RESUMO
OBJECTIVES: This report investigated whether annual changes in body mass index (BMI) are associated with the opposite changes in prostate-specific antigen (PSA). Previous studies have confirmed lower PSA levels among men with higher BMI. METHODS: Normal linear mixed models were used to characterize annual PSA, BMI and the ratio of PSA to BMI profiles for 2641 men undergoing prostate cancer screening for up to 8 years as part of a San Antonio screening study. RESULTS: Among the 1898 participants (71.9%) who never received a prostate biopsy during the study and the 585 participants (22.1%) who had one or more biopsies, all negative for prostate cancer, BMI was higher for Hispanics than other racial groups, lower for older men at study entry, and increased every year during the study; and PSA and PSA/BMI ratios were higher for older men at study entry and increased each year on study (all P values<.05). Among the 158 men (6.0%) eventually diagnosed with prostate cancer, no trends in BMI were statistically significant, but PSA and PSA/BMI ratios were higher on average for older men at study entry and increased each year on study (both P values<.05). Correlations between BMI and PSA changes per year were negative but not statistically significantly different from zero. CONCLUSIONS: The individual man scrutinizing his PSA and weight year to year can expect a slight annual increase in both, but changes in PSA from one year to the next cannot be attributed to weight gain or loss.