Morbid prognostic features in patients with chronic liver failure undergoing nonhepatic surgery

Abstract

Background: Although the risk of portal decompression surgery is accurately predicted by objective scoring systems (Child classification and Pugh score), few useful prognostic criteria exist regarding nonhepatic surgery in patients with chronic liver failure.

Objective: To evaluate the clinical findings associated with perioperative mortality in patients with chronic liver failure undergoing nonhepatic surgery.

Design: A retrospective cohort study.

Setting: University teaching hospitals.

Patients: Forty consecutive patients with an International Classification of Diseases, Ninth Revision (ICD-9), diagnosis of chronic liver failure and one or more of the following: jaundice, cirrhosis, chronic hepatitis, or alcoholism.

Interventions: Forty operations, including 28 abdominal procedures, 2 coronary artery bypass grafts, 5 orthopedic procedures, and 5 miscellaneous procedures.

Main outcome measures: Thirty-day mortality as related to 19 preoperative clinical and laboratory variables.

Results: Eleven (28%) of the patients died within 30 days of surgery. By univariate analysis, the following variables were significantly (P < .05, pearson chi 2 test for categorical data or Mann-Whitney U test for continuous data) associated with nonsurvival: encephalopathy, congestive heart failure, the need for emergent surgery, infection, hyperbilirubinemia, international normalized ratio greater than 1.6, hypoalbuminemia, and an elevated creatinine level. By multiple logistic regression analysis, an international normalized ratio greater than 1.6 and encephalopathy were associated with a greater than 10- and 35-fold increased mortality risk, respectively. Child classification and Pugh score failed to predict 30-day mortality.

Conclusions: We identified 8 clinical and laboratory variables associated with death within 30 days in patients with chronic liver failure undergoing nonhepatic surgery. Two factors-international normalized ratio greater than 1.6 and encephalopathy-independently predicted mortality by multivariate analysis. Neither Child classification nor Pugh score was prognostically helpful. Nonhepatic surgery confers a substantial mortality risk in patients with chronic liver failure.