Abstract
Objective:
We compared the rates of hospital-onset secondary bacterial infections in patients with coronavirus disease 2019 (COVID-19) with rates in patients with influenza and controls, and we investigated reports of increased incidence of Enterococcus infections in patients with COVID-19.
Design:
Retrospective cohort study.
Setting:
An academic quaternary-care hospital in San Francisco, California.
Patients:
Patients admitted between October 1, 2019, and October 1, 2020, with a positive SARS-CoV-2 PCR (N = 314) or influenza PCR (N = 82) within 2 weeks of admission were compared with inpatients without positive SARS-CoV-2 or influenza tests during the study period (N = 14,332).
Methods:
National Healthcare Safety Network definitions were used to identify infection-related ventilator-associated complications (IVACs), probable ventilator-associated pneumonia (PVAP), bloodstream infections (BSIs), and catheter-associated urinary tract infections (CAUTIs). A multiple logistic regression model was used to control for likely confounders.
Results:
COVID-19 patients had significantly higher rates of IVAC and PVAP compared to controls, with adjusted odds ratios of 4.7 (95% confidence interval [CI], 1.7–13.9) and 10.4 (95 % CI, 2.1–52.1), respectively. COVID-19 patients had higher incidence of BSI due to Enterococcus but not BSI generally, and whole-genome sequencing of Enterococcus isolates demonstrated that nosocomial transmission did not explain the increased rate. Subanalyses of patients admitted to the intensive care unit and patients who required mechanical ventilation revealed similar findings.
Conclusions:
COVID-19 is associated with an increased risk of IVAC, PVAP, and Enterococcus BSI compared with hospitalized controls, which is not fully explained by factors such as immunosuppressive treatments and duration of mechanical ventilation. The mechanism underlying increased rates of Enterococcus BSI in COVID-19 patients requires further investigation.
Secondary bacterial infections contribute to excess morbidity and mortality in patients with influenza and other viral lower respiratory tract infections.1 At the time of hospital admission, bacterial coinfection in patients with coronavirus disease 2019 (COVID-19) is uncommon, with most studies reporting rates of 3%–8%.2–7 In contrast, hospital-onset secondary bacterial infections in COVID-19 patients appear to be a more significant problem, with incidence estimates as high as 87% for ventilator-associated pneumonia (VAP).3,4,6,8,9 Given reported associations between secondary bacterial infections and adverse outcomes including increased mortality, further investigation of incidence in COVID-19 patients is needed.3,10
Studies of secondary infections performed to date have used varying VAP definitions, making comparisons of rates, and interpretation of results, challenging. Studies of bloodstream infection (BSI) in COVID-19 patients have reported incidences as high as 40%–68%,11,12 but few studies have included a control group, and those that did have variously reported that rates in COVID-19 patients are lower13 or higher14 than in controls, or the same as in influenza patients.5 Moreover, many have not adjusted for potential confounders including baseline immunosuppression, receipt of immunosuppressive treatments, and duration of mechanical ventilation, leaving uncertainty about the relative contribution of these factors to secondary infection risk in COVID-19 patients.
Several surveillance studies have reported elevated rates of Enterococcus BSI in COVID-19 patients, but it remains unclear why Enterococcus spp are often among the most frequent BSI pathogens identified.12,14–16 In the 2 prior reports to address this question, nosocomial transmission was either suspected11 or proven.17 Therefore, whether risk of Enterococcus infection is truly elevated in COVID-19 or simply the result of regional infection control practices remains unclear. Furthermore, most secondary bacterial infection surveillance studies in COVID-19 patients have been performed outside of North America, where treatment protocols and antibiotic prescribing patterns may differ, resulting in unclear applicability to settings within the United States.
To address these gaps and advance understanding of hospital-onset secondary bacterial infection risk among patients with COVID-19, we used standardized US Centers for Disease Control and National Healthcare Safety Network (NHSN) definitions to evaluate the incidence of infection-related ventilator-associated complications (IVACs), possible ventilator-associated pneumonia (PVAP), BSIs, BSIs with Enterococcus, and catheter-associated urinary tract infections (CAUTIs) in patients hospitalized for COVID-19, compared to patients with influenza or to a hospitalized control group.
Methods
We performed a retrospective cohort study of adults admitted to the University of California, San Francisco Medical Center (UCSF) between October 1, 2019, and October 1, 2020, by evaluating hospital electronic health records under institutional review board protocol 17-24056 (Table 1). COVID-19 and influenza diagnoses were identified based on SARS-CoV-2 or influenza virus RT-PCR positivity within 2 weeks before or after the day of hospital admission. Controls included patients who were admitted to UCSF Medical Center and who did not have a positive severe acute respiratory coronavirus virus 2 (SARS-CoV-2) or influenza test during the study period. Patients with length of stay <2 days were excluded. Notably, it became institutional practice to test all patients for COVID-19 upon admission in April 2020.
Table 1A.
