Change in International Prostate Symptom Score, prostrate‐specific antigen and prostate volume in patients with benign prostatic hyperplasia followed longitudinally

Abstract

The natural history of benign prostatic hyperplasia (BPH) in Japan resembles Western studies in that symptoms worsen, improve or stabilize in equal proportions of patients. We sought to determine if this pattern persisted in men seeking care at a urology referral center and if worsening of symptoms was due to increase in prostate volume (PV). We reviewed the records of all BPH patients who attended the Urology Clinic of Sapporo Medical University Hospital, during December 2003 and June 2004 with the inclusion criterion that they have at least two PV and lower urinary tract symptoms measurements using the International Prostate Symptom Score (IPSS). Patients who had prostate cancer or who underwent hormone therapy or prostate surgery between the two visits were excluded. Correlation (Spearman's rank) was used to assess interrelationships among variables at baseline and follow up; the strength of association between change in IPSS and change in PV were modeled by multiple linear regression. Sixty-seven patients were eligible. Baseline PV correlated with residual urine volume (r = 0.37, P < 0.05) and prostate-specific antigen (PSA; r = 0.65, P < 0.001) but not IPSS (r =-0.16). PV increased in 46 (70%) men, remained the same in 10 and decreased in 11; in the former group, the mean prostate enlargement generally increased as baseline PV increased. In multiple linear regression models that included baseline IPSS, correlation between change in IPSS and change in PV was 0.47 (P = 0.05) based on 25 patients with measures at concurrent visits. Change in PV was also correlated with change in quality of life score (0.46, P = 0.02) but not with change in PSA (r = 0.38, P = 0.07, maximum flow rate (-0.24) or residual urine volume (-0.06). IPSS were not correlated with any laboratory measure of urinary function at baseline; however, change in IPSS was associated with change in PV. PV was also moderately correlated to PSA levels and residual urine volume at baseline.