Smoking and diabetes—The double health hazard!

Article Outline

• Abstract
• 1. Introduction
  • 1.1. Smoking and the incidence of diabetes
• 2. Risk of smoking for mortality
• 3. Risk of smoking for fatal and non-fatal cardiovascular events
• 4. Pathophysiological mechanisms linking smoking and diabetes
• 5. Effects estimated by smoking cessation programmes
• 6. Conclusion
• Conflicts of interest
• References
• Copyright

Abstract 

Smoking is a predictor of the transition from normoglycaemia to impaired fasting glucose and increases the risk of type 2 diabetes, independent from possible confounders. In patients with diabetes as in non-diabetics, smoking is a significant and independent risk factor for all-cause mortality and for mortality from cardiovascular disease and corononary heart disease, as well as for aggregates of fatal and non-fatal cardiovascular events. There is little doubt that smoking is a risk factor for coronary heart disease, but this risk appears to be stronger than the risk for stroke in diabetics. Pathophysiological mechanisms by which smoking causes glucose intolerance and worsens clinical outcomes in established diabetes include greater insulin resistance, impaired beta-cell function and insulin secretion, chronic low-grade inflammation, endothelial dysfunction, as well as interacting indirectly with other factors known to aggravate diabetes and lifestyle factors. Smoking cessation programs are of great importance for primary care specialists dealing with diabetes.

Keywords: Cardiovascular disease, Coronary heart disease, Diabetes, Mortality, Smoking, Stroke

Back to Article Outline

1. Introduction 

Smoking and diabetes are two important hazards to the health of many individuals and contribute substantially to the global burden of disease in various ways. Smoking can not only aggravate the diabetes complications such as macro- or micro-vascular disease, but has also been shown to deteriorate glucose metabolism in normal subjects and thereby may provoke the onset of type 2 diabetes [1].

Smoking is associated with increased mortality and cardiovascular events in diabetes. To investigate these effects of smoking, and the interaction between the effects of smoking and diabetes, we conducted a literature search with the PubMed computerised database for relevant studies and examined the reference lists of the original articles and reviews on the topic. Selection criteria for inclusion were as follows: studies on subjects with type 2 diabetes or with undefined diabetes (assuming a majority of patients with type 2 diabetes), with or without inclusion of non-diabetic subjects, which assessed the relative risk of smoking, taking into account age, sex and other relevant covariates in unequivocal multivariable analyses.

As most patients with diabetes have type 2 and are followed within primary health care, it is of great importance not only to investigate these relationships but also to find effective methods for smoking cessation, e.g. based on invitations based on smoking status in medical records [2]. This should be one of the main targets for preventive efforts as carried out by the primary health care diabetes team.

1.1. Smoking and the incidence of diabetes 

Cigarette smoking is one of the most health hazardous lifestyle habits known in medicine. Besides the well-documented risk for cardiovascular disease, cancer and chronic obstructive pulmonary disease (COPD) there is also evidence for an increased risk of development of type 2 diabetes in heavy smokers, as shown in several epidemiological studies from various populations. One of the first observational studies was that of Rimm et al. [3]. These authors investigated the diabetes risk in a health professionals’ follow-up study across the United States consisting of more than 40,000 male health professionals aged 40–75 years and free of diabetes, cardiovascular disease, and cancer at baseline, followed up for six years. The incidence of type 2 diabetes was studied in 509 men diagnosed with diabetes. After controlling for known risk factors men who smoked 25 or more cigarettes daily had a relative risk (RR) of diabetes of 1.94 (95% confidence interval; CI: 1.25–3.03) compared with non-smokers.

In another study by Manson et al. [4], 21,068 US male physicians aged 40–84 years, initially free of diagnosed diabetes mellitus, cardiovascular disease, and cancer, were studied in the Physicians’ Health Study. During follow-up a total of 770 new cases of type 2 diabetes mellitus were identified. Smokers had a dose-dependent increased risk of developing type 2 diabetes mellitus: compared with never smokers, the age-adjusted RR was 1.2 (95% CI: 1.0–1.4) for ex-smokers, 1.4 (95% CI: 1.0–2.0) for current smokers of <20 cigarettes per day, and 2.1 (95% CI: 1.7–2.6) for current smokers of ≥20 cigarettes per day. After multivariable adjustment for body mass index, physical activity, and other risk factors, the RRs were 1.1 (95% CI: 1.0–1.4) for past smokers, 1.5 (95% CI: 1.0–2.2) for current smokers of <20 cigarettes per day, and 1.7 (95% CI: 1.3–2.3) for current smokers of ≥20 cigarettes per day. Total pack-years of cigarette smoking was also associated with the risk of diabetes (P for trend <0.001).