Baseline Characteristics of Full Study Cohort
| Variable | COVID-19 (n = 314) | Influenza (n = 82) | Control (n = 14,332) | P Valuea |
|---|---|---|---|---|
| Age, mean y (SD) | 57.9 (18.6) | 58.6 (21.4) | 55.8 (18.6) | .12 |
| Male, no. (%) | 176 (56.0) | 39 (47.6) | 6,479 (45.2) | <.01 |
| Race, no. (%) | ||||
| White | 100 (31.8) | 33 (40.2) | 7,760 (54.1) | <.01 |
| Black | 23 (7.3) | 10 (12.2) | 1,453 (10.1) | .21 |
| Asian | 57 (18.1) | 24 (29.3) | 2,587 (18.0) | .03 |
| Other | 127 (40.4) | 16 (19.5) | 2,752 (19.2) | <.01 |
| Unknown | 14 (4.5) | 2 (2.4) | 184 (1.3) | <.01 |
| Ethnicity, no. (%) | ||||
| Hispanic/Latino | 120 (38.2) | 17 (20.7) | 2,305 (16.1) | <.01 |
| Immunocompromised, no (%) | 100 (31.8) | 30 (36.6) | 5,129 (35.8) | .35 |
| Immunosuppressive medications in hospital, no. (%) | 144 (45.9) | 37 (45.1) | 5,963 (41.6) | .26 |
| ANC<500, no. (%) | 4 (1.3) | 4 (4.9) | 502 (3.5) | .08 |
| ALC<500, no. (%) | 71 (22.6) | 36 (43.9) | 2,466 (17.2) | <.01 |
| LOS, mean d (SD) | 13.6 (14.1) | 8.4 (9.9) | 6.9 (9.7) | <.01 |
| ICU admission, no. (%) | 126 (40.1) | 26 (31.7) | 2,702 (18.8) | <.01 |
| Central-line days, mean d (SD) | 6.8 (19.1) | 5.3 (17.4) | 2.7 (9.2) | .02 |
| Vent days, mean d (SD) | 3.7 (9.5) | 2.0 (5.8) | 0.3 (2.7) | <.01 |
| Urinary catheter days, mean d (SD) | 4.8 (11.0) | 2.5 (6.2) | 1.4 (4.1) | .37 |
| Blood cultures, mean no. (SD) | 3.4 (5.0) | 3.1 (3.6) | 1.0 (2.4) | <.01 |
| Cultures per patient day, mean no. (SD) | 0.3 (0.2) | 0.4 (0.4) | 0.1 (0.3) | <.01 |
| Admit SOFA7 score, mean (SD) | 3.6 (2.9) | 4.1 (3.1) | 2.4 (2.3) | <.01 |
We extracted information on baseline patient characteristics, admissions, treatments, and outcomes directly from the electronic medical record. Sequential Organ Failure Assessment (SOFA) score,18 included as a measure of severity of illness, was defined as the maximum SOFA score within the first 24 hours of admission. Immunocompromising conditions were identified using a preselected list of ICD-10 codes (Supplementary Table S1 online), and immunosuppressive medications were defined as listed in Supplementary Table S2 (online). We evaluated the incidence of IVAC, PVAP (a subcategory of IVAC), and CAUTI using the NHSN definitions.19 We identified BSIs using the NHSN Bloodstream Event definitions,20 and we included both primary and secondary BSIs. Multiple logistic regression was employed to analyze differences in IVAC, PVAP, BSI, and CAUTI between groups. We performed subgroup analyses in patients who were admitted to the intensive care unit (ICU) and in patients who required mechanical ventilation.
Covariates were chosen a priori based on known predisposing factors for bacterial infections, and included age, sex, race or ethnicity, admission SOFA score, underlying immunocompromising conditions, presence of neutropenia and/or lymphopenia, and receipt of immunosuppressive medications while hospitalized. Length of stay, duration of mechanical ventilation, central-line days, and (for CAUTI) urinary catheter days were also included. The number of blood cultures collected during admission was included as a covariate to account for possible differences in sampling frequency between groups. Days of therapy with IV antibiotics was included to account for a possible impact of differences in empiric antibiotic coverage at admission. Baseline differences between groups were evaluated using the χ2 test for categorical variables or the Kruskal-Wallis test for continuous variables. Logistic regression modeling was performed with R software (R Foundation for Statistical Computing, Vienna, Austria).
Given that COVID-19 patients in our hospital were placed in cohorts on a small number of designated units, we further investigated whether BSIs with Enterococcus spp had a common nosocomial source by performing whole-genome sequencing (WGS) of blood isolates. Illumina WGS was carried out on a MiSeq instrument following previously described methods.21 Raw sequences were adapter trimmed and quality controlled with fastp version 0.20.0 software and were analyzed using the SNP Pipeline for Infectious Disease (SPID) software.22 SPID aligned samples against reference genome Enterococcus faecalis strain OG1RF using minimap2, followed by Samtools to perform an mpileup.21,23 Julia code was then run to call a consensus allele at each position, and the SNP instances were computed between every pair of samples. Phylogenetic analysis was performed using Randomized Axelerated Maximum Likelihood (RAxML).24 Phylogenetic trees were further visualized with ETE Python API.25
Results
In total, 14,728 admissions, including 314 for COVID-19, 82 for influenza, and 14,332 control patient admissions, were evaluated. Given falling rates of influenza hospitalizations during the pandemic, the influenza admissions were concentrated from October 2019 through March 2020. Clinical and demographic characteristics are summarized in Table 1. World Health Organization (WHO) COVID-19 Ordinal Scale for Clinical Improvement scores26 were available for 126 (40.1%) of the COVID-19 patients. The median score was 5 (range, 3–8), corresponding to need for noninvasive ventilation or high-flow oxygen. Rates of IVAC, PVAP, BSI, BSI with Enterococcus spp, and CAUTI are provided in Table 2. The median WHO ordinal scale in COVID-19 patients with secondary infection was 7, corresponding to need for mechanical ventilation and additional organ support, suggesting that secondary infections were associated with critical illness.
Table 2.
