Editor's Commentary - The effect of demographic and clinical factors on the relationship between BMI and PSA levels

BERKELEY, CA (UroToday.com) - Dr. Jonathan Wright and colleagues from the University of Washington report in The Prostate how confounding variables influence obesity and PSA levels.

It is reported in numerous other studies that PSA levels may be influenced by diabetes mellitus (DM), BPH, medications, race, lifestyle factors and use of calcium supplementation. This study sought to establish what factors are independently associated with PSA levels. The study cohort consisted of men without self-reported history of prostate cancer (CaP) who participated in a population-based case-control study of CaP risk factors. A final total of 787 men had complete data and PSA levels. Current medication use was recorded and finasteride users were excluded. BMI was calculated.

In the first analysis, the relationship between obesity and PSA was assessed for confounding factors. The base model included only age and BMI. The second analysis determined if variables were independently associated with PSA values and a forward stepwise algorithm was performed. Increase in PSA was associated with age, and age-adjusted PSA was associated with a family history of CaP and history of BPH. Aspirin use and combined use of aspirin and non-steroidal drugs (NSAIDs) correlated with lower geometric mean PSA levels in age-adjusted models. PSA was not significantly altered by a history of DM, NSAIDs use alone or thiazide usage. DM and statin use were both more common in overweight and obese men relative to men with a normal BMI. Statin and aspirin use increased with increasing age. Age-adjusted modeling showed that obese men had a 17% reduction in mean PSA compared to normal weight men. BPH had the greatest effect on the relationship between BMI and PSA, followed by statin use, DM and any other NSAID use. Following adjustment for these confounding factors, the relationship between BMI and PSA level was no longer significant. They found that race, family history of CaP and thiazide use did not significantly change the estimate for BMI. In the model that included all variables, the relationship between BMI and PSA was also non-significant. Variables independently associated with geometric mean PSA level were age, statin use, BPH and any NSAID or aspirin use. A history of BPH was associated with a 48% increase in the geometric mean PSA. Use of statins or any NSAID had a 15% decrease in PSA. No other variable, including BMI, was independently associated with PSA levels.

Wright JL, Lin DW, Stanford JL



Prostate. 2011 Mar 22. Epub ahead of print.

PubMed Abstract
PMID: 21432865

UroToday.com Prostate Cancer Section

email news signup