Associations Between Prenatal Cannabis Exposure and Birth Outcomes: Results from a Prospective Cohort Study

This prospective cohort study demonstrates that prenatal cannabis exposure is independently associated with reduced birthweight and an increased risk of being small for gestational age, with these adverse effects significantly intensified when combined with nicotine exposure.

The Gist

Imagine a pregnancy as a long journey on a boat, where the baby is the precious cargo. The goal is to arrive at the destination (birth) with the cargo as healthy and well-provisioned as possible. For a long time, scientists have worried that if the captain (the mother) smokes a certain plant called cannabis during the trip, the cargo might arrive a bit lighter or smaller than expected.

However, there's been a tricky problem: many captains who smoke this plant also smoke tobacco. It's like trying to figure out if the boat is rocking because of the wind (cannabis) or the waves (tobacco), when in reality, the captain is often dealing with both at the same time. Previous studies were like trying to guess the weather without a thermometer; they often couldn't tell which substance was doing the most damage.

The New Study: A Better Compass
This new research acts like a high-tech weather station. The scientists followed a group of 448 pregnant women, specifically looking at those who had used cannabis before or during pregnancy. But here's the clever part: instead of just asking the women, "Do you smoke?" they used a "chemical detective" (urine tests) to check for two things:

1. Cannabis markers: To see if the plant was in the system.
2. Nicotine markers (cotinine): To see if tobacco was also present.

This allowed them to separate the effects of the "wind" from the "waves."

What They Found
The results painted a clear picture of the cargo's condition:

• The Cannabis Effect: Even when they accounted for tobacco, babies exposed to cannabis during pregnancy tended to be born with slightly lower birthweights. Think of it as the boat arriving with a few fewer crates of supplies than it should have. The more often the cannabis was used (like a frequent storm), the lighter the cargo arrived.
• The "Double Trouble" Effect: The situation got much worse when cannabis and tobacco were mixed. This is like a boat caught in a hurricane and a tsunami. Babies exposed to both substances were significantly smaller and more likely to be classified as "Small for Gestational Age" (meaning they were smaller than the standard size for their age in the womb).
• The Hidden Danger: A surprising finding was that many women who said they didn't smoke tobacco still tested positive for nicotine. The researchers suspect this happened because of "blunts" (cannabis rolled in tobacco leaves) or secondhand smoke. It's like thinking you're only sailing in a light breeze, but the chemical tests reveal you're actually in a gale.

The Takeaway
The main lesson from this study is that cannabis isn't just a "harmless herb" for a developing baby; it acts as its own distinct risk factor, independent of tobacco. While mixing it with tobacco makes the problem much worse, cannabis alone still seems to weigh down the baby's growth.

The researchers are essentially saying to expectant parents and doctors: "Don't just ask about tobacco. We need to check for cannabis, and we need to understand that even without tobacco, this plant can change the course of the journey." It's a call to be more precise in our medical advice, ensuring that every baby gets the best possible start, regardless of what the captain is consuming.

Technical Summary

1. Problem Statement

Cannabis use during pregnancy is rising globally, yet professional guidelines recommend against it. While previous literature suggests an association between prenatal cannabis exposure (PreCE) and reduced birthweight, these findings have been critically limited by confounding variables, specifically concurrent nicotine use. Many prior studies relied solely on self-reporting, which often fails to capture nicotine exposure (e.g., via blunts or secondhand smoke) or accurately distinguish between cannabis-only and cannabis-plus-nicotine exposure. Consequently, the independent effect of cannabis on neonatal outcomes remains unclear.

2. Methodology

The study utilized data from the Cannabis Use During Early Life and Development (CUDDEL) study, a prospective longitudinal cohort.

• Study Population: The sample consisted of 448 pregnant women (ages 18–41), predominantly non-Hispanic Black (85.3%).
• Exposure Classification:
  • PreCE Group ($n=297$): Women with a lifetime history of cannabis use who either self-reported use or tested positive for the cannabis metabolite THC-COOH in urine at any trimester.
  • Unexposed Group ($n=151$): Women who reported no use and tested negative for THC-COOH.
• Nicotine Assessment: Crucially, the study employed urinary cotinine testing as a biochemical marker for nicotine exposure, moving beyond self-reported tobacco use.
• Statistical Analysis:
  • Models: Linear regression (for birthweight) and modified Poisson models (for Small for Gestational Age [SGA]).
  • Covariates: Adjustments were made for sociodemographic factors (Area Deprivation Index), gestational age at birth, maternal age, BMI, and urinary cotinine status.
  • Stratification: Analyses were stratified by:
    1. Cannabis use frequency (High: $\ge$ weekly vs. Low: $\le$ monthly).
    2. Nicotine co-exposure status (Cotinine-positive vs. Cotinine-negative).

3. Key Contributions

• Biochemical Validation: Unlike many prior studies, this research used objective urine toxicology (THC-COOH and cotinine) to classify exposure, significantly reducing misclassification bias.
• Disentangling Confounders: By stratifying based on cotinine status, the study isolates the specific impact of cannabis from the impact of nicotine, addressing a major gap in the existing literature.
• Frequency Analysis: The study differentiates between high-frequency and low-frequency cannabis use to determine dose-response relationships.

4. Key Results

• Overall Association: In adjusted models, PreCE was significantly associated with lower birthweight ($\beta = -0.08$, $p_{adj} = 0.041$).
• Frequency Effects:
  • High-frequency use ($\ge$ weekly) was associated with significantly lower birthweight compared to unexposed pregnancies ($\beta = -0.13$, $p_{adj} = 0.03$).
  • Low-frequency use did not show a statistically significant difference in birthweight compared to unexposed groups.
• Nicotine Interaction:
  • The most severe reduction in birthweight was observed in the PreCE + Nicotine group ($\beta = -0.20$, $p < 0.001$) compared to unexposed controls.
  • However, PreCE without nicotine (cotinine-negative) still showed birthweight reductions, indicating that cannabis acts as an independent risk factor, though nicotine exacerbates the effect.
• Small for Gestational Age (SGA): PreCE was associated with a higher likelihood of SGA ($\le$ 5th percentile). The risk was highest in the PreCE + Nicotine group compared to both unexposed and PreCE-only groups.
• Prevalence of Co-exposure: Despite exclusion criteria for self-reported tobacco use, biochemical testing revealed high rates of nicotine exposure in the PreCE group (45.5% via blunts and 36% via secondhand exposure).

5. Significance and Conclusion

This study provides robust evidence that prenatal cannabis exposure is an independent perinatal risk factor for reduced birthweight and SGA, even in the absence of concurrent nicotine use. However, the presence of nicotine (detected via cotinine) significantly intensifies these adverse outcomes.

The findings highlight two critical clinical and research implications:

1. Independent Risk: Cannabis should be counseled against during pregnancy regardless of tobacco use, as it carries its own risks for fetal growth restriction.
2. Assessment Necessity: In populations where "blunt" use (cannabis wrapped in tobacco leaf) or secondhand smoke is common, self-reporting is insufficient. Biochemical cotinine assessment is essential to accurately characterize exposure and understand the synergistic risks of dual substance use.

Note: As this is a preprint posted on medRxiv, the findings have not yet undergone peer review and should not currently be used to guide clinical practice.