Haemostatic Efficacy of Topical Agents During Liver Resection: A Network Meta-Analysis of Randomised Trials

• Published in: World journal of surgery Vol. 44; no. 10; pp. 3461 - 3469
• Main Authors: Wells, Cameron I., Ratnayake, Chathura B. B., Mentor, Keno, Sen, Gourab, Hammond, John S., French, Jeremy J., Wilson, Colin H., Manas, Derek, White, Steve, Pandanaboyana, Sanjay
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01-10-2020; Springer Nature B.V
• Subjects: Abdominal Surgery; Bayes Theorem; Bayesian analysis; Bile; Bleeding; Blood transfusion; Cardiac Surgery; Clinical trials; Collagen; Complications; Fibrin; Fibrin Tissue Adhesive - therapeutic use; General Surgery; Hemostasis; Hemostatics - therapeutic use; Hepatectomy; Hepatectomy - adverse effects; Hepatectomy - methods; Humans; Liver; Medicine; Medicine & Public Health; Meta-analysis; Network Meta-Analysis; Postoperative Complications - etiology; Randomization; Randomized Controlled Trials as Topic; Scientific Review; Surgery; Thoracic Surgery; Transfusion; Vascular Surgery
• ISSN: 0364-2313; 1432-2323
• DOI: 10.1007/s00268-020-05621-z

Background
Hepatic resection carries a high risk of parenchymal bleeding both intra- and post-operatively. Topical haemostatic agents are frequently used to control bleeding during hepatectomy, with multiple products currently available. However, it remains unknown which of these is most effective for achieving haemostasis and improving peri-operative outcomes.
Methods
A systematic review and random-effects Bayesian network meta-analysis of randomised trials investigating topical haemostatic agents in hepatic resection was performed. Interventions were analysed by grouping into similar products; fibrin patch, fibrin glue, collagen products, and control. Primary outcomes were the rate of haemostasis at 4 and 10 min.
Results
Twenty randomized controlled trials were included in the network meta-analysis, including a total of 3267 patients and 7 different interventions. Fibrin glue and fibrin patch were the most effective interventions for achieving haemostasis at both 4 and 10 min. There were no significant differences between haemostatic agents with respect to blood loss, transfusion requirements, bile leak, post-operative complications, reoperation, or mortality.
Conclusions
Amongst the haemostatic agents currently available, fibrin patch and fibrin glue are the most effective methods for reducing time to haemostasis during liver resection, but have no effect on other peri-operative outcomes. Topical haemostatic agents should not be used routinely, but may be a useful adjunct to achieve haemostasis when needed.