2009 H1N1 pandemic period or earlier. We describe antiviral prescribing during influenza admissions in Canadian pediatric centres after the pandemic and identify factors associated with antiviral use.

Methods

Active surveillance was performed for laboratory-confirmed influenza hospitalizations among children aged ≤16 years at the 12 Canadian Immunization Monitoring Program Active (IMPACT) hospitals from 2010–2011 to 2018–2019. Logistic regression analyses were used to identify factors associated with antiviral use.

Results

Among 7,545 patients, 57.4% were male; median age was 3 years (IQR 1.1 to 6.3). Overall, 41.3% received antivirals; 72.8% received antibiotics. Antiviral use varied across sites (range 10.2% to 81.1%) and influenza season (range 19.9 to 59.6%) and was more frequent in children with one or more chronic health condition (52.7% versus 36.7%; p <0.001). Children who received antivirals had increased markers of disease severity (median hospitalization duration 4 versus 2 d, p <0.001; intensive care unit [ICU] admission, 27.8% versus 8.7%, p <0.001; influenza-related mortality, 0.9% versus 0.2%, p <0.001). On multivariable analysis, factors associated with antiviral use included older age (adjusted odds ratio [aOR] 1.04, 1.02 to 1.05), more recent season (highest aOR 9.18, 6.70 to 12.57, for 2018–2019), admission during peak influenza period (aOR 1.37, 1.19 to 1.58), availability of local treatment guideline (aOR 1.54, 1.17 to 2.02), timing of laboratory confirmation (highest aOR 2.67, 1.97 to 3.61, for result available before admission), presence of chronic health conditions (highest aOR 4.81, 3.61 to 6.40, for cancer), radiographically confirmed pneumonia (aOR 1.39, 1.20 to 1.60), antibiotic treatment (aOR 1.51, 1.30 to 1.76), respiratory support (aOR 1.57, 1.19 to 2.08), and ICU admission (aOR 3.62, 2.88–4.56).

Conclusion

Influenza antivirals were underused in Canadian pediatric hospitals, including among children with high-risk chronic health conditions. Prescribing varied considerably across sites, increased over time, and was associated with patient- and hospital-level characteristics. Multifaceted hospital-based interventions are needed to strengthen adherence to influenza treatment guidelines and antimicrobial stewardship practices, especially among children with high-risk conditions.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):60.

SP03. Jamestown Canyon and snowshoe hare virus seroprevalence in New Brunswick

Jacqueline Mincer ^1,^✉, Stefanie Materniak ², Kristina Dimitrova ³, Heidi Wood ³, Mahmood Iranpour ³, Antonia Dibernardo ³, Courtney Loomer ³, Michael A Drebot ³, Robbin Lindsay ³, Duncan Webster ^1,^2,⁴

Objectives

Jamestown Canyon virus (JCV) and snowshoe hare virus (SSHV) are wide-ranging mosquito-borne arboviruses within the California serogroup (CSG) that are known to circulate in New Brunswick. Despite the potential for debilitating central nervous system manifestations, the prevalence of human exposure to these viruses in New Brunswick is unknown. The goal of this study was to quantify rates of human exposure in New Brunswick to these neglected arboviruses.

Methods

A retrospective, anonymized provincial serosurvey was performed using a stratified random sample of residual sera submitted between May 2015 and August 2016 for routine screening. The selection process was stratified for age, gender, and regional health care zone to ensure proportionate sampling. To determine the seroprevalence of JCV and SSHV, competitive enzyme-linked immunosorbent assay–positive samples were confirmed positive using plaque-reduction neutralization testing (PRNT).

Results

A total of 452 serum samples were screened. The CSG virus seroprevalence in New Brunswick was estimated to be 31.6%. PRNT results indicated that most single virus exposures were due to JCV versus SSHV.

Conclusion

The prevalence of human exposure to CSG viruses is high but comparable to rates observed in other Atlantic Canadian jurisdictions. Studies such as this provide important baseline epidemiological data with regard to the risk of human exposure to these neglected arboviruses. This information plays a central role in the development and refinement of public health policies for the management of mosquito-borne infections in New Brunswick.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):60–61.

SP04. Evaluation of the identification of Staphylococcus pseudintermedius from human wound cultures

Helen L Bibby ^1,^✉, Kristen L Brown ^1,²

Objectives

Staphylococcus pseudintermedius, an emerging zoonotic pathogen in humans, can easily be mistaken for Staphylococcus aureus using phenotypic methods. We began confirming the identification of all coagulase-positive staphylococci isolated from wound cultures with matrix-assisted laser desorption ionization–time of flight mass spectrometry and looked to determine the incidence of S. pseudintermedius since the change and at what cost.

Methods

A retrospective review was performed on all wound swab cultures from which coagulase-positive staphylococci were isolated 7 months before and after the change in identification procedure.

Results

A total of 49 S. pseudintermedius isolates were identified. The number of S. pseudintermedius isolates from wound cultures as a proportion of isolated coagulase-positive staphylococci increased significantly from the before (0.2%) to the after (1.0%) period (difference 0.9%; 95% CI 0.56% to 1.3%; p <0.0001)). Susceptibility testing was performed in 44 isolates; only 1 had a minimum inhibitory concentration of 0.5–2.0 μg/mL, the range in which if the isolate was misidentified as S. aureus, a very major error (VME) in susceptibility interpretation would occur. The increase in cost of the change in identification procedure was $14,400 per year in our laboratory, which performs microbiology testing for community and acute care patients in a zone servicing nearly 1.7 million people.

Conclusion

The overall incidence of S. pseudintermedius in wound cultures is very low, and misidentification as S. aureus would result in a VME in susceptibility reporting within an acceptable range (VME <1.5%). Although we will only continue to learn more about this emerging pathogen if we make attempts to properly identify it in clinical cultures, the additional time and cost involved may be unacceptably high in some laboratories. With continual pressures to minimize laboratory costs and turnaround times, we must closely examine our practices.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):61.

SP05. Getting to the heart of the matter: A case of Bartonella quintana endocarditis with serological cross-reactivity to Coxiella burnetii

Heather Glassman ^1,^✉, Stanley C Houston ¹, Anas Alrohimi ¹, Peter Dobrowolski ¹, Madeline Oosthuizen ², Justin Z Chen ¹

Objectives

Bartonella quintana is a known cause of culture-negative endocarditis commonly diagnosed by serology. We present a case of B. quintana endocarditis with serological cross-reactivity to B. henselae and Coxiella burnetii.

Methods

A 31-year-old Caucasian female teacher working in a remote Indigenous community was found collapsed and confused in her classroom. Previous history included a repaired aortic valve (AV) and ventricular septal defect at birth with residual moderate aortic regurgitation and stenosis. Examination revealed neither fever nor peripheral stigmata of endocarditis. Exposure history included a pet cat and a visit to a petting zoo.

Results

Brain MRI demonstrated multifocal strokes. Transesophageal echocardiogram revealed AV vegetations. Serial blood cultures were negative without antibiotic influence. She developed a single fever of 38.3 °C 4 days after admission. Ceftriaxone was prescribed for presumed culture-negative endocarditis. Serology results returned 4 weeks later: C. burnetii immunoglobulin (Ig) M positive, Phase I IgG <1:32, Phase II IgG 1:64, B. henselae IgG 1:1,024, and B. quintana IgG 1:1,024. Therapy was changed to hydroxychloroquine, doxycycline, and gentamicin. Blood polymerase chain reaction (PCR) was then requested, and results were positive for B. quintana and negative for B. henselae and C. burnetii. Hydroxychloroquine was stopped, and a 6-week course of doxycycline was completed. She remains well and without significant neurologic or cardiac sequelae at 1-year follow-up.

Conclusion

B. quintana is a rare cause of infective endocarditis. Known risk factors of homelessness and chronic alcoholism were not seen in this patient. Serological cross-reactivity occurred with C. burnetii, which has only rarely been described in the literature. This case illustrates the values and limitations of serological testing for culture-negative endocarditis. In such cases, PCR may be used to confirm the diagnosis.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):61–62.

SP06. Difference between age- and weight-based approach to determine adequate volume of blood culture volume drawn from pediatric patients

Eugene YH Yeung ^1,^2,^✉, Nicole MA Le Saux ³

Objectives

The optimal blood culture volume required from the pediatric population is often a mystery because there is no unified guidance. A recent review summarized the age-dependent and weight-dependent approaches on the basis of the established guidance. It is unknown whether the pediatric blood culture volumes collected at our children’s hospital were compliant with the guidance. It is also unknown whether the age-dependent or weight-dependent approaches would lead to too few blood culture bottles being taken.

Methods

A retrospective audit was conducted on all pediatric patients with blood culture ordered from November 1 to November 7, 2020, at our 167-bed pediatric teaching hospital. Patients aged ≥10 years or weighing ≥20 kg were determined to require at least two blood culture bottles in each order, as per established guidance.

Results

Of the 61 blood culture orders, 3 (mean patient age 5.51 y) were positive. Using the age-based approach, we found that 8 orders (13.11%) had too many blood culture bottles collected, and all showed no growth; 6 orders (9.84%) had too few blood culture bottles collected, and all showed no growth. Using the weight-based approach, we found that 6 orders (9.84%) had too many blood culture bottles collected, and all showed no growth; 9 orders (14.75%) had too few blood culture bottles collected, and all showed no growth. The age-based and weight-based approaches disagreed with each other in 9 orders (14.75%).

Conclusion

Increasing the number of blood culture bottles collected would possibly improve the positive culture yield among pediatric patients, especially if aged ≥10 years or weighing ≥20 kg. However, it is unknown whether the age-based or weight-based approach is superior in detection of pathogens. We need a better unified approach to improve compliance to the established guidance to determine adequate volume of blood culture collected.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):62.

SP07. Invasive mould disease in fatal COVID-19: A systematic review of autopsies

Brittany E Kula ^1,^✉, Cornelius J Clancy ², M Hong Nguyen ², Ilan S Schwartz ¹

Objectives

Invasive mould disease (IMD)—most commonly pulmonary aspergillosis—is reported to affect up to one-third of critically ill coronavirus disease 2019 (COVID-19) patients. Most reported cases are diagnosed with probable or putative COVID-19– associated pulmonary aspergillosis (CAPA) on the basis of a combination of non-specific clinical, radiographic, and mycological findings, but the clinical significance—and whether these cases represent true invasive disease—is unresolved. We performed a systematic review of an autopsy series of decedents with COVID-19 for tissue-proven evidence of IMD.

Methods

We searched PubMed, Web of Science, OVID (Embase), and MedRχiv for English- or French-language case series published from January 1, 2019, to September 26, 2020. We included series describing lung histology of three or more decedents, and authors were contacted for missing information as necessary. We recorded patient-level data when available.

Results

We identified 51 case series describing autopsies of 702 decedents. Individual-level data were available for 430 decedents. The median age was 72 (IQR 61 to 80) years. Diabetes mellitus, pre-existing lung disease, and immunocompromising conditions were reported for 129 (32%), 95 (22%), and 25 (6%) decedents, respectively. The median hospitalization length was 10 (IQR 5–22) days. Of decedents, 51.6% had received mechanical ventilation for a median of 9 (IQR 5–20) days. Treatment included immunomodulation in 60 decedents (most often steroids or tocilizumab) and antifungals in 41 decedents. Eleven decedents (1.6%) had autopsy-confirmed IMD (6 with CAPA, 4 with invasive pulmonary mycosis not specified, and 1 with disseminated mucormycosis). Among 173 decedents who received mechanical ventilation, 5 had IMD (2.9%).

Conclusion

Autopsy-proven IMD, including CAPA, is uncommon in fatal COVID-19.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):62–63.

SP08. Risk factors and protective measures for health care worker infection during highly infectious viral respiratory epidemics: A systematic review and meta-analysis

Chenchen Tian ^1,^✉, Olivia Lovrics ², Alon Vaisman ³, Ki Jinn Chin ⁴, George Tomlinson ⁵, Yung Lee ⁶, Marina Englesakis ⁷, Matteo Parotto ^4,⁸, Mandeep Singh ^4,⁸

Objectives

Health care workers (HCWs) are susceptible to emerging infectious diseases as a result of occupational exposures to patients and contaminated environments. This systematic review and meta-analysis investigated risk factors for HCW infection in viral respiratory pandemics (severe acute respiratory syndrome coronavirus 2, Middle East respiratory syndrome, severe acute respiratory syndrome coronavirus 1, influenza A H1N1, influenza H5N1) to improve understanding of HCW risk management amid the coronavirus disease 2019 (COVID-19) pandemic.

Methods

MEDLINE, EMBASE, CINAHL, and Cochrane CENTRAL databases were searched from conception until July 2020 for studies comparing infected HCWs (cases) and non-infected HCWs (controls) and risk factors for infection. Outcomes included HCW type, infection prevention practices, and medical procedures. Pooled effect estimates stratified meta-analysis and inverse variance meta-regression analysis were completed. GRADE framework was used to rate certainty of evidence.

