Comparative efficacy of reinforced suturing, transanal drainage tube, and no additional intervention in preventing anastomotic leakage after rectal cancer surgery: a network meta-analysis
摘要
Anastomotic leakage (AL) is a severe complication after rectal cancer surgery. This network meta-analysis (NMA) compares reinforced suturing (RS), transanal drainage tube (TDT), and no additional intervention (NRT) for AL prevention.
MethodsAn NMA was conducted according to PRISMA-NMA guidelines. PubMed, Web of Science, and Embase were searched for randomized controlled trials and observational studies comparing RS, TDT, or NRT in adults undergoing anterior resection for rectal cancer, with AL as the primary outcome. Secondary outcomes included Grade C AL, stricture, bleeding, ileus, and wound infection.
Results16 studies (3 RCTs, 11 RNCTs, and 2 PNCTs; n = 4562) were included. For overall AL incidence, both RS (OR 0.32, 95% CrI 0.16–0.62) and TDT (OR 0.47, 95% CrI 0.33–0.63) significantly reduced AL vs. NRT. RS ranked highest (SUCRA 0.93). Although RS had the highest SUCRA for overall AL, the RS–TDT contrast was not statistically significant(OR 1.44, 95% CrI 0.68–3.09), so ranking should not be over-interpreted as proof of superiority. For Grade C AL, RS significantly reduced risk versus both TDT (OR 5.01, 95% CrI 1.33–28.67) and NRT (OR 0.10, 95% CrI 0.02–0.32; SUCRA 0.99). No significant differences were found among interventions for anastomotic bleeding, ileus, or wound infection. TDT showed a trend toward reduced anastomotic stricture risk (SUCRA 0.73), but the effect was not statistically significant (TDT vs. NRT: OR 0.68, 95% CrI 0.19–2.27). Sensitivity analysis restricted to larger studies (≥ 100 patients/group) confirmed the robustness of primary outcomes.
ConclusionsBoth RS and TDT were associated with a reduction in overall AL risk compared to NRT. Network estimates suggested that RS may be more effective than TDT in preventing the more severe Grade C AL; however, this finding is based on indirect comparisons with wide credible intervals and requires confirmation in future head-to-head trials. The choice of intervention may therefore depend on patient risk profile and clinical context.