Tobacco Documents Online

LQ'S ~r r a h1 .~}4,1 G,,/ 1Z2 y~:.c 16'' ~ Cl . ~- . ( . ~, 06t.r, 1~`7 Z- /, hr 4Z y- ¢4-3. PASSSIVE SMOKING AND CORONARY ARTERY DISEASE. BIOLOGICAL PLAUSIBILI'TY AND SEVERITY OF EFFEGT G. CR$PAT Institut Universitaire de Technologie, University of Dijon„ BP 510, 21014 Dijon C6dez, France AffiIRACr A number of small! pieces of incriminating evidence, apparently painting in the same direction, do not neeessarily prove a crime. Therefore, before reviewing the various mechanisms suggested to incriminate ETS in CHD' incidence, a brief reminder oE' what is actually well known about the atherosclerosis process is necessary. We shall stress what progress has, been made over the last decade and what risk factors are now being considered. The role of active smoking, which has long been set in evidence by epidemiologie studies, is now fairly well' understood scientifically_ The mechanisms of cardio-vascular attack are triggered by two primary stimuli: nicotine and earbon monoxide. Induced effects on eathecholamines, platelets, earbozyhemoglobin; frbrinogen, lipoprotein metabolism, etc... help understand their incidence on CHD! Everything however, is not: altogether clear nor consistent. Concerning ETS, the ten epid'emiologic studies investigating its associatiom with heart disease mortality, produce mean RR values ranging from 1.2 to 1.4 i,n both sexes (Glhntz). This fact and corresponding criticisms will' not be dealt with here. We shall however concentrate on a detailed study of the mechanisms suggested to explain the effects of passive smoking and compare them with those of'~ active smoking. If ETS brings into an exposed non-smoker's blood sueh primary stimuli as nicotine, CO, benzene, PAH allow the biological plausibility of CHD attack, but careful consideration of their mode of action and magnitude cannot, as far as is known„incriminate ETS'as the third cause of mortality (Wells). INTRODUCTION There are now 10 epidemiological studies [1-10) on the relationship between exposure to environmental tobacco smoke (ETS) in the home and the risk of coronary heart disease (CHD) (table 1). Most of these studies have reported relative risks greater than 11.0. 429 ,,.k . -... ~,.r. t Ail .; .-, ~

Table 1. Rel'ative risks for heart disease death from passive smoking epidemiologic studies. Author Type Total cases Relative 95 96 risk confidence interval Country Fl.rla1CS Hinyama (1984)' P 494 1.2 0.9-1.4 Japan Gillis d a!. (1984) P 21 3.6 0.9-13.8 Scotland Garland et aG (1985) P 19 2.7 0.9-13.6 California Leed aL (1986) C 77' 0.9 0.5-1.6 United Kingdom Helsing et aL (1988) P 988' 1.2 1.1-1.4 Maryland He (1989) C 34' 1.5 1.3-1.8 China Humble d al. (1990) P 76 1.6 1.0-2.6 Georgia Butler (1990) P 64' 1.4 0.5-3.8 California 1f7Slts Gillis et al. (1984) P 32 1.3 0.7-2.6 Scotland Leed'a!. (1986) ' C 41 1.2 0.5-2.6 United Kingdom Svendsen et ol: (1987) P 13 2.1 0.7-6.5 United States Helsing d al. (1988) P 370 1.3 1.1-1.6 Maryland Hok ct aL (1989) P 84 2.0 1.2-3.4 Scotland CHD : Coronary Heart Disease P: Prospective cohort C: Case control studies In spite of large differences in study design and type of heart disease considered (ischaemic heart disease, death of any origin, myocardial „infarction death, non fatal coronary symptoms including angina); all results have been pooled giving an overall relative risk estimate of 1.23 (limits 1.1-1.4) for 6 studies in women an& 1.31 (limits 1.1-1.6)~ for 4 studies in men (Wells, [11']). When these values are used to calculate CHD mortality rate in the US:A, they give the very high figure of 32,00U deaths/year [11i: Glantz [12j takes up this figure an& collects evidence from a number of more or less relevant studies all tending to prove that this finding is plausible. We shall first consider the formation process of atherosclerosis. It is a complex, multifactorial slowly developing phenomenon. It is therefore biologically plausible that ETS exposure may cause some blood factors to vary and accelerate the atherosclerosis process. It remains to be proven, however, that the amplitude of such variations occurring during actual ETS ezposure produces an effect leading to a 30'b risk increase [12]. That is why, after a brief reminder of what is known to date about the mechanism of atherosclerosis, we shall examine what are the aggravating factors due to active tobacco smoking: Then, we shall attempt to evaluate the impact of ETS induced stimuli compared to those affecting an~ active smoker who is also passively exposed to his own smoke and that of other smokers.

