By Philip R. Fischer, MD, DTM&H

Professor of Pediatrics, Department of Pediatric and Adolescent Medicine, Mayo Clinic, Rochester, MN; Department
of Pediatrics, Sheikh Shakhbout Medical City, Abu Dhabi, United Arab Emirates

SYNOPSIS: For children with acute vomiting (but not diarrhea), rapid rectal swab testing for bacterial and viral pathogens has potential for helping clinicians confirm a diagnosis of gastroenteritis and, thus, avoid unnecessary evaluation for other diagnoses.

SOURCE: Freedman SB, Xie J, Lee BE, et al. Microbial etiologies and clinical characteristics of children seeking emergency department care due to vomiting in the absence of diarrhea. Clin Infect Dis 2021; May 16:ciab451. [Online ahead of print].

Many children with viral gastroenteritis have a day or two of vomiting prior to the onset of diarrhea. However, many laboratories choose not to even test normal formed stools for gastrointestinal pathogens. Children with isolated vomiting (i.e., without associated diarrhea) are not usually evaluated for infectious gastrointestinal diseases, and the infectious etiologies of isolated vomiting have not been well studied.

With that awareness, over a four-year period, Freedman and colleagues in Alberta, Canada, studied intestinal testing (rectal swabs if no diarrhea, stool if diarrhea) in 2,695 children (younger than 18 years of age) with at least three daily bouts of vomiting and/or diarrhea for not more than seven days in two children’s hospital emergency departments (one in Calgary, one in Edmonton). They also tested children who presented to the same emergency departments with health concerns unrelated to possible infections. Testing was done for common viral, bacterial, and parasitic pathogens of children. Clostridioides difficile was not considered to be a pathogen when isolated from children younger than 2 years of age.

At the time of presentation, 11% of children had isolated diarrhea, 49% had vomiting and diarrhea, and 40% had isolated vomiting. The median age of study participants was 2 years. Of children with isolated vomiting, 12% had a non-gastrointestinal chronic medical condition. Isolated vomiting was most common in the fall months. Nearly half (45%) of children with isolated vomiting developed diarrhea within a few days of the initial visit.

Of children with both vomiting and diarrhea, a pathogen was identified in 81%. Children with isolated vomiting had pathogens recovered in 55% of cases. Interestingly, 20% of asymptomatic control children had a pathogen identified, usually norovirus. Bacterial pathogens were found in 6% of children with isolated vomiting, usually C. difficile.

As compared to children with diarrhea, children with isolated vomiting were more likely to undergo testing of urine and imaging with abdominal X-rays and/or ultrasounds. Four percent of children with isolated vomiting were found to have a non-gastrointestinal diagnosis, usually urinary tract infection.

Bacterial pathogens were found in 6% of children with isolated vomiting and in 17% of children with isolated diarrhea. C. difficile, Salmonella, and Aeromonas were the most commonly identified bacterial pathogens in children who neither initially had nor subsequently developed diarrhea.

Children with isolated vomiting had similarly long stays in the emergency department (3.4 hours) to those with both diarrhea and vomiting (3.3 hours).

Rapid viral and bacterial testing of rectal swabs in children with isolated vomiting might be able to reduce a fruitless search for other pathologic etiologies of the vomiting, and testing might, thus, prompt cost savings. Whether microbiology results will be available quickly enough to alter care and whether physicians would alter diagnostic testing based on the identification of rectal pathogens are issues that deserve further study.


Identifying a microbiological cause for a patient’s symptoms is useful if there will be changes in intervention and outcome based on knowledge of the results. Since childhood vomiting and/or diarrhea usually is caused by viruses, and since these infections usually resolve spontaneously without requiring medical care beyond provision of oral hydration, knowing an etiology would not necessarily alter microbiologic management or outcomes.

Last month in Infectious Disease Alert, we reviewed a paper showing that widespread use of microbial testing of stool for children with diarrhea did not alter overall costs or lengths of hospitalization.1 However, we suggested that focused testing (for children with bloody diarrhea, for instance) might potentially prove to be worthwhile. This month, we are considering a newer paper that looked at rectal swab testing of vomiting children who did not (yet, at least) have diarrhea as part of their acute illness. With rapid availability of results, it is possible that knowledge of a viral source of the vomiting might prompt reduction in laboratory and radiologic testing to look for other diagnoses.

With this framework, it is fascinating to see that Freedman and colleagues showed that more than half of children presenting for emergency care with vomiting (without diarrhea) had a viral pathogen identified on rectal swabs. In contrast, only 6% had a potentially treatable bacterial pathogen. X-rays, ultrasounds, and urine tests were done commonly in children with isolated vomiting, but such testing revealed alternate diagnoses (usually urinary tract infection) in only 5.7% of the children. Limiting diagnostic testing in children with isolated vomiting and an identified enteropathogen potentially could shorten stays in emergency departments and reduce the costs for unnecessary testing.

To achieve such cost and time savings from microbiological testing of vomiting children, results would have to be available quickly, and clinicians would need to act differently based on knowledge of the results. Test results now are available within a few hours, potentially quickly enough to alter prescribing habits. However, a recent controlled study of rapid pathogen testing during 931 emergency department encounters for children with respiratory symptoms showed that identification of specific pathogens (positive findings in 85% of subjects) did not significantly alter medication prescriptions.2 Just because results could or should alter diagnostic and therapeutic interventions does not mean that clinicians actually will alter their behavior based on these results. Whether rectal swab testing, when used, would alter testing or treatment in emergency departments remains to be seen. 


  1. Cotter JM, Thomas J, Birkholz M, et al. Clinical impact of a diagnostic gastrointestinal panel in children. Pediatrics 2021;147:e2020036954.
  2. Rao S, Lamb MM, Moss A, et al. Effect of rapid respiratory virus testing on antibiotic prescribing among children presenting to the emergency department with acute respiratory illness: A randomized clinical trial. JAMA Netw Open 2021;4:e2111836.