Development of a composite endpoint for randomized controlled trials in liver surgery

• Published in: British journal of surgery 2011-08, Vol.98 (8), p.1138-1145
• Main Authors: van den Broek, M. A. J., van Dam, R. M., van Breukelen, G. J. P., Bemelmans, M. H., Oussoultzoglou, E., Pessaux, P., Dejong, C. H. C., Freemantle, N., Olde Damink, S. W. M.
• Format: Article
• Language: English
• Subjects: Aged; Biological and medical sciences; Consensus; Endpoint Determination; Female; General aspects; Human health and pathology; Humans; Life Sciences; Liver Diseases - surgery; Male; Medical sciences; Middle Aged; Postoperative Complications - diagnosis; Randomized Controlled Trials as Topic
• ISSN: 0007-1323; 1365-2168
• DOI: 10.1002/bjs.7503

Background: The feasibility of randomized controlled trials (RCTs) in liver surgery using a single‐component clinical endpoint is low as such endpoints require large sample sizes owing to their low incidence. A liver surgery‐specific composite endpoint (CEP) could solve this problem. The aim of this study was to develop a liver surgery‐specific CEP with well‐defined components. Methods: Components of a liver surgery‐specific CEP were selected based on a systematic literature search and consensus among 28 international hepatopancreatobiliary (HPB) surgeons. As an example, two prospective cohorts of patients who had undergone liver surgery in high‐volume HPB centres were used to assess the event rate and effect of implementing a liver surgery‐specific CEP. Results: Components selected for the liver surgery‐specific CEP were ascites, postresectional liver failure, bile leakage, intra‐abdominal haemorrhage, intra‐abdominal abscess and operative mortality, all with a Clavien–Dindo grade of at least 3 and occurring within 90 days after initial surgery. The incidence of this liver surgery‐specific CEP was 19·2 per cent in one cohort and 10·7 per cent in the other. These rates led to an approximately twofold reduction in the theoretical sample size required for an adequately powered RCT in liver surgery using the CEP as primary endpoint. Conclusion: The proposed liver surgery‐specific CEP consists of ascites, postresectional liver failure, bile leakage, intra‐abdominal haemorrhage, intra‐abdominal abscess and operative mortality. It has a considerably higher event rate than any of its components. Its use as the primary endpoint will increase the feasibility and comparability of RCTs in liver surgery. Copyright © 2011 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. Consensus of a feasible end point