Tobacco Documents Online

3022 C; vt"\ a Letters to the Editor will be published. if suitable, as space permits. Thev should not exceed 1000 words . (t}ped double-spaced) in length and mav be.subject to editing or abridgment. :lnxietY and Risk of Coronary Heart Disease , cwdme the level of exposure for not being relevant to levels To r1)e Ediror: --~ encountered by passive smokers, Coggins claimed that the total Kawychi et'a0 report that anxiety may increase the risk of fatall suspended particulates (TSP) was some 300 times higher than coronar4 heart disease through h;perven[tlation during an acure., would be encountered in a smoking home. Penn and Snyder attack. wlaich could in turn induce coronary spasm, or an acute coun[ered by arguing that if the higher published concentran6ns attack of a'ni.ietv, [ripgonng an episode of fatal ventricular arrhyth- for TSP are used. then the TSP concentration in their study is only mia. The ne'probiology is suggested by studies linking brairtstem ?0 times that reported for public places. cardiovasculak control. cardiovascular reactivity in challenging Clearly, whatever value is taken. the cockerels were exposed to tasks, and the\induction of breathin¢ panic, and vasospasm to levels that passive smokers would not normally encounter. Srre- dopamine iater ized to the right hemisphere =•% This hypotHesis is spective of the level of exposure. however, we would submit that supported by an norrnal parahippocampal region and its afferent there are more important questions that need to be answered and efferent connections in patients vulnerable to lactate-induced before the findings of Penn and Snyder are accepted as evidence of panic.' It is also supported by optimal response organiuationat a duect effect of environmental tobacco smoke (ETS) on the intermediate dopamike tone in a medsal-frontal-strratal activation development of plaques. system' and by a ene(aal role for the medial-frontal cortex in ' The important finding by the ,.athors was that cockerels (from 66 emotional experience aitiJd expression and in oomple[e sympathetic to 22 weeks of age) exposed to sidestream smoke for 6 hours per activation of cardiovascular responses to both severely and mildly day, 5 days per week, showed significantly enhanced growth of stressful stimuli•` These t~tsdings suggest that subclmical impair- aortic plaques. They also found, however, that neither the number ment of lung airways in patients with panic disorder might be due nor distribution of plaques was affected by the exposure. The main to the dysregulation of airw)qy smooth muscle tone by dopamine ' conciusion is thus very dependent on acceptance of the validity of abnormalttiesN They also poit to the existence of deep and lawful the method of measuring plaque growth. -fhe crucial point here is mental structures governing co nitive and emotional functioning, that plaque areas of the exposed and unexposed groups were not reflecting properties of neuronaactivtty.and firing."-it -compared direcdy,. Rather, they were first standardized by dividing ., `'bl •, 3ozZ-7o'a-3,\4 S1^ Correspondence plaque area by luminal circumference to produce a plaque index. Ernest H. Friedman, MD All of the data on plaque development and the statistical analyses DepanmXnts oj ,Nedicine and Prychiany relate to the indices, not to raw data. A plaque index based on se Western Reserve Gnrversrtv erc mference however ma not reflect absolute changes unless u y 1 I I I I I I I I Clevefand, Ohio -. further standardizations are made. For example, it plaque areas were the same for the two groups L Kawachi 4 Sparrow D, Vokonas PS, issSt Symptoms of apxiety and risk of coronary heart disease: [ Norrnative Aging Study. but the vessels of the exposetl group were slightly smaller. then the Circufarion. 1994:90:?]?5-2129. net effect would be an increase in the plaque index. implying 1 Friedman EH, Owens JF, Matthews - Stoney CM, Berga SL inereaseddevelopmen[whennosuchinereaeeoccurted-Ofcourse, 4lenopause and blood pressure. Circrilari . 1994;89:294Z hetter. in such a case, plaques in the smaller vessels would be relatively i. Pitchot W. Ansseau M, Moreno AG. Han nne M. von Frenckell laraer.'Plaaue area-however- may be a function of age (the same in R. Dopaminergie funetion in panic'diso ec comparlson with bothgrou s)andinde endent withinlimits)ofvesselsize.Nodata major and minor depressioa 8io( Pnrhiurn•. l99?:32:1004-101 L p p f 4. Reiman EM, Raichle ME- Robins E, Butler F Herscovitch P• Fox are provided on the sizes of vessels or, as a surrogate, the weights P, Perlmuner S. The application oPpoiitiov e ission tomography of the cockerels in the exposed and unexposed groups. Did the to the study of panic aisorder. Am I Prvchrany. 