|
Back to 2015 Annual Meeting Program
the Impact of Chronic Liver Disease on the Risk Assessment of ACS NSQIP Mortality and Morbidity After Hepatic Resection
Victor M. Zaydfudim*, Matthew J. Kerwin, Florence E. Turrentine, Todd W. Bauer, Reid Adams, George J. Stukenborg
Department of Surgery, University of Virginia, Charlottesville, VA
Objectives: The American College of Surgeon (ACS) National Surgical Quality Improvement Program (NSQIP) aims to provide risk-adjusted metrics of mortality and morbidity to reduce potential risk and improve outcomes. The current ACS NSQIP risk adjustment model for patients who require hepatic resection does not include metrics of underlying chronic liver disease such as fibrosis or cirrhosis and relies on Current Procedural Terminology codes to adjust for extent of resection. The applicability of this limited risk adjustment for patients requiring hepatic resection is under debate. This study aims to identify additional clinically meaningful metrics of mortality and morbidity in the hepatectomy population.
Methods: The study population includes all cases of hepatic resection performed at a tertiary academic medical center between 2006 and 2013. Case specific data, including metrics of chronic liver disease and extent of resection using the Brisbane classification, were abstracted for each patient and linked with the ACS NSQIP Participant User File data, including estimated probability of mortality and morbidity. Sequential general linear models were used to estimate ACS NSQIP probabilities of mortality and morbidity adjusting for clinically meaningful covariates.
Results: 526 hepatic resections were performed during the study period. Patient population included 91 (17.3%) patients with fibrosis, 38 (7.2%) patients with cirrhosis, and 81 (15.4%) patients with steatosis/steatohepatitis; 171 (32.5%) of the patients required hemi-hepatectomy or extended hepatectomy. The mean ACS NSQIP 30-day probability (converted to proportions) of mortality was 1.7% (±2.4%) and mean probability of morbidity was 23.5% (±10.9%). There were no significant differences in mean probability of mortality for patients with fibrosis, cirrhosis, steatosis and steatohepatitis (all p>0.05). Hemi-hepatectomy or extended resections were associated with higher probability of mortality, compared to resection of <3 segments (p<0.001). Ascites and thrombocytopenia were associated with higher probability of mortality (both p≤0.038). Extent of resection and fibrosis were associated higher probability of morbidity (both p<0.001). Underlying liver cirrhosis was not associated with post-operative morbidity (p=0.059).
Conclusions: In patients selected for hepatic resection, majority of clinical metrics of chronic liver disease were not associated with differences in ACS NSQIP estimated probability of mortality. However, ascites and thrombocytopenia were associated with significant differences in ACS NSQIP risk stratification model estimated risks of post-operative mortality. Estimates of post-operative mortality and morbidity after hepatic resection can be potentially improved by including selected metrics of underlying chronic liver disease.
Back to 2015 Annual Meeting Program
|
