Rauchfrei-sg.ch

Passive smoking and risk of coronary heart
disease and stroke: prospective study with
cotinine measurement
Peter H Whincup, Julie A Gilg, Jonathan R Emberson, Martin J Jarvis, ColinFeyerabend, Andrew Bryant, Mary Walker and Derek G Cook 2004;329;200-205; originally published online 30 Jun 2004; BMJdoi:10.1136/bmj.38146.427188.55 Updated information and services can be found at: References
This article cites 24 articles, 18 of which can be accessed free at: 3 online articles that cite this article can be accessed at: Rapid responses
33 rapid responses have been posted to this article, which you can accessfor free at: Email alerting
Receive free email alerts when new articles cite this article - sign up in the box at the top right corner of the article To order reprints of this article go to: We thank the many people who supported the conduct of this (Prepared for the Gulf War Illness Advisory Committee, Department of study: representatives of the armed services; the British Legion, National Defence, by Goss Gilroy Inc. Management Consultants.) in particular Terry English; National Gulf Veterans and Families www.dnd.ca/site/reports/Health/vol1_TOC_e.htm (accessed 10 May2004).
Association; Gulf Veterans Association; and, most importantly, Ishoy T, Andersson AM, Suadicani P, Guldager B, Appleyard M. Major the study participants for the time and effort involved. We also reproductive health characteristics in male Gulf veterans. Dan Med Bull acknowledge the skills and commitment of those who worked on the study, particularly Tommy Clarke, Graham Davies, Kang H, Magee C, Mahan C, Lee K, Murphy F, Jackson L, et al. Pregnancy Haydon Hughes, Juliet Jain, Sam Lewis, Margo Pelerin, Susan outcomes among U.S. Gulf veterans: a population-based survey of 30,000veterans. Ann Epidemiol 2001;11:504-11.
Prior, Darren Reed, Patrick Sampson, and Janet Sullivan. For Araneta MR, Schlangen KM, Edmonds LD, Destiche DA, Merz RD, supplying cohort data and for invaluable help with queries, we Hobbs CA, et al. Prevalence of birth defects among infants of Gulf veter- thank all members of the Gulf Veterans Illness Unit (now the ans in Arkansas, Arizona, California, Georgia, Hawaii, and Iowa, Veterans Policy Unit) at the Ministry of Defence, in particular 1989-1993. Birth Defects Res Part A Clin Mol Teratol 2003;67:246-60.
Nick Blatchley, John Graham, Philip Bolton, Linda Walpole, and Sim M, Abramson M, Forbes A, Glass D, Ikin J, Ittak P, et al. Australian Gulfveterans’ health study 2003. Vol 2. Commonwealth Department of Chris Baker. We also thank Steve McManus and colleagues at Veterans’ Affairs, 2003. www.dva.gov.au/media/publicat/2003/gulfwarhs the British Forces Post Office for providing valuable information on serving status and addresses.
Doyle P, Maconochie N, Davies G, Maconochie I, Pelerin M, Prior S, et al.
Miscarriage, stillbirth and congenital malformation in the offspring of UK veterans of the first Gulf war. Int J Epidemiol 2004;33:74-86.
Funding: The study was funded by the UK Ministry of Defence 10 Maconochie N, Doyle P, Davies G, Lewis S, Pelerin M, Prior S, et al. The and administered by the UK Medical Research Council.
study of reproductive outcome and the health of offspring of UKveterans of the Gulf war: methods and description of the study population. BMC Public Health 2003;3:4.
Ethical approval: West Midlands multi-centre research ethics 11 World Health Organization. WHO laboratory manual for the examination of committee; London School of Hygiene and Tropical Medicine human semen and sperm-cervical mucus interaction. 3rd ed. Cambridge, UK:Cambridge University Press on behalf of WHO, 1992.
12 Breslow NE, Day NE. Statistical methods in cancer research. Vol II. The analysis of cohort studies. Lyon: International Agency for Research on Penman AD, Currier MM, Tarver RS. No evidence of increase in birth defects and health problems among children born to Persian Gulf veter- 13 Abou-Donia MB, Suliman HB, Khan WA, Abdel-Rahman AA. Testicular ans in Mississippi. Mil Med 1996;161:1-6.
germ-cell apoptosis in stressed rats following combined exposure to Cowan DN, DeFraites RF, Gray GC, Goldenbaum MB, Wishik SM. The pyridostigmine bromide, N-diethyl m-toluamide (DEET) and permeth- risk of birth defects among children of Persian Gulf veterans. N Engl J rin. J Toxicol Environ Health A. 2003;66:57-73.
