Elective and emergency hepatic resection. Determinants of operative mortality and morbidity.

To determine the reasons for improved mortality and morbidity rates after major hepatic resection, five variables were analyzed retrospectively in 300 patients operated on over a 27-year period: (1) the indication for surgery, (2) the surgical approach, (3) the urgency with which surgery was performed, (4) the nature of the surgical procedure, and (5) the experience of the surgeon. The operative mortality rate decreased from 19% between 1962 and 1979 to 9.7% between 1980 and 1988 (p less than 0.05). The operative mortality rates for patients undergoing resection for benign hepatic neoplasms was 3.4%; for metastatic tumors, 6.3%; for primary hepatic malignancies, 19%; and for trauma, 33%. Fifty-seven percent of operations before 1980 were performed through a thoracoabdominal exposure as compared with 19% after 1980. Overall a thoracoabdominal exposure of the liver was associated with a 20% mortality rate as compared with 8.6% for operations with abdominal exposure of the liver (p less than 0.02). Elective operations accounted for 65% of hepatic resections before 1980, as compared with 90% after 1980, and were associated with an 8.8% mortality rate as compared with 30.7% for urgent and emergency operations (p less than 0.001). Segmental and wedge resections were associated with a 5.3% mortality rate as compared with 14.7% for major hepatic resections (p less than 0.05), but this difference did not affect overall operative mortality rates because there was no change in the proportion of major hepatic resections after 1980. Surgical experience was not a determinant of operative mortality or morbidity rates in elective operations. Although there was no reduction in the complication rate after 1980, there was a reduction in postoperative stay from 26 days before 1980 to 16 days after 1980 (p less than 0.001). A reduction in the incidence of postoperative sepsis and a change in its management was associated with improved operative mortality rates.