RESUMEN
OBJECTIVE: To assess predictors of diarrhea and dehydration and to investigate the role of diarrhea in mortality among children with complicated severe acute malnutrition. STUDY DESIGN: A prospective cohort study, nested in a probiotic trial, was conducted in children with complicated severe acute malnutrition. Children were treated according to World Health Organization and national guidelines, and diarrhea and dehydration were assessed daily. Multiple linear and log-linear Poisson regression models were used to identify predictors of days with diarrhea and dehydration, respectively, and multiple logistic regression was used to assess their role in mortality. RESULTS: Among 400 children enrolled, the median (IQR) age was 15.0 months (11.2-19.2 months), 58% were boys, and 61% had caregiver-reported diarrhea at admission. During hospitalization, the median (range) number of days with diarrhea was 5 (0-31), the median duration of hospitalization was 17 days (1-69 days), and 39 (10%) died. Of 592 diarrhea episodes monitored, 237 were admission episodes and 355 were hospital acquired. During hospitalization, young age was associated with days with diarrhea, and young age and HIV infection were associated with dehydration. Both days with diarrhea and dehydration predicted duration of hospitalization as well as mortality. The odds of mortality increased by a factor of 1.4 (95% CI, 1.2-1.6) per day of diarrhea and 3.5 (95% CI, 2.2-6.0) per unit increase in dehydration score. CONCLUSIONS: Diarrhea is a strong predictor of mortality among children with complicated severe acute malnutrition. Improved management of diarrhea and prevention of hospital-acquired diarrhea may be critical to decreasing mortality.