Optimal strategies for pancreatic stump management in distal pancreatectomy: A network meta-analysis of randomized controlled trials

Abstract

Postoperative pancreatic fistula (POPF) is the most common and clinically significant complication following distal pancreatectomy (DP). Despite numerous randomized trials evaluating stump closure and reinforcement techniques, the optimal strategy remains uncertain. We conducted a Bayesian network meta-analysis of randomized controlled trials (RCTs) to compare various pancreatic stump management strategies in DP. Pancreatic stump management in DP involves several components: transection, duct closure, margin closure, and coverage. We simplified these into three main categories: primary stump closure (suture or stapler), adjunctive coverage (none, mesh, or autologous patch), and reinforced staplers. We performed comparisons using Bayesian network meta-analysis with rank probability estimation. The primary outcome was clinically relevant POPF. Secondary outcomes included biochemical leakage, intra-abdominal fluid collection, intra-abdominal abscess, estimated blood loss, operative time, hospital stay, and overall complications. Our analysis included 20 RCTs, comprising 2,868 patients. For clinically relevant POPF, reinforced staplers were associated with a lower risk compared to suture-based closure (odds ratio [OR] 0.41, 95% credible interval [CrI] 0.17-0.95) and ranked highest among all strategies. For intra-abdominal fluid collection, autologous patch-based techniques demonstrated a protective effect (autologous patch stapler OR 0.15, 95% CrI 0.02-0.91; autologous patch suture OR 0.28, 95% CrI 0.09-0.74). However, clinically relevant POPF persisted across all evaluated strategies, indicating a residual risk that we conceptualize as the "POPF grey zone." Future technical advancements should focus on mitigating this grey zone by improving protection of the pancreatic stump margin.