14 Military Health Research Advisory Group (MHRAG). MRC review of Araneta MRG, Moore CA, Olney RS, Edmonds LD, Karcher JA, research into UK Gulf veterans’ illnesses 2003. London: Medical Research McDonough C, et al. Goldenhar syndrome among infants born in Council, 2003. www.mrc.ac.uk/pdf-gulf-illness-review.pdf (10 May 2004).
military hospitals to Gulf veterans. Teratology 1997;56:244-51.
Health study of Canadian forces personnel involved in the 1991 conflict in thePersian Gulf. Vol 1. Canadian Department of National Defence, 1998.
Passive smoking and risk of coronary heart disease and
stroke: prospective study with cotinine measurement
Peter H Whincup, Julie A Gilg, Jonathan R Emberson, Martin J Jarvis, Colin Feyerabend,
Andrew Bryant, Mary Walker, Derek G Cook
Abstract
Hazard ratios (for cotinine 0.8-14.0 v ≤ 0.7 ng/ml) were particularly increased during the first (3.73, 1.32 Objective To examine the associations between a
to 10.58) and second five year follow up periods (1.95, biomarker of overall passive exposure to tobacco 1.09 to 3.48) compared with later periods. There was smoke (serum cotinine concentration) and risk of no consistent association between cotinine Design Prospective population based study in general
Conclusion Studies based on reports of smoking in a
practice (the British regional heart study).
partner alone seem to underestimate the risks of Participants 4729 men in 18 towns who provided
exposure to passive smoking. Further prospective baseline blood samples (for cotinine assay) and a studies relating biomarkers of passive smoking to risk detailed smoking history in 1978-80.
of coronary heart disease are needed.
Main outcome measure Major coronary heart
disease and stroke events (fatal and non-fatal) during20 years of follow up.
Introduction
Results 2105 men who said they did not smoke and
Meta-analyses examining the effect of living with a who had cotinine concentrations < 14.1 ng/ml were cigarette smoker on risk of coronary heart disease divided into four equal sized groups on the basis of (CHD) among non-smokers have shown an overall cotinine concentrations. Relative hazards (95% increase in risk of about one quarter, after adjustment confidence intervals) for coronary heart disease in the for potential confounding factors.1 2 Passive smoking second (0.8-1.4 ng/ml), third (1.5-2.7 ng/ml), and fourth (2.8-14.0 ng/ml) quarters of cotinineconcentration compared with the first ( ≤ 0.7 ng/ml)were 1.45 (1.01 to 2.08), 1.49 (1.03 to 2.14), and 1.57 This is the abridged version of an article that was posted on (1.08 to 2.28), respectively, after adjustment for bmj.com on 30 June 2004: http://bmj.com/cgi/doi/10.1136/ established risk factors for coronary heart disease.
BMJ VOLUME 329 24 JULY 2004
Living with someone who smokes accounts for less than half of the variation in cotinine concentration among non-smokers4 and does not take account of additional exposure in workplaces and in public places (particularly pubs and restaurants).5 Biomarkers of passive exposure to smoking, particularly cotinine (a nicotine metabolite), can provide a summary measure of exposure from all these sources.6 Although cotinine concentration in non-smokers has been strongly related to prevalent CHD,7 there are no published reports of the prospective associations between serum cotinine con- centration and risk of CHD and stroke in non-smokers.
0-0.9 1-1.9 2-2.9 3-3.9 4-4.9 5-5.9 6-6.9 7-7.9 8-8.9 9-9.9 We examined these associations in the British regional heart study, a prospective study of cardiovascular disease in middle aged men, using retained baseline samples for Fig 1 Distribution of serum cotinine concentrations among current
non-smokers; lifelong non-smokers and former smokers are shown retrospective measurement of cotinine.
