Hepatic resection in the elderly.

Abstract

From 1986 to 1995 a total of 97 patients > 65 years of age underwent hepatic resections at the Department of General Surgery, Hospital Lainz, Vienna, Austria. The population consisted of 39 men and 58 women with a mean age of 74.0 +/- 5.5 years. Primary neoplasia of the liver was the cause of resection in 35 patients, gallbladder cancer in 16 patients, and metastatic disease to the liver (due to colorectal cancer in 70%) in 40 patients. The rate of major resections (> or = 3 liver segments) was 96% for primary neoplasia of the liver, 70% for metastatic disease to the liver, and 50% for gallbladder cancer; the associated mortality rates were 23%, 2.5%, and 25%, respectively. The magnitude of the resection had a significant influence on survival for gallbladder cancer (p = 0.02) and for primary neoplasia of the liver (p = 0.002) but not for metastatic disease to the liver. This reflects the high rate of cirrhosis in hepatocellular and cholangiocellular carcinoma (88%) and gallbladder cancer (37.5%). Both pre- and postoperative severe liver dysfunction had a significantly higher risk for postoperative mortality and morbidity, which showed an incremental risk with age. Another organ system able to predict outcome at the beginning of treatment by its moderate severe dysfunction were the lungs. Overall, only right and extended right lobectomies carried a significantly higher risk for postoperative mortality and morbidity. Postoperative complications were recorded in 43% of our patients, with infection the most frequent problem in nearly all of these patients (95%). Pneumonia was the leading complication associated patient survival. All patients who developed pneumonia as a late complication during a complicated postoperative course died postoperatively. The postoperative Goris score of the patients who died was 6.9 +/- 2.9 (range 3-11), whereas the surviving patients' score averaged 2.2 +/- 1.9 (range 0-9), which was significantly different (p = 0.0003). None of the 54 patients with a GORIS score < or = 2 died postoperatively, whereas 5 of 6 patients with a score > or = 9 died (p = 0.0001). Severe liver dysfunction rather than the extent of resection influences clinical mortality. Patients > 80 years of age with a preoperative severe liver dysfunction showed a postoperative mortality of 57%, and all of these patients developed postoperative complications. Therefore resection cannot be recommended for those patients. Cirrhosis led to an unacceptable mortality of 44% after hepatic resection of > or = 5 liver segments for primary neoplasia of the liver. Major resections cannot be recommended in the aged with gallbladder cancer because 50% of the patients died after such operations. Overall, only resection of > or = 5 liver segments with segments I to III or less remaining were found to pose a major risk for clinical mortality and morbidity, but the cause of death was preexisting liver dysfunction and cirrhosis in all of these patients. Major resections of large neoplasia of the liver can be recommended even in the aged, but a preoperative preselection of patients with respect to liver function and pulmonary function preoperatively may help lower the postoperative morbidity and mortality, especially in patients who will undergo resection of > or = 5 liver segments. Major hepatic resection for metastatic disease to the liver in the elderly carries no additional survival risk. Patients > 65 years of age and especially those > 80 years of age are more liable to succumb to postoperative organ failure and complications, especially infections.