Preterm birth and caesarean risk are associated with antirheumatic drugs but it could come down to type and timing, according to a large cohort study.
Pregnant women taking drugs for psoriatic arthritis are at more risk of preterm birth and caesarean delivery, especially if they’re taking biologics.
A large Swedish study, published in Arthritis & Rheumatology, highlighted the correlation between antirheumatic treatment and increased risks, but suggested that it was not necessarily the drugs that were the cause.
“The increased risk of adverse outcomes in pregnancies with maternal antirheumatic treatment is probably attributed to disease severity rather than an effect of the medication itself,” researchers said.
In the register-based cohort study, over 900 pregnant women with PsA were identified over a ten-year period. They were matched 1:10 to over 9000 non-PsA pregnancies, factoring in maternal age, year and parity. As a proxy for disease severity, the presence, type and timing of antirheumatic treatment was tracked.
The findings confirmed current understanding that psoriatic arthritis (PsA) is associated with poorer pregnancy outcomes. Preterm birth in PsA vs. non-PsA-pregnancies attracted an adjusted odds ratio (aOR) of 1.69. Caesarean delivery risk for PsA pregnancies was aOR 1.77 for elective caesareans and aOR 1.42 for emergency caesareans.
An increased risk for preterm birth differed with presence, timing and type of antirheumatic treatment.
Compared with non-PsA pregnancies, women with PsA who took antirheumatic treatment during pregnancy had double the risk for preterm birth, increasing to four-fold increased risk among women taking bDMARDs. Among the bDMARD users, the majority were exposed to TNFi as monotherapy or in combination with corticosteroids.
Risk of preterm birth was primarily increased in first pregnancies with treatment before, but not during, pregnancy.
Researchers focused on underlying disease activity as the indicator of increased risk rather than agent-specific effects.
“It may be that corticosteroid and/or csDMARD treated pregnancies have a better controlled maternal disease activity leading to more favourable outcomes compared to pregnancies that need treatment with bDMARD,” they said.
Pregnant women with PsA were also more likely to be smokers, more likely to be obese and have a history of hypertension and diabetes. Comorbidities and confounders were accounted for in the study design.
An important study limitation was the absence of disease activity measures such as DAS28CRP or HAQ. Instead, the researchers considered the use of bDMARDs as proxy for higher disease activity, potentially limiting the interpretation of results. The authors also noted that poor adherence to dispensed medication may have misclassified pregnancies as exposed to antirheumatic medication when they were not.
Given their increased risk of pre-term birth, the authors recommended that special attention be paid to first pregnancies. They also advised pre-pregnancy counselling and individual monitoring for all pregnant women with PsA.
Arthritis Rheumatol 2021, online October 20