Rates of Infectious Outcomes in the Full Study Cohort and in the ICU Subgroup
| Group | Total | IVAC, No. (%) | PVAP, No. (%) | BSI, No. (%) | Enterococcus BSI, No. (%) | CAUTI, No. (%) |
|---|---|---|---|---|---|---|
| Full cohort | ||||||
| COVID-19 | 314 | 11 (3.5) | 4 (1.3) | 24 (7.6) | 8 (2.6) | 3 (1.0) |
| Influenza | 82 | 3 (3.7) | 1 (1.2) | 3 (3.7) | 0 (0.0) | 0 (0.0) |
| Control | 14,332 | 30 (0.2) | 5 (<0.03) | 430 (3.0) | 48 (0.3) | 43 (0.3) |
| ICU subgroup | ||||||
| COVID | 126 | 11 (8.7) | 4 (3.2) | 20 (15.9) | 8 (6.4) | 3 (2.4) |
| Flu | 26 | 3 (11.5) | 1 (3.8) | 3 (11.5) | 0 (0.0) | 0 (0.0) |
| Control | 2,702 | 30 (1.1) | 5 (0.2) | 160 (5.9) | 18 (0.7) | 25 (0.9) |
Note. ICU, intensive care unit; IVAC, infection-related ventilator-associated complications; PVAP, probable ventilator-associated pneumonia; BSI, bloodstream infection; CAUTI, catheter-associated urinary tract infection.
Compared with controls, the unadjusted odds ratios (OR) for COVID-19 patients were 17.3 (95% confidence interval [CI], 8.6–34.9) for IVAC, 37.0 (95% CI, 9.9–138.3.0) for PVAP, 2.7 (95% CI, 1.8–4.1) for BSI, 7.8 (95% CI, 3.6–16.6) for BSI with Enterococcus, and 4.4 (95% CI, 1.4–14.6) for CAUTI. Compared with influenza, unadjusted ORs for COVID-19 patients were1.0 (95% CI, 0.3–3.5) for IVAC, 1.1 (95% CI, 0.1–9.5) for PVAP, and 2.2 (95% CI, 0.6–7.4) for BSI. Odds ratios could not be calculated for Enterococcus BSI or CAUTI because there were no events in the influenza group.
Adjusted ORs based on logistic regression incorporating multiple covariates for COVID-19 patients versus controls (covariates listed in Table 1) remained significantly increased: 4.7 (95% CI, 1.7–13.2) for IVAC, 10.4 (95% CI, 2.1–52.1) for PVAP, and 3.8 (95% CI, 1.5–9.4) for Enterococcus BSI (Table 3A). The adjusted ORs were 1.0 (95% CI, 0.6–1.8) for BSI and 0.9 (95% CI, 0.2–3.9) for CAUTI. The adjusted ORs for COVID-19 compared with influenza were not significant for any infectious outcome (Table 3B).
Table 3.
Adjusted Odds Ratios (95% CI) for Secondary Bacterial Infections in (3a) COVID-19 versus controls and in (3b) COVID-19 Versus Influenza
Table 3a. COVID-19 Versus Controls
| Independent Variable | IVAC, aOR (95% CI) | PVAP,aOR (95% CI) | BSI,aOR (95% CI) | Enterococcus BSI,aOR (95% CI) | CAUTI,aOR (95% CI)a |
|---|---|---|---|---|---|
| COVID-19 | 4.73 (1.70–13.86)b | 10.40 (2.08–52.09) | 1.01 (0.56–1.81) | 3.75 (1.49–9.41) | 0.85 (0.18–3.94) |
| Age, per year | 1.00 (0.98–1.02) | 1.00 (0.96–1.05) | 1.02 (1.01–1.03) | 1.01 (0.99–1.03) | 1.01 (0.99–1.03) |
| Sex, male | 1.10 (0.51–2.36) | 1.23 (0.27–5.62) | 1.33 (1.07–1.67) | 1.01 (0.56–1.81) | 0.28 (0.14–0.59) |
| Race or ethnicity | |||||
| Black | 1.50 (0.45–4.97) | 2.25 (0.20–24.88) | 0.96 (0.65–1.43) | 1.44 (0.57–3.59) | 1.57 (0.52–4.71) |
| Asian | 1.53 (0.62–3.79) | 1.81 (0.28–11.61) | 0.86 (0.65–1.14) | 0.73 (0.31–1.73) | 1.31 (0.57–2.98) |
| Hispanic/Latino | 0.66 (0.23–1.92) | 1.30 (0.21–8.29) | 0.87 (0.63–1.21) | 1.23 (0.58–2.60) | 1.52 (0.66–3.47) |
| Baseline IC | 0.54 (0.23–1.92) | 0.14 (0.01–1.42) | 0.92 (0.73–1.17) | 1.30 (0.71–2.41) | 1.90 (0.98–3.69) |
| IS meds | 1.27 (0.57–2.81) | 1.36 (0.31–5.99) | 0.65 (0.50–0.83) | 0.65 (0.34–1.22) | 3.45 (1.53–7.79) |
| ALC<500 | 1.26 (0.52–3.02) | 1.97 (0.41–9.53) | 2.01 (1.56–2.58) | 1.61 (0.81–3.18) | 0.90 (0.41–1.99) |
| LOS, per day | 1.00 (0.98–1.04) | 1.01 (0.95–1.07) | 0.99 (0.97–1.00) | 1.01 (0.99–1.03) | 1.03 (1.01–1.04) |
| Central-line days, per day | 0.98 (0.95–1.00) | 0.98 (0.94–1.03) | 0.97 (0.95–0.98) | 0.95 (0.93–0.97) | 0.97 (0.95–0.99) |
| Ventilator days, per day | 1.09 (1.05–1.13) | 1.04 (0.97–1.12) | 0.91 (0.88–0.93) | 0.98 (0.94–1.02) | 1.01 (0.97–1.06) |
| No. of blood cultures | 1.04 (0.96–1.12) | 1.03 (0.87–1.22) | 1.58 (1.51–1.66) | 1.27 (1.18–1.37) | 1.06 (0.98–1.16) |
| Antibiotic days, per day | 1.03 (1.01–1.04) | 1.01 (0.98–1.05) | 1.02 (1.01–1.03) | 1.02 (1.00–1.03) | 1.00 (0.98–1.03) |
| SOFA | 1.19 (1.08–1.32) | 1.33 (1.11–1.60) | 1.07 (1.03–1.11) | 1.09 (1.00–1.20) | 1.00 (0.90–1.11) |
Note. CI, confidence interval; IVAC, infection-related ventilator-associated complications; PVAP, probable ventilator-associated pneumonia; BSI, bloodstream infection; CAUTI, catheter-associated urinary tract infection; IC, immunocompromised; IS meds, immunosuppressive medications; ALC<500, absolute lymphocyte count <500 cells/μL at least once during encounter; LOS, length of stay; SOFA, Sequential Organ Failure Assessment score.
