Impact of Delivery Rate on the Acute Response to Intravenous Nicotine: A Human Laboratory Study with Implications for Regulatory Science

Abstract

Faster delivery rate enhances the abuse potential of drugs of abuse, yet systematic studies on the impact of delivery rate on the acute effects of nicotine in humans are lacking. Using an intravenous (IV) nicotine infusion procedure that allows precise control of rate of delivery, we examined the impact of nicotine delivery rate on the positive subjective drug effects, smoking urges, withdrawal, heart rate, blood pressure and attention function in smokers. Twenty-four male and female (ages 21 to 35) dependent smokers attended five experimental sessions, following overnight abstinence from smoking. Using a crossover design, participants attended 5 sessions, where they were assigned to a random sequence of saline infusion or 1 mg nicotine delivered over 1, 2.5, 5, or 10 minutes at rates of 1, 0.4, 0.2, or 0.1 mg/minute, respectively. The positive subjective effects of nicotine were most robust under the two faster delivery rate conditions, 1 and 0.4 mg nicotine/minute. In contrast, all nicotine delivery rates were equally more effective than saline in alleviating urges to smoke. Likewise, nicotine-induced heart rate increases did not vary with the rate of nicotine delivery. Lastly, the cognitive enhancing effects of nicotine were observed only under the two slowest delivery rate conditions – 0.1 and 0.2 mg nicotine/minute. Collectively, these findings support the critical role of delivery rate in optimizing nicotine’s abuse potential vs. potential therapeutic effects and have timely implications for developing novel therapeutics for nicotine dependence, as well as for tobacco regulatory science.

1. INTRODUCTION

Convergent evidence indicates that the reinforcing effects of drugs of abuse increase with faster delivery to the brain. In controlled human studies the rate of delivery positively correlates with the abuse potential of various drugs, such as cocaine ¹, morphine ², and diazepam ³. In preclinical studies, even small increases in the delivery rate of cocaine and nicotine can have a host of neurobiological consequences, such as enhancing the blockage of dopamine transporters, facilitating the expression of immediate early genes (e.g., c-fos and arc) and promoting psychomotor sensitization ^4–6. Other studies indicate that rapid delivery of drugs such as cocaine may increase drug-seeking behavior by promoting neuroplasticity within reward pathways ^7–10; or by evoking brain changes which, following abstinence, intensify drug craving ¹¹.

Further, clinical evidence highlights that increasing the delivery rate of some therapeutic drugs (e.g., psychostimulants and opioids), by switching from the oral to the intravenous route of administration, can dramatically increase their addictive properties ^12,13. Conversely, slowing the delivery rate of drugs have been utilized as a strategy to reduce their abuse potential (e.g., psychostimulants) ¹⁴. Collectively, these findings underscore the need to systematically study the relationship between the delivery rate of drugs and their abuse potential.

Accumulating preclinical data shows that nicotine has both addictive and therapeutic effects depending on the rate of delivery ^15–17. In humans, when delivered rapidly by cigarette smoking or electronic cigarettes (e-cigarettes), nicotine is highly reinforcing and addictive. In contrast, products delivering nicotine slower than cigarettes or e-cigarettes while achieving comparable blood levels (e.g., nicotine patch, lozenge or spray), have negligible addictive effects but, among abstinent smokers, retain their ability to alleviate craving and suppress withdrawal ¹⁸. Such discrepancy may be explained by the diverse mechanisms underlying the various effects of nicotine ¹⁹. Notably, the nicotinic acetylcholine receptor (nAChR) subtypes that mediate the reinforcing, withdrawal-suppressing, cognitive enhancing, and physiological effects of nicotine show significant variation in their distribution pattern, agonist affinity, and desensitization threshold (Figure 1) ²⁰. These differences at the neurobiological level may influence the response to changes in the delivery rate of nicotine at the behavioral level. However, systematic human studies examining the impact of delivery rate on the abuse potential vs. therapeutic effects (i.e., alleviation of craving and withdrawal) of nicotine are lacking. A clearer understanding of how the rate of nicotine delivery impacts these outcomes may help to develop strategies that minimize the risk of addiction and improve treatment outcomes.

Figure 1. Neurobiology of the subjective, physiological, and cognitive effects of nicotine.

