Tobacco Documents Online

e(yPiN- „rQs4n TC. al -To/.;qJr iUrvw- CSe9S;~... ~ n J r~ . IwVC .w~ol ~~~wco Sw.Ae ' Sc:a...< MA-- `tx1 tL 1A .A... or , ~ y , 1 CORONARY HEART DISEASE AND INVOLUNTARY SL4DKiNG Joachim Thiery. N..D. ar.d Peter Cremer. M.D. Institute for Clinical Chemistry, Klinikam GroBhadern. University of Munich, :^larch_oninistraBe 15. D-8000 Munich 70, Germany. The coronary hoart disease (CHD) is one of the major causes of death in the western world. There is general consensus on the central role of the chalesteruli concentrations in hiood, reflecting the levels of athc_ogenic 1_uoproteins, especially of low density lipoproteihs in the developmen= of atherosclerosis. Other risk factors for coronary heazt disease like hypertensi'on, diabetes and active smoking potantiate thP risk for atherosclerosis in hypercholesterolemic patients. These additional risk factors - one leading risk factor is active smoking - enhance the coronary risk parti:ularly in patients :ri-A only mild hypercholesteroLemia (LDL-cholesterol 120-190 mg/dl). vops.ver, in sr.•re.re hypercholiesterolemi~a additional risk factors sLch as a:ctivE smoking do not appear to aggravate corDnary risk any further These findings of our follow-up study on in:idence and prevalence ofcDronary heart disease in 6000 industrial workers are in sgreement Wk-tt other popullation studies [2. 31. InWite of th3 central role of plasma llipoproteins in the development of =ronary heart disease and': of the risk factor 'smoking' it is also w=+nwhile to focus our attention on the effects of involuntary smdding and its relationship to the process of atherogenesis. • ltt:itrosclerosis can be described as a 'response to injury' of the az>saial vall. The initial events causing endothelial injury may vary, =butllipoprotei.ns themyrlves contribute to such injury. This could be -"etcs increased permeabiLity of the endothelium for plasma -2Qpq>roteins, the release of growth factors and chemotactic siabtancea. In addition hypertension, diabetes„ oxidized liipoproteins„

2 higrh plasma fi5rinogen levels and plasma viscosity, active smoking and distress may be further important causes for the initial injury of the vessel wall (1. 5). To determine rizether a relationship between passive smoking and atheroscleroe_s exi~sts, we should first ask for a pathobiochemicaL concept and erperimental investigations to explain the proposed toxiC effects of in.vluutary smoking. Sowever, there a_e no experimental studies that c:an demonrtrate a definite effect of passive smoking on the developmen: of atherosclerosis. But there are a few epidemiologicti invest:gations on the incidence of coronary heart disease in active and passive smokers.-Eiowever,, all of these epidemiological inveytigations reveal a number of weaknesses in design and execution so that the results should be reRarded with a certain amount of caution. The name& deficiencies may be explained by the fact that these studies originally were aimed at investigating other issues and the resuLta have subsequently been analyzed for an association between passivo smoking and CaD. The following deficiencies can be found in literally all population stad:es on pa:,sive smoking and actual CHD. 1. Active smol;ing could not be ruled out with certainty for all the subjects invectigeted who claimed to be passive smokers. 2. Passive smO;ci~ng exposure was inadequately assessed,, and was mostly based on secoudary information about the spouse's smoking habit, while the spouse was often not questioned personally. 3. No assessment of the subjects' smoking behaviour or their passi've smoke exposure was carried out during the study periode.. 4. There was no clear definitioa of the target event (=rmally death from CHD) and no indication of the findings required for the target dixgnosis. -

3 S. The influei:ce of other risk factors generally known to be strongly associated with coronary heart disease, such as hypercholesterolemia. hypertension end'psychosocial factors on the results of the studies has not been taken into account. According to the data published by HIRAYAMA (1981) [6) the risk of death from CHD:was completely identical for women exposed to passive smoking as for those who were not exposed. This analysis was based on 406 deaths from CHD in the observation period of 13 to 15 years and a total of more than 90.000 subjects investigated'. What is striking is the fact that in a reanalysis of the same group carried out in 1984. (7), which was based on only 88 additional, deaths, the same author found a risk, compared to non-exposed'wowen„ which was significantly increased~by a factor of 1.1 for women married to ex-smokere and for women married to smokers with a daily cigarette consumption of 1-19, while women married to heavier smokers showed a risk significantly increased by =.3 times. When we look at the change in the findings between 1981 and 1984, together with the general deficiencies mentioned above, the results of this study must be regarded as quzstionable. In another stu3y by GILLIES (8], from the total study group of 16.000 subiects approximately 1.0ao non-exposed men and women were compared with about 1.700 passive smokers for the risk of death from CHD or stroke. The results were inconsistent; While in women the incidence rates for the vascular diseases stated above were slightly higher for passive smoke=s than for the non-exposed subjects, the incidence rates in men were higher in the non-exposed subjects than in those exposed to passive smoking. Incidence rates always differed by less than one case per 1.000 subjects and year. In view of these marginal differences aad the rather low absolute number of cases, the authors did not see any point in calculating the•significance lievel. In an update of this report by David Hole and coworkers (8j, the