Epidemiology andPublic Health, RoyalFree and University reviews of patients’ records every two years throughout The British regional heart study is a prospective study follow up.8 The analyses presented are based on all first of cardiovascular disease in 7735 men aged 40-59 major CHD or stroke events during the follow up Martin J Jarvisprofessor of health period to December 2000, with an average follow up of Baseline assessment—In 1978-80, research nurses 18.5 years for men who had no myocardial infarction administered a questionnaire on present and previous smoking habits (cigarettes, cigar, pipe), alcohol intake, Definition of baseline smoking status—Men were physical activity, and medical history (including angina, classified as “current non-smokers” at baseline if they myocardial infarction, stroke, and diabetes diagnosed reported that they did not smoke cigarettes, cigars, or a by a doctor). Participants also completed a question- pipe and had a serum cotinine concentration < 14.1 naire on chest pain. Blood pressure, non-fasting total ng/ml.10 Among these men, “lifelong non-smokers” serum cholesterol concentration, and high density were those who reported never having smoked lipoprotein (HDL) cholesterol were measured. Serum cigarettes, cigars, or a pipe. For comparison purposes, samples were placed in long term storage at − 20°C in “light active smokers” were men who reported the last 18 study towns. In 2001-2, these were thawed smoking 1-9 cigarettes a day, irrespective of cotinine Follow up—All men were followed up for all cause Statistical methods—We used Cox proportional haz- mortality and cardiovascular morbidity. Deaths were ards models, stratified by town of residence, to examine flagged by the NHS central registers. We obtained the independent contribution of serum cotinine information on non-fatal CHD events and strokes concentration to the risks of CHD and stroke. These from general practitioners’ reports, and from regular produced relative hazards, adjusted for age and other Hazard ratios for cotinine group and risk of coronary heart disease (CHD) over 20 years of follow up Cotinine concentration (ng/ml)
2.8-14.0
Active smokers (1-9/day)
P value for trend*
Excluding former smokers
*From tests for linear trend between log (cotinine) concentration and CHD hazard across passive smoking groups.
†For CHD: adjustment 1 stratified by town and adjusted for age; adjustment 2 additionally adjusted for systolic blood pressure, diastolic blood pressure, total cholesterol, HDL cholesterol, FEV , height, and pre-existing CHD; adjustment 3 additionally adjusted for BMI, triglycerides, white cell count, diabetes, physical activity (none, occasional, light, moderate or more), alcohol intake(none/occasional, light/moderate, heavy), and social class (I, II, III non-manual, III manual, IV, V, and Armed Forces).
BMJ VOLUME 329 24 JULY 2004
risk factors, for each quarter of the distribution ofserum cotinine concentration compared with the lowest. We carried out overall tests of the association, fitting the continuous relation between log cotinine concentration and risk of CHD (see bmj.com). Wefitted age, body mass index, height, systolic blood pres- sure, diastolic blood pressure, serum total cholesterol,high density lipoprotein cholesterol, white cell count, lung function (forced expiratory volume in one second (FEV )), and triglyceride concentration as continuous variables. Physical activity was fitted as a factor with four levels, alcohol intake with three levels, and social class with seven levels. History of cigarette smoking,pre-existing CHD, and diabetes were fitted as dichoto- Fig 2 Proportion of men with major CHD by years of follow up in
each smoking group. “Light passive” refers to lowest quarter ofcotinine concentration among non-smokers (0-0.7 ng/ml), “heavypassive” to upper three quarters of cotinine concentration combined (0.8 to 14.0 ng/ml), “light active” to men smoking 1-9 cigarettes aday In the last 18 towns of the study, 5661 men took part(78% response rate). For 4729 of these we had data onhistory of smoking and blood samples for cotinine Influence of follow up period—In a Kaplan-Meier plot analysis. These men resembled the whole study popu- showing the cumulative proportions of men with lation in reported smoking habits and risks of CHD major CHD over time among three groups (light pas- and stroke. A total of 2158 men reported that they sive (lowest cotinine quarter), heavy passive (upper were current non-smokers, of whom 2105 (97.5%) had three cotinine quarters), and light active (1-9 serum cotinine concentrations < 14.1 ng/ml. Of these, cigarettes/day)) we found that the heavy passive and 945 men were classified as lifelong non-smokers, the light active groups diverged rapidly from the light pas- remaining 1160 as former smokers. The cotinine sive group during the first years of follow up but distributions of these two groups (fig 1) were skewed, remained almost parallel during later years (fig 2). The with a slightly higher geometric mean cotinine among corresponding hazard ratios for cotinine and risk of former smokers than among lifelong non-smokers CHD in separate five year follow up periods were high- (1.49 v 1.18 ng/ml). Few men in either group had coti- est in the early years of follow up and declined with nine concentrations close to the 14.1 ng/ml cut off.
