The Influence of a Mouthpiece-Based Topography Measurement Device on Electronic Cigarette User’s Plasma Nicotine Concentration, Heart Rate, and Subjective Effects Under Directed and Ad Libitum Use Conditions

Abstract

Introduction:

Electronic cigarettes e-cigarettes aerosolize a liquid solution often containing nicotine. e-cigarette nicotine delivery may be influenced by user puffing behaviors (“puff topography”). E-cigarette puff topography can be recorded using mouthpiece-based computerized systems. The present study sought to examine the extent to which these systems influence e-cigarette nicotine delivery and other e-cigarette associated acute effects under ad libitum use conditions.

Methods:

Plasma nicotine concentration, heart rate, and subjective effects were assessed in 29 experienced e-cigarette users using their preferred e-cigarette battery and liquid (≥12mg/mL nicotine) in two sessions differing only by the presence of a mouthpiece-based device. In both sessions, participants completed a directed e-cigarette use bout (10 puffs, 30-s interpuff interval) and a 90-min ad libitum bout. Puff topography was recorded in the session with the topography mouthpiece.

Results:

Plasma nicotine, heart rate, and subjective effects, aside from “Did the e-cigarette Taste Good?” were independent of topography measurement (higher mean taste ratings were observed in the no topography condition). Mean (SEM) plasma nicotine concentration following the ad libitum bout was 34.3ng/mL (4.9) in the no topography condition and 35.7ng/mL (4.3) in the topography condition. Longer puff durations, longer interpuff intervals, and larger puff volumes were observed in the ad libitum relative to the directed bout.

Conclusions:

E-cigarette use significantly increased plasma nicotine concentration and heart rate while suppressing abstinence symptoms. These effects did not differ when a topography mouthpiece was present. Future studies using ad libitum e-cigarette use bouts would facilitate understanding of e-cigarette toxicant yield.

Implications:

No prior study has examined whether mouthpiece-based topography recording devices influence e-cigarette associated nicotine delivery, heart rate, or subjective effects under ad libitum conditions or assessed ad libitum puff topography in experienced individuals using their preferred e-cigarette battery and liquid with a mouthpiece-based computerized device. E-cigarette use significantly increased plasma nicotine concentration and heart rate while suppressing abstinence symptoms. These effects did not differ when a topography mouthpiece was present. Ad libitum puff topography differed from puff topography recorded during directed puffing. These findings suggest that future studies using ad libitum use bouts would facilitate better understanding of e-cigarette toxicant yield.

Introduction

Electronic cigarettes (e-cigarettes) are a class of alternative tobacco products that are increasing in popularity worldwide.^1–4 E-cigarettes use an electrical heating element to aerosolize a solution consisting of flavorants, solvents like propylene glycol and/or vegetable glycerin, and, often, nicotine. The ability of these products to deliver nicotine to users varies considerably. For example, relative to newer e-cigarette models, older e-cigarette designs that are intended to resemble tobacco cigarettes deliver less nicotine to the blood and also are less effective at suppressing tobacco/nicotine abstinence symptoms in cigarette smokers.^5,6 Also, liquid nicotine concentration is related directly to nicotine delivery.^7,8 Furthermore, relative to experienced e-cigarette users, e-cigarette naive cigarette smokers are less effective at obtaining nicotine.⁹ These differences in nicotine delivery based on user experience may reflect differences in puffing behavior.

The measurement of puff number, duration, volume, interpuff interval (IPI), and flow rate—collectively referred to as “puff topography” measurement—has been vital to understanding toxicant exposure from a variety of tobacco products including “low-yield” cigarettes¹⁰ and waterpipe or hookah.¹¹ For e-cigarettes several studies have measured e-cigarette puff topography using a mouthpiece-based topography recording device designed originally for measuring the puff topography of tobacco cigarettes and generally found puff duration values comparable to those observed in cigarette smokers (ie, 1.8–3s).^12–15 These differing results across studies may be due to differential sensitivity across topography devices.¹⁶ However, two studies using observational methods^17,18 and two studies using a mouthpiece-based topography recording device designed to accommodate low flow rate puffs associated with e-cigarette use^8,16 revealed that experienced e-cigarette users exhibited longer puff durations and larger puff volumes relative to tobacco cigarette smokers and to e-cigarette naive cigarette smokers, providing a possible explanation for the higher plasma nicotine concentrations observed in experienced e-cigarette users.⁹