The effects of nicotine are mediated by nicotinic acetylcholine receptors (nAChR) found in the cortical and subcortical sensory, limbic, and associative areas of the brain and peripheral organs. These receptors are widely distributed in the prefrontal cortex (PFC), which mediates nicotine-induced cognitive enhancement; the nucleus accumbens (NAC) and the ventral tegmental areas (VTA), which underlie nicotine’s reinforcing effects; the habenula (HB), which regulates nicotine withdrawal; and the brainstem (BST) and the sympathetic ganglion chain (SGC), which control nicotine’s cardiovascular effects – in conjunction with the hormones epinephrine and norepinephrine, secreted by the adrenal medulla. The solid black lines represent ascending pathways, and the dashed black lines represent descending pathways. The nAChR subunits responsible for mediating the effects of nicotine show significant variation in distribution pattern, agonist affinity, and desensitization threshold. Taken together, these differences at the neurobiological level may influence the response to changes in the delivery rate of nicotine at the behavioral level for the subjective, cognitive, and withdrawal/craving suppressing effects of nicotine.

In a pilot study using an intravenous (IV) nicotine infusion procedure, which provides precise control over the dose and rate of delivery, we examined the impact of delivery rate of nicotine on the subjective drug effects and urges to smoke in dependent smokers (n=18) ²¹. We found that 1 mg of nicotine – the average amount of nicotine delivered by smoking a cigarette – produced greater positive subjective effects (indicating higher abuse potential) when it was infused IV over 1 minute, compared to the same amount of nicotine delivered more slowly (i.e., over 5 or 10 minutes ²¹). Conversely, 1-minute and 5-minute nicotine delivery rates were equally effective in alleviating urges to smoke. Based on these promising findings, we designed the current study, which built on our prior work to include a larger sample size; administer additional nicotine delivery rate conditions, thereby emulating a broader range nicotine delivery systems; include a broader range of clinically relevant outcomes; and collect biomarkers of nicotine use. Notably, since cognitive deficits are one of the hallmarks of tobacco withdrawal, the present study includes measures of cognitive performance as a function of the nicotine delivery rate ^22,23.

We hypothesized that the abuse potential of nicotine, indexed by its positive subjective effects, would gradually decline with slower delivery rates. We further hypothesized that the nicotine delivery rate of nicotine would have minimal effects on urges to smoke.

2. METHODS

2.1. Participants

Participants were recruited from the New Haven, Connecticut, area through newspaper and on-line advertisements. The study was approved by the VA Connecticut Healthcare System and the Yale University Institutional Review Boards. Written informed consent was obtained from all participants before enrollment in the study, and all participants were compensated. A total of 23 participants completed all study procedures and 1 participant provided partial data (13 male and 11 female). The mean age of the participants was 28.1 (SD=3.4, range 21–35) years. Participants reported smoking more than 5 cigarettes/day for the past year and were not seeking smoking cessation therapies. Daily smoking status was confirmed by having urine cotinine levels higher than 100 ng/mL ²⁴, and abstinence from recent drug use, other than nicotine, was confirmed by a urine drug immunoassay (Table 1). The participants had no current major medical problems, as determined by medical history, physical exam, and laboratory testing. The Structured Clinical Interview for DSM-IV (SCID-IV) was used to determine the presence of an exclusionary psychiatric disorders, including serious mental illnesses – schizophrenia, major depressive disorder, or bipolar disorder – or substance use disorders other than nicotine. Finally, as a safety measure, women who were pregnant or breastfeeding were excluded from the study.

Table 1:

Demographics, Smoking, and Tobacco Product Use

	M (SD)	n (%)
Demographics
Age (years)	28.1 (3.4)
Sex
Male		13(54.1)
Female		11(45.8)
Race
African American		14(58.3)
Caucasian		7 (29.1)
Native American		1 (4.2)
Other		2 (8.3)
Smoking severity and tobacco product use at baseline
FTND	2.8 (0.9)
Average cigarettes per day	7.7 (2.6)
Years of smoking	17.7 (2.7)
Serum cotinine level (ng/mL)	206 (141.1)
Serum 3-OH cotinine level (ng/mL)	66 (49.6)
3-OH cotinine / cotinine ratio	0.37 (0.1)

Abbreviations: FTND: Fagerstrom Test of Nicotine Dependence; CO: carbon monoxide.

3-OH cotinine: 3-hydroxycotinine..