Results

Fifty-four comparative studies were included (n = 191,004 HCWs). Compared with non-front-line HCWs, front-line HCWs were at increased infection risk (odds ratio [OR] 1.66; 95% CI 1.24 to 2.22), and risk was greater for HCWs involved in endotracheal intubations (risk difference 35.2%; 95% CI 21.4 to 47.9). Use of gloves, gown, surgical mask, N95 respirator, face protection, and infection training were strongly protective against infection. Meta-regression showed reduced infection risk among front-line HCWs working in facilities with infection-designated wards (OR −1.04; 95% CI −1.53 to −0.33, p = 0.004) and performing aerosol-generating medical procedures in designated centres (OR −1.30; 95% CI −2.52 to −0.08; p = 0.037).

Conclusion

During highly infectious respiratory pandemics, widely available protective measures such as use of gloves, gowns, and face masks are strongly protective against infection and should be instituted, preferably in dedicated settings, to protect front-line HCWs during waves of respiratory virus pandemics. In the midst of the COVID-19 pandemic, these findings draw attention to important risk factors associated with HCW infection and provide direction for future research to better protect front-line HCWs.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):63.

SP09. Recovering influenza genomes from wild bird habitats for better avian flu surveillance

Kevin S Kuchinski ^1,^✉, Jun Duan ¹, Michelle Coombe ^1,², Chelsea Himsworth ^1,², William Hsiao ¹, Natalie Prystajecky ^1,³

Objectives

Avian influenza viruses (AIVs) cause severe disease in poultry and dangerous zoonotic infections in humans. Outbreaks in livestock are costly and create opportunities for novel flu viruses to adapt to human hosts, threatening future pandemics and new types of seasonal flu. A cornerstone of outbreak prevention and pandemic preparedness is surveillance of AIVs in wild waterfowl, their natural animal reservoir. Current surveillance efforts focus on testing captured or dead birds but rarely provide warning of dangerous strains. We have developed an alternative surveillance strategy using sediment specimens from wetlands habitats where virus-laden wild bird feces accumulate. Here, we report a custom laboratory method to recover influenza genomes from sediment using targeted genomic sequencing.

Methods

A proof-of-concept experiment was conducted on four sediment specimens collected near Vancouver, British Columbia, during fall 2016. Total RNA was extracted from these sediments and prepared into metagenomic sequencing libraries. Hemagglutinin, neuraminidase, and matrix segments of the influenza A virus genome were enriched using 600 custom hybridization probes, then sequenced on an Illumina MiSeq. Consensus sequences of hemagglutinin and neuraminidase genome segments were generated by mapping sequencing reads to subtype-specific references using the Burrow-Wheeler Aligner. Reference sequences were selected by querying the reads from each library against a BLAST database containing 8,555 AIV sequences.

Results

Seventeen hemagglutinin and neuraminidase segments were recovered from these sediment samples, 9 of which were at least 80% complete. Sequence analysis identified six subtypes of hemagglutinin (H3, H4, H6, H9, H10, H11) and five subtypes of neuraminidase (N1, N2, N6, N8, and N9). Phylogenetic analysis revealed they were closely related to other AIV sequences previously isolated from wild birds in the Pacific Northwest in recent years.

Conclusion

Targeted genomic sequencing of wetlands sediment has demonstrated its potential as a viable complement to current AIV surveillance practices.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):63–64.

SP10. Nine-month evaluation of a prenatal universal screening program for hepatitis C in Alberta, Canada

L Alexa Thompson ^1,^✉, Carmen L Charlton ^1,²

Objectives

Hepatitis C virus (HCV) can be vertically transmitted during pregnancy and cause liver complications. Currently, pregnant women are only screened for HCV if they demonstrate risk behaviours for infection (risk-based screening). However, it is unclear how effective risk-based screening is at diagnosing HCV in pregnant women. We evaluated 9 months of a prenatal HCV universal screening program to estimate prevalence and determine which screening program is optimal for diagnosing prenatal HCV.

Methods

A prenatal HCV universal screening program was implemented in Alberta in 2020. Routine prenatal testing and HCV antibody records were retrospectively analyzed 9 months before implementation of the universal program to estimate the prenatal prevalence from risk-based screening. Prenatal testing was assessed 9 months after universal implementation to estimate prevalence from universal screening. The absolute risk and projected number of women diagnosed with HCV/year were calculated by comparing programs. Demographic characteristics were evaluated for all prenatal patients positive for HCV.

Results

The estimated prevalence of prenatal HCV from the risk-based screening program was 0.06% (0.045% to 0.097%) and the prevalence from the universal screening program was 0.10% (0.07% to 0.13%). The absolute risk between programs was 0.04% (0.00% to 0.07%), translating to approximately 22 more prenatal women/year diagnosed with HCV through universal screening, with up to 18 of these women being chronic carriers. Most prenatal positives were aged 26–35 years and were from the Calgary or Edmonton areas. Of the 62 confirmed prenatal positives, 6 were co-infected with syphilis (9.68%), 2 with hepatitis B virus (3.22%), and none with HIV, although 4 declined testing (6.45%).

Conclusion

Universal screening identified more prenatal women infected with HCV in Alberta compared with risk-based screening. Permanently implementing universal screening could increase diagnoses among prenatal women and improve downstream outcomes, although cost-effectiveness should be considered. Women with co-infections should be monitored closely during pregnancy for pre-term complications.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):64.

SP11. Evaluating the efficacy of the DMC40-defined microbial community on Clostridiodes difficile infection

Katya S Douchant ^1,^2,^✉, Mabel Guzman-Rodriguez ¹, Shu Mei He ¹, Curtis Noordhof ¹, Stephen Vanner ^1,², Gregory Gloor ³, Emma Allen-Vercoe ⁴, Elaine Petrof ^1,², Prameet M Sheth ^1,^2,⁵

Objectives

Defined microbial communities (DMCs) have shown great promise as therapeutic options for Clostridioides difficile infection (CDI), including the highly efficacious 33-member community MET-1. However, the mechanisms by which DMCs work is unclear. DMCs are hypothesized to restore bacterial diversity in perturbed gastrointestinal populations. Here we investigate the efficacy of a diverse DMC of 40 organisms, called DMC40, against CDI.

Methods

An antibiotic-perturbed mouse model of C. difficile was used to assess the therapeutic efficacy of DMC40. C57/Bl6 mice received a cocktail of antibiotics (kanamycin, gentamicin, colistin, metronidazole, and vancomycin) for 3 days. Mice were orally gavaged with either saline, DMC40, C. difficile ribotype-027 (CD-027), or DMC40 +CD027 (n = 6 in each group). C. difficile disease severity was evaluated using clinical signs consistent with C. difficile disease in this model including weight loss, colonic histopathology, and C. difficile toxin A (TcdA) and toxin B (TcdB) activity using fibroblast rounding assays and toxin stool concentrations.

Results

Treatment with DMC40 did not protect mice from significant weight loss after C. difficile infection, with DMC40-exposed mice experiencing similar weight loss to mice in the CD027 group (p = 0.4). However, stool concentrations of TcdA were significantly lower in the DMC40 group compared with the CD-027 group (8.31 versus 17.48 ng/mL; p <0.001), whereas TcdB concentrations were the same (15.02 ng/mL versus 10.40 ng/mL; p = 0.09). In addition, DMC40 was not associated with reduced toxin activity in fibroblast rounding assays (CD027, 51%, versus DMC40, 51%; p = 0.99).

Conclusion

Co-administration of DMC40 with CD027 in mice resulted in detectable levels of active TcdA and TcdB and was not associated with protection from clinical CDI. These data suggest that despite the increased diversity of the DMC40 community, designing efficacious DMCs remains a challenge, and further work needs to be undertaken to elucidate the mechanisms associated with microbial community efficacy.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):64–65.

SP12. Clinical characterization of string-test–positive Klebsiella pneumoniae invasive disease compared with string-test–negative K. pneumoniae invasive disease controls

Isabella F McNamara ^1,^2,^✉, Kevin R Barker ³, Bryn Hazlett ¹, Tony T Mazzulli ^1,⁴, Susan M Poutanen ^1,⁴

Objectives

Hypervirulent Klebsiella pneumoniae (hvKp) is an emerging microbial agent that has been reported to cause tissue-invasive disease. The string test is a simple phenotypic laboratory method to identify hvKp strains. It is of interest to explore the epidemiology and clinical syndrome and validate diagnostic techniques to ensure rapid and adequate diagnosis.

Methods

From September 29, 2015, to May 31, 2019, all unique K. pneumoniae isolates from sterile sites were identified from hospitals in Toronto, Ontario, Canada. Cases were defined as patients with string-test–positive K. pneumoniae isolates, and controls were patients with string-test–negative K. pneumoniae. Controls were matched to cases (2:1) on the basis of sex, age, hospital site, and ward where the culture was drawn. Demographics, hospital course, and outcomes were collected from patient medical records. Statistical analysis was done in Excel (Microsoft Corp, Redmond, Washington, USA).

Results

During the study, 500 K. pneumoniae isolates were identified; 51 (10.2%) were identified as positive string tests; 36 were included as cases as a result of adequate matches. String-test–positive patients were more likely to have lower Charlson Comorbidity Index scores. Of the 36 cases, 16 (44.4%) had abdominal abscesses, 2 (5.6%) had pulmonary consolidations, and 6 (16.7) had both. Among the 72 controls, 7 (9.7%) cases of abdominal abscess and 15 (20.8%) pulmonary consolidations were identified. In cases and controls, 7-day mortality was 3.2% and 6.9%, respectively, and 30-day mortality was 6.5% and 13.8%, respectively. Significant rates of hospital readmission were noted in each group. Although not significant, those with diagnosed hvKp were more likely to be readmitted for related complications (odds ratio 1.72, 95% CI 0.62 to 4.76).

Conclusion

Overall, patients with string-test–positive K. pneumoniae isolates demonstrated higher rates of tissue-invasive infections in a healthier population. Invasive infections identified in the string-test–negative group could represent variant hvKp strain, and more should be done to characterize these isolates.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):65.

SP13. Challenges of pediatric prospective audit and feedback during the COVID-19 pandemic

Khaled Alsager ^1,^✉, Deonne Dersch-Mills ², Joseph Vayalumkal ^3,⁴, Cora Constantinescu ⁵

Objectives

Prospective audit and feedback (PAF) at Alberta Children’s Hospital (ACH) is typically done through face-to-face communication with health care providers using a handshake stewardship model. Since the start of the coronavirus disease 2019 pandemic, the PAF process has been done virtually. Our aim was to compare the proportion of successful PAF interventions in 2020 compared with the previous 3 years.

Methods

PAF is implemented for 1 month (during November–December) each year at the ACH by infectious disease fellows supervised by an antimicrobial stewardship physician and pharmacist. Data were collected prospectively and compared with 2017–2019 data. Analysis was completed to summarize the number of interventions, location and primary attending service of patients requiring interventions, and success rate in accepting the interventions.

Results

A total of 40 antimicrobial stewardship interventions were completed during the 2020 PAF period, compared with 15–27 interventions per month in the previous 3 years. On average, 60 patient records were reviewed daily, which was stable over the 4 years. Distribution of interventions according to location and primary attending service in 2020 was 35% general pediatrics, 27.5% surgical specialties, 17.5% neonatal intensive care unit, and 7.5% oncology. In 2020, only 30% of interventions were fully accepted compared with an average of 61% for the previous 3 years (see Figure SP13-1).

Figure SP13-1: — Outcome of antimicrobial stewardship program interventions by year

Conclusion

Acceptance rate of antimicrobial stewardship program interventions was notably lower in 2020 compared with the previous 3 years. This may be related to fewer face-to-face interactions and use of electronic documentation (compared with standard handwritten notes in the patient charts). Further study is required to understand whether other factors contributed to the decline in acceptance of PAF interventions and how PAF can be improved in the setting of hospital pandemic restrictions.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):66.

SP14. Impact of compliance with inpatient antibiotic audit and feedback on patient and economic outcomes

Sydney London ^1,^✉, Gerald McDonald ², Peter Daley ¹

Objectives

Prospective audit and feedback among inpatient antibiotic prescriptions is associated with decreased duration of treatment, decreased length of stay, and decreased cost without increase in mortality. Some audit and feedback recommendations are not followed, and this may reduce the impact of the intervention.

Methods

We performed prospective audit and feedback among tertiary care inpatients between February 1, 2020, and September 22, 2020, at day 3 of prescription for piperacillin–tazobactam or carbapenems and left recommendations in the chart for the attending physician. We retrospectively assessed compliance with recommendations as complete, partial, or none. We compared duration of piperacillin–tazobactam or carbapenem treatment, length of stay, readmission rate, and mortality rate between compliance groups. Follow-up duration was between 15 and 49 weeks.

Results

Three hundred seventy-three prescriptions were included. Mean patient age was 62.3 years, 37.5% were female, and 53.4% of admissions were to medical services. A total of 201/373 (53.4%) of recommendations were followed, 51/373 (13.7%) were partially followed, and 121/373 (32.4%) were not followed. Rate of any compliance was 252/373 (67.5%). Overall duration of target antibiotic treatment after audit and feedback was 3.7 (SD 4.7) days, length of stay was 33.9 (SD 50.4) days, readmission percentage was 42.9%, and mortality was 30.6%. Any compliance with audit and feedback reduced the duration of target antibiotic treatment from 6.2 days to 2.5 days (p <0.001), reduced length of stay from 34.1 days to 33.6 days (p = 0.89), reduced readmission rate from 47.9% to 40.5% (p = 0.18), and did not significantly increase mortality rate (31.0% versus 29.8%, p = 0.90).

