Cardiometabolic Morbidity and Mortality with Smoking Cessation, Review of Recommendations for People with Diabetes and Obesity

Purpose of Review Obesity is closely linked with the pathogenesis of type 2 diabetes (T2DM) and cardiovascular disease (CVD), and whilst smoking cessation is associated with weight gain, there are concerns that this weight gain may offset the benefit of CVD risk reduction especially in those with considerable post-cessation weight gain. The aim of this narrative review is to evaluate recent evidence on smoking cessation and cardiometabolic outcomes and discuss limitations of current knowledge and studies. Recent Findings Nicotine is a key player in modulating energy balance by influencing lipid storage in adipose tissue by affecting lipolysis, energy input by modulating appetite and energy output by increasing sympathetic drive and thermogenesis. It also increases insulin resistance and promotes abdominal obesity. The CVD risk and mortality associated with cigarette smoking potentiate the CVD risks in patients with diabetes. Evidence supports the benefit of quitting cigarette smoking regardless of any subsequent weight gain. Data suggests that the cardiometabolic risk is limited to the first few years and that cardiovascular health and mortality benefit of smoking cessation outweighs the harm related to weight gain. This weight gain can be limited by nicotine replacement of which e-cigarettes (vaping) are increasingly popular if it is not an alternative to cigarette smoking. However, long-term health data on e-cigarettes is needed prior to formal recommendation for its use in smoking cessation. Summary The recommendation for cessation of cigarette smoking is justified for those at high risk of weight gain and diabetes. However, for most benefit, consideration should be given for personalized weight management to limit weight gain. Awareness of a ‘lean paradox’ by which lower weight is associated with increased CVD risk may help to improve motivation and insight into the bias of smoking, health and body composition otherwise known to epidemiologists as the ‘obesity paradox’.


Introduction
Smoking is pro-atherogenic and accelerates the progression of macro and microvascular disease and is an independent risk factor for premature mortality in patients with diabetes [1]. People with diabetes are a high-risk group for cardiovascular disease (CVD) even without smoking, mainly attributed to the chronic effects of hyperglycaemia. The American Diabetes Association and the United Kingdom's National Institute of Health and Care (NICE) recommendation for subjects with diabetes is cigarette smoking cessation [2,3]. However, whilst some weight (re)gain is inevitable, the efficacy of abstinence is greatly compromised by the unwillingness of smokers to gain even modest amounts of weight with quitting [4]. Obesity itself is a major independent risk factor for type 2 diabetes (T2DM) and cardiovascular disease to the extent of an exponentially increase of T2DM risk with each unit of BMI [5,6]. The increasing rates of obesity in the USA have partially been attributed to a decline in smoking popularity [7]. Cigarette smoking rates in the USA were 45% in 1965 and 15.5% in 2016, and 5-6% of current smokers use e-cigarettes [1,8,9]. Smoking has been associated with the bias explaining the 'obesity paradox' [10]. Smoking increases adipose tissue breakdown (lipolysis), induces insulin resistance and leads to weight loss. When not accounting for smokers, the general population with obesity may appear to have lower CVD mortality than lean subjects such as those who smoke.
The conundrum of post-cessation weight gain and health has stimulated research and discussions on the cardiovascular benefit of smoking cessation in those who gain weight in excess [11]. This review explores the evidence of the interplay of diabetes and CVD morbidity and mortality with smoking cessation in the context of obesity and explores any potential bias. It challenges the hypothesis that weight gain after quitting smoking worsens diabetes and cardiovascular risk disproportionally to the extent that health-care professionals should be cautious about recommending smoking cessation.