increasing duration of follow up (bmj.com). These pat- Serum cotinine and cardiovascular risk factors— terns were little affected by adjustment for cardiovascu- Among current non-smokers, cotinine concentrations lar risk factors, and again the hazard ratios for the were not consistently related to age, total cholesterol heavier passive smoking groups were comparable with concentration, physical activity score, or prevalent those of light active smokers. Restriction of these CHD but showed graded positive associations with analyses to lifelong non-smokers or to men with no mean body mass index, systolic and diastolic blood evidence of pre-existing CHD had no material effect pressure, high density lipoprotein cholesterol, white cell count, and triglycerides (weakly) and positive asso- Serum cotinine exposure and stroke—There was no ciations with the prevalence of former smoking, heavy strong association between cotinine concentration and drinking, and manual occupation. Cotinine concentra- stroke among non-smokers, either before or after tions were inversely associated with FEV , prevalence adjustment for major cardiovascular risk factors (see of low alcohol intake, and residence in southern bmj.com). Analyses based on lifelong non-smokers England. These associations were generally little affected when we excluded former smokers. Lightactive smokers had lower mean body mass index, Discussion
diastolic blood pressure, and FEV and a higher mean white cell count than men who did not smoke.
We believe this is the first published report to relate Serum cotinine concentration and CHD risk—We passive exposure to smoking, based on cotinine meas- examined the association between quarters of the coti- urements, prospectively to the risks of CHD and stroke, nine distribution and CHD hazard ratios among all although cotinine concentration among non-smokers 2105 current non-smokers using the complete follow has been related to prevalence of CHD.7 Our study was up period (table). The risks in the upper three cotinine conducted in a geographically and socially representa- groups were markedly higher than the risk in the tive sample of middle aged men.8 We were able to lowest group, with a relative hazard of 1.61 in the adjust for a wide range of potential confounding vari- highest group in the simplest model, a hazard estimate ables, and this had little effect on the results, in agree- similar to that of light active smokers. The association ment with earlier reports.1 2 A greater concern is the between cotinine concentration and CHD seemed possibility that men in the higher cotinine groups were graded and was not markedly affected by adjustment smoking cigarettes on an intermittent basis. However, for other cardiovascular risk factors. When we study participants had no incentive to provide inaccu- examined the overall association between cotinine rate information, there was close agreement between concentration and CHD, we found that a doubling of reported smoking and cotinine concentration, and the cotinine concentration was associated with a hazard increase of 16% (95% confidence interval 6% to 27%).
approached, even among former smokers. In addition, BMJ VOLUME 329 24 JULY 2004
almost all surviving current non-smokers (99%) contin- What is already known on this topic
ued to report that they were non-smokers in responsesto postal questionnaires five and 12 years later.