O1!ERVIM ON TSE' MECHANI5M Epidemiologic studies from a variety of countries throughout the world have clearly established a relationship between the development and/or the progression of atherostatic process and many lifestyle factors: sedentarity, obesity, dietary intake (excess calories, saturated fat, salt...), cholesterol, stress, cigarette smoking... etc. Risk factors Since Framingham and his team published their findings 15 years ago [13], it has been known that the three major CHD risk factors are: hypercholesterolemia, high blood pressure and cigarette smoking (Fig. 1) j14]. Combination of two or three of such risk factors has been found' to increase CHD incidence nine times for two and 16 times for three factors [14]. Figure 1. Increase in risk of coronary heart disease as a result of smoking, hypertension and hypercholorestemia, relative to a 45- year-old non-smoking man with a systolic blood pressure (SBP) of 110 mm Hg and total cholesterol of 185 mg dl-1. Drawing made on the basis of data from the Framingham study (from Kannel W.B. [14]).. 431

It is now known that some of these factors should be better defined. With cholesterol, for instance, it is the byperlipoproteinemic anomaly which must be considered (table 2). Cholesterol carriers such as LDL (Low Density Lipoprotein) represent a risk and VLDL (Very Low Density Lipoprotein) an additional risk whereas HDL (High Density Lipoprotein) counters the risk. Apolipoproteins are better correlated with CHD risk: Apo B100 (contained in LDL - VLDL) represent the risk, whereas Apo Al (contained in HDL) counters the risk [15]. Bloodstream lipoproteinic particles (16) now appear to be even more related to CHD risk. LpB or Lp(a+) [I17]I represent the risk whereas Lp Al are the protecting agents [16]. These factors are generally of the genetic risk type, whereas LDL modified (18) by oxidation, acetylation, glycosylation and MDA LDL conjugation with MDA (Malondialdehyde) are due to metabolic and environmental chemicali alterations [19,20). Destruction of modified LDL by the "scavenger" pathway contributes through different mechanisms to atherosclerosis development [21]. This helps understand the aggravating effect of such oxidising substances in blood as free radicals or the protecting role of direct' or indirect antioxydants such as vit. C, vit. E, Se.. [22). Table 2. Atherosclerosis and risk factors due to lipoproteinemic anomaly: Risk increasing faetors Risk decreasing factors Total Cholesterol *'4 Total Cholbsterol L Cholesterol LDL Cholesterol i 1 LDLC %m VLDL Cholesterol 7I VLDLC 1 LDL jM LDL 1 Lipoproteins VLDL *A VLDL 1 HDL 1 IiDL l• Apoproteins Apo Bi00 M Apo B100: y Apo A1 1 Apo A1 1 4B r LpA1 J~ Lp (a+) 1~ Lipid particles o: LDL 7~ Glyc LDL q MDA LDL 1' other factors free radicals, etc.. Vit C, Vit E, Selenium, etc:.., Patbogenesis of atherosclerosis Taking into account the different theories and hypotheses on lipid infiltration, endothelial injury, and platelet role, we can state that: atherosclerosis is characterized by increased endothelial permeability, monocyte infiltration, internal smooth muscle cell (SMC) proliferations platelet aggregation and accumulation of lipids, Ca**' and extra cellular