Importantly, many studies that have examined e-cigarette acute effects and puff topography have used controlled puffing conditions (ie, 10 puffs, 30-s IPI).^7,8,19,20 These standardized puffing parameters are similar to puffing parameters observed in non-abstinent tobacco cigarette smokers under ad libitum conditions,^21,22 thus enabling comparisons across products. However, standardizing e-cigarette puffing conditions in this manner may not reflect actual e-cigarette use. In addition, no study in which e-cigarette user puff topography was measured allowed use of participants’ preferred e-cigarette liquid and battery, both of which may be important to facilitating naturalistic e-cigarette topography.

The present study expands on a preliminary report¹⁶ that examined the influence of e-cigarette acute effects under directed puffing conditions only, by reporting results when e-cigarette users used their preferred liquid and battery in a directed and an ad libitum bout while assessing nicotine delivery, heart rate, and subjective effects, and also recording topography using a mouthpiece-based device designed specifically for e-cigarettes in one of the two conditions. While other reports have used 60-min ad libitum e-cigarette-use bouts,⁹ a 90-min bout was used in the present study because normal e-cigarette use patterns are not well understood and a longer use period may allow more opportunity for natural use behavior to be revealed. This study design enabled the first examination of the extent to which the presence of a mouthpiece-based topography recording device interferes with e-cigarette acute effects under ad libitum conditions.

Methods

Participants

Participants were recruited for this study, approved by Virginia Commonwealth University’s (VCU’s) institutional review board, by advertisements and word of mouth. Fifty participants provided informed consent for this study. However, only 29 experienced e-cigarette users (22 males, 24 White) were included in the final analyses: 11 withdrew prior to completing the study, five exhibited elevated blood pressure during screening, two were discontinued because venous access could not be maintained, and one displayed an elevated heart rate during screening. Two additional participants, using the same e-cigarette liquid, completed the study but their data were not included here because subsequent analysis detected no nicotine in their e-cigarette liquid, although the liquid was advertised as containing 12mg/mL nicotine.

Participants were deemed eligible if they self-reported being healthy, aged 18–55, using 5 or less cigarettes daily, using at least 1mL of e-cigarette liquid daily, using e-cigarette solution with a nicotine concentration at least 12mg/mL, and using their e-cigarette for at least 3 months. Thus, these criteria allowed for dual users to be eligible to participate as long as they did not use more than 5 cigarettes/day (two participants reported being dual users, both reporting 1 cigarette/day on average). Participants were excluded for self-reported: history of chronic disease or psychiatric condition, regular use of a prescription medication (aside from birth control), marijuana use more than 10 days and alcohol use more than 25 days in the past 30 days, and use of other illicit drugs (eg, cocaine, opioids, benzodiazepines, and methamphetamine) in the past 30 days. Additionally, a positive pregnancy test (by urinalysis) at screening was exclusionary for women.

Materials

Participants used their preferred e-cigarette battery and liquid for each experimental session. Listed in Supplementary Table 1 are the e-cigarette models, liquid nicotine concentrations (according to product labeling), and labeled flavors. The liquid vehicle in which nicotine is dissolved in e-cigarette liquids may influence nicotine yield²³ as does battery voltage.²⁴ Therefore, the advertised ratios of propylene glycol to vegetable glycerin for each participant’s e-cigarette liquid and the advertised battery voltage are also listed when available. All experienced e-cigarette users in this study normally used a “tank” system and an e-cigarette powered by a rechargeable battery. Variable voltage batteries were set to participants preferred setting and kept consistent across conditions. However, there were several instances in which participants’ voltage setting was not recorded (Supplementary Table 1). While participants were permitted to use their normal e-cigarette battery, standardization of the cartomizer used was necessary, as the sizes of tank and cartridge systems vary considerably at the mouth end, and thus may not have be compatible with the mouthpiece-based topography system used in this study. Thus, “510” style cartomizers with 1.5 Ω resistance and dual heating coils (produced by SmokTech, Shenzhen, China) were used for each participant. Each participant provided their preferred e-cigarette battery, while their preferred e-cigarette liquid was purchased by laboratory staff from the participants’ typical source (ie, either Internet or local retailer).

