Vaping disrupts ventilation-perfusion matching in asymptomatic users

Abstract

Inhalation of e-cigarette’s aerosols (vaping) has the potential to disrupt pulmonary gas exchange, but the effects in asymptomatic users are unknown. We assessed ventilation-perfusion (V̇_A/Q̇) mismatch in asymptomatic e-cigarette users, using magnetic resonance imaging (MRI). We hypothesized that vaping induces V̇_A/Q̇ mismatch through alterations in both ventilation and perfusion distributions. Nine young, asymptomatic “Vapers” with >1-yr vaping history, and no history of cardiopulmonary disease, were imaged supine using proton MRI, to assess the right lung at baseline and immediately after vaping. Seven young “Controls” were imaged at baseline only. Relative dispersion (SD/means) was used to quantify the heterogeneity of the individual ventilation and perfusion distributions. V̇_A/Q̇ mismatch was quantified using the second moments of the ventilation and perfusion versus V̇_A/Q̇ ratio distributions, log scale, LogSDV̇, and LogSDQ̇, respectively, analogous to the multiple inert gas elimination technique. Spirometry was normal in both groups. Ventilation heterogeneity was similar between groups at baseline (Vapers, 0.43 ± 0.13; Controls, 0.51 ± 0.11; P = 0.13) but increased after vaping (to 0.57 ± 0.17; P = 0.03). Perfusion heterogeneity was greater (P = 0.04) in Vapers at baseline (0.53 ± 0.06) compared with Controls (0.44 ± 0.10) but decreased after vaping (to 0.42 ± 0.07; P = 0.005). Vapers had greater (P = 0.01) V̇_A/Q̇ mismatch at baseline compared with Controls (LogSDQ̇ = 0.61 ± 0.12 vs. 0.43 ± 0.12), which was increased after vaping (LogSDQ̇ = 0.73 ± 0.16; P = 0.03). V̇_A/Q̇ mismatch is greater in Vapers and worsens after vaping. This suggests subclinical alterations in lung function not detected by spirometry.

NEW & NOTEWORTHY This research provides evidence of vaping-induced disruptions in ventilation-perfusion matching in young, healthy, asymptomatic adults with normal spirometry who habitually vape. The changes in ventilation and perfusion distributions, both at baseline and acutely after vaping, and the potential implications on hypoxic vasoconstriction are particularly relevant in understanding the pathogenesis of vaping-induced dysfunction. Our imaging-based approach provides evidence of potential subclinical alterations in lung function below thresholds of detection using spirometry.

INTRODUCTION

E-cigarette use, or vaping, is increasingly common, with reports of current use ranging from 0.2% to 27% of the adult populations in Europe (1). This is significantly higher in adolescents and young adults, with nearly one in five 18- to 29-yr-olds and >29% of high school seniors reporting regular use in the United States (2). Despite this increasing prevalence, the health effects of vaping are poorly understood. Some national advisory committees suggest that vaping is likely to be lower risk than traditional combustible cigarettes (3), partly due to the absence of the toxic byproducts of combustion. However, the Forum of International Respiratory Societies (FIRS), composed of nine scientifically based organizations who promote respiratory health, has not supported e-cigarettes as a means of harm reduction because of potential adverse effects on health (4) and the skyrocketing rates of use in adolescent and young adult never-smokers (5).

The recent outbreak of e-cigarette or vaping product use-associated lung injury (EVALI) has focused public attention on potential risks of vaping on the lung, including inflammation, pneumonia, and tissue damage (6), but these are not well understood. The Centers for Disease Control and Prevention has identified associations between vaping tetrahydrocannabinol oils adulterated with Vitamin E acetate and the development of EVALI (7, 8). However, aside from this severe injury, there is little information on the effects of commercially available e-cigarette aerosols on lung function in humans, particularly in asymptomatic individuals. Studies using traditional pulmonary function tests to investigate acute effects on lung function have been largely inconclusive (9, 10), whereas population studies have shown that e-cigarette use may be associated with increased risk of developing respiratory disease (2, 11).

Vaping has the potential to affect lung function, pulmonary gas exchange, and ventilation-perfusion matching by affecting both ventilation and perfusion. Nicotine within e-liquids acutely increases heart rate, systolic blood pressure, and cardiac output (12), and may alter the distribution of blood flow to the pulmonary capillary beds. Additives and contaminants in e-liquids, such as propylene glycol, flavoring agents, and heavy metals, as well as thermal byproducts, such as formaldehyde, can directly affect the airways, leading to inflammation (13) and increased airway resistance (10), potentially altering ventilation. Assessing alterations in ventilation-perfusion matching may thus allow the identification of early-stage changes in lung function that are not detected using spirometry (14).

