Ambient Air Pollution Is Associated With Increased Risk of Hospital Cardiac Readmissions of Myocardial Infarction Survivors in Five European
http://www.100md.com
《循环学杂志》
the Institute of Epidemiology (S.v.K., A.P., H.L.) and the Institute of Health Economics and Health Care Management (A.H.), GSF–National Research Center for Environment and Health, Neuherberg, Germany
Department of Physical Sciences (P.A., M.K.), University of Helsinki, Helsinki, Finland
Department of Occupational and Environmental Health (T.B., N.B.), Stockholm County Council, Stockholm, Sweden
Institute of Environmental Medicine (T.B., N.B.), Karolinska Institutet, Stockholm, Sweden
Department of Epidemiology (D.D., S.P., F.F.), ASLRME-Local Health Authority, Rome, Italy
IMIM-Municipal Institute for Medical Research (R.E., J.S.), Barcelona, Spain
the Department of Public Health and General Practise (J.P.), University of Kuopio, Kuopio, Finland
Unit of Environmental Epidemiology (T.L., J.P.), KTL-National Public Health Institute, Kuopio, Finland.
Abstract
Background— Ambient air pollution has been associated with increases in acute morbidity and mortality. The objective of this study was to evaluate the short-term effects of urban air pollution on cardiac hospital readmissions in survivors of myocardial infarction, a potentially susceptible subpopulation.
Methods and Results— In this European multicenter cohort study, 22 006 survivors of a first myocardial infarction were recruited in Augsburg, Germany; Barcelona, Spain; Helsinki, Finland; Rome, Italy; and Stockholm, Sweden, from 1992 to 2000. Hospital readmissions were recorded in 1992 to 2001. Ambient nitrogen dioxide, carbon monoxide, ozone, and mass of particles <10 μm (PM10) were measured. Particle number concentrations were estimated as a proxy for ultrafine particles. Short-term effects of air pollution on hospital readmissions for myocardial infarction, angina pectoris, and cardiac causes (myocardial infarction, angina pectoris, dysrhythmia, or heart failure) were studied in city-specific Poisson regression analyses with subsequent pooling. During follow-up, 6655 cardiac readmissions were observed. Cardiac readmissions increased in association with same-day concentrations of PM10 (rate ratio [RR] 1.021, 95% CI 1.004 to 1.039) per 10 μg/m3) and estimated particle number concentrations (RR 1.026 [95% CI 1.005 to 1.048] per 10 000 particles/cm3). Effects of similar strength were observed for carbon monoxide (RR 1.014 [95% CI 1.001 to 1.026] per 200 μg/m3 [0.172 ppm]), nitrogen dioxide (RR 1.032 [95% CI 1.013 to 1.051] per 8 μg/m3 [4.16 ppb]), and ozone (RR 1.026 [95% CI 1.001 to 1.051] per 15 μg/m3 [7.5 ppb]). Pooled effect estimates for angina pectoris and myocardial infarction readmissions were comparable.
Conclusions— The results suggest that ambient air pollution is associated with increased risk of hospital cardiac readmissions of myocardial infarction survivors in 5 European cities.
Key Words: air pollution epidemiology hospital readmission myocardial infarction multicenter study
Introduction
Ambient air pollution has been associated with increases in acute morbidity and mortality.1,2 Studies have suggested that patients with chronic obstructive pulmonary disease,3 congestive heart disease,4,5 or previous myocardial infarction or diabetes6 are at higher risk of death on days with high concentrations of air pollution. Patients who have survived a myocardial infarction are at increased risk for recurrent ischemic events7 and heart failure.8 The association of acute hospitalization with air pollution, to the best of our knowledge, has not been specifically assessed in predefined cohorts of vulnerable myocardial infarction survivors.
The Health Effects of Particles on Susceptible Subpopulations (HEAPSS) Study followed cohorts of first myocardial infarction survivors in 5 European cities: Augsburg (Germany), Barcelona (Spain), Helsinki (Finland), Rome (Italy), and Stockholm (Sweden). For the same period, measurements of combustion-related air pollutant concentrations included mass of particles <10 μm (PM10), nitrogen dioxide (NO2), and carbon monoxide (CO). Moreover, the particle number concentration (PNC) was estimated as a proxy for ultrafine particles. The secondary pollutant ozone was also considered. The objective of the present work was to assess the association of cardiac hospital readmission with urban air pollution in this potentially susceptible subpopulation.