In 2007, a meta-analysis on this topic was published by Willi et al. [5]. In total, 25 prospective cohort studies (with 1.2 million participants) were found after systematic searches. These studies reported 45,844 incident cases of diabetes during a study follow-up period ranging from 5 to 30 years. Of the 25 studies, 24 reported adjusted RRs greater than 1 (range for all studies, 0.82–3.74). The pooled adjusted RR was 1.44 (95% CI: 1.31–1.58). Results were consistent and statistically significant in all subgroups. The risk of diabetes was greater for heavy smokers (≥20 cigarettes/day; RR, 1.61; 95% CI: 1.43–1.80) than for lighter smokers (RR, 1.29; 95% CI: 1.13–1.48) and lower for former smokers (RR, 1.23; 95% CI: 1.14–1.33) compared with active smokers, consistent with a dose-response phenomenon. The authors concluded that active smoking is associated with an increased risk of type 2 diabetes and added that future research should attempt to establish whether this association is causal and to clarify its mechanisms [4].

In a prospective study in Malmö, Sweden, Lyssenko et al. [6] recently published that smoking is an independent predictor of incident type 2 diabetes based on data from a very large cohort of 16,061 Swedish and 2770 Finnish subjects. Type 2 diabetes developed in 2201 (11.7%) of these subjects during a median follow-up period of 23.5 years. Smoking remained a predictive factor, RR 1.39 (95% CI: 1.19–1.61), after adjusting for both biological and genetic predictors. A further gene–environment interaction analysis is planned.

Finally, it has also been shown that smoking is a predictor of the transition from normoglycaemia to impaired fasting glucose (IFG) in a US study by Rafalson et al. [7]. In total, 1455 healthy participants (mean age, 56.5 years; range, 35–79 years) were investigated at baseline and later re-examined. Incident IFG was defined as a subject whose baseline fasting plasma glucose was <100mg/dL (normoglycaemic) and between 100 and 125mg/dL at follow-up. Of the 1455 participants, 924 were normoglycaemic at baseline: 101/924 converted to IFG over 6 years. Compared with those who remained normoglycaemic, converters to IFG were at baseline older, had a larger body mass index, more likely to be hypertensive, currently smoke, and have a family history of type 2 diabetes mellitus (all P<0.05). Multivariable logistic regression analysis demonstrated that compared with subjects who remained normoglycaemic, the odds ratio of incident IFG among former and current smokers (versus never smokers) was 1.68 (95% CI: 0.99–2.80) and 2.35 (95% CI: 1.17–4.72) (P for trend=0.008), respectively.

In summary, it is now well established that smoking increases the risk for deterioration of glucose metabolism and also type 2 diabetes in normoglycaemic individuals independent of possible confounders. Some potential mechanisms have been proposed (Table 1). However, there could well exist some rest-confounding bias associated with poor lifestyle (diet, physical activity) as these important variables are often difficult to measure in large population-based studies. It is for example well known that smokers tend to prefer a different diet than non-smokers, containing more saturated fat, coffee and alcohol, and less fruit, fibre and vegetables [8]. This strengthens arguments to stop smoking especially in subjects with a positive family history of type 2 diabetes.

Table 1. Suggested mechanisms for the influence of smoking on risk of type 2 diabetes.