Urinary catheter days were included as a covariate for the CAUTI outcome only; aOR, 1.06 (1.03–1.09).
Covariates significant in the model are in bold.
Table 3b.
COVID-19 Versus Influenza
| Independent Variable | IVAC,aOR (95% CI) | PVAP,aOR (95% CI) | BSI,aOR (95% CI) | Enterococcus BSI,aOR (95% CI)a | CAUTI,aOR (95% CI)a |
|---|---|---|---|---|---|
| COVID-19 positive | 0.37 (0.05–2.65) | 1.06 (0.07–16.64) | 2.04 (0.41–10.04) | … | … |
| Age, per year | 0.97 (0.91–1.03) | 0.98 (0.91–1.05) | 1.05 (1.01–1.09)b | … | … |
| Male sex | 4.59 (0.61–34.68) | 4.16 (0.25–70.14) | 1.24 (0.41–3.80) | … | … |
| Race or ethnicity | |||||
| Black | … | … | 0.92 (0.16–5.35) | … | … |
| Asian | 0.84 (0.06–10.93) | 4.68 (0.16–139.55) | 0.29 (0.06–1.42) | … | … |
| Hispanic/Latino | 1.09 (0.17–7.13) | 3.65 (0.24–56.35) | 0.47 (0.12–1.85) | … | … |
| Baseline IC | 1.11 (0.18–6.96) | 0.80 (0.04–14.82) | 0.61 (0.18–2.07) | … | … |
| IS meds | 2.98 (0.48–18.69) | 0.24 (0.02–2.58) | 1.20 (0.38–3.75) | … | … |
| ALC<500 | 1.28 (0.19–8.77) | 4.98 (0.32–79.58) | 0.66 (0.19–2.25) | … | … |
| LOS, per day | 0.97 (0.88–1.08) | 0.94 (0.78–1.14) | 1.02 (0.94–1.11) | … | … |
| Central-line days, per day | 0.99 (0.93–1.05) | 0.95 (0.86–1.05) | 0.96 (0.92–1.00) | … | … |
| Ventilator days, per day | 1.18 (1.06–1.32) | 1.05 (0.88–1.24) | 0.85 (0.78–0.94) | … | … |
| No. of blood cultures | 0.87 (0.71–1.07) | 1.14 (0.81–1.63) | 1.55 (1.26–1.91) | … | … |
| Antibiotic days, per day | 1.06 (0.99–1.13) | 1.05 (0.94–1.17) | 1.05 (0.99–1.11) | … | … |
| SOFA | 1.05 (0.84–1.31) | 1.34 (0.96–1.87) | 1.19 (1.01–1.41) | … | … |
Note. CI, confidence interval; IVAC, infection-related ventilator-associated complications; PVAP, probable ventilator-associated pneumonia; BSI, bloodstream infection; CAUTI, catheter-associated urinary tract infection; IC, immunocompromised; IS meds, immunosuppressive medications; ALC<500, absolute lymphocyte count <500 cells/μL at least once during encounter; LOS, length of stay; SOFA, Sequential Organ Failure Assessment score.
Unable to evaluate given no events in the influenza group.
Covariates significant in the model are in bold.
The ICU subgroup analysis included 126 COVID-19 patients, 26 influenza patients, and 2,702 control patients; characteristics are shown in Table 1B and outcomes are shown in Table 2. Logistic regression analysis revealed similar findings to those in the entire cohort (Table 4).
Table 1B.
Baseline Characteristics of the ICU Cohort
| Variable | COVID-19 (n = 126) | Influenza (n = 26) | Control (n = 2,702) | P Valuea |
|---|---|---|---|---|
| Age, mean y (SD) | 58.1 (17.9) | 59.8 (18.8) | 60.0 (16.4) | .22 |
| Male, no. (%) | 85 (67.5) | 12 (46.1) | 1,493 (55.3) | .02 |
| Race, no. (%) | ||||
| White | 30 (23.8) | 10 (38.5) | 1,528 (56.5) | <.01 |
| Black | 6 (4.8) | 5 (19.2) | 233 (8.6) | .05 |
| Asian | 26 (20.6) | 7 (26.9) | 455 (16.8) | .22 |
| Other | 59 (46.8) | 3 (11.5) | 505 (18.7) | <.01 |
| Unknown | 8 (6.4) | 2 (7.7) | 36 (1.33) | <.01 |
| Ethnicity, no. (%) | ||||
| Hispanic/Latino | 50 (39.7) | 3 (11.5) | 416 (15.4) | <.01 |
| Immunocompromised, no. (%) | 29 (23.0) | 5 (19.2) | 908 (33.6) | .02 |
| Immunosuppressive medications in hospital, no. (%) | 86 (68.2) | 10 (38.5) | 2,702 (55.1) | <.01 |
| ANC<500, no. (%) | 0 (0.0) | 1 (3.8) | 83 (3.1) | .13 |
| ALC<500, no. (%) | 45 (35.7) | 12 (46.2) | 617 (22.8) | <.01 |
| LOS, mean d (SD) | 22.1 (15.5) | 14.7 (14.0) | 12.2 (13.7) | <.01 |
| Central-line days, mean d (SD) | 13.6 (19.9) | 14.8 (28.7) | 7.3 (16.8) | <.01 |
| Vent days, mean d (SD) | 9.3 (13.2) | 6.3 (9.1) | 1.7 (6.0) | <.01 |
| Foley days, mean d (SD) | 10.8 (13.8) | 7.9 (9.1) | 4.1 (7.7) | <.01 |
| Blood cultures, mean no. (SD) | 6.4 (6.5) | 5.7 (5.1) | 2.2 (3.9) | <.01 |
| Cultures per patient day, mean d (SD) | 0.3 (0.2) | 0.4 (0.3) | 0.2 (0.3) | <.01 |
| Admit SOFA score, mean (SD) | 5.2 (3.4) | 6.3 (3.6) | 4.0 (3.4) | <.01 |
Note. ICU, intensive care unit; ANC<500, absolute neutrophil count <500 cells/μL at least once during encounter; ALC<500, absolute lymphocyte count <500 cells/μL at least once during encounter; LOS, length of stay; SD, standard deviation; SOFA, Sequential Organ Failure Assessment.
Significant P values in bold.
Table 4.
ICU Subgroup Analysis: Adjusted Odds Ratios (95% CI) for Secondary Bacterial Infections in COVID-19 Versus Controls (a) and in COVID-19 Versus Influenza (b).
Table 4a. COVID-19 Versus Controls (ICU subgroup)
| Independent Variable | IVAC,aOR (95% CI) | PVAP,aOR (95% CI) | BSI,aOR (95% CI) | Enterococcus BSI,aOR (95% CI) | CAUTI,aOR (95% CI)a |
|---|---|---|---|---|---|
| COVID-19 | 4.27 (1.61–11.31)b | 8.04 (1.59–40.64) | 1.22 (0.60–2.46) | 7.91 (2.53–24.76) | 0.48 (0.09–2.66) |
| Age, per year | 1.00 (0.97–1.02) | 1.00 (0.96–1.05) | 1.02 (1.01–1.03) | 1.01 (0.98–1.04) | 1.01 (0.99–1.04) |
| Male sex | 0.91 (0.43–1.90) | 1.03 (0.23–4.56) | 1.30 (0.90–1.89) | 0.91 (0.36–2.29) | 0.26 (0.10–0.66) |
| Race Black | 1.83 (0.56–6.03) | 2.61 (0.25–27.64) | 1.18 (0.63–2.22) | 1.63 (0.33–8.12) | 1.37 (0.27–6.86) |
| Race Asian | 1.72 (0.70–4.21) | 1.75 (0.27–11.31) | 0.88 (0.55–1.41) | 0.45 (0.10–2.11) | 0.74 (0.18–2.93) |
| Hispanic/Latino | 0.92 (0.33–2.55) | 1.55 (0.24–9.94) | 0.96 (0.57–1.62) | 1.28 (0.42–3.94) | 3.53 (1.33–9.36) |
| Baseline IC | 0.60 (0.25–1.42) | 0.21 (0.02- 1.95) | 0.84 (0.56–1.24) | 0.87 (0.31–2.44) | 1.16 (0.48–2.78) |
| IS meds | 0.81 (0.37–1.78) | 0.97 (0.22–4.27) | 0.66 (0.45–0.97) | 0.44 (0.16–1.21) | 4.09 (1.20–13.99) |
| ALC<500 | 1.11 (0.48–2.58) | 1.79 (0.38–8.45) | 1.89 (1.28–2.79) | 1.82 (0.65–5.11) | 0.82 (0.29–2.35) |
| LOS, per day | 0.99 (0.95–1.03) | 0.99 (0.92–1.08) | 0.99 (0.98–1.01) | 1.02 (1.00–1.05) | 1.04 (1.01–1.06) |
| Central line days, per day | 0.98 (0.96–1.01) | 0.99 (0.94–1.04) | 0.97 (0.95–0.99) | 0.95 (0.92–0.98) | 0.98 (0.95–1.00) |
| Ventilator days, per day | 1.07 (1.04–1.11) | 1.04 (0.96–1.12) | 0.94 (0.91–0.97) | 0.96 (0.91–1.01) | 1.05 (1.00–1.10) |
| No. of blood cultures | 1.01 (0.94–1.09) | 1.02 (0.87–1.20) | 1.41 (1.33–1.50) | 1.20 (1.09–1.33) | 1.00 (0.90–1.10 |
| Antibiotic-days, per day | 1.04 (1.02–1.06) | 1.01 (0.97–1.06) | 1.02 (1.00–1.03) | 1.03 (1.01–1.06) | 0.98 (0.95–1.02) |
| SOFA | 1.06 (0.96–1.17) | 1.18 (0.98–1.43) | 1.06 (1.01–1.11) | 1.11 (0.98–1.26) | 0.99 (0.87–1.12) |
Note. ICU, intensive care unit; CI, confidence interval; IVAC, infection-related ventilator-associated complications; PVAP, probable ventilator-associated pneumonia; BSI, bloodstream infection; CAUTI, catheter-associated urinary tract infection; IC, immunocompromised; IS meds, immunosuppressive medications; ALC<500, absolute lymphocyte count <500 cells/μL at least once during encounter; LOS, length of stay; SOFA, Sequential Organ Failure Assessment score.
Urinary catheter days were included as a covariate for the CAUTI outcome only; aOR, 1.05 (1.01–1.08).
Covariates significant in the model are in bold.
Table 4b.
COVID-19 Versus Influenza (ICU Subgroup)
| Independent Variable | IVAC,aOR (95% CI) | PVAP,aOR (95% CI) | BSI,aOR (95% CI) | Enterococcus BSI,aOR (95% CI)a | CAUTI,aOR (95% CI)a |
|---|---|---|---|---|---|
| COVID-19 | 0.25 (0.03–2.19) | 0.67 (0.03–17.27) | 0.74 (0.13–4.15) | … | … |
| Age, per year | 0.97 (0.91–1.03) | 0.98 (0.91–1.06) | 1.06 (1.01–1.10)b | … | … |
| Sex, male | 5.43 (0.66–44.62) | 5.42 (0.21–139.57) | 1.89 (0.51–6.99) | … | … |
| Race or ethnicity | |||||
| Asian | 1.00 (0.08–12.93) | 4.42 (0.14–140.28) | 0.27 (0.05–1.41) | … | … |
| Hispanic/Latino | 1.42 (0.22–9.33) | 7.30 (0.31–174.41) | 0.60 (0.14–2.62) | … | … |
| Baseline IC | 1.60 (0.21–12.05) | 1.02 (0.04–29.04) | 0.36 (0.08–1.71) | … | … |
| IS meds | 2.60 (0.37–18.17) | 0.17 (0.01–2.08) | 1.03 (0.29–3.64) | … | … |
| ALC<500 | 0.99 (0.12–8.17) | 5.68 (0.24–137.05) | 1.12 (0.30–4.12) | … | … |
| LOS, per day | 0.96 (0.87–1.08) | 0.94 (0.77–1.14) | 1.02 (0.93–1.11) | … | … |
| Central-line days, per day | 1.00 (0.94–1.06) | 0.96 (0.86–1.07) | 0.97 (0.92–1.02) | … | … |
| Ventilator days, per day | 1.17 (1.05–1.31) | 1.06 (0.89–1.26) | 0.89 (0.81–0.98) | … | … |
| No. of blood cultures | 0.87 (0.71–1.05) | 1.09 (0.78–1.51) | 1.40 (1.16–1.70) | … | … |
| Antibiotic days, per day | 1.05 (0.98–1.11) | 1.03 (0.93–1.14) | 1.04 (0.98–1.10) | … | … |
| SOFA | 1.00 (0.79–1.26) | 1.19 (0.83–1.70) | 1.12 (0.94–1.34) | … | … |
Note. ICU, intensive care unit; CI, confidence interval; IVAC, infection-related ventilator-associated complications; PVAP, probable ventilator-associated pneumonia; BSI, bloodstream infection; CAUTI, catheter-associated urinary tract infection; IC, immunocompromised; IS meds, immunosuppressive medications; ALC<500, absolute lymphocyte count <500 cells/μL at least once during encounter; LOS, length of stay; SOFA, Sequential Organ Failure Assessment score.
Unable to evaluate given no events in the influenza group.
Covariates significant in the model are in bold.
The analyses were repeated on the subgroup of patients who required mechanical ventilation, and findings were again similar (characteristics, outcome rates, and logistic regression analyses in Supplementary Tables S3–S5 online).
The 8 Enterococcus infections in the COVID-19 group comprised 6 vancomycin-susceptible Enterococcus faecalis, 1 vancomycin-susceptible Enterococcus faecium, and 1 vancomycin-resistant Enterococcus faecium. The patients had a mean age of 60.6±15.8, similar to the COVID-19 patients as a whole. Notable comorbidities included diabetes mellitus (50%) and history of renal transplant (25%). All 8 patients were in the ICU with central lines in place at the time of bacteremia: 1 patient was on extracorporeal membrane oxygenation. Also, 4 patients received immunosuppressive medications in the hospital, and 6 patients had received broad-spectrum antibiotics prior to BSI onset. In 2 patients, a central line was suspected as the source; in 1 of these patients, the source was a lower-extremity abscess; and in the other 5 cases, the source of bacteremia was unclear. Enterococcus BSIs occurred, on average, 13.1 days (SD, ±15.1) into the hospital stays. The WGS of Enterococcus isolates from the primary COVID-19 ICU revealed that isolates were genetically distinct, thus ruling out a nosocomial outbreak (Supplementary Fig. S1 online).
We further characterized the microbiology of the bloodstream infections in COVID-19 and control patients (Fig. 1). The 3 bloodstream infections in the influenza patients were Staphylococcus aureus, group A streptococcus, and viridans group Streptococcus. In the COVID-19 group, the 5 PVAPs included 5 distinct organisms: 2 methicillin-susceptible Staphylococcus aureus and 1 each of methicillin-resistant Staphylococcus aureus, Pseudomonas fluorescans, Haemophilus influenzae, and Burkholderia gladioli.
Fig. 1.
Organisms isolated from blood culture in patients with COVID-19 patients versus controls.
Discussion
We report markedly increased rates of IVAC and PVAP in the setting of SARS-CoV-2 infection compared to controls, even after adjusting for potential confounders. Our findings are concordant with 3 recent reports of increased VAP in mechanically ventilated COVID-19 patients compared to controls.27–29 Rates of IVAC and PVAP did not differ significantly between COVID-19 and influenza admissions, emphasizing that although influenza patients have a higher rate of bacterial coinfection at the time of admission,2 there may be a similarly high risk of secondary bacterial pulmonary infections in both diseases. In patients with influenza, animal models have suggested that increased susceptibility to bacterial infections is due in part to viral induction of type I interferon signaling resulting in suppressed antibacterial immune responses.30 Whether SARS-CoV-2, which has also been demonstrated to cause significant immune dysregulation,31 leads to increased VAP via a similar mechanism will require further study.
The incidence of IVAC and PVAP that we report in our critically ill COVID-19 cohort, 8.7% and 3.2%, respectively, are markedly lower than the 38.5%–62% incidence previously reported in the literature.2,3,27,28,32,33 We suspect that this difference is related to the use of strict NHSN definitions rather than clinical pneumonia definitions, combined with our study setting in a center where resources, including both personnel and personal protective equipment, were readily available. However, another small study that also used NHSN surveillance definitions reported a PVAP rate of 54% in patients with COVID-19.34 Given the significant overlap between the clinical presentations of severe COVID-19 pneumonia and bacterial VAP, further investigation into the true rates of VAP in these patients is needed.
The BSI rates in our COVID-19 cohort were 7.6% overall and 15.9% in the ICU subgroup, with no significant difference in the multivariable model compared to either controls or influenza patients. These results are in line with published reports of BSI rates in COVID-19 patients, which have ranged from 3% to 68%3,5,11–16 depending on the cohort studied. We also found no difference between groups in rates of CAUTI, concordant with other published data.35
Enterococcus BSI was markedly higher in COVID-19 patients compared to controls and was the most common organism in these patients, a finding in line with 2 prior studies that reported Enterococcus to be the most frequently identified BSI organism isolated from COVID-19 patients.15,16 A third report also identified higher-than-expected incidence of Enterococcus BSI in COVID-19 patients compared to controls, although in that study the patients had an epidemiologic link and the possibility that the increased incidence was due to a nosocomial outbreak could not be excluded.11 We, in contrast, found that Enterococcal BSI events had no clear epidemiologic association, and further WGS analysis demonstrated genetically distinct isolates, ruling out a common nosocomial source. Whether our finding is the result of SARS-CoV-2 enterocyte tropism36 or systemic inflammatory responses leading to gut translocation, or the result of other factors, needs further investigation.
Our study has several limitations. The study groups had baseline differences in severity of illness, duration of mechanical ventilation, and other factors. Although we attempted to control for these by including them as covariates in the regression analysis, there may have been unmeasured differences between the groups that remained unaccounted for, and this may have affected the magnitude of the effects we observed. Additionally, we used NHSN surveillance definitions for IVAC, PVAP, BSI, and CAUTI, which may have missed some clinician-suspected infections. However, given that these definitions are the standard used across the United States, they enable a consistent comparison across healthcare centers; furthermore, many studies have highlighted the low reliability of provider-based VAP diagnoses.37–39 The influenza and COVID-19 cohorts were hospitalized during different periods, a necessity given the low rates of influenza virus infection during the COVID-19 pandemic. Finally, this was a single-center study, conducted in a setting where personal protective equipment was sufficient and hospital infection prevention practices closely enforced, and findings may not be generalizable to other settings.
In summary, we report that COVID-19, similar to influenza virus infection, confers a significantly increased risk of hospital-onset secondary bacterial infections, a finding with important implications for infection prevention and clinical management during the ongoing COVID-19 pandemic.
Acknowledgments
We thank Lynn Ramirez for their infection prevention expertise and guidance, and Jennifer Babik for discussions about coinfections in patients with COVID-19.
Supplementary material
For supplementary material accompanying this paper visit http://dx.doi.org/10.1017/ice.2021.391.
click here to view supplementary material
Financial support
This work was supported by the National Heart, Lung, and Blood Institute (grant no. K23HL138461-01A1 to C.L.) and philanthropic contributions from Mark and Carrie Casey, Julia and Kevin Hartz, Carl Kawaja and Wendy Holcombe, Eric Keisman and Linda Nevin, Martin and Leesa Romo, Three Sisters Foundation, Diana Wagner, and Jerry Yang and Akiko Yamazaki.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
References
- 1.Shah NS, Greenberg JA, McNulty MC, et al. Bacterial and viral co-infections complicating severe influenza: incidence and impact among 507 US patients, 2013–14. J Clin Virol 2016;80:12–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Youngs J, Wyncoll D, Hopkins P, Arnold A, Ball J, Bicanic T.Improving antibiotic stewardship in COVID-19: bacterial coinfection is less common than with influenza. J Infect 2020;81(3):e55–e57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Soriano MC, Vaquero C, Ortiz-Fernández A, Caballero A, Blandino-Ortiz A, de Pablo R.Low incidence of coinfection, but high incidence of ICU-acquired infections in critically ill patients with COVID-19. J Infect 2020. doi: 10.1016/j.jinf.2020.09.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Garcia-Vidal C, Sanjuan G, Moreno-García E, et al. Incidence of coinfections and superinfections in hospitalized patients with COVID-19: a retrospective cohort study. Clin Microbiol Infect 2020. doi: 10.1016/j.cmi.2020.07.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hughes S, Troise O, Donaldson H, Mughal N, Moore LSP.Bacterial and fungal coinfection among hospitalized patients with COVID-19: a retrospective cohort study in a UK secondary-care setting. Clin Microbiol Infect 2020;26:1395–1399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Langford BJ, So M, Raybardhan S, et al. Bacterial coinfection and secondary infection in patients with COVID-19: a living rapid review and meta-analysis. Clin Microbiol Infect 2020. doi: 10.1016/j.cmi.2020.07.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Vaughn VM, Gandhi T, Petty LA, et al. Empiric antibacterial therapy and community-onset bacterial coinfection in patients hospitalized with COVID-19: a multihospital cohort study. Clin Infect Dis 2020. doi: 10.1093/cid/ciaa1239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Cai Q, Huang D, Ou P, et al. COVID-19 in a designated infectious diseases hospital outside Hubei Province, China. Allergy 2020;75:1742–1752. [DOI] [PubMed] [Google Scholar]
- 9.Schmidt M, Hajage D, Lebreton G, et al. Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome associated with COVID-19: a retrospective cohort study. Lancet Respir Med 2020;8:1121–1131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gamberini L, Tonetti T, Spadaro S, et al. Factors influencing liberation from mechanical ventilation in coronavirus disease 2019: multicenter observational study in fifteen Italian ICUs. J Intensive Care 2020;8:80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bonazzetti C, Morena V, Giacomelli A, et al. Unexpectedly high frequency of enterococcal bloodstream infections in coronavirus disease 2019 patients admitted to an Italian ICU: an observational study. Crit Care Med 2021;49(1):e31–e40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kokkoris S, Papachatzakis I, Gavrielatou E, et al. ICU-acquired bloodstream infections in critically ill patients with COVID-19. J Hosp Infect 2021;107:95–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sepulveda J, Westblade LF, Whittier S, et al. Bacteremia and blood culture utilization during COVID-19 surge in New York City. J Clin Microbiol 2020;58(8). doi: 10.1128/JCM.00875-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Engsbro AL, Israelsen SB, Pedersen M, et al. Predominance of hospital-acquired bloodstream infection in patients with COVID-19 pneumonia. Infect Dis (London) 2020;52:919–922. [DOI] [PubMed] [Google Scholar]
- 15.Giacobbe DR, Battaglini D, Ball L, et al. Bloodstream infections in critically ill patients with COVID-19. Eur J Clin Invest 2020;50(10):e13319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cataldo MA, Tetaj N, Selleri M, et al. Incidence of bacterial and fungal bloodstream infections in COVID-19 patients in intensive care: an alarming “collateral effect.” J Glob Antimicrob Resist 2020;23:290–291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kampmeier S, Tönnies H, Correa-Martinez CL, Mellmann A, Schwierzeck V.A nosocomial cluster of vancomycin resistant enterococci among COVID-19 patients in an intensive care unit. Antimicrob Resist Infect Control 2020;9:154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jones AE, Trzeciak S, Kline JA.The Sequential Organ Failure Assessment score for predicting outcome in patients with severe sepsis and evidence of hypoperfusion at the time of emergency department presentation. Crit Care Med 2009;37:1649–1654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.NHSN VAE. Centers for Disease Control and Prevention website. https://www.cdc.gov/nhsn/pdfs/pscmanual/10-vae_final.pdf. Published 2021. Accessed August 30, 2021.
