Table II shows the operative characteristics and
postoperative morbidities of patients who have
undergone hepatic resection. For all patients who
underwent hepatic resection, hepatectomy
consisted of an attempted curative resection. Sixty
patients underwent a bisegmentectomy or a more
extended hepatectomy (major hepatic resection)
and 92 patients underwent a segmentectomy (minor
hepatic resection). Major hepatic resection
was performed in 10 patients and minor hepatic
resection in 12 patients positive for SSI. SSI was
absent in 50 patients who had undergone major
hepatic resection and in 80 patients who had
undergone minor hepatic resection. There was no
correlation between SSI and the operation-related
parameters including operative procedure, operation
duration, and the perioperative amount of
red blood cells in mannitoleadenineephosphate
solution (RC-MAP). Intraoperative estimated blood
loss volume and the perioperative amount of fresh
frozen plasma was significantly larger in cases
positive for SSI. When blood glucose for postoperative
hyperglycaemia was controlled using the
programmed insulin method, the incidence of SSI
was significantly lower than when it was controlled
by the sliding scale insulin method (P ¼ 0.045). The
incidence of SSI was significantly higher when
postoperative bile leakage of organ/space was
detected in a patient than when it was not
(P ¼ 0.018). ROC curve analysis revealed an
optimal cut-off value of 810 mL for intraoperative
blood loss, yielding 72.7% sensitivity and 60.8%
specificity for the occurrence of SSI (P ¼ 0.001).
According to ROC curve analysis there was no
correlation between the incidence of SSI and the
amount of fresh frozen plasma (P ¼ 0.105). Patients
found negative for SSI required a significantly
shorter hospital stay duration than patients found
SSI positive (P < 0.001). There was no readmission
of patients in the present study.