Background/Purpose: The thiopurines azathioprine (AZA) and 6-mercaptopurine (6-MP) are effective treatments for many rheumatic and autoimmune conditions. These medications have a category D classification from the US FDA. In humans, few studies have evaluated their safety in pregnancy, and most of them have focused on women with Inflammatory Bowel Disease (IBD). We performed an updated meta-analysis to study their effect on birth outcomes for any indication, not only IBD.

Methods: We searched Medline and the Cochrane database from inception to May 2013 for articles reporting birth outcomes in all women who took thiopurines within 3 months prior to conception and/or during pregnancy. Outcomes of interest were incidence of congenital abnormalities (CAs), low birth weight (LBW; <2,500 g) and/or preterm birth (PTB; <37 weeks gestation). We calculated the Relative Risk (RR) for each outcome. Data were combined using a random effects model, and heterogeneity was assessed using an I² score.

Results: Fifteen case-control studies met inclusion criteria. We could not identify any study exclusively including patients with rheumatic conditions. A total of 1184 pregnancies were exposed to thiopurines, and 5955 disease-matched pregnancies were not exposed. There were 13 studies reporting on CAs, 10 on LBW, and 15 on PTB. Analysis of the studies that reported thiopurine exposure indicated an association with increase in CAs, but did not reach statistical significance (RR 1.35; 95% CI 0.99-1.85; I² = 0%). No statistically significant difference was found for LBW (RR 1.17; 95% CI 0.63-2.18; I² = 79%). Similarly, there was no significant increase of PTB (RR 1.28; 95% CI 0.88-1.86; I² = 69%).

Conclusion: We found that thiopurine use at the time of conception and/or during pregnancy in women with autoimmune conditions is not associated with CAs, LBW or PTB. There was a positive association with increased risk of congenital malformations, but this might be due to potential confounders, such as disease activity or use of other immunosuppressants. Overall, the results are reassuring, but welcome larger prospective studies reporting more data on disease activity during pregnancy. Interestingly, although only 2 trials included patients with rheumatic diseases, it is reassuring that all 15 studies used the same doses of thiopurines that rheumatologists use in practice. These results could be helpful in discussing therapy with patients considering pregnancy.

Table 1. Characteristics of included studies

*: includes IBD, rheumatic diseases, Myasthenia Gravis, Interstitial lung disease, glomeruloneprhitis and others.

** disease-matched pregnancies were either on other immunosuppressant or on no medication.

RC: Retrospective Cohort; PC: Prospective Cohort; AZA: Azathioprine; 6-MP: 6-Mercaptopurine. IBD: Inflammatory Bowel Disease; KT: kidney Transplant; AIH: Autoimmune Hepatitis