Background/Purpose: Pregnancy is the only condition in which Rheumatoid Arthritis (RA) shows spontaneous remission. Both older and more recent studies showed improvement of disease activity, which was more pronounced and occurred earlier in pregnancy in the older studies. An explanation for the difference in improvement could be that the older studies measured RA disease activity retrospectively using self-assessed questionnaires, whereas the more recent studies measured disease activity  prospectively using a joint score. The Disease Activity Score in 28 joints (DAS28)  and the self-assessed Rheumatoid Arthritis Disease Activity Index (RADAI) are both valid tools to determine the disease activity in RA. However the influence of pregnancy on the RADAI score has not been determined. Insight into this may contribute to a better understanding of the difference in outcome between older and more recent studies. Should the RADAI prove to be a valid instrument during pregnancy, it could be used for studies on RA and pregnancy. The aim of this study is to determine the validity of the RADAI during pregnancy in patients with RA by determining the correlation and agreement of the RADAI to the DAS28.

Methods: Pregnant RA patients were visited at their home-address (once before conception, during every trimester and three times post partum). During these visits the disease activity was measured by the RADAI and the DAS28. Correlation coefficients were determined for each time point. Furthermore, patients were stratified according to three disease states (high, intermediate, low and remission combined) based upon DAS28 and RADAI, with cut off points for the RADAI that are thought to be equivalent to those used for the DAS28.

 Results: Disease activity determined by RADAI as well as by DAS28 showed a decrease during pregnancy and a flare after delivery (p< 0.01) (fig. 1).   The correlations between DAS28 and RADAI were good (pre conception rho = 0.49, all other time points 0.61 < rho < 0.71 (p<0.01). When patients were stratified according to disease states, it was shown that according to the RADAI more patients showed remission or low disease activity during first trimester. (1^st trim. RADAI 42,1%;DAS28 26.3%), whereas during the 3^rd trimester these percentages were comparable (RADAI 43.3%; DAS 41.3%). After delivery, for both the RADAI and the DAS28, highest disease activity was observed at 12 weeks after delivery. Despite this the agreement on the disease activity states between RADAI and DAS28 was low (0.28 < kappa < 0.51).

Conclusion: The RADAI and DAS28 show moderate to good correlations. The RADAI might be a useful questionnaire to determine the disease activity for pregnant patients with RA. The study also shows that the RADAI score drops earlier than the DAS28.  This might explain why earlier studies that used self-assessed questionnaires show earlier improvements as compared to recent studies.

Fig .1