RESUMO
AIM: This study aims to describe and compare the spectrum, course, seasonality and outcome of children with virus-associated respiratory symptoms (VARS) admitted to two paediatric intensive care units (PICUs) in the United Kingdom (UK) and South Africa (SA). METHODS: Cross-sectional study of routinely collected data on subjects admitted to PICU with respiratory symptoms and positive respiratory viral polymerase chain reaction between July 2009 and July 2011. RESULTS: Six hundred forty-six samples yielding 765 viral isolates (74% from SA) from 599 patients (53% male; median (interquartile range) age 6.0 (2.3-16.5) months) were included. Rhinovirus, respiratory syncytial virus and adenovirus were most commonly isolated. Adenovirus was more prevalent in SA (24.3% vs. 16.8%, P = 0.03). Possible or likely nosocomial viral acquisition occurred in 78% of isolates in SA versus 48% in the UK (P < 0.0001).Total mortality was 13.5%; 17% in SA versus 4% in the UK (P < 0.0001). Mortality for community acquired VARS was 8.4% versus 16.1% in those with possible nosocomial viral acquisition (P = 0.009). Factors independently associated with mortality were: SA study site (adjusted odds ratio (OR) 3.4, 95% confidence interval (CI) 1.4-8.5; P = 0.008); age (months) (OR 1.0, 95% CI 1.0-1.02; P = 0.001); Paediatric Index of Mortality 2 score (%) (OR 1.0, 95% CI 1.01-1.03; P = 0.0002) and isolation of adenovirus (OR 3.0, 95% CI 1.8-5.0; P < 0.0001). CONCLUSIONS: The outcome of children with VARS was worse in SA compared with the UK PICU. Nosocomial VARS was highlighted as an important concern and requires further investigation.