Methods
Study Population
Incident myocardial infarction survivors were recruited in the 5 cities during 1992 to 2000 (Table 1). Data sources were population-based myocardial infarction registries in Augsburg (Cooperative Health Research in the Augsburg Region9) and Barcelona and administrative hospital discharge registers in Helsinki, Rome, and Stockholm. Enrollment was restricted to residents of the above cities aged 35 years or more (Augsburg, 35 to 74 years; Barcelona, 35 to 79 years) who had their first myocardial infarction (index event) during the recruitment period. Diagnostic criteria were established for Augsburg and Barcelona within the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) framework.10 Definite and possible myocardial infarctions and successfully resuscitated cardiac arrests according to the MONICA criteria were included.9 Using the registers’ databases, it was possible to exclude recurrent events. In Helsinki, Rome, and Stockholm, subjects with a primary myocardial infarction diagnosis in the hospital discharge record were included (International Classification of Diseases [ICD], 9th revision [ICD-9] code 410; ICD, 10th revision [ICD-10} codes I21 and I22) if they had no myocardial infarction during the 3 years preceding the index event and no diagnosis that indicated a previous myocardial infarction (ICD-9 412, ICD-10 I25.2). For all centers, the study was approved by the respective institutional review boards.
Follow-Up
Subsequent first cardiac rehospitalizations within the study area were recorded from the 29th day after the index event until the center-specific end of the follow-up period. Readmissions of interest were those with primary diagnoses of acute myocardial infarction (ICD-9: 410; ICD-10: I21, I22), angina pectoris (ICD-9: 411, 413; ICD-10: I20, I24), dysrhythmia (ICD-9: 427; ICD-10 I46.0, I46.9, I47–I49, R00.1, R00.8), and heart failure (ICD-9: 428; ICD-10: I50). Follow-up was performed by record linkage with the hospital admissions databases (Helsinki, Rome, and Stockholm), a telephone interview (Barcelona), or a mailed questionnaire sent to all cohort members or their relatives after the end of follow-up, followed by a search in the Central Clinic and in the Myocardial Infarction Registry (Augsburg). Vital status and place of residence at the end of the follow-up period were ascertained via comprehensive national (Helsinki and Stockholm) or regional (Barcelona) mortality registries or via municipal registries (Augsburg and Rome).
Air Pollution and Meteorology Data
Air pollution data from fixed monitors were collected for each city according to the APHEA (Air Pollution and Health, a European Approach) procedures11 during the center-specific recruitment and follow-up periods.12 Daily average concentrations of CO, NO2, and PM10 and the maximum 8-hour average of ozone were calculated for each monitor, when at least 75% of the observations were available. For each pollutant, daily city means were calculated by averaging over all monitors with a modified APHEA procedure.11
In Augsburg, total suspended particles were measured until 1999. PM10 was assessed thereafter with the same device. PM10 was derived for 1995 to 1999 by scaling down total suspended particles by a factor of 0.83, a locally validated procedure,13 which yielded 84% of PM10 data. For Barcelona, 20% of PM10 data (years 1996 and 1997) were estimated with a linear prediction model that used total suspended particles and black smoke, with adjustment for trend and season, based on periods for which all 3 pollutants were available. The R2 of the model was 0.55, and the correlation (observed versus predicted values) was 0.77.
PNC was assessed retrospectively for the HEAPSS project as an indicator of the number concentration of ultrafine particles (diameter <100 nm). In each center, condensation particle counters (CPC 3022A, TSI) were set up in 2001 to measure PNC for at least 1 year.12 On the basis of these data, PNC was fitted with other air pollutant and meteorological data in a regularized linear prediction model, and applied to the HEAPSS study period.14
Meteorological variables measured included daily mean temperature, dew point temperature, average barometric pressure at sea level, and relative humidity. Dates of temporary decreases in the local populations at risk due to vacations or holidays were recorded for each center.
