Session Information
Session Type: ACR Poster Session B
Session Time: 9:00AM-11:00AM
Background/Purpose: Given the limited data on impacts of perinatal medication use, our objective was to investigate the association between conventional synthetic DMARD (csDMARD) use preconception and during pregnancy and risk of preterm deliveries.
Methods: We linked population-based health data in British Columbia, Canada from 2002 to 2012 on all physician visits, hospital admissions, and dispensed medications with a perinatal registry. Essential data from this registry included date of conception according to ultrasound-confirmed gestational age or last menstrual period. We created a pregnancy cohort of women with inflammatory arthritis using a case definition of 2 outpatient physician ICD9 codes for RA, ankylosing spondylitis, JIA, PsA, or connective tissue diseases and adult systemic vasculitides, ³ 2 months and ² 2 years apart. We grouped csDMARDs according to accepted safety profiles – antimalarials, gold, CSA, and SSZ (Group 1); and MTX, LEF, AZA, CYC, chlorambucil, penicillamine, MMF, and minocycline (Group 2) – and determined perinatal exposure over two periods: as binary use (yes/no) during the 90 days preconception and as time-dependent use during pregnancy from date of conception until delivery. We used Cox proportional hazards models to evaluate the association between csDMARDs exposure preconception and/or during pregnancy and risk of preterm delivery, defined as live births with gestational age (GA) < 37 weeks, while adjusting for baseline and time-dependent covariates. Subgroup analyses were conducted according to csDMARD groups.
Results: There were 610 pregnancies in 512 women and 5639 pregnancies in 4078 women with and without csDMARDs exposure, respectively (Table 1). There were 562 (10%) and 138 (23%) preterm deliveries in unexposed and exposed pregnancies, respectively, with mean GA at delivery at 38.5 weeks (SD 2.06) in unexposed pregnancies, 38.3 (2.46) in Group 1, and 37.1 (2.96) in Group 2. The adjusted hazard ratio (HR) for exposure to csDMARDs preconception and during pregnancy and risk of preterm delivery was 2.12 (95%CI 1.63-2.76) (Table 2). In subgroup analyses, the risk of preterm delivery was highest in those exposed to Group 2 csDMARDs during pregnancy (HR 3.92, 95%CI 1.45-10.62).
Conclusion: Our findings suggest a risk of preterm delivery associated with exposure to csDMARDs preconception and during pregnancy. This may have been impacted by disease activity and differences in inflammatory arthritis types between exposure groups. Nonetheless, with mean GA of neonates approximating term delivery, it is important to consider the risks and benefits of medication use during pregnancy.
|
Table 1. Characteristics of pregnant women with inflammatory arthritis exposed and unexposed to csDMARDs preconception and during pregnancy
|
||
|
Characteristics
|
Exposed to csDMARDs N (%)
|
Unexposed to csDMARDs N (%)
|
|
512 women, 610 pregnancies |
4078 women, 5369 pregnancies |
|
|
Current pregnancy
|
|
|
|
Age at delivery (mean (SD))
|
32 (5) |
31 (5) |
|
Multiparous
|
335 (55) |
3356 (60) |
|
Obstetrical history
|
||
|
Prior preterm delivery |
50 (8) |
308 (5) |
|
Prior spontaneous abortion |
161 (27) |
1424 (25) |
|
Prior neonatal death |
<5 |
34 (0.6) |
|
Prior stillbirth |
17 (3) |
62 (1) |
|
Prior low birth weight |
26 (4) |
162 (3) |
|
Prior congenital anomaly |
6 (1) |
50 (1) |
|
Inflammatory arthritis type
|
|
|
|
Rheumatoid arthritis |
387 (63) |
1402 (25) |
|
Psoriasis/psoriatic arthritis |
77 (13) |
3366 (60) |
|
Juvenile idiopathic arthritis |
28 (5) |
73 (1) |
|
Connective tissue diseases and adult systemic vasculitides |
280 (46) |
786 (14) |
|
Ankylosing spondylitis |
38 (6) |
378 (7) |
|
csDMARD use*
|
|
|
|
Group 1 |
499 (82) |
– |
|
Group 2 |
189 (31) |
– |
|
Other medication use#
|
|
|
|
Biologics |
26 (4) |
44 (0.8) |
|
Glucocorticoids |
211 (35) |
211 (4) |
|
Traditional NSAIDs |
158 (26) |
525 (9) |
|
COX2 NSAIDs |
45 (7) |
62 (1) |
|
Antidepressants |
100 (17) |
598 (11) |
|
Anxiolytics |
49 (8) |
295 (5) |
|
Comorbidities
|
|
|
|
Mood disorders |
32 (5) |
250 (4) |
|
Anxiety |
63 (10) |
548 (10) |
|
Asthma |
3 (0.5) |
30 (0.5) |
|
*percentages do not add to 100 as each pregnancy can be exposed to more than one category of csDMARDs #other medication use during 90 days preconception and/or during pregnancy |
||
|
Table 2. Multivariable models of csDMARD use preconception and/or during pregnancy and risk of preterm deliveries
|
|||
|
Models*
|
Exposure groups
|
Hazard ratio
|
95% Confidence Interval
|
|
Model 1 All csDMARDs
|
Preconception only vs. unexposed |
1.37 |
1.03-1.82 |
|
During pregnancy only vs. unexposed |
1.93 |
0.80-4.65 |
|
|
Preconception AND during pregnancy vs. unexposed |
2.12 |
1.63-2.76 |
|
|
Model 2 Group 1 csDMARDs
|
Preconception only vs. unexposed |
1.40 |
1.02-1.90 |
|
During pregnancy only vs. unexposed |
1.86 |
0.77-4.48 |
|
|
Preconception AND during pregnancy vs. unexposed |
1.87 |
1.40-2.49 |
|
|
Model 3 Group 2 csDMARDs
|
Preconception only vs. unexposed |
1.58 |
1.02-2.44 |
|
During pregnancy only vs. unexposed |
3.92 |
1.45-10.62 |
|
|
Preconception AND during pregnancy vs. unexposed |
2.64 |
1.75-3.98 |
|
|
*All models were adjusted for baseline covariates including maternal characteristics (maternal age, neighborhood income quintile, parity, number of visits to rheumatologist, hospitalizations, and maternal body mass index); obstetrical history (prior premature births, spontaneous abortions, low birth weight births, congenital anomalies, neonatal deaths, stillbirths, total prior hospital admissions, number of antenatal visits); use of medications (glucocorticoids, traditional NSAIDs, COX2 NSAIDs, biologic DMARDs, anxiolytics, antidepressants); and comorbidities (asthma, depression, anxiety). All models were also adjusted for time-dependent covariates during pregnancy, namely use of medications (traditional NSAIDs, COX2 NSAIDs, biologic DMARDs, anxiolytics, antidepressants) |
|||
To cite this abstract in AMA style:
Tsao NW, Sayre EC, Sadatsafavi M, Avina-Zubieta JA, Ensworth S, De Vera MA. Risk of Preterm Delivery Associated with Perinatal Exposure to Dmards in Women with Inflammatory Arthritis: A Population-Based Cohort Study [abstract]. Arthritis Rheumatol. 2017; 69 (suppl 10). https://acrabstracts.org/abstract/risk-of-preterm-delivery-associated-with-perinatal-exposure-to-dmards-in-women-with-inflammatory-arthritis-a-population-based-cohort-study/. Accessed .« Back to 2017 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/risk-of-preterm-delivery-associated-with-perinatal-exposure-to-dmards-in-women-with-inflammatory-arthritis-a-population-based-cohort-study/