Conclusion

Compliance with audit and feedback was moderate. Audit and feedback significantly reduced the use of target antibiotics but did not have a significant impact on patient and economic outcomes. With a larger sample size and longer follow-up, readmission rate may be significantly reduced.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):66–67.

SP15. Retrospective cohort comparison of outpatient antibiotic use for seniors, across British Columbia and Ontario, from 2000 to 2018

Ariana Saatchi ^1,^✉, Jennifer Reid ², Marcus Povitz ³, Salimah Shariff ², Michael Silverman ⁴, Andrew M Morris ⁵, David M Patrick ^6,⁷, Fawziah Marra ^1,⁶

Objectives

Antimicrobials remain among the most prescribed medications in Canada, with more than 90% prescribed in outpatient settings. Older adults (aged ≥65 y) prescribed antimicrobials are particularly vulnerable to adverse drug events and antimicrobial resistance. In response to observed increasing trends in antimicrobial prescribing, the province of British Columbia initiated a formal community stewardship program in 2005, with increased focus on asymptomatic bacteriuria in 2013. Unlike British Columbia, Ontario promotes stewardship but does not have a formal program. This study compared the annual prevalence of outpatient antimicrobial prescribing to seniors in two provinces in Canada to examine trends over time in the context of distinct provincial stewardship programs.

Methods

Population-based analyses using linked administrative health databases were conducted in the provinces of British Columbia and Ontario. All outpatient, oral antimicrobials dispensed to older adults were identified, from 2000 to 2018. Antimicrobials were limited to acute use only (continuous <30 d supply). Prescription rates were calculated per 1,000 older adult residents, and annual trends were examined using a Poisson regression model.

Results

Antimicrobial prescribing remained steady in both British Columbia and Ontario; British Columbia prescribed at slightly elevated rates (range 790 to 930/1,000 residents) in comparison with Ontario (range 745 to 785/1,000 residents) throughout the study period (Figure SP15-1). Per annum, rates of antimicrobial use decreased by 2.9/1,000 residents (95% CI −3.0 to −2.8; p <0.05) in British Columbia, with Ontario observing a marginal annual decrease of 0.43/1,000 residents (95% CI −0.49 to −0.36; p <0.05].

Figure SP15-1: — Ontario and British Columbia prescription rates per 1,000 seniors

Conclusion

A larger decline in outpatient antimicrobial prescribing has occurred over time in British Columbia compared with Ontario, with an observed consistent decline from 2014 onward. Despite a formal stewardship initiative in place for British Columbia, prescribing for older adults is comparable across both provinces, with more conservative use in Ontario. Further analyses to delineate patient demographics and indications for use are underway.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):67.

SP16. Extensive stomatitis caused by herpes simplex virus 1 in a small-cell lung cancer patient undergoing chemoradiotherapy: A case report and literature review

Carson Ka-Lok Lo ^1,^✉, Alison E Sumner ², Calvin Ka-Fung Lo ³, Shariq Haider ¹

Objectives

Mucositis or stomatitis caused by herpes simplex virus (HSV) reactivation is well reported in hematologic malignancies undergoing myeloablative chemoradiotherapy regimens, supporting guideline recommendations of antiviral prophylaxis in seropositive patients. However, evidence to support similar recommendations in solid tumours is lacking. We report a case of herpetic stomatitis in a lung cancer patient post-chemoradiotherapy. We also reviewed literature on HSV infections in solid tumour patients receiving chemoradiotherapy.

Methods

A comprehensive search was conducted on all English-language articles using EMBASE, Ovid MEDLINE, and PubMed.

Results

Excluding case reports, only 10 publications on HSV oral infections were found on solid tumour patients undergoing chemoradiotherapy. One randomized controlled trial showed no benefit of acyclovir prophylaxis versus placebo and overall low rates of herpetic stomatitis (5%) in head–neck cancer (HNC) patients. Nine cohort studies found no correlation between HSV shedding and chemoradiotherapy-induced mucositis. However, eight of nine cohort studies evaluated only HNC patients receiving one chemotherapy regimen (cisplatin). Case reports of HSV infections included encephalitis (n = 40), oral or genital lesions (n = 2), and other sites (n = 9; eg, esophagitis); the chemoradiotherapy regimens used before developing HSV infections varied. Our case described a 65-year-old man with extensive-stage small-cell lung cancer receiving cyclophosphamide, doxorubicin, and vincristine, a myeloablative regimen with similar drugs used for treatment of non-Hodgkin lymphoma. After his second cycle of chemotherapy, he developed febrile neutropenia with gram-negative bacteremia and otomastoiditis treated with antibiotics. He later developed perioral lesions with extensive stomatitis (Figure SP16-1) that swabbed positive for HSV-1 by polymerase chain reaction. He was successfully treated with acyclovir followed by prophylaxis and remained lesion free pursuing chemotherapy.

Figure SP16-1: — Extensive perioral lesions and stomatitis upon diagnosis (A); lesions started to crust over without new lesions developing on acyclovir (B). (Permission/consent obtained from patient before use)

Conclusion

HSV oral infections are not well studied in solid tumour patients. Larger epidemiologic studies looking at specific myeloablative regimens and their risk of viral reactivation in different solid tumours are needed to better evaluate the role of antiviral prophylaxis.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):68.

SP17. Evaluation of human antibodies of the immunoglobulin class A and G against SARS-CoV-2 in serum: A provincial public health laboratory experience

Laura Gilbert ^1,^2,^✉, Robert Needle ^1,², George Zahariadis ^1,³, Yang Yu ^1,³, Hedy Dalton-Kenny ¹, Rodney S Russell ², Peizhong Peter Wang ^2,⁴, Catherine Donovan ^2,⁵, Sandy Hookey ¹, Lei Jiao ^1,³

Objectives

The initial focus of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing has been on molecular methods using real-time reverse transcription polymerase chain reaction (RT-qPCR) to detect SARS-CoV-2 RNA. This methodology indicates the presence of viral RNA, not necessarily the presence of viable viral particles. In addition, the absence of viral RNA does not indicate past infection, recovery, or undetectable levels of virus. The ability to detect antibodies to SARS-CoV-2 is currently under investigation with various performance characteristics and indications for use, including identifying past infection. In an effort to provide comprehensive public health and microbiological services, we evaluated the performance of three serological assays to establish their utility.

Methods

We assessed the ability of the Abbott SARS-CoV-2 IgG (A-IgG), EuroImmun SARS-CoV-2 ELISA IgG (EI-IgG), and EuroImmun SARS-CoV-2 ELISA IgA (EI-IgA) kits, to detect evidence of previous infection with SARS-CoV-2. These assays were selected because our laboratory has Abbott Architect and EuroImmun platforms for the standard catalogue of public health serological tests. We tested 49 known coronavirus disease-19 (COVID-19) patients and 111 pre-pandemic stored serology specimens.

Results

We found sensitivities of 95.9%, 100.0%, and 91.3% and specificities of 98.2%, 98.2%, and 90.8%, respectively, for A-IgG, EI-IgG, and EI-IgA, using manufacturer recommended cut-offs after inconclusive results were excluded. If a two-tiered algorithmic approach was applied (ie, testing with A-IgG followed by E-IgG), 100% specificity and sensitivity could be obtained after excluding inconclusive results. Cross-reactivity of hepatitis C virus seropositive specimens was observed, resulting in false positives (p <0.05).

Conclusion

Performance characteristics in our study demonstrate the superior performance of immunoglobulin G class antibodies for investigating previous infections. In addition, using a second antibody test for supplementary testing may significantly enhance performance, particularly in lower prevalence settings. Serological test results must be evaluated in concert with clinical and epidemiological information given the potential for cross-reactivity.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):68–69.

SP18. Antibiotic usage in a neonatal intensive care unit before and after the introduction of an antibiotic stewardship intervention

Marika Hirtle-Lewis ^1,^✉, Roger Chafe ¹, Natalie Bridger ¹, Julie Emberley ¹, Cheryl Foo ¹

Objectives

Antibiotics are the most commonly used medication in a neonatal intensive care unit (NICU). Efforts are being adopted across Canada to decrease antibiotic use in neonatal care. We examined the impact of an antibiotic stewardship intervention on antibiotic usage in a NICU.

Methods

The study setting is a 26-bed level III NICU with 300–400 annual admissions. The intervention began in 2017 and consisted of once-weekly discussions between infectious disease and neonatal physicians of all patients on antibiotics for >36 hours. Using Canadian Neonatal Network data, we compared pre-intervention antibiotic usage (2015–2017) with post-intervention antibiotic usage (2017–2019). Antibiotic usage was measured as length of therapy expressed as calendar days for each infant receiving antibiotics and number of patients started on antibiotics.

Results

A total of 1,380 infants were admitted to the NICU during the study period, 715 (51.8%) during the pre-intervention period and 665 (48.2%) in the post-intervention period. There was a significant reduction in the number of infants treated with antibiotics in the post-intervention cohort (78.6% versus 61.1%, p <0.001), with a relative reduction of 22.2% (95% CI 16.4% to 27.6%). The median length of therapy was 13 days pre–antibacterial stewardship program (ASP) and 11 days post-ASP, but this change did not reach statistical significance (p = 0.12).

Conclusion

The introduction of our antibiotic stewardship intervention did not require significant additional expenditures but was associated with a reduction in antibiotic initiation in the NICU. The adoption of similar programs should be explored in other NICUs.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):69.

SP19. Are we overreacting to cutaneous diphtheria? A 10-year review of public health investigations and diphtheria toxin testing across Alberta

Natalie C Marshall ^1,^2,^✉, Maulik Baxi ³, Clayton MacDonald ^1,², Christopher Sikora ⁴, Gregory J Tyrrell ^1,²

Objectives

Classical diphtheria is a potentially fatal respiratory disease mediated by the diphtheria toxin of Corynebacterium diphtheriae. As a result of high vaccination rates against this toxin, Canadian rates of respiratory diphtheria are near zero; however, diphtheria toxin testing has become more frequent. In this study, we show that cutaneous diphtheria investigations have increased demand for toxin testing, although the toxin plays no role in this form of disease. Accordingly, we propose an alternative public health strategy to manage uncomplicated cutaneous diphtheria.

Methods

Over the past 10 years, we retrospectively determined the number of C. diphtheriae isolates identified in Alberta, the disease state of each individual tested (disease versus asymptomatic carrier), the source (eg, cutaneous, respiratory), and the number of toxin tests performed.

Results

Between 2010 and 2019, zero cases of respiratory diphtheria and three cases of cutaneous diphtheria were identified. Even in the absence of respiratory disease, diphtheria toxin testing in Alberta steadily increased, with cutaneous disease investigation protocols mandating the collection of 315 specimens from 92 individuals. These specimens revealed low rates of toxigenic C. diphtheriae colonization and extremely low rates of toxigenic C. diphtheriae disease.

Conclusion

This study challenges the concept that diphtheria toxin testing adds value for uncomplicated cutaneous disease in a highly vaccinated population. Cutaneous diphtheria investigations currently demand disproportionate lab and public health resources. Therefore, we recommend excluding toxin testing and focusing instead on ensuring adequate vaccination of patients and contacts in uncomplicated cutaneous diphtheria investigations. This would spare resources while customizing the public health response around the involvement of the diphtheria toxin in each form of disease.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):69–70.

SP20. Exploring next-generation sequencing limits of detection for antimicrobial resistance genes in genomes and metagenomic samples

Ashley M Rooney ^1,^2,^✉, Amogelang R Raphenya ³, Roberto G Melano ^1,⁴, Derek R MacFadden ⁵, Noelle Yee ², Andrew G McArthur ³, Pierre HH Schneeberger ⁶, Bryan Coburn ^1,²

Objectives

The detection limits of Illumina sequencing for the identification of antimicrobial resistance (AMR) genes in genomes and metagenomes have not been established and are needed to understand the sensitivity of the assay. We aimed to determine minimum sequencing depths needed to detect AMR genes in an Escherichia coli ST38 isolate spiked into a synthetic microbial community.

Methods

A multi-drug-resistant E. coli ST38 isolate with gyrA and parC point mutations and a bla_CTX-M-15 was combined with a microbial consortium (MET-1) at a range of relative abundances (90%, 50%, 10%, 1%, and 0.1%). The combined samples and controls were sequenced on the NovaSeq 6000 at 2 × 150 bp with an average read depth of 69 million reads ± 30 million. From each sample, sequences were subsampled from 5,000,000 to 10,000 reads and bootstrapped 100 times per subsample. Resistance genes were identified using the Resistance Gene Identifier to align either assemblies or sequences against the Comprehensive Antimicrobial Resistance Database.

Results

Using an assembly-based method for AMR gene detection, a depth of 300,000 reads or approximately 15× coverage was sufficient to detect resistance genes in E. coli ST38, as well as an additional 961 E. coli genomes. Analysis of the combined metagenomic samples demonstrated that at a range of E. coli relative abundances, 15× coverage was sufficient to detect the known resistance genes in metagenomic samples. The relative abundance of the E. coli ST38 isolate must be taken into consideration to maintain the minimum genome coverage needed to detect the point mutations and bla_CTX-M-15.

