Background: The outcomes after elective surgery in patients with cirrhosis have not been well studied.
Study design: We used the Nationwide Inpatient Sample (NIS) to identify all patients undergoing elective surgery for four index operations (cholecystectomy, colectomy, abdominal aortic aneurysm repair, and coronary artery bypass grafting) from 1998 to 2005. Elixhauser comorbidity measures were used to characterize patients' disease burden. Three distinct groups were created based on severity of liver disease: patients without cirrhosis (NON-CIRR), those with cirrhosis (CIRR), and patients with cirrhosis complicated by portal hypertension (PHTN). In-hospital mortality was the primary endpoint.
Results: There were 22,569 patients with cirrhosis (of whom 4,214 had PHTN) who underwent 1 of the 4 index operations compared with approximately 2.8 million patients without cirrhosis having these operations. Patients with CIRR or PHTN were more likely to be women (49.5% versus 44.0%, p < 0.0001) and were less likely to be treated in a large hospital (62.8% versus 67.6%, p < 0.0001) than NON-CIRR patients. Length of hospital stay and total charges per hospitalization increased with severity of liver disease for all operations (p < 0.001, respectively). Adjusted mortality rates increased with increasing liver disease for each operation (cholecystectomy: CIRR hazard ratio [HR] 3.4, 95% CI 2.3 to 5.0; PHTN HR 12.3, 95% CI 7.6 to 19.9; colectomy: CIRR HR 3.7, 95% CI 2.6 to 5.2; PHTN HR 14.3, 95% CI 9.7 to 21.0; coronary artery bypass grafting: CIRR HR 8.0, 95% CI 5.0 to 13.0, PHTN HR 22.7, 95% CI 10.0 to 53.8; abdominal aortic aneurysm: CIRR HR 5.0, 95% CI 2.6 to 9.8, PHTN HR 7.8, 95% CI 2.3 to 26.5).
Conclusions: In-hospital mortality, length of stay, and total hospital charges are significantly higher after elective surgery in cirrhotic patients, even in the absence of portal hypertension. Careful decision-making about surgery in these patients is critical as the nationwide increase in hepatitis C and cirrhosis continues.