651

Introduction: The purpose of this review was to systematically identify and characterize randomized controlled trials (RCTs) in pediatric critical care. Hypothesis: We hypothesize that RCTs in critically ill children are limited in number, size and methodologic quality. Methods: We searched MEDLINE, EMBASE, LILACS and CENTRAL (from inception to July 1, 2012). We included published RCTs and quasi-randomized trials that administered any intervention to children in a pediatric critical care unit. We excluded trials enrolling exclusively newborns, cross-over trials and those only published as abstracts. We used no language restrictions. Pairs of reviewers independently screened studies for eligibility and abstracted data. Discrepancies were resolved by consensus. Results: We included 226 RCTs out of 7280 unique citations screened: 39 (17.3%) were multicentered and 10 (4.4%) were multinational. The most frequently studied populations were mixed critically ill (43.4%), post-operative cardiac surgery (16.4%) and bronchiolitis (7.1%). The most frequently evaluated interventions were medications (59.7%), devices (10.6%), nutrition (9.3%) and mechanical ventilation (8.0%). Almost half, 105 (46.5%), reported blinding. Of the 93 (41.2%) trials that reported an a priori sample size, 32 (34.4%) were stopped early and 20 (21.5%) were stopped for futility, funding or recruitment problems. The median number of children randomized per trial was 47.5 and varied from 6 to 1199. Of the 133 (58.9%) trials reporting the mean age children enrolled, the mean was 4.3 years and varied from 1 month to 14.3 years. Of the 135 (59.7%) trials reporting mortality outcomes, the median mortality rate was 6.4%, varying from 0% to 88.2%. The frequency of RCT publication increased at a mean rate of 1.2 RCT/year (p<0.001) from 1988 to 2011, but the median sample size did not change significantly (p=0.31). Conclusions: The majority of RCTs in pediatric critical care focused on medications, were single-centered and of variable methodological quality. The number of published pediatric critical care trials is increasing but the sample size is not. Significant challenges still exist in completing these studies.Funding: The Canadian Institutes of Health Research.