Passive smoking (generally defined as living with Passive smoking and CHD
someone who smokes) is associated with an Although the study was modest in size with limited increase in risk of coronary heart disease risk of precision, our results suggest that the relative risk of CHD associated with high levels of passive exposure to Passive smoking may also increase the risk of smoke is greater than that estimated for partner smok- ing alone, even at exposure levels of 20 cigarettes a dayor more.2 The similarity of relative risks among partici- Living with someone who smokes is not the only pants with substantial exposure to passive smoking relevant source of passive smoking, but few studies and those who were light active smokers exposure is have taken account of all sources of exposure by consistent with earlier findings.1 The high relative risk relating biomarkers such as cotinine to disease associated with exposure to passive smoking in our study probably reflects the wide range of cotinine con-centrations observed among non-smokers; the average What this study adds
concentrations of the lowest and highest cotinine Higher concentrations of serum cotinine among quarters differed by almost 10-fold. This range seems non-smokers are associated with an excess risk of considerably wider than that which might be expected coronary heart disease of about 50-60%, but show from partner smoking alone, which in recent data has been associated with average increases in cotinine con-centration of about 3.6-fold.4 A 3.6 fold difference in Risks associated with passive smoking are cotinine concentration would, on the basis of the log widespread among non-smokers in this study cotinine-CHD association defined in the present study, be associated with a relative risk of CHD of about 1.32(95% confidence interval 1.12 to 1.55)—a figure similar to the estimates of the effect of a partner smoking concentration and coronary heart disease seems obtained in the earlier systematic reviews.1 2 to decline with time since assessment of exposure, The marked attenuation in the relative risks suggesting that studies examining the association relating cotinine and CHD with increasing duration of of passive smoking to coronary heart disease over follow up suggests that a single measurement of long follow up periods may have underestimated cotinine is a weak measure of the long term effects of passive smoking. In the present study population, ahigh initial cotinine concentration probably provides asystematic overestimate of the true longer term passiveexposure to smoke, which has declined markedly in sive exposure to tobacco smoke and risk of stroke sug- Britain during the follow up period—partly because of gests that the association between passive smoking and a declining prevalence of active smoking11 and partly CHD is specific and is not simply due to selection bias because passive exposure to smoke in public places placing high risk participants in the higher cotinine and work places has declined.12 In studies with a single assessment of exposure, analyses based on the first Conclusion
decade or so of follow up may well provide a better High overall exposure to passive smoking seems to be indication of true relative risks than those based on associated with a greater excess risk of CHD than part- ner smoking and is widespread in non-smokers, As well as providing evidence that the relative risks suggesting that the effects of passive smoking may have associated with overall passive exposure to tobacco been underestimated in earlier studies. Our results add smoke are higher than those associated with partner to the weight of evidence suggesting that exposure to smoking alone, the results indicate that important passive smoking is a public health hazard and should degrees of exposure to passive smoke were widespread in the present cohort, in which three quarters of non-smokers had an increased risk of CHD associated with increased exposure to cotinine. In contrast, the Funding: The analysis of cotinine measurements was supported prevalences of partner smoking in studies of by a Project Grant from the BUPA Foundation. The British non-smoking men in earlier reports have often been regional heart study is funded by the British Heart Foundationand the Department of Health.
well below 50%.13 14 Taken in combination with the higher estimated relative risks, the high prevalence of Ethical approval: Ethical approval was provided by the Medical exposure suggests that the contribution of passive Research Council and subsequently by the relevant local and smoking to the population attributable risk of CHD multicentre research ethics committees.
Passive smoking and stroke
Law MR, Morris JK, Wald NJ. Environmental tobacco smoke exposureand ischaemic heart disease: an evaluation of the evidence. BMJ Our results do not support the earlier suggestion from a retrospective case-control study that passive tobacco He J, Vupputuri S, Allen K, Prerost MR, Hughes J, Whelton PK. Passivesmoking and the risk of coronary heart disease—a meta-analysis of smoke exposure increases risk of stroke,3 though the epidemiologic studies. N Engl J Med 1999;340:920-6.
confidence intervals around the hazard estimates are Bonita R, Duncan J, Truelsen T, Jackson RT, Beaglehole R. Passive smok-ing as well as active smoking increases the risk of acute stroke. Tob Control wide. The lack of an apparent association between pas- BMJ VOLUME 329 24 JULY 2004
Jarvis MJ, Feyerabend C, Bryant A, Hedges B, Primatesta P. Passive smok- 10 Jarvis MJ, Tunstall-Pedoe H, Feyerabend C, Vesey C, Saloojee Y.
ing in the home: plasma cotinine concentrations in non-smokers with Comparison of tests used to distinguish smokers from nonsmokers. Am J smoking partners. Tob Control 2001;10:368-74.
Public Health 1987;77:1435-8.
Jarvis MJ, Foulds J, Feyerabend C. Exposure to passive smoking among 11 Jarvis MJ, Goddard E, Higgins V, Feyerabend C, Bryant A, Cook DG.
bar staff. Br J Addict 1992;87:111-3.
