A single centre retrospective study of the long‑term weight and pregnancy-related outcomes in women of reproductive age before and after liraglutide exposure

BACKGROUND: The impact of GLP‑1 receptor agonist exposure prior to pregnancy on pregnancy-related outcomes remains unclear, with studies reporting mixed findings. METHODS: We conducted a retrospective study evaluating bodyweight and pregnancy-related outcomes among women of reproductive age who were initiated on liraglutide for obesity management and had 2-5 years of follow-up. We compared pregnancies occurring before and after liraglutide exposure in the same women. RESULTS: A total of 556 women met the inclusion criteria of which 97 (17.4%) conceived and 459 (82.6%) did not conceive after liraglutide exposure. Baseline characteristics were comparable. After 2-5 years, the mean final weight change was significantly different between women who conceived compared to those who did not (+1.1 ± 11.3% vs. -6.7 ± 9.9%, p < 0.001). The nadir weight reduction on liraglutide in women who conceived was -6.2 ± 7.8%. Gestational weight change was similar in pre- and post-exposure pregnancies (8.0 ± 10.9% vs. 9.2 ± 11.8%, p = 0.554). Additionally, there were no differences in gestational diabetes (32.1% vs. 37.7%, p = 0.629), hypertensive disorders of pregnancy (9.2% vs. 0%, p = 0.063), need for c-section (33.3% vs. 38.9%, p = 0.549), pre-term delivery (3.7% vs. 1.9%, p = 0.500), large for gestational age (5.5% vs. 5.5%, p = 1) or small for gestational age infants (11.1% vs. 9.3%, p = 1) in pre- vs. post-exposure pregnancies. Subgroup analysis demonstrated a lower gestational weight gain with metformin use during pregnancy but was not statistically significant (5.8 ± 9.1 kg vs. 8.2 ± 9.5 kg, p = 0.369). The proportion of women with inconsistent obesity medication use during the study period was higher in those who conceived compared to those who did not (100% vs. 36.5%). CONCLUSION: Preconception liraglutide exposure, followed by discontinuation before pregnancy, was associated with weight regain but not with excess gestational weight gain in our within-individual design. Despite a small sample size, we did not observe increased adverse gestational outcomes in the post-exposure pregnancies.