2.2. Study Medication

Nicotine bitartrate was obtained from Interchem Corporation (Paramus, NJ), and solutions for IV injection were prepared by U.S. Specialty Formulations (Bethlehem, PA). An Investigational Tobacco Product (ITP) application was approved by the United States Food and Drug Administration (FDA).

2.3. Laboratory procedures

Following overnight abstinence from smoking – verified by expired air carbon monoxide levels of ≤ 10 parts per million – participants attended 5 experimental sessions, separated by at least 24 hours to limit carryover effects. At the beginning of each session, participants had an indwelling intravenous catheter placed in their ventral forearm. The five nicotine infusion conditions were saline or 1 mg/70 kg nicotine, infused at one of the four different rates: 1, 0.4, 0.2, and 0.1 mg nicotine/minute, corresponding to approximately 16.7, 6.7, 3.3 and 1.7 μg nicotine/second, respectively. The total nicotine dose was 1 mg/70 kg body weight, which is within the range of nicotine delivered by smoking either a tobacco cigarette or an e-cigarette ²⁵. The fastest nicotine delivery rate (1 mg/minute) is close to what a smoker might encounter while smoking a cigarette, 0.1 mg/ of nicotine intake over 2 seconds per each puff ²⁶. The slowest nicotine delivery rate, 0.1 mg/minute was chosen to approximate the nicotine intake rate, which an individual might encounter from an e-cigarette, which ranges from 0.8 to 1.4 mg over 10 minutes ^27,28. The two middle nicotine delivery rates of 0.4 and 0.2 mg nicotine/minute were selected to approximate the rate of delivery encountered while using newer generation e-cigarettes. Notably, 3 out of 4 nicotine delivery rates used in this study – 1, 0.2 and 0.1 mg/min – were piloted in our previous work ²¹.

The total infusion duration of 10 min across each session, was achieved by adding saline infusions of variable durations after nicotine delivery. A research nurse, who did not participate in data collection, controlled 2 infusion pumps to infuse nicotine or saline solutions, labeled by the research pharmacy as drug A or B. For the 10-minute nicotine infusion condition (0.1 mg nicotine/minute), participants received only nicotine. For the 1-, 2.5-, and 5-minute infusion conditions (1, 0.4 and 0.2 mg nicotine/minute) participants first received nicotine for 1, 2.5 and 5 minutes, followed by saline infusions for a total infusion duration of 10 minutes. For the saline condition, participants received exclusively saline for 10 minutes. Both the research team and participants were blind to the randomization.

2.4. Primary outcomes

2.4.1. Subjective Effects

Subjective drug effects were assessed using the Drug Effects Questionnaire (DEQ) assessed at 1, 3, 5, 8, and 10 minutes after the initiation of each infusion. The DEQ is composed of 9 items measured on a 1 to 100 visual analog scale anchored in the statements “not at all” and “extremely”, and subsequently converted to a numeric score. Building on our previous work ^29–31, DEQ ratings were grouped into three categories: 1) stimulatory effects, the average of “feel stimulated”, “feel drug effects” and “feel high”; 2) pleasurable effects, the average of “like”, “feel good ” and “want more”; and 3) aversive effects, the average of “feel anxious”, “feel down” and “feel bad”.

2.4.2. Urges to Smoke and Nicotine Withdrawal

Cigarette craving was assessed using the Questionnaire of Smoking Urges – Brief (QSU-B), administered just before and at 5, 30, 60, and 120 minutes after the initiation of each infusion. The QSU-B is composed of 10 items, each rated with values from 0 to 7, anchored in the statements “strongly disagree” and “strongly agree”. Also building on our prior work, the QSU-B scores were analyzed as two factors, whose respective items reflect a desire to smoke to obtain positive/reinforcing effects (Factor 1), or to reduce negative affect (Factor 2) ³².

Lastly, nicotine withdrawal was assessed using the Minnesota Nicotine Withdrawal Scale (MNWS), administered just before and then 30, 60, 90, and 120 minutes after the initiation of infusion. The MNWS includes 8 items, rated on a 0 to 100 visual analogue scale, anchored in the statements “not at all” and “extremely”, and subsequently converted the sum of total numeric score.

2.5. Secondary outcomes

2.5.1. Physiological Effects

Heart rate was measured at 1 min and every 2 minutes for 20 minutes; and then at 25, 30, 40, 50, 60, 90, and 120 minutes after the initiation of infusions. Systolic and diastolic blood pressure were measured at 5-minute time point and every 5 minutes for 30 minutes then at 30, 40, 50, 60, 90, and 120 min after the initiation of infusions.

