The findings suggest that hormonal changes precede the onset of RA in men and influence the phenotype of the disease, Mitra Pikwer, MD, of Lund University in Malmö, Sweden, and co-authors reported online in Annals of Rheumatic Diseases.
"Since this is the first major study of testosterone and related hormones in the preclinical phase of RA, our findings should be verified in other populations," the authors concluded.
The risk of RA involves interplay among genetic, environmental, and hormonal factors. The disease predominately affects women in the premenopausal period, but the sex difference decreases with increasing age.
Cross-sectional studies have shown low testosterone in men and women with RA as compared with people unaffected by the disease. Because proinflammatory cytokines suppress the hypothalamic-pituitary-gonadal axis, low testosterone levels might occur as a consequence of RA-associated inflammation, the authors noted in their introduction.
A single prospective study has examined the relationship between testosterone and RA in men and showed no significant associations. However, the study involved a small number of men and did not control for confounders, the authors continued.
Evaluation of other hormones, such as follicle-stimulating hormone and luteinizing hormone, is necessary to interpret differences in testosterone levels, they wrote. Measurement of sex hormone-binding globulin is required to calculate levels of free testosterone.
Given the lack of information about hormonal influences and the risk of RA in men, Pikwer and colleagues reviewed data from the Malmö Preventive Medicine Program, which involved 22,444 men born in Malmö from 1921 to 1949 and 10,902 women born from 1925 to 1938.
At enrollment, study participants had the option to provide fasting blood samples, which were frozen for future evaluation.
Investigators previously identified study participants who developed RA by the end of 2004. Pikwer and colleagues determined that 104 of the patients had contributed serum samples, and they were matched with 174 Malmö Preventive Medicine Program participants who did not develop RA and who had provided serum samples.
The median time from program screening to RA diagnosis was 12.7 years. The RF status at diagnosis or afterward could be ascertained for 83 of the cases. At enrollment in the study, the patients and control group had a mean age of 45 to 46, and age at RA diagnosis was 59 for both groups.
The RA group had a lower body mass index (BMI) at enrollment in the program. BMI had a negative correlation with testosterone and free testosterone levels (P0.001 for both) but not with other hormones. Smoking was associated with RA and with higher levels of all the hormones measured (testosterone, free testosteone, follicle-stimulating hormone, leutinizing hormone, and sex hormone-binding globulin).
The cases had lower levels of all hormones except free testosterone -- which was the same -- and sex hormone-binding globulin, which was higher in the cases.
The authors found that 22 of 83 cases had RF-negative RA.
Unadjusted analyses yielded a trend toward an inverse association between testosterone (free and total) and RF-negative RA and a significant positive association between follicle-stimulating hormone levels and RF-negative RA (13.1 IU/L versus 0.80 IU/L for all 104 cases, 95% CI 1.70 to 100).
After adjustment for BMI and smoking status, the investigators found inverse associations of both testosterone and free testosterone with RF-negative RA (OR 0.31 and OR 0.38 per standard deviation, respectively). Follicle-stimulating hormone levels had a significant positive association with RF-negative RA (OR 11.5, 95% CI 1.46 to 91.1).
Multivariate analysis showed a trend toward a negative association between testosterone levels (total and free) and any diagnosis of RA. Follicle-stimulating hormone levels had a negative association with RF-positive RA (OR 0.42, 95% CI 0.20 to 0.88).
The authors acknowledged the predominately Caucasian-Scandinavian heritage of the population that could limit the applicability of the findings to other ethnic groups or populations in other geographic areas.
The study was supported by the Swedish Research Council, the Swedish Rheumatism Association, Lund University, and the County of Skane.
The authors reported no potential conflicts of interest.
Reference: Pikwer M, et al "Association between testosterone levels and risk of future rheumatoid arthritis in men: a population-based case-control study" Ann Rheum Dis 2013; DOI: 10.1136/annrheumdis.2012.202781.