Skip to main content
. 2016 Jan 15;18(1):79–87. doi: 10.1016/j.hpb.2015.07.006

Table 2.

Univariate survival statistics – all patients

II-A. Univariate survival – Patient characteristics Median (months) 3 Yr (%) 5 Yr (%) p
Overall survival
 n = 881 26.6 41.5 27.1
Univariate demographics
Age (10 years), HR (95% CI) 1.02 (0.96, 1.09) 0.5100
Gender <0.0001
 Male 31.3 37.4 20.0
 Female 21.3 44.7 33.0
Comorbidities 0.0468
 None 30.8 44.50 31.43
 One or more 21.1 36.61 27.07



II-B. Univariate tumor characteristics Median (months) 3 Yr (%) 5 Yr (%) p
Grade <0.0001
 1–2 32.3 46.9 31.8
 3–4 21.4 33.7 19.4
Surgical margin <0.0001
 Negative 31.0 45.4 30.4
 Micro or macro positive 19.6 29.4 14.1
 Involved – not specified 15.3 27.3 14.8
Regional nodes examined 0.0108
 0 29.0 45.4 30.2
 1–3 29.1 43.0 28.1
 ≥4 21.4 30.7 17.7
Lymph nodes positive <0.0001
 0 36.5 50.5 34.7
 1–3 15.0 18.1 5.8
 4–11 15.4 10.5 0.0
Tumor size 0.0004
 <5 cm 40.2 54.1 36.3
 ≥5 cm 23.8 37.9 24.3



II-C. Univariate – operative outcome Median (months) 3 Yr (%) 5 Yr (%) p
Length of stay (10 fold increase), HR(95% CI) 2.13 (1.35, 3.36) 0.0012
Readmission 30 days 0.3494
 None 29.0 43.5 31.7
 Readmission 26.6 37.7 22.3



II-D. Univariate adjuvant therapy Median (months) 3 Yr (%) 5 Yr (%) p
Systemic treatment 0.2177
 Chemotherapy only 22.6 36.8 18.8
 Chemotherapy and radiation 27.8 44.3 22.0
 None 27.8 42.5 30.5

Missing data reported in Table 1. Kaplan–Meier curves of categorical variables compared with log rank test of homogeneity or trend test for 3 or more ordered levels. Continuous variables tested with Cox proportion hazards regression. HR, hazard ratio.