2.5.2. Cognitive Effects

Cognitive performance was assessed with Continuous Performance Test (CPT), using the Automated Neuropsychological Assessment Metrics (ANAM) software. The cognitive task was administered prior to infusion then at 30, 60, 90, and 120 minutes after the initiation of infusions. The CPT is used to assess sustained attention, concentration, and working memory and is sensitive to both tobacco withdrawal and nicotine delivery ^22,23. The running memory CPT requires participants to press the left button on an external mouse when the displayed number (i.e., 0–9) matches the number which immediately preceeded it. If the displayed number is not the same as the preceeding number, the participant was asked to press the right button on the external mouse. Out of 80 trials, 50% displayed a number that matched the preceeding number and for the other 50% the displayed number did not not match the preceeding number. Each number was displayed for 2500 ms and the participant had 3000 ms to respond before the next number was presented. Consistent with prior research ³³, CPT administrations with low accuracy (<56%) were excluded from the final analysis. The main outcome measures for the CPT were the throughput score, percent correct responses, and reaction time (52). The throughput score has been used in our prior work ^34,35 and includes information on both the speed and accuracy of task performance ³⁶. Finally, biomarkers of nicotine use, including plasma nicotine and cotinine concentration, were collection prior to each infusion and used to confirm overnight abstinence and assess baseline nicotine intake, respectively.

2.6. Data Analysis

Each dependent variable was analyzed using a mixed effects model with the within-subject factors of delivery rate (0.1, 0.2, 0.4, 1 mg nicotine/min; saline only), time (since the initiation of the infusion), the interaction between delivery rate and time; and a main effect of session. Random effects for subject and structured variance-covariance matrices for the repeated measures on each subject within-session were used to account for the correlation structure of the data. The best-fitting structure for each outcome was selected based on the Schwartz-Bayesian Information Criterion (BIC). Least square means and standard errors were calculated to describe the patterns of means for each outcome. Tests of effect slices were used to describe significant interactions. Pairwise comparisons of least square means were used to describe significant main effects of delivery rate and interactions of delivery rate with time (i.e., minute after the infusion initiation).

3. RESULTS

3.1. Subjective Effects

For stimulatory effects, we found significant main effects of delivery rate (F(4,67.1)=8.46, p<.0001) and time (F(4,100)=8.31, p<.0001), as well as an interaction of delivery rate and time (F(16,100)=3.05, p=.0003). While the stimulatory effects were significantly lower under saline condition than all other delivery rates, the slowest delivery rate (0.1 mg/minute) was associated with significantly lower stimulatory effects than the two fastest delivery rate conditions: 0.4 mg/minute (Mean difference (Mdiff) (SE) = −11.02 (4.08), df=68.7, t=−2.70, p=.009) and 1 mg/min (Mdiff (SE) = −11.72 (3.98), df=66, t=−2.94, p = .005). The peak stimulatory effects were observed at 3 minutes and significant differences among delivery rate conditions were observed at 1 minute, 3 minutes and 5 minutes after the initiation of infusions (p-values < .001, Figure 2A).

Figure 2. Impact of nicotine delivery rate on subjective drug effects.

Mean scores for subjective (A) stimulatory, and (B) pleasurable effects, indexed by the Drug Effects Questionnaire (DEQ), in response to infusion of saline and nicotine (total dose, 1.0 mg per 70 kg body weight), infused over 1 min (1 mg nicotine/minute), 2.5 minutes (0.4 mg nicotine/minute), 5 minutes (0.2 mg nicotine/minute) and 10 min (0.1 mg nicotine/minute). Error bars reflect standard error of the mean. * p < 0.05 for pairwise comparisons of the nicotine condition with the saline condition. For details, please see section 3.1 of the manuscript.

