Pediatric stool culture tests: reducing workload and costs

Source: Church DL, Cadrain G, Kabani A et al: Practice guidelines for ordering stool cultures in a pediatric population. Am J Clin Pathol 1995; 103: 149-153

Summary: Canadian Medical Association Journal 1996; 154: 528

The practice of ordering multiple sequential stool cultures regardless of the likelihood of a bacterial cause has made stool tests a high-volume, labour-intensive task in microbiology laboratories. Because the rate of positive findings from cultured stool samples from children with acute diarrheal illness is generally very low, researchers conducted a comprehensive utilization review of all stool cultures performed at Alberta Children's Hospital during a 3-year period in order to establish safe and sustainable practice guidelines for ordering stool cultures in this population.

The patient population included 6700 children in hospital, 36 000 who presented to the emergency department and 60 000 who were seen at clinics and physicians' offices. Stool cultures were surveyed from 3420 of these children, 294 (9%) of whom were found to have enteric bacterial infections. The peak incidence of diarrhea caused by enteric bacterial pathogens occurred in the summer, Escherichia coli 0157:H7 and Salmonella sp. accounting for most positive cultures.

Of the stool tests ordered for children in hospital 16% were multiple culture requests, although only 3% of cases of confirmed enteric infection would have been missed had only single samples been submitted. All cases were confirmed in the first 4 days after admission to hospital; the likelihood of obtaining positive cultures from samples ordered after this time was negligible. Multiple tests were commonly ordered in the pediatric oncology unit, where two or more cultures were submitted for 47% of the children regardless of whether previous culture results were negative. Only 2% of these children had positive culture results.

Among the children who presented to the emergency department with acute diarrhea, 20% had positive stool culture results; however, physicians in the emergency department were the least likely to order multiple tests. By contrast, 81% of cultures ordered by the gastroenterology clinic were repeat tests, despite the very low incidence rate of enteric bacterial infection (2%) among these children. Of the children who attended clinics or physician's office, 7% had positive stool samples, all clinically significant cultures resulting from the first sample.

Based on these results, two guidelines were implemented:

  1. Only one stool culture test should be ordered initially for all children regardless of location.
  2. Stool cultures are discouraged for children in whom diarrhea develops after their fourth day in hospital.
Implementation of the guidelines reduced laboratory workload by 36% and supply costs by 30%.

In almost all cases of pediatric diarrheal illness a single stool test is diagnostic, regardless of the type of bacteria isolated. Transporting samples in buffered media will maximize the yield of isolation. Diarrheal illness in most children in hospital is caused by viral gastroenteritis or Clostridium difficile toxin-mediated diarrhea associated with antibiotic use; it can also be associated with chemotherapy and graft-versus-host disease. Therefore, positive stool culture results are not likely to be found after the first 4 days in hospital. It is recommended that physicians consider epidemiologic data such as time of year and recent travel abroad when treating children with acute diarrhea.

| CMAJ February 15, 1996 (vol 154, no 4) |