We hypothesized that vaping acutely disrupts ventilation-perfusion matching by altering both ventilation and perfusion distributions in the lung. To test this, we used proton magnetic resonance imaging (MRI) techniques developed in our laboratory to quantitatively assess the distribution of ventilation (15, 16), perfusion (17–19), and ventilation-perfusion mismatch (20, 21). Estimates of ventilation-perfusion mismatch measured by proton MRI have shown an excellent relationship with those derived from the multiple inert gas elimination technique (21), which is arguably the gold standard to assess ventilation-perfusion relationships. In this study, we evaluated a population of asymptomatic young e-cigarette users, with a history of vaping for at least 1 yr and actively vaping daily. Using proton MRI, we evaluated these subjects after a 6+ h abstinence and then acutely after a single ad libitum vaping session. We also compared baseline ventilation-perfusion mismatch in these individuals to a population of healthy normal control subjects without any history of vaping, smoking, or exposure to other inhalants.

METHODS

Study Population

Study protocols were approved by the Institutional Review Board at the University of California, San Diego (IRB:190844). Subjects provided written informed consent before participation in the study. Nine subjects (six males and three females) with a daily use pattern, at least 1-yr vaping history, and no previously diagnosed cardiovascular or pulmonary disease (Vapers) were recruited for this study. They abstained from any form of tobacco, including e-cigarettes and other inhaled substances, for a minimum of 6 h before their study visit. Seven (four males and 3 females) nonsmoking, nonvaping, healthy normal subjects (Controls) were controls. Spirometry was performed in both groups before imaging. The Vapers were imaged twice, once at baseline and then immediately after vaping using their personal devices, whereas Controls were imaged only at baseline.

All subjects vaped using commercially available nicotinic products and reported using multiple brands and compositions. To quantify nicotine intake, we computed a pack-year-equivalent metric, comparable to the pack-year metric commonly used to define smoking history. The pack-year-equivalent was defined as follows:

$$\mathrm{P}\mathrm{a}\mathrm{c}\mathrm{k}\text{-}\mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r}\text{-}\mathrm{e}\mathrm{q}\mathrm{u}\mathrm{i}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{ }=\mathrm{ }\frac{\mathrm{n}\mathrm{i}\mathrm{c}\mathrm{o}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{ }\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{ }\left(\frac{\mathrm{m}\mathrm{g}}{\mathrm{m}\mathrm{L}}\right)\times \mathrm{d}\mathrm{a}\mathrm{i}\mathrm{l}\mathrm{y}\mathrm{ }\mathrm{u}\mathrm{s}\mathrm{e}\mathrm{ }\left(\frac{\mathrm{m}\mathrm{L}}{\mathrm{d}\mathrm{a}\mathrm{y}}\right)\times \mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{s}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{v}\mathrm{a}\mathrm{p}\mathrm{i}\mathrm{n}\mathrm{g}}{\mathrm{A}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{ }\mathrm{n}\mathrm{i}\mathrm{c}\mathrm{o}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{ }\mathrm{o}\mathrm{n}\mathrm{e}\mathrm{ }\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{k}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{c}\mathrm{i}\mathrm{g}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{t}\mathrm{t}\mathrm{e}\mathrm{s}\mathrm{ }(\sim 30\mathrm{ }\mathrm{m}\mathrm{g})}$$

Oxygen saturation and heart rate were monitored during imaging with pulse oximetry (7500FO, Nonin, MN). Spirometry was not repeated after vaping to minimize the time between vaping and imaging.

Imaging Protocols

Imaging was performed on a 1.5-Tesla Signa HDx TwinSpeed Proton MRI system (GE Medical Systems, Milwaukee, WI). Subjects were fitted with a face mask (7600 series Oro-nasal Mask, Hans-Rudolph) and imaged supine while wearing an eight-element torso coil (GE 8 channel cardiac array coil, GE Medical Systems). A flow-bypass device (22) was used to supply medical oxygen required for the specific ventilation imaging protocols (23), at a flow rate of ∼120 L/min so as to exceed the maximal inspiratory flow rate. Silicone phantoms provided reference standards for absolute quantification of perfusion and lung density (24). Image acquisition was limited to the right lung. Data were acquired at functional residual capacity (resting expiratory lung volume) in four nonoverlapping sagittal slices of 15 mm thickness, covering ∼80% of the right lung, starting in the medial lung adjacent to the heart and progressing laterally. Density and perfusion images were acquired before ventilation images to prevent any confounding effects of the 100% oxygen, required for specific ventilation imaging, on the perfusion measurements (25). The location of the spine was used as a reference to landmark the slice locations for the baseline and postvaping scans in the Vapers. Proton density images were collected using both the body coil built into the scanner and the torso coil, allowing for construction of a coil sensitivity profile for each subject, which was used to correct heterogeneity in perfusion images, occurring based on proximity to the coil elements.

