Cannabis in pregnancy: meta-review expands on surgeon general's data

  • Singh S & al.
  • BJOG
  • 23 Sep 2019

  • International Clinical Digest
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.

Takeaway

  • Prenatal cannabis use is prevalent, but declines through pregnancy.
  • Paucity of data on quantification of cannabis exposure by gestational age, frequency, recreational prenatal cannabis use/legalization.

Why this matters

  • In August 2019, the US Surgeon General issued a national warning against cannabis use during pregnancy and lactation, citing potential effects on the developing brain.
    • Supporting data were limited; metadata provide greater context with larger overall study/participant numbers.
  • Discuss potential harms of cannabis, other substances with pregnant patients.

Key results

  • 41 studies.
  • Prenatal cannabis use prevalence: 0.24% to 22.6%; especially ages
  • Cannabis use increased during the first trimester (n=1 study): from 5.7% to 12.1%, 2002-2003 to 2016-2017, respectively (difference, 6.4%; 95% CI, 2.6%-10.3%); overall data demonstrated consistent declines in second, third trimesters.
  • Concurrent alcohol in 7.6%-64.6% of cannabis users; concurrent illicit drug and tobacco use in 0.5%-61.5% and 27.4%-77.3%, respectively.
  • Neonatal outcomes:
    • Low birth weight (n=3/7): aOR, 1.5 (95% CI, 1.1-2.1) to 2.72 (95% CI, 1.67-4.43).
    • Increased ICU admissions (n=4/6): adjusted relative risk (aRR), 1.40 (95% CI, 1.36-1.44) to aOR, 2.0 (95% CI, 1.7-2.5).
    • Preterm birth (n=3/9): aRR, 1.41 (95% CI, 1.36-1.47) to aOR, 2.2 (95% CI, 1.9-2.5).
    • Small size (n=5/10): aOR, 1.30 (95% CI, 1.03-1.62) to 2.2 (95% CI, 1.8-2.7).

Study design

  • Systematic meta-review of recreational prenatal cannabis use, health outcomes in high-income countries, 2000-2019.
  • Funding: Canadian Institutes of Health Research.

Limitations

  • Limited generalizability.
  • Select neonatal, maternal outcomes.
  • Observational, residual confounding.