In utero exposure to electronic cigarette carriers alters craniofacial morphology

Abstract

Objectives: Despite the popularity of electronic nicotine delivery systems (ENDS), there is currently a lack of regulation and consistency regarding the formulation of the e-liquids that undergo combustion in use. The two main constituents of most e-liquids are the humectants propylene glycol (PG) and glycerol (vegetable glycerin, VG). E-liquids consist of a ratio of these two components with PG utilized to increase the "throat hit" effect and VG used to increase visible vapor. As PG-based e-liquids are known to generate more carcinogenic carbonyls and increase the uptake of nicotine, many commercial products have moved toward a more VG-centric formulation to reduce potential harm. The purpose of this study was to test the hypothesis that a common VG-based formulation (30/70 PG/VG) would result in fewer negative effects on craniofacial growth compared to an evenly concentrated formulation (50/50 PG/VG) in the absence of nicotine.

Materials and methods: Adult breeder mice were utilized to generate in utero ENDS component exposed litters including free air exposure (control), 30/70 PG/VG, and 50/50 PG/VG groups. The resulting pups were assessed at postnatal day 14 for skull morphology.

Results: Data demonstrate significant reductions in body weight, facial, and cranial dimensions, where there was a significant reduction in growth for the 30/70 PG/VG exposed group. There were no significant differences found between control and 50/50 PG/VG.

Conclusions: These results suggest the overall movement to a more VG-centric ENDS formulation may not result in reduced profile for health concerns. Further, it suggests that PG/VG are not a harmless carrier and now popular nicotine-free ENDS formulation may not be considered safe for use in pregnant populations.

Fig 1. Sample size and weight measures.

A. Sample Size denoted as litters and enumeration of postnatal 14-day pups. B. Kruskal Wallis test of independent samples revealed significant differences in weight by exposure. Post-Hoc Bonferroni analyses revealed the 30/70 PG/VG pups weighed significantly less than Free Air Control, p = 0.015, and 50/50 PG/VG, p < 0.001 respectively. * = p < 0.05 compared to control; ### = p < 0.001 compared to 50/50 PG/VG. Data represents weight in grams for the 3 exposure groups. Median is indicated by dark line, box represents 25th to 75th percentile, T-Bars represent 1.5 times the height of the box. Points outside the T Bars are indicated as outliers. Note the single outlier within the 30/70 PG/VG group.

Fig 2. Change in skull form due to ENDS carrier exposure.

TOP: Lateral 3D rendering of stereotypical skulls of in utero ENDS carrier exposures. MIDDLE: Superior View 3D rendering of stereotypical skulls of in utero ENDS carrier exposures. BOTTOM: Coronal (left) and sagittal overlays of stereotypical skulls of in utero ENDS carrier exposures. White/Grey = Control; Yellow = 30/70 PG/VG; Green = 50/50 PG/VG. Coronal overlay utilized sella turcica for anatomical registration. Sagittal overlay utilized C1 for anatomical registration.

Fig 3. Change in skull form due to ENDS carrier exposure (Anterior and inferior view).

TOP: Anterior 3D rendering of stereotypical skulls of in utero ENDS carrier exposures. MIDDLE: Inferior View 3D rendering of stereotypical skulls of in utero ENDS carrier exposures. BOTTOM: Coronal Facial overlays of stereotypical skulls of in utero ENDS carrier exposures. White/Grey = Control; Yellow = 30/70 PG/VG; Green = 50/50 PG/VG. Coronal Facial overlay utilized the rostral edge of the first molar for anatomical registration.

Fig 4. Cranial cephalometric measures.

A. Kruskal Wallis test of independent samples for cranial length revealed no statistically significant differences by exposure. B. Analysis of Variance revealed significant differences in cranial width by exposure. Post-Hoc Bonferroni analyses revealed the 30/70 PG/VG pups had significantly reduced cranial width than Free Air Control, p = 0.002, and 50/50 PG/VG, p = 0.022 respectively. C. Analysis of Variance with Welch’s correction revealed significant differences in cranial height by exposure. Post-Hoc Bonferroni analyses revealed the 30/70 PG/VG pups had significantly reduced cranial height than Free Air Control, p < 0.001, and 50/50 PG/VG, p < 0.001 respectively. D. Kruskal Wallis test of independent samples for cranial base length revealed no statistically significant differences by exposure. E. Analysis of Variance with Welch’s correction for cranial base width revealed no statistically significant differences by exposure. F. Kruskal Wallis test of independent samples for craniofacial length revealed no statistically significant differences by exposure. ** = p < 0.01 and ***p < 0.001 compared to control; # = p < 0.05 and ### = p < 0.001 compared to 50/50 PG/VG. Boxplots are provided for additional context of growth variables. Data represented in millimeters for each growth variable for the 3 exposure groups. Median is indicated by dark line, box represents 25th to 75th percentile, T-Bars represent 1.5 times the height of the box. Points outside the T Bars are indicated as individual subjects statistically identified as outliers.

Fig 5. Facial cephalometric measures.

A. Kruskal Wallis test of independent samples for anterior facial width revealed statistically significant differences by exposure. Post-Hoc Bonferroni Analyses revealed 30/70 PG/VG pups had significantly reduced width compared to Free Air Control, p = 0.045. B. Analysis of Variance revealed significant differences in midfacial width by exposure. Post-Hoc Bonferroni analyses revealed the 30/70 PG/VG pups had significantly reduced cranial width than Free Air Control, p < 0.001, and 50/50 PG/VG, p = 0.008 respectively. C. Analysis of Variance revealed significant differences in posterior facial width by exposure. Post-Hoc Bonferroni analyses revealed the 30/70 PG/VG pups had significantly reduced width than Free Air Control, p < 0.001, and 50/50 PG/VG, p = 0.004 respectively. D. Kruskal Wallis test of independent samples for facial length revealed statistically significant differences by exposure. Post-Hoc Bonferroni analyses revealed the 30/70 PG/VG pups had significantly reduced facial length than Free Air Control, p = 0.004, and 50/50 PG/VG, p = 0.016 respectively. E. Kruskal Wallis test of independent samples for facial height revealed statistically significant differences by exposure. Post-Hoc Bonferroni Analyses revealed 30/70 PG/VG pups had significantly reduced width compared to Free Air Control, p = 0.006. F. Analysis of Variance revealed significant differences in nasal length by exposure. Post-Hoc Bonferroni analyses revealed the 30/70 PG/VG pups had significantly reduced width than Free Air Control, p < 0.001, and 50/50 PG/VG, p < 0.001 respectively. * = p < 0.05, ** = p < 0.01, ***p < 0.001 compared to control; and # = p < 0.05, ## = p < 0.01, ### = p < 0.001 compared to 50/50 PG/VG. Boxplots are provided for additional context of growth variables. Data represented in millimeters for each growth variable for the 3 exposure groups. Median is indicated by dark line, box represents 25th to 75th percentile, T-Bars represent 1.5 times the height of the box. Points outside the T Bars are indicated as outliers.