matrix components such: as collagen, elastin an& proteoglycans in vessel wall. EndotJulial cell Wury (Fig. 2)' Factors and forces promoting such damage are quite undefined and may be of physical, metabolic, hormonal, cellular, molecular or genetic nature. Among identifie& factors we can list: hi& blood pressure, anoxia, immune activation, turbulent blood factors, increases in oxidized LDL, Lp(a) or free radicals... etc. Response from irritated' endotheliaL cells induces an increase in permeability for plasma compounds into the subendothelial space. HypecaDokrsiokmi.e - oxLDL IRRCCA'IWG S'ISP7ULT Hypertemion-Ywtukntblood flov Anozti - CO - Piee-Rad"b, et... . I2TiI2+1A O PLATELETS AGGREGA7E ON ENDOZfffiZIUM SURFACE AND SECRETE : 12 HETE,IXA2, PGDF(AH)~ MEDIA SMC: 'ContrecnBe phenotype. sc.r (ru.) ~ MONOCYTES infiha9on MACROPHAGES ACCUMULAT7ON FOAM CELLS AND EX7RUDED LIPIDS : 'FATTY STRP.AICS' ~ LESION GROWS : SMC PROLIFERA77ON COLLAGEN SECRETTON ELAS'PiN',PROTEOLYCAN, GLYCOAMINOGLYCAN, CHOLES'lEROTL azd C... Qeposleon Figure 2. Pathogenesis of atherosclerosis. 433 LDL OXLDL Lp(a) ,cn: (vis) u SMC PROLIFERA'IiON 'Synmetic pAenatqpe•

lnfUtrotion of plaima components (PiQ. 2) Endothelial injury enhances infiltration of monocytes which differentiate into macrophages, LDL and possibly Lp(a) which are oxidized by damaged endothelial cells [27J1 Macrophages and oxidized LDL are chemostatic for monoeytes which further penetrate into subendothelial space [22]. :'1'3tEE`-~AIlTCkLiz?`:: j O= 2RA1tSPOrrr t ooM PC2t - TXAz 8 AL"CE I PLA7ZLFf AOaRLOA7iON CB2C[TLAnNa t CATECHOiAMINES t!!A tLDL 1FDL t MONOClZB WFII.IRAI7ON ~ ,t MAi3ROPHAdEB i =MC,DDORAY7ON PROL!l132A7ION WPLUX LDL EPlLUDC HDL EALANCE t OX LDL T A'IFIFROSCLPROSiS Figure 3. Smoking, CO, Nicotine, free radicals, PARs and atherosclerosis. Foam cell'accamulation and platelet aggregation (Fig: 2) The modified LDLs are taken up via the scavenger receptor pathway by macrophages which turn into foam cells [21,231. Continued accumulation of macrophages in the presence of high cholesterol leads to extrusion of lipids into the arterial wall interstitial space and formation of the "fatty streak" which has become one of the pathological hallmarks of the atherosclerotic process. Several more or less known factors cause platelets to adhere to the injured endothelium. Macrophages and platelets release growth factors 434

PGDF(AB) from platelets and PGDF(BB) from macrophages (Heldin) [25] which are chemostatic and mitogenic for vascular smooth muscle cells. Vascular SMC are assumed'to migrate from the media to the intima where they convert from a contractile into a synthetic phenotype, proliferate and secrete growth factors (PGDFAA) [26]. At this stage, we must stress the importance of the balance between prostacycline (PG12) secreted by injured endothelial cells which inhibits platelet aggregation and thromboxane (TXA2) secreted by the platelets which stimulates aggregation [27]. Progression of atherosclerosis (Fig. 2) The hallmark in the progression of atherosclerosis is the proliferation of SMC and accumulation of extra cellular matrix in the intimal layer: collagen, elastin, proteoglpcans (PGs), glycoaminoglyeans (GAGs). These molecules combine with LDL, modified LDL, Lp(a); celli debris, Ca++ deposits to form, an atheroma gradually weakening and narrowing the artery, encouraging the formation of a thrombus which may develop into myocardiali infarction. Finally, there is normally a balance between cholesterol influx (LDL, Lp(a) into the membrane and cholesterol efflux (HDL2) out of the plasma membrane. When influx of cholesterol! exceeds efflux, cholesterylesters are stored by the cells. On the ot'her hand, the lesion progression is also dependent on SMC proliferation and consequently on the balance between growth promoters (PGDF; TXA2, 12 HETE) and growth inhibitors (PGI2). This is just an overview of atherosclerosis pathogenesis which does not cover the abundant literature documenting such factors as : Lipoperoxydation of polyunsaturated fatty acids leading to MDA-LDL [181; Free radicals and antioxidising role of plasma [28] selenium, The role of Ca++ as a second messenger involved in regulating processes in the vessel wall [27] promoting LDL receptor binding, inducing monocyte and SMC chemotaxis and stimulating secretion of collagen and other components. A number of epidemiologic studies have brought evidence of association between active smoking and atherosclerosis development but the physio- biochemical mechanisms suggested are not yet definite as many smoke constituents are likely to be involved. Mainstream smoke chemical compwaition About 4,000 components have been identified in mainstream smoke. In the gas phase the major constituents are: carbon dioxide, carbon monoxide, 435