Procedures

Participants completed two approximately 2.5-h sessions at VCU’s Clinical Behavioral Pharmacology Laboratory on two separate days. Sessions were separated by a minimum of 48h and differed only by the presence of a mouthpiece-based topography recording device. Condition order was counter-balanced across participants to control for possible order effects. Participants were instructed to abstain from nicotine/tobacco for at least 12h prior to each session. Abstinence from combustible tobacco was verified via participants’ expired air carbon monoxide (CO) concentration (BreathCO monitor; Vitalograph, Lenexa, KS; ≤10ppm).²⁵ Abstinence from nicotine delivered by noncombustible means such as e-cigarettes was verified by examining participants’ baseline plasma nicotine concentration retrospectively. Ultimately, five of the 29 participants included in the final analyses were suspected to have not abstained from nicotine for at least 12h, as evidenced by baseline plasma nicotine concentrations at least 5.2ng/mL (this threshold was determined arbitrarily). However, these five non-abstainers were included in the final analyses because none of the results changed upon excluding them, and the higher N improved statistical power. After recording expired air CO, physiological monitoring of heart rate and blood pressure was initiated. A catheter was then inserted intravenously and 7mL of blood was collected. Next, participants completed several computerized subjective questionnaires (see below). Thirty minutes after this initial blood draw, participants completed a directed e-cigarette-use bout (10 puffs, 30-s IPI). While puff number and IPI were held constant in this bout, puff duration, volume, and flow rate were not controlled. Immediately after the 10th puff, another 7mL blood was sampled and the same questionnaires were administered a second time. After 10 and 20 additional minutes, two more 7mL blood samples were collected. Following this fourth blood sample, participants completed a 90-min ad libitum e-cigarette-use session in which they were instructed to take as many puffs as they liked, whenever they liked. During the ad libitum bout, three additional 7mL blood samples were taken, one every 30min. Immediately after the ad libitum bout, participants completed the same subjective questionnaires for a third time. Fifteen minutes after the ad libitum bout was completed, the eighth and final blood sample occurred, the catheter was removed, and participants were compensated (US $100 after first session, US $150 after second).

Outcome Measures

Physiological Measures

After being centrifuged and stored at −70°C, all plasma samples were sent to VCU’s Bioanalytical Analysis Core Laboratories for analysis of nicotine concentration. The limit of quantitation (LOQ) for these analyses was 2ng/mL; further details can be found elsewhere.²⁶ Heart rate was monitored continually every 20s using Criticare Systems model 507 (fitted with pulse oximeter).

Subjective Questionnaires

Four questionnaires were administered using a computerized visual analog scale (VAS), consisting of a word or phrase centered on a horizontal line with “not at all” on the left and “extremely” on the right. Participants recorded their responses by moving a mouse cursor and clicking at any point on the horizontal line, and their scores were expressed as a percentage of total line length (0–100). Nicotine abstinence symptoms and subsequent withdrawal suppression were evaluated using a modified version of the Hughes–Hatsukami withdrawal scale (11 items, lacking two items from the original: “Increased eating,” and “Insomnia/Disturbed sleep”).²⁷ The direct effects of nicotine²⁸ and e-cigarette use^29,30 were also measured, both of which contain 10 items. A fourth subjective questionnaire assessing the extent to which the topography mouthpiece interfered with normal e-cigarette-use behavior was administered in the topography mouthpiece condition and consisted of six VAS items.²¹ Items on each of these four questionnaires were modified such that when the words “cigarette” or “smoking” appeared in the original versions, “e-cigarette” or “vaping” appeared in this study.

Puff Topography

E-cigarette puff topography was recorded using a mouthpiece-based topography device developed and manufactured at the American University of Beirut. As with commercially available products designed for cigarettes,²¹ the device’s mouthpiece is attached at two points to a pressure transducer, and flow-induced changes in pressure across the two points are amplified, digitized, and sampled every 100ms. Importantly, the orifice dimensions of the mouthpiece and the sensitivity of the transducer were chosen so that they can measure accurately puff velocities as low as 3mL/s (further details can be found elsewhere¹⁶).

