Systemic Effects of Smoking^* FREE TO VIEW

Tobacco smoking is one of the most potent and prevalent addictive habits, influencing behavior of human beings for > 4 centuries. Smoking is now increasing rapidly throughout the developing world and is one of the biggest threats to current and future world health.¹ Furthermore, while the prevalence of tobacco use has declined among men in some high-income countries, it is still increasing among young people and women.² Cigarette smoking is the most common type of tobacco use. In average, to date 47.5% of men and 10.3% of women are current smokers. Tobacco continues to be the second major cause of death in the world. By 2030, if current trends continue, smoking will kill > 9 million people annually.³

Tobacco smoking affects multiple organ systems resulting in numerous so-called tobacco-related diseases. The well-known health risks in tobacco smoking pertain to diseases of the respiratory tract such as COPD and cancer, particularly lung cancer and cancers of the larynx and tongue.⁴⁵

While the adverse effects of cigarette smoke on lung health are well established, it is becoming more evident that smoke has an important extrapulmonary toxicity. Injury in the lung, primary target of inhaled smoke, can be explained by the direct chemical exposure to cigarette smoke, but effects causing chronic diseases in other organ systems are likely to be the result of indirect consequences of the exposure. However, despite the overwhelming amount of studies demonstrating the relationship of smoking with numerous widespread “systemic” diseases such as atherosclerosis and COPD, the precise mechanisms how smoke potentiates its systemic effects need to be clarified. In a article by van der Vaart et al,⁶ the local and systemic effects of acute smoke exposure on oxidative stress and inflammatory mediators were reviewed. In the present article, we aim to review systemic effects of long-term smoking exposure in humans. In particular, traditional markers of generalized response to smoking such as systemic oxidative stress and systemic inflammation are reviewed. In addition, in view of findings suggesting the relationship between some of these markers and atherosclerosis, aspects of hemostatic and coagulation systems are discussed in relation to tobacco smoking.

Cigarette smoke contains approximately 10¹⁷ oxidant molecules per puff.⁷ This oxidative stress can be registered in several different ways, either by direct measurements of the oxidative burden (reactive oxygen species [ROS] production by peripheral blood cells) or by the effects of oxidative stress on target molecules (lipid peroxidation products and oxidized proteins), or as the responses to the oxidative stress (antioxidant capacity of plasma)⁸ [Table 1 ].

Only a few studies⁹¹⁰¹¹ have used ROS production by blood cells extracted from the circulation of smokers. Perhaps more important than the presence of oxidative stress are the effects of this oxidative stress on a variety of vital target molecules. Numerous markers for oxidative damage have been proposed, including oxidation and nitration of proteins. For example, nitration of tyrosine residues of proteins leads to the production of 3-nitrotyrosine, which may be considered as a marker of nitric oxide (NO)-dependent oxidative damage. Indeed, NO-mediated and peroxynitrite-mediated formation of 3-nitrotyrosine is elevated in plasma and platelets of chronic smokers.¹²¹³ Furthermore, Pignatelli and coworkers¹⁴ reported significantly higher levels of nitrated and oxidized fibrinogen, transferrin, plasminogen, and ceruloplasmin in smokers than in nonsmokers.

Free radicals from cigarette smoke also cause peroxidation of the polyunsaturated fatty acids of cell membranes that amplify oxidative stress during smoking. The F₂-isoprostanes, prostaglandin-like compounds, are products of free radical-catalyzed lipid peroxidation of arachidonic acid. Several studies¹⁵¹⁶¹⁷ have reported an increased level of isoprostane 8-iso-prostaglandin F2 (PGF2)α formation in smokers. It has been found that urinary 8-epi-PGF2α excretion was significantly increased in long-term current and former smokers compared with the age- and sex-matched nonsmoking control subjects.¹⁷ In addition, a dose-response relationship was observed between the number of cigarettes smoked and both urinary cotinine and urinary 8-epi-PGF2α.¹⁶ The biological role of isoprostanes is not clear yet. It has been shown that F₂ isoprostane levels are significantly increased in atherosclerotic plaques compared with normal vascular tissue, suggesting that these compounds may play a role in the pathogenesis of the disease.¹⁸ The observation of elevated levels urinary 8-iso-PGF2α in patients with coronary heart disease strengthens this hypothesis.¹⁹