Conclusion

We found that 15× coverage or 300,000 reads is sufficient to detect AMR genes in an E. coli ST38 isolate. Maintaining 15× coverage of the E. coli ST38 isolate in a metagenome is sufficient to detect known AMR genes across a range of E. coli relative abundances (100% to 10%).

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):70.

SP21. Dimensions of poverty as risk factors for antimicrobial-resistant organisms in Canada: A structured narrative review

Teagan King ^1,^✉, Richelle Schindler ^1,², Swati Chavda ¹, John Conly ^1,^3,^4,⁵

Objectives

This is the first review of its kind in Canada to summarize the literature examining the relationship between living in poverty and risk of infection with bacterial antimicrobial-resistant organisms (AROs) from 1990 to 2020 to understand whether poverty is a risk factor for ARO infections. Poverty was broadly defined by the Statistics Canada Dimensions of Poverty (https://www.statcan.gc.ca/eng/topics-start/poverty).

Methods

A structured narrative review methodology was used. A systematic search of three databases—MEDLINE, EMBASE, and Web of Science—was performed. Two reviewers conducted screening on the basis of inclusion and exclusion criteria. Of 1,136 initial articles, 43 articles were selected for final inclusion, then critically appraised using the Joanna Briggs framework and assigned a quality rating. Findings were summarized by organism. Any bacterial ARO with any resistance pattern was included.

Results

Of the 43 studies, 15 related to methicillin-resistant Staphylococcus aureus (MRSA), with 1 study showing that higher income was protective against MRSA infection. Remaining studies showed an elevated prevalence of MRSA among homeless and Indigenous persons, with higher risk influenced by factors such as housing quality and injection drug use. Only 1 other study directly measured income and its relationship to resistance. It found higher income and higher maternal education were protective against resistant otitis media infections among children awaiting myringotomy. Of the 43 studies, 20 pertained to tuberculosis (TB), clearly showing that the burden of resistant TB is in the foreign-born population. None of these studies used income or other dimensions of poverty in analyses, making it difficult to link economic marginalization to risk of resistant TB.

Conclusion

Our findings suggest there is an association between the dimensions of poverty and ARO infections in Canada, most strongly evidenced for MRSA. Higher income was associated with reduced resistant acute otitis media. Data have yet to confirm whether poverty is a risk factor for resistant tuberculosis or other resistant infections.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):70–71.

SP22. Cost-effectiveness analysis of the BioFire FilmArray Blood Culture Identification panel molecular rapid diagnostic test compared with conventional methods for identification of Escherichia coli bloodstream infections

Kwadwo Mponponsuo ^1,^✉, Jeanine Leal ^2,^3,^4,⁵, Eldon Spackman ^1,⁵, Ranjani Somayaji ^1,^2,^3,^5,⁶, Daniel Gregson ^1,^6,^7,⁸, Elissa Rennert-May ^1,^2,^3,^5,⁶

Objectives

Gram-negative pathogens including Escherichia coli are common causes of bloodstream infections (BSI) and increasingly demonstrate antimicrobial resistance. Delays in appropriate therapy for gram-negative BSIs increase patient morbidity and mortality. Molecular rapid diagnostic tests offer faster pathogen identification and susceptibility results but higher costs compared with conventional methods. We determined the cost-effectiveness of the BioFire FilmArray Blood Culture Identification (BCID) panel compared with conventional methods of identifying E. Coli BSIs.

Methods

We constructed a decision-analytic model comparing the BCID panel with conventional methods in the identification and susceptibility testing of hospitalized patients with E. coli BSIs from the perspective of the public health care payer. The projected life expectancy of an individual was used as the time horizon. Model inputs were obtained from published literature. In the base-case analysis, cost-effectiveness was calculated by determining the per-patient admission cost, the quality-adjusted life-years (QALYs) gained, and the incremental cost-effectiveness ratio (ICER). Deterministic and probabilistic sensitivity analyses were conducted to assess the robustness of the model. All costs reflect 2019 US dollars.

Results

In the base-case analysis, the BCID panel resulted in savings of $2783.52/QALY. One-way sensitivity analyses revealed length of stay and cost per day of hospitalization to have the biggest impact on the ICER. Across varying willingness-to-pay thresholds up to $100,000/QALY, the BCID panel continued to be a cost-effective method with probabilities between 0.8 and 0.92 compared with conventional methods with probabilities between 0.07 and 0.17.

Conclusion

Our decision-analytic model found that compared with conventional testing, the BCID panel resulted in cost savings when used to assess E. Coli BSI. The ICER was most influenced by length of stay and cost per day of hospitalization and suggested that the use of BCID decreased both of these variables. Modification of this model and incorporation of local data will improve its applicability to local and national health regions.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):71.

SP23. Management and outcome of febrile neutropenia in admitted presumed immunocompetent patients with suspected viral illness

Estelle Morin ^1,^✉, Catherine Corriveau-Bourque ², Jessica Foulds ³

Objectives

Febrile neutropenia (FN) creates concern in pediatric practice because of the risk of rapidly progressing serious bacterial infections (SBIs). Protocols with empiric antibiotics designed for hematology and oncology are often applied with healthy children with FN despite lower rates of SBI in this population. This study quantified rates of proven infections in presumed immunocompetent children hospitalized with suspected viral illnesses and FN.

Methods

This was a retrospective chart review of healthy children admitted to Stollery Children’s Hospital from 2007 and 2017 with fever, absolute neutrophil counts <0.5 × 10⁹/L and viral symptoms. Primary outcomes were the incidence of SBI and radiographically confirmed bacterial pneumonia.

Results

A total of 383 patient encounters were reviewed, with 96 admissions for 82 patients meeting inclusion criteria. Eighty-eight (91.7%) encounters were managed with empiric antibiotics. Viruses were identified in naso-pharyngeal swabs and stool for 42% of patients encounters. Three blood cultures were positive for coagulase-negative Staphylococcus and one for Coryneforms, all of which were interpreted as contaminants. There were two proven urinary tract infections and a possible third with written documentation of positive cultures but no microbiologic record. Two patients were treated for possible bacterial pneumonia. Of patients, 83% had documented normalization of neutrophil counts, with a median neutropenia duration of 3.2 months (range 1 d to 9.6 y). Relevant follow-up diagnoses included chronic benign neutropenia of childhood (n = 17) and diagnosed or suspected rheumatologic or autoimmune conditions (n = 3).

Conclusion

Our results support previous findings of low rates of invasive bacterial infections in healthy patients with FN. With a SBI rate of 3.1% and few patients later found to have a pathologic etiology for their neutropenia, prospective studies would be valuable to evaluate whether a change in practice is needed regarding antibiotic use in low-risk patients with suspected viral-induced neutropenia.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):71–72.

SP24. Assays to measure the impact of previous coronavirus infection on the immune response generated from SARS-CoV-2 infection and vaccination

Fang Fang Li ^1,^✉, Aaron Liu ², Ebrima Gibbs ³, Guadalein Tanunliong ⁴, Jun Duan ^1,⁴, Natalie Prystajecky ^1,⁴, Mel Krajden ^1,⁴, Muhammad Morshed ^1,⁴, Soren Gantt ⁵, Neil Cashman ³, Inna Sekirov ^1,⁴, Agatha Jassem ^1,⁴

Objectives

The “original antigenic sin” (OAS) hypothesis describes the immune system’s preferential use of memory when it encounters a slightly different version of a pathogen, leading to a blunted adaptive response. Therefore, previous exposures to seasonal coronaviruses (HCoV) or severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) may hinder subsequent responses to SARS-CoV-2 infection or vaccination, leading to worsened outcomes of infection or protection. We assessed the performance of three pan-coronavirus immunoassays to determine their utility in addressing the role of OAS in coronavirus disease 2019 (COVID-19) clinical course and vaccination.

Methods

Pre- and post-pandemic clinical samples previously submitted to the public health laboratory were tested with three immunoassays. COVID-19 cases were confirmed by SARS-CoV-2 polymerase chain reaction (PCR). The Meso Scale Diagnostics (MSD) multiplex immunoassay detects antibodies against Spike for the HCoVs and Spike, RBD, NTD, and Nucleocapsid for SARS-CoV-2 by electrochemiluminescence. VirScan-CoV displays entire HCoV and SARS-CoV-2 proteomes on phage surfaces as peptides and identifies captured antibodies through next-generation sequencing. Surface plasmon resonance imaging (SPRi) is a biosensor technology that detects high- and low-affinity antibodies against SARS-CoV-2 Spike RBD and Nucleocapsid and Spike of HCoVs and characterizes the isotypes by subclasses and avidity.

Results

Of the 50 samples tested so far, 3 were SARS-CoV-2 PCR-positive, 34 were SARS-CoV-2 PCR-negative, and 13 were of unknown status. Preliminary results demonstrate high concordance between PCR status and the MSD assay and SPRi platform (diagnostic sensitivity of 100%). Preliminary VirScan-CoV data revealed positive signals across negative samples, suggesting that optimization to increase specificity is required.

Conclusion

Pan-CoV immunoassays are expected to become powerful tools in elucidating the contribution of OAS in COVID-19 disease severity and SARS-CoV-2 vaccine response. More samples must be tested to establish complete assay validations. Future work will include workflow analysis (turnaround time, hands-on time, scalability, etc.).

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):72.

SP25. Endocarditis due to Bartonella quintana, the etiological agent of trench fever, among four individuals linked to the same homeless shelter

Carl Boodman ^1,^✉, Terence Wuerz ¹, Philippe Lagacé-Wiens ¹

Objectives

To describe a cluster of Bartonella quintana endocarditis cases that occurred from February to May 2020.

Methods

Inpatient evaluation, outpatient follow-up, and chart review were performed for patients with culture-negative endocarditis demonstrating molecular evidence, serologic evidence, or both of B. quintana infection.

Results

Four individuals with endocarditis demonstrated positive serology to B. quintana with immunoglobulin G titres >1:8,192. In all cases, echocardiography revealed large mitral vegetations ranging in size from 16 × 9 mm to 25 × 9 mm. Among the two individuals who underwent valvular replacement surgery, 16S rRNA sequencing from valvular tissue demonstrated B. quintana genetic material with 100% homology over 731 and 741 base pairs, respectively. Two individuals presented with intracerebral hemorrhage due to ruptured mycotic aneurysms. During the 6 months before hospitalization, three individuals had presented to the emergency room with body lice ectoparasitosis, weight loss, left upper quadrant pain, chest pain, and/or neurologic symptoms. All four individuals were men aged between 33 and 62 years and had experienced homelessness in the preceding 18 months. All had used the same homeless shelter. Two individuals were from the same northern remote community. All patients received at least 6 weeks of antimicrobial therapy. Three patients were treated with combinations of doxycycline and either ceftriaxone or gentamicin. Renal impairment precluded the use of gentamicin for two patients. All patients recovered and remained well at 3-month follow-up. One patient was left with severe neurologic impairment.

Conclusion

B. quintana, transmitted by body lice, may be an underdiagnosed cause of endocarditis among people who experience homelessness in Canada. Seroprevalence and ectoparasite studies are needed to understand its burden among urban and rural communities plagued by inadequate access to housing.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):72–73.

SP26. Genomic epidemiology of carbapenemase-producing Enterobacterales at a health care facility in Toronto, Ontario, Canada

Alainna Jamal ^1,^✉, Laura Mataseje ², Victoria Williams ³, Jerome Leis ³, Nathalie Tijet ⁴, Sandra Zittermann ⁴, Roberto Melano ⁴, Michael Mulvey ⁵, Kevin Katz ³, Vanessa Allen ⁴, Allison McGeer ^1,⁶

Objectives

At a health care system (H1) in Toronto, we determined whether whole-genome sequencing (WGS) altered the initial interpretation of CPE transmission made using conventional epidemiology alone.

Methods

Patients with CPE colonization–infection identified by population-based surveillance in Toronto–Peel from October 2007 to August 2018 were included if they received health care at H1 in the year before or after CPE detection. H1 reported transmission clusters determined using conventional epidemiology. Single nucleotide variant (SNV) analysis and plasmid characterization were assessed with epidemiological data.

Results

This study included 85 patients. Fifty-one (60%) patients were first detected as CPE-colonized at H1, and 34 (40%) were first detected at another local hospital but received health care at H1 in the year before or after. H1 identified seven epidemiological transmission clusters (A–G) that included 24/85 (28%) patients. SNV analysis confirmed transmission clusters C, D, and F and suggested two additional cases belonging to cluster A. The first additional case was travel related and the likely cluster A index case (0–6 SNVs from other isolates); this case stayed on the same unit as the initially presumed index case 4 months before the initially presumed index case’s detection on another unit. The second additional case stayed in a room previously occupied by five cluster A cases. Plasmid sequence analysis excluded a case from cluster A and suggested that clusters E and G were potentially one single cluster. SNV analysis also revealed a case identified at another hospital that was 16–18 SNVs away from cluster B; no direct epidemiological link between this case and cluster B was identified, suggesting undetected inter-hospital transmission.