For pleasurable effects of nicotine, we found statistically significant main effects of delivery rate (F(4,89.9)=5.35, p=.0007) and time (F(4,384)=9.24, p<.0001). The rating of pleasurable effects was significantly lower under the saline conditions than the 3 nicotine delivery rate conditions: 0.2 mg/minute (Mdiff (SE) = −13.30 (4.19), df=89.4, t=−3.17, p=.002), 0.4 mg/minute (Mdiff (SE) = −17.00 (4.34), df=91, t=−3.92, p=. 0002) and 0.1 mg/minute M(SE) = −15.71 (4.17), df=90.4, t=−-3.76 p=. 0003). Further, the slowest delivery rate of 0.1 mg/min was associated with lower pleasurable effects than those observed for the second fastest delivery rate of 0.4 mg/min (Mdiff (SE) = −9.20 (4.19), df=90.3, t=−2.19, p=.003). The peak stimulatory effects were observed at 3 minutes and significant differences among delivery rate conditions were observed at 1-, 5- and 7-minute time points (p-values < .05) (Figure 2B).

3.2. Smoking Urges and Nicotine Withdrawal

For smoking urges, the Factor 1 of the QSU-B showed a significant interaction between delivery rate and time (F(20,103)=1.98, p=.014) and a significant main effect of time (F(5,102)=12.94, p<.0001). There were no significant differences among delivery rate conditions at any of the time points (Figure 3A).

Figure 3. Impact of nicotine delivery rate on smoking urges.

Mean scores for (A) Factor 1– urges to smoke to obtain positive/reinforcing effects – and (B) Factor 2 (i.e., urges to smoke to relieve negative affect) of the Questionnaire on Smoking Urges-Brief (QSU-B), in response to saline and nicotine (total dose, 1.0 mg per 70 kg body weight) infused over 1 minute (1 mg nicotine/minute), 2.5 minutes (0.4 mg nicotine/minute), 5 minutes (0.2 mg nicotine/minute) and 10 min (0.1 mg nicotine/minute). Error bars reflect standard error of the mean. * p < 0.05 for pairwise comparisons of the nicotine condition with the saline condition. For details, please see section 3.2 of the manuscript.

For Factor 2, we found significant main effects of delivery rate (F(4,77.3)=3.00, p=.02) and of time (F(5,104)=6.28, p<.0001). QSU-B Factor 2 scores were significantly higher under saline condition than under the two slower delivery rates of 0.1 mg/min, (Mdiff (SE) = 2.56 (1.01), df=76.5, t=2.53 p=. 001), 0.2 mg/min, (Mdiff (SE) = 3.47 (1.04), df=76, t=3.34p=.001), and the fastest delivery rate of 1 mg/min (Mdiff (SE) = 2.07 (1.03), df=76.2, t=2.01 p=.05). As expected, smoking urges were also higher at the beginning of the infusion, compared to subsequent time points and significantly lower at 3 minutes, compared to subsequent time points (p-values < .05, Figure 3B). The delivery rate conditions were not associated with changes in MNWS scores.

3.3. Heart Rate and Blood Pressure

For heart rate, we observed significant main effects of delivery rate (F(4,84.5)=7.40, p<.0001) and time (F(16,1657)=56.18, p<.0001) and an interaction between delivery rate and time (F(64, 1657)=4.27, p<.0001). Heart rate was significantly higher under all nicotine delivery conditions than the saline condition (p-values < .001); however, there were no differences among the nicotine delivery conditions. The difference between the 4 nicotine delivery conditions from saline was significant from 3 to 40 minutes, during which heart followed an inverted U-shape, such that participants experienced an increase, followed by the heart rate’s gradual return to baseline (Figure 4).

Figure 4. Impact of nicotine delivery rates on heart rate.

The mean change in heart rate in response to saline and nicotine (total dose, 1.0 mg per 70 kg body weight), infused over 1 minute (1 mg nicotine/minute), 2.5 minutes (0.4 mg nicotine/minute), 5 minutes (0.2 mg nicotine/minute) and 10 min (0.1 mg nicotine/minute). Error bars reflect standard error of the mean. * p < 0.05 for pairwise comparisons of the nicotine condition with the saline condition. For details, please see section 3.3 of the manuscript.

For systolic blood pressure (Figure 5), we found a significant main effects of nicotine delivery rate (F(4,83.4)=3.63, p=0.009) and time (F(10,1042)=4.58, p<.0001), as well as an interaction between nicotine delivery rate and time (F(40, 1041)=1.54, p=0.02). Systolic blood pressure was significantly higher for all nicotine delivery rate conditions than saline (p-values < .05) – except for the 1 mg/minute delivery rate, and there were no other differences among nicotine delivery rate conditions. The difference between the 4 nicotine delivery conditions from saline was significant from 10 to 40 minutes during the infusion (p-values < .05).

