On the prediction of the histologic composition of benign prostatic hyperplasia based on clinical and MRI parameters

BACKGROUND The histologic composition of prostate adenoma seems related to the development of clinical benign prostatic hyperplasia (BPH). Therefore, a new noninvasive prediction model as an alternative for biopsies was investigated. METHODS In 19 patients, the data of a routine preoperative workup for transurethral resection (TURP) and of an additional MRI‐examination were related to the results of morphometry on TURP‐tissue. RESULTS Statistical analysis identified age of the patient and MRI‐volumetrics of the prostate adenoma as best predictors of the epithelial fraction, with a 95% confidence interval of at least 5% (range, 9–14%) (R2 = 50%). CONCLUSIONS This prediction model is sufficiently accurate to categorize a population of patients into histologic subgroups. It seems very likely that this method will be of use as an investigative tool in medical trials to provide insight into the pathogenesis of clinical BPH and into treatment strategies for the individual patient. Prostate 32:179–187, 1997. © 1997 Wiley‐Liss, Inc.