Conclusion

WGS can identify cases belonging to clusters that conventional epidemiology missed and can also exclude cases from clusters. Integration of routine WGS into epidemiological transmission investigations has the potential to inform prevention and control practices and more rapidly halt transmission chains.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):73.

SP27. An imperative of flu vaccination: A systematic review of the clinical characteristics of influenza–COVID-19 co-infection

Karan Varshney ^1,^2,^3,^✉, Rutvin Kyada ³, Jenna Adalbert ^3,⁴

Objectives

Influenza is a respiratory pathogen that results in hundreds of thousands of deaths annually. Since late 2019, the world has dealt with the emergence of another deadly respiratory pathogen: coronavirus disease 2019 (COVID-19). Although it is recognized that COVID-19 has a higher mortality rate than influenza, the two infections also have numerous similarities. Hence, distinguishing between the two pathogens and their respective infections can be very difficult. Moreover, how co-infection of these pathogens clinically manifests in patients is not currently well known. For this reason, the purpose of our work was to systematically review the literature to describe the clinical characteristics of influenza–COVID-19 co-infection.

Methods

Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, searches were conducted in PubMed, EBSCOhost Academic Source, ScienceDirect, Medrχiv, and World Health Organization COVID-19 database. Original articles that included patients infected with both COVID-19 and influenza and provided a description of clinical characteristics for patients were eligible.

Results

Searches generated 1,787 articles, of which 26 were eligible for inclusion in this review. In total, there were 544 co-infected patients. In addition to COVID-19 infection, 85.8% of patients were infected with influenza A and 14.2% were infected with influenza B. Of the patients, 47.4% were male, and mean patient age was 59.5 (SD 8.7) years; 13.8% of patients had at least one comorbidity. The most frequently reported symptoms were fever, cough, and shortness of breath. A total of 92 deaths (16.9%) were reported.

Conclusion

Co-infection with COVID-19 and influenza has been shown to result in many symptoms that are similar to infection with either one of these pathogens. However, the results of this work also indicate that co-infection may lead to a considerably higher risk for mortality. These findings emphasize how critical it is to ensure that as many people as possible receive the annual influenza vaccination amid the COVID-19 pandemic.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):73–74.

SP28. A dramatic shift in the nasal microbiota of individuals infected with SARS-CoV-2

Emily Moslinger ^1,^2,^✉, Kyla Tozer ^1,³, Katya Douchant ^1,⁴, Curtis Nordhoff ^1,², Shu-Mei He ^1,², Henry Wong ⁵, Prameet Sheth ^1,^2,^3,⁵

Objectives

The upper respiratory tract (URT) harbours a diverse population of microorganisms collectively referred to as the nasal microbiota. Perturbations in the URT have been seen after viral infections such as influenza and have been associated with an increase in abundance of pathogenic organisms, including Pseudomonadales, Streptococcus, and Staphylococcus in the URT. The impact of the newly emerged severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) on the URT remains unclear, and insight may provide evidence of why SARS-CoV-2 viral shedding in the URT differs among infected patients.

Methods

SARS-CoV-2–positive and –negative patients were enrolled. Extracted RNA was used to evaluate the nasal microbiome through 16S rRNA Illumina next generation sequencing (NGS). The 16S rRNA hypervariable region was targeted through standard polymerase chain reaction using V3-V4 primers. After sample purification, specimens were anonymized using an index algorithm for sample library preparation. To validate sample concentration, Qubit technology was used before sample dilution and pooling. Data were analyzed using PRISM 9.0.0 software and Microbiome Analyst.

Results

SARS-CoV-2–positive and –negative patients (n = 45/cohort) were evaluated. SARS-CoV-2–infected individuals had significantly higher alpha-diversity Shannon Index (2.059 versus 1.176; p <0.001) and beta-diversity (35.2% versus 10.1%; p <0.001) at the genus level. Specifically, the URT in SARS-CoV-2 patients had an enrichment in the genera Streptococcus and Staphylococcus (p <0.0001 for both) and a depletion of Bifidobacterium and Moraxella (p <0.0001 for both) in the URT.

Conclusion

This is the first study to evaluate changes in the URT microbiome in SARS-CoV-2 patients. Analysis revealed a statistically significant shift in the URT microbiome in SARS-CoV-2 patients and an enrichment of inflammatory species such as Streptococcus and the depletion of anti-inflammatory species such as Bifidobacterium in the URT of SARS-CoV-2 patients. These data suggest that further research needs to evaluate the impact of URT changes on SARS-CoV-2 viral shedding.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):74–75.

SP29. Empiric antimicrobial prescribing, microbiology, and antibiotic stewardship needs in early breast cancer patients in the emergency department

Victoria Bugaj ^1,^✉, Michael Guirguis ², Carlo DeAngelis ^1,^3,⁴, Christine Peragine ³

Objectives

Antibiotic stewardship is an important quality improvement practice. Solid-tumour patients are likely to have high rates of antibiotic exposure; however, there is a paucity of real-world data describing antibiotic use and infection management practices among such patients presenting to emergency departments. This research describes empiric antibiotic orders, microbiology findings, and antimicrobial stewardship needs in a retrospective sample of breast cancer patients presenting to the emergency department with probable infection.

Methods

Ambulatory breast cancer patients who received curative anthracycline- or taxane-based chemotherapy, or both (August 2013 and July 2019) and presented to the emergency department with suspected infection, neutropenia, or fever during active therapy were identified. Encounter-level data were extracted from electronic medical records and descriptively summarized. An antimicrobial stewardship pharmacist reviewed the first 70 cases and identified antibiotic-related drug therapy problems (DTPs).

Results

Three hundred thirty-nine ED encounters were identified. Of these, 71% (241/339) presented with fever, 37% (125/339) presented with neutropenia, and 42% (141/339) were treated as inpatients. Empiric antibiotics were provided in 172/339 (51%) of cases; piperacillin–tazobactam (22%; 76/339), fluoroquinolones (11%; 37/339), first-generation cephalosporins (11%; 36/339), aminoglycosides (11%; 36/339), and ceftriaxone (10%; 33/339) were the agents most frequently prescribed. A single carbapenem order was identified. Blood cultures were collected in 73% (246/339) of cases, with a 5% positivity rate (12/246). Urine cultures were collected in 60% (205/339) of cases, with a 25% positivity rate (52/205). Antibiotic-related DTPs were identified in 61% (43/70) of reviewed cases. The indication for antibiotics was unclear in 63% (27/43) of DTPs identified.

Conclusion

Although relatively low rates of empiric piperacillin–tazobactam and carbapenem use were found, the antimicrobial stewardship pharmacist identified antibiotic-related DTPs in 61% of reviewed cases. Additional research to inform and improve antibiotic use and infection management practices in patients with solid tumours is warranted.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):75.

SP30. Impact of automated WASPLab incubation on plate contamination and turnaround times for cerebrospinal fluids

Xena X Li ^1,^2,^✉, Mark Gaskin ³, Cheryl Main ^1,^2,³

Objectives

Cerebrospinal fluid (CSF) specimens were added to the WASPLab system in our regional microbiology laboratory on March 9, 2020. We investigated whether the semi-automated workflow decreased plate contamination rates and improved turnaround time (TAT) for clinically significant results. We also described the problems encountered with workflow changes.

Methods

We retrospectively reviewed all native and device-related CSF specimens in our laboratory 4 months pre– and post–WASPLab implementation. We collected data on patient demographics and CSF type, along with collection, receipt, Gram stain, and culture reporting times. Plate contamination was defined as any one of the following: organism growth outside main inoculum, growth on single plate, non-pathogenic organism in native CSF, or as determined by medical laboratory technologist or microbiologist.

Results

A total of 778 (78% native, 22% device-related) specimens were manually processed before WASPLab implementation, and 625 (72% native, 28% device-related) specimens were processed afterward. Positive Gram stain rates went from 1.7% to 0.8% (p = 0.15), and positive cultures rates went from 5.1% to 3.8% (p = 0.25). There was no change in plate contamination rates at 6.6% (51/778 and 41/625, respectively; p = 0.93). TAT for Gram stain went from 48 minutes to 44 minutes (p = 0.81). TAT for all culture results decreased from 9,792 to 7,205 minutes (p <0.05), and the TAT for positive culture reports decreased from 3,987 to 3,521 minutes (p <0.05). Major workflow errors were noted, primarily missed plate reads and missed thioglycolate broth reads. Of the samples, 27% contained at least one workflow or documentation issue.

Conclusion

The WASPLab system did not decrease plate contamination rates but did decrease TAT in positive culture reports on CSF specimens. However, increased errors were seen as a result of workflow changes. Careful monitoring when implementing automation and ongoing audits of potential workflow errors help to ensure a smooth transition.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):75–76.

SP31. COVID-19–associated pulmonary aspergillosis (CAPA) in critically ill patients: A systematic review

Ruwandi M Kariyawasam ^1,^2,^✉, Tanis C Dingle ^1,², Wendy Sligl ^3,⁴, Brittany E Kula ³, Ilan S Schwartz ³

Objectives

Invasive pulmonary aspergillosis (IPA) is a complication of critical coronavirus disease 2019 (COVID-19) infection, but the true prevalence and optimal diagnostic algorithm for COVID-19–associated pulmonary aspergillosis (CAPA) are uncertain.

Methods

We conducted a systematic review of the incidence, diagnostics, treatments, and mortality of CAPA in patients in intensive care units (ICUs). Three electronic databases (PubMed, Embase, Web of Science) and the grey literature were searched to October 26, 2020, for cohort studies reporting CAPA in the ICU. Two reviewers assessed for study eligibility, selected studies, and extracted data. Spearman’s rank correlation was performed to assess the relationship between diagnostic algorithms.

Results

After removing duplicates, we identified 81 relevant titles; after abstract review, 35 full texts were eligible for screening. Twenty studies met inclusion criteria, including 11 with patient-level details. Reported incidence rates of CAPA ranged from 3.8% to 35%; the median was 21% (interquartile range 15% to 27%). The three diagnostic algorithms for CAPA (Verweij, White, and Koehler) had poor to moderate correlation with one another (p = 0.22 to 0.45). Only 62% of the 101 patients reported to have CAPA met the most recent consensus (ie, Koehler) definitions. Mould-active antifungals were prescribed for 60 (59%) patients. The overall mortality rate was 44% (45/101), including 50% (30/60) among those treated with antifungals.

Conclusion

In this systematic review of critically ill patients with COVID-19, the incidence of CAPA ranged widely, diagnostic criteria varied, different algorithms correlated poorly, and mortality remained high despite antifungals. More research is needed on this emerging complication of COVID-19.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):76.

SP32. Comparative analysis of SARS-CoV-2 diagnosis specificity and sensitivity using three sampling methods by quantitative PCR and droplet digital PCR

Erin Collins ^1,^2,^✉, Khatereh Shir-Mohammadi ¹, Jian-Jun Jia ¹, H Chaim Birnboim ³, Curtis Cooper ^1,², Marc-André Langlois ¹

Objectives

Limitations of naso-pharyngeal (NP) collection for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) diagnosis include patient discomfort, need for professional collection, and supply shortages. Saliva and oropharyngeal (OP) sample collection have been found to be reliable alternatives, with potential to improve test compliance and facilitate population-level testing. We performed a prospective diagnostic accuracy study using NP, OP, and saliva samples from 20 SARS-CoV-2–positive cases.

Methods

NP, OP, and saliva samples were collected from 20 adults with recent positive SARS-CoV-2 nucleic acid amplification test, ≤10 days from symptom onset or positive test date, whichever was first. Saliva was collected using the OMNIgene•ORAL Kit (DNA Genotek, Ottawa, Ontario, Canada). Negative controls were collected from 10 asymptomatic adults. Viral RNA was extracted using a QIAamp Viral RNA Mini Kit (52906; Qiagen, Hilden, Germany). Quantification cycle (Cq) values (quantitative polymerase chain reaction [qPCR]) and genomic equivalent copy number determination (ddPCR) were performed for SARS-CoV-2 (Wuhan) N1 and E genes and the RPP30 internal control.

Results

ddPCR yielded one (1/10) false-positive NP swab. qPCR and ddPCR results were otherwise accurate for all positive and negative NP, OP, and saliva samples. One hundred percent sensitivity was achieved for saliva samples using both quantification methods. Although NP and saliva copy numbers and Cq values were comparable, N1 and E target gene amplification were approximately 100-fold less efficient for OP swabs. Detection of N1 demonstrated slightly higher sensitivity than the viral E gene.

Conclusion

Saliva vRNA detection sensitivity was comparable to NP sensitivity and superior to OP sensitivity. These results support the use of saliva in widespread SARS-CoV-2 testing and for population-scale epidemiological studies. Early morning collection may have favorably contributed to the performance of saliva testing. Going forward, (1) these collection strategies will be evaluated in an automated and high-throughput RNA extraction system and (2) NP, OP, and saliva viral titres will be compared and correlated with coronavirus disease 2019 symptom severity.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):76.

SP33. Community-acquired MRSA pericarditis in a pregnant woman

Omar Mourad ^1,^✉, Deborah Koh ¹, Philippe El-Helou ¹

Objectives

Community-acquired methicillin-resistant Staphylococcus aureus (MRSA) purulent pericarditis is a rare, life-threatening disease. We aim to review the current literature and present our case of MRSA pericarditis.

