Short halt in vaping modifies cardiorespiratory parameters and urine metabolome: a randomized trial

Abstract

Propylene glycol and glycerol are e-cigarette constituents that facilitate liquid vaporization and nicotine transport. As these small hydrophilic molecules quickly cross the lung epithelium, we hypothesized that short-term cessation of vaping in regular users would completely clear aerosol deposit from the lungs and reverse vaping-induced cardiorespiratory toxicity. We aimed to assess the acute effects of vaping and their reversibility on biological/clinical cardiorespiratory parameters [serum/urine pneumoproteins, hemodynamic parameters, lung-function test and diffusing capacities, transcutaneous gas tensions (primary outcome), and skin microcirculatory blood flow]. Regular e-cigarette users were enrolled in this randomized, investigator-blinded, three-period crossover study. The periods consisted of nicotine-vaping (nicotine-session), nicotine-free vaping (nicotine-free-session), and complete cessation of vaping (stop-session), all maintained for 5 days before the session began. Multiparametric metabolomic analyses were used to verify subjects' protocol compliance. Biological/clinical cardiorespiratory parameters were assessed at the beginning of each session (baseline) and after acute vaping exposure. Compared with the nicotine- and nicotine-free-sessions, a specific metabolomic signature characterized the stop-session. Baseline serum club cell protein-16 was higher during the stop-session than the other sessions (P < 0.01), and heart rate was higher in the nicotine-session (P < 0.001). Compared with acute sham-vaping in the stop-session, acute nicotine-vaping (nicotine-session) and acute nicotine-free vaping (nicotine-free-session) slightly decreased skin oxygen tension (P < 0.05). In regular e-cigarette-users, short-term vaping cessation seemed to shift baseline urine metabolome and increased serum club cell protein-16 concentration, suggesting a decrease in lung inflammation. Additionally, acute vaping with and without nicotine decreased slightly transcutaneous oxygen tension, likely as a result of lung gas exchanges disturbances.

Keywords: electronic nicotine delivery systems; metabolomics; nicotine; pneumoproteins; transcutaneous oxygen tension.

Fig. 1.

Typical course of the three experimental periods. The periods consisted of regular nicotine e-cigarette vaping for 5 days before the experimental session (nicotine-session) (A); nicotine-free-vaping for 5 days before the experimental session (nicotine-free-session) (B); and complete cessation of e-cigarette vaping for 5 days before the experimental session (stop-session) (C). Biological/clinical cardiorespiratory parameters (serum/urine pneumoproteins, continuous hemodynamic parameters, transcutaneous gas tensions, and skin microcirculatory blood flow) were assessed at the beginning of each session and after acute vaping exposure: 10 nicotine puffs (acute nicotine-vaping in the nicotine-session) (A); 10 nicotine-free puffs (acute nicotine-free-vaping in the nicotine-free-session) (B); and 10 sham puffs (acute sham-vaping in the stop-session) (C).

Fig. 2.

Baseline values of serum nicotine (A), semiquantitative assessment of urine cotinine (COT) (B), serum propylene glycol (PG) assessed by gas chromatography with a flame ionization detector (C), area under the curve (AUC) of serum PG assessed by ¹H-NMR spectroscopy (D), AUC of urine PG (¹H-NMR spectroscopy) (E), and exhaled carbon monoxide (CO) (F). Horizontal brackets represent the P values for the comparison between baseline values in each session. A mixed-effects linear model analysis was performed with experimental sessions and time points as fixed effects and baseline values as random effects (random intercept model). **P < 0.01, ***P < 0.001. Data represent the mean ± SE.

Fig. 3.

Baseline values of peripheral pulse oximetry ($\text{S}{\text{p}}_{{\text{O}}_2}$) (A), heart rate (B), forced expiratory flow at 25% (FEF-25%) (C), and club cell protein-16 (D). Horizontal brackets represent the P values for the comparison between baseline values in each session. A mixed-effects linear model analysis was performed with experimental sessions and time points as fixed effects and baseline values as random effects (random intercept model). *P < 0.05, **P < 0.01, ***P < 0.001. Data are presented as the mean ± SE (A and B) or the median [interquartile range, P25–P75] (C and D).

Fig. 4.

Absolute changes over time in transcutaneous partial pressures of oxygen (TcpO₂) (A) and carbon dioxide (TcpCO₂) (B) after acute nicotine-vaping (nicotine-session) (red triangles), acute nicotine-free-vaping (nicotine-free-session) (black squares), and acute sham-vaping (stop-session) (blue dots). Compared with acute sham-vaping, acute nicotine- and nicotine-free-vaping were associated with slight decreases in TcpO₂ for 10 min postexposure. Compared with acute sham-vaping, acute nicotine-vaping was associated with significant decreases in TcpCO₂ that persisted for 40 min postexposure. The red asterisks represent significant P values from the comparison of acute nicotine-vaping vs. acute sham-vaping, and the black asterisks represent significant P values from the comparison of acute nicotine-free-vaping vs. acute sham-vaping. A mixed-effects linear model analysis was performed with experimental sessions and time points as fixed effects and baseline (BSL) values as random effects (random intercept model). *P < 0.05. Data are the mean ± SE.

