The Interplay Between Addiction to Tobacco Smoking and Sleep Quality Among Healthy Adults

Scientific background Tobacco smoking is a major health problem, leading to considerable morbidity and mortality due to cancer, pulmonary illnesses, and cardiovascular diseases (Taghizadeh, Vonk \& Boezen, 2016). The main psychoactive and addictive substance in tobacco is nicotine (Zaparoli \& Galduroz, 2012). Nicotine addiction is a complex phenomenon that involves both physical and psychological dependence (Cohrs et al., 2014), which causes not only difficulty in quitting but also a strong tendency to return to smoking after having quit for a long time) Zaniewska, Przegalinski \& Fillip, 2009).

A common theoretical model of addiction (Koob \& Volkow, 2016) holds that the transition from occasional smoking to addiction involves an upregulation of neurobiological stress systems. Consequently, even a brief period of abstinence from smoking, leading to reduced concentration of nicotine in the body, induces both emotional withdrawal symptoms (anxiety, restlessness, irritability, anhedonia) and cognitive withdrawal symptoms (diminished memory and attention) that may in turn produce a compulsive urge to smoke again, in order to ease the unpleasant sensations (Koob \& Volkow, 2016). Although some studies suggest reduced quality of sleep is also among the consequences of smoking and smoking cessation (Cohrs et al., 2014; Colrain, Trinder \& Swan, 2004) this issue was not fully explored, and the contribution of the reduced quality of sleep to negative emotional situations and to the motivation to smoke is unclear.

Smoking and sleep. Smokers report more sleep disturbances such as insomnia (i.e., a variety of complaints reflecting dissatisfaction with the ability to initiate and maintain sleep, along with a significant reduction in total sleep time followed by daytime sleepiness) (Kaneita et al., 2005; Phillips \& Danner, 1995), and a correlation was found between the level of addiction to cigarettes and reduced quality of sleep (Palmer, Harrison \& Hiorns, 1980; Patten et al., 2000). A deleterious effect of smoking on the quality of sleep could partially result from disturbed pulmonary function (Simon-Tuval et al., 2011). Indeed, heavy smokers are at high risk for Chronic Obstructive Pulmonary Disease (COPD) (Tarasiuk et al., 2006), which causes disturbed sleep leading to reduced quality of life (Scharf et al., 2011; Won \& Kryger, 2014). Moreover, even in young, seemingly healthy, moderate smokers there is a clear reduction in pulmonary function, and a negative correlation was found in healthy adults between pulmonary functioning and sleep quality (Phillips et al., 1989). However, this factor could not solely explain the relationship between smoking and quality of sleep, as studies have shown an increased number of awakenings in the course of the night during early phases of smoking cessation (Hatsukami, Hughes \& Pickens, 1985; Hatsukami et al., 1988).

The reliability of subjective sleep measures is not unequivocal and needs support from objective measurements (Pillar, Malhotra \& Lavie, 2000). To date, only a handful of studies compared the sleep architecture of active and quitting smokers to that of nonsmokers using polysomnographic (PSG) tests (consisting of brain wave, muscle tension, eye movement, and other measurements). These studies confirmed the deleterious effects of smoking and smoking cessation on the quality of sleep. Zhang and colleagues (2006) found that compared to nonsmokers, the sleep of smokers was characterized by shorter duration, longer time to reach (rapid eye movement) REM sleep, longer time spent in light sleep (stage 1), less time spent in deep sleep (stages 3 and 4, slow-wave sleep-SWS), and lower sleep efficiency (percentage of actual sleep time of the total time spent in bed). Similarly, sleep duration among smokers was shorter than among nonsmokers (Jaehne et al., 2012). Likewise, PSG tests conducted on individuals undertaking smoking cessation demonstrated increases in the number of awakenings at night (Prosise et al., 1994), shortening of REM latency, shortening of SWS sleep duration, and extension of the phases of light sleep (stages 1 and 2) (Moreno-Coutiño, Calderón-Ezquerro \& Drucker-Colín, 2007; Wetter et al., 2000), all indications of reduced quality of sleep. Yet, these PSG studies also had limitations as they were usually conducted on a small number of patients on a single night, under the artificial conditions of sleep laboratories.

