Background Data suggest infertility is increased in women with rheumatoid arthritis (RA) compared to healthy women. Therefore, it is possible that diminished ovarian reserve and ovulatory dysfunction may be more common among women with RA.
Objectives To compare differences in ovarian reserve and ovulatory frequency, as well as in self-reported infertility, between women with and without RA.
Methods Women with RA aged 20–40 seen in a university clinic without a history of ovarian surgery or prior exposure to possible ovary-toxic medications were invited to participate in a cross-sectional survey. Healthy controls were women aged 20–40 without an autoimmune disease, matched for age and current use of hormonal contraceptives. Infertility was defined as a patient reporting physician-diagnosed infertility or being unable to get pregnant after 12 months of trying. Ovarian reserve was assessed by measuring anti-Müllerian hormone (AMH). In women who were not taking hormonal contraceptives, progesterone level was measured from a serum sample drawn between days 21 and 23 of the menstrual cycle. Anovulation was defined as a progesterone level <3 ng/mL. In descriptive comparisons, differences in proportions were determined by Fisher's exact test, and ANOVA determined differences in means for continuous variables. Multivariate linear models estimated the effect of RA on AMH. In women with RA, the effect of RA medication use on AMH and anovulation was explored.
Results There were 75 RA patients (83% Caucasian, 4% Hispanic, 77% with at least a college education) and 75 controls (64% Caucasian, 5% Hispanic, 88% with at least a college education). The majority of RA patients were married (60%), compared to 31% of controls. The mean age of both RA patients and controls was 32 years. Mean AMH in RA patients was 3.0 (SD: 2.6) compared to 3.9 (SD: 3.9) in controls (p-value: 0.1). In linear regression models adjusted for age, hormonal contraceptives and race (nonwhite vs. white), RA patients had a lower AMH than healthy controls (β: -1.05; 95% CI: -2.09, -0.005; p=0.05). There was no observed difference in the proportion of RA patients and controls with anovulation (19% in RA and 21% in controls). Infertility was reported by 12% of RA patients and 7% of controls (p=0.4).
Among RA patients, 81% reported having ever used methotrexate (MTX). The mean AMH for MTX users was 2.8 (SD: 2.4) compared to 4.0 (SD: 3.1) in never users (p=0.1). In linear regression models adjusted for age, hormonal contraceptives and race, RA patients who had ever taken MTX had a lower AMH than those who had never taken MTX (β: -1.49; 95% CI: -2.83, -0.15; p=0.03). However, when the cumulative dose of MTX was analyzed, there was no effect of cumulative MTX and AMH. Ever use of prednisone or NSAIDs did not appear to affect AMH levels in RA patients. Methotrexate, prednisone, and NSAIDs use did not have an observed effect on anovulation.
Conclusions In this cross-sectional study, women with RA appeared to have a lower AMH level than healthy controls, suggesting ovarian reserve may be lower in these patients. In RA patients, previous use of methotrexate was associated with lower AMH, although no dose response of cumulative methotrexate exposure was observed. We did not observe a difference in anovulation between RA patients and controls. This suggests women with RA may have reduced fertility for reasons other than anovulation.
Acknowledgements This study was funded by Pfizer.
Disclosure of Interest M. Clowse Grant/research support from: Pfizer, Janssen, Consultant for: UCB, G. McDaniel: None declared, A. Eudy: None declared