Use of serum creatinine to predict pathologic stage and recurrence among radical prostatectomy patients.

OBJECTIVES: To assess serum creatinine as a putative marker for staging/prognosis in localized prostate cancer. Although clinical stage, tumor grade, serum prostate-specific antigen (PSA), and tumor volume assessment by biopsy positivity are established prognostic markers in prostate cancer, the need for additional serum markers is clear. In a prior neural network analysis by our group, serum creatinine appeared to improve staging and prognosis. Even though serum creatinine is one of the most common blood tests used by practicing urologists, it has not been tested rigorously as a potential staging/prognosis marker in localized prostate cancer. METHODS: The data on 409 patients who underwent radical prostatectomy at the Walter Reed Army Medical Center between 1990 and 1996 were analyzed. Logistic regression analysis was used to evaluate the ability of serum creatinine to predict the pathologic stage. The ability of creatinine to predict PSA recurrence was also assessed using Cox regression analysis. In multivariable analyses, creatinine was assessed while simultaneously controlling for race, age, prostate weight, clinical stage, Gleason (World Health Organization) grade, prostatism history, treatment of benign prostatic hyperplasia, and pretreatment PSA level. RESULTS: Creatinine ranged from 0.1 to 2.3 mg/dL (mean and median 1.1 mg/dL). The relationship of creatinine to pathologic stage was significant (P = 0.050). As the level of creatinine increased, the proportion of patients with extraprostatic disease generally decreased. In multivariable logistic regression analysis, creatinine was not a significant predictor (P = 0.270). The relationship of the creatinine level to PSA recurrence was not significant in the univariate or multivariable analysis. CONCLUSIONS: Creatinine did not provide independent information for predicting pathologic stage or disease recurrence in patients with early prostate cancer.