Early postoperative non-steroidal anti-inflammatory drugs and anastomotic leakage after oesophagectomy

Background Previous studies have suggested that postoperative non-steroidal anti-inflammatory drug (NSAID) use may increase the risk of anastomotic leakage after colorectal surgery. However, the association between NSAIDs and anastomotic leakage after oesophagectomy is unclear. The aim of this retrospective study was to assess the effect of early postoperative NSAID use on anastomotic leakage after oesophagectomy. Methods The Data of patients who underwent oesophagectomy for cancer between July 2010 and March 2019 were extracted from a Japanese nationwide inpatient database. Stabilized inverse probability of treatment weighting (IPTW), propensity score matching, and instrumental variable analyses were performed to investigate the association between NSAID use in the early postoperative period (defined as the day of and the day after surgery) and short-term outcomes, adjusting for potential confounders. The primary outcome was anastomotic leakage. The secondary outcomes were acute kidney injury, gastrointestinal bleeding, and mortality. Results Among 39 418 eligible patients, early postoperative NSAIDs were used by 16 211 individuals (41 per cent). Anastomotic leakage occurred in 5729 patients (15 per cent). In stabilized IPTW analyses, NSAIDs were not associated with anastomotic leakage (odds ratio 1.04, 95 per cent c.i. 0.97 to 1.10). The proportions of acute kidney injury and gastrointestinal bleeding, as well as 30-day mortality and in-hospital mortality, did not differ according to NSAID use. Propensity score matching and instrumental variable analyses demonstrated similar results. Conclusion Early postoperative NSAID use was not associated with anastomotic leakage or other complications in patients who underwent oesophagectomy. In this nationwide inpatient database study investigating 39 418 oesophagectomies for oesophageal cancer, a stabilized inverse probability of treatment weighting analysis showed that early postoperative non-steroidal anti-inflammatory drug use was not associated with anastomotic leakage. Propensity score matching and instrumental variable analyses demonstrated similar results.