Background/Purpose:   While rheumatoid arthritis (RA) has historically improved in pregnancy, recent studies suggest the improvement may not be dramatic now that we are able to control the disease better outside of pregnancy. TNF inhibitors are now routinely continued during pregnancy for women with inflammatory bowel disease (IBD) because active IBD is associated with pregnancy morbidity.  The link between RA activity and pregnancy outcomes, however, is less clear.  We sought to explore the role that RA activity and medications play in pregnancy outcomes. 

Methods:   Pregnancies in women with RA from a prospective registry were reviewed to determine the extent that RA activity and medications in the 1^st and 2^nd trimesters impacted pregnancy outcomes. Disease activity was divided into 2 levels based on the worst DAS-CRP3 and/or physician’s global assessment in the first 24 weeks of pregnancy.  Chi-square and non-parametric tests were used for univariate analysis.   A general estimating equation was used in multivariate analysis to account for multiple pregnancies in some women. 

Results: A total of 31 pregnancies in 25 women with RA or JIA were enrolled in the registry before 24 weeks gestation.  Seven pregnancies were in women with JIA and 24 with adult-onset RA.  Two ended with first trimester miscarriages; both in women with low RA activity without prednisone or TNF inhibitor exposure, but one took methotrexate in pregnancy.  Of the remaining 29 pregnancies, 15 (51.7%) had RA that was either mild or in remission throughout the 1^st and 2^nd trimesters and 14 (48.3%) had RA that was moderately to severely active during this period.  6 of 29 (20.7%) live births had poor outcomes: 4 with preterm delivery, 1 with preeclampsia, and 1 with preterm preeclampsia. 

The rate of Sulfasalazine (SSZ), hydroxychloroquine (HCQ), and TNF inhibitor use was not statistically different for women with low vs high RA activity.  Women with high RA activity, however, were more likely to take prednisone (57.1% vs 13.3%, p=0.02).

Significantly more women with preterm birth and/or preeclampsia had moderate/severe RA in early pregnancy (see table).  While not statistically significant, more pregnancies with poor outcomes were exposed to prednisone and fewer to TNF inhibitors early in pregnancy.   SSZ and HCQ were not associated with pregnancy outcomes.  Methotrexate was associated with preterm birth. 

A logistic regression model demonstrated that lower RA activity and use of a TNF inhibitor in the 1^st and 2^nd trimesters were associated with term birth without preeclampsia.  Taking prednisone in the first half of pregnancy did not appear to impact pregnancy outcomes.  

Conclusion: In this era of treat-to-target management of RA, our paradigm for RA pregnancy management may need adjusting.  By controlling RA activity with medications considered relatively safe in pregnancy, we may be able to improve both the pregnancy experience and pregnancy outcomes.