HOME
Return to Table of Contents
MORBIDITY AND MORTALITY FOLLOWING CORONARY ARTERY BYPASS GRAFT SURGERY IN PATIENTS WITH CIRRHOSIS: AN ANALYSIS OF US HOSPITALIZATIONS FROM 1998 TO 2004
AAM Shaheen, GG Kaplan, RP Myers
Liver Unit, Division of Gastroenterology, University of Calgary, Calgary, Alberta
BACKGROUND: Most studies describing the outcome of cardiac surgery in patients with cirrhosis are limited by small sample sizes and referral bias. Our objective was to describe the outcomes of coronary artery bypass graft surgery (CABG) in a large population of cirrhotics using a nationally representative database.
METHODS: Patients who underwent CABG (ICD-9-CM codes 36.1-36.3) between 1998 and 2004 were identified using the Nationwide Inpatient Sample, a stratified survey of 20% of U.S. hospitals. The impact of cirrhosis (ICD-9-CM 571.2, 571.5, 571.6) on outcomes and independent predictors of mortality among cirrhotics were determined using logistic regression analyses.
RESULTS: Out of 403,094 patients who underwent CABG, 711 (0.18%) were cirrhotic. The median age was 62 years (IQR 54-69), 73% were male, and 9% had ascites or encephalopathy. CABG indications included acute MI in 19% and chronic ischemic heart disease in 76%. Compared with non-cirrhotics, cirrhotic patients were younger (P<0.0001), but had more comorbid conditions (P<0.0005) including diabetes mellitus (37% vs. 31%), pulmonary disease (30% vs. 18%), renal failure (7% vs. 3%), alcohol abuse (21% vs. 2%), and coagulopathy (30% vs. 5%; P<0.005 for all). Cirrhotic patients also had greater median length of stay (9 vs. 6 days), median hospital charges per admission ($69,697 vs. $50,122), and in-hospital mortality (17% vs. 3%; odds ratio [OR] 6.17; 95% CI 5.05-7.53) (P<0.0001 for all). After adjusting for comorbidities, independent predictors of mortality among cirrhotics included age >60 years (OR 2.52 [95% CI 1.41-4.50]), Hispanic race (vs. whites: 2.47 [1.09-5.60]), and the presence of ascites or encephalopathy (4.66 [2.34-9.29]). A risk score generated by combining these 3 variables with the indication for CABG (MI vs. non-MI), was highly predictive of in-hospital death (mortality rates for scores of 0, 1, 2, and 3: 8.6%, 14.4%, 31.4%, and 64.7%; P<0.0001).
CONCLUSION: CABG surgery among cirrhotic patients is associated with substantial in-hospital mortality (17%) particularly among older patients, Hispanics, and those with features of hepatic decompensation.