Lung Density

Lung water density was calculated in sagittal slices acquired using a fast multiecho gradient echo sequence using the body coil. This technique has previously been described (26) and validated, showing high reliability and excellent agreement with gravimetric measures of lung water (24). This imaging sequence obtains signal intensity information at two time points to determine the local decay rate of the signal (T2*) and back-extrapolates the signal to a time point of zero on a voxel-by-voxel basis assuming a monoexponential decay. Proton density imaging sequence parameters were as follows: repetition time (TR) = 10 ms, flip angle = 10°, slice thickness = 15 mm, field of view = 40 cm, receiver bandwidth = 125 kHz, and a full acquisition matrix of 64 × 64 (reconstructed by scanner to 256 × 256). The signal at a time point of zero is calibrated by the signal derived from the phantom (of known signal characteristics), to obtain regional lung proton (water) density in units of milliliter H₂O per cubic centimeter lung, subsequently referred to as lung density.

Pulmonary Perfusion

Regional pulmonary perfusion was assessed using two-dimensional (2-D) arterial spin labeling (ASL) with a flow-sensitive alternating inversion recovery with an extra radiofrequency pulse (FAIRER) imaging sequence and a half-Fourier acquisition single-shot turbo spin-echo (HASTE) data collection schema (17). This technique has been previously described (18) and used in multiple studies (27–29) by our group. A pair of images is acquired with a delay of 80% of one R-R interval (individually set for each subject): a spatially selective inversion pulse and a spatially nonselective inversion. Imaging sequence parameters were as follows: inversion time (TI) = 600–800 ms (based on subject’s heart rate), echo time (TE) = 21.3 ms, field of view = 40 cm, slice thickness = 15 mm. The collected image matrix size was 256 × 128 (reconstructed by scanner to 256 × 256). The two images are subtracted, and after correction for coil heterogeneity and quantification (17, 18), the difference image is reflective of the total blood delivered to the imaged lung slice in one cardiac cycle. This technique has been validated and shows excellent reliability (30, 31).

Specific Ventilation Imaging

Regional specific ventilation was measured as described in a study by Sá et al. (23) by acquiring images at functional residual capacity while the subject was breathing room air, followed by breathing 100% oxygen, in alternating 20 breath cycles with five air-oxygen cycles acquired for robustness. Oxygen supply was switched during subject expiration using the actuation of a three-way valve, to produce a stepwise change in inspired oxygen concentration before the subsequent inspiration. An additional 20 breaths were added to the first 100% oxygen cycle, to increase the ability to quantify slower equilibrating units, requiring a total of 220 breaths to complete the protocol. The concentration of oxygen alters the longitudinal relaxation time (T₁) of lung tissue. In the T₁-weighted images acquired in this protocol, a change in the concentration of inspired oxygen induces a change in the signal intensity. Following a change in inspired oxygen concentration (from room air to 100% O₂, or vice versa), the rate of change of the signal intensity is therefore reflective of the rate at which resident gas in the lung is replaced by inhaled gas. This rate of replacement is determined by specific ventilation (23). The two-dimensional T₁-weighted images required for specific ventilation imaging were acquired with an inversion recovery single-shot fast spin echo (SSFSE) sequence using an eight-element torso coil. During each breath hold, four images were consecutively acquired after a global inversion pulse, with a unique inversion recovery times of 1,100, 1,335, 1,570, and 1,805 ms (from medial to lateral). Each image had a 40-cm field of view with a 128 × 128 acquisition matrix (reconstructed to 256 × 256). The specific ventilation imaging technique shows good test-retest reliability and has shown to be validated against the multiple-breath washout technique (16).

Image Analysis

Image analysis was performed using MATLAB (Mathworks, Natick, MA). The density and specific ventilation maps were coregistered, smoothed to an effective spatial scale of ∼1 cm³, and combined, giving spatial maps of alveolar ventilation as described previously (23). Quantified perfusion maps (17) were smoothed to the same spatial scale. Relative dispersion was calculated as the ratio of standard deviation over mean for each of the ventilation and perfusion images. It should be noted that as a result of the spatial smoothing required for quantification of ventilation-perfusion ratio, the values of relative dispersion reported here for perfusion are lower than similar measures reported in previous studies (28, 31).