Data Analysis
Poisson regression models were used to analyze the association of air pollution with hospital readmissions within the cohort. End points considered in the analyses were readmissions within the study area for (1) myocardial infarction, (2) angina pectoris, and (3) cardiac causes, a combined end point defined as the first readmission for acute myocardial infarction, angina pectoris, heart failure, or dysrhythmia. A person was considered to be at risk for each specific end point from the 29th day after the index event until the event of interest, death, migration out of the study area, loss to follow-up, or end of follow-up. Model specification was done for each city and outcome individually. All models included the natural logarithm of the number of persons at risk each day as an offset. The outcome was the daily number of readmissions. Generalized additive models were applied to allow for penalized spline functions of the continuous confounders with R software (the R Foundation for Statistical Computing, version 2.0.1) with the package "mgcv" (version 1.1-8).15 To ensure model selection was independent of the results, potential confounders were selected before the addition of air pollution concentration as an independent variable. The covariates tested were penalized regression splines of trend (to control for long-term trends, seasonality, and changes in the baseline rate) and meteorology (daily temperature, relative humidity, air pressure, and the difference between current day temperature and the mean temperature of the previous 3 days), as well as indicators of weekdays, vacation periods, or holidays. Trend and 1 temperature term had to remain in the model. Criteria for the inclusion of other covariates were probability value (P<0.1), and minimization of generalized cross validation score and the absolute value of the sum of partial autocorrelation coefficients.
Next, the lag structure of the effect of each pollutant was assessed with unconstrained distributed lag models that considered air pollution concentrations for the same day and the 3 previous days.16 City-specific results were pooled considering the variance covariance matrix of the estimates.17 Given the results, same-day air pollution concentrations were considered subsequently. City-specific single pollutant estimates were pooled with random effect models if the 2 test suggested heterogeneous city-specific estimates (P0.1).17 Otherwise, variance-weighted averages of city-specific regression coefficients were calculated.
The association of readmissions with ozone was analyzed taking into account only the warm months (April to September), because levels and correlations with other pollutants depend on season, and health effects of ozone have generally been associated with summertime levels. Pollutants were first considered separately, in single-pollutant models, and then jointly, in 2-pollutant models, with subsequent pooling. Two-pollutant models were not performed for PNC, CO, and NO2 because of high correlation between these pollutants. Sensitivity analyses were performed that compared the results of the final models with the estimates achieved when modifying the smooth functions (natural cubic splines or loess instead of penalized splines), setting more stringent convergence criteria,18 or changing confounders. For instance, temperature was replaced by apparent temperature19 or dew point temperature. The results were robust to these changes. Analyses that used a time-stratified case-crossover design yielded comparable results.
Rate ratios (RRs) and 95% CIs for PM10 are expressed for an increase of 10 μg/m3 to present results comparable to previous studies. Estimates for PNC, CO, NO2, and O3 are shown for an increase of 10 000/cm3, 0.2 mg/m3 (0.172 ppm), 8 μg/m3 (4.16 ppb), and 15 μg/m3 (7.5 ppb), respectively, equivalent to the mean of city-specific interquartile ranges multiplied by 0.5 (ratio of 10 μg/m3 and the study mean of PM10 interquartile ranges: 20 μg/m3).
Results
Study Population and Follow-Up
Overall, during the follow-up periods, 2321 hospital readmissions for myocardial infarction, 3541 for angina pectoris, and 6655 for cardiac events were observed among the 22 006 cohort members (Table 1).
Air Pollution Levels
The highest levels of PNC were estimated for Barcelona and Rome, the 2 most polluted study centers (Table 2). PM10 and CO concentrations were on average also higher in Rome and Barcelona than in Helsinki and Stockholm, with Augsburg showing medium concentrations. NO2 levels were comparably high in Augsburg, Barcelona, and Rome. Ozone in the warm months had a similar range in all 5 cities.
PNC was highly correlated with CO and NO2 in Augsburg and Rome, with CO in Barcelona, and with NO2 in Helsinki and Stockholm (Table 2). NO2 and CO were moderately correlated in Augsburg, Rome, and Stockholm. Other correlations were below 0.5 in all centers other than Augsburg. Ozone concentrations in the summer were negatively correlated with all other pollutants.