Figure 5. Impact of nicotine delivery rates on blood pressure.

The mean change in systolic (A) and diastolic (B) blood pressure in response to saline and nicotine (total dose, 1.0 mg per 70 kg body weight), infused over 1 minute (1 mg nicotine/minute), 2.5 minutes (0.4 mg nicotine/minute), 5 minutes (0.2 mg nicotine/minute) and 10 min (0.1 mg nicotine/minute). Error bars reflect standard error of the mean. * p < 0.05 for pairwise comparisons of the nicotine condition with the saline condition. For details, please see section 3.3 of the manuscript.

Conversely, we did not find significant main or interactive effects of time or nicotine delivery rates for changes in diastolic blood pressure.

3.4. Cognitive Effects

For CPT throughput scores (Figure 6), which incorporates information from both accuracy and reaction time (higher throughput scores indicate ‘better’ task performance), there was a significant main effect of delivery rate (F(4,123) = 2.48, p = .05) and a significant main effect of testing session (F(4,131) = 3.89, p = .005). Throughput scores were highest at the 0.2 mg nicotine/minute condition (M = 107.4, SD = 23.08), across all time points, and this condition was significantly different from saline (M = 100.6, SD = 22.00, p = .007), 0.4 mg (M = 102.2, SD = 22.8, p = .05), and 1 mg (M = 102.6, SD = 24.6, p = .05) delivery rate conditions. The 0.1 mg (M = 104.4, SD = 23.3) nicotine/minute condition was also significantly higher than the saline condition (p = .04). Throughput scores were also highest on the final testing session, although the second highest value was on the second testing session. The second testing session was significantly higher than the first (p = .009) and the final testing session was significantly higher than the first (p = .0002) and third session (p = .03), indicating improved task performance with repeated sessions (e.g., possible practice effect).

Figure 6. Impact of nicotine delivery rates on CPT performance.

This figure illustrates the throughput scores (higher scores indicate better performance) on a Continuous Performance Task (CPT) under different nicotine delivery rates. Participants received saline and nicotine (total dose, 1.0 mg per 70 kg body weight), infused over 1 minute (1 mg nicotine/minute), 2.5 minutes (0.4 mg nicotine/minute), 5 minutes (0.2 mg nicotine/minute) and 10 min (0.1 mg nicotine/minute). Error bars reflect standard error of the mean. * p < 0.05 for pairwise comparisons of the nicotine condition with the saline condition. For details, please see section 3.4 of the manuscript.

There was a significant main effect of time for CPT percent correct (F(4,311) = 3.97, p = .004). The percent of correct responses significantly decreased over time within each session with the 90- and 120-minute time points significantly lower than the previous time points. Similarly, for CPT reaction time, there was a significant main effect of time (F(4,308) = 4.84, p = .0008) as well as a significant main effect of testing session (F(4,115) = 10.74, p < .0001). There was a significant decrease in reaction time within each session with reaction times lower at all time points compared to baseline. Additionally, reaction times decreased from session to session with the lowest reaction time at the final session

4. DISCUSSION

This study evaluated the impact of delivery rate on nicotine’s acute effects on multiple outcome domains in dependent smokers. We found that the positive subjective effects of nicotine were most robust under the two faster delivery rate conditions. In contrast, all nicotine delivery rates were equally more effective than saline in alleviating urges to smoke in abstinent smokers. Likewise, nicotine-induced increases in heart rate increases did not vary as a function of delivery rate either. Lastly, the cognitive enhancing effects of nicotine were observed only under the two slowest delivery rate conditions. Collectively, these findings, in conjunction with our previous work ²¹, support the central role of delivery rate in determining the trade-off between the abuse potential and the therapeutic effects of nicotine. Namely, that nicotine delivered at faster rates has higher abuse potential; conversely, nicotine delivered at slower rates not only has lower abuse potential, but may also provide adequate – or even superior, in the case of cognitive performance – therapeutic effects.