Increased levels of malondialdehyde, which are degradation product of lipid peroxides, have been found associated with current smoking status in population-based studies.²⁰²¹ Similarly, higher levels of thiobarbituric acid reactive substances (TBARS) have been found in smokers compared to nonsmokers.²² New evidence for the association of systemic oxidative stress with pulmonary function comes from population-based study²³ conducted in the New York State (n = 2,346). Ochs-Balcom and coworkers²³ showed an inverse association of TBARS with percentage of predicted FEV₁ and percentage of predicted FVC in men but not in women, suggesting gender differences in the relation of oxidative stress to pulmonary function.

Exposure to oxidant chemicals in smoke is associated with depletion of endogenous levels of antioxidants in the systemic compartment. Numerous studies have reported that smoking results in low antioxidant concentrations in plasma. The total plasma Trolox-equivalent antioxidant capacity (TEAC) was significantly lower in smokers than in nonsmokers.¹²²⁴ However, no relationship was found between spirometric end points (FEV₁ or FEV₁/FVC) and plasma levels of TEAC in healthy smokers.²⁴ The third National Health and Nutrition Examination Survey (NHANES) and other studies¹⁷²⁵ reported that smokers have significantly lower serum levels of vitamin C, α-carotene, β-carotene, β-cryptoxanthin, melatonin, α-tocopherol, and lutein/zeaxanthin. However, diet could also influence levels of antioxidants, and independent effects of smoking have only been shown on plasma levels of vitamin C and β-carotene.²⁵²⁶²⁷ In addition, an inverse relationship between cigarette consumption and plasma levels of vitamin C and β-carotene corrected for habitual dietary intake has been found.²⁷²⁸ Such decreases in plasma antioxidants can disturb the normal oxidative-antioxidative balance in smokers. Remarkably, numerous studies²⁹³⁰ have shown that antioxidant supplements provide at best only a limited protection.

Glutathione (GSH) is a major antioxidant used to eliminate peroxides to nontoxic hydroxyl fatty acids and/or water and to maintain vitamins C and E in their reduced and functional forms. Cigarette smoke contains ROS that oxidize GSH to disulfide form (oxidized glutathione), resulting in decreased plasma GSH levels.³¹ Similar processes are responsible for an even more extensive oxidation of the cysteine (Cys)/oxidized cysteine (CySS) redox couple and reduced Cys levels showing that smoking has additional effects on sulfur amino acid metabolism. Taking into account that cysteine is the critical molecule for normal GSH synthesis, this observation suggests that evaluation of the Cys/CySS redox couple may be a new sensitive marker of oxidative stress in smokers. In summary, the oxidative burden in the systemic compartment of smokers is mainly characterized by elevated levels of peroxides (isoprostanes and TBARS) and decreased levels of traditional plasma antioxidants (vitamins A and C), whereas GSH-related antioxidants are affected to a lesser extent.

Activation and release of inflammatory cells into the circulation, and an increase in circulating inflammatory mediators such as acute phase proteins and proinflammatory cytokines, characterize the systemic inflammation.