Association of Air Pollution and Hospital Readmissions
Distributed lag models suggested immediate effects of all air pollutants on cardiac readmissions (Figure 1). No pattern of a delayed or cumulative effect was seen. Therefore, same-day air pollution levels were considered in the subsequent analyses. Whereas models in Figure 1 included lag day 0 to 3 of the analyzed pollutant, Figure 2 shows single lag 0 pollutant estimates. Same-day levels of air pollutants were significantly associated with cardiac readmissions. Although city-specific results differed, the effects were homogeneous from a statistical point of view. Combined effect estimates for myocardial infarction readmissions were higher for PNC, PM10, and CO but had broader confidence intervals (Table 3). Associations of angina pectoris admissions with PM10 and CO were considerably lower than for cardiac or myocardial infarction admissions.
In 2-pollutant models, the effect of PNC, CO, and NO2 on cardiac readmissions remained stable when controlling for PM10 or ozone (results available on request). The association with PM10 was only slightly attenuated by the traffic-related air pollutants. The effect of ozone remained, or was even stronger, when we adjusted for the other pollutants.
Discussion
In this multicenter cohort study of myocardial infarction survivors, an increased risk of cardiac readmissions during days with elevated concentrations of several markers of urban air pollution (PNC, PM10, CO, NO2, and ozone) was observed. The associations were similar overall for angina pectoris and reinfarction, but most had broader confidence intervals because of smaller numbers of cases in the follow-up.
Previous studies on air pollution health effects on cardiovascular morbidity primarily defined cardiac admissions more broadly (ICD-9 390 to 429) than the present study, and none investigated the association with PNC. There is consistent evidence of an association between PM10 and cardiac hospital admissions in the general population,20–25 the elderly,18,21,24,26,27 and subgroups with comorbidities.26,27 Among these, 2 large multicenter studies, the US NMMAPS (National Morbidity, Mortality, and Air Pollution Study)18 and the European APHEA study,21 reported 1% and 0.5% increases in cardiac hospital admission, respectively, in the general population per 10 μg/m3 PM10 compared with an increase of 2.1% in the present study. For CO, increases in risk of 0.1% to 0.5% per 0.2 mg/m3 were found,22,23,28,29 which were lower than in the present work. For NO2 and cardiac admissions, only 1 study reported a higher effect size than the present study,22 whereas others reported similar or lower effects.20,23,24,29 Associations with ozone in the warm months were inconsistent, with significant negative estimates,20 nonsignificant estimates,23,24 or, as here, significant positive20,22,25 estimates.
Previously reported associations of air pollution (CO, NO2, and ozone) with angina pectoris admissions30 were weaker and statistically nonsignificant compared with the present study. Associations of myocardial infarction readmissions were positive but statistically nonsignificant for all air pollutants except ozone in the present study. The few previously conducted studies that examined associations of myocardial infarction with CO, NO2, O3, and particles have shown positive associations,30–35 positive but not significant associations,32,36–38 and occasionally negative38 associations.
The present study, designed to examine health effects of air pollution in a cohort of subjects potentially vulnerable with respect to recurrent coronary events, showed positive and generally larger effects than studies conducted in the general population. Therefore, myocardial infarction survivors might indeed be a susceptible subpopulation.
Various biological mechanisms have been proposed that associate particulate air pollution with cardiac end points. We hypothesize that these mechanisms are most likely related to changes in plaque vulnerability and blood rheology. It has been observed that rapidly translocated particles or PM components might induce endothelial dysfunction that contributes to plaque instability.2 Ambient particle inhalation may induce oxidative stress and inflammatory responses in the lung. Subsequent systemic inflammation may promote atherosclerosis progression, plaque instability, sudden arterial vasoconstriction, and a thrombosis-favoring blood rheology.2 Furthermore, the cardiac autonomic balance might be disturbed via lung neural reflex arcs or via components of PM that directly pass the pulmonary epithelium. The present results suggest the immediacy of such effects, at least among myocardial infarction survivors.