nitrogen oxides, nitrosocompounds, hydrogen cyanide, formaldehyde, PAH and free radicals have also been investigated. atherosclerosis development are nicotine„ carbon monoxide (CO) [54], while specific chemical substance. The primary stimuli involved in using all mainstream smoke or tars, which cannot lead to incriminate any A number of investigations of smoke toxicity have been carried out in found in the particulate phase: nicotine, phenol, benzo(a)pyrene, nitrosamines, pyrene, naphtalene, etc... acrolein and benzene. The biologically active compounds, however, are prostanoid synthesis PG12-TXA2 by decreasing PGI2 synthesis by vascular endothelial cells [38], increasing release of TXA2 platelets [40] an& incidentally platelet' aggregation and SMC constriction. In fact, the direct role of nicotine and its action as a function of dose and in presence of CO iss still controversial [41). 6. Carbon monoxide: Carbon monoxide forms carboxyhaemoglobin; reduces blood oxygen-carrying capacity and causes hypoxemia. Mean COHb Lipid(s)) which cause LDL oxidation and then pathogenesis of smoking- induced atherosclerosis [37].. 5. PGI2-TXA2 Balance: Cigarette-smoking affects the balance of vascular smoking induces increased PAF-LL (PlateleU-Activating-Factor-Like- which can produce the same effect in vitro. On the other hand, cigarette platelet aggregabili'ty [35, 36]• The responsible agent is likely to be nicotine 4. Platelets: Cigarette smoking induces a marked,, transient increase in reversed in just 30 days after cessation of smoking (Moffsat R.) (33). smoking on HDL cholesterol does not seem to be cumulative and cam be correlates positively with HDL cholesterol subfraction. Finally, the effect of decrease associated with cigarette smoking. Alcohol consumption, however, controversial and recent studies [34]' Snd' no statistically significant HDL fraction [33]. The data on association between CHD and HDL subfraction is HDL tend to be lower and LDL slightly elevated. The most important aspect could be the sharp decrease in the anti-atherogenic HDL2 cholesterol Lipoproteins: a number of studies have shown that in smokers' plasma heart and therefore increases oxygen consumption [32]. sympathoadrenal stimulation [31] : but increases the delivery of FFA to the lipolysis. FFA mobilisation from adipose tissue is a consequence of injected nicotine raises plasma concentration of FFA through enhanced 3. Effects on lipids: Increase in free fatty acids (FFA): intravenously myocardial contractile function [30],. explained by direct action of nicotine or by cathecol6mine action on 2. Increase in blood pressure and heart'rate: this effect can be readily adrenal glands or cardiac tissues [521. increase sympathetic nervous activity and': release catecholamines from 1. Nicotine alone or in conjunction with cigarette smoking is known~ to Maiastream smoke (MS) stimuli (k}g. 3) 436