Back to Article Outline

2. Risk of smoking for mortality 

Smoking has been shown to be a significant and independent risk factor for all-cause mortality [9], [10] and for mortality from cardiovascular disease [10], [11] and coronary heart disease [9], [12] in diabetes. Smokers die on average 8–10 years younger than non-smokers as age is entered into most multi-regression analyses. Ford and DeStefano [9] assessed the effect of smoking in 602 diabetics and 12,562 non-diabetics in the NHANES study. The adjusted RR for all-cause mortality of current smokers versus never smokers amounted to 1.79 (95% CI: 1.10–2.91) in diabetics and to 1.60 (95% CI: 1.34–1.90) in non-diabetics. The RRs for coronary heart disease mortality were, respectively, 2.49 (95% CI: 0.94–6.59) and 1.71 (95% CI: 1.19–2.47). The statistical interactions of smoking and diabetes were not significant, indicating that the strength of the associations between smoking and mortality did not differ significantly among persons with and without diabetes. Finally, mortality risk was similar in former smokers and never smokers. The risk of current cigarette smoking for coronary heart disease mortality in diabetics was confirmed in a case control study by DeStefano et al. [12]. Smoking was associated with a RR of 1.5 (95% CI: 0.9–2.5) in younger diabetic men and women and of 1.8 (95% CI: 1.3–2.6) in older diabetics, after controlling for age, sex, race, education, marital status, income, state of residence and other risk factors within each age group.

Al-Delaimy et al. assessed the relationship between cigarette smoking and mortality among 7401 women with type 2 diabetes in the Nurses’ Health Study [10]. In multivariable analyses, adjusting for age, history of high blood pressure and high cholesterol, and other cardiovascular risk factors, compared with never smokers, the RRs of mortality were 1.31 (95% CI: 1.11–1.55) for past smokers, 1.43 (95% CI: 0.96–2.14) for current smokers of 1–14 cigarettes/day, 1.64 (95% CI: 1.24–2.17) for smokers of 15–34 cigarettes/day, and 2.19 (95% CI: 1.32–3.65) for smokers of ≥35 cigarettes/day (P for trend <0.001). The test for interaction between diabetes status and smoking categories was not significant (P=0.83). The RRs for cardiovascular mortality were, respectively, 1.30 (95% CI: 1.00–1.68), 1.58 (95% CI: 0.82–3.07), 2.56 (95% CI: 1.69–3.88) and 1.85 (95% CI: 0.74–4.64) (P<0.001). It is noteworthy that women who had stopped smoking for ≥10 years had only a slight non-significant increase in mortality risk (1.11 (95% CI: 0.92–1.35)) compared with never smokers. These findings suggest a dose-dependent effect of smoking and indeed length of smoking abstinence in reducing risk, which implies causality.

Fuller et al. [11] reported on cardiovascular mortality in the WHO multinational study of vascular disease in diabetes mellitus. In multivariable analyses of RRs for cardiovascular disease mortality, including age, duration of diabetes, systolic blood pressure, serum cholesterol, smoking, proteinuria, retinopathy and ECG abnormalities, the RR of smokers versus non-smokers amounted to 1.6 (95% CI: 1.1–2.4) in 1502 men and 1.7 (95% CI: 1.2–2.4) in 1677 women with type 2 diabetes.

In conclusion, just as is seen in nearly all other patient groups, current smoking is a significant risk factor for all-cause mortality and for mortality from cardiovascular disease and coronary heart disease. Despite smokers dying earlier than non-smokers, the lifetime risk of diabetes in smokers is higher than in non-smokers suggesting a direct causal relationship with smoking and diabetes.

Back to Article Outline

3. Risk of smoking for fatal and non-fatal cardiovascular events 

Al-Delaimy et al. [13] and Cederholm et al. [14] investigated the effect of smoking on composite cardiovascular disease that is stroke and coronary heart disease, whichever came first, in patients with type 2 diabetes. In 6547 women from the Nurses’ Health Study, the multivariable RRs of cardiovascular disease were 1.03 (95% CI: 0.86–1.25) for past smokers, 1.46 (95% CI: 1.02–2.10) for current smokers of 1–14 cigarettes/day and 2.42 (95% CI: 1.94–3.02) for smokers of ≥15 cigarettes/day (P for trend<0.001) [13]. An analysis on 11,646 female and male patients from the Swedish National Diabetes Register revealed that the RR of smoking, including those who stopped smoking within the past 3 months, amounted to 1.49 (95% CI: 1.31–1.69), with adjustment for age, sex, diabetes duration, HbA1c, body mass index, systolic blood pressure, and antihypertensive and lipid-lowering treatment [14]. Therefore these 2 studies identify smoking as a significant and independent risk factor for composite cardiovascular disease.