Smoking, Smoking Cessation and Weight
Smokers lose subcutaneous body fat through increased fat tissue breakdown (lipolysis) and a reduction in adipose tissue synthesis (lipogenesis) [12]. The fat distribution however appears to change unfavourably with an increase of the abdominal fat depot. Abdominal obesity is a parameter of the metabolic syndrome used in cardiovascular risk assessment, albeit the abdominal circumference parameter was a later addition to the original syndrome description by Reaven who, when first describing it, linked the pathogenesis of cardiovascular disease with insulin resistance [13,14]. The degree of change in the abdominal fat distribution appears to have an association with the amount of cigarettes smoked as shown in a crosssectional population study of 6123 Caucasians with the observation that with increasing cigarette usage, the odds ratio (OR) for abdominal obesity in men increased to 1.28 (95% confidence interval (CI): 0.78-2.10) for moderate smokers and 1.94 (1.15-3.27) for heavy smokers (P = 0.03 for trend) and 1.07 (0.72-1.58) and 2.15 (1.26-3.64) in female moderate and heavy smokers, respectively (P < 0.01 for trend) [15].
Apart from carcinogenic tar, cigarettes contain nicotine, which suppresses appetite as confirmed in rat experiments by a decrease in their meal size and lengthening of intervals between meals [16]. This is mediated by nicotine affecting appetite regulating hormones (peptide YY, orexin, neuropeptide Y), activation of proopiomelanocortin (POMC) and changes in expression of various hypothalamic receptors [12]. Nicotine also affects adipose tissue metabolism directly by an increase in lipolysis and affects lipases which regulate deposition of free fatty acids into adipose tissue as well as the secretion of fat hormones such as adiponectin [12]. A resulting surplus of FFA leads to lipotoxicity and contributes to insulin resistance.
Furthermore, nicotine increases neurotransmitter release and stimulates the sympathetic drive through which it can activate thermogenesis of brown adipose tissue and which further stimulates weight loss. It acts through an upregulation of cholinergic nicotinic receptors (nACHRs) peripherally and centrally. These receptors are also expressed in adipose tissue and have been discussed as potential drug targets for the treatment of obesity [17]. Nicotine also stimulates dopaminergic neurons of the lateral hypothalamus which are part of the reward system and induces addiction [18].
With quitting smoking and without treatment or drugs to assist cessation, a meta-analyses of 62 studies reported an average weight gain of 4.67 kg (95% confidence interval: CI 3.96-5.38) 12 months after continuous abstinence [19]. Results from the same meta-analysis showed that the weight gain is noticeable from the first months of cessation and progressively gained in the first year of quitting [19], suggesting that the weight gain is a direct result of abstinence. Longitudinal studies relating to cardiometabolic risk typically consider the weight regain at 6 to 8 years post-cessation [20••, 21••], albeit about 60% of weight regain occurs in the first year [19,22].
Greater weight gain is found in previously heavier smokers (high Fagerström Score for nicotine dependence) and those with higher BMI before quitting, in women and those below age of 55 [19,[23][24][25]. More prone to post-cessation weight gain are smokers of low socio-economic status. Weight gain also depends on geographical region (higher in North America than Asia) and is more frequent in former smokers who control negative emotions with temporary gratification [12,26,27] Frequent unsuccessful attempts to quit smoking could lead to weight cycling and higher weight regain after every episode [26]. Reports on the proportion of people who gain more than 10 kg range from 13 to 15% [19, 21••, 24]. A study using selfreported weight gain showed that of those classified as historic binge eaters gained on average 11.4 kg (SD = 11.2) in comparison to 5.0 kg (SD = 6.9) in non-binge eating obese subjects, suggesting that former smokers may have replaced binge eating with smoking as a coping strategy [28].
Twin studies refute a contribution of genetic factors in these weight changes [29]. The weight change with smoking and its cessation is mediated largely through the cumulative actions of nicotine (or its withdrawal) [12]. Nicotine withdrawal increases calorie intake in the range of 250-300 kcals per day [26]. As mentioned above, nicotine affects fat hormones and storage in adipose tissue [30], and its withdrawal slows the metabolic rate, resets appetite and promotes addictive behaviour [12].

Smoking and Metabolic Risk/Diabetes
A large meta-analysis including 88 prospective cohort studies concluded that active smoking was associated with an increased risk of incident T2DM, with a pooled relative risk (RR) of 1.37 (95% CI: 1.3-1.4) compared with never smokers [31]. Similar increases were found in reviews and reports from European, Asian and Japanese populations for incident diabetes as well as prediabetes [32,33]. Globally, about 12-19% of T2DM in men and 3-5% in women were estimated to be attributable to smoking [31,34]. About 27 million diabetes cases worldwide were linked directly to active smoking [31]. A meta-analysis of 22 prospective observational studies in Japanese subjects showed that T2DM increased by 16% for each increment of 10 cigarettes smoked per day [34]. A systematic review and meta-analysis concluded that people with diabetes who smoke have a lower HbA1C and less favourable lipid profile [35].
From experimental studies, we learn that smoking cigarettes decreases insulin sensitivity such as in young healthy individuals, smokers and non-smokers and worsens insulin resistance in those with existing type 2 diabetes [36][37][38].
In vivo experiments in mice exposed to nicotine had reduced insulin sensitivity [39]. Nicotine-mediated activation of AMPKα2 kinase and the respective increase in lipolysis in conjunction with weight loss is thought to contribute to the increased insulin resistance of cigarette smoking [10,39]. Whilst there may be several factors to explain dyslipididaemia in humans [40], a clinical study on 11 heavy smokers who used transdermal nicotine replacement for cessation showed that insulin sensitivity only improved after nicotine withdrawal irrespective of weight gain [41], suggesting nicotine as a key player in its pathogenesis.

