Skip to main content
. Author manuscript; available in PMC: 2016 Aug 11.
Published in final edited form as: Clin Genitourin Cancer. 2014 Oct 25;13(2):e65–e72. doi: 10.1016/j.clgc.2014.08.012

TABLE 5.

Univariate and multivariate regression models of factors predicting adverse pathologic outcomes in prostate cancer patients following radical prostatectomy at University of Pennsylvania, 1990–2012.

Univariate analysis of factors predicting seminal vesicle invasion
Variables HR 95% CI p value
Age 1.06 1.02 to 1.10 0.003
African-American Race 2.34 1.37 to 3.99 0.002
Serum PSA 1.13 1.09 to 1.16 <0.001
Clinical stage 1.6 1.14 to 2.27 0.007
Clinical gleason score ≤ or > 6 9.1 5.21 to 16 <0.001
Year of prostatectomy 1.02 0.97 to 1.07 0.51
Body mass index, continuous 1.06 1.00 to 1.12 0.07

Multivariate forward stepwise model of factors predicting seminal vesicle invasion
Variables HR 95% CI p value

Age 1.06 1.01 to 1.12 0.02
Serum PSA 1.12 1.07 to 1.17 <0.001
Clinical stage 1.86 0.91 to 3.80 0.089
Clinical gleason score ≤ or > 6 5.18 2.67 to 10.1 <0.001
Year of prostatectomy 1.07 1.01 to 1.16 0.055
African-American Race 2.00 1.01 to 3.94 0.046

Multivariate forward stepwise model of factors predicting extraprostatic extension
Variables HR 95% CI p value

Serum PSA 1.12 1.08 to 1.16 <0.001
Clinical stage 1.69 1.08 to 2.64 0.02
Clinical gleason score ≤ or > 6 2.54 1.62 to 3.98 <0.001
Body mass index, continuous 2.81 1.02 to 1.10 0.005

Multivariate forward stepwise model of factors predicting positive surgical margin
Variables HR 95% CI p value

Serum PSA 1.08 1.04 to 1.11 <0.001
Clinical gleason score ≤ or > 6 1.67 1.02 to 2.73 0.04
Year of prostatectomy 1.05 1.01 to 1.09 0.024
Body mass index, continuous 1.05 1.00 to 1.09 0.025

Multivariate forward stepwise model of factors predicting Lymph node involvement
Variables HR 95% CI p value

Age 0.89 0.78 to 1.03 0.13
Serum PSA 1.2 1.07 to 1.34 0.003
Clinical gleason score ≤ or > 6 7.22 0.92 to 56.58 0.06
Year of prostatectomy 3.31 1.34 to 8.17 0.01

NOTE. Boldfaced values represent statistically significant differences between groups.

Abbreviations: PSA- Prostate-specific antigen, CI- confidence intervals,

Adverse pathologic features: seminal vesicle invasion, extraprostatic extension, positive surgical margin, and/or lymph node involvement. P values were derived from a forward-stepwise logistic regression model. Predictors of adverse pathologic features were examined using logistic regression models. Age, PSA, and year of surgery were examined as continuous variables. T-stage (T1a–c vs. T2), clinical Gleason score, and race were examined as categorical variables.