198fi:143:469-477. exposed group grow at the same rate as the unexposed group over 5. Frysz[ak RJ, Neafsey EJ. The effect ofroedial fr tal cortex lesions „ the 4-month period? , - on cardiovascular conditioned emotional respons in the rat. Brain If. nn the other hand, the vessels, or eauivalent aortic se¢ments Res. 1994:643:181-193, in Penn and Sn der's stu b. Pema G. Marconi C, Battaglia M, Bertani A. Panz Bellodi Y.s dy, weree all fairly similar in s12e, would it la ue nall the raw data on t additi b tt t h b q p y ave een e er o presen o L. Subolinical impairment of lung auways in patie ts wi[h panic not disorder. Biol Prychtany. t994:36:606605. area? That would have done two things. It would have allowed the ', Friedman EH. Bowel dysfunction in young women--OuL 1994; reader to gain an appreciation both of the absolute change in 35:57?. Letter. . plaque size and of the variability between segments and between 8. Tanji J. Shima K- Role for supplementary motor ar cells in animals. planning several movements ahead. Namre. t9941371:4f l6. is crucial to achieving a lin e ade uate sam i - I g , q p n our exper enc 9, Urena 3, Femandez-Chacon R. Benot AR, Alvarez de T edo G. t-opez-Bameo 7• Hypoxia induces voltage-dependent CAenery representative view of the tme state of plaque stze and distribution. and quantal dopamine secretion in carotid body glomuscell Pra In this respect, it is not clear why Penn and Snyder sampied 30 •Varl Acad Sci USA. I994:91:I0208-10211. .. \ animals idthe exposed group but only 12 in the unexposed group. t0. Friedman EH. Karasawa Y. Neurobiology of passive avoid:_M\ce This bias-was exacerbated in the sampling of plaques in each impairment after ischemia. Stroke. 199425:15?6. !-etrer. segment, where, for the con[rols. 3 to 10 samples were taken (for a 11. Friedman EH. Neurobiology of cigarette smoking in peptic ulce(. total of 50), and for the smoke-exposed animals. 6 to 24 samples diseazeJ C1in GastroenteroL 7494:I8:89-90. Le[ten . - --- were taken (for a total of t53)• Although different sample sizes are .~. - - - generally acceptable if certain conditions ar< met or particuiarN Sidestream. Cigarette Smoke and Arteriosclerosis future investigations are envisaged, in this case there may be cause~.,/ To the Editor.• . .. .. , for concem because of the skewed distribution of plaque s¢e. Am Coggins' criticized the study by Penn andSnyder' (Inhalation pf few large plaques could strongly influence the mean value, andc4 sidestteam smoke accelerates development of arteriosclerotic - undersampling may result in the mean and median values beingGo plaques. Clrculanon. 1993:88:18?0-1825.) on several grounds.. in- . closer together. In Fig 2 of the article by Penn and Snyder. thecA) . . . .. _. . . . . .... . ~~ w a. ~3

Correspondence I I I I I I I I differences between the mean and median values for the control values were still smaller than rhe crorrespondine means tor all eight group(n=12) arenoticeablysmatlerthanfortheexposedgroup se[sofplaquesegments.(Fortwoo[hersetsofseemen[s.eachwith (n=301. Moreover, in the exposed group- the two large means for only two plaque samples. no meammedian ratio was determined.l segments'_ and 10 may have high leverane in the analysis and hence In the smoke-exposed eroup. the means also exceeded the medians. <ubstannalh• influence the conclusions. These mac also be infiu- this time in eight of ninese[s.. The tenth set had only tuo samples. enced by missing.values such as the medians forsegmenr 10 of the In addition, on a segmenrbv-segment basis- mean cross-sectional exposed group and of segments 8 and 9 of the unexposed group. area measurements were still larger for smoke-exposed cockerels to a combination . than for air controls in every case- oP dtm t p I ne ~b sss and d ffer nce; tnmeslsel g s b ize'. With the published The suegetion that we present all the raw data is unrealistic. We informatton. this is not possible to ascertain. Since there are could have provided summary statistics for each animal and each apparently only a total of 203 plaque mdes dara points. n.wouidbe segment. as the writers suggest. Ho.e.er. we decided that our helpful to see the individual area and circumferer•ce value or at manner of presentation was clear and direct. Apparently, the least summare statisncs for each animal and for each segment. reviewers agreed. The issue of enhanced plaque ¢rowth is,importanrs since there is The number of animals chosen for the experiment was based on tncreasine debate over the effects.of ETS on health. In Glantz and results from many previous experiments tie. on esperience). For Parmley's review' of the effects of passive smoktn¢ qn hean our subsequent `sidestream smoke from one cigare[te' study disease, the epidemtoloetcal and btologtcal data are used to (Circularron. 1994:90:1363-1367.1. a power study was carried out reinforce each other to reach the conclusion that. based on the based on oui'-five-ciearette studc' to help_ us determine the combined weight of eeidenoe. ETS causes heart disease. The - approprfate numbers of animals to test. Regarding sampling bias, eptdemiological evidence. however, is not strong. and the small Drs Merrileesand Mengersen are incorrect to say "... for the increase in relative risk (1.3) has been challenged' as not beingfree controls. 