E. L. Godfrey2, M. L. Kueht1, A. Rana1, S. Awad3,4 1Baylor College Of Medicine,Department Of Surgery, Division Of Abdominal Transplantation And Hepatobiliary Surgery,Houston, TX, USA 2Rice University,Department Of Bioengineering,Houston, TX, USA 3Baylor College Of Medicine,Department Of Surgery,Houston, TX, USA 4Michael E. DeBakey Veterans Affairs Medical Center,Department Of Surgery/Critical Care,Houston, TX, USA
Introduction: Recently, the Sodium-Model for End-Stage Liver Disease (NaMELD) score has been shown to be a superior measure of liver disease severity in transplantation, but has not yet been applied extensively in non-transplant surgery. We aimed to analyze NaMELD scores and outcomes of cirrhotic patients that underwent emergency surgery with the hypothesis that there would be a discrete NaMELD score threshold at which outcomes would be significantly worse.
Methods: We conducted a retrospective chart review of all patients with cirrhosis who underwent emergency surgery at our institution between January 2001 and April 2013. Univariate and multivariate regression was used to identify predictors of peri-operative outcomes: 30-day mortality, peri-operative morbidity, and disposition at time of discharge (home or need for transitional care). NaMELD scores were analyzed at 1-unit increments to determine risk thresholds.
Results: 85 patients with cirrhosis underwent emergency surgery. Univariate threshold analysis identified NaMELD risk cutoffs of 19, 17, and 12 for predictors of 30-day mortality (OR=3.44), peri-operative morbidity (OR 3.08), and discharge to home (inverse relationship, OR=0.31), respectively. Multivariate analysis revealed independent predictors of intraoperative complications to be congestive heart failure (OR=11.65) and serum creatinine (OR=2.25). Independent predictors of morbidity and discharge to home were estimated blood loss in surgery (OR=1.01) and the presence of a post-operative complication (OR=0.21), respectively. When patients were grouped by NaMELD score, most complication types occur more frequently in higher scored groups.
Conclusion: Although emergency surgery in patients with cirrhosis can be life-saving, knowledge of the significant peri-operative risks should drive the discussion with the patient and family. While further study is needed to develop a definitive threshold of NaMELD scores to predict negative outcomes of surgery in cirrhotics, this analysis shows an NaMELD of 17 is associated with increased peri-operative complications, 19 with higher 30-day mortality, and 12 with increased need for transitional care after discharge.
