Can a baseline prostate specific antigen level identify men who will have lower urinary tract symptoms later in life?

Abstract

We evaluated the relationship between baseline prostate specific antigen (PSA) and subsequent lower urinary tract symptom development during 3 decades in unselected men in the Baltimore Longitudinal Study of Aging. Urinary questionnaires were used to evaluate lower urinary tract symptoms in 704 men during 3 decades. The number of repeat evaluations was 1 to 18. We divided subjects into age groups of younger than 50 and 50 to 69.9 years at the time of the first PSA evaluation. Subjects were divided into 3 PSA groups based on initial PSA below the 25th, 25th to 75th and above the 75th percentile. A mixed effects Poisson model was used to test whether there was a significant relationship between PSA grouping and symptom score with time. There was no statistically significant difference in symptom score distribution across PSA percentiles in men younger than 50 years (p = 0.87) or 50 to 69.9 years old (p = 0.59). When age was used as an independent variable in the model, there was no statistically significant relationship between baseline PSA and symptom score (p = 0.38). These data suggest that PSA is not a useful predictor of the development of lower urinary tract symptoms in unselected, asymptomatic men.