Data Preparation and Analysis

As reported elsewhere,^7,8,16,20,31 plasma nicotine values below the LOQ were replaced with the LOQ (2ng/mL), and heart rate data were averaged for the 5min prior to each e-cigarette bout or blood sampling. Replacing plasma nicotine values below the LOQ with the LOQ is a more conservative approach than assuming values below the LOQ were zero when, in fact, they could range from 0 to 2.0ng/mL. For topography data, two data cleaning procedures were performed by the device’s software to correct for transducer noise prior to analyses. These cleaning procedures consisted of fusing any two puffs that were separated by 300ms or less into one puff and deleting any puffs with a duration less than 300ms. These cleaning procedures are performed automatically in real time and typically only involve a few milliseconds of data at any given time during a recording session; no record of these corrections was maintained. Remaining data for each topography variable were averaged for each participant within each bout, meaning participants ultimately had two topography values for each measure.

Repeated measures analysis of variance (ANOVAs) were used to examine plasma nicotine, heart rate, and subjective effect data. A power analysis conducted prior to beginning the study revealed that 15 participants were required to provide greater than 80% power, assuming a medium effect size and repeated measures correlations greater than 0.80 for plasma nicotine concentration.³² A two (session) by eight (time) repeated measures ANOVA was used to examine plasma nicotine. A two by nine repeated measures factorial ANOVA was used to examine heart rate data, as there was one additional timepoint for this outcome. For subjective measures that were administered in both conditions (ie, Hughes–Hatsukami, direct effects of e-cigarette use, and direct effects of nicotine), two (session: with and without mouthpiece) by three (time) ANOVAs were conducted for each item. Violations of sphericity were adjusted using Huynh-Feldt corrections, and Tukey’s Honestly Significant Difference (HSD) test was used to analyze any significant main effects and interactions. Tukey’s HSD was used for post hoc analysis because it is considered to be one of the most conservative post hoc tests for making pairwise comparisons and because the assumptions of Tukey’s HSD are that of omnibus F, known to be robust in the face of potential violations of normality.^33,34 Furthermore, the Huynh-Feldt correction accounts for potential violations of sphericity and the MSerror term from the corrected F statistics were used to perform all corresponding Tukey tests.

The six items assessing the influence of topography equipment, administered in the mouthpiece condition only, were analyzed individually using a paired t test with two levels of time as the within-subjects factor. Pre-e-cigarette use scores were not relevant for this questionnaire, as participants were unable to evaluate the extent to which the topography equipment influenced their normal behavior prior to using it.

Results

Statistical analyses (main effects and interactions) for all measures are displayed in Table 1. The main effect of time and the interaction between time and condition were of greatest interest, as the main effect shows the influence of e-cigarette use, and the interaction shows the extent to which the effects of e-cigarette use over time were affected by the presence of a topography mouthpiece. As displayed in Table 1, a main effect of time was observed for 21 measures, a main effect of condition was observed for one item, and an interaction of time and condition were observed for two subjective items.

Table 1.

Statistical Analyses Results for Plasma Nicotine, Heart Rate, and Subjective Measures (Directed Bout + Ad Libitum Bouts)

