A Systematic Review of Randomized Trials Evaluating Regional Techniques for Postthoracotomy Analgesia

Abstract

Thoracotomy induces severe postoperative pain and impairment of pulmonary function, and therefore regional analgesia has been intensively studied in this procedure. Thoracic epidural analgesia is commonly considered the "gold standard" in this setting; however, evaluation of the evidence is needed to assess the comparative benefits of alternative techniques, guide clinical practice and identify areas requiring further research. In this systematic review of randomized trials we evaluated thoracic epidural, paravertebral, intrathecal, intercostal, and interpleural analgesic techniques, compared to each other and to systemic opioid analgesia, in adult thoracotomy. Postoperative pain, analgesic use, and complications were analyzed. Continuous paravertebral block was as effective as thoracic epidural analgesia with local anesthetic (LA) but was associated with a reduced incidence of hypotension. Paravertebral block reduced the incidence of pulmonary complications compared with systemic analgesia, whereas thoracic epidural analgesia did not. Thoracic epidural analgesia was superior to intrathecal and intercostal techniques, although these were superior to systemic analgesia; interpleural analgesia was inadequate. Either thoracic epidural analgesia with LA plus opioid or continuous paravertebral block with LA can be recommended. Where these techniques are not possible, or are contraindicated, intrathecal opioid or intercostal nerve block are recommended despite insufficient duration of analgesia, which requires the use of supplementary systemic analgesia. Quantitative meta-analyses were limited by heterogeneity in study design, and subject numbers were small. Further well designed studies are required to investigate the optimum components of the epidural solution and to rigorously evaluate the risks/benefits of continuous infusion paravertebral and intercostal techniques compared with thoracic epidural analgesia.