Incidence and outcomes of acute gastroenteritis in Italian children
BACKGROUND AND AIMS: Little is known about the epidemiology of acute gastroenteritis (AGE) and intussusception (IS) or gastrointestinal obstruction (GO) in Italy. We conducted a retrospective study to assess AGE incidence and symptoms, 1-month hospitalization risk, and IS incidence among Italian children. METHODS: A dynamic cohort study was conducted using the PEDIANET general practice research database. The study population comprised 79,949 children aged <10 years, from September 2001 to September 2004. AGE, GO, and IS cases were identified from diagnoses, hospitalizations, and free-text searches in patient diaries, and were individually validated. AGE outcomes were assessed. Episodes with ascertained rotavirus (RV) infection and incidence rates with 95% confidence intervals (CIs) were calculated. RESULTS: Overall, 13,978 AGE episodes were identified, and 20 cases of GO including 9 IS cases. The overall AGE incidence rate was 76.1 (74.8-77.4) per 1000 person-years (PY), dropping with age from 124/1000 PY among 0-12 months old to 97/1000 PY among 13-48 months old, and 45/1000 PY among 5-10 years old. Male/female AGE rates were similar. The overall rate of GO including IS was 11.2 per 100,000 PY (95% CI: 7.1-16.9). The overall rate of IS alone was 5.0 per 100,000 PY (95% CI: 2.5-9.2). The rate of GO decreased substantially with increasing age from 32 per 100,000 PY (95% CI 17-56) among children 0-12 months of age to 5.5 per 100,000 PY (95% CI 2.1-12.1) among children 5-10 years of age. Most children with AGE presented with diarrhea (91%), about 30% had vomiting, and 29% had fever. The hospitalization risk for children with AGE was 3.0% (95% CI: 2.70-3.27) within 30 days after start of AGE. CONCLUSIONS: AGE is a substantial disease burden in primary care, especially among children aged <5 years. AGE etiology is rarely identified. Most of the cases occurred during the winter period and in children of 6-24 months of age, suggesting that rotavirus could be frequently involved. IS constitutes a rare event in this population, although underreporting might have occurred.