The present data suggest an association of traffic-related air pollutants (NO2, CO, and PNC) with cardiac readmissions. PNC, dominated by fresh combustion particles that originate primarily from mobile sources but that possibly include some particles generated by photochemical processes from gaseous pollutants, might be the culprit from a pathological point of view. NO2, which also originates mainly from motor vehicle emissions, has no known effect on the cardiovascular system. CO ambient concentrations, which are also mainly traffic-related, were low for causing acute poisoning, but associations have been shown at similar levels.28 Because CO and PNC were highly correlated, which is due to the common source but additionally induced in the present study by the estimation process, future studies with measurements of both parameters might disentangle the effects of these traffic-related pollutants. The fact that NO2, CO, and PNC were associated with the outcome in the present study and that the association was independent of PM10 and ozone, as shown in the 2-pollutant models, lends further support to evidence showing that traffic exposure might be linked to myocardial infarction onset 1 hour later.39 The effect of PM10, which is produced in combustion processes but which also comes from soil, tire fragmentation, and secondary particle formation, was only slightly attenuated in 2-pollutant models when the traffic markers were considered. The significant and independent association of ozone with hospital readmissions for angina pectoris and cardiac causes is a new finding that deserves consideration in future studies.
The retrospective PNC estimation is clearly a limitation of the present study. The first concern is the quality of the models. Models fitted PNC data relatively well, with R2 of 0.77, 0.80, 0.58, 0.84, and 0.81 for Augsburg, Barcelona, Helsinki, Rome, and Stockholm, respectively.14 The most important predictor variables were nitrogen oxides. Models were robust with respect to inclusion of irrelevant explanatory variables and were highly tolerant of missing data. The modeling process was fairly reliable, as seen when the estimated coefficients of half of the measurement period were applied to the other half. In these "split-halves" analyses, correlations of fitted and measured PNC were above 0.8 in all cities except Helsinki (r=0.72).
Second, we could not detect possible diverging time trends of PNC and the predictors used in the model, because the 1-year measurement period after April 2001 was used to estimate PNC levels as far back as 1992 (Barcelona). In Erfurt, Germany, time trends of PNC and other pollutants differed after the unification of East and West Germany. This was explained by changes in sources over time.40 In Western European cities, similar changes might have occurred a decade earlier, before the HEAPSS study period.
Additionally, there are issues on exposure assessment of PNC in time series studies that should be considered.41,42 Because ultrafine particles are mostly produced by local traffic, PNC is more heterogeneous spatially and depends more on distance from the roadway than other PM size fractions. The use of central monitors may limit the ability to detect a real PNC effect when data on average personal exposure for a population are lacking. Parallel measurements often showed a good overall correlation and suggested that background sites might well characterize average population exposure to ultrafine particles if the site is chosen carefully.12,43
For Barcelona, 2 years of PM10 data had been imputed with black smoke and total suspended particle data. Misclassification of PM10 might explain in part the finding of no association of cardiac readmissions with PM10 in this center. Sensitivity analyses that excluded the 2 years of imputed PM10 data showed higher but not statistically significant point estimates.
Although the centers differed with respect to myocardial infarction diagnosis in recruitment and in the follow-up, as described specifically in Methods, these diagnoses have been found to be highly comparable.44–46 Differences in susceptibility due to sex, age, and risk factor prevalence and use of drug therapies in these coronary patients could not be evaluated systematically because of the lack of specific data and the limited power of the study. Despite these differences, pooled estimates suggested an association between urban air pollution and hospital readmissions for cardiac events in myocardial infarction survivors in this multicenter study that used a population-based prospective approach and applied a common protocol. These results might be generalized to Europe, because an overall effect was observed despite different inclusion criteria and pollution levels across centers.
Conclusions
The results suggest that ambient air pollution is associated with increased risk of hospital cardiac readmissions for myocardial infarction survivors in 5 European cities. The study highlights the potential health effects of pollutants that originate from mobile sources in Europe. Likewise, an association with secondary pollutants, including ozone, was indicated.
