Meta-analysis of predictive factors and outcomes for failure of non-operative management of blunt splenic trauma.

BACKGROUND This study aimed to analyse predictive factors and outcomes of failure of non-operative management (NOM) following blunt splenic trauma. METHODS A systematic review of the literature was performed for studies comparing failed NOM (fNOM) to successful NOM (sNOM) in adults (≥ 16 years). The main endpoints were fNOM and associated mortality. Between-study heterogeneity was assessed. Meta-analysis of high quality studies, identified using the Newcastle-Ottawa Scale, was performed using fixed or random models. RESULTS Four prospective and 21 retrospective studies were included. From 24,615 unselected patients, 3025 experienced fNOM (12%, range 4-52%). Meta-analysis of the high quality studies revealed that mortality was significantly higher with fNOM in unselected age groups (odds ratio 1.93, 95% confidence interval 1.04-3.57, p = 0.04, I(2) = 0%), in those <55 years old (OR 3.42, 95% CI 1.73-6.77, p = 0.02, I(2) = 0%) and in those ≥ 55 years old (OR 2.65, 95% CI 1.20-5.82, p = 0.02, I(2) = 0%). There was a significant improvement in sNOM following introduction of angioembolisation protocols (OR 0.26, 95% CI 0.13-0.53, p<0.002, I(2) = 51%), although these five studies were non-randomised. American Association for the Surgery of Trauma injury grades 4-5, the presence of moderate or large haemoperitoneum, increasing injury severity score and increasing age were all significantly associated with increased risk of fNOM. fNOM led to significantly longer intensive care unit and overall lengths of stay. CONCLUSIONS fNOM leads to increased resource use and increased mortality. Methods of preventing fNOM, such as angioembolisation, warrant further assessment. Patients with increasing age, AAST scores and moderate or large haemoperitoneums may benefit from closer monitoring.