Meta‐analysis of the impact of postoperative complications on survival after oesophagectomy for cancer

BACKGROUND: Oesophagectomy has a high risk of postoperative morbidity. The impact of postoperative complications on overall survival of oesophageal cancer remains unclear. This meta‐analysis addressed the impact of complications on long‐term survival following oesophagectomy. METHODS: A search of PubMed and Cochrane Library databases was undertaken for systematic review of papers published between January 1995 and August 2016 that analysed the relation between postoperative complications and long‐term survival. In the meta‐analysis, data were pooled. The main outcome was overall survival (OS). Secondary endpoints included disease‐free (DFS) and cancer‐specific (CSS) survival. RESULTS: A total of 357 citations was reviewed; 21 studies comprising 11 368 patients were included in the analyses. Overall, postoperative complications were associated with significantly decreased 5‐year OS (hazard ratio (HR) 1·16, 95 per cent c.i. 1·06 to 1·26; P = 0·001) and 5‐year CSS (HR 1·27, 1·09 to 1·47; P = 0·002). Pulmonary complications were associated with decreased 5‐year OS (HR 1·37, 1·16 to 1·62; P < 0·001), CSS (HR 1·60, 1·35 to 1·89; P < 0·001) and 5‐year DFS (HR 1·16, 1·00 to 1·33; P = 0·05). Patients with anastomotic leakage had significantly decreased 5‐year OS (HR 1·20, 1·10 to 1·30; P < 0·001), 5‐year CSS (HR 1·81, 1·11 to 2·95; P = 0·02) and 5‐year DFS (HR 1·13, 1·02 to 1·25; P = 0·01). CONCLUSION: Postoperative complications after oesophagectomy, including pulmonary complications and anastomotic leakage, decreased long‐term survival.