Coregistered ventilation and perfusion images were then combined to give ventilation-perfusion images as previously described (20, 21). Regional ventilation and perfusion were plotted as a function of the ventilation-perfusion ratio on a log scale using the 50-compartment model (20, 21, 32). The widths of these distributions, exclusive of shunt and deadspace compartments, were quantified using the second moments of the ventilation (LogSDV̇), and perfusion (LogSDQ̇) distributions plotted against ventilation-perfusion ratios in the log scale, as in the multiple inert gas elimination technique (14, 33), against which this technique has been validated (21).

Statistical Analysis

Data were compared between Controls and Vapers at baseline using unpaired t tests. In the vaping cohort, paired t tests were used to compare baseline and postvaping data. Two-tailed α threshold of 0.05 was used to determine statistical significance. The Benjamini–Hochberg procedure (34) was applied to minimize the potential error of false positives in performing multiple t tests. Considering the small number of samples and comparisons, the false discovery rate was set to 10% and P values were only reported as significant if they were less than the calculated critical value (35).

RESULTS

Subject Demographics

Subject demographics across the two groups were similar, with only height significantly different between the two groups (P = 0.05). The ratio of forced expiratory volume in the first second to the forced vital capacity (FEV₁/FVC) was greater in Vapers compared with Controls (P = 0.006), but all spirometry measurements were within normal predicted values for both groups (Table 1).

Table 1.

Subject demographics and pulmonary function tests

Parameter	Controls (n = 7)	Vapers (n = 9)	P Values
Sex ratio	4M/3F	6M/3F
Age, yr	23 (5)	21 (2)	0.12
Height, cm	167 (8)	175 (6)	0.05*
Weight, kg	69 (15)	73 (12)	0.55
BMI, kg/m2	25 (5)	24 (4)	0.74
FVC, L	4.4 (0.7)	5.0 (0.9)	0.20
FVC, % pred	97 (8)	99 (8)	0.77
FEV1, L	3.5 (0.4)	4.3 (0.9)	0.05*
FEV1, % pred	94 (6)	100 (10)	0.17
FEV1/FVC	0.80 (0.03)	0.86 (0.04)	0.006*
FEV1/FVC, % pred	95 (3)	102 (5)	0.01*

All values are presented as means (SD). BMI, body mass index; FVC, forced vital capacity; FEV₁, forced expiratory volume in the first second; % pred, percent of predicted value. P values reported are based on two-tailed unpaired t tests.

^*Significance after correcting for 10% false discovery rate in multiple comparisons using the Benjamini–Hochberg procedure.

Inhalant History

Vapers reported a vaping history ranging from 18 mo to over 4 yr, with 5–20 vaping sessions per day. Average vaping history was estimated to be 3.3 ± 2.3 pack-years equivalent. All subjects used flavored disposable e-cigarettes. During the acute exposure, one subject used an e-liquid with 6 mg/mL nicotine concentration. All other subjects used disposable devices manufactured by Puff Bar, which uses an e-liquid with nicotinic salts and is advertised to contain nicotine concentrations of 50 mg/mL.

Vapers also reported a varied history of other inhalant use. Seven subjects reported previous cigarette use (average history of 2.1 ± 1.8 pack-years) and five subjects reported infrequent hookah use. Eight out of nine subjects reported marijuana use (one reported vaping, three reported smoking, and four reported both methods of use). These use patterns are consistent with typical trends reported in population studies in this age group. No subject currently smoked cigarettes.

Metabolic Data

In the vaping cohort, there was no significant difference in oxygen saturation between baseline and postvaping measurements, but heart rate increased acutely after vaping (from 60 ± 7 beats/min to 72 ± 11 beats/min, P = 0.0005).

Ventilation

Mean alveolar ventilation, within the imaged voxels, was similar between Controls and Vapers at baseline (P = 0.21) but was significantly decreased in Vapers after vaping (P = 0.04, Table 2). Ventilation heterogeneity (Fig. 1A) as measured by the relative dispersion, was also similar between the two groups at baseline (P = 0.18) but was significantly increased after vaping (0.43 ± 0.13 at baseline vs. 0.57 ± 0.17 after vaping; P = 0.03).

Table 2.