Quality of sleep, stress, and addiction to smoking. It is well accepted that sleep plays a vital role in health as well as behavioral and emotional stability (Scharf et al., 2010; Tarasiuk et al., 2005). Those who suffer from poor sleep quality exhibit higher rates of psychological stress, depression, and various anxiety symptoms than the general population (Fernández-Mendoza et al., 2009; Ohayon, 2005). The tendency toward poor sleep quality among smokers and quitters, and the correlation between proper sleep and behavioral and emotional stability raise the hypothesis that the impairment in sleep quality among smokers and quitters affects their psychological functioning and their smoking behavior. The few studies that investigated this issue demonstrated that abstaining from cigarette smoking for 48 hours increased the subsequent rate of smoking (Hamidovic\& de Wit, 2009), and that sleep disruptions during withdrawal have a negative effect on the success of smoking cessation (Jaehne et al., 2009; Persico, 1992). These findings seem to integrate well into the aforementioned theoretical model, which holds that exposure to nicotine leads to the development of aversive psychological symptoms when the bodily nicotine concentration drops, which leads to a compulsive urge to smoke to alleviate these unpleasant sensations (Cohen \& George, 2013).

A variety of findings suggest that these aversive psychological symptoms are due to an increase in the activity of neurobiological systems that regulate stress responses (Cohen \& George, 2013). Stress responses are regulated by two key neuroendocrinological systems: the sympathetic nervous system and the HPA (hypothalamus-pituitary-adrenocortical) axis. The sympathetic system is responsible for increasing the level of arousal in situations of danger and acute stress. It involves mainly the secretion of adrenaline by the adrenal gland. Studies indicate that level of the α-amylase enzyme in the saliva is an indirect but reliable measure of sympathetic activity (Rohleder et al., 1994). It has been demonstrated that the α-amylase levels before and after smoking cessation predicted the measure of success in persisting with abstinence from smoking over time (Duskova et al., 2010). The HPA system regulates physiological reactions related to coping with ongoing stressors. Specifically, it induces secretion of the Corticotrophin Releasing Hormone (CRH) from the paraventricular nucleus, located in the hypothalamus, to the anterior pituitary gland, which in response secretes the AdrenoCorticoTropic Hormone (ACTH) that stimulates the release of cortisol by the adrenal gland (Ma et al., 2011). Cortisol levels in the body generally rise sharply in response to stressors, but in people who are in chronic stress situations, regulation of the HPA system may be disrupted, which is manifested by reduced secretion of cortisol both in the basal situation and in response to stress (Miller, Chen \& Zhou, 2007). This sub-regulation may reflect difficulty in coping with stress (Miller, Chen \& Zhou, 2007): low levels of cortisol are associated with an increased level of depression (Moraes et al., 2016), aggression (Granger, 1998), and impulsiveness (Blomqvist et al., 2007). Note that while acute exposure to nicotine increases levels of cortisol temporarily (Newhouse et al., 1990; Winternitz \& Quillen, 1977), secretion of cortisol in chronic smokers in response to stress is reduced relative to that of nonsmokers (Kirschbaum, Strasburger \& Langkrär, 1993; Rohleder \& Kirschbaum, 2006). Moreover, studies indicate a decrease in cortisol levels during smoking cessation (Steptoe \& Ussher, 2006; Targovnik,1989), which correlates with the intensity of the somatic and emotional withdrawal symptoms and with the level of the urge to smoke (Cohen, al'Absi \& Collins, 2004; Targovnik,1989); it also correlates with the level of success in persisting with smoking cessation over time (al'Absi et al.,2004; Frederick et al., 1998). This may be due to the development of sub-regulation by the HPA system, similarly to the situation of those suffering from chronic stress (Richardson et al., 2008), which in turn leads to the sensitization of the mechanisms of stress in the brain, such as increasing the activity of CRH in the limbic system (Vendruscolo et al., 2012). The literature partially supports the possibility that an increase in the activity of neurobiological stress mechanisms (due to chronic exposure to nicotine) contributes to disruption of the quality of sleep. For example, increased sympathetic activity during the day, as reflected in high levels of α-amylase, is highly correlated with insomnia (Nater et al., 2007). However, various studies that have examined the relationship between different measures of cortisol secretion throughout the day and night, and various indices of sleep, produced conflicting results (Elder et al., 2014). Recent studies indicate that cortisol is produced at a higher rate during REM sleep; therefore, prolonged sleep, which includes longer REM sleep, produces higher cortisol levels (Van Lenten \& Doane, 2016). Note that whereas an increase in the levels of stress may disrupt sleep, poor sleep quality may in itself cause stress (Fernández-Mendoza et al., 2009; Ohayon, 2005).

