Page 98 - Fertility in Women with Rheumatoid Arthritis Vruchtbaarheid van vrouwen met reumatoïde artritis

Chapter 6
high AMH levels with a longer TTP, which could be explained by patients with high AMH levels more often suffering from anovulation or other characteristics of the polycystic ovary syndrome, which is known to cause subfertility.24 In our study, AMH levels as a continuous variable showed a non-signi cant shorter TTP with increasing AMH levels. Splitting the AMH levels in quintiles comparing the lowest and highest AMH groups to the middle group, did not con rm the ndings of these previous studies (data not shown).
Although AMH levels have not clearly been shown to predict a woman’s TTP, many studies have shown that AMH levels are a reliable indicator for ovarian response after stimulation for arti cial reproductive techniques, such as in vitro fertilization (IVF).40- 42. No studies on ovarian stimulation results in RA patients have been available thus far, but given the current results, patients with established RA undergoing IVF may have a lower oocyte yield than would be expected by age alone.
As we have shown previously, disease activity and the preconceptional use of NSAIDs, did have a signi cant effect on TTP, as did not having had a previous pregnancy.2 However, in the previous study also periconceptional use of prednisone had a signi cant effect on TTP, which we could not con rm in the current analysis. The estimated hazard ratio for the occurrence of pregnancy in prednisone users versus non-users in the current study appeared higher than in the previous study, also when repeating the analysis with the same covariates as in the previous study (data not shown). A possible explanation for this difference lies in the selection of patients. Where the former study included also rst trimester visits, the current study only included preconceptional visits. This selection was chosen, because AMH levels are known to be decreased during pregnancy.14-17 Comparing the baseline characteristics for our two studies, signi cantly more patients in the current study used prednisone or NSAIDs during the periconceptional period, and signi cantly less women used no anti-rheumatic drugs. Since TTP and the percentage of women without a previous pregnancy did not differ between the two studies, our current cohort does not seem to represent a less fertile selection than the previous study. Therefore, it is dif cult to give a clear explanation for the difference in effect of prednisone on TTP. The advantage of the current exclusively preconceptional cohort is that such a study probably suffers less from recall bias of TTP and preconceptional use of anti-rheumatic drugs. Furthermore, possible effects of pregnancy on disease activity and serum AMH levels are excluded in this cohort.
96