BACKGROUND: The relationship between serum prealbumin and the risk of all-cause mortality after hepatectomy in patients with hepatocellular carcinoma (HCC) needs to be evaluated. METHODS: We conducted a retrospective study. A Cox proportional hazards regression model was used to adjust for potential confounders. Prealbumin level was transformed by Z-scores and categorized into quartiles (Q1: <147 mg/L, Q2: 147-194 mg/L, Q3: 194-239 mg/L, Q4: >239 mg/L). We assessed the dose-response relationship between serum prealbumin and the risk of all-cause mortality using a restricted cubic spline model. RESULTS: Data were included from 2,022 HCC patients who underwent hepatectomy at Guangxi Medical University Cancer Hospital in China between January 2006 and January 2016. The adjusted hazard ratios (HRs) for increasing quartiles of serum prealbumin were 0.78 [95% confidence interval (CI): 0.64-0.95] for Q2, 0.66 (0.53-0.81) for Q3, and 0.51 (0.41-0.64) for Q4 in the Cox model (all P < 0.001). Serum prealbumin showed an L-shaped, non-linear dose-response relationship with the risk of all-cause mortality (P < 0.001). Among patients whose serum prealbumin was below 250 mg/L, risk of all-cause mortality decreased by 27% (95% CI: 18-36%) per increase of one standard deviation (69.8 mg/L) in serum prealbumin. CONCLUSIONS: Levels of serum prealbumin under 250 mg/L may be considered dangerous with respect to all-cause mortality after hepatectomy in HCC patients. Serum prealbumin may be useful as a prognostic marker in HCC patients undergoing hepatectomy.
BACKGROUND: The relationship between serum prealbumin and the risk of all-cause mortality after hepatectomy in patients with hepatocellular carcinoma (HCC) needs to be evaluated. METHODS: We conducted a retrospective study. A Cox proportional hazards regression model was used to adjust for potential confounders. Prealbumin level was transformed by Z-scores and categorized into quartiles (Q1: <147 mg/L, Q2: 147-194 mg/L, Q3: 194-239 mg/L, Q4: >239 mg/L). We assessed the dose-response relationship between serum prealbumin and the risk of all-cause mortality using a restricted cubic spline model. RESULTS: Data were included from 2,022 HCC patients who underwent hepatectomy at Guangxi Medical University Cancer Hospital in China between January 2006 and January 2016. The adjusted hazard ratios (HRs) for increasing quartiles of serum prealbumin were 0.78 [95% confidence interval (CI): 0.64-0.95] for Q2, 0.66 (0.53-0.81) for Q3, and 0.51 (0.41-0.64) for Q4 in the Cox model (all P < 0.001). Serum prealbumin showed an L-shaped, non-linear dose-response relationship with the risk of all-cause mortality (P < 0.001). Among patients whose serum prealbumin was below 250 mg/L, risk of all-cause mortality decreased by 27% (95% CI: 18-36%) per increase of one standard deviation (69.8 mg/L) in serum prealbumin. CONCLUSIONS: Levels of serum prealbumin under 250 mg/L may be considered dangerous with respect to all-cause mortality after hepatectomy in HCC patients. Serum prealbumin may be useful as a prognostic marker in HCC patients undergoing hepatectomy.