4 authors investigated about further 8000 subjects recruited from a Westscotish popular.ion for the effects of passive smoking in relati~on to cardiovascular symptoms. Passive smokers were groupad in terms uf high and low exposure. The high exposure group showed a large number of symptoms and: an increased mortality rate in coronary heart disease by 2.01. The fact that only 801 of the population cohort were investigated and that the studv was confined to subjects tietween 45-60 years of age, so that only insufficient information on the smoking habits of all persons living in the household was obtained, may be regarded as a li__ting facto_ cf this investigation. Moreover, additional interviews revealed that 5: of the men and 211 of the women, categorized as controls in the study by Hole, lived indeed together with current smokers, who, however, were not incLud'ed in the study. Another s.rious shortcoming maa be seen in the relatively small number of observed deaths; conseqr1ently, confidence intervals are rather large, so that a significant d_fference could'oniy be found for dcaths on ischemic heart disease. with the relative risk of passive smokers of 2.01, cvming CLU6e to that of active smokers by 2.27. 3owever, the relative risk for card_ovascular symptoms in passive smokers r.ompared with controls was sint significantly different. Garland [10] analyzed the risk of death from CHD in 695 married wosen who were categorized according to the spouses' smoking habits, broken down into 'never smokers'. 'ex-smokers', and 'current smokers'. In women married to ex-smokers, the risk of death from CHD was found to be 3.6 times inc€eased, while in women married to current smokers the risk was 2.7 times increased as against non-exposed womsn. Apart from the deficiency mentioned above, it is striking that in this study women married'to ex-smokers are alled'ged to have a higher risk than women married to current smokers. This is hardly in keeping with epidemioloQical findings according,to which even in active smokers after giving up smoking the risk of death from CHD showe a relatively rapid decline to the level of never smokers. Hence, there is hardly

5 1 any pathophye:ologiCal explanation for tne findings of Garland, so that the susp_.cion of misclassification arises. It should be added that the analviis of the study was based on only 19 deaths from CAD, so already one singl'e case of death more or feaer in this or that subgroup of the study population could produce diametrically opposed results. Consequently, statistical significance reaches only a Level' of 901. although, a 952 level is usually required. A further stucLy by Lee [111 is the only non-prospective and the onlyy case control szudy among the investigations dealt with here. L.ae did not find any relationship between passive smoking and coronary heart disease or st_oke. f3owever, it should be pointed out that this study can also not eztirelly be excluded from the criticism mentioned above. As a part of the Mul~tiple Risk Factor Intervention Trial Svendson and coworkers (121 compared a total of 1.3-00 non-smoking men with either smoking or non-smoking wives, with regard to the risk of CHD and death from CHD. Agn_y the study was based only on a small number of cases. The authors fu-ind for those exposed to Fassive smoke in an observation period of 6-8 years a 1.6 times increased risk of newly developed coronary hear: disease and a 2.2 times increased risk of deaths from CHD:. However, the differences in the coronary risk did not reach significance Level. Although the study reveals certain advantages as clearly definrd target events, validnticn of diagnosis of target events, consideration of other risk factors, and assessment of the subjects smok~:ng habits during the study period, these are outweighed by the fact thst the subjects were primarily participating in the CHD intervention study including drastic medical care to reduce the risk factors of CHD, such as a dietary advice or antihypertensive medication. This led to a large number cf confounding factors likelx to affect the results of the study. A particular weakness is the fact that the study also failed to assess the spouses' smoking habita, especially the extent of tobacco smoke exposure during the study period.

