Clinical features as predictors of positive stool culture in children with acute diarrhea

Abstract

Background: Acute diarrhea is a common problem in pediatrics. Stool culture shows low favorable rates, high cost, and time for  results.
Objective: This study aimed to identify clinical features to predict positive stool culture in children with acute diarrhea at Hatyai Hospital.
Method: We conducted research with retrospective cross-sectional data collection of children with acute diarrhea at Hatyai Hospital from 1st January 2017 to 31st December 2020
Result: Of 758 children who were diagnosed with acute diarrhea, the incidence of positive stool culture in our study was 20%. In 155 culture-positive patients, Salmonella spp. (80%) was detected most common. Multivariable logistic regression demonstrated that age below 12-months (OR 3.18, 95% CI: 1.78–5.66, p-value <0.001), frequency of diarrhea > 5 times in 24 hours (OR 2.6, 95%CI: 1.93-6.44, p-value <0.001), no comorbidity (OR 2.03, 95%CI: 1.11-3.69, p-value 0.019), no moderate dehydration (OR 2.06, 95% CI: 1.14-3.73, p-value 0.015) and fecal leukocyte (OR 6, 95% CI: 3.35–10.73, p-value <0.001) were associated with positive stool culture with statistical significantly. Using these factors for scoring system to predict positive stool culture by calculating the area under the receiver operating characteristic curve and considered optimal cut-off point for assessing diagnostic accuracy test. If the score of 9 or more predicted positive stool culture. This scoring system is 59% sensitivity and 84% specificity.
Conclusion: Using this scoring system to predict positive stool culture may lead to cost-effective utilization and rationalizing antibiotics use