The HEAPSS Study Group
Italy: F. Forastiere, D. D’Ippoliti, P. Michelozzi, C.A. Perucci, S. Picciotto, M. Stafoggia, G. Accetta, P. Compagnucci, G. Cattani, A. Marconi. Germany: A. Peters, S. von Klot, H. Lwel, A. Hrmann, C. Meisinger, J. Cyrys, C. Harmath, C. Greschik, K. Zetsche, U. Kaup, A. Schuler, D. Lukitsch, C. Winter, G. Zimmermann, P. Heilander, J. Schnelle-Kreis. Spain: J. Sunyer, J. Marrugat, R. Elosua, M. Gil, G. Sanz, L. Molina, V. Valle, J. Julvez, C. Rebato, M. Cabaero, L. Muoz. Finland: J. Pekkanen, T. Lanki, P. Tiittanen, V. Salomaa, T. Koskentalo, M. Kulmala, P. Aalto, P. Paatero, K. Hmeri. Sweden: F. Nyberg, N. Berglind, T Bellander, G. Pershagen, A. Bess, B. Sjvall, T. Jonsson, C. Johansson, M. Kster.
Acknowledgments
The study was funded by the European Union (QLK4-2000-00708). The funding source had no role in design or conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript. Finnish patient data were provided by the National Research and Development Centre for Welfare and Health, and vital status was ascertained using the mortality register at Statistics Finland. German data were provided by the GSF-Research Center of Environmental and Health funded by the German Ministry of Research and Science and the State of Bavaria. Swedish patient data were provided by Statistics Sweden and the Swedish National Board of Health and Welfare. Dr Peters was partially supported by the US Environmental Protection Agency STAR center grant R-827354.
References
Brunekreef B, Holgate ST. Air pollution and health. Lancet. 2002; 360: 1233–1242.
Brook RD, Franklin B, Cascio W, Hong Y, Howard G, Lipsett M, Luepker R, Mittleman M, Samet J, Smith SC Jr, Tager I. Air pollution and cardiovascular disease: a statement for healthcare professionals from the Expert Panel on Population and Prevention Science of the American Heart Association. Circulation. 2004; 109: 2655–2671.
Sunyer J, Schwartz J, Tobias A, Macfarlane D, Garcia J, Anto JM. Patients with chronic obstructive pulmonary disease are at increased risk of death associated with urban particle air pollution: a case-crossover analysis. Am J Epidemiol. 2000; 151: 50–56.
Goldberg MS, Burnett RT, Bailar JC III, Tamblyn R, Ernst P, Flegel K, Brook J, Bonvalot Y, Singh R, Valois MF, Vincent R. Identification of persons with cardiorespiratory conditions who are at risk of dying from the acute effects of ambient air particles. Environ Health Perspect. 2001; 109 (suppl 4): 487–494.
Kwon HJ, Cho SH, Nyberg F, Pershagen G. Effects of ambient air pollution on daily mortality in a cohort of patients with congestive heart failure. Epidemiology. 2001; 12: 413–419.
Bateson TF, Schwartz J. Who is sensitive to the effects of particulate air pollution on mortality A case-crossover analysis of effect modifiers. Epidemiology. 2004; 15: 143–149.
Jokhadar M, Jacobsen SJ, Reeder GS, Weston SA, Roger VL. Sudden death and recurrent ischemic events after myocardial infarction in the community. Am J Epidemiol. 2004; 159: 1040–1046.
Hellermann JP, Goraya TY, Jacobsen SJ, Weston SA, Reeder GS, Gersh BJ, Redfield MM, Rodeheffer RJ, Yawn BP, Roger VL. Incidence of heart failure after myocardial infarction: is it changing over time Am J Epidemiol. 2003; 157: 1101–1107.
Lwel H, Meisinger C, Heier M, Hrmann A, Kuch B, Gostomzyk J, Knig W. Sex specific trends of sudden cardiac death and acute myocardial infarction: results of the population based KORA/MONICA-Augsburg register 1985 to 1998. Dtsch Med Wochenschr. 2002; 127: 2311–2316.
Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, Rajakangas AM, Pajak A. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project. Registration procedures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents. Circulation. 1994; 90: 583–612.
Katsouyanni K, Schwartz J, Spix C, Touloumi G, Zmirou D, Zanobetti A, Wojtyniak B, Vonk JM, Tobias A, Ponka A, Medina S, Bacharova L, Anderson HR. Short term effects of air pollution on health: a European approach using epidemiologic time series data: the APHEA protocol. J Epidemiol Community Health. 1996; 50: S12–S18.