Outcome measures	Condition (C)	p value	Time (T)	p value	C × T	p value
	F value		F value		F value
Plasma nicotinea	0.5	ns	36.5	<.001*	0.7	ns
Heart rateb	0.1	ns	19.8	<.001*	0.7	ns
Subjective measures
Hughes–Hatsukamic
Anxious	1.2	ns	9.0	<.001*	1.1	ns
Craving	2.24	ns	51.7	<.001*	1.0	ns
Depression	0.4	ns	5.4	<.05*	1.0	ns
Difficulty concentrating	1.1	ns	3.1	ns	1.8	ns
Drowsy	0.3	ns	0.7	ns	1.4	ns
Hunger	0.2	ns	9.9	<.001*	0.4	ns
Impatient	0.0	ns	5.0	<.05*	4.0	<.05*
Irritable	1.1	ns	13.4	<.001*	0.7	ns
Restless	2.3	ns	3.3	<.05*	1.0	ns
Sweets	0.3	ns	0.5	ns	0.0	ns
Urge to vape	1.4	ns	47.6	<.001*	0.2	ns
Direct effects of nicotinec
Confused	4.4	<.05*	1.4	ns	0.5	ns
Dizzy	0.1	ns	6.3	<.01*	0.2	ns
Headache	0.6	ns	0.1	ns	0.7	ns
Heart pound	0.5	ns	2.3	ns	1.6	ns
Lightheaded	0.2	ns	6.8	<.01*	1.2	ns
Nauseous	0.3	ns	0.4	ns	0.7	ns
Nervous	0.1	ns	1.4	ns	1.3	ns
Salivation	0.5	ns	2.0	ns	1.0	ns
Sweaty	0.7	ns	3.0	ns	0.7	ns
Weak	0.0	ns	0.4	ns	0.8	ns
Direct effects of vapingc
Awake	1.0	ns	10.6	<.001*	1.4	ns
Calm	1.6	ns	25.2	<.001*	0.4	ns
Concentrate	0.0	ns	8.1	<.001*	0.0	ns
Dizzy	0.0	ns	6.0	<.01*	0.2	ns
Pleasant	0.8	ns	139.8	<.001*	2.8	ns
Reduce hunger	0.3	ns	8.2	<.01*	0.3	ns
Right now	0.4	ns	22.7	<.001*	0.1	ns
Satisfying	0.2	ns	127.6	<.001*	2.4	ns
Sick	0.0	ns	0.3	ns	1.3	ns
Taste good	3.6	ns	130.4	<.001*	3.9	<.05*

ns = nonsignificant. Asterisks (*) denote statistically significant F values (p < .05).

ªdf C = (1,28); df T = (7,196); df C × T (7,196).

^bdf C = (1,27); df T = (8,216); df C × T (8,216).

^Cdf C = (1,28); df T = (2,56); df C × T (2, 56).

Participant Characteristics

Participants in this study were: aged 18–55 (Mean = 29.6, SD = 7.7), used 5 or less cigarettes daily (Mean = 0.1, SD = 0.3), used at least 1mL of e-cigarette liquid daily (Mean = 2.7, SD = 1.4), used e-cigarette solution with a nicotine concentration at least 12mg/mL (Mean = 18.9, SD = 5.9; see Supplementary Table 1), and had been using their e-cigarette for at least 3 months (Mean = 10.2, SD = 9.2).

Physiological Measures

Plasma Nicotine

As Table 1 indicates, a significant main effect of time and no significant main effect of or interaction with condition was observed for plasma nicotine [F (7,196) = 36.5, p < .001]. Figure 1 shows the mean data for each condition and timepoint. Collapsed across condition, mean plasma nicotine concentration immediately after the directed e-cigarette-use bout (20.6, SEM = 2.8ng/mL) was significantly greater relative to baseline (4.0, SEM = 0.7ng/mL) and 10-min post-directed bout (11.7, SEM = 1.4ng/mL). Collapsed across condition, mean plasma nicotine concentration at 30 (25.7, SEM = 3.7ng/mL), 60 (31.2, SEM = 3.9ng/mL), and 90 (35.0, SEM = 4.6ng/mL) min of the ad libitum bout was also significantly greater relative to baseline and 5-min pre-ad libitum bout (9.6, SEM = 1.3ng/mL).

Figure 1.

Mean data (±SEM) for plasma nicotine across conditions (N = 29). Electronic cigarette (e-cigarette)-experienced participants completed a 10-puff e-cigarette use bout (30-s IPI) and a 90-min ad libitum use period in two conditions: with (circles) and without (triangles) a topography mouthpiece. Black bars beneath the x-axis indicate when blood plasma was sampled. Filled symbols indicate a significant difference from baseline (−30; the first timepoint). All ps < .05; Tukey’s Honestly Significant Difference. There were no significant between condition differences at any timepoint.

Heart Rate

A significant main effect of time [F (8,216) = 19.8, p < .001] and no significant main effect of or interaction with condition was observed for heart rate (Table 1). Figure 2 shows the mean data for each condition and timepoint. Collapsed across condition, mean values immediately after the directed bout (73.3, SEM = 1.3 bpm) and at 30 (73.9, SEM = 1.5 bpm), 60 (73.6, SEM = 1.6 bpm), and 90 (74.4, SEM = 1.7 bpm) minutes after the onset of the ad lib bout were significantly greater relative to baseline (66.3, SEM = 1.3 bpm).