Commonly used nicotine delivery products (e.g., cigarettes, e-cigarettes, or NRT) do not allow precise control on the delivery rate; therefore, they are not suitable to examine the impact of delivery rate on the acute effects of nicotine. By using a well-validated IV nicotine infusion procedure, we were able to systematically examine the impact of nicotine delivery rate on various clinically relevant outcomes. Notably, the greater pleasurable and stimulatory effects produced by faster delivery rates (1mg nicotine/minute and 0.4 mg nicotine/minute), compared to those produced by slower delivery rates (0.2 mg nicotine/minute and 0.1mg nicotine/minute), are consistent with the results of our previous study ²¹. Altogether, these findings unequivocally demonstrate the impact of delivery rate on nicotine’s rewarding and potentially addictive effects in humans. In contrast, the alleviation of smoking urges was less dependent on the nicotine delivery rate than nicotine’s subjective effects: IV nicotine reduced the Factor 2 (urges to smoke to relieve negative affect), but not Factor 1 (urges to smoke to achieve positive affect) of the QSU-B ³². Further, the 1mg nicotine/minute, 0.2 mg nicotine/minute, and 0.1mg nicotine/minute conditions were all similarly effective in reducing urges to smoke to relieve negative affect, indexed by Factor 2 of QSU-B. Similarly, in our previous study, IV nicotine alleviated only Factor 2 of the QSU-B²¹. The underlying mechanisms for these findings are unclear. In previous studies with abstinent smokers, different nicotine replacement products were similarly effective in suppressing urges to smoke to relieve negative affect as well as to achieve positive affect (Davies, Willner et al. 2004, Evans, Blank et al. 2006). Notably, the present findings add support to the notion that the delivery rate can be leveraged to optimize the trade-off between risks (i.e., abuse liability) and benefits (i.e., suppression of craving) of nicotine. For instance, a slow delivery of 1 mg nicotine – over 10 minutes – alleviated urges to smoke, and the pleasurable effects produced were equivalent those obtained with saline.

The impact of nicotine delivery rate on reinforcement vs. craving suppression is likely due to differences in neurobiological mechanisms underlying these outcomes. Nicotine reinforcement is mediated by increases in phasic dopamine (DA) release in the nucleus accumbens, as a result of stimulation of post-synaptic nAChR on DA neurons in the ventral tegmental area (VTA) ³⁷. In contrast, for urges to smoke and withdrawal, a key mechanism is reduction of tonic DA release in the nucleus accumbens, which involves nAChR in the habenula-interpeduncular area (Figure 1). The specific contributions of nAChR subtypes – which differ in agonist affinity and desensitization threshold – need to be elucidated in further mechanistic studies investigating the relationship between neurobiological and behavioral correlates of the nicotine delivery rate ^20,38,39.

Although nicotine-induced heart rate increases were higher across all nicotine delivery conditions than saline, systolic blood pressures increase were significantly higher from saline for all delivery conditions, except the 1mg nicotine/minute condition. These results differ from our previous study, in which the 1 mg nicotine/minute condition produced the greatest heart rate increases ²¹. Notably, in the present study the magnitude of heart rate and systolic blood pressure increase was small – about 4–5 bpm and 4–6 mmHg, respectively –, and the nicotine, at any delivery rate, did not increase diastolic blood pressure more than saline. Such apparent inconsistencies in the findings could be due to additional data-collection time points and a larger sample in the present study. Nicotine, as a stimulant drug, activates the sympathetic nervous system, resulting in increased heart rate, blood pressure, and the circulating catecholamines epinephrine and norepinephrine ⁴⁰. Given the potentially harmful cardiovascular effects of nicotine, the present study highlights the delivery rate as an important factor in estimating the overall health effects of nicotine products ⁴¹.

Among abstinent smokers, difficulty concentrating and easy distractibility are commonly and consistently reported ⁴². Nicotine is also known to have cognitive-enhancing effects, especially for attentional function, although this is not consistently reported across studies ⁴³. We found that the two slower nicotine delivery conditions – 0.1 mg nicotine/minute and 0.2mg nicotine/ minute – , but not the faster conditions – 0.4 mg nicotine/minute and 1 mg nicotine/minute – were associated with improved attention, indexed by higher throughput scores on the CPT ^22,23. There were no differences among nicotine delivery rate conditions for the accuracy (number of correct responses) and speed (reaction time for correct responses) measures from the CPT. However, the throughput score is recommended as a primary outcome measure for this task since it is more effective at capturing holistic task performance than either the speed or accuracy measures are, in isolation. This is because the task would be expected to capture some degree of speed-accuracy trade-off (i.e., one could improve accuracy by sacrificing response speed or improve speed by sacrificing accuracy). Therefore, throughput is the outcome measure which can most meaningfully capture overall optimization of task performance.

