Definition of at-risk patients: baseline variables.

Benign prostatic hyperplasia (BPH) represents a significant burden in ageing men due to frequently associated lower urinary tract symptoms (LUTS), which may impair their quality of life. Some men will have progressive disease, mainly characterized by symptom deterioration of > or = 4 points on the International Prostate Symptom Score, but also by the occurrence of acute urinary retention (AUR) and BPH-related surgery. Identifying those at risk of unfavourable outcomes is important to optimize their management. Community-based longitudinal studies provide excellent data on the natural history of BPH. Baseline variables such as age, severe LUTS, low peak flow rate, high postvoid residual urine volume (PVR), enlarged prostate and high serum prostate-specific antigen (PSA) levels, have been identified as risk factors for AUR and BPH-related surgery in such studies. Placebo arms of controlled studies have more limited value for assessing the natural history, due to strong selection criteria which generate a regression to the mean artefact and narrow the applicability to the general population. Nevertheless, in these controlled studies, baseline serum PSA level and to a lesser extent prostate size consistently predicted the risk of AUR and BPH-related surgery. Conversely, quantitative variables such as baseline symptom severity and peak flow rate behaved paradoxically, probably as a consequence of strict inclusion criteria, resulting regression to the mean, and 'ceiling' effects. Results from the Medical Therapy of Prostatic Symptoms study and the Alfuzosin Long-Term Efficacy and Safety Study also suggest that using the PVR in clinical practice needs to be reconsidered as a predictor of BPH progression. Further research is needed to clarify the role of chronic inflammation in the process of BPH progression.