In sum, several studies suggest that smoking and, even more so smoking cessation, disrupt sleep, and that the reduced quality of sleep may contribute to the motivation to smoke and reduce the prospects for quitting. However, the number of these studies is relatively small, and many of them are based on subjective reports, whose reliability is not unequivocal. The minority of studies that used objective measurements usually tested a small number of participants, under the artificial conditions of the sleep laboratory. Notably, the quality of sleep can be measured objectively and reliably for extended periods of time in the participants' natural environment using an actigraph device that is worn on the wrist and analyzes movement (Ancoli-Israel et al., 2015; Sadeh et al., 1989). However, actigraphy has not been used thus far for assessing the sleep quality of smokers. Moreover, to date there has been no systematic study of the complex relations between sleep, pulmonary function, stress systems functioning, and smoking as they relate to the development and persistence of tobacco smoking addiction.

We hypothesize that chronic tobacco smoking impairs sleep quality, which in turn enhances the activity of stress mechanisms, and thus induces negative emotional states. The processes that harm sleep quality and the processes that increase responsiveness to stress and negative emotional symptoms thus reciprocally nourish each other and ultimately increase the urge to smoke as a means for stress relief. Our preliminary findings support this hypothesis by demonstrating that active smokers suffer from diminished quality of sleep (assessed by actigraphy and questionnaires), compared to nonsmokers, an effect that was correlated with markers of stress activation, namely levels of cortisol and α-amylase in the participants' saliva.

Research objectives and expected significance

Our research will explore the role of sleep disturbances in tobacco smoking addiction. Based on our preliminary findings and the available literature, the main hypothesis of this proposal is that smoking, and to a greater degree early phases of smoking cessation, will be associated with reduced quality of sleep. The reduced quality of sleep will, in turn, predict the severity of negative symptoms and the extent of difficulty in abstaining from smoking. Successful abstinence from smoking will lead to normalization of sleep. Experiments to investigate this hypothesis will address the following specific aims:

1. To examine physiological and psychological factors predicting reduced quality of sleep among smokers. The goal is to explore the deleterious effects of smoking on sleep quality and to explore whether the reduced sleep quality is predicted by the following factors: poor pulmonary function (FEV1, FEF), degree of nicotine dependence, emotional symptoms (anxiety and depression), and altered regulation of stress systems (HPA axis and the sympathetic nervous system). We expect that, compared to nonsmokers, smokers will exhibit poorer sleep quality. The level of disruption to the quality of sleep will be related to the intensity of nicotine dependence, poor pulmonary functioning, the levels of negative emotionality, sympathetic activation, and deregulation of HPA axis.
2. To explore the impact of smoking cessation on sleep quality and related symptoms. We will examine whether smokers who try to quit experience a worsening sleep quality, and whether this poor sleep quality can predict the magnitude of their stress response (level of cortisol and α-amylase) and the severity of emotional and cognitive symptoms as measured by the psychometric tests.
3. To explore the role of sleep quality on smoking motivation and relapse to smoking. Experiments will determine whether sleep can predict the urge to smoke (in current smokers) and the likelihood of abstinent smokers to relapse.
4. To explore whether prolonged abstinence from nicotine improves sleep quality. The aim here is to examine whether smoking cessation ultimately induces normalization of sleep and related factors, including pulmonary functioning, stress response (level of cortisol and α-amylase), and emotional and cognitive functioning. Smoking abstinence will be verified by measurement of cotinine in the participant's saliva.

Expected Significance. Experiments will show how reduced quality of sleep may result from chronic smoking and interfere with attempts to quit smoking. This study will shed light on the interrelated physiological and psychological mechanisms that mediate the interplay between smoking addiction and sleep, including psychological distress, dysfunctional stress mechanisms, and reduced pulmonary functioning.

