Predicting prognosis in hepatocellular carcinoma after curative surgery with common clinicopathologic parameters

BMC Cancer

Table 2 Effect of demographic and clinical parameters on survival outcome in the initial training set (N = 272)

Variable Name	Coefficient*	p-value	Variable Name	Coefficient	p-value
Poly (age,2)	-	0.58	Tumor recurrence	1.47	5.7e-8
age	-0.04	0.37	Gender, male	0.154	0.59
age²	0.0004	0.33	Family history	-0.082	0.77
Liver function parameters			Smoking
AFP [log₁₀] (ng/mL)	0.229	0.002	No	-0.464	0.13
SGPT (U/L	-0.003	0.2	Moderate	-0.385	0.24
SGOT (U/L)	0.0022	0.22	Heavy	ref	ref
BILIRUBIN (μM)	-0.011	0.4	Alcohol drinking
ALBUMIN (mg/mL)	-0.042	0.02	No	0.750	0.05
Venous infiltration	1.13	8.4e-07	Moderate	0.817	0.05
Non-tumorous liver histology		0.628	Heavy	Ref	ref
Tumor size	0.0612	0.004	No. of tumor nodule(s)	-
pTNM stage			1	ref	ref
I	-16.56	1.0	2	0.822	0.023
II	Ref	Ref	3~6	0.989	0.10
III	1.05	7.0e-05	Multiple >6	0.841	0.0016
IV	1.30	1.4e-05
AJCC stage			Child's grade, B	-0.141	0.84
I	Ref	ref
II	0.78	8.9e-03	Edmondson grade	-	0.15
III & IV	1.46	7.0e-08	HBsAg +	0.357	0.31

* The regression coefficient in the Cox proportional hazards model. The hazard rate ratio can be calculated as the exponentiation of the regression coefficient.

ISSN: 1471-2407