LISBON – Advanced age appears to trump disease activity or antirheumatic drug treatment as the driving force behind miscarriage in women with rheumatoid arthritis.
“The risk for miscarriage in rheumatoid arthritis is age dependent,” Dr. Jenny Brouwer said at the annual meeting of the European Society of Human Reproduction and Embryology.
She reported on 162 evaluable pregnancies in 239 Dutch women enrolled during 2002-2010 in the nationwide Pregnancy-Induced Amelioration of Rheumatoid Arthritis (PARA) study, the world’s largest prospective cohort of pregnant RA patients. They had a mean age of 32.3 years and a mean RA duration of 4.9 years. Overall, 49.4% were nulliparous and 13.6% had suffered a previous miscarriage.
Prednisone (37%), sulfasalazine (33%), and NSAIDs (29%) were most commonly used in the periconceptional period. Women eligible for PARA had to have stopped methotrexate for at least 3 months prior to attempting to conceive, a parameter included because it is an indicator for more severe disease.
Varying reports have shown no increased miscarriage rate in women with RA, while others have reported a significantly higher risk. These studies, however, were cross-sectional or retrospective in nature and lacked detailed information on preconception disease activity and antirheumatic drug use, observed Dr. Brouwer of Erasmus Medical Center, Rotterdam, the Netherlands.
Among the 162 pregnancies, 28 miscarriages occurred (17.3%), roughly half in weeks 5-9.
Reassuringly, the miscarriage rate is comparable with that in the general population, she said. Indeed, a recent study from the Nurses Health Study II also reported a miscarriage rate of 17.3%.
Disease Activity Score based on 28 joints (DAS28) was higher 6 and 12 weeks after miscarriage than before pregnancy, with 33% of women having a disease flare after miscarriage.
“This can be a sign that women who have more active disease have a higher risk for miscarriage and should be monitored more closely when their disease activity in the preconceptional period increases,” Dr. Brouwer said.
Women who miscarried were significantly older than were women with an ongoing pregnancy (mean 33.9 years vs. 32 years; P = .022). They also were more likely to be positive for anti-citrullinated protein antibodies (ACPA) (82% vs. 60%; P = .058), have higher DAS28 scores (3.92 vs. 3.59; P = .166), and have used methotrexate in the past (82% vs. 68%; P = .174).
No significant association was observed between miscarriage and corticosteroids, anti-inflammatories, or biological therapies, she said.
Because of the low frequency of miscarriages in the study, logistic regression showed only a trend toward higher miscarriage risk in association with increasing age (OR, 1.12; P = .065 per year of age increase) and ACPA positivity (OR, 2.47; P = .092), Dr. Brouwer said.
“Since the confidence interval is 0.99-1.25, there clearly seems to be a trend toward a higher risk for occurrence of miscarriage with increasing age, which is only logical since in the general population we also see an increase in miscarriage rate with increasing age,” she said in an interview. “My expectation is that, in a larger RA cohort, with a larger absolute number of miscarriages, the association will be significant.”
Follow-up revealed that, within 1 year of miscarriage, 68% of women became pregnant again, 14% stopped trying to conceive, and 11% were lost to follow-up. The live birth rate in the subsequent pregnancy was 90%. A follow-up study of these children did not find an increased rate of congenital malformations, despite the use of RA medications during pregnancy, Dr. Brouwer said.
The audience questioned whether the investigators are suggesting that rheumatoid arthritis has no role in miscarriage, observing that the study outcomes look very similar to patients without RA. Dr. Brouwer responded that there seems to be a higher sensitivity in patients with ACPA using methotrexate in the past but that this failed to reach significance.
“The results found in this study: some association with more severe disease – ACPA positivity and past methotrexate use – although not significant are a confirmation to us that extra care should be given to suppressing active disease already before actively trying to achieve a pregnancy,” she told this publication.
For more details, the full manuscript has been published (Arthritis Rheumatol. 2015;67:1738-43).
Reumafond, the Dutch Arthritis Foundation, funded the study. Dr. Brouwer reported no financial conflicts.