Smoking Cessation, Diabetes and Weight Change
The risk of new-onset diabetes continues to increase in the first few years after smoking cessation compared to never smokers and then begins to decline, and the risk becomes comparable to those who never smoked. In the metaanalysis of Pan et al. which included 10 prospective studies (1,086,608 participants) comparing never smokers with quitters, the risk of diabetes was increased, with a RR = 1.54 (95% CI: 1.36-1.74) for those who had quit smoking less than 5 years before baseline, decreased progressively for those who stopped for 5-9 years to a RR of 1.18 and to a RR of 1.11 for those abstaining for ≥ 10 years [31]. This is similar to the findings from a meta-analysis of Japanese employees (n = 343,573) which showed a decline of diabetes risk reaching an RR of 1.00 (95% CI: 0.88-1.13) for long-term quitters (≥ 10 years) in comparison to never smokers; however, the influence of weight change was not explored [34]. This shortterm elevated risk for T2DM is higher in the Asian population than in North Americans or Europeans [31].
The analysis of data from US populations including the Nurses' Health Study (NHS), NHS II and the Health Professionals Follow-Up Study (HPFS) by Hu et al. on progression of diabetes in those without prevalent diabetes, CVD or any chronic disease at baseline [21••] found that the diabetes risk is in direct proportion to weight gain after smoking cessation (weight change within 6 years of cessation explained 68% of elevated diabetes risk in the mediation analysis). This risk was highest in the quitters who gained more than 10 kg (hazard ratio (HR) = 1.59 (95% CI: 1.36-1.85)), peaked at 5 to 7 years from quitting and was overall higher in former smokers than in current smokers (HR of 1.22, 1.12-1.32). The weight gain did not attenuate the apparent benefits of smoking cessation on reducing cardiovascular mortality or extending longevity (see Table 1). The diabetes risk was not increased in those without weight gain [21••].
When considering glycaemic control, a Chinese prospective study showed that the risk of poor glycaemic control was higher in smokers, and this risk normalized after at least 10 years of abstinence [46]. The same pattern was found for fasting glucose and HbA1C [46]. In comparison, studies of shorter duration did not find any difference in glycaemic control after stopping smoking for 1 year [47]. Evidence reviews report that quitting smoking reduces macrovascular diabetes complications and nephropathy; the evidence however is less conclusive in regard to other microvascular complications [48].

Smoking Cessation and CVD in Relation to Weight Change
Cigarette smoking is independently associated with worsening atherosclerosis/hypertension, dyslipidaemia and increased cardiovascular morbidity and mortality. Smoking cessation reduces these risks [49]. The multiple effects of nicotine on adipose tissue itself play an important role in the pathogenesis of atherosclerosis [50].
Liu and colleagues examined the risk of CVD and mortality in relation to weight gain after smoking cessation in healthcare professionals with T2DM from the Nurses' Health Study   had diabetes at baseline) with data collected from 1984 to 2011 for the 6-year incidence of CVD events in smokers and quitters who lost or gained weight. Due to the small numbers of events, the subgroup analyses considering weight gain was limited but suggested a CVD risk benefit in diabetic participants regardless of weight gain qualitatively similar to non-diabetic subjects; however, this did not reach statistical significance, possibly due to limited study power [44]. A study in postmenopausal women in the Women's Health Initiative study (WHI, n = 161,808; see Table 1) was similarly not able to make conclusions in the diabetes subgroups (5-7% of the population had diabetes) but showed improvement in CHD with smoking cessation in people with no diabetes regardless of weight gain [45]. A substantial improvement was also reported in all-cause mortality regardless of post-cessation weight. Liu et al. report all-cause mortality among long-term quitters > 6 years since smoking cessation with a multivariate-adjusted HR of 0.69 (95% CI: 0.58-0.82) among those without weight gain, 0.57 (95% CI: 0.45-0.71) among those with weight gain of 0.1-5.0 kg and 0.51(95% CI: 0.42-0.62) among those with weight gain of > 5·0 kg [20••]. Key studies considering post-cessation weight gain are shown in Table 1.