Figure 2.

Mean data (±SEM) for heart rate across conditions (N = 28). A malfunction of the equipment resulted in incomplete data collection for one participant out of the 29 completers who were used for all other analyses. Black bars beneath the x-axis indicate the timepoints at which heart rate was recorded. In other respects, the figure is identical to Figure 1. There were no significant between condition differences at any timepoint.

Subjective Measures

Hughes–Hatsukami Withdrawal Scale

Significant main effects of time and no significant main effect of or interaction with condition were observed for the items “Anxious,” “Urge to use an e-cigarette,” “Craving an e-cigarette,” “Impatient,” “Depression,” “Hunger,” “Irritable,” and “Restless.” Collapsed across condition, mean ratings of the items with the three largest F values were: “Craving” (Baseline: 57.6, SEM = 5.4; post-directed: 25.5, SEM = 3.7; post-ad libitum: 13.7, SEM = 2.7), “Urge to use an e-cigarette” (Baseline: 59.1, SEM = 5.2; post-directed: 27.0, SEM = 3.3; post-ad libitum: 16.6, SEM = 3.0), and “Irritable,” (Baseline: 10.4, SEM = 2.7; post-directed: 2.2, SEM = 1.1; post-ad libitum: 1.7, SEM = 1.0). A similar pattern was observed for the items for which a significant main effect of time was observed. A significant condition by time interaction was observed for the item “Impatient” [F (2,56) = 4.0, p < .05]. Mean values for the three timepoints in the no mouthpiece condition for the item “Impatient” were: baseline (7.3, SEM = 3.0), post-directed (5.5, SEM = 2.4), post-ad libitum (4.8, SEM = 2.3) while mean VAS scores in the mouthpiece condition were: baseline (10.6, SEM = 3.0), post-directed (3.0, SEM = 1.6), post-ad libitum (3.9, SEM = 2.0). However, post hoc tests (Tukey’s HSD) revealed no differences between conditions at any timepoint.

Direct Effects of Nicotine

Main effects of time and no significant main effect of or interaction with condition were observed on VAS items evaluating “Dizzy” and “Lightheaded.” Post hoc tests (Tukey’s HSD) revealed that collapsed across condition, the mean post-directed bout VAS scores of these items were significantly greater relative to baseline. A significant main effect of condition was observed for the item “Confused” [F (1,28) = 4.4, p < .05]. Collapsed across time, higher mean ratings were observed in the no mouthpiece condition (1.7, SEM = 0.7) relative to the mouthpiece condition (0.6, SEM = 0.2) for this item.

Direct Effects of E-Cigarette Use

Main effects of time and no significant main effect of or interaction with condition were observed for nine of the 10 items. Collapsed across condition, mean scores of the three items with the largest F values were: “Pleasant” (baseline: 8.5, SEM = 4.2; post-directed: 73.3, SEM = 3.6; post-ad libitum: 79.0, SEM = 3.7), “Satisfying” (baseline: 7.2, SEM = 3.4; post-directed: 71.8, SEM = 4.2; post-ad libitum: 77.4, SEM = 4.4), and “Calm” (Baseline: 6.3, SEM = 3.6; post-directed: 38.3, SEM = 4.9; post-ad libitum: 39.7, SEM = 5.5). The remaining items with which main effects of time were observed had mean values that followed a similar pattern. A significant condition by time interaction was also observed for the item “Taste Good” [F (2,56) = 3.9, p < .05]. The interaction between condition and time is explained by a significantly greater post-directed bout mean score in the no mouthpiece condition (83.2, SEM = 3.5) relative to the condition with the topography mouthpiece present (72.7, SEM = 5.0; Tukey’s HSD, p < .05).