There is little doubt that smoking is a risk factor for coronary heart disease in patients with type 2 diabetes [13], [15], [16], [17]. In the London cohort of the WHO multinational study of vascular disease in diabetics (n=497) the adjusted RR of smoking for all new ischaemic heart disease was 1.9 (95% CI: 1.1–3.4) [15]. In 2693 patients from the United Kingdom prospective diabetes study (UKPDS), the adjusted hazard ratio of smoking for coronary heart disease, that is myocardial infarction and angina pectoris associated with ECG changes, amounted to 1.41 (95% CI: 1.06–1.88) [16]. In a subsequent analysis on 4540 patients from the same study the risk of smoking was estimated at 1.35 (95% CI: 1.11–1.50) [9].

Kothari et al. [18] reported on the risk of stroke in the UKPDS study, and identified duration of diabetes, age, sex, smoking, systolic blood pressure, total cholesterol to high-density lipoprotein cholesterol ratio and presence of atrial fibrillation as significant independent risk factors. The risk of smoking at diagnosis of diabetes for stroke was 1.55 (95% CI: 1.08–2.01). By contrast, smoking was not a significant risk factor for stroke in another smaller study [15].

The report by Al-Delaimy et al. [13] is of particular interest because these investigators analysed the RR of smoking for coronary heart disease and stroke in the same study population (Nurses’ Health Study). As shown in Table 2 risks for both outcomes were not different between past smokers and never smokers. In smokers who smoked 1–14 cigarettes/day the risk was significant for coronary heart disease but not for stroke. In those who smoked ≥15 cigarettes/day, the RR for coronary heart disease was substantially stronger than the risk for stroke. This trend was confirmed in a more recent analysis of the Swedish National Diabetes Register, in which the RR associated with smoking was higher for myocardial infarction (2.33 (95% CI: 1.74–3.11)) than for stroke (1.12 (95% CI: 0.76–1.65)) in middle-aged patients (30–59 years old), but this was not the case in patients aged 60–74 years, in whom the RR for both outcomes was about 1.4 (P<0.01 for each outcome) [19].

Table 2. Risk of smoking for coronary heart disease (CHD) and stroke in the Nurses’ Health Study [adapted from reference [13]).

	CHDa	Stroke
Never smokers	1.00	1.00
Past smokers	1.21 (0.97–1.51)	0.69 (0.48–1.00)
Current smokers
1–14cig/day	1.66 (1.10–2.52)	1.04 (0.50–2.17)
≥15cig/day	2.68 (2.07–3.48)	1.84 (1.21–2.81)
P for trend	<0.001	=0.004

These observational studies were not designed to explain why the observed relative risk of smoking tended to be higher for myocardial infarction than for stroke and did not attempt to explain the difference.

Back to Article Outline

4. Pathophysiological mechanisms linking smoking and diabetes 

Several mechanisms have been proposed and also investigated to explain the association between smoking and diabetes risk, or the aggravation of diabetes complications, as summarized in a review by Eliasson [20] and summarized in Table 1. It is well known that smokers as a group tend to be leaner than non-smokers, but on the other hand there is an increased tendency for accumulation of abdominal fat. This is related to increased production of adipokines, cytokines and other markers of chronic inflammation. In addition, it has been shown that smokers are generally more insulin resistant than non-smokers and exhibit several aspects of the insulin resistance syndrome [21]. This could contribute to the increased risk for type 2 diabetes. Cigarette smoking increases the risk for diabetic nephropathy, retinopathy, and neuropathy, probably via its metabolic effects in combination with increased inflammation and endothelial dysfunction. This association seems to be the strongest in type 1 diabetic patients. The increased risk for macro-vascular complications, coronary heart disease, stroke, and peripheral vascular disease, is however most pronounced in patients with type 2 diabetes and based on increased risk of atherosclerosis.