E-cigarette Smoking
E-cigarettes are gaining popularity with 60% of smokers in the USA using e-cigarettes at least on some days if not regularly [52,53]. A UK survey found that about 6% of people used e-cigarettes in 2017, most of them (52%) were previous smokers [54]. E-cigarettes also known as electronic nicotine delivery systems (ENDS) are battery-driven devices that vaporize the nicotine present in a refill liquid. Vaping has become the most popular method for smoking cessation [9]. A worldwide online survey in 10 languages showed that many people use ecigarettes in combination to cigarette smoking as dual use [55]. The same survey found that only 3.5% of smokers use non-nicotine containing vapes [55].
Nicotine replacement has been used since the 1980s commonly as dermal patches, inhalers, gum, etc. and found effective for improving the symptoms of smoking cessation [12]. This has since been replicated with the use of nicotine containing e-cigarettes which ameliorate post-cessation weight gain [56] and maybe more effective than traditional nicotine replacement therapies in quitting success as shown in smokers attending UK NHS smoking services who achieved a 1-year abstinence of 18% when nicotine e-cigarettes were combined with behaviour support [57]. A New Zealand survey showed that nicotine e-cigarettes lead to abstinence in more people (7.3%) at 6 months than placebo e-cigarettes (risk difference 3.16 (95% CI: − 2·29 to 8.61) [58]. Cohort studies however report only a modest quit rate which is no different in dual users [59,60]. A recent meta-analyses [61] concluded that there is only very limited evidence regarding their impact on smoking cessation [54]. This is in agreement with a recent survey in England, which shows that E-cigarette use is associated with only a modest increased quit rate [62].
Recent surveys show that many quitters are not aware of potential weight benefits with use of nicotine e-cigarettes [53,63]. The quit rate especially in people with diabetes is low, and whether the awareness of weight benefit could improve quit rates remains to be shown. A survey of English smokers concluded that one in eight people would stop smoking if they were convinced that vaping prevented gain weight [63]. Several studies on vaping in regards its weight benefit are ongoing, and some researchers discuss nicotine vaping as intervention for the management of obesity per se [63][64][65].
Challenges for the assessment of e-cigarettes' efficiency and metabolic effects are the appreciation of nicotine dosage. Their content varies, and regulations differ among countries, e.g. a Juul pod in the USA can contain up to 59 mg/mL of nicotine compared with 18 mg/mL in the UK, whilst a cigarette contains average of 10-12 mg, of which only 1% is estimated to be inhaled [66]. In addition, it is difficult to estimate the rate of absorbance of nicotine as this depends on the vaping technique, and modern devices appear to allow a larger absorbance [55]. Such a direct comparison of any vaping 'unit' with corresponding number of cigarettes for equal absorption of nicotine is not possible.
Vaping carries health risks as demonstrated by increasing number of reports of acute lung injury (> 1900 in the USA). These present as infiltrates on chest imaging and are attributed to various toxic compounds and ingredients [67][68][69]. The exact pathophysiology of acute lung inflammation is yet unclear nor are long-term health effects of vaping [70]. This is particularly important for those choosing vaping for the long term rather than for smoking cessation per se and increasing number of people not using e-cigarettes for this purpose [20••, 21••]. As outlined above, whilst nicotine impairs glucose tolerance and influences the vascular system with an increased sympathetic activation, concerns have been raised regarding worsening glucose impairment and cardiovascular risk with nicotine replacements [41,71]. Reports on mice and humans so far do not detect changes in insulin resistance with ecigarette use [72]; however, this may be nicotine dose dependent, and smokers using nicotine containing e-cigarettes may expose themselves with lower doses than with tobacco containing classical cigarettes. E-cigarettes are not approved by the FDA at time of writing this review, and the NICE review in 2018 advises the need of more evidence on long-term safety prior to a possible recommendation [73].

