RESUMO
BACKGROUND: Epidurals provide excellent analgesia for cardiac surgery and may reduce complications. However, their use has been tempered because of concern of the rare, but serious complication of epidural haematoma. The aim of this meta-analysis was to assess the effect of epidural on survival and the risk estimate of epidural haematoma. METHODS: A systematic review of the literature (Pubmed, Embase, Scopus and the Cochrane Register) and a meta-analysis of the available randomized and case-matched studies were performed to estimate the effect on survival. An international, directed and viral anonymous survey was performed to identify the incidence of haematomas with a corresponding estimate of the number of epidurals performed. RESULTS: Of 66 randomized and case-matched studies, 57 trials including 6383 patients reported the incidence of all-cause mortality at the longest follow up available, with a significant reduction with epidurals (59/3123 [1.9%] vs 108/3260 [3.3%] in the control arm, RR 0.65 [95% CI 0.48-0.86], P=0.003, NNT=70). No epidural haematoma was reported in these 66 trials (3320 epidurals). All other literature revealed nine haematomas in 13,100 patients. Through the anonymous, web-based, viral, international survey, we identified 16 further, non-published, epidural haematomas from 72,400 positioned epidurals. Therefore, a total of 25 haematomas have been identified from an estimate of 88,820 positioned epidurals, producing an estimated risk of 1:3552 (95% CI 1:2552-1:5841). CONCLUSIONS: The use of epidural analgesia in cardiac surgery is associated with a reduction in mortality (NNT=70), and with an estimated risk of epidural haematoma of 1:3552.
