Postoperative bleeding after complex abdominal wall reconstruction: A post hoc analysis of a randomized clinical trial.

Background: Abdominal wall reconstruction requires extensive dissection of the abdominal wall, exposure of the retroperitoneum, and aggressive chemoprophylaxis to reduce the risk of thromboembolic complications. The need for early anticoagulation puts patients at risk for bleeding. We aimed to quantify postoperative blood loss, incidence of transfusion and reoperation, and associated risk factors in patients undergoing complex abdominal wall reconstruction.

Methods: All patients underwent a posterior component separation with transversus abdominis release and placement of retromuscular mesh for ventral hernias <20 cm wide and were enrolled in a clinical trial assessing the utility of trans-fascial mesh fixation. A post hoc analysis was performed to quantify postoperative hemoglobin drop, blood transfusions, and procedural interventions for ongoing bleeding during the first 30 postoperative days. Multivariate logistic regression was used to identify predictors of transfusion.

Results: In 325 patients, hemoglobin decreased by 3.61 (±1.58) g/dL postoperatively. Transfusion incidence was 9.5% (n = 31), and 3.1% (n = 10) required a surgical intervention for bleeding. Initiation of therapeutic anticoagulation postoperatively resulted in a higher likelihood of requiring surgical intervention for bleeding (odds ratio 10.4 [95% confidence interval 2.75-43.8], P < .01). Use of perioperative therapeutic anticoagulation was associated with higher rates of transfusion (odds ratio 3.51 [95% confidence interval 1.34-8.53], P < .01). Neither intraoperative blood loss nor operative times were associated with an increased transfusion requirement or need for operative intervention.

Conclusion: Patients undergoing transversus abdominis release are at a high risk of postoperative bleeding that can require transfusion and reoperation. Patients requiring postoperative therapeutic anticoagulation are at particularly high risk.