levels in smokers are about 5 % but may reach 10 % and more in heavy smokers. CO can affect permeability of endothelial wall, fibrinogen retention by arterial wall and PGIZ-TXA2 balance [41J. 7. Polycyclic Aromatic Hydrocarbons (PAHs): By weekly PAH injections in pectoral muscles of white carneau pigeons, Revis [42) showed that PAH such as Benzo(a)' pyrene (BaP) with the exclusion of BeP, might be the only potential atherogen in avian atherosclerosis. Randerath [43J also demonstrated on mice dermally treated with cigarette tar presumably containing aromatic compounds like BaP, induced lesions in heart DNA in a tissue specific manner. However, the administration route, the doses and the species, cannot convincingly lead to the conclusion that it is an atherosclerosis risk for a smoker. 8. Free radicals: We know that smoke contains free radicals and that free radicals are found in the atheroma plaque. Free radicals have been, implicated in cardiac ischaemic artery [44] and congestive heart failure [451. Free radicals can cause lipoperoxidation of unsaturated fatty acids and then form MDA (Malondialdehyde). LDL malonisation, then leads to increased fixation on macrophages with foam cell' production [57). 1 To sum up (Fig. 3); consistent evidence is now available to explain the aggravating effect of tobacco smoke on atheroma plaque. Nicotine and carbon monoxide are the identified primary stimuli causing a chain of biochemical reactions accelerating the atherosclerosis process [54]. Free radicals and polycyclic aromatic hydroxarbons are among the molecules recently incriminated but their precise role and mode of action require further investigation. A AND PASSIVE SMOHIIVG ETS exposure has no marked effects on atherosclerosis parameters. This is due to the fact that amounts of active compounds which penetrate into the body are on the whole very small even for heavy exposures, as is most convincingly demonstrated by Scherer [46) (Table 3). The findings provide experimental evidence that for passive smoking, exposure to the gas phase of ETS is more important' than to the particulate phase. In contrast to smoking, uptake of tobacco smoke derived particles during passive smoking seems to be very low and not detectable by usual methods [46]! Therefore, nicotine and cotinine in smokers reflect smoke particle exposure whereas in passivee smokers these parameters indicate exposure mainly to tobacco smoke vapour phase. Let us consider the blood stimuli generated by ETS, likely to contribute to atherosclerosis process. co.ooHb As Carbon monoxide is mainly a vapour phase compound of sidestream. 437

Table 3. Estimated dose ratio between smoking and passive smoking from G. Scherer [46]. Tobacco smoke constituents Smoking S (20 eig/d)i Passive smoking PS (81Jd)~ Doae ratio SJPS V-Phase CO (mg) 40-400' 14.4-96 2.7-4.2 Volatile nitrosamines (ug) 0.05-1.0 0.03-0.4' 1.5-2,6 Benzene (ug) 200-1200 40-400 3-5 Particulabe matter Particles (mg) 75-300 0:024-0.24 1250-3000 Nicotine (mg) 7.5-30 0:08-0.4 75•90 Benzo(a)-pyrene (ug) 0.15-0.75 OA01,0.011 75-150 Tobaccospecific nitrosamines (ug) 4.5-45 0.002-0:010 2300-4500 smoke, an exposed non-smoker shows a significant increase in CoHb after heavy ETS exposure. However, CO uptake is 2.7 to 4.2 times lower than in an active smoker (table 3) [46] CoHb levelk obtained range from 0.5 to 1.5% (National Research Council 81, Aronow 78, Wald 81 [47], Davis [48]; though Sherer [46] found a higher CoHb value 6 % after 8 hours''exposure. In fact, 1%. CoHb is considered to be representative of average tobacco smoke exposures,, which is not far from levels observed in exposures to other CO sources: cooking, heating, exhaust fumes, etc... (3 % CoHb in non-smoking taxi drivers in London): In active smokers, however, CoHb levels are much higher: 5% and more. Moskowitz [47] found that whole blood 2-3 diphosphoglycerate (2-3 DPG) was higher in~ smoke-exposed than in unexposed' children, which shows that the organism attempts to compensate for hypoxia by increasing 2-3 DPG level in blood' to meet tissue oxygen requirements. However, the results are significant for boys only. Nicotine In a non-smoker, plasma nicotine rises so faintly after exposure to tobacco smoke that' variations observed are sometimes not significant. Regarding significant quantities absorbed, Sherer [46] has recently shown (table 3) that an active smoker's (S) uptake is 75 to 90 times that of a passive smoker (PS). Regarding plasma, salivary or urinary concentrations, Jarvis [49]ihas found that the ratio is about 100. I'n these circumstances, direct or indirect action of nicotine on an ETS expose& non smoker can only be very weak. Lipids Only a few studies have investigated this aspect. Moskovit'z [47] found that High Density Lipoprot'ein (HDL) cholesterol' was lower in ETS expose& children ; the HDL2 cholesterol subfraotion was decreased but in boys only 438'