The systemic inflammatory response is characterized by the stimulation of the hematopoietic system, specifically the bone marrow resulting in the release of leukocytes and platelets into the circulation. Numerous studies³²³³³⁴ have shown that long-term cigarette smoking increases total WBC counts, mainly due to an increase in polymorphonuclear neutrophil (PMN) counts in the circulation of smokers. A large population-based study³⁵ of 6,902 men and 8,405 women performed in Great Britain revealed that current smoking had a stronger effect on mean total WBC than cumulative exposure as measured by pack-years. However, other authors³⁶ reported a dose-response relationship with pack-years smoked and WBC. This inflammatory response in smokers is characterized not only by an increase in the number of circulating cells but also by phenotypic changes. Indeed, neutrophilia related to chronic smoking was associated with an increase in numbers of circulating band cells, a hallmark of early bone marrow release of PMNs, and an increase in L-selectin expression, a cell adhesion molecule, constitutively highly expressed on maturating PMNs.³² L-selectin could initiate the adherence of PMNs to endothelium and has been shown to be important for the recruitment of PMNs to inflamed tissue.³⁷ In addition, PMNs from smokers have higher levels of myeloperoxidase, an enzyme produced at the early stages of PMN proliferation.³²³⁸ Overall, these findings suggest that smoking causes bone marrow stimulation and the release of younger cells from bone marrow. van Eeden and colleagues³⁹ speculated that circulating cytokines like interleukin (IL)-1β and IL-6 may be responsible for bone marrow stimulation induced by lung inflammation. Indeed, the same authors³⁹ have shown that IL-6 cytokine also potently stimulates the bone marrow to release leukocytes and platelets.

Reports⁴⁰ have underlined the role of T-lymphocytes as a potentially important factor in the systemic inflammatory process associated with smoking-induced diseases like COPD. Some studies⁴¹⁴²⁴³ have reported increased total numbers of circulating T-lymphocytes in humans exposed to cigarette smoke. Studies investigating the influence of smoking on different lymphocyte subsets have produced conflicting data. In heavy smokers, a decrease in CD4+ cells (helper T-cells) and increase in CD8+ cells (suppressor T-cells) with subsequent decrease in CD4+/CD8+ ratio have been reported.⁴¹ In contrast, Tollerud and colleagues⁴⁴ found that cigarette smoke was associated with an increase in leukocyte count with a selective increase in CD4+ cells, resulting in significant increase in the CD4+/CD8+ ratio in healthy white subjects. Other studies⁴⁵⁴⁶ support these findings. Further analysis has shown that smokers have higher absolute numbers of peripheral blood memory T-cells and naive T-cells as compared with nonsmokers.⁴³⁴⁶ Analysis of T-cell subpopulations in heavy and light-to-moderate smokers revealed that numbers of memory T-cells were significantly correlated with daily cigarette consumption.⁴² Overall, the results of these studies suggest that smoking may exert a selective influence on subsets of T-cells. Therefore, taking into account the role of lymphocytes in a number of inflammatory conditions associated with smoking, additional studies would be relevant to better understand the response of circulating lymphocytes to cigarette smoke.

Activated inflammatory cells produce a great variety of inflammatory mediators in response to cigarette smoke, first of all, acute-phase proteins (APPs) and cytokines. Conditions that commonly lead to substantial changes in the plasma concentrations of APPs and cytokines include infection, trauma, surgery, burns, tissue infarction, various immunologically mediated inflammatory conditions, and cancer. In recent years, these inflammatory mediators have been studied as potential markers of subtle and persistent systemic alterations. Many studies have reported changes in levels of inflammatory mediators not only in the lungs but also in the circulation of healthy smokers. Several studies⁴⁷⁴⁸⁴⁹ have reported strong associations between cigarette smoking and different APPs such as C-reactive protein (CRP) and fibrinogen (Table 2 ).