Evidence Limitations
Limitations of studies on smoking cessation and cardiometabolic risk include fairly small sample sizes, especially in subgroups of high weight gain and diabetes. Some studies use point prevalence measures of non-smoking status rather than capturing dynamic changes in smoking status. Even those who manage abstinence for 1 year, about 30% have been reported to relapse in the following decade [74]. Conditions which affect weight change and lifestyle behaviour need to be excluded to avoid bias by reverse causality such as the 'sick quitter effect' [11]. Abrupt smoking cessation and complete abstinence may be more typical of a person with a new diagnosis of a health condition. Many studies are based on self-reporting; this can include smoking behaviour; however, using objective measure of smoking, e.g. cotinine levels, is preferable. Data on weight changes and often the presence and duration of diabetes are also frequently obtained by self-reporting. The diabetes control at time of smoking cessation should be taken into account as this will influence future weight.
With nicotine being a key player in modulating weight change, we propose that the dosage of nicotine should be taken into account when evaluating efficacy and risks. Using the basic energy balance formulation used in the nutrition and physiology literature and adjusting it to nicotine (see Fig. 1) [75], its independent effects on all variables of energy balance become apparent: energy intake (by reducing appetite), energy expenditure (e.g. by increasing sympathetic drive) and energy storage (by affecting adipose tissue's fat deposition enzymes). Mathematical modelling studies of changes in energy balance and respective weight change need to accept the physiological impact of nicotine on energy metabolism. An obese smoker weighs less with heavy cigarette smoking, and the leaner smoking state is associated with higher CVD morbidity and mortality than the obese state (lean paradox = the lean habitus does not appear to offer protection from CVD otherwise found with lower BMI). There is a natural weight gain with ageing which should be taken into account when referring to quitting induced weight gain. The weight differences with smoking cessation are notified by: A which refers to the weight gain with smoking cessation which is not taking account the weight difference from the pre-smoking weight shown with B. C takes account for the age-related weight gain. *Energy balance equation adjusted to smoking in which R is the rate of kcal/d that is stored or lost (and thus defines weight change), I is the intake of kcal/d, E is the rate of kcal/d expended and N is the nicotine dose (which does not reach 0 as nicotine is a natural ingredient of vegetables) Use of smoking cessation strategies even if they have been used only intermittently or combined needs to be appreciated when referring to weight or glucose metabolism studies. Whilst no increased CVD risk has been observed with the use of drugs assisting smoking cessation such as bupropion and varenicline [33], they may affect post-cessation weight gain.

Conclusion and Future Perspective
Stopping cigarette smoking reduces cardiovascular risk and mortality even in those with weight gain as result of smoking cessation, albeit ameliorating the benefit when considerate weight is gained. Smoking cessation thus benefits people with and without diabetes. As shown in Fig. 1, in order to estimate the cardiometabolic benefit, the ill effects of smoking need to be taking into account. As regards post-cessation weight gain, weight changes should to be related to the pre-smoking baseline weight corrected for weight gain with ageing once quitting commences. According to epidemiological studies, the weight gain through the middle adult age is estimated to be 0.5 to 1 kg per year [76]. The obesity paradox by which obesity is favourable over the lean smoking state holds true with smoking and CVD risk. Epidemiological studies have to correct for smoking as a strong statistical confounder to avoid bias in the relationship of BMI and cardiometabolic health [10]. In doing so, the assumption of an 'obesity paradox' can also be removed in ageing populations [77]. The considerable effect of nicotine on energy balance and appetite, by which it influences body weight, and the risk of diabetes, especially in heavy smokers (> 25 cigarettes a day), becomes apparent by this paradoxical BMI and CVD risk correlation [10,23]. To the individual smoker, however, the term 'lean paradox' might be easier to understand and aid to the acceptance of post quitting weight gain by explaining that the lower smokers' body weight is unfavourable and as the evidence presented suggests carries a higher CVD risk. Patients can be reassured that whilst some weight gain with smoking cessation is physiological, it is modifiable and the 'lesser evil' in comparison to cigarette smoking.
A proactive approach is needed with appropriate personalized cessation strategies to limit weight regain, especially for those at particular risk such as for heavy smokers with obesity and diabetes, and subjects who require emotional control are prone to binge eating. Studies are underway to trial scalable psychological interventions such as cognitive behaviour therapies and mindfulness-based therapies [78]. Whilst initial findings do not support a worsening in insulin resistance, little is yet known about the cardiometabolic and health risks of long-term e-cigarette use [73].
Emerging evidence from a meta-analysis shows that smoking depresses pulmonary immune function and increases progression of the COVID-19 infection with an OR of 1.91 (CI: 1.42-2.59) [79]. As this appears to be mediated by the nicotinergic nACHR7 receptor, it cannot be excluded that the use of nicotine-based products in general increases this risk [80]. Time will show whether the COVID-19 pandemic will motivate to healthier lifestyles and for quitting.

Compliance with Ethical Standards
Conflict of Interest K Kos does declare that there is no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.