Topography

Paired samples t tests were also conducted to compare all puff topography measures between the directed and ad libitum bouts within the mouthpiece condition. A significant difference was observed for puff duration [t(28) = 3.55, p < .01], indicating that participants took longer puffs, on average, during the ad libitum bout (5.3, SEM = 0.4s) relative to the directed bout (4.5, SEM = 0.3s; see Table 2). In addition, significant differences were observed for puff volume [t(28) = 2.5, p < .05] and IPI [t(28) = 6.7, p < .001] demonstrating that participants also exhibited larger mean puff volumes (148.5, SEM = 22.2mL) and IPIs (102.7, SEM = 11.7s) in the ad libitum compared to the directed bout (volume: 124.6, SEM = 16.6mL; IPI: 25.1, SEM=0.3s). No differences were observed between bouts for flow rate. Additionally, a paired samples t test revealed no difference between the two conditions (ie, no mouthpiece and mouthpiece) for puff number observed in the ad libitum bouts (Table 2).

Table 2.

Mean (SD) Puff Parameters for Directed and Ad Libitum E-Cigarette-Use Bouts

Condition	Duration, s	Volume, mL	Flow rate, mL/s	IPI, s	Puff number
Mouthpiece
Directed	4.51a (1.55)	124.56a (89.13)	27.78 (19.48)	25.19a (1.55)	9.97 (0.12)
Ad libitum	5.29 (2.08)	148.52 (119.6)	27.47 (22.63)	102.77 (63.07)	62.55 (32.34)
No mouthpiece
Directed	N/A	N/A	N/A	N/A	10
Ad libitum	N/A	N/A	N/A	N/A	62.10 (39.99)

N/A indicates values that were not measured for that particular topography variable and condition.

^aSignificant difference between the directed and ad libitum bouts in the topography condition.

Topography Equipment Questionnaire

Within the mouthpiece condition, none of the six topography equipment items administered (“Altered e-cigarette use behavior,” “Made vaping less likely,” “Reduced enjoyment,” “Affected e-cigarette taste,” “Increased awareness,” and “Increased vaping difficulty”) differed between the post-directed and post-ad libitum bout values.

Discussion

The primary purpose of this study was to determine whether the presence of a mouthpiece-based topography recording device influenced plasma nicotine concentration, heart rate, or subjective effects in experienced e-cigarette users using their preferred battery and liquid. Importantly, unlike a preliminary study with fewer participants,¹⁶ the present study reports both 10-puff directed and 90-min ad libitum e-cigarette-use bouts. Similar to preliminary findings that only included one directed bout,¹⁶ results of the present study do not support the notion that the mouthpiece-based device affected any outcomes systematically, even under ad libitum e-cigarette-use conditions. Also as reported elsewhere,³¹ reliable increases in plasma nicotine concentration and suppression of abstinence symptoms were observed after participants used their preferred e-cigarette battery and liquid. Furthermore, plasma nicotine concentrations observed appeared to be pharmacologically active, as indexed by increased heart rate during e-cigarette use. Interestingly, the mean plasma nicotine concentrations observed in this study during the ad libitum bout were among the highest seen in experienced e-cigarette users using their preferred battery and liquid.

Another purpose of this study was to compare puff topography variables under directed and ad libitum e-cigarette use conditions using a mouthpiece-based device that was designed specifically to measure e-cigarette topography. Results indicated that participants took longer and larger puffs with greater IPIs during ad libitum use. These findings suggest that controlled e-cigarette puffing parameters (eg, 10 puffs, 30-s IPI) used in this study and others^7,8,20 may alter other puff topography outcomes such as puff duration, even when these other outcomes are not controlled experimentally. Thus, while controlling certain puffing characteristics during e-cigarette use in the laboratory can facilitate comparisons to cigarette smoking, this level of control may not mimic real-world e-cigarette use precisely. Importantly, when user puff topography is mimicked precisely and used to produce aerosol from other tobacco products (eg, waterpipe), significant correlations between toxicant yield and user toxicant exposure are observed.¹¹ Thus, understanding toxicant exposure in e-cigarette users may require recording puff topography under more natural use conditions.³⁵

Interestingly, nicotine delivery observed in e-cigarette users during the directed bout and during ad libitum use was comparable to that observed in cigarette smokers using their own brand of cigarettes under similar conditions. Generally, when smoking a single cigarette, smokers take an average of 10 puffs while exhibiting roughly 30-s IPIs^21,22 and achieve an increase in plasma nicotine of approximately 16ng/mL.²⁰ Under these same puffing parameters (ie, 10 puffs, 30-s IPI), e-cigarette users in the present study using their preferred battery and liquid also increased their plasma nicotine concentration on average by 16ng/mL. Furthermore, at the end of the period when e-cigarette users used their preferred device battery and liquid ad libitum, mean peak plasma concentrations (collapsed across condition) of 35ng/mL were observed (SEM = 4.6). This mean concentration is comparable to that seen in cigarette smokers smoking their preferred brand of cigarettes ad libitum over an extended period (eg, 27.0ng/mL, SEM = 4.9, N = 30²⁹; 29.2ng/mL, SEM = 2.2, N = 24³⁶). With regard to nicotine delivery, these results are consistent with others that indicate that at least some e-cigarettes can be as effective as combustible tobacco cigarettes at delivering this dependence-producing stimulant.^7,8

There were several limitations to this study. First, the laboratory setting may have influenced the puff topography of experienced e-cigarette users who are accustomed to puffing in their natural environment. Ambulatory measurement and recording of e-cigarette user puff topography,³⁵ accompanied with comparisons to clinical laboratory studies, is likely one way of determining the extent to which the laboratory setting influences study outcomes. Second, because the majority of puff topography variables (eg, puff duration, volume) were not recorded in the condition without the mouthpiece-based device, determining the extent to which the mouthpiece altered participants’ topography, as has been seen with cigarette smokers,^37,38 was challenging. However, given that participants’ plasma nicotine and puff number did not differ between the two conditions (ie, with and without the topography mouthpiece), there is little evidence to suggest that other puff topography variables such as puff duration and puff volume differed significantly between conditions. Third, participants were not able to use their preferred tank or cartomizer, as standardizing the cartomizer was necessary to accommodate the topography mouthpieces used. Given the variability in e-cigarette design, topography recording devices that are compatible with more than one style of cartomizer may be needed. Fourth, prior to the directed bout, participants were abstinent from nicotine for at least 12h but they began the ad libitum bout without this level of abstinence (ie, having completed the directed bout 30min before). These different abstinence periods resulted in participants beginning the ad libitum bout with a slightly higher, but not significantly different, mean plasma nicotine concentration. Future studies using directed and ad libitum bouts may benefit from standardizing participants’ period of abstinence from nicotine for both bouts, as differential abstinence periods may affect study outcomes including puff topography. Fifth, results would have been informed further by a comprehensive analysis of each participant’s preferred liquid, including actual nicotine concentration and propylene glycol:vegetable glycerin ratio, as well as pH levels. Future studies using participant-provided liquid may benefit from this analysis, as well as an assessment of how much liquid is consumed during laboratory studies. Last, conventional methods for verifying compliance of presession combustible product abstinence (eg, expired air CO) cannot be used to verify e-cigarette abstinence, as e-cigarettes do not produce CO. In this study, compliance was confirmed retrospectively using plasma nicotine concentration (the criterion for abstinence was set arbitrarily at less than 5.2ng/mL). This retrospective method can be problematic on several levels (eg, study sensitivity; research costs) if a high proportion of participants are found to have been noncompliant after recruitment has ended. Future work that requires nicotine/tobacco abstinence from experienced e-cigarette users may require a sensitive method for detecting noncompliance in this population immediately before a session commences.

In conclusion, the present study provided no evidence that the presence of a mouthpiece-based topography recording device influenced e-cigarette-associated nicotine delivery, heart rate changes, or the majority of subjective measures, and it confirmed reports that some e-cigarette device/liquid combinations are as effective as a combustible cigarette at nicotine delivery. E-cigarette users in the present study exhibited longer puff durations/IPIs and larger puff volumes when using their device ad libitum relative to controlled puffing conditions. Further understanding of e-cigarette user puff topography under natural use conditions is an essential step in understanding the toxicant content of e-cigarette aerosols that users inhale.

Supplementary Material

Supplementary Table 1 can be found online at http://www.ntr.oxfordjournals.org

Funding

Research reported in this publication was supported by the National Institute on Drug Abuse of the National Institutes of Health under Award Number P50DA036105 and F31DA040319 and the Center for Tobacco Products of the US Food and Drug Administration. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Food and Drug Administration.

Declaration of Interests

None declared.