Aalto P, Hmeri K, Paatero P, Kulmala M, Bellander T, Berglind N, Bouso L, Castano-Vinyals G, Cattani G, Cyrys J, von Klot S, Lanki T, Marconi A, Nyberg F, Pekkanen J, Peters A, Sjvall B, Sunyer J, Zetzsche K, Forastiere F. Aerosol particle number concentration measurements in five European cities using TSI-3022 condensation particle counter over a three year period during HEAPSS. J Air Waste Manag Assoc. 2005; 55: 1064–1076.
The Council of the European Union. Council directive 1999/30/EC relating to limit values for sulphur dioxide, nitrogen dioxide and oxides of nitrogen, particulate matter and lead in ambient air. April 22, 1999.
Paatero P, Aalto P, Picciotto S, Bellander T, Castano-Vinyals G, Cattani G, Cyrys J, Kulmala M, Lanki T, Nyberg F, Pekkanen J, Peters A, Sunyer J, Forastiere F. Estimating time series of aerosol particle number concentrations in the five HEAPSS cities on the basis of measured air pollution and meteorological variables. Atmos Environ. 2005; 39: 2261–2273.
Wood SN, Augustin NH. GAMs with integrated model selection using penalized regression splines and applications to environmental modelling. Ecol Modell. 2002; 157: 157–177.
Schwartz J. The distributed lag between air pollution and daily deaths. Epidemiology. 2000; 11: 320–326.
Berkey CS, Laird NM. Nonlinear growth curve analysis: estimating the population parameters. Ann Hum Biol. 1986; 13: 111–128.
Schwartz J, Zanobetti A, Bateson T. Morbidity and mortality among elderly residents of cities with daily PM measurements. In: Revised Analyses of the National Morbidity, Mortality, and Air Pollution Study (NMMAPS), Part II. Boston, Mass: Health Effects Institute; 2003.
Kalkstein LS, Valimont KM. An evaluation of summer discomfort in the United States using a relative climatological index. Bull Am Meteorological Soc. 1986; 67: 842–848.
Wong CM, Atkinson RW, Anderson HR, Hedley AJ, Ma S, Chau PY, Lam TH. A tale of two cities: effects of air pollution on hospital admissions in Hong Kong and London compared. Environ Health Perspect. 2002; 110: 67–77.
Le Tertre A, Medina S, Samoli E, Forsberg B, Michelozzi P, Boumghar A, Vonk JM, Bellini A, Atkinson R, Ayres JG, Sunyer J, Schwartz J, Katsouyanni K. Short-term effects of particulate air pollution on cardiovascular diseases in eight European cities. In: Revised Analyses of Selected Time-Series Studies. Boston, Mass: Health Effects Institute; 2003.
Yang CY, Chen YS, Yang CH, Ho SC. Relationship between ambient air pollution and hospital admissions for cardiovascular diseases in Kaohsiung, Taiwan. J Toxicol Environ Health A. 2004; 67: 483–493.
Biggeri A, Baccini M, Bellini P, Terracini B. Meta-analysis of the Italian studies of short-term effects of air pollution (MISA), 1990–1999. Int J Occup Environ Health. 2005; 11: 107–122.
Morgan G, Corbett S, Wlodarczyk J. Air pollution and hospital admissions in Sydney, Australia, 1990 to 1994. Am J Public Health. 1998; 88: 1761–1766.
Chang CC, Tsai SS, Ho SC, Yang CY. Air pollution and hospital admissions for cardiovascular disease in Taipei, Taiwan. Environ Res. 2005; 98: 114–119.
Zanobetti A, Schwartz J, Gold D. Are there sensitive subgroups for the effects of airborne particles Environ Health Perspect. 2000; 108: 841–845.
Zanobetti A, Schwartz J. Cardiovascular damage by airborne particles: are diabetics more susceptible Epidemiology. 2002; 13: 588–592.
Burnett RT, Cakmak S, Brook JR, Krewski D. The role of particulate size and chemistry in the association between summertime ambient air pollution and hospitalization for cardiorespiratory diseases. Environ Health Perspect. 2000; 105: 614–620.
Moolgavkar SH. Air pollution and hospital admissions for diseases of the circulatory system in three U.S. metropolitan areas. J Air Waste Manag Assoc. 2000; 50: 1199–1206.
