• Hepatocellular carcinoma;
  • hepatic resection;
  • blood transfusion;
  • hospital mortality;
  • hospital cost


From September 1981 to December 1988, 163 patients underwent hepatic resection for hepatocellular carcinoma. The patients were divided into two groups: those operated on from September 1981 to March 1985 (n = 55) and those operated on from April 1985 to December 1988 (n = 705). There was an increase in the number of relatively small hepatocellular carcinomas in 1987–88. Differences in the incidence of accompanying liver cirrhosis (72 versus 62 per cent) were not statistically significant; however, values of the indocyanine green test (21·5 versus 77·0 per cent, P < 0·01) aided in strict patient selection. In more recent years, initial hepatic hilar dissection for control of vascular structures was undertaken and an ultrasonic dissector was used in about three-quarters of these patients. Consequently, the mean estimated blood loss (2500 versus 1300ml, P<0·001) and mean intraoperative blood replacement (2200 versus 560 ml, P < 0·001) were significantly less than in the earlier period. Among the 58 patients treated in the early period, hospital morbidity and mortality rates were 52 and 29 per cent respectively. In contrast, the rates were 23·8 and 1·9 per cent respectively among the 105 patients operated on during the recent period (P<0·01). The decline in hospital mortality is attributed to the careful selection of patients, use of modern tools, and a diminished blood loss.