OBJECTIVES

The aim of the study was to determine whether intravenous fluid administration is independently associated with a reduction in unscheduled emergency department (ED) revisits within 7 days.

METHODS

We conducted a single-center, retrospective observational cohort study in a pediatric ED in Toronto, Canada. Participants were younger than 18 years, diagnosed as having gastroenteritis, and discharged home between July 2003 and June 2008. Multivariable regression models were used to determine the associations between the exposures (intravenous rehydration, triage severity score, age) and ED revisits and revisits with intravenous rehydration. Accuracy was assessed using bootstrap analysis.

RESULTS

There were 22,125 potentially eligible visits; 3346 were included in our final cohort. A total of 497 children (15%) received intravenous rehydration and 543 (16%) had an unscheduled revisit. Regression analysis included 2874 children with complete data, and identified 5 independent predictors of an ED revisit: intravenous rehydration (odds ratio [OR] 1.76; 95% confidence interval [CI] 1.36-2.26); number of vomiting episodes (1.20; 95% CI 1.04-1.28/5 episode increase); days of diarrhea (OR 0.92; 95% CI 0.88-0.97/day increase); frequency of diarrhea (1.19; 95% CI 1.03-1.38/5 episode increase); and age (OR 0.94; 95% CI 0.91-0.98/year). Bootstrap methodology identified intravenous rehydration, age, number of vomiting episodes, days of diarrhea, and number of diarrheal stools a minimum of 500 of 1000 iterations.

CONCLUSIONS

Intravenous rehydration is associated with unscheduled ED revisits after adjustment for clinical findings. Although children experiencing revisits were likely more unwell, our data do not support the provision of intravenous fluids to prevent unscheduled ED revisits in children with mild-to-moderate dehydration.