Postoperative Nonsteroidal Anti-inflammatory Drugs Do Not Increase Postoperative Anastomotic Leakages in Colorectal Surgery: A Meta-analysis

Abstract

Background- Nonsteroidal anti-inflammatory drugs are cornerstones in multimodal analgesia and recent data suggest an association with an increased incidence of anastomotic leakage in surgical colorectal cancer patients. Objective- Aim of this systematic review and meta-analysis was to evaluate any effect of post-operative NSAIDs administration on the risk of anastomotic leakages and to carry out a scatter analysis between promoting dosages and fistulas. Data sources- Indexed and unindexed published articles from 1990 to 2022 were identified by electronic databases researches. Study selection- Two-arm studies about adult colorectal cancer patients undergoing surgery and assigned to postoperative non-steroidal anti-inflammatory drug-based analgesia (experimental cohorts) vs control ones. Intervention- Non-steroidal anti-inflammatory drug administration in the postoperative period. Endpoints- Primary endpoint was the evaluation of the incidence of anastomotic leakages (overall cumulative effect, subgroup pharmacological classes). In the second part of the analysis, a comparative analysis between drugs’ dosage and postoperative fistula was assessed. Results- According to PRISMA ® 2020 flow diagram, fourteen studies were included for a total of 36,330 enrolled patients (15,051 experimental vs 21,279 control) (Cohen’s k: 0.87). Overall postoperative anastomotic leakage incidence was 4.69%. NSAIDs do not increas rge risk of anastomotic leakage after colorectal surgery (OR: 1.24, 95%CI: 0.93–1.66; p = 0.14). At subgroup analysis, non-selective NSAIDs resulted safer than selective ones (OR: 0.51, 95% CI: 0.40–0.66, p < 0.001). No differences among drugs and fistula risk was found (Diclofenac vs Ibuprofen vs Ketorolac: OR: 1.99, p = 0.09 vs OR: 0.91, p = 0.82 vs OR: 1.14, p = 0.63). At the scatter plot analysis, excellent pharmacological safety profiles at maximal dosage for all the molecules investigated was reported. Limitations- Four studies lacked an exhaustive methodological report on protocols and dosages. Subgroup analysis for selected cohorts for type of drug administration could lead to a type 2 statistical error. Conclusions- Postoperative administration of NSAIDs in colorectal cancer surgery is safe and does not increase the risk for postoperative anastomotic dehiscences. However, selective drugs could predispose to dehiscence when compared to non-selective ones.