6 Finally. Sand':er and coororkers 113) studied the frequency of deaths from CEiD in passive smokers by estimating the deoree of exposure ca the basis of the smoker anamneses of all persons living with the subjects in the aame households. The author compared CHD death rates in these pass:.•re smokers with those in non-exposed subjects and found a 1.2 to 1.3 times elevated risk, with the increase just reaching si~gnificance, durinS the 12 year period of follox-up. A dose-effect relationship couid, however, not be demonstrated!. °urthermore, the critical pnin-s mentioned above apply to this stldy as well. There was no assessment :if passive Rmoke exposure during the period of follow-up and it is unl:-Aely that all persons in a household would maintain the same smoking b.haviour over a period of 12 years. The authors themselves have considerable reservation on this point. Taking into ac:ount the small increase in coronary risk in passive smokers as compared to non-exposed subjects and also the low vaiicity and small numaer of epidemiologicaI studies available and~the fact that their resalta are at least inconsistent, a relationship between passive smoking and cardiovascular diseases cannot be established on these data. The American 5lrgeon General. in his reFort from 1986 arrives to the same ccnclusion by stating [14): 'The magnitude of risk associated with involuntary smoking exposure is uncertain. Sample sizes in most studies are not large, the point estimates of effect are unstable, and confidence limits are broad and generally overlap from one ^tudy to another.' It should be added that the current epidemiologicaL methods seem to be not sensitive enough to reveal a relaticnship betveen passive smoking and CEiD, and :o distinguish it from the effects of major risk factors„ such as hyper+:holesterolemia. Furthermore, a review by the National Research Council (15J estimated

7 that tha relative risk of coronary heart disease in nonsmokers exposed to environmental tobacco smoke as compared with that in true nonsmokers would be approximately 1.02 - an increment difficult to detect or est_mate reLiably in nonexperimental studies. The question L_iseS. whether evidence can be provided on a pathophysiological basis, pointing to a role of passive smoking in the development of cardiovascular diseases. It has widely been accepted that it is pr_marily nicotine and carbon monoxide which may account for an increaFed tendency towards atherosclerotic changes of the arterial wall in active smokers. H'owever, it has been demonstrated that under normal life conditions the blood levels for carbon monoxide, nico_ine and cotinine measured in passive smokers hardly differ from those found'in non-exposed subjects at all, whereas smokers show concentrations of these substances which are many tiIIies higher as compared to the controls [16, 17]. Several active mechanisms are in discussion, at least for nicotine, by whi.h the substance may cor.tribute to _he development of atherosclerosis [18J. However, in view of these low cuncentrati'ons, it can be ruled out as an important factor for an elevated risk of CHD to passive smoking. If therefore nicotine and carbon monoxide cannot be made responsible for an assumed elevated risk of CHD in paesfve smokers, which of the substances wi:l then account for this as toxic agent? So far there a_e no well-documented relevant findings from appropriate experiments or epidemiological studies whatsoever. However, one further experimental consideration could be the determination of chemieally modified lipoproteins and the investigation of the oxidation of _ipoprotein particles in the plasma of passive smokingg subjects. Modified lipoproteins have been found to be preaent in atherosclerotsc plaques and products of oxidation have been seen in early and late lesions [19]. There ia increasing evidence that oxidized lipop_oteins.may play an important role i'n atherogenesie. It has been shown that oxidized LDL enhances monozyte-endothelial 0

8 interactions and can accumuLate in macraphages [20)~. Previous studies could demonstrate that oxid'ation of lipcproteins in plasma derived from smokers is fac'_litated compared to non-smoking controls. However, it can be expected that the concentratit•n of inhaled oxidizing substances and free radicals during pas,ive smoking will be very low. In conclusion it should be pointed out that on the basis of published studies and d3ta available a relationship between passive smoking and coronary heart disease cannot be established. In contrast to these inconsistent Dbservations, it is a matter of fact, that a pria+ary and secondary prevention of caronary heart ciseaee is possibTe by changing life habits including a change of nutrit,ionaL habits, normalizing hypercholesr.erolemia and high blood preESUre and to stop active smoking [21, 22, 23].