Summary of measurements

Parameter	Controls (n = 7)	Vapers (baseline n = 9)	Vapers (post vaping n = 9)	P, Controls vs. Vapers at baseline	P, Vapers, post vs. baseline
Heart Rate (beats/min)	55 (6)	60 (7)	72 (11)	0.2	0.0005*
SpO2 (%)	98 (2)	98 (1)	99 (1)	0.9	0.1
Alveolar Ventilation
Mean, mL/min/mL	1.6 (0.6)	2.1 (0.8)	1.6 (0.6)	0.21	0.04*
Relative dispersion	0.51 (0.11)	0.43 (0.13)	0.57 (0.17)	0.18	0.03*
Perfusion
Mean, mL/min/ml	2.3 (0.4)	2.1 (0.3)	2.6 (0.6)	0.43	0.02*
Relative dispersion	0.44 (0.10)	0.53 (0.06)	0.42 (0.07)	0.04*	0.005*
LogSDV̇	0.49 (0.09)	0.67 (0.13)	0.80 (0.10)	0.01*	0.006*
LogSDQ̇	0.43 (0.12)	0.61 (0.12)	0.73 (0.16)	0.01*	0.03*

All values are presented as means (SD). LogSDV̇ and LogSDQ̇, second moments of ventilation and perfusion histograms plotted against V̇_a/Q̇ ratios in the log scale, respectively. Relative dispersion, standard deviation/mean; SpO₂, peripheral oxygen saturation measured using pulse oximetry during image acquisition; P values reported are based on two-tailed t tests.

^*Significance after correcting for 10% false discovery rate in multiple comparisons using the Benjamini–Hochberg procedure.

Figure 1.

Heterogeneity of ventilation and perfusion measured using relative dispersion, in seven normal Controls (first column) and nine Vapers, at baseline (second column) and immediately after vaping (third column). Paired data in second and third columns are connected to show changes in heterogeneity for each individual subject. Vapers at baseline had ventilation heterogeneity comparable to Controls, but elevated perfusion heterogeneity. Acutely after vaping, ventilation heterogeneity was increased, whereas perfusion heterogeneity decreased consistent with impairment of hypoxic pulmonary vasoconstriction. *Statistically significant differences after correcting for a 10% false discovery rate in multiple comparisons using the Benjamini–Hochberg procedure.

Perfusion

Mean perfusion was also similar between the two groups at baseline (P = 0.43), but in Vapers increased significantly after vaping (P = 0.02; Table 2), consistent with the increase in heart rate. Perfusion heterogeneity (Fig. 1B) was significantly elevated in Vapers at baseline (0.53 ± 0.06) compared with Controls (0.44 ± 0.10, P = 0.04), and significantly decreased after vaping (0.42 ± 0.07, P = 0.005) in this group.

Ventilation-Perfusion Matching

The spatial distribution of ventilation, perfusion, and V̇_A/Q̇ ratios in one sagittal slice from a representative subject are shown in Fig. 2. The top row corresponds to spatial maps at baseline and the bottom row represents the maps of the same variables after vaping. Figure 3 shows examples of the changes in ventilation and perfusion histograms plotted as a function of the local V̇_A/Q̇ ratio in a 50-compartment model for the same subject. The extent of ventilation-perfusion heterogeneity in the Vapers at baseline was at the upper limit of normal reported for healthy subjects (36) and Vapers had significantly greater ventilation-perfusion heterogeneity relative to Controls (Fig. 4) (LogSDV̇, 0.49 ± 0.09 in Controls vs. 0.67 ± 0.13 in Vapers, P = 0.01; LogSDQ̇, 0.43 ± 0.12 in Controls vs. 0.61 ± 0.12 in Vapers, P = 0.01). Following vaping, Vapers also had significant increases in ventilation-perfusion heterogeneity (LogSDV̇, 0.80 ± 0.10, P = 0.006; LogSDQ̇, 0.73 ± 0.16 post vaping, P = 0.03), and the extent of the disruption reached a level observed in some patients with chronic obstructive pulmonary disease (COPD) (14).

Figure 2.

Spatial maps of ventilation, perfusion, and ventilation-perfusion (V̇_A/Q̇) ratios, at baseline and after vaping, in a single sagittal slice of a representative subject. The top row corresponds to maps acquired at baseline, and the bottom row corresponds to maps acquired after vaping. Black regions within the lung field represent regions removed from analysis, including conducting airways and vessels, and regions with poor signal to noise characteristics.

Figure 3.

Sample plots of ventilation (V̇_A) and perfusion (Q̇) histograms plotted against ventilation-perfusion (V̇_A/Q̇) ratio in a 50-compartment model from the same subject shown in Fig. 2. A: the histograms of ventilation and perfusion at baseline. B: corresponds to the respective plots post vaping. There is a shift in distributions toward lower V̇_A/Q̇ ratios and an increase in the width of the distributions post vaping. The arrow in B indicates a mode of increased perfusion to regions of low V̇_A/Q̇ ratio.

Figure 4.

Ventilation-perfusion heterogeneity measured using second moments of ventilation (LogSDV̇) (A) and second moments of perfusion (LogSDQ̇) (B) in seven normal Controls and nine Vapers, at baseline and immediately after vaping. Changes in data for Vapers are shown using connected points. Both metrics were elevated in Vapers at baseline compared with Controls, and increased significantly after vaping. *Statistically significant differences after correcting for a 10% false discovery rate in multiple comparisons using the Benjamini–Hochberg procedure.