The proposed study will utilize a variety of powerful methods and an interdisciplinary collaboration of experts in the fields of sleep, addiction, and respiratory medicine to explore the interplay of sleep and tobacco smoking. Results of this study will provide novel insight on the role of sleep in nicotine addiction. Findings of the proposed research are expected to promote the use of sleep quality enhancement techniques in smoking cessation interventions. Methodology Participants. Participating in the study will be 150 healthy volunteers, men and women, aged 18-30, with no history of mental illness or drug abuse. The number of participants was established by power analysis, based on our preliminary results, with addition of 20% to compensate for possible drop-outs during the study. Fifty participants will be nonsmokers, and the rest regular smokers . Half of the smokers will be with a stated interest in quitting and half with no such interest (non-quitting control). Participants will receive a compensation fee of 400 NIS (roughly 100 US dollars). All procedures used in this study have been approved by Yezreel Valley College (YVC) Institutional Review Board.

Research procedure. Participants will be recruited from the student body of the college and, via advertisement, from the surrounding communities. All study sessions will start between 7:00 AM and 9:00 AM (up to 30 minutes following awakening) either at the YVC Psychobiology Laboratory or in the participant's home.

The study will include 3 groups (N=50): smokers attempting to quit (smoking cessation group, SCG), nonsmokers group (NSG), and a group of smokers not attempting to quit (SNCG). The latter group is needed in order to assure that changes in certain measures following smoking cessation are not due to time-related events that are not related to abstinence from tobacco. The study design includes 4 stages: A) Baseline. B) First week of nicotine cessation. C +D) Follow up tests three months and six months following the initiation of smoking cessation. Note that although the 2 control groups (i.e., the nonsmokers group (NSG) and the group of smokers not attempting to quit (SNCG)) will not undergo smoking cessation, they will be evaluated with the same tests and at the same time points as the smoking cessation group.

Baseline Phase: At the first session, all the participants will be asked to sign a consent form and provide background information. After assessing their smoking status by checking the level of carbon monoxide (CO) in their breath, they will be asked to give a saliva sample (to test the levels of cortisol and α-amylase) by spitting into a test tube (1.5 mL minimum). To further validate the participants' smoking status, the saliva samples will also be used to detect the levels of cotinine, the primary metabolite of nicotine. Participants will be asked to complete a broad spectrum of questionnaires, evaluating their level of anxiety, depression, quality of sleep, and smoking dependence (questionnaires are described in detail below), as well as completing the Cognitive Assessment Battery (CAB), a computerized neurocognitive test designed to assess a large range of cognitive skills related to executive functions. In addition, their pulmonary functioning will be assessed via spirometry. Finally, participants' sleep will be continuously monitored over a 1-week period by a miniature wrist-worn actigraph, and will be recorded during the last two nights of this week by PSG.

Smoking cessation phase: At the end of the baseline week, participants of the smoking cessation study group will begin abstaining from smoking, while participants on the 2 control groups (i.e., the nonsmokers group (NSG) and the group of smokers not attempting to quit (SNCG) will carry on with their regular routine. In the morning of the first day of smoking cessation, all participants will provide saliva samples and complete the same set of questionnaires as in the baseline phase. All participants will then be asked to wear the actigraphy device during the second week as well, and to return to the lab for additional sessions after 48 hours, 72 hours, 5 days, and one week (altogether four times). At each of these sessions, the smoking status of the participants will be assessed by means of the exhalation test, and participants will again be asked to provide a saliva sample, and complete the same set of questionnaires. On the last two nights of this week, participants' sleep will also be recorded by PSG. On the morning of day 7 of abstinence, all participants will also complete (again) the CAB neurocognitive test. Follow up phase: Will be conducted on all participants three months and six months into the smoking cessation process of the smoking abstinence group. At each of these time points, all participants will be summoned for an additional data collection session in which it will be determined who of the smoking abstinence participants relapsed back to smoking. In addition, all participants will again be asked to complete the set of questionnaires and the CAB neurocognitive test. The pulmonary functioning of the participants will again be measured by spirometry and they will be asked to wear the actigraph device for one week. In addition, their sleep will be recorded for two consecutive nights by PSG. Saliva samples collected during the study will destroyed following their ELISA analysis. All participants will be assured of confidentiality and anonymity, and all staff involved in the study will maintain the confidentiality of the study. All material collected will be kept locked in a designated cabinet.