Antagonistic Effects of Smoking and Maternal Glycemia on Fetal Growth: A Retrospective Study Among 13,958 French Pregnant Women

Emmanuel Cosson ^1,2* Lionel Carbillon ³ Sopio Tatulashvili ^1,2 Hélène Bihan ² Eric Vicaut ⁴ Ines Barka ² Sara Pinto ² Imen Rezgani ² Mohamed Zerguine ² Jean-Jacques Portal ^1,4 Marion Fermaut ³ Jardena J Puder ⁵ Amélie Benbara ³

• ¹ Université Sorbonne Paris Nord and Université Paris Cité, INSERM, INRAE, CNAM, Center of Research in Epidemiology and StatisticS (CRESS), Nutritional Epidemiology Research Team (EREN), Bobigny, France
• ² AP-HP, Avicenne Hospital, Paris 13 University, Sorbonne Paris Cité, Department of Endocrinology-Diabetology-Nutrition, CRNH-IdF, CINFO, Bobigny, France
• ³ AP-HP, Jean Verdier Hospital, Paris 13 University, Sorbonne Paris Cité, Department of Obstetrics and Gynecology, Bondy, France
• ⁴ AP-HP, Unité de Recherche Clinique St-Louis-Lariboisière, Université Denis Diderot, Paris, France
• ⁵ Obstetric service, Department Woman-Mother-Child, Lausanne University Hospital, Lausanne, Switzerland

Smoking and hyperglycemia first-diagnosed during pregnancy (H1inP) have opposing effects on fetal growth. The aim of this study was to explore adverse pregnancy outcomes, particularly fetal growth, by smoking and H1inP status.We included 13,958 women from a large French dataset (2012-2018). Using multivariable regression analyses, we retrospectively evaluated the risk for large-for-gestational-age babies (LGA) and other adverse outcomes according to H1inP and smoking status in four groups: no H1inP/non-smoker (group A: n=10,454; 88.2%), no H1inP/smoker (group B: n=819; 5.9%), H1inP/non-smoker (group C: n=2,570; 18.4%), H1inP/smoker (group D: n=115; 0.8%).The rates of LGA were 8.9, 4.0, 14.6 and 8.7% in groups A, B, C and D, respectively (global anova p<0.0001; factor H1inP p=0.0003, factor smoking p=0.0002, interaction p=0.48). After adjustment for potential confounders including age, body mass index, employment, ethnicity, parity, hypertension before pregnancy, gestational weight gain and alcohol and drug consumption, H1inP was associated with a higher risk (Odds ratio 1.50 [95% confidence interval 1.30-1.74]) and smoking with a lower risk (0.35 [0.25-0.50]) for LGA. Additionally, H1inP was associated with a lower total gestational weight gain and less small-for-gestationalage babies (SGA), but higher rates of hypertensive disorders, more frequent caesarean sections and admissions in neonatal intensive care units; smoking was associated with higher rates of SGA, including severe SGA (<3 rd centile), and this despite a higher total gestational weight gain. Smoking increased the risk of hypertensive disorders only in women with H1inP.Smoking among H1inP women may mask the risk of maternal hyperglycemia for LGA babies. This may provide a false sense of security for H1inP women who smoke, particularly when 5 assessing for LGA alone, but these women still face other risks to their health such as hypertensive disorders and the health of the fetus.