For example, the large-scale, population-based NHANES III study revealed a strong independent dose-response relationship between cigarette smoking and elevated levels of CRP and fibrinogen.⁵⁰ This analysis was based on > 4,187 current smokers, 4,791 former smokers, and 8,375 never-smokers with smoking status based on cotinine levels and was adjusted for different confounders. These findings are in accordance with results of two other studies⁴⁷⁵¹ of APPs in smokers. Data from several prospective studies support the hypothesis that CRP, and fibrinogen levels in particular, are primarily related to lifetime exposure of smoking (pack-years) and not to years since quitting smoking. In the MONICA Augsberg Study⁴⁹ and the Speedwell Study,⁵² CRP was still significantly raised 10 years after smoking cessation. Data from the British Regional Heart Study³³ indicate that smoking cessation results in a rapid reduction in hemostatic and inflammatory markers, but CRP levels remained significantly raised after 10 to 19 years and did not revert to that of never-smokers until after 20 years; for CRP, the reduction was dependent on the number of cigarettes smoked. A dose-effect relationship between the number of cigarette smoked per day and plasma fibrinogen concentrations have also been reported.⁵³⁵⁴ Interestingly, the NHANES III study (of 7,685 participants) confirmed that cigarette smoking contributes significantly to low-grade systemic inflammation and, furthermore, shows that reduced lung function by itself is also associated with increased odds of elevated CRP, fibrinogen, and blood leukocytes; but having both risk factors suggests an additive effect contributing to higher levels of systemic inflammation in susceptible individuals.⁵⁵ Furthermore, a large-scale population-based study⁵⁶ confirmed that low FVC was associated with higher plasma levels of fibrinogen, α₁-antitrypsin, haptoglobin, ceruloplasmin, and α₁-acid glycoprotein, and with increased incidences of myocardial infarction and cardiovascular death. However, the nature of the observed associations between decreased lung function and APPs induction is still unclear.

The role of other APPs is less extensively investigated. It has been reported that concentrations of α₁-acid glycoprotein, ceruloplasmin, and α₂-macroglobulin are increased in the plasma of smokers as compared to nonsmokers by 39%, 28%, and 12%, respectively.⁵⁷ Furthermore, the large prospective Malmö Preventive Project study⁵⁸ with 18-year follow-up revealed that all measured APP levels (α₁-acid glycoprotein, α₁-antitrypsin, haptoglobin, fibrinogen, and ceruloplasmin) increased significantly with increasing cigarette consumption in healthy adult men, independent of other known cardiovascular risk factors.

While these blood changes in smokers may simply be markers of smoking-induced tissue damage, it is also possible that high APP levels may have a direct effect on the promotion of cardiovascular diseases. Increased levels of CRP and fibrinogen have been associated with risk for subsequent cardiovascular events in several large prospective studies.⁵⁹⁶⁰ Another study⁶¹ showed that CRP might be not only a biomarker of different cardiovascular diseases but may have direct effects on the pathogenesis of atherosclerosis and endothelial dysfunction. For example, CRP stimulates IL-6 and endothelin-1 production and upregulates adhesion molecules, promoting a cascade of events that can lead to clot formation and even promotes atherosclerosis in apolipoprotein E-deficient mice.⁶² However, the exact role of CRP in development of cardiovascular diseases is still under discussion.⁶³ Some causative speculations are also discussed concerning the role of fibrinogen in atherogenesis. Probably, fibrinogen may promote cardiovascular diseases through effects on blood viscosity, platelet aggregation, and fibrin formation.⁶⁴ In conclusion, these data clearly indicate that CRP and fibrinogen levels are markedly increased in smokers, possibly contributing to the proinflammatory and proatherogenic effects of chronic smoking exposure.

Raised levels of plasma APPs may partly reflect elevations of inflammatory cytokines such as IL-6 and tumor necrosis factor (TNF)-α, which are major inductors of APP and, therefore, regulators of systemic inflammatory response. Similarly to APPs, increased levels of proinflammatory cytokines like TNF-α and IL-6 have been shown to be a risk factor and predictor for myocardial infarction, coronary heart disease, and stroke.⁶⁵⁶⁶

Several studies¹⁷⁵¹ have shown increased levels of TNF-α and IL-6 in smokers. The population-based MONICA III North Glasgow study⁶⁷ revealed that mean IL-6 levels were substantially raised in current smokers (by approximately 46% compared to never-smokers), while ex-smokers had similar levels of IL-6 to never-smokers. Furthermore, a significant correlation was found between IL-6 and fibrinogen, and IL-6 and WBC counts, reflecting the major role of IL-6 as inducer of fibrinogen. The Women’s Health Study⁵¹ from the United States showed a trend toward increasing IL-6 levels across never, former, and current smoking women. Further, Wirtz and colleagues⁶⁸ reported a trend for higher baseline TNF-α levels among healthy smokers. However, a study conducted by Gander and coworkers⁶⁹ failed to show significant effects of smoking on TNF-α plasma levels.

