To evaluate the relationships among age, total prostate-specific antigen level (PSA), free PSA level, and prostate volume.

A total of 656 patients complaining of lower urinary tract symptoms who attended our urology outpatient department were enrolled. The standard assessment for lower urinary tract symptoms was applied, including serum total and free PSA determinations and transabdominal prostate volume measurement. Patients with a history of transurethral surgery, prostate cancer, and conditions other than benign prostatic hyperplasia that could affect the PSA levels were excluded. A linear regression model was used to estimate the prostate volume. Receiver operating characteristic curves were constructed to evaluate the ability of serum PSA and free PSA to estimate threshold prostate volumes and to select the optimal serum PSA and free PSA cutoff values.

The linear regression model included age (P < .000), total PSA (P < .006), and free PSA (P < .000) as independent predictors of prostate volume. Consequently, an easy to use equation was developed to estimate the prostate volume. Free PSA performed better than total PSA at predicting the prostate volume. An area under the curve of 0.668 ± 0.022 at predicting prostate volume >40 cm³ with total PSA increased to 0.721 ± 0.021 with free PSA. Moreover, free PSA with a cutpoint of 0.495 ng/mL correctly estimated a prostate volume of >40 and <40 cm³ in 71% and 66% of the cases, respectively.

The prostate volume can be estimated using easily obtained serum PSA levels, and free PSA had a better performance.