BACKGROUND/AIMS: Hepatic failure after hepatic resection is a lethal compli
cation. Various factors affecting the occurrence of hepatic failure were ex
amined.
METHODOLOGY: The subjects were 315 patients who underwent hepatic resection
for hepatocellular carcinoma during the 11-year period between 1985 and 19
95. Univariate analyses of 14 variables were performed among living and dea
d patients after hepatic resection. With the significant prognostic variabl
es obtained in the multivariate analysis, the predicted probability of deat
h (PPD) was calculated for each patient.
RESULTS: There were 291 survivors and 24 patients with post-operative liver
failure. Among the factors showing statistical or near significance in the
univariate analysis, KICG and blood loss were disclosed to be factors inde
pendently correlating with survival. PPD was calculated for each patient ac
cording to the following equation:
PPD = 1 / Exp(1.6766 - 0.0004394 x blood loss + 16.69 x KICG) + 1
Assessing the goodness-of-fit model by Hosmer-Lemeshow test indicated the m
odel seemed to fit quite well.
CONCLUSIONS: Minimizing the blood loss during hepatic resection is importan
t to avoid post-operative liver failure. Careful hemostatic procedure is ne
cessary for patients with unexpected massive blood loss during hepatic rese
ction so as to prevent post-operative bleeding.