Taken as a whole, these data suggest that there are limited data yet on circulating concentrations of IL-6 and TNF-α in healthy smokers. Taking into account the possible predictive effects of major cytokines for cardiovascular diseases, it would be worth to address large-scale population-based studies to investigate potential relationship of cytokine plasma levels with traditional cardiovascular risk factors including smoking.

The biological mechanism linking smoking and atherogenesis, the process leading to cardiovascular diseases, is complex and not fully understood. Besides inflammation, proposed potential mechanisms by which smoking increases the risk of cardiovascular pathology include several other pathways: vascular endothelial dysfunction, systemic hemostatic and coagulation disturbances, and lipid abnormalities. Many of these indexes including fibrinogen (marker of coagulation), fibrin d-dimer (a marker of cross-linked fibrin turnover), and tissue plasminogen activator antigen (t-PA, marker of endothelial dysfunction) have been identified as independent predictors of subsequent cardiovascular events in prospective studies⁷⁰⁷¹ conducted in healthy subjects. In addition, platelet hyperaggregation and activation, plasma viscosity, and plasminogen activator inhibitor (PAI) type I (marker of impaired fibrinolysis) levels have also been associated with cardiovascular morbidity and mortality in prospective studies.⁷²⁷³ The effect of smoking on these variables has also been investigated in several cross-sectional studies, as will be discussed below (Table 2).

Endothelial dysfunction is mainly caused by diminished production or availability of NO.⁶¹ It has been demonstrated that the serum concentration of nitrate and nitrite, metabolic end-products of NO, is significantly decreased in smokers relative to that in nonsmokers.⁷⁴ In cigarette smokers, low-density lipoprotein (LDL) is more prone to oxidation due to higher level of ROS and reactive nitrogen species.⁷⁵ Oxidatively modified LDL limits the bioactivity of endothelium-derived NO; and, in turn, the loss of NO bioactivity is associated with increased inflammatory cell entry into the arterial wall.⁷⁶ Oxidatively modified LDL is taken up by macrophage scavenger receptors, promoting cholesterol ester accumulation and foam cell formation.

Most recently, upregulation of the CD40/CD40L dyad and increased platelet/monocyte aggregation have been proposed as potential contributors to the atherothrombotic consequences of smoking.⁷⁷ CD40-CD40 ligand couples, members of TNF family, are coexpressed by all of the major cellular players in atherosclerosis. In particular, smokers appeared to have elevated plasma levels of soluble CD40 and increased surface expression of CD40 on monocytes together with increased CD40 ligand on platelets. Furthermore, plasma cotinine concentrations correlated with CD40 and CD40 ligand expression, and with rate of platelet-monocyte aggregations. A recent study suggests that oxidatively modified LDL may play the role of initial trigger for CD40/CD40L expression in human endothelial and smooth-muscle cells.⁷⁸

Dysfunctional endothelial cells lose their critical physiologic property of nonadherence to circulating immune effector cells (monocytes, macrophages, T-lymphocytes, platelets). Some adhesion molecules are known to be elevated in plasma of smokers. Several groups reported significantly higher levels of soluble intracellular adhesion molecule (ICAM)-1 and P-selectin and E-selectin in current smokers than in nonsmokers among healthy women.⁵¹⁷⁹ A dose-dependent relationship was observed between plasma ICAM-1 concentration and daily cigarette consumption, plasma cotinine level, and exhaled carbon monoxide level.⁸⁰ Generally, impaired endothelial function caused by cigarette smoking may lead to increased susceptibility of vasculature to atheroma formation and can be considered as an early feature of atherogenesis in humans.⁸¹

There is increasing evidence that blood levels of rheologic variables are associated with subsequent cardiovascular events.⁷⁰ These indexes include whole-blood viscosity and its main determinants: hematocrit and plasma viscosity, principally composed by plasma fibrinogen and lipoproteins.⁶⁷ Several studies³³⁶⁷ revealed that current smokers have increased blood viscosity, associated with increased hematocrit or/and plasma viscosity resulting in a procoagulant condition. Increased plasma viscosity may be caused by higher levels of fibrinogen reported in plasma of smokers, as have been discussed before in this review.