Children’s exposure to passive smoking in England since the 1980s: coti- Jarvis M, Tunstall-Pedoe H, Feyerabend C, Vesey C, Salloojee Y.
nine evidence from population surveys. BMJ 2000;321:343-5.
Biochemical markers of smoke absorption and self reported exposure to 12 Walker A, O’Brien M, Traynor J, Fox K, Goddard E, Foster K. Living in passive smoking. J Epidemiol Community Health 1984;38:335-9.
Britain: results from the 2001 general household survey. London: Stationery Tunstall-Pedoe H, Brown CA, Woodward M, Tavendale R. Passive smok- ing by self report and serum cotinine and the prevalence of respiratory 13 Sandler DP, Comstock GW, Helsing KJ, Shore DL. Deaths from all causes and coronary heart disease in the Scottish heart health study. J Epidemiol in non-smokers who lived with smokers. Am J Public Health Community Health 1995;49:139-43.
Walker M, Shaper AG, Lennon L, Whincup PH. Twenty year follow-up of 14 Hole DJ, Gillis CR, Chopra C, Hawthorne VM. Passive smoking and car- a cohort based in general practices in 24 British towns. J Public Health Med diorespiratory health in a general population in the west of Scotland. BMJ Feyerabend C, Russell MA. A rapid gas-liquid chromatographic method for the determination of cotinine and nicotine in biological fluids. JPharm Pharmacol 1990;42:450-2.
Acquisition of Helicobacter pylori infection after outbreaks of
gastroenteritis: prospective cohort survey in institutionalised
young people
Rémi Laporte, Philippe Pernes, Pascale Pronni, Frédéric Gottrand, Pascal Vincent
The exact mode of spread of Helicobacter pylori is still detected by using the non-invasive HpSA stool unknown. Transmission during transit disorders of the antigen test (Meridian; 91% sensitivity and 93% gastrointestinal tract has been suggested, although specificity).4 Stool samples were stored and trans- there is no evidence to date of transmission during ported at 4°C within 48 hours and then frozen. The outbreaks of gastroenteritis.1–3 We determined whether residents were monitored for one year. Events and gastroenteritis in young people infected with H pylori clinical data were recorded daily by nurses. Gastroen- can lead to acquisition of the bacterium by peers.
teritis was defined as a sudden outbreak of liquidstools in more than two residents concurrently. For Centre Antoine deSaint-Exupéry,Vendin-le-Vieil, Participants, methods, and results
each patient with H pylori infection, we defined oneday of potentially infective diarrhoea or vomiting as Our study took place in a French institution for that when at least one liquid stool or vomitus was neurologically handicapped children and adolescents.
emitted. Stool samples were collected every week for The young people had been institutionalised for each resident who was free of infection at the study several years and resided across five housing sections outset, and at the end of the study we compared the (A to E). We included all 112 residents in May 2001. H last sample with the first sample. When conversion pylori infection present at the outset of the study was was observed, we identified the oldest positive stool sample collected from that patient during follow Person days of infective diarrhoea per week The prevalence of H pylori infection was high; 47 Diarrhoea outbreak level limit (95% limit Poisson distribution) of the 112 residents (42%) were positive for H pylori at Person days of infective vomitus per week the study outset. Seven of the 65 residents who were initially negative for H pylori showed conversion Individual living in another section but in close contact with during follow up (figure). Five of the seven young peo- ple lived in section E, and our records showed that the two other residents had frequent contacts with the infected patients from section E during physiotherapy and entertainment sessions. Vomitus was rare in all sections. The frequency of diarrhoeal stools from the infected patients varied across sections; residents of sections A and B had 475 and 338 person days of potentially infective diarrhoea over the year, respec-tively. Acute diarrhoea was rare in these sections, and no outbreaks of gastroenteritis were recorded. A lower May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May frequency of diarrhoeal stools was observed in sections C, D, and E; 34, 104, and 164 person days ofpotentially infective diarrhoea over the year, respec- Relation between new cases of Helicobacter pylori infection andexposure to infective diarrhoea and vomitus in housing section E tively. The frequency was always low in sections C andD, where no outbreaks of gastroenteritis were BMJ VOLUME 329 24 JULY 2004

Source: http://www.rauchfrei-sg.ch/downloads/studien/Passiv_smoking_and_CHD_and_stroke_2004.pdf