Skip to main content
. 2024 Apr 5;14:8034. doi: 10.1038/s41598-024-58205-5

Table 2.

Uni- and multi-variable logistic regression analyses to identify factors associated with PHLF based on preoperative data.

Variable Univariable regression Multivariable regressiona
OR (95% CI) P OR (95% CI) P
Age (years) 1.012 (0.981–1.044) 0.459
Male 0.316 (0.072–1.388) 0.127 - -
Positive HBsAg 0.548 (0.200–1.500) 0.242 - -
Positive anti-HCV 2.667 (0.469–15.157) 0.269 - -
Platelet count (109/L) 0.988 (0.982–0.995)  < 0.001
Albumin (g/L) 0.891 (0.837–0.948)  < 0.001
Total bilirubin (μmol/L) 1.092 (1.034–1.153) 0.002
AST (U/L) 1.011 (1.003–1.019) 0.005
Creatinine (μmol/L) 1.000 (0.990–1.011) 0.979 - -
Prothrombin time (sec) 1.762 (1.315–2.360)  < 0.001 1.477 (1.069–2.042) 0.018
CSPH 5.221 (2.375–11.479)  < 0.001 1.979 (0.698–5.615) 0.199
ALBI score 4.728 (2.271–9.845)  < 0.001 2.457 (1.012–5.970) 0.047
FIB-4 index 1.304 (1.151–1.478)  < 0.001 1.215 (1.043–1.415) 0.012
Tumor size (cm) 1.061 (0.975–1.154) 0.172 - -
Tumor number 0.888 (0.342–2.309) 0.808 - -
Major resection 2.364 (1.087–5.140) 0.030 2.433 (1.007–5.879) 0.048

ALBI albumin-bilirubin, AST aspartate aminotransferase, CSPH clinically significant portal hypertension, FIB-4 fibrosis-4, HBsAg Hepatitis B surface antigen, HCV Hepatitis C virus, MELD model for end-stage liver disease, PHLF post-hepatectomy liver failure.

aTo avoid collinearity, platelet count, albumin, total bilirubin, and AST were not enrolled into the multivariable regression model.