Methods

Here we report on a 34-year-old woman who was 24 weeks pregnant and had a known history of intravenous drug use who presented to our hospital with 3 days shortness of breath and pleuritic chest pain. Initial electrocardiogram on day 0 of admission showed diffuse ST elevation, and her transthoracic echocardiogram (TTE) showed a small pericardial effusion. She was initially started on prednisone. She did not respond to therapy, and ibuprofen was added.

On day 15 of admission, she developed rapid atrial fibrillation requiring cardioversion. A repeat TTE on day 21 showed a moderate-size pericardial effusion and fibrinous stranding. A third TTE on day 38 of admission showed a worsening effusion, and she subsequently underwent a pericardiocentesis on day 42 of her admission that drained purulent material.

Results

Pericardial fluid cultures grew MRSA, and she was treated with daptomycin. Blood cultures drawn at the same time were negative. Because of the nature of her disease, she underwent a pericardial window that was done on day 47 of admission. She responded very well to the surgery and completed a course of 8 weeks of daptomycin after her pericardiocentesis with full resolution of her symptoms to date.

Conclusion

MRSA pericarditis is a rare, life-threatening disease that requires both medical and surgical management. There are limited data on the clinical presentation of the disease as well as recommendations with regard to its management.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):77.

SP34. In vitro susceptibility of common bacterial pathogens causing respiratory tract infections in Canada to lefamulin, a new pleuromutilin

Robert M Taylor ^1,^✉, James A Karlowsky ^1,², Melanie R Baxter ¹, Heather J Adam ^1,², Andrew Walkty ^1,², Phillipe Lagacé-Wiens ^1,², Lionel Mandell ³, George G Zhanel ¹

Objectives

Community-acquired bacterial pneumonia (CABP) represents a significant global health concern. With increasing resistance to first-line agents for CABP worldwide, the need for new and improved antibacterial agents is paramount. Lefamulin, a novel pleuromutilin, was approved for the treatment of adults with CABP in Canada in July 2020. It is reported to be highly active against many lower respiratory tract bacterial pathogens and is available in both oral and intravenous formulations. This study assessed the in vitro activity of lefamulin against bacterial isolates associated with CABP in Canada.

Methods

Lower respiratory bacterial isolates cultured by hospital laboratories across Canada and submitted to the annual CANWARD surveillance study’s coordinating laboratory in Winnipeg, Manitoba, Canada, from January 2015 to October 2018 were tested. A total of 709 bacterial isolates from respiratory sources were tested against lefamulin and comparator agents using the Clinical and Laboratory Standards Institute reference broth microdilution method. Lefamulin minimum inhibitory concentrations (MICs) were interpreted using current US Food and Drug Administration breakpoints. MICs for all other agents were interpreted using current CLSI M100 breakpoints.

Results

One hundred percent of Streptococcus pneumoniae tested from respiratory (n = 315) specimens were susceptible to lefamulin (MIC ≤0.5 μg/mL), including isolates resistant to penicillin, clarithromycin, doxycycline, trimethoprim–sulfamethoxazole, as well as multi-drug-resistant isolates. Lefamulin also inhibited 99% of Haemophilus influenzae isolates (regardless of beta-lactamase production) (n = 99; MIC ≤2 μg/mL) and 95.7% of methicillin-susceptible Staphylococcus aureus (MSSA) (n = 70; MIC ≤0.25 μg/mL) at their susceptible breakpoints. Lefamulin exhibited an MIC₉₀ (concentration inhibiting 90% of isolates) value of 0.12 μg/mL versus Moraxella catarrhalis. Lefamulin demonstrated MIC₉₀ values of 0.25 μg/mL versus both MSSA and methicillin-resistant S. aureus (n = 130).

Conclusion

Lefamulin demonstrated potent in vitro activity against all respiratory isolates tested and may represent a significant advancement in empiric treatment options for CABP in Canada.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):77–78.

SP35. Atypical diagnosis of typical HUS with off-site molecular stool testing

Peter Gregory ^1,^✉, Emily MacAdam ¹, Glenn Patriquin ^1,²

Objectives

We highlight the importance of diagnostic clarity for adult patients with a classic presentation of typical hemolytic–uremic syndrome (HUS) and negative stool cultures.

Methods

A 27-year-old otherwise healthy woman was admitted to a tertiary care centre with acute kidney injury and thrombotic microangiopathy. Eight days before admission, she had acute onset of diarrhea (20 episodes/day) and was assessed in the emergency department. Her bloodwork was unremarkable. A computed tomography scan of the abdomen revealed right-sided colitis most in keeping with an infectious etiology. Clostridium difficile test was negative, and stool culture was negative for Salmonella, Shigella, Yersinia, Campylobacter, and Escherichia coli O157.

Under the care of hematology, she was initially managed with plasmapheresis and urgent dialysis. Thrombotic thrombocytopenic purpura was ruled out with a normal ADAMTS13 level. Typical HUS was thought to be ruled out with her negative stool culture, leaving a working diagnosis of atypical HUS. Before starting eculizumab, infectious diseases (ID) was consulted for vaccination and prophylaxis recommendations. With this diagnosis of atypical HUS, the patient would likely require long-term eculizumab therapy and would be an unlikely kidney transplant candidate, if required. To seek more diagnostic clarity, the ID team obtained one of the original stool samples. The stool sample was referred to a pediatric hospital for molecular testing, using a syndromic gastrointestinal pathogen panel. This test was not available at the primary laboratory because of the low incidence of HUS in the adult population.

Results

The sample was reported as positive for shigatoxin-producing E. coli. Serotyping performed at a reference lab confirmed E.coli O121 H nil.

Conclusion

This case highlights the importance of recognizing the non-O157 causes of HUS and associated laboratory limitations. Improper diagnosis in this case would have had major implications for duration, cost, and toxicities of therapy, as well as for organ transplant candidacy.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):78.

SP36. Comparative evaluation of four vaporized hydrogen peroxide–based systems to decontaminate N95 respirators

Shawn T Clark ^1,^✉, Herman Ng ², Ge Wu ², Devika Jain ³, Rita Kandel ^1,⁴, Tony Mazzulli ^1,²

Objectives

The coronavirus disease 2019 (COVID-19) pandemic has strained health care supply chains and created shortages in essential resources, including personal protective equipment (PPE). This generated interest in reprocessing single-use PPE items, such as N95 respirators, to conserve supplies. After decontamination, respirators are evaluated for filtration performance and fit; however, the efficiency of the decontamination process is not assessed. Here, we evaluated the ability of hydrogen peroxide–based methods to eliminate gram-positive and gram-negative organisms from artificially contaminated N95 respirators.

Methods

Four hydrogen peroxide–based decontamination systems, STERRAD (Advanced Sterilization Products, Irvine, California, USA), Bioquell (ECOLAB, St Paul, Minnesota, USA), Stryker (Stryker Corporation, Kalamazoo, Michigan, USA), and Clēan Works (Clēan Works, St Catharines, Ontario, Canada), were evaluated in this study. Each system was challenged with up to five 3M^TM N95 respirator models (1860, 1860S, Aura^TM 1870+, 8210, 8110S, or VFlex^TM 9105S or 1805) inoculated with suspensions of six gram-positive (S. aureus (methicillin-susceptible and methicillin-resistant Staphylococcus aureus strains, Baccilus subtilis and Enterococcus faecium [VRE]) and gram-negative (Escherichia coli, Pseudomonas aeruginosa) bacteria in biological triplicate. Inoculated areas of treated masks were cut, placed into BHI broth and incubated for 48 hours at 37˚C and sub-cultured if turbid. Colonies cultured from N95 respirators were identified by matrix-assisted laser desorption ionization–time of flight (VITEK^® MS, bioMérieux Inc, St Laurent, Quebec, Canada), and counts were compared with untreated controls to assess decontamination efficacy.

Results

Gram-negative bacteria were not recovered from any treated respirator (100% efficiency). The STERRAD and Stryker systems achieved a reduction of ≥10⁶ CFU but did not consistently eliminate gram-positive organisms. All gram-positive organisms were cultivatable from STERRAD-decontaminated respirators, ranging from 44% to 100% of inoculated areas by model. The Bioquell device consistently eliminated both gram-positive and gram-negative organisms, including VRE (100% efficiency).

Conclusion

Differences in decontamination efficacy across devices may reflect the operating properties of each device and characteristics of the bacteria and respirators examined. Further evaluation is needed to determine the clinical significance of post-decontamination organism burden.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):78–79.

SP37. Clinical characteristics, management, and outcomes of splenic abscesses

Rupinder Mangat ^1,^✉, Jonathan Gray ¹, Ted Louie ¹

Objectives

Splenic abscesses are an uncommon but challenging clinical entity, with a diverse range of etiologies. Traditional modality of treatment has been splenectomy; however, percutaneous drainage has become increasingly popular in clinical practice, especially for patients with multiple comorbidities who have a higher risk of surgical complications. Given the lack of literature related to this disease process, we designed a study to evaluate clinical characteristics, management, and outcomes of splenic abscesses at our institution.

Methods

Retrospective chart review was performed for diagnosed cases of splenic abscesses within a 5-year period. Patient characteristics, symptoms before diagnosis, imaging, and microbiological data were evaluated. Treatment modalities and patients’ outcomes were also reviewed.

Results

Twenty-one patients were found to have splenic abscesses during the study time period. Of these patients, 66.7% had a fever on presentation, and 47.6% endorsed left upper quadrant pain; 42.8% of patients were bacteremic at the time of diagnosis, and Escherichia coli was identified in culture for 19%. All patients had a computed tomography of the abdomen at the time of diagnosis. A total of 25% of patients were managed medically, and 36.8% underwent splenectomy. The remaining patients underwent aspiration or catheter drainage with a prolonged course of antibiotics. Seventeen patients recovered, three died, and one was lost to follow-up.

Conclusion

Although we observed strong trends in our study patients, our limited data set was underpowered to provide any statistically significant correlations. Notable predisposing factors include diabetes mellitus and a history of malignancy. Of the patients, 42.8% were bacteremic at the time of diagnosis; however, no striking microbiological concordance was demonstrated as the causative agent of disease. Different treatment modalities were seen in the three deceased patients: medical management alone, fine needle aspiration, and splenectomy. Given our limited data set, more studies are needed to evaluate outcomes on the basis of treatment modalities.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):79.

SP38. Prevention of COVID-19 in internally displaced persons camps in the eastern Democratic Republic of the Congo: A mixed-methods study

Qaasim N Mian ^1,^✉, Kasereka M Claude ², Muyisa S Serge ², Kahindo K Alexis ³, Michael T Hawkes ^1,^4,^5,^6,⁷

Objectives

The global coronavirus disease 2019 (COVID-19) pandemic poses a threat to refugees and internally displaced persons (IDPs). We examined knowledge, attitudes, and practices with respect to COVID-19 prevention among IDPs in the war-torn eastern Democratic Republic of the Congo (DRC).

Methods

Mixed-methods study with qualitative (focus group discussions [FGDs]) and quantitative (52-item survey questionnaire) data collection and synthesis.

Results

FGDs (total 23 participants) and survey questionnaires (164 IDPs from three camps and 143 control participants from neighboring villages) were conducted in May 2020. FGDs unearthed harrowing narratives recounting violence that IDPs had fled. IDPs differed from control participants in terms of larger household size, more extreme poverty, lower educational attainment, and lower access to information through media and Internet (p <0.05 for all comparisons). IDPs had a high level of awareness (99%) and fear (98%) of COVID-19 but lower specific knowledge (15% sufficient knowledge versus 30% among controls; p <0.0001), which remained significant in a multivariable model adjusting for potential confounding variables. IDPs faced seemingly insurmountable barriers to implementing COVID-19 prevention measures. Physical distancing was impossible for IDPs in crowded shelters, and 70% had come in close contact with someone other than a family member within the past 24 hours (compared with 56% of controls; p = 0.014). Frequent movements in and out of the camp for subsistence needs left IDPs vulnerable to introduction of COVID-19; 61% left the camp on a daily basis, and 65% had received a visitor in the past month. Despite acceptance of hand hygiene for prevention, 92% lacked soap (compared with 65% of controls; p <0.001). Pleas for peace and a return to their native homes, where COVID-19 precautions could be feasibly implemented, overshadowed perceived benefits of a COVID-19 vaccine.

Conclusion

These findings provide empiric evidence supporting the exquisite vulnerability of IDPs to COVID-19 and a call for action to protect neglected displaced populations.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):79–80.

SP39. Comprehensive genomic insight into Shiga-toxin-producing Escherichia coli (STEC) to understand virulence and detect priority STEC serotypes

Khadidja Yousfi ^1,^2,^✉, Rufaida H Khan ¹, Chrystal Landgraff ³, Katrina Lanyon ³, Isabelle Bernaquez ¹, Florence Doualla-Bell ¹, David Roy ¹, Sadjia Bekal ¹

Objectives

We exploited the genomic approach to identify virulence genes and stable potential virulence signatures in Shiga-toxin-producing Escherichia coli (STEC) that can provide potential candidates for diagnosis and knowledge of their role in pathogenesis. Moreover, we investigated the presence of bacteriocins that could play a role in enhancing bacterial pathogenicity.