To our knowledge, this is the first study to examine the impact of delivery rate on the cognitive effects of nicotine, so the mechanisms by which nicotine’s delivery rate influences its attention-enhancing effects remain to be elucidated. The nAChRs in the prefrontal cortex (PFC) are implicated in improving attention. Fast or transient neuromodulatory acetylcholine (ACh) release in the PFC is believed to mediate cue-detection, whereas slower neuromodulatory ACh release may underlie attentional function ⁴⁴. Thus, it remains to be determined whether slower delivery rates of nicotine may lead to prolonged ACh release, thereby leading to greater improvement in attentional function.

Despite its notable strengths, this study has some limitations. First, a single dose of nicotine of 1 mg/70 kg of body weight was tested. Hence, whether delivery rate similarly affects higher or lower doses of nicotine remain to be determined in future studies. Second, participants were tested about 10 hours following abstinence from smoking. While sufficient to assess urges to smoke, longer durations of abstinence may be necessary for the severity of some withdrawal symptoms to reach peak level. Third, the participants were dependent cigarette smokers, and the results may not be generalizable to non-dependent or non-daily smokers or users of other tobacco products Fourth, the addictive potential of smoking tobacco may be by factors beyond nicotine-induced subjective effects – such as by interoceptive and external drug-related stimuli ⁴⁵. Lastly, the sample size was too small to examine individual differences including sex/gender and race/ethnicity on the study outcomes. It will be important to evaluate these factors in future studies.

Current tobacco control strategies focus primarily on limiting or reducing the total amount of nicotine delivered by tobacco products with a high abuse potential – such as cigarettes and e-cigarettes. Early preclinical studies have suggested that rapid delivery of nicotine, especially in the context of intermittent and spiking drug intake, tend to facilitate incentive sensitization, a hallmark of addictive behavior. Such behavior can be promoted by not only striatal dopamine release, but also by selectively fostering neurobehavioral plasticity in mesocorticolimbic circuits^46,47. A more recent proposal by Shihadeh and Eissenberg focused on the rate of nicotine yield from an e-cigarette (“nicotine flux”) as the key factor in determining its abuse potential ⁴⁸. Accordingly, if the nicotine flux of an e-cigarette is above a certain – yet empirically undetermined – threshold, it can have abuse potential and either lead to the initial development of or the increased severity of nicotine addiction. In contrast, if the nicotine flux is “optimum”, the e-cigarette may have a low abuse potential (i.e., by lowering its positive reinforcing properties), while providing sufficient nicotine delivery to support smoking cessation, by alleviating urges to smoke and reducing tobacco withdrawal symptoms (i.e., providing sufficient negative reinforcement). Our findings provide empirical support for this hypothesis and further suggest that in addition to limiting the total amount of nicotine delivered, reducing the delivery rate of the nicotine may be a potential target to balance the abuse potential vs. therapeutic effects of nicotine. Newer tobacco products like e-cigarettes provide multiple ways to control the delivery rate of nicotine ⁴⁹, including hardware/software (e.g., puff duration and power of the device) and e-liquid ingredients (e.g., nicotine level). For example, limiting the a) maximum nicotine levels of e-liquids; b) maximum puff duration (per puff) of devices (e.g., perhaps by requiring an auto-shut off function the battery which limits how long the coil is heated following each activation (by inhalation or button press)); and c) maximum heat or power settings of devices, would be expected to not only reduce the amount of nicotine delivered to the user, but also the rate of delivery. Factors limiting the nicotine flux could also have secondary safety benefits; for example, reducing the maximum heat of the coil could also reduce the other toxins generated by the device. Altogether, these features of electronic nicotine delivery systems offer opportunities for novel therapeutics and for materializing clinically relevant targets for tobacco regulatory science ⁵⁰.

5. CONCLUSION

We have shown that the acute effects of nicotine among shorter-term abstinent smokers are modulated by the nicotine delivery rate, such that slower delivery rates favor suppression of urges to smoke over abuse liability. Against the backdrop of increasingly available and potent e-cigarettes, our findings have timely implications for regulatory science. Further, recognizing that the drug delivery rate has a central role in establishing and maintaining addiction to nicotine underscores the need to subject e-cigarettes – and other new introductions to the tobacco product marketplace – to rigorous experimental evaluation.