Poloniecki JD, Atkinson RW, de Leon AP, Anderson HR. Daily time series for cardiovascular hospital admissions and previous day’s air pollution in London, UK. Occup Environ Med. 1997; 54: 535–540.
Mann JK, Tager IB, Lurmann F, Segal M, Quesenberry CP Jr, Lugg MM, Shan J, Van Den Eeden SK. Air pollution and hospital admissions for ischemic heart disease in persons with congestive heart failure or arrhythmia. Environ Health Perspect. 2002; 110: 1247–1252.
D’Ippoliti D, Forastiere F, Ancona C, Agabiti N, Fusco D, Michelozzi P, Perucci CA. Air pollution and myocardial infarction in Rome: a case-crossover analysis. Epidemiology. 2003; 14: 528–535.
Zanobetti A, Schwartz J. The effect of particulate air pollution on emergency admissions for myocardial infarction: a multi-city case-crossover analysis. Environ Health Perspect. 2005; 113: 978–982.
Linn WS, Szlachcic Y, Gong H Jr, Kinney PL, Berhane KT. Air pollution and daily hospital admissions in metropolitan Los Angeles. Environ Health Perspect. 2000; 108: 427–434.
Peters A, Dockery DW, Muller JE, Mittleman MA. Increased particulate air pollution and the triggering of myocardial infarction. Circulation. 2001; 103: 2810–2815.
Sullivan J, Sheppard L, Schreuder A, Ishikawa N, Siscovick D, Kaufman J. Relation between short-term fine-particulate matter exposure and onset of myocardial infarction. Epidemiology. 2005; 16: 41–48.
Peters A, von Klot S, Heier M, Trentinaglia I, Cyrys J, Hrmann A, Hauptmann M, Wichmann H, Lwel H. Part I: Particulate Air Pollution, Personal Activities, and Onset of Myocardial Infarction in a Case-Crossover Study. Boston, Mass: Health Effects Institute; 2005. Report 124.
Ruidavets JB, Cournot M, Cassadou S, Giroux M, Meybeck M, Ferrieres J. Ozone air pollution is associated with acute myocardial infarction. Circulation. 2005; 111: 563–569.
Peters A, von Klot S, Heier M, Trentinaglia I, Hrmann A, Wichmann H-E, Lwel H. Exposure to traffic and the onset of myocardial infarction. N Engl J Med. 2004; 351: 1721–1730.
Kreyling WG, Tuch T, Peters A, Pitz M, Heinrich J, Stlzel M, Cyrys J, Heyder J, Wichmann H-E. Diverging long-term trends in ambient urban particle mass and number concentrations associated with emission changes caused by the German unification. Atmos Environ. 2003; 37: 3841–3848.
Sioutas C, Delfino RJ, Singh M. Exposure assessment for atmospheric ultrafine particles (UFPs) and implications in epidemiologic research. Environ Health Perspect. 2005; 113: 947–955.
Pekkanen J, Kulmala M. Exposure assessment of ultrafine particles in epidemiologic time-series studies. Scand J Work Environ Health. 2004; 30 (suppl 2): 9–18.
Forastiere F, Stafoggia M, Picciotto S, Bellander T, D’Ippoliti D, Lanki T, von Klot S, Nyberg F, Paatero P, Peters A, Pekkanen J, Sunyer J, Perucci CA. A case-crossover analysis of out-of-hospital coronary deaths and air pollution in Rome, Italy: the HEAPSS study. Am J Respir Crit Care Med. 2005;Jun 30. Epub ahead of print.
Ferrario M, Cesana G, Vanuzzo D, Pilotto L, Sega R, Chiodini P, Giampaoli S. Surveillance of ischaemic heart disease: results from the Italian MONICA populations. Int J Epidemiol. 2001; 30 (suppl 1): S23–S29.
Hammar N, Nerbrand C, Ahlmark G, Tibblin G, Tsipogianni A, Johansson S, Wilhelmsen L, Jacobsson S, Hansen O. Identification of cases of myocardial infarction: hospital discharge data and mortality data compared to myocardial infarction community registers. Int J Epidemiol. 1991; 20: 114–120.
Rapola JM, Virtamo J, Korhonen P, Haapakoski J, Hartman AM, Edwards BK, Heinonