Abstract
The objective of this study was to compare the aetiologic yield of standard of care microbiologic testing ordered by physicians with that of a multiplex PCR platform. Stool specimens obtained from children and young adults with gastrointestinal illness were evaluated by standard laboratory methods and a developmental version of the FilmArray Gastrointestinal Diagnostic System (FilmArray GI Panel), a rapid multiplex PCR platform that detects 23 bacterial, viral, and protozoal agents. Results were classified according to the microbiologic tests requested by the treating physician. A median of 3 (range 1-10) microbiologic tests were performed by the clinical laboratory during 378 unique diarrhoeal episodes. A potential aetiologic agent was identified in 46% of stool specimens by standard laboratory methods and in 65% of specimens tested using the FilmArray GI Panel (P<0.001). For those patients who only had Clostridium difficile testing requested, an alternative pathogen was identified in 29% of cases with the FilmArray GI Panel. Notably, 11 (12%) cases of norovirus were identified among children who only had testing for C. difficile ordered. Among those who had C. difficile testing ordered in combination with other tests, an additional pathogen was identified in 57% of stool specimens with the FilmArray GI Panel. For patients who had no C. difficile testing performed, the FilmArray GI Panel identified a pathogen in 63% of cases, including C. difficile in 8%. Physician-specified laboratory testing may miss important diarrhoeal pathogens. Additionally, standard laboratory testing is likely to underestimate co-infections with multiple infectious diarrhoeagenic agents.
Keywords: Clostridium difficile, gastrointestinal illness, acute gastroenteritis, FilmArray, children
BACKGROUND
Despite advances in sanitation, food safety, and immunization, diarrhoeal diseases continue to cause substantial disease and mortality in children living in both high and low resource settings.(1, 2) A 2011 report from the United States Centres for Disease Control and Prevention estimated that 178.8 million acute diarrhoeal illnesses occur annually in the U.S., resulting in 473,832 hospitalizations and 5,072 deaths.(3) The changing epidemiology and ongoing outbreaks of old and emerging pathogens including Clostridium difficile, Cryptosporidium spp., Cyclospora spp. and diarrhoeagenic Escherichia coli underscore the need for accurate diagnostics and improved surveillance for infectious diarrhoea.(4, 5)
Correctly diagnosing the aetiology of infectious diarrhoea can improve clinical care and public health surveillance.(6) One major obstacle in diagnosing infectious diarrhoea is the large and growing number of viruses, bacteria, and protozoa that are recognized to cause diarrhoea. Testing for specific pathogens is complex, requiring a variety of methods. Clinicians must choose the appropriate test but this is challenging because of substantial overlap in clinical and epidemiologic features. Testing therefore can be expensive and inefficient. For some important pathogens such as diarrhoeagenic Escherichia coli, norovirus and sapovirus, testing is not readily available. To address this, there has been interest in developing multiplex platforms that can simultaneously detect a range of diarrhoeal pathogens.(7-12)
The objective of this study was to compare the diagnostic yield of standard microbiologic testing ordered by the treating physician with the FilmArray Gastrointestinal Diagnostic System (FilmArray GI Panel, BioFire Diagnostics Inc., Salt Lake City, UT), a multiplex polymerase chain reaction (PCR) system that simultaneously detects 23 diarrhoeal pathogens. We hypothesized that physician-specified microbiologic testing would not accurately predict the pathogens present.
METHODS
Human Subjects Protection
This study was approved and granted a waiver of informed consent by the University of Utah and Intermountain Healthcare (Intermountain) Institutional Review Boards (IRB #45464).
Setting and Study Population
Stool samples were collected from children and young adults 1-25 years of age with symptoms of acute gastrointestinal illness (e.g. fever, vomiting, abdominal pain, and diarrhoea) who received medical care at Primary Children's Hospital (PCH). Stool specimens that conformed to the shape of the cup were submitted for standard laboratory testing at the request of the treating physician and residual samples were stored frozen at −80°C until tested with the FilmArray GI Panel. We collected specimens from August 2010 through December 2012.
Of 1504 diarrhoeal episodes, 378 episodes were selected for this study. This convenience sample was based on adequate residual specimen volume and was enriched for those who had multiple standard laboratory tests, including C. difficile testing, and for patients in whom a pathogen was detected. We designed a sampling scheme based on the number of standard laboratory tests ordered. We randomly selected 125 specimens that had one standard laboratory test ordered, 118 specimens that had two or three tests ordered, and 135 that had four or more tests ordered. A single stool specimen from each of the 378 episodes was tested with the FilmArray GI Panel. All microbiologic results from physician-ordered tests performed within ±72 hours of the collection of the specimen tested using the FilmArray GI Panel were included in the analysis. The selection of specimens for inclusion in this study was designed to evaluate the potential impact of a multiplex, PCR-based diagnostic platform in comparison to physician-specified standard laboratory testing and not to describe the epidemiology of diarrhoea.
Standard Laboratory Testing
Standard laboratory testing was performed at the discretion of the treating physician. Routine stool cultures identified Salmonella spp., Shigella spp., Aeromonas spp., E. coli O157H7 and O121, Bacillus cereus, and Campylobacter jejuni and C. coli using trypticase soy agar (TSA), MacConkey II agar, sorbitol MacConkey agar, Hektoen Enteric agar, and Campy CVA agar. When Yersinia enterocolitica testing was ordered, stool was plated on cefsulodin–irgasan–novobiocin (CIN) agar. C. difficile was detected using the illumigene C. difficile amplification assay (Meridian Bioscience, Inc., Cincinnati, OH). Rotavirus and adenovirus F 40/41 were detected using commercial immunoassays (Immunocard STAT!® Rotavirus and Meridian Premier™ Adenoclone®, Meridian Bioscience, Inc.). Shiga toxin-producing Escherichia coli (STEC) was sought for all bloody specimens and when requested using a rapid immunoassay for shigatoxin (Meridian Premier™ EHEC, Meridian Bioscience, Inc.) on specimens grown 24 hours in nutrient broth. Giardia lamblia and Cryptosporidium parvum were detected using the MERIFLUOR antigen detection immunoassay (Meridian Bioscience, Inc.). Other protozoa were identified by routine ova and parasite examination when requested. Laboratory testing for norovirus by PCR was introduced during the last six months of the study; however, norovirus testing was not ordered by the physician for any of the specimens evaluated in this study. No standard laboratory tests were available for the detection of astrovirus or sapovirus.
FilmArray GI Panel Pathogen Detection
The FilmArray rapid multiplex PCR platform(13, 14) used in this study was a development version of the FilmArray GI Panel, which simultaneously detects 23 diarrhoeagenic bacterial, viral, and protozoal agents in less than one hour (Table 1) using pathogen-specific virulence genes or gene signatures in housekeeping genes. Identification of C. difficile was based upon detection of the genes that encode an enterotoxin (tcdA) and a cytotoxin (tcdB). The pathotypes of pathogenic E. coli were identified using pathotype-specific genetic markers: STEC by detection of Shiga toxin 1 or 2 genes (stx1 or stx2), EPEC by detection of the intimin gene (eae), ETEC by genes encoding heat-labile (lt) or heat-stable (st) enterotoxin, EIEC and Shigella by the invasion plasmid antigen H gene (ipah), EAEC by pAA virulence plasmid carried genes encoding the AAF biogenesis transcription regulator (aggR) or outer membrane protein (aatA). The FilmArray GI Panel identifies Salmonella spp., Aeromonas spp., Cryptosporidium spp., and pathogenic species of Vibrio and Campylobacter (V. cholerae, V. parahaemoliticus, V. vulnificus, V. mimicus, V. alginolyticus, V. fluvialis, C. jejuni, C. coli, and C. upsaliensis). Extensive performance evaluation of the developmental version of the FilmArray GI Panel was conducted on 1721 clinical specimens and has been reported previously.(15) FilmArray testing was performed while blinded to standard testing results and patient characteristics.
Table 1.
Detection of bacterial, viral, and parasitic diarrhoeal pathogens from 378 pediatric stool specimens evaluated by standard laboratory methods and the FilmArray GI Panel.
| Organism | Standard Laboratory Methods No. Positive / No. Tested (%) | FilmArray GI Panel No. Positive / No. Tested (%) |
|---|---|---|
| Bacterial pathogens | ||
| Clostridium difficile | 77/273 (28%) | 83/378 (22%) |
| EPEC a | NA | 37/378 (10%) |
| All STEC b | 19/193 (10%) | 30/378 (8%) |
| E. coli O157 | 12/193 (7%) | 18/378 (5%) |
| Non-O157 | 7/193 (4%) | 12/378 (3%) |
| Salmonella spp. | 24/189 (13%) | 27/378 (7%) |
| Campylobacter spp. | 6/189 (3%) | 15/378 (4%) |
| Shigella / EIEC c | 5/189 (3%) | 11/378 (3%) |
| EAEC d | NA | 10/378 (3%) |
| Aeromonas spp. | 2/189 (1%) | 9/378 (2%) |
| ETEC e | NA | 7/378 (2%) |
| Yersinia enterocolitica | 0/36 (0%) | 2/378 (1%) |
| Plesiomonas shigelloides | 0/0 (0%) | 1/378 (<1%) |
| Vibrio cholerae | 0/0 (0%) | 1/378 (<1%) |
| Viral pathogens | ||
| Norovirus GI/GII | 0/0 (0%) | 43/378 (11%) |
| Adenovirus F 40/41 | 6/73 (8%) | 16/378 (4%) |
| Rotavirus A | 17/98 (17%) | 16/378 (4%) |
| Sapovirus | NA | 11/378 (3%) |
| Astrovirus | NA | 8/378 (2%) |
| Parasitic pathogens | ||
| Giardia lamblia | 11/125 (9%) | 18/378 (5%) |
| Cryptosporidium spp. | 2/118 (2%) | 6/378 (2%) |
| Cyclospora cayetanensis | 0/0 (0%) | 0/378 (0%) |
| Entamoeba histolytica | 0/0 (0%) | 0/378 (0%) |
Enteropathogenic E. coli (EPEC)
Shigatoxigenic E. coli (STEC)
Enteroinvasive E. coli (EIEC)
Enteroaggregative E. coli (EAEC)
Enterotoxigenic E. coli (ETEC); standard laboratory testing not available (NA).
Statistical Analysis
To characterize findings by physician ordering practice, we divided patients into three groups for the analysis: 1) those for whom the physician only ordered testing for C. difficile; 2) those for whom testing was ordered for C. difficile and other pathogens; and 3) those for whom testing was only ordered for pathogens other than C. difficile. Descriptive statistics were used to characterize the study population and microbiologic testing results. Rates and proportions were compared using the χ2-test or Fisher's exact test, as appropriate. Non-parametric data were compared using the Wilcoxon rank sum test. Alpha was set equal to 0.05; all P-values are 2-sided. Analyses were performed in Stata 11.2 (StataCorp LP, College Station, TX) and R 3.0 (R Foundation for Statistical Computing, Vienna, Austria).
To assess the concordance between standard laboratory testing and the FilmArray GI Panel we evaluated the sensitivity, specificity, positive predictive value, and negative predictive value for C. difficile, rotavirus, Salmonella, and , and STEC using specimens tested by both methods (Supporting Information 1-4).
RESULTS
Patient Characteristics
FilmArray testing was performed on stool specimens submitted during 378 diarrhoeal episodes at PCH from August of 2010 through December of 2012. There were 339 unique patients with a median age of 6 (interquartile range: 3-12) years (40% were <5 years of age); 58% were male. The majority were White (77%), Hispanic (11%), and Black (4%). Patient encounters occurred in the outpatient setting (68%), inpatient (20%), and the emergency department (12%).
Standard Laboratory Testing
A median of 3 (range 1-10) tests were ordered for each diarrhoeal episode. A gastrointestinal pathogen was detected by standard laboratory methods in 175 (46%) of 378 diarrhoeal episodes (Figure 1 and Interactive Data Visualization). Co-infections were identified in 1.6% of diarrhoeal episodes by standard laboratory methods. The most commonly detected pathogens included: C. difficile (77 of 273 tested; 28%), Salmonella (24/189; 13%), rotavirus (17/98; 17%), and STEC (19/193; 10%).
Figure 1.
Clostridium difficile and other diarrhoeal pathogens detected from pediatric stool specimens tested by standard laboratory methods and the FilmArray GI Panel.
The number of tests ordered varied according to the setting. Outpatient encounters had the lowest number of tests ordered (median 2; interquartile range [IQR] 1-3), followed by emergency department visits (median 3; IQR 2-4), and inpatient encounters (median 3; IQR 1-5). A pathogen was detected by standard laboratory methods in 44% of outpatient encounters, 50% of emergency department visits, and 46% of inpatient encounters. C. difficile was detected by standard laboratory testing in 22 of 64 (34%) outpatients, 7 of 47 (15%) emergency department patients, and 48 of 162 (30%) inpatients.
FilmArray GI Panel
When C. difficile testing was performed upon the same specimen using both standard laboratory methods and the FilmArray GI Panel, the sensitivity was 95% (95% confidence interval [CI]: 87-99%) and the specificity was 99% (95% CI: 96-100%) (Supporting Information 1).
A pathogen was identified in stool specimens from 244 of 378 (65%) diarrhoeal episodes. (Table 1 and Figure 1). Multiple pathogens were detected in 77 (20%). The most common pathogens detected included C. difficile (83/378; 22%), norovirus (43/378; 11%), enteropathogenic E. coli (37/378; 10%), and STEC (30/378; 8%).
The aetiologic agents detected varied based upon the setting of the encounter (Table 2 and Supporting Information 5). C. difficile was more frequently detected among outpatients (25/100; 25%) and inpatients (49/202; 24%) than from patients in the emergency department (9/76; 12%) (P = 0.02). However, the overall proportion of specimens with a pathogen identified did not vary between the outpatient, emergency department, and inpatient settings (P = 0.8).
Table 2.
Diarrhoeal pathogens detected from pediatric stool specimens collected in outpatient, emergency, and inpatient settings with the FilmArray GI Panel.
| Pathogen | FilmArray GI Panel |
||
|---|---|---|---|
| Outpatient (n = 100) | Emergency (n = 76) | Inpatient (n = 202) | |
| Bacterial pathogens | |||
| Clostridium difficile | 25 (25%) | 9 (12%) | 49 (24%) |
| EPEC a | 10 (10%) | 10 (13%) | 17 (8%) |
| All STEC b | 7 (7%) | 10 (13%) | 13 (6%) |
| E. coli O157 | 4 (4%) | 5 (7%) | 9 (4%) |
| Non-O157 | 3 (3%) | 5 (7%) | 4 (2%) |
| Salmonella spp. | 3 (3%) | 14 (18%) | 10 (5%) |
| Camplyobacter spp. | 5 (5%) | 5 (7%) | 5 (2%) |
| Shigella / EIEC c | 3 (3%) | 4 (5%) | 4 (2%) |
| EAEC d | 0 (0%) | 6 (8%) | 4 (2%) |
| Aeromonas spp. | 3 (3%) | 5 (7%) | 1 (1%) |
| ETEC e | 2 (2%) | 3 (4%) | 2 (1%) |
| Yersinia enterocolitica | 0 (0%) | 0 (0%) | 2 (1%) |
| Plesiomonas shigelloides | 0 (0%) | 0 (0%) | 1 (1%) |
| Vibrio cholerae | 1 (1%) | 0 (0%) | 0 (0%) |
| Viral pathogens | |||
| Norovirus GI/GII | 13 (13%) | 9 (12%) | 21 (10%) |
| Adenovirus F 40/41 | 4 (4%) | 5 (7%) | 7 (3%) |
| Rotavirus A | 1 (1%) | 3 (4%) | 12 (6%) |
| Sapovirus | 3 (3%) | 4 (5%) | 4 (2%) |
| Astrovirus | 1 (1%) | 1 (1%) | 6 (3%) |
| Parasitic pathogens | |||
| Giardia lamblia | 9 (9%) | 4 (5%) | 5 (2%) |
| Cryptosporidium spp. | 3 (3%) | 2 (3%) | 1 (1%) |
| Negative | 37 (37%) | 16 (21%) | 81 (40%) |
NOTE: Numbers presented within cells are the number of positive specimens and in parentheses are the percentages for each diarrheal pathogen. Pathogens with no detections in all three clinical settings are not featured in this table.
Enteropathogenic E. coli (EPEC)
Shigatoxigenic E. coli (STEC)
Enteroinvasive E. coli (EIEC)
Enteroaggregative E. coli (EAEC)
Enterotoxigenic E. coli (ETEC).
Use of the FilmArray improved the diagnostic yield from 46% to 65% as compared to standard laboratory testing methods (P < 0.001). Similar improvements were identified among patients who had one, two or three, or four or more standard laboratory tests ordered (18%, 23%, and 14%, respectively).
Findings by Physician Test Selection
Only C. difficile Testing Requested
Standard laboratory testing for C. difficile was the only test ordered in 91 episodes (Figure 2). C. difficile was detected by standard laboratory methods in 42 episodes (46%) and by the FilmArray GI Panel in 40 (44%) (P = 0.8). An additional pathogen was detected by FilmArray in 26 (29%); 11 (12%) as a co-infection with C. difficile. Co-detections among the 11 patients with C. difficile included 7 with norovirus, 1 with Salmonella, and 1 with Campylobacter spp. Pathogens detected among the 15 children who tested negative for C. difficile included norovirus (4), astrovirus (2), sapovirus (1), and G. lamblia (1). In this group, the proportion of episodes without an identified pathogen declined from 54% to 40% using the FilmArray GI Panel (P = 0.05). A complete listing of the pathogens detected with standard laboratory testing and the FilmArray GI Panel is featured in Supporting Information 6.
Figure 2.
Physician ordering patterns and their impact upon diarrhoeal pathogen detection from pediatric stool specimens tested with standard laboratory methods and the FilmArray GI Panel.
The majority (59%) of stool samples in which only C. difficile testing was ordered were collected during inpatient encounters. Of these, 50% tested positive for C. difficile by standard laboratory methods. Similarly, 46% were positive for C. difficile using the FilmArray GI Panel, although an additional 19% had another pathogen detected.
Among children who only had C. difficile testing requested, the greatest increase in pathogen detection was among children 1-4 years of age (n = 42) (Supporting Information 7). In this sub-group, 20/42 (48%) children had C. difficile detected by standard laboratory testing methods while 30/42 (71%) had a pathogen detected by the FilmArray GI Panel (P = 0.03). The additional pathogens detected included: norovirus (n = 7), EPEC (n = 3), Campylobacter (n = 2), sapovirus (n = 2), G. lamblia (n = 1), and P. shigelloides (n = 1).
C. difficile Requested in Combination with Other Tests
In 182 of 273 (67%) episodes C. difficile testing was requested along with one or more additional tests. Of these, 51% had one or more pathogens detected by standard laboratory tests compared to 68% when tested with the FilmArray GI Panel (P = 0.001) (Figure 2). The most common pathogens detected using standard laboratory tests were C. difficile (35; 19%), Salmonella (15; 9%), STEC (13; 7%), and rotavirus (9; 5%). Co-infections were detected in 4 (2%) of specimens. Using the FilmArray GI Panel, C. difficile was detected in 35 (19%), norovirus in 28 (15%), EPEC in 22 (12%), STEC in 20 (11%), and Salmonella in 16 (9%). In 15 of 35 (43%) patients in whom C. difficile was detected an additional pathogen was also identified. The most common co-infecting pathogens were norovirus (n = 7), EPEC (n = 5), Campylobacter spp. (n = 2), and Salmonella (n = 1). When C. difficile was not detected, the FilmArray GI Panel detected a pathogen in 60% (88/147), including norovirus (21/147; 14%), STEC (20/147; 14%), Salmonella (15/147; 20%), and EPEC (17/147; 12%). When compared to standard laboratory methods, the proportion of specimens with more than one diarrhoeal pathogen identified rose from 2% to 25% (P < 0.001).
Only Tests Other than C. difficile Requested
In 105 (28%) episodes of diarrhoea, stool specimens were submitted for standard laboratory testing for one or more diarrhoeal pathogens, but did not have C. difficile testing. Of these, 40 (38%) had a pathogen detected using standard laboratory testing methods. With the FilmArray GI Panel, a pathogen was detected in 63% (P < 0.001); multiple pathogens were detected in 19 (18%).
The most common pathogens detected by standard laboratory testing methods were G. lamblia (10/105; 10%), Salmonella (8/105; 8%), rotavirus (7/105; 7%), and STEC (6/105; 6%). Using the FilmArray GI Panel, G. lamblia (11/105; 10%), Salmonella (10/105; 10%), STEC (10/105; 10%), C. difficile (8/105; 8%), and rotavirus (7/105; 7%) were the most common. The proportion of specimens with multiple pathogens rose from 2% to 18% (P < 0.001). The proportion of specimens that had a pathogen identified did not significantly differ by age (P = 0.07) or patient location (P = 0.1).
DISCUSSION
Correctly diagnosing the aetiology of infectious diarrhoea depends both on the physician's decision to order the correct test and the sensitivity of the testing method. In this study, we compared the detection of infectious pathogens by physician-selected standard tests to a multiplex PCR assay that simultaneously detects 23 bacterial, viral and protozoal pathogens. In this sample, the identification of a pathogen increased from 46% of episodes to 65%. Co-detection of multiple diarrhoeal pathogens increased from 2% to 20%. Applying the FilmArray GI Panel to patents that clinicians had decided warranted diagnostic testing identified 72 additional viral infections, including 43 cases of norovirus, 13 additional protozoal pathogens and 100 additional potential bacterial infections. The bacterial infections that were not detected by physician-selected testing included many with clear clinical importance, including: 11 additional STEC, 9 Campylobacter, 7 Aeromonas, 7 ETEC, 6 Shigella, and 6 Salmonella. This study demonstrated some limitations of physician-specified testing for patients with diarrhoea. Among children in whom C. difficile testing was not ordered, 8% had C. difficile detected using the FilmArray GI Panel. Conversely, when only C. difficile was sought, additional pathogens were detected in 28%.
C. difficile is an important and frequent cause of nosocomial and antibiotic-associated diarrhoea in adults and is increasingly recognized as an important pathogen among children.(16-19) Kim et al. reported a 53% increase in the incidence of C. difficile among 22 free-standing children's hospitals across the U.S. from 2001 to 2006.(20) In a recent propensity matched cohort study, the morbidity and costs attributable to C. difficile infections in hospitalized children were substantial; $18,900 per episode of community onset C. difficile infection and $93,600 for an episode of hospital-onset C. difficile infection.(21) These findings underscore the importance of detecting C. difficile in a variety of clinical settings. It is noteworthy that we detected C. difficile in 8% of diarrhoeal episodes where C. difficile testing was not sought.
To be useful, diagnostic tests for diarrhoeal pathogens should impact clinical care, infection prevention, and public health.(6) Routine stool cultures identify a diarrhoeal pathogen infrequently, with estimates from several studies ranging from 1.5-5.8%.(22-24) Use of the FilmArray GI Panel identified a substantial number of additional bacterial infections for which treatment may be helpful (9 Campylobacter, 7 ETEC, 7 Aeromonas, 6 Shigella, 2 Y. enterocolitica and 1 Plesiomonas shigelloides) and where the use of antibiotics may be undesirable (11 STEC and 3 Salmonella).(6)
Norovirus is the leading cause of foodborne disease outbreaks and is increasingly recognized as a nosocomial pathogen, sometimes mimicking C. difficile.(25-27) With multiplex testing, we identified norovirus in 11% of diarrhoeal episodes. Interestingly, using the same CDC definitions designed for C. difficile, 3 hospital-onset cases of norovirus were identified, defined as detection more than 72 hours after admission. Routine detection of norovirus could facilitate infection control efforts in the hospital and outbreak detection in the community.(27)
Limited data exist describing the extent to which co-infections complicate the clinical presentation of paediatric diarrhoea.(28) Tvede et al. evaluated 32 Swedish children who were hospitalized with C. difficile infection and found that 44% were concurrently infected with another bacterial pathogen, including Campylobacter spp., Salmonella, Y. enterocolitica, and E. coli.(29) In this study, 20% of specimens tested were positive for two or more diarrhoeal pathogens using the FilmArray GI Panel. Among patients with C. difficile infection, 35% had at least one additional diarrhoeal pathogen identified, including norovirus in 17%. It is unclear if these co-infections impacted disease severity.
We frequently detected genes associated with EPEC and EAEC; however, detection of these genes may not prove that they are present in a single organism. The complex molecular pathogenesis and overlapping virulence genes complicates the interpretation of pathogenic E. coli detection.(30) However, in a case control study in a paediatric emergency room in Seattle, Denno et al. demonstrated that EAEC was significantly associated with acute diarrhoea.(31)
The results of this study are subject to several limitations. First, this was a convenience sample of diarrhoeal stool specimens designed to look at test ordering patterns and not a random or sequential sample of diarrhoeal episodes. Therefore, the detection of pathogens by standard methods was higher in this study than would be expected in routine testing. For example, in our clinical laboratory, a pathogen is identified in about 12% of all specimens by routine physician-selected testing, compared to 46% in this study. Second, this study was laboratory-based and analysed de-identified samples; therefore limited clinical data were available for review. Most importantly, no control group of asymptomatic patients was included making it impossible to establish whether detection of a specific pathogen or co-infections with multiple diarrhoeal pathogens was associated with disease.
The diagnosis and management of paediatric diarrhoea is complicated and current testing strategies are expensive and inefficient. Our data support the use of the FilmArray GI Panel or other multiplex testing platforms to simultaneously detect a wide spectrum of diarrhoeal pathogens. Future studies will need to evaluate the clinical and epidemiologic utility, efficiency, accuracy, and cost effectiveness of multiplex molecular testing.
Supplementary Material
Acknowledgments
Funding
This study and the development of the FilmArray GI Panel were supported by NIH Grant #5R01AI089489.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interest
MR, BH, MV, RC, MP, and ST are employees of BioFire Diagnostics, Inc., the maker and manufacturer of the FilmArray GI Panel.
Disclaimer
This publication contains information regarding assays that have not been cleared for the FDA for in vitro diagnostic use.
REFERENCES
- 1.Cheng AC, McDonald JR, Thielman NM. Infectious diarrhea in developed and developing countries. Journal of clinical gastroenterology. 2005;39(9):757–73. doi: 10.1097/01.mcg.0000177231.13770.07. Epub 2005/09/08. [DOI] [PubMed] [Google Scholar]
- 2.Walker CL, Rudan I, Liu L, Nair H, Theodoratou E, Bhutta ZA, et al. Global burden of childhood pneumonia and diarrhoea. Lancet. 2013;381(9875):1405–16. doi: 10.1016/S0140-6736(13)60222-6. Epub 2013/04/16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Scallan E, Griffin PM, Angulo FJ, Tauxe RV, Hoekstra RM. Foodborne illness acquired in the United States--unspecified agents. Emerging infectious diseases. 2011;17(1):16–22. doi: 10.3201/eid1701.P21101. Epub 2011/01/05. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Steiner TS, Samie A, Guerrant RL. Infectious diarrhea: new pathogens and new challenges in developed and developing areas. Clinical infectious diseases. 2006;43(4):408–10. doi: 10.1086/505874. Epub 2006/07/14. [DOI] [PubMed] [Google Scholar]
- 5.Buss SN, Alter R, Iwen PC, Fey PD. Implications of culture-independent panel-based detection of Cyclospora cayetanensis. Journal of clinical microbiology. 2013;51(11):3909. doi: 10.1128/JCM.02238-13. Epub 2013/08/30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Guerrant RL, Van Gilder T, Steiner TS, Thielman NM, Slutsker L, Tauxe RV, et al. Practice guidelines for the management of infectious diarrhea. Clinical infectious diseases. 2001;32(3):331–51. doi: 10.1086/318514. Epub 2001/02/15. [DOI] [PubMed] [Google Scholar]
- 7.Liu J, Gratz J, Amour C, Kibiki G, Becker S, Janaki L, et al. A laboratory-developed TaqMan Array Card for simultaneous detection of 19 enteropathogens. Journal of clinical microbiology. 2013;51(2):472–80. doi: 10.1128/JCM.02658-12. Epub 2012/11/24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Navidad JF, Griswold DJ, Gradus MS, Bhattacharyya S. Evaluation of Luminex xTAG gastrointestinal pathogen analyte-specific reagents for high-throughput, simultaneous detection of bacteria, viruses, and parasites of clinical and public health importance. Journal of clinical microbiology. 2013;51(9):3018–24. doi: 10.1128/JCM.00896-13. Epub 2013/07/16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Panchalingam S, Antonio M, Hossain A, Mandomando I, Ochieng B, Oundo J, et al. Diagnostic microbiologic methods in the GEMS-1 case/control study. Clinical infectious diseases. 2012;55(Suppl 4):S294–302. doi: 10.1093/cid/cis754. Epub 2012/11/28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Haque R, Roy S, Siddique A, Mondal U, Rahman SM, Mondal D, et al. Multiplex real-time PCR assay for detection of Entamoeba histolytica, Giardia intestinalis, and Cryptosporidium spp. The American journal of tropical medicine and hygiene. 2007;76(4):713–7. Epub 2007/04/12. [PubMed] [Google Scholar]
- 11.Paton AW, Paton JC. Multiplex PCR for direct detection of Shiga toxigenic Escherichia coli strains producing the novel subtilase cytotoxin. Journal of clinical microbiology. 2005;43(6):2944–7. doi: 10.1128/JCM.43.6.2944-2947.2005. Epub 2005/06/16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Vidal M, Kruger E, Duran C, Lagos R, Levine M, Prado V, et al. Single multiplex PCR assay to identify simultaneously the six categories of diarrheagenic Escherichia coli associated with enteric infections. Journal of clinical microbiology. 2005;43(10):5362–5. doi: 10.1128/JCM.43.10.5362-5365.2005. Epub 2005/10/07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Poritz MA, Blaschke AJ, Byington CL, Meyers L, Nilsson K, Jones DE, et al. FilmArray, an automated nested multiplex PCR system for multi-pathogen detection: development and application to respiratory tract infection. PloS one. 2011;6(10):e26047. doi: 10.1371/journal.pone.0026047. Epub 2011/11/01. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Blaschke AJ, Heyrend C, Byington CL, Fisher MA, Barker E, Garrone NF, et al. Rapid identification of pathogens from positive blood cultures by multiplex polymerase chain reaction using the FilmArray system. Diagnostic microbiology and infectious disease. 2012;74(4):349–55. doi: 10.1016/j.diagmicrobio.2012.08.013. Epub 2012/09/25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rogatcheva M, Harrel B, Vaughn M, Crisp R, Li C, Wallace R, Trauscht R, Healy T, Pavia AT, Barney T, Daly J, Pang XL. Detection of Viral Diarrheal Pathogens by the FilmArray GI Panel.. The 30th Annual Clinical Virology Symposium; Daytona Beach, FL. April 27-30, 2014; Poster #29. [Google Scholar]
