Carbon monoxide levels, smoking and adverse pregnancy outcomes

Identifying pregnant smokers is crucial for cessation support and increased fetal surveillance, but some patients may not disclose their smoking. Biochemical markers like breath carbon monoxide (CO) can improve detection, yet the optimal CO threshold for predicting smoking-related risks remains unknown. Our objective was to assess the relationship between smoking, CO levels, and adverse pregnancy outcomes.

This retrospective cohort study analyzed 1 year of pregnancies (2023) in Oxford. Birthweight and adverse outcomes, small for gestational age (SGA), preterm birth (PTB), and extended perinatal mortality (EPM) were compared across CO categories and between self-reported smokers and nonsmokers with CO ≤2 and >2 ppm. Statistical analyses included changepoint analysis, one-way ANOVA, the Cochran-Armitage test for trend, binary logistic regression, and univariate linear regression.

Of 6963 pregnancies, 5041 (72.4%) had recorded CO levels. The mean gestation at birth was 39 + 6 weeks, and the mean birthweight (BW) was 3439 g (560); the mean BW centile was 54.25 (28.00). The prevalence of SGA, PTB, and EPM was 6.7%, 5.0%, and 0.6%, respectively. CO levels were >2 ppm in 11.7% of the cohort, including 5.0% of self-reported non-smokers. Changepoint analysis identified 2 ppm as the threshold for mean birthweight. Above this level, each 1 ppm increase in CO was associated with a 69.35 g decrease in BW and a 3.15-point decrease in BW centile, indicating a dose-response relationship. Above 2 ppm, the odds of adverse outcomes were significantly increased: for SGA, OR 2.05-3.37; for PTB, OR 1.50-3.21; and for EPM, OR 2.52-4.22. Compared to non-smokers with low CO, smokers with high CO had the highest risk of all outcomes. Non-smokers with high CO had increased risks, but not significantly.

At a threshold of >2 ppm, CO was associated with lower mean birthweight and higher rates of SGA and PTB, with a dose-response relationship. Universal CO testing could help identify and quantify risk in pregnancy.