Fig. 5.

Absolute changes over time in saturation of hemoglobin with oxygen as measured by pulse oximetry oxygen ($\text{S}{\text{p}}_{{\text{O}}_2}$) (A), end-tidal carbon dioxide (EtCO₂) (B), and respiratory rhythm (RR) (C) in case of acute nicotine-vaping (nicotine-session) (red triangles), acute nicotine-free-vaping (nicotine-free-session) (black squares), and acute sham-vaping (stop-session) (blue dots). In comparison to acute sham-vaping, neither acute nicotine-vaping nor nicotine-free-vaping modified $\text{S}{\text{p}}_{{\text{O}}_2}$, EtCO₂, or RR. A mixed-effects linear model analysis was performed with experimental sessions and time points as fixed effects and baseline (BSL) variables as random effects (random intercept model). Data are presented as the mean ± SE.

Fig. 6.

Absolute changes over time in skin microcirculatory blood flow (SkBF) (A) and cutaneous vascular conductance (CVC) (B) in case of acute nicotine-vaping (nicotine-session) (red triangles), acute nicotine-free-vaping (nicotine-free-session) (black squares), and acute sham-vaping (stop-session) (blue dots). In comparison to acute sham-vaping, neither acute nicotine-vaping nor acute nicotine-free-vaping modified SkBF and CVC. A mixed-effects linear model analysis was performed with experimental sessions and time points as fixed effects and baseline (BSL)variables as random effects (random intercept model). Data are presented as the mean ± SE. PU, perfusion unit (arbitrary unit).

Fig. 7.

Absolute changes over time in systolic blood pressure (SBP) (A), diastolic blood pressure (DBP) (B), and heart rate (HR) (C) after acute nicotine-vaping (nicotine-session) (red triangles), acute nicotine-free-vaping (nicotine-free-session) (black squares), and acute sham-vaping (stop-session) (blue dots). In comparison to acute sham-vaping, acute nicotine-vaping increased SBP, DBP, and HR for 40, 30, and 60 min postexposure, respectively. In comparison to acute nicotine-free-vaping, acute nicotine-vaping increased SBP, DBP, and HR for 30, 30, and 60 min postexposure, respectively. The red asterisks represent significant P values from the comparison of acute nicotine-vaping vs. acute sham-vaping, the black asterisks represent significant P values from the comparison of acute nicotine-free-vaping vs. acute sham-vaping, and the gold asterisks represent significant P values from the comparison of acute nicotine- and nicotine-free-vaping. A mixed-effects linear model analysis was performed with experimental sessions and time points as fixed effects and baseline (BSL) values as random effects (random intercept model). *P < 0.05, **P < 0.01, ***P < 0.001. Data represent the mean ± SE.

Fig. 8.

A–D: partial least squares discriminant analysis (PLS-DA) on baseline urine spectra. Patient clusters defined by classes: nicotine-session (red triangles), nicotine-free-session (black squares), and stop-session (blue dots). PLS-DA performed on baseline urine metabolome allowed us to split up the nicotine- and the stop-sessions (C) and the nicotine-free- and the stop-sessions (D) but not the nicotine- from the nicotine-free-session (B) or the other 3 sessions (A). A: model parameters: R²Xcum = 0.247; R²Ycum = 0.211; Q2cum = −0.00412; Hotelling T2 = 0.95, Two proposed principal components. B: model parameters: R²Xcum = 0.266; R²Ycum = 0.257; Q²cum = −0.21; Hotelling’s T2 = 0.95, Two proposed principal components. C: model parameters: R²Xcum = 0.306; R²Ycum = 0.464; Q²cum = 0.135; Hotelling’s T2 = 0.95, two proposed principal components. D: model parameters: R²Xcum = 0.253; R²Ycum = 0.511; Q²cum = 0.155; Hotelling’s T2 = 0.95, two proposed principal components.

Fig. 9.

Heatmap plot using metabolites with variables of importance (VIP) values ≥ 1. 3-Hydroxyisoval, 3 – Hydroxyisovalerate; Ach, acethylcholine; Bet, betaine; DMA, dimethylamine; Hipp, hippurate; Methylhist, π-Methylhistidine; N-PAG, N- phenylacetylglycine; PG, propylene glycol; Pyr, pyruvate; TMAO, trimethylamine oxide.