- 20.NHSN bloodstream infection event. Centers for Disease Control and Prevention website. https://www.cdc.gov/nhsn/pdfs/pscmanual/4psc_clabscurrent.pdf. Published 2021. Accessed August 30, 2021.
- 21.Crawford E, Kamm J, Miller S, et al. Investigating transfusion-related sepsis using culture-independent metagenomic sequencing. Clin Infect Dis 2020;71:1179–1185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kamm, Jack. SNP pipeline for infectious disease. Github website. https://github.com/czbiohub/Spid.jl. Published 2021. Accessed August 30, 2021.
- 23.Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N.The sequence alignment/map format and SAMtools. Bioinforma (Oxford) 2009;25. doi: 10.1093/bioinformatics/btp352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Stamatakis A.RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinforma (Oxford) 2014;30:1312–1313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Huerta-Cepas J, Serra F, Bork P.ETE 3: reconstruction, analysis, and visualization of phylogenomic data. Mol Biol Evol 2016;33:1635–1638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.COVID-19 therapeutic trial synopsis. World Health Organization website. https://www.who.int/publications-detail-redirect/covid-19-therapeutic-trial-synopsis. Accessed August 8, 2021.
- 27.Maes M, Higginson E, Pereira-Dias J, et al. Ventilator-associated pneumonia in critically ill patients with COVID-19. Crit Care 2021;25. doi: 10.1186/s13054-021-03460-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Razazi K, Arrestier R, Haudebourg AF, et al. Risks of ventilator-associated pneumonia and invasive pulmonary aspergillosis in patients with viral acute respiratory distress syndrome related or not to coronavirus 19 disease. Crit Care (London) 2020;24:699. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Rouzé A, Martin-Loeches I, Povoa P, et al. Relationship between SARS-CoV-2 infection and the incidence of ventilator-associated lower respiratory tract infections: a European multicenter cohort study. Intensive Care Med 2021. doi: 10.1007/s00134-020-06323-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Shahangian A, Chow EK, Tian X, et al. Type I IFNs mediate development of postinfluenza bacterial pneumonia in mice. J Clin Invest 2009;119:1910–1920. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Sarma A, Christenson S, Mick E, et al. COVID-19 ARDS is characterized by a dysregulated host response that differs from cytokine storm and is modified by dexamethasone. Res Sq 2021. doi: 10.21203/rs.3.rs-141578/v1. [DOI] [Google Scholar]
- 32.Feng Y, Ling Y, Bai T, et al. COVID-19 with different severities: a multicenter study of clinical features. Am J Respir Crit Care Med 2020;201:1380–1388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Blonz G, Kouatchet A, Chudeau N, et al. Epidemiology and microbiology of ventilator-associated pneumonia in COVID-19 patients: a multicenter retrospective study in 188 patients in an un-inundated French region. Crit Care 2021;25. doi: 10.1186/s13054-021-03493-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Moretti M, Van Laethem J, Minini A, Pierard D, Malbrain MLNG.Ventilator-associated bacterial pneumonia in coronavirus 2019 disease, a retrospective monocentric cohort study. J Infect Chemother. 2021. doi: 10.1016/j.jiac.2021.01.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Fakih MG, Bufalino A, Sturm L, et al. Coronavirus disease 2019 (COVID-19) pandemic, central-line-associated bloodstream infection (CLABSI), and catheter-associated urinary tract infection (CAUTI): the urgent need to refocus on hardwiring prevention efforts. Infect Control Hosp Epidemiol 2021. doi: 10.1017/ice.2021.70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Lamers MM, Beumer J, van der Vaart J, et al. SARS-CoV-2 productively infects human gut enterocytes. Science 2020;369:50–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Klompas M.Does this patient have ventilator-associated pneumonia? JAMA 2007;297:1583–1593. [DOI] [PubMed] [Google Scholar]
- 38.Klompas M.Interobserver variability in ventilator-associated pneumonia surveillance. Am J Infect Control 2010;38:237–239. [DOI] [PubMed] [Google Scholar]
- 39.Klompas M, Kulldorff M, Platt R.Risk of misleading ventilator-associated pneumonia rates with use of standard clinical and microbiological criteria. Clin Infect Dis 2008;46:1443–1446. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
For supplementary material accompanying this paper visit http://dx.doi.org/10.1017/ice.2021.391.
click here to view supplementary material