Furthermore, elevated levels of markers of fibrinolysis have been reported in healthy smokers. t-PA, the main fibrinolytic activator, converting plasminogen to plasmin is synthesized by endothelial cells. In vivo studies⁸² have demonstrated major impairment of t-PA release from the vascular endothelium of smokers. The primary inhibitor of fibrinolysis is PAI-I, which inhibits plasminogen activation by binding with t-PA to form the PAI/t-PA complexes. Current smoking is associated with a significant increase in t-PA antigen, which represents mainly the circulating PAI/t-PA complexes and indicates impaired fibrinolytic activity in smokers.³³³⁴ Supporting previous findings,³⁴⁸³ plasma PAI-1 antigen and/or activity is significantly higher in smokers than in nonsmokers and is correlated with pack-years smoked.

Plasmin promotes the degradation of fibrin within the thrombus, disintegrating clots, and hence maintains vascular patency. Fibrin d-dimer is a degradation product of cross-linked fibrin that is related to cardiovascular diseases risk.⁵² As been reported, smoking is positively associated with fibrin d-dimer.³³⁸⁴ The increased d-dimer in smokers probably reflects increased coagulation activation because this antigen is present in several degradation products from the cleavage of cross-linked fibrin by plasmin.⁸⁵ Taking into account possible adverse effects of abnormal fibrinolysis and excess coagulation on vascular health, further studies are essential to evaluate the impairment of the fibrinolytic system in smokers.

Overall, data presented in this review suggest that smoking is one of the major lifestyle factors influencing levels of a number of novel inflammatory, coagulation, and hemostatic markers associated with common widespread diseases in population-based prospective studies. Systemic oxidative stress followed by low-grade inflammation and endothelial dysfunction caused by chronic smoking exposure could be one of “real-working” mechanisms that explain increased prevalence of common diseases of modern civilization like coronary heart disease, peripheral vascular disease, and COPD in smokers. However, this hypothesis has to be taken cautiously because not all smokers acquire one or more of these diseases. It clearly suggests the existence of other mechanisms influencing common disease development, and, beyond doubt, genetic susceptibility is one of them.⁸⁶ Different approaches like case-control and whole-genome association studies, linkage analysis of extended pedigrees, and affected sibling pairs are used to dissect genetic component of complex traits. For genetically complex disease like cardiovascular diseases, common disease-common variant hypothesis has been put forward, which assumes that common disease susceptibility or resistance variants are expected to be few at each genetic locus, relatively common in the human population and enriched in the coding and regulatory sequence of genes.⁸⁷ Despite the small effects of such genes individually, the magnitude of their attributable risk may be large because they are quite frequent in the population, making them of public health significance. However, there is a possibility that many rare variants, each with small contribution, underlie the susceptibility to common human disease.⁸⁸ Nevertheless, it seems that complete model of complex disease initiation and progression is based on cooperation where these multiple genes are likely to operate through interactions with many environmental factors where smoking is only one but a very important variable.

The possible biological mechanisms responsible for the observed association of smoking with various diseases and global mortality are numerous and, in spite of a many attempts to find causative relationships, are still unclear. It is a great scientific task to unravel exact pathways through which smoking affects human health. Although the effects of smoking on inflammatory markers may persist for many years, a majority of the adverse health effects of smoking are reversible. Therefore, quitting smoking avoids much of the excess health-care risk associated with smoking and allows increasing life expectancy.