Methods

Genomes of 80 STEC clinical isolates (O157 and non-O157), collected from children aged younger than 5 years, were analyzed using different bioinformatics software. Neptune software was used to identify potential virulence signatures highly specific to STEC and non-O157:H7. Bacteriocin production was tested using the stab and overlay method.

Results

The majority of the detected genes were clinically significant virulence markers, some considered to be highly specific to the STEC–enterotoxigenic E. coli pathotype. The prevalence of virulence genes espP, hlyA, hlyB, hlyD, stx1, nleG9, espM1, ospG-like, espN, espV, espW, espX7, nleC-like, tccP2, ecf1-4, bacteriocin D (cda), and several hypothetical proteins encoding genes were found in the majority of the STEC genomes because Neptune produced signatures of these genes, suggesting that they were highly common among all STEC population. The bacteriocins identified were mostly found in the non-O157 strains. The cba, cda, and celb genes were observed mainly in the Big Six isolates, whereas cvaC, mcbA, colicinV, and Microcin B17 genes were mostly found in O52 serotype. The majority of the bacteriocinogenic isolates inhibited at least one of the isolates tested.

Conclusion

We identified virulence genes that are more often associated with STEC. These findings provide fundamental information that will lead to a better characterization of STEC, which may contribute to the building of an efficient monitoring system. Our results suggest that the presence of bacteriocins in Big Six genomes make it highly prevalent after O157:H7.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):80.

SP40. WITHDRAWN

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):80.

SP41. Sapovirus molecular epidemiology in Canadian children: Insights from a 4-year province-wide prospective study

Xiaofeng Ding ^1,^✉, Ran Zhuo ¹, Luong Jasper ¹, Jianling Xie ², Stephen Freedman ², Ying Wu ¹, Linda Chui ^1,³, Bonita Lee ¹, Samina Ali ¹, Xiaoli Pang ^1,³

Objectives

The objective of this study was to determine the incidence, genetic diversity, and seasonality of sapovirus and the clinical characteristics of sapovirus infections in children in the province of Alberta, Canada, between December 2014 and August 2018.

Methods

We analyzed stool specimens and rectal swabs from 3,347 children aged younger than 18 years with acute gastroenteritis (AGE) who were enrolled through two pediatric emergency departments (EDs) and a nursing telephone advice line in Alberta. Reverse transcription real-time polymerase chain reaction (PCR) was used for sapovirus detection. Sapovirus VP1 genotype was determined by a nested PCR-gel electrophoresis-Sanger sequencing approach. The Modified Vesikari Scale scores (≥11 defined as severe) was used to characterize clinical severity.

Results

Sapovirus was identified in 9.5% (317/3,347) of the children. GI.1 (36%) was the most common genotype identified, followed by GI.2 (18%), GII.5 (8%), GII.3 (6%), GII.1 (5%), GII.2 (4%), GV.1 (3%), GII.4 (1%), GIV.1 (1%), GI.3 (0.3%), and GI.7 (0.3%). Sapovirus was detected year-round, with the peak of detection in the winter months (November–January), except when GI.2 overtook GI.1 and became the predominant strain during 2016–2017 with the resultant sapovirus peak persisting into March. Severe gastroenteritis occurred in 62% of sapovirus cases with no difference observed by age group, sex, or genogroup; however, sapovirus AGE was more severe in patients who presented to the ED.

Conclusion

Sapovirus was shown to be an important pathogen of childhood gastroenteritis in Alberta. It is necessary to conduct surveillance of circulating sapovirus genotypes to fully comprehend the role of sapovirus as a viral pathogen in AGE.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):80–81.

SP42. Overuse of piperacillin–tazobactam in patients admitted to medicine wards

Emily MacAdam ^1,^✉, Peter Gregory ¹, Paul Bonnar ^1,²

Objectives

Piperacillin–tazobactam is commonly used empirically on admission to medicine wards. We sought to collect the local indications for its use and further evaluate compliance with local guidelines and assessment of appropriateness.

Methods

A chart audit was performed on patients admitted to medicine wards and prescribed piperacillin–tazobactam during a 3-week time period. A pharmacy daily report of active antibiotic orders was used to identify patients. Data were entered into the National Antimicrobial Prescribing Survey (NAPS), and this tool was used to summarize indications, compliance with guidelines, and appropriateness.

Results

Twenty-seven unique patients were identified. Of these patients, 18 (67%) were admitted to the medical teaching unit, and 9 (33%) were admitted to hospitalist wards. The most common indications were community-acquired pneumonia (30%), unknown indication (22%), and sepsis (15%). Of those labelled as sepsis, urinary was the most common source documented despite negative urine cultures. Overall, 70% of piperacillin–tazobactam prescribing was non-compliant with guidelines, and the majority of use was inappropriate (78%). We also looked at review or stop dates and found that this was documented in only 26% of cases.

Conclusion

We were able to confirm and characterize the overuse and inappropriate use of piperacillin–tazobactam in an academic hospital. With these local data, we were able to provide education to prescribers and residents with teaching sessions, review order sets, and promote local antimicrobial stewardship resources. The NAPS tool facilitated a standardized assessment of piperacillin–tazobactam prescribing.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):81.

SP43. Treatment of bloodstream infections caused by ceftriaxone-resistant Escherichia coli, Proteus mirabilis and Klebsiella species: Review of extended-spectrum beta-lactamase production and patient outcomes in a low-prevalence setting

Shawn Smith ^1,^✉, Hilal Al Sidairi ^2,³, Emma K Reid ⁴, Carolyn Smith ⁵, Glenn Patriquin ⁵, Paul Bonnar ^3,⁴, Ross Davidson ⁵

Objectives

Invasive infections due to extended-spectrum beta-lactamase (ESBL)–producing Enterobacterales may have reduced susceptibility to beta-lactam–beta-lactamase inhibitor (BLBLI) antibiotics, contributing to broad-spectrum carbapenem use. Recent Infectious Diseases Society of America guidelines recommend that carbapenem or non-BLBLI antibiotics be used preferentially in the treatment of non-cystitis ESBL infections, even in cases in which BLBLI susceptibility is demonstrated through in vitro laboratory testing. We aimed to characterize the impact of ESBL production and antibiotic choice on clinical outcomes in patients with ceftriaxone-resistant Enterobacterales bacteremia in order to advise local practices.

Methods

Ceftriaxone-resistant Escherichia coli, Klebsiella pneumoniae, K. oxytoca, and Proteus mirabilis were obtained from adult inpatient blood cultures between March 2016 and June 2020 and subsequently frozen. Isolates were regrown from frozen stock, and ESBL status was confirmed with phenotypic disc testing. For piperacillin–tazobactam-sensitive isolates, patient charts were reviewed for admission and antimicrobial treatment data, including outcomes of death, bacteremia relapse, and hospital readmission.

Results

Of 84 ceftriaxone-resistant isolates, 62 were ESBL positive (73.8%), and 60 were susceptible to piperacillin–tazobactam. Of 51 ESBL-producing isolates in this cohort, 23 patients received definitive BLBLI treatment (piperacillin–tazobactam or amoxicillin–clavulanate), and 28 received definitive non-BLBLI treatment, primarily carbapenems. Preliminary 30-day outcomes suggest similar rates of death (13.0% versus 10.7%), relapse (8.7% versus 7.1%), and readmission (17.4% versus 25.0%) among ESBL-positive infections treated with BLBLI and non-BLBLI respectively.

Conclusion

Approximately 74% of ceftriaxone-resistant Enterobacterales blood isolates were ESBL positive, suggesting that ceftriaxone resistance was a pragmatic proxy for ESBL positivity without need for phenotypic testing. Preliminary clinical outcomes were similar for BLBLI- and non–BLBLI-treated ESBL infections, but small numbers limit generalizability. Implementing modified BLBLI sensitivity reporting in ceftriaxone-resistant isolates in favour of carbapenems should affect only a relatively small number of patients, approximately one to two per month in our centre.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):81–82.

SP44. Prediction of cytomegalovirus antiviral resistance using next-generation sequencing in a clinical virology laboratory

Samuel D Chorlton ^1,^✉, Gordon Ritchie ^1,², Tanya Lawson ^1,², Elizabeth McLachlan ³, Marc G Romney ^1,², Nancy Matic ^1,², Christopher F Lowe ^1,²

Objectives

Cytomegalovirus (CMV) infection causes a large burden of disease in immunocompromised patients, and antiviral drug resistance (AVDR) testing is critical to guide management for those patients with antiviral treatment failure. We developed a next-generation sequencing (NGS) assay for CMV AVDR genotyping, enabling faster turnaround time.

Methods

We performed a retrospective review of CMV AVDR testing requests at our laboratory between 2017 and 2019. For the NGS approach, amplicons were generated of UL97 and UL54, sequenced on a MinION sequencer and analysed with a novel cloud bioinformatics pipeline (BugSeq). Results were compared with traditional Sanger sequencing at the national reference laboratory.

Results

Twenty patient samples were included in analysis and sequenced on a MinION. NGS had 100% sensitivity for AVDR variants detected with Sanger sequencing. Six samples had no AVDR mutations detected using either NGS or Sanger. In the remaining 14 samples, NGS identified additional mutations conferring AVDR in UL97 and/or UL54 in 8 (57%) samples as compared with Sanger sequencing. These mutations were minority variants present in 10%–30% of NGS reads.

Conclusion

Amplicon NGS using a MinION enables accurate AVDR prediction for CMV resistance. Pairing this technology with a novel cloud platform can potentially enable NGS in clinical laboratories for rapid AVDR genotyping.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):82.

SP45. WITHDRAWN

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):82.

SP46. Oral administration of defined microbial community (DCM-18) protects against Clostridioides difficile infection in mice

Mabel Guzman ^1,^✉, Curtis Noordfhof ¹, Shu-Mei He ¹, Emma Allen-Vercoe ², David Hurlbut ³, Gregory B Gloor ⁴, Stephen J Vanner ¹, Elaine O Petrof ¹, Prameet M Sheth1, ^3,⁵

Objectives

Defined microbial communities (DMCs) and fecal microbial transplants (FMTs) have both shown great promise in the treatment of Clostridioides difficile infection (CDI). However, unlike FMTs, safety profiles, donor screening, and donor variation are all mitigated when using DMCs. Here we present data on the performance of a DMC, called DMC-18, as a frozen preparation (F-DMC-18) in an antibiotic-induced model of C. difficile disease.

Methods

C57BL/6 female mice were exposed to a cocktail of antibiotics (kanamycin 0.4 mg/mL, gentamicin 0.035 mg/mL, colistin 850 U/mL, metronidazole 0.215 mg/mL, and vancomycin 0.045 mg/mL) in drinking water for 3 days; then 4×10¹⁰ CFU of DMC-18 was orally administered 24 hours before CD ribotype-027 (1×10⁵ CFU) exposure. Control animals received saline 24 hours before CD ribotype-027 exposure. Daily animal weights, colon histological assessment, enzyme-linked immunosorbent assay of peripheral blood cytokines and toxins from stool pellets, followed by cytotoxic assays on fibroblast, were performed to follow disease progression.

Results

Mice pre-exposed to F-DMC-18 were protected from weight loss 48 hours after CDI (they gained weight, 11.23%, compared with the CDI group; p <0.01). F-DMC-18 pre-treatment protected mice from cecum shortening or colonic bleeding, and they had intact stool pellets, similar to uninfected control mice. Reduced tissue damage, neutrophil infiltration, and inflammation in the colon were observed as well in mice that received F-DMC-18 compared with control mice exposed to CD027 (p <0.05). F-DMC-18 pre-exposure was associated with lower systemic levels of interleukin 1a, G-CSF, and KC (all p <0.05) and reduced concentrations of TcdB in stool compared with CD mice (reduction of 75%; p <0.001).

Conclusion

We demonstrate that oral pre-exposure to F-DMC-18 resulted in reduced weight loss, systemic and local inflammation, and toxin levels in mice exposed to CD. These data suggest that DMCs developed in laboratory environments following freezing protocols, key in their long-term stability, hold great promise as microbial therapeutics against CDI.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):82–83.

SP47. Quantifying the effect on case numbers of public health intervention in Canada’s COVID-19 pandemic

Matthew T Crocker ¹, Paul Ioudovski ^1,^✉, Wendy Wobeser ^1,², Vernon Hoeppner ³

Objectives

Previous models showed that epidemics could be mitigated when public health interventions (PHIs) were applied early. One objective of PHIs was to preserve health care services. We used the Farr method, based on local reported numbers, to project cases and dates. The objective was to quantify the effect of PHIs and examine key observations from the coronavirus disease 2019 pandemic for future application.

Methods

We extracted reported cases from the Public Health Agency of Canada. We obtained PHI and ease dates from publicly accessible websites. We used the Farr method, which is based on percent change and rate of change of reported data, to project cases and dates. We used the model projections to compare cases without PHIs. The recording interval was 7 days.