- 16.Cohen SH, Gerding DN, Johnson S, Kelly CP, Loo VG, McDonald LC, et al. Clinical practice guidelines for Clostridium difficile infection in adults: 2010 update by the society for healthcare epidemiology of America (SHEA) and the infectious diseases society of America (IDSA). Infection control and hospital epidemiology. 2010;31(5):431–55. doi: 10.1086/651706. Epub 2010/03/24. [DOI] [PubMed] [Google Scholar]
- 17.Gorbach SL. Antibiotics and Clostridium difficile. The New England journal of medicine. 1999;341(22):1690–1. doi: 10.1056/NEJM199911253412211. Epub 1999/11/26. [DOI] [PubMed] [Google Scholar]
- 18.Sammons JS, Toltzis P, Zaoutis TE. Clostridium difficile Infection in children. JAMA pediatrics. 2013;167(6):567–73. doi: 10.1001/jamapediatrics.2013.441. Epub 2013/03/06. [DOI] [PubMed] [Google Scholar]
- 19.Sandora TJ, Fung M, Flaherty K, Helsing L, Scanlon P, Potter-Bynoe G, et al. Epidemiology and risk factors for Clostridium difficile infection in children. The Pediatric infectious disease journal. 2011;30(7):580–4. doi: 10.1097/INF.0b013e31820bfb29. Epub 2011/01/15. [DOI] [PubMed] [Google Scholar]
- 20.Kim J, Smathers SA, Prasad P, Leckerman KH, Coffin S, Zaoutis T. Epidemiological features of Clostridium difficile-associated disease among inpatients at children's hospitals in the United States, 2001-2006. Pediatrics. 2008;122(6):1266–70. doi: 10.1542/peds.2008-0469. Epub 2008/12/03. [DOI] [PubMed] [Google Scholar]
- 21.Sammons JS, Localio R, Xiao R, Coffin SE, Zaoutis T. Clostridium difficile infection is associated with increased risk of death and prolonged hospitalization in children. Clinical infectious diseases. 2013;57(1):1–8. doi: 10.1093/cid/cit155. Epub 2013/03/28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Guerrant RL, Shields DS, Thorson SM, Schorling JB, Groschel DH. Evaluation and diagnosis of acute infectious diarrhea. The American journal of medicine. 1985;78(6B):91–8. doi: 10.1016/0002-9343(85)90370-5. Epub 1985/06/28. [DOI] [PubMed] [Google Scholar]
- 23.Van Gilder T, Christensen D, Shallow S, et al. Variations in stool handling and culturing practices among clinical microbiology laboratories within the Foodborne Active Surveillance Network (FoodNet): do we need practice guidelines?. 99th American Society for Microbiology; Chicago, IL. July 1999. [Google Scholar]
- 24.Slutsker L, Ries AA, Greene KD, Wells JG, Hutwagner L, Griffin PM. Escherichia coli O157:H7 diarrhea in the United States: clinical and epidemiologic features. Annals of internal medicine. 1997;126(7):505–13. doi: 10.7326/0003-4819-126-7-199704010-00002. Epub 1997/04/01. [DOI] [PubMed] [Google Scholar]
- 25.Bresee JS, Widdowson MA, Monroe SS, Glass RI. Foodborne viral gastroenteritis: challenges and opportunities. Clinical infectious diseases. 2002;35(6):748–53. doi: 10.1086/342386. Epub 2002/08/31. [DOI] [PubMed] [Google Scholar]
- 26.Koo HL, Ajami NJ, Jiang ZD, Dupont HL, Atmar RL, Lewis D, et al. A nosocomial outbreak of norovirus infection masquerading as clostridium difficile infection. Clinical infectious diseases. 2009;48(7):e75–7. doi: 10.1086/597299. Epub 2009/02/28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Said MA, Perl TM, Sears CL. Healthcare epidemiology: gastrointestinal flu: norovirus in health care and long-term care facilities. Clinical infectious diseases. 2008;47(9):1202–8. doi: 10.1086/592299. Epub 2008/09/24. [DOI] [PubMed] [Google Scholar]
- 28.McFarland LV, Brandmarker SA, Guandalini S. Pediatric Clostridium difficile: a phantom menace or clinical reality? Journal of pediatric gastroenterology and nutrition. 2000;31(3):220–31. doi: 10.1097/00005176-200009000-00004. Epub 2000/09/21. [DOI] [PubMed] [Google Scholar]
- 29.Tvede M, Schiotz PO, Krasilnikoff PA. Incidence of Clostridium difficile in hospitalized children. A prospective study. Acta paediatrica Scandinavica. 1990;79(3):292–9. doi: 10.1111/j.1651-2227.1990.tb11459.x. Epub 1990/03/01. [DOI] [PubMed] [Google Scholar]
- 30.Nataro JP, Mai V, Johnson J, Blackwelder WC, Heimer R, Tirrell S, et al. Diarrheagenic Escherichia coli infection in Baltimore, Maryland, and New Haven, Connecticut. Clinical infectious diseases. 2006;43(4):402–7. doi: 10.1086/505867. Epub 2006/07/14. [DOI] [PubMed] [Google Scholar]
- 31.Denno DM, Shaikh N, Stapp JR, Qin X, Hutter CM, Hoffman V, et al. Diarrhea etiology in a pediatric emergency department: a case control study. Clinical infectious diseases. 2012;55(7):897–904. doi: 10.1093/cid/cis553. Epub 2012/06/16. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.