Finally, it should be kept in mind that smoking is often also a marker for many adverse and diabetes risk-influencing factors such as poor lifestyle, adverse social background factors, and eventually a more risk-taking behaviour pattern. Even early life programming could be of importance as it has been documented that smoking mothers tend to have offspring growing up to be smokers themselves and facing increased health risks [22].

Back to Article Outline

5. Effects estimated by smoking cessation programmes 

In a recent systematic review 33 trials were included based on data from hospitalized patients, some of them supposedly with diabetes [23]. Smoking counselling was evaluated that began during hospitalization and included supportive contacts for more than 1 month after discharge. This programme increased smoking cessation rates at 6–12 months (pooled odds ratio [OR], 1.65; 95% confidence interval [CI], 1.44–1.90). No benefit was found for interventions with less post-discharge contact. Counselling was effective when offered to all hospitalized smokers and to the subset admitted for cardiovascular disease. Adding nicotine replacement therapy to counselling produced a trend toward efficacy over counselling alone (OR, 1.47; 95% CI, 0.92–2.35). One study added bupropion hydrochloride to counselling, which had a non-significant result (OR, 1.56; 95% CI, 0.79–3.06). The authors concluded that offering smoking cessation programmes to all hospitalized smokers is effective as long as supportive contacts continue for more than 1 month after discharge. Adding nicotine replacement therapy to counselling may further increase smoking cessation rates and should be offered when clinically indicated, especially to hospitalized smokers with nicotine withdrawal symptoms [23].

Regretfully, there are no published studies on the clinical outcomes in diabetes patients engaged in smoking cessation programmes, and therefore such interventions should be encouraged.

Back to Article Outline

6. Conclusion 

A summary of data indicates that smoking is associated with increased risk of type 2 diabetes, and that this effect could be mediated by insulin resistance, beta-cell dysfunction, and chronic inflammation. The risk of complications in patients with established diabetes is increased, especially for micro- and macro-vascular complications. The RR increase is however not more pronounced in patients with diabetes as compared to non-diabetic controls, even if the absolute risk is higher in diabetes. The conclusion of this is that all smokers should try to stop their hazardous habit, but especially those with a positive family history of diabetes. Programmes for more effective smoking cessation in primary health care are strongly warranted and should be tested within randomised, controlled studies, for example comparing outcomes of individual contacts with that of group-based interventions. There is a strong need to incorporate smoking cessation advice and information in the periodic review of patients with diabetes, both by physicians and diabetes nurses.

Back to Article Outline

Conflicts of interest 

RHF has no conflict of interest. PMN is a regular lecturer for most pharmaceutical companies with anti-diabetes drugs and in addition also for companies with drugs to support smoking cessation.