DISCUSSION

This study presents evidence that ventilation-perfusion matching is markedly abnormal at baseline in young, healthy, “asymptomatic” e-cigarette users with normal spirometry. Ventilation-perfusion matching worsens acutely after a single vaping session, to a degree of disruption seen in some patients with COPD and arises because of spatially uncorrelated changes in the regional distributions of both ventilation and perfusion. Combined, this suggests that vaping alters both ventilation and perfusion distributions in the lung and may possibly inhibit the ability of the pulmonary circulation to respond to alterations in ventilation.

Acute Effects of Vaping on Ventilation

In the present study, the heterogeneity of ventilation, as measured by the relative dispersion, increased after vaping because of both a decrease in mean and an increase in the standard deviation of alveolar ventilation. Inhalation of e-cigarette vapors has been linked to several mechanistic processes that may alter regional ventilation. In animal models, chronic exposure to e-cigarette vapor compounds is associated with increased inflammatory cytokines (37, 38), decreased mucociliary clearance, and microvascular leaks, leading to airway obstruction and constriction (39, 40), and potentially small airway collapse. Vaping, with or without nicotine, induces airway epithelial injury in human bronchial epithelial cells (41), and vaping with nicotine is associated with airway hyperreactivity, distal airspace enlargement, and tissue destruction in mice (42). These alterations can result in an increase in airway resistance and alter air flow dynamics in a dose-dependent manner, particularly in small airways exposed to the aerosols in e-cigarettes.

Although the observed decrease in mean alveolar ventilation after vaping may be attributed to either physiological effects of the inhaled aerosols or to psychosomatic effects of vaping, the overall increase in both the standard deviation and the relative dispersion of ventilation suggests an alteration in the spatial distribution of ventilation. This may arise because of variability in aerosol distribution since alveolar airspaces and small airways that are exposed to greater concentrations of inhaled aerosols may experience greater changes in airway resistance through mechanisms such as bronchoconstriction (40). Consequently, the heterogeneity of ventilation induced by vaping is subject to factors affecting the distribution of the aerosols, including but not limited to posture, airway morphometry and the pattern of inhalation (43), temperature of the heating element (44–46), composition of the liquid (47, 48), and spatial patterns of bronchoconstriction (49), which were not evaluated here.

Acute Effect of Vaping on Perfusion

We found an increase in mean perfusion, and a concomitant decrease in perfusion heterogeneity, as assessed by the relative dispersion, post vaping. This may be explained by two potential mechanisms. First, vaping with e-liquids containing nicotine is known to impact cardiovascular hemodynamics, increasing heart rate, systolic blood pressure (12), and cardiac output. In keeping with this, we saw an increase in mean pulmonary blood flow and heart rate seen in our subjects, post vaping. This increase in pulmonary perfusion may affect recruitment and dilation of any vessels not already maximally dilated, thus leading to more uniform blood flow and reduced perfusion heterogeneity. This is consistent with previous work that showed a reduction in perfusion heterogeneity as measured by the relative dispersion when light exercise was used to increase pulmonary perfusion (50). However, it is not expected that this in itself would be sufficient to cause an increase in the LogSDV̇ and LogSDQ̇.

Second, vaping may alter perfusion distribution by affecting local hypoxic vasoconstriction. There is some ventilation heterogeneity even in the normal lung. In the presence of relatively poor local ventilation and alveolar hypoxia, vasoconstriction of the vessels feeding the associated capillary beds leads to a reduction of local perfusion, with redistribution of blood flow to regions of relatively well-ventilated lung. This preserves ventilation-perfusion matching but may also contribute to some associated perfusion heterogeneity. Conversely, if any baseline hypoxic pulmonary vasoconstriction is released, a reduction in perfusion heterogeneity is expected, as was seen in the present study. Nicotine may significantly inhibit the oxygen sensing mechanisms that drive vasoconstriction (51) and impair the ability of the vasculature to respond to locally altered alveolar oxygen levels. In humans, inhaled cigarette smoke has been associated with vasodilation, consistent with a release of local vasoconstriction that may be present due to variations in local oxygen concentration (52), and thermal by-products of glycerin and glycol in e-cigarettes may also have acute effects on smooth muscle tone and vascular function in systemic circulation, even in the absence of nicotine (53). Thus, alterations in hypoxic pulmonary vasoconstriction may play a role in the observed changes in perfusion heterogeneity, although other mechanisms cannot be ruled out.