3 to 10 samples were taken ....'- We `took" all samples. of bias and confounders. Demonstration of a direct effect on lesion Every measurement was made double-blind. There were actually growth is thus potentially very important for resohtng whether or . three sets of codes that had to be broken before any of the not E7S is a causative a¢ent. It is therefore particularly important calculated plaque index values were assigned to control or treated that data supporting a direct effect are scrutiniied carefully and groups. There were 2.5 times as many cockerels in one group as in analvzed correctly and all alternative ihteroretations ruled out. It is the other and there were 25 times as many total plaque-containing for this reasoh that while the quesnonsraised in this letter may . segments in one group as in the uther. There is no evidence of seem trivial. they are of central impoitance to the conclusions of sampling bias. the studv in question and the issue in getierai. The last paragraph of the letter raises some important points ' - with which we concur but omits reference to iwo recent. relevant Mervyn J. Merrilees, PhD publications. One by Wells (.1 Am Cofl Cardiol. 199424:546-554.) • Department of Anatomy discusses epidemiological evidence relating ETS exposures to heart p'nirersirvofAuckland disease dea[hs. Wells's calculations show that every year in the Auckland, New Zealand United States, over 80 000 excess deaths from heart disease occur Kerrie Mengersen, PhD as a direct result of earlier exposures to ETS. The secomf study was Department of Srarisrics our own'sidestream smoke from onecigarette" study noted in the . . Queensland UnirersiLy bf Technology paragraph above. Again, there is a significant size increase in the . B.isbane, A¢st.alia population of plaque-containing segments from smoke-exposed cockerels compared with controls. The ew-onmental relevance I. Cuggins CRE Sidestream cigarerte smoke. Circulation. 1994: was confirmed by showing that carbon monoxide levels (a surrogate y9;+9y3 ?or ETS) in taverns where heavy smoking was taking piace equaled ]. Penn A- Snyder CA Inhalation of sidestream cigarette smoke accef- or exceeded the carbon monoxide levels faced by the cockerels in eraces de.elopment ofa[heroseterotic plaques. Circula+ion:.1993:88: the one-cigarette exposure group. Finally, we never say that "ETS i820-18'_s. -- is a causative agent." We do say that it promotes plaque develop- 3. Glantz SA. Parmley W W, Passive smoking and heart disease: epi- ment. The distinction between "causation" and "promotion" is not qa-L.de,...alo gy,-..L, alog AbstractA y, physiology, and biochemisuy. Cixufarion. 1991:83(suppl semantic and reflects the mechanism by which ETS and other . . . 4. Lee PN. Environmenml Tobacco Smoke and :NonaliN. Baset- Swi[- environmental agents act iolhis system. zerland. S Rarger: 199? ' Arthur Penn, PhD Response -~ Res Prof Environ Med We appreciate the concern of Drs Merrilees and Mengersen Carrolf A. Snyder, PhD regarding questions of standardized values and sampling bias Res Prof Environ Med values. We recognize that there are definite advantages topre- - NYU Medical Center senting data without normalizing values. In some of our earliest - •'/elson lnsarure o(Environmental Medicine studies. we reported plaque sizes as a0ss-sectional areas. (In those . Anthony J Lanza Research Laboraroder studies, as in the case of the present sidestream smoke studies, the Turedo, NY treatments had no significant effect on body weight.) However, morphometrv specialis[s pointed out that without normalizing Hemodynamic Effects of Sympathetic Pic - plaque size to a separate measure of artery wall dimension, we Efferent Reinnervation After Orthotopic would be liable to introduce errors due.to sectioning angle differ- Cardiac Transplantation . ences from one artery segment to the next. By including artery To the Editor: circumference in every calculation of plaque size, we help ensure We were interested in the elegant demonstration of the hemor~,~ that sectioning affects plaque and the underlying artery wall dynamic effects of sympathetic relnnervation more than 4 month9tY similarly. Subsequently, we were commended for taking this pre- after cardiac transplantation, reported in Circulation by Burke ac caution by reviewers of one ofourgrant proposals. .' alJ The same group has previously demonstrated Punetional rein4A To help alleviate the concern of Drs Merrilees and Mengersen, nervation of the sinus node in a similar group of subjects.z %J we reevaluated our data in terms of plaque s¢e only, without We have also investigated the functional consequences of sym-0 consideratfon of artery wall dimensions, For the controls, median pathetic efferent reinnervation of the sinus node in 16 humanW_S ~ ~ ~