1 Cremer ? E Muche P: GtSttinger Risiko-, Inzidenz- und PrBvaleazstudie (GRIPS). Therapeutische Umschau 47 (6)r 482-491 (1990) 2 Stamler J, Wentworth D. Neaton JD for the MRFIT Research Group, Minneap~)lis: Is relationship between serum cholesterol and risk of premature death from coronary keart disease continuous and graded? JAHA 256 (20); 2823-2828 (1986) 3 Castelli WP. Garrison RJ, Wilson FWF. Abbott B.D., Kalouedian S. Kannel 'WB: Incidence of coronary i.eart disease and Iipoprotein cholest_rol levels. JAMA 256 (20): 2835-2838 (1986) 4 Steinberg D: Lipoproteins and the pathogenesis of atheros:lerosis. Circulation 76: :•08-514 (1987) 5 Ross R: The pathogenesis of atherosclerosis: an update. N Engl J Med 314: 488-500 (11986) 6 Hirayama T: Non-smoking wives of f,eavy smokers have a higher risk of lung cancero a study from Japan. Er Med J 282: 183-185 (1981) 7 Hirayama T: Lung cancer in Japan: Effects of nutrition and passive smoking. In: Mi'zeLl H & Correa M(Eds) Lung Cancer: Causes and Prevention. ProceedingE of the International Lung Cancer Upd.te Conference, New OrlEans, Louisiana. Verlag Chemie International, 175-195 (1984) 8 Gillis :R, Hole DJ, Flawthorne VM, Boyle P: The effect of environTental tabacco smoke in two urban communities in the west of Scotland. In: Rylander R. Peterson Y & Sne1La H-C (Ede) ETS- EnviroaTent Tabacco Smoke. Report from aworkshop on Effects and Ezposure Levels. University of Geneva. Switzerland, 121-126 (1983) 9 Hole DJ, Gillis CR, Chopra C, Hawthorne VY.: Passive emo3cing and cardiorespiratory health in a general population in the Ves: of Scotland. Br Med J 299: 423-427 (1989) 10 Garland C. Barrett-Connor E, Suarez L. Criqui MS„ Wingard DL: Effects of passive smoking on ischemic heart disease mortality of nonsmokers. A prospective study. Am J EpidemioL 121: 643-650 (1985) 11 Lee PN,, Chamberlain J, Alderson MP.: Relationship of passive smoking to risk of lung cancer and other smoking-assoeiated diseases. Br J Cancer 54: 97-105 /1986)

12 Svendscn KH, Kuller L8, Martin MJ, Ockene JR: Effects of passive smoking in the multiple risk factcr intervention trial. i+m J Eoiaemi3l :26: 783-795 (1987) 13 Sandler DP, Helsing KJ, Comstock GWr Heart disease mortality in persons living with smokers. Indocr Air '87. Proceedings of the 4th International Conference on ILdonr Air Quality and Climate. Vol. 2, Beri~in, 29-33 (1987) 14 US Department of Health and Human Services: The health consequences of involuntary smoking: a report of the Surgeon Generall, Washington D.C., Guvernmental Printing Office, Publicacion Nr. DHES, CDC, 87-839E (1986) 1S Fieldin6 JE. Kenneth JP: Health effects of involuntary smoking. N.?ngi J Med 319 (22): 1+52-59 (1589) 16 Jarvis 'sJ, Russell MAE, Feyerabeac C: ,Abaorption of nicotine and carbon -nonoxide from passive smoking under natural conditions of expos:xra. Thorax 38: 829-833 (1198_ ) 17 Weld NJ. Boreham J, Bailey A, Ritchie C. Haddow JE, Knight G: Urinary cotinine as marker of breithing other people's tabacco smoke. The Lancet 1:: 230-231 (198L) 18 Allen D3, Browse NL, Rutt DL: Effects of cigarette smoke, carbon monoxide and nicotine on the uptake of fitrinogen by the canine arterial wall. Atherosclerosis 77: 83-88 (11989) 19 Eaberland ME, Fong D, Cheng L: MalondiaLdshyd.-altered protein occurs in atheroma of Watanabe heritable hyperlipidemic rabbits. Science 86: 1376-1376 (1986) 20: Steinberg D, Parthasarathy 5. CazEw Th E, Rhoo JC, Witztum JL: Beyond cholesterol. Modifications of low-density lipoproteir. that increases its atherogenicity. N Engl J MQd 320 (14): 913- 924 (1939) 21 Lipid Researeh Clinics Program: The Lipid Research Clinics Cororiary Primary Prevention Trial results. I. Reduction in incidence of coronary heart disease. JAMA 251s 331-364 (1984) 22 Thiery J, Armstrong VW. Eisenhauer Th, Ad'am R, Janning G,. Crtutzfeldt iI, Kreuzer H, Seidel D: Combination of Simvastatin and Heparin induced LDLIFi~brinogen Precipitation (HELP) in the Treatment of Hypercholesterol~emia in CAD-Patients. In: Crepaldi 6,.Gotto AM, Manzato E (Eds) Atheroclerosis VIII, Excerpta Medica. Amsterdam, 831-835 (1989)