Acute Effect of Vaping on Ventilation-Perfusion Matching

In Vapers, we found an acute deterioration in ventilation-perfusion relationships after a single vaping session, as indicated by an increase in both LogSDV̇ and LogSDQ̇. We found that mean perfusion was significantly increased, whereas mean alveolar ventilation decreased. These bulk changes are not expected to affect the heterogeneity of ventilation-perfusion distributions as assessed by LogSDV̇ and LogSDQ̇ per se, but rather shift the distributions to an overall lower mean. The spatial distribution of both ventilation and perfusion are altered after vaping, as seen in Fig. 2. After vaping, there is a shift toward lower V̇_A/Q̇ ratios and an increase in the width of the distributions, as seen by the plots in Fig. 3. There is also evidence of increased blood flow to regions of low V̇_A/Q̇ ratio, as seen by the increase in the “Q” histogram for V̇_A/Q̇ ≤ 0.1 (marked by the arrow). This suggests that ventilation-perfusion mismatch is increased, at least in part, due to the increase in perfusion to regions of relatively poor ventilation.

In addition, we found that ventilation heterogeneity increased after vaping, whereas perfusion heterogeneity decreased. If the redistribution of perfusion were spatially matched to the changes in ventilation, the heterogeneity measured by relative dispersion would increase for both distributions, and ventilation-perfusion matching would be preserved. Conversely, relative dispersion can remain the same even when flow is spatially redistributed as long as mean flow and standard deviation across the whole lung are unchanged.

In our study, the combined changes in heterogeneity of ventilation and perfusion with an increase in ventilation-perfusion mismatch suggest that changes in the distribution of perfusion did not compensate for the increased ventilation heterogeneity after vaping. Despite these alterations in ventilation-perfusion matching, we did not find any significant differences in oxygen saturation measured using pulse oximetry. Previous work has shown a transient change in transcutaneous oxygen tension, with a mean decrease of ∼12 mmHg after 30 min (41). The lack of a significant change in oxygen saturation at a similar time interval after vaping is likely because these changes are small, and overall gas exchange remained on the flat part of the oxygen-hemoglobin dissociation curve. Therefore, oxygen saturation measured using pulse oximetry is likely to be a poor indicator of pulmonary gas exchange efficiency in these conditions.

Chronic Effects of Vaping

Vaping is characterized by chronic intermittent use, representing a potential pattern of repeated acute insults. Subjects participating in our study reported vaping up to 20 times daily, and some subjects reported inhaling short, quick puffs every hour throughout the day. It is possible that the time periods between the last vaping session and the time of our study (>6 h) were insufficient to allow any residual changes to resolve in our subjects. Alternately, repeated acute insults may result in chronic changes that affect pulmonary gas exchange efficiency. In asymptomatic Vapers at baseline, ventilation heterogeneity was comparable to Controls, whereas perfusion heterogeneity and ventilation-perfusion heterogeneity was significantly greater. It is possible that repeated vaping results in pulmonary vascular dysfunction leading to ventilation-perfusion mismatch before any alterations affecting the airways and in keeping with this, spirometry in the Vapers was normal in the present study. A similar pattern is observed in patients with early-stage COPD, in whom alterations to the pulmonary vasculature preceded airway abnormalities and ventilation-perfusion inequalities were greater earlier in the disease than expected from spirometry (14).

Although our baseline measurements in Vapers were made after abstinence for at least 6 h, the extent of ventilation-perfusion mismatch in these subjects was substantial. The MRI-based techniques used in the present study have previously been validated against the multiple inert gas elimination technique (MIGET) and are shown to provide comparable values for LogSDV̇ and LogSDQ̇ (21). In healthy nonsmoking subjects, the upper limit measured by MIGET for LogSDQ̇ is ∼0.6 and for LogSDV̇ is 0.65 in subjects <30 yr of age. LogSDQ̇ and LogSDV̇ increases to 0.7 and 0.73, respectively, in subjects > 60 yr of age (54). Figure 5 depicts LogSDQ̇ data acquired using MIGET across a range of ages in normal subjects, reproduced from a study by Cardús et al. (54), with the data from our MRI-based study superimposed on the same plot. At baseline, the majority of the data for Vapers are at or above the upper limits for healthy subjects of comparable age. On average, the extent of ventilation-perfusion mismatch was equivalent to or worse than in healthy subjects aged ≥60 yr (54), whereas Controls, with one exception, overlie the reference values for healthy subjects. After vaping, the extent of ventilation-perfusion mismatch in these young Vapers approached that seen in patients with mild COPD (14).

Figure 5.

