(BSPOT22) Associations between Dose of Prescribed Opioid Analgesic Exposure During Pregnancy and Adverse Birth Outcomes and Early Mortality: A Population-Based Study

Background:
Studies on prescribed opioid analgesic (POA) use in pregnancy have primarily focused on examining dichotomous use rather than dose-response associations. Limited research has suggested a dose-response association between prenatal POA exposure and some adverse outcomes (e.g., preterm birth [PTB]), but has not elucidated the association with mortality.

Objectives:
To quantify associations between dose of prenatal POA exposure and risk for PTB, small for gestational age (SGA), and mortality adjusted by measured covariates and using different comparison groups to address confounding by indication.

Methods: This retrospective cohort study used a population-based sample of 1,468,504 offspring, born 2007-2020 in Sweden obtained from national registers that included information on ~99% of births. Exclusion criteria included maternal opioid use disorder (OUD) and OUD medication use. POA exposure was defined with filled POA prescriptions (Rxs, ATC codes starting with N02A) in the first 2 trimesters. Cumulative oral morphine equivalent dose in mg was dichotomized as either low (≤156) or high (≥157). Outcomes were PTB ( < 37 gestational weeks), SGA (birth weight ≤ 2 standard deviations below expected fetal weight for gestational age), and mortality (stillbirth or death ≤ 27 days after birth). Covariates included a wide range of maternal, paternal, and socioeconomic characteristics. Poisson regression was used to estimate associations between low and high dose (compared to unexposed) with outcomes while adjusting for covariates. Additional comparisons included unexposed vs exposed births to those with diagnosed painful conditions and births to those with POA Rxs before but not during pregnancy.

Results:
The PTB rate was highest among offspring with high OME (8.6%), followed by low OME (6.6%) and was lowest among unexposed (4.7%). We observed a dose-response association in covariate-adjusted models (low dose relative risk [RR] 1.25, 95% CI 1.18-1.32, high dose RR 1.38, 95% CI 1.31-1.46). Associations were only slightly attenuated with additional comparison groups (e.g., before pregnancy Rx group: high RR 1.24, 95% CI 1.17-1.32). High dose only was associated with SGA (RR 1.16, 95% CI 1.06-1.27) though the association was attenuated when compared to offspring of individuals used POAs before pregnancy (RR 1.11, 95% CI 1.00-1.23). Neither low nor high dose was statistically significantly associated with mortality, though the analyses cannot rule out small associations.

Conclusions:
These analyses suggest that risk of PTB increases as amount of exposure to POAs increases. After adjusting for potential confounders, the results do not support an association between POA exposure and SGA or mortality.