Back to Article Outline

References 

1. Haire-Joshu D, Glasgow R, Tibbs T American Diabetes Association. Smoking and diabetes. Diab. Care. 2004;27(Suppl. 1):S74–75
   • View In Article
2. Persson L, Hjalmarson A. Smoking cessation in patients with diabetes mellitus: results from a controlled study of an intervention programme in primary healthcare in Sweden. Scand. J. Prim. Health Care. 2006;24:75–80
   • View In Article
   • MEDLINE
   • CrossRef
3. Rimm E, Chan J, Stampfer M, et al. Prospective study of cigarette smoking, alcohol use, and the risk of diabetes in men. Br. Med. J. 1995;310:555–559
   • View In Article
4. Manson J, Ajani U, Liu S, et al. A prospective study of cigarette smoking and the incidence of diabetes mellitus among US male physicians. Am. J. Med. 2000;109:538–542
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (86 KB)
   • CrossRef
5. Willi C, Bodenmann P, Ghali W, et al. Active smoking and the risk of type 2 diabetes: a systematic review and meta-analysis. J.A.M.A. 2007;298:2654–2664
   • View In Article
   • CrossRef
6. Lyssenko V, Jonsson A, Pulizzi N, et al. Clinical risk factors, DNA variants, and the development of type 2 diabetes. N. Engl. J. Med. 2008;359:2220–2232
   • View In Article
   • CrossRef
7. Rafalson LL, Donahue R, Dmochowski J, et al. Cigarette smoking is associated with conversion from normoglycemia to impaired fasting glucose: The Western New York Health Study. Ann. Epidemiol. 2009;(Epub ahead of print April 1)
   • View In Article
8. Nuttens M, Romon M, Ruidavets J, et al. Relationship between smoking and diet: the MONICA-France project. J. Int. Med. 1992;231:349–356
   • View In Article
9. Ford E, DeStefano F. Risk factors for mortality from all causes and from coronary heart disease among persons with diabetes. Findings from the National Health and Nutrition Examination Survey I Epidemiologic Follow-up Study. Am. J. Epidemiol. 1991;133:1220–1230
   • View In Article
   • MEDLINE
10. Al-Delaimy W, Willett W, Manson J, et al. Smoking and mortality among women with type 2 diabetes: The Nurses’ Health Study cohort. Diab. Care. 2001;24:2043–2048
    • View In Article
11. Fuller J, Stevens L, Wang S. Risk factors for cardiovascular mortality and morbidity: the WHO Multinational Study of Vascular Disease in Diabetes. Diabetologia. 2001;44(Suppl. 2):S54–64
    • View In Article
    • CrossRef
12. DeStefano F, Ford E, Newman J, et al. Risk factors for coronary heart disease mortality among persons with diabetes. Ann. Epidemiol. 1993;3:27–34
    • View In Article
    • MEDLINE
    • CrossRef
13. Al-Delaimy W, Manson J, Solomon C, et al. Smoking and risk of coronary heart disease among women with type 2 diabetes mellitus. Arch. Int. Med. 2002;162:273–279
    • View In Article
14. Cederholm J, Eeg-Olofsson K, Eliasson B, et al. on behalf of the Swedish National Diabetes Register. Risk prediction of cardiovascular disease in type 2 diabetes: a risk equation from the Swedish National Diabetes Register (NDR). Diab. Care. 2008;31:2038–2043
    • View In Article
15. Morrish N, Stevens L, Fuller J, et al. Risk factors for macrovascular disease in diabetes mellitus: the London follow-up to the WHO Multinational Study of Vascular Disease in Diabetics. Diabetologia. 1991;34:590–594
    • View In Article
    • MEDLINE
    • CrossRef
16. Turner R, Millns HH, Neil H, et al. Risk factors for coronary artery disease in non-insulin dependent diabetes mellitus: United Kingdom Prospective Diabetes Study (UKPDS: 23). Br. Med. J. 1998;316:823–828
    • View In Article
17. Stevens R, Kothari VV, Adler A, et al. United Kingdom Prospective Diabetes Study (UKPDS) Group The UKPDS risk engine: a model for the risk of coronary heart disease in Type II diabetes (UKPDS 56). Clin. Sci. 2001;101:671–679
    • View In Article
18. Kothari V, Stevens R, Adler A, et al. UKPDS 60 risk of stroke in type 2 diabetes estimated by the UK Prospective Diabetes Study risk engine. Stroke. 2002;33:1776–1781
    • View In Article
    • CrossRef
19. Nilsson P, Cederholm J, Eeg-Olofsson K, et al. Smoking as an independent risk factor for myocardial infarction or stroke in type 2 diabetes: a report from the Swedish National Diabetes Register (NDR). for the Swedish National Diabetes Register (NDR) Eur. J. Cardiovasc. Prev. Rehab. 2009;16:506–512
    • View In Article
20. Eliasson BB. Cigarette smoking and diabetes. Prog. Cardiovasc. Dis. 2003;45:405–413
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (71 KB)
    • CrossRef
21. Nilsson P, Lundgren H, Söderström M, et al. Effects of smoking cessation on insulin and cardiovascular risk factors—a controlled study of 4 months duration. J. Int. Med. 1996;240:189–194
    • View In Article
22. Nilsson P, Hofvendahl S, Hofvendahl E, et al. Smoking in pregnancy in relation to gender differences in offspring adult mortality risk. Scand. J. Public Health. 2006;34:660–664
    • View In Article
    • MEDLINE
    • CrossRef
23. Rigotti NA, Munafo MR, Stead LF. Smoking cessation interventions for hospitalized smokers: a systematic review. Arch. Int. Med. 2008;168:1950–1960
    • View In Article