Results

The PHIs prevented 95% or 8.6 million cases that were projected to occur without PHIs. The curve began to flatten in 5 weeks when PHIs were applied to 2,194 cases in Wave 1. A total of 9,899 additional cases might have been prevented if PHIs had been maintained at baseline. The curve began to flatten in 11 weeks when PHIs were applied to 12,000 cases in Wave 2. A total of 247,877 additional cases might have been prevented if PHIs had been applied to 3,456 cases in Wave 2.

Conclusion

PHIs were highly successful in mitigating the epidemic. An additional 258,000 cases might have been prevented by maintaining them in Wave 1 and using earlier application in Wave 2. The additional prevention was important in maintaining already stressed health care services.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):83.

SP48. Decreasing the incidence of external ventricular drain infections: Has standardizing insertion and management practices made a difference?

Anna Cvetkovic ^1,^✉, Anjali Shroff ¹, Cheryl Main ¹, Dominik Mertz ¹

Objectives

External ventricular drain (EVD)–related infections are associated with significant mortality. To address high infection rates, we created evidence-based, standardized protocols for EVD insertion and management. Changes to previous practice included pre-procedure hair clipping, use of aseptic technique in dedicated procedures rooms, and eliminating routine cerebrospinal fluid (CSF) sampling. Protocol development started in 2013 and was fully implemented by April 2016. The purpose of this study was to assess the impact of these standardized protocols on EVD-related infections.

Methods

Retrospective chart review of EVD-related infections at a tertiary care centre in Hamilton, Ontario. We compared infection rates 1 year pre-standardization of insertion and management techniques (April 2012–March 2013) with rates 1 year after full implementation (April 2016–July 2017). Two independent reviewers assessed all positive EVD cultures. Discrepant cases were adjudicated by a third reviewer. Cases were deemed infections if there were fevers, elevated CSF cell counts, consistent imaging, or treatment of EVD-related infections. The EVD insertion denominator was determined by counting dictated EVD procedure notes. Aggregate event rates between the two time periods were compared using a χ² test, and a time series using statistical process control charts was run (SPC; p-chart).

Results

A total of 75 and 54 positive EVD cultures were reviewed from April 2012 to March 2013 and April 2016 to July 2017, respectively. Of these, 16 (21%) and 12 (22%), respectively, were deemed infections, producing an infection rate of 16/141 EVD insertions (11.3%) before and 12/224 EVD insertions (5.3%) after implementation (odds ratio 0.44; 95% Cl 0.2 to 0.97; p = 0.036). No infections occurred from January to July 2017. The change in rates resulted in a special cause change in the SPC with 7 time points below the baseline average.

Conclusion

Full implementation of the EVD insertion and maintenance bundle was associated with a >50% reduction in EVD-related infections compared with the pre-implementation period.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):83–84.

SP49. Creation of a cumulative antibiogram for local intensive care units reveals significant resistance

D Brody Duncan ^1,^✉, Deborah Yamamura ^1,², Dominik Mertz ^1,², Neal Irfan ^1,²

Objectives

Bacterial infections in intensive care units (ICUs) have different epidemiology and resistance profiles compared with the rest of the hospital, and it has been recommended that unit-specific antibiograms in addition to a hospital cumulative antibiogram be available to guide empiric antibiotic choices. The objective was to create an ICU-specific cumulative antibiogram to determine local resistance patterns and communicate these to clinicians.

Methods

Microbiological data from 2019 were analyzed from three adult tertiary care hospital ICUs to create a cumulative antibiogram. The ICUs included medical and surgical patients. All clinical isolates were included. Clinical and Laboratory Standards Institute M39 guidelines were followed, and only the first isolate per species per patient was included.

Results

A total of 738 unique isolates were identified for inclusion. Between the three hospitals’ ICUs, rates of ESBL production ranged from 23.3 to 42.4% in Escherichia coli and 10.3% to 38.5% in Klebsiella pneumoniae. The frequency of multi-drug-resistant Pseudomonas aeruginosa was 9.7% to 19.2%, and no single beta-lactam agent had better than 85% coverage. Rates of cloxacillin resistance in Staphylococcus aureus (25.3%–37.5%) were not significantly different from the hospital-wide antibiograms (28%–31%). Less than 10% of Enterococcus were vancomycin resistant. A syndromic antibiogram including only respiratory isolates revealed that one hospital’s ICU had a significantly higher prevalence of chromosomal AmpC-producing Enterobacterales species compared with the other sites (37.8% versus 8.3%, p <0.01).

Conclusion

The creation of cumulative antibiograms quantified the rate of antimicrobial resistance in our local ICUs. High rates of beta-lactam resistance among Enterobacterales and variable beta-lactam resistance among P. aeruginosa suggest that patients with ICU-acquired sepsis, and especially septic shock, may benefit from empiric carbapenem or combination antibiotic therapy. Further study is ongoing on the implementation of these ICU antibiograms in collaboration with our antimicrobial stewardship program.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):84.

SP50. Evaluating correlates of viral shedding patterns in SARS-CoV-2 patients

Kyla M Tozer ^1,^2,^✉, Emily Moslinger ^1,³, Jasmine Lam ⁴, Amal Khalil ⁵, Paige Beddoe ⁵, Katya Douchant ^1,², Henry Wong ⁶, Prameet Sheth ^1,^2,^3,⁶

Objectives

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the newest member of the Coronavirinae family. After its identification in late 2019, the virus spread around the world, resulting in a global pandemic. Clinical severity of infection has been associated with higher viral loads; however, a significant proportion of people infected with SARS-CoV-2 remain asymptomatic and contribute to viral dissemination. Preliminary data suggest that both symptomatic and asymptomatic infected individuals shed SARS-CoV-2 virus after symptom resolution, although the duration of viral shedding and biological factors associated with shedding duration are not clearly understood. Here we investigate the correlates of SARS-CoV-2 viral shedding.

Methods

Naso-pharyngeal swabs were tested using a laboratory-developed dual-target (E-gene and 5’ UTR) polymerase chain reaction specific for SARS-CoV-2. Patient information (SARS-CoV-2 result, testing date, age, sex, and postal codes) were anonymized using a pseudo-script algorithm to protect patient privacy and analyzed using Microsoft Excel, PRISM, and IBM SPSS Statistics.

Results

A total of 88,420 specimens from 67,863 patients were analyzed. Cohort positivity rate was 0.5% (352/67,863), with gender data available for 270/352 (77%; 45% females and 31.7% males). Age stratification of all positive cases illustrated that pediatric cases accounted for 14/352 (4%), whereas those aged >65 years accounted for 77/352 (22%). Persistent shedding stratified according to patient sex; duration of viral shedding in females was significantly shorter than in males (mean 36 [range 10–151] days versus 40 [range 10–147] days, respectively; p = 0.0001).

Conclusion

Duration of viral shedding after infection varied and, in our cohort, appears to be associated with sex. Our cohort had higher infection rates among women than among men aged 19–64 years. These preliminary data findings need further validation and suggest that host biological variables may play a role in duration of SARS-CoV-2 shedding.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):84–85.

SP51. Fever and runny nose in a healthy adult patient: An interesting case of bacterial meningitis with normal CSF parameters

Catherine Guenther ^1,^✉, Wendy Sligl ², Geoffrey Shumilak ¹

Objectives

Describe a case of Streptococcus pneumoniae meningitis with absence of cerebrospinal fluid (CSF) abnormalities in a healthy adult patient.

Methods

Case report.

Results

We present the case of a healthy 25-year-old woman who presented to a hospital emergency department with a 48-hour history of fever, non-productive cough, and a headache. Aside from a temperature of 38.2°C and a heart rate of 105, she had a benign full physical exam. Lab work was significant only for mild leukocytosis. A lumbar puncture was done and revealed completely normal CSF parameters and a negative Gram stain. The patient was subsequently discharged with symptomatic management recommendations for a viral upper respiratory tract infection.

The patient re-presented to the emergency department 24 hours later after being found unresponsive in her home. Upon arrival, the patient was intubated for decreased GCS. A computed tomography scan revealed severe cerebral edema, leptomeningeal enhancement, and early tonsillar herniation. At the time of her re-presentation, the CSF culture from her prior lumbar puncture returned positive for organisms resembling S. pneumoniae. After emergent placement of an external ventricular drain, repeat CSF analysis again revealed normal parameters, but a Gram stain showed abundant gram-positive diplococci again suggestive of S. pneumoniae.

Aggressive medical management of bacterial meningitis was continued in the intensive care unit. Preliminary investigations for immunosuppression were unremarkable, including a negative HIV test, normal immunoglobulin levels, and normal T cell subsets. Extensive testing was not done because, unfortunately, the patient met clinical criteria for brain death within 12 hours of hospitalization.

Conclusion

This case highlights an adverse patient outcome despite appropriate clinical management based on published recommendations and evidence. Bacterial meningitis in the absence of any CSF abnormalities is a very rarely described entity with scarce case reports in medical literature. This case challenges a dogmatic approach to the interpretation of investigations in patients with suspected bacterial meningitis.

Off J Assoc Med Microbiol Infect Dis Can. 2021 Jul 21;6(Suppl):85.

SP52. Acceptability of self-collected samples for COVID-19 diagnosis in outpatients

Emma Finlayson-Trick ^1,^✉, Peter Tilley ^1,², Ghada N Al-Rawahi ^1,², Jocelyn A Srigley ^1,², Jeffrey M Pernica ³, Linda MN Hoang ^1,⁴, David M Goldfarb ^1,²

Objectives

In the context of rapid scale-up of molecular testing for severe acute respiratory syndrome coronavirus 2, one of the key weaknesses in the diagnostic cycle has been the challenge of sample acquisition, which has led to the implementation of self-collected sampling. We sought to compare the user acceptability of multiple self-collected sample types with the reference standard of health care worker (HCW)–collected naso-pharyngeal swabs (NPS).

Methods

Outpatients who had recently been diagnosed with coronavirus disease 2019 or who were identified as symptomatic household contacts of confirmed cases were approached for participation. Various self-collected samples and a HCW-collected NP flocked swab were obtained, and participants were asked about sample acceptability, using a 5-point Likert scale ranging from 1 (least acceptable) to 5 (most acceptable). Mean acceptability measures of all sample types were compared using repeated-measures analysis of variance, with post hoc significant pairwise differences determined using Tukey’s Honestly Significant Difference test. Statistical significance was set at p <0.05.

Results

Of the 43 participants who provided samples, 40 (aged 8–71 years) provided responses for the acceptability rating scale; 55% were female. Each participant provided a mean of 4.7 different sample types (range 3–6). Mouth rinse–gargle samples had the highest acceptability score (mean 4.9, SD 0.3), followed by saliva samples (mean 4.4, SD 1.1), throat samples (mean 4.1, SD 0.93), nose–throat samples (mean 4.0, SD 1.4), midturbinate samples (mean 3.4, SD 1.3), and finally HCW-collected NPS (mean 3.1, SD 1.3). The following self-collected sample types had significantly higher scores than HCW-collected NPS: mouth rinse–gargle (mean difference [MD] 1.8), saliva samples (MD 1.3), and combined anterior nares–throat swabs (MD 1.2).

Conclusion

There is significant variability in user-reported acceptability of self-collected sample types. These data should be considered when implementing self-collected sample testing programs.

PERMALINK

2020 Annual Conference Conférence Annuelle

A01. Incidence and timing of prosthetic joint infections after primary and revision arthroplasty: A population-based retrospective cohort study

Christopher Kandel

Nick Daneman

Jessica Widdifield

Bettina E Hansen

Richard Jenkinson

Allison McGeer

Objectives

Methods

Results

Conclusion

A02. Clinical characteristics, management, and outcomes of patients with indeterminate Clostridioides difficile tests using a modified two-step algorithm

Laura E Burnes

Stephanie W Smith

Kinga Kowalewska-Grochowska

Justin Z Chen

Objectives

Methods

Results

Conclusion

A03. Urinary tract infection investigation and treatment in older adults presenting to the emergency department with confusion: A health record review of local practice patterns

Rhiannan AM Pinnell

Tim Ramsay

Han X Wong

Pil Joo

Objectives

Methods

Results

Conclusion

A04. Impact of Cobas® PCR media freezing on SARS-CoV-2 viral RNA integrity and whole genome sequencing analyses

Patrick Benoit

Floriane Point

Simon Gagnon

Daniel Kaufmann

Cécile Tremblay

Richard Paul Harrigan

Isabelle Hardy

François Coutlée

Simon Grandjean-Lapierre

Objectives

Methods

Results

Conclusion

B01. Verification of the new bioMérieux piperacillin–tazobactam ETEST® against reference broth microdilution

Philippe Lagacé-Wiens

Heather Adam

Christine Turenne

Markus Stein

Andrew Walkty

James Karlowsky

Objectives

Methods

Results

Conclusion

B02. Performance of matrix-assisted laser desorption/ionization time of flight mass spectrometry for identification of filamentous moulds using several different extraction methods

Lisa Li

Elizabeth Bryce

Billie Velapatino

Ramin Najafi

Leane Kishi

Bing Wang

Claire Swanston

Linda Hoang

Vincent Tang

Marthe Charles

Objectives

Methods

Results

Conclusion

B03. Evaluation of isavuconazole gradient strips for susceptibility testing of Mucorales isolates

Lillian Feng

LeeAnn Turnbull

Brad Jansen

Tanis C Dingle

Objectives

Methods

Results

Conclusion