Abstract
As the most common cancer in men in the United States, risk factors for prostate cancer (PCa) need to be identified. Serum prostate specific antigen (PSA) levels are used for the screening of prostate cancer due to its association with the disease. Investigations have indicated that the risk of prostate cancer determined based on PSA can be further stratified on the basis of total PSA (tPSA) and f/t PSA. Further, the red blood cell distribution width-to-albumin ratio (RAR) has recently been identified as a novel biomarker for multiple inflammatory diseases. The relationship between RAR and PSA remains unclear. Here, we intended to study the association between RAR and PSA. National Health and Nutrition Examination Surveys (NHANES) represents a cross-sectional observational study within the United States. We obtained clinical data throughout the 2003 - 2010 NHANES study period. In 41,156 NHANES men, we selected 5,992 men aged 40 years or older. Missing data were imputed using multiple imputation. The association between RAR and PSA was assessed using multivariable adjusted linear regression analysis. Variance inflation factor (VIF) values were also calculated to exclude collinearity of independent variables. The association of the threshold effects was assessed using inflection points. The effect of RAR levels on PSA was significant in 5,992 subjects after adjusting the confounders (β = 1.13, 95% CI: 0.59-1.67). The notion of a threshold level was supported by the presence of inflection point at RAR = 3.762. The effect of a 1 unit increase in the RAR was a consistently increasing function of quartile of RAR. For instance, in the highest quartile of RAR, if RAR rises by 1 unit, PSA rises by 1.36 (β = 1.36, 95% CI: 0.90-1.83), suggesting a non-linearity of the two. For example, when RAR is below 3.762, higher RAR levels seem associated with higher PSA levels. This is important for understanding the factors that may play an important role in the occurrence and development of prostate cancer. Future studies must do assessments of prostate cancer incidence within the cohorts described.