Variations in second moments of perfusion (LogSDQ̇) with age. Open circles and the line of linear regression fit represent data from multiple inert gas elimination technique (MIGET) studies on healthy normal subjects, reproduced from Ref. 54. Controls, shown in grey, largely fall within the normal range for their ages. Vapers have higher mean LogSDQ̇ at baseline, in black, and most are at or near the upper limits of values expected for healthy individuals of their age.

We are unaware of any data on the acute effects of smoking on V̇_A/Q̇ matching in humans. In a mouse model of cigarette smoke exposure, chronic exposure was associated with an increase in standard deviation of V̇_A/Q̇ ratios (measured as the width of the V̇_A/Q̇ histogram) corresponding to increased heterogeneity (55). Positron emission tomography has also shown that acute exposure to cigarette smoke results in increased V̇_A/Q̇ heterogeneity in sheep after 2 h of exposure (56). Also, inhalation of wood smoke acutely increases blood flow distribution to areas of low V̇_A/Q̇ with an increase in V̇_A/Q̇ mismatch and a decreased sensitivity to inhaled nitric oxide (57) in this species. Combined, these results suggest that some of the observed physiological response patterns may be common to lung dysfunction associated with different types of exposure.

Study Considerations

Our study has several strengths. This is the first study to assess the acute impact of vaping directly on ventilation-perfusion mismatch in the lungs and to detect alterations in young asymptomatic subjects at baseline. Assessing ventilation-perfusion matching potentially provides a more sensitive approach to detect early stage changes that are below the threshold of detection using traditional tools such as spirometry. The MRI protocols and metrics of ventilation-perfusion matching used in this study have been applied in multiple research studies and have been extensively evaluated and shown to be reliable and reproducible (16, 21, 30). The noninvasive nature and lack of ionizing radiation of our imaging allows us to perform repeated image acquisitions and potentially perform longitudinal follow-ups, without the need for intravenous contrast or ionizing radiation.

There are several limitations of our study. We studied a small population of young subjects that may not be representative of the general population at large. Image acquisition was restricted to the right lung to minimize artifacts from cardiac motion. Assessing long-term impact of vaping presents a unique challenge, partly due to the continuously evolving device design, liquid composition, and the wide range of flavors and additives available in the market. Younger users are also more likely to dual-use with marijuana or traditional cigarettes, or transition between products (2). In this study, we did not control for delivery device, puff topography, brands, flavors, or nicotine concentrations. Conversely, the data presented here represents the acute changes for each subject occurring with their normal use patterns. Some of our subjects also reported a previous history of smoking traditional cigarettes and/or a current marijuana use, and these patterns of dual-use or switching between products is common among e-cigarette users (2). It is difficult to isolate the relative contribution of vaping alone (compared with other inhalant use) to the abnormalities in ventilation-perfusion matching observed at baseline, and this is a limitation of our study. Although all subjects abstained from vaping for at least 6 h before the study, some of the effects on lung function may be transient and reversible, and it is possible that a longer period of abstinence may have normalized baseline perfusion and ventilation-perfusion values in the vaping cohort. Because the half-time of the effects of vaping on ventilation and perfusion are not well known, future longitudinal studies to follow the impairment of gas exchange with vaping history and patterns of use will provide a better understanding of the chronic effects of vaping and relative risk factors compared with smoking combustible cigarettes.

Conclusions

Vaping acutely impairs ventilation-perfusion matching in the lung through an alteration of both ventilation and perfusion distributions, and potentially, an inhibition of the hypoxic pulmonary vasoconstrictive response. Even at baseline, young, asymptomatic e-cigarette users with normal spirometry have increased ventilation-perfusion mismatch, to an extent comparable to much older individuals. Our data suggest that changes in pulmonary perfusion may precede potential airway abnormalities in response to long-term vaping and/or other inhalant use. Our study suggests that the use of e-cigarettes may have significant consequences for public health, especially in young subjects who may be exposed for several years.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute (NHLBI) Grants R01HL129990 and R01HL119201.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.S.K.P., A.R.E., R.C.S., L.E.C., and S.R.H. conceived and designed research; A.S.K.P., A.R.E., R.C.S., K.M.A., and S.R.H. performed experiments; A.S.K.P., A.R.E., R.C.S., K.M.A., L.E.C.A., and S.R.H. analyzed data; A.S.K.P., A.R.E., R.C.S., L.E.C.A., and S.R.H. interpreted results of experiments; A.S.K.P. and S.R.H. prepared figures; A.S.K.P., A.R.E., R.C.S., L.E.C.A., and S.R.H. drafted manuscript; A.S.K.P., A.R.E., R.C.S., K.M.A., L.E.C.A., and S.R.H. edited and revised manuscript; A.S.K.P., A.R.E., R.C.S., K.M.A., L.E.C.A., and S.R.H. approved final version of manuscript.