Background Bisphosphonates (BPs) are sometimes prescribed to young women in childbearing age for bone disorders or corticosteroid induced osteoporosis. However their effects on pregnancy outcome have been poorly studied.
Objectives The aim of our study was to determine pregnancy and newborn outcomes in women exposed to BPs during pregnancy.
Methods Cases and controls were collected from the CRAT database (French Reference Centre of Teratogenic Agents),an ongoing prospective follow-up of drug-exposed pregnant women. Cases were all patients that received BPs within the 6 weeks before or during pregnancy between 1987 and 2014. Since pregnancy outcomes may be affected by the underlying maternal disease two sub-groups were defined: pregnant women with inflammatory diseases treated mainly for steroids-induced osteoporosis and pregnant women with bone diseases. Likewise, two control groups were selected from the CRAT database: pregnant women with inflammatory diseases and not exposed to BPs and healthy pregnant women non-exposed to teratogenic drugs. For each case, 4 controls were assigned. Pregnancy outcomes were compared between inflammatory cases and controls, and between bone disease cases and healthy controls.
Results 39 women were identified as having be exposed to BPs during pregnancy. Median age was 33 [27;36] years and 20% of patients were primiparous. Among them, 23 had inflammatory diseases including systemic lupus erythematosus (n=5), rheumatoid arthritis (n=5), anti-phospholipid syndrome (n=1), vasculitis (n=6) or other connective/auto-immune disease (n=6) and 16 had bone diseases including osteoporosis (n=9), malignancy (n=3) and other bone disease (n=4). Characteristics are reported in table 1. Compared to their 92 matched controls, women with inflammatory diseases exposed to BPs were more likely to receive steroids (78% vs 47%, p=0.009), methotrexate (26% vs 5% p=0.008), and colchicine (17% vs 2%, p=0.014) but less likely to receive anti-malarial drugs (p=0.019). 26% of cases vs 13% of the controls had voluntary or therapeutic pregnancy termination (p=0.195, OR=2.33, IC95 [0.63; 7.94]). Among the remaining patients, the rate of live births did not differ (94% vs 80%, p=0.29, OR=3.96, IC95 [0.53; 177.94]). No difference was found on birth weightssizes or congenital malformations (9% vs 4%, p=0.345, OR 2.08 IC95 [0.18;15.7]). In women with a bone disease the rate of therapeutic or voluntary abortions tended to be higher than in controls (19% vs 3%, p=0.054). However, among the remaining women, those treated with BPs had a lower live birth rate than healthy controls (69% vs 100%, p=0.0005), partly due to 3 spontaneous abortions in the BPs group. No difference on birth weights, sizes, or congenital malformations (p=1.00) was observed. In both groups, no neonatal hypocalcemia occurred.
Conclusions Despite its limited sample size, this study is to our knowledge the largest one on pregnancy outcomes among women exposed to bisphosphonates during pregnancy, especially during the first trimester. Results are reassuring and did not show any major teratogenic effect of BPs. However, we observed a higher rate of spontaneous abortions in patients with bone disease treated with BPs, potentially related to an underlying malignancy. Collecting characteristics of child's follow-up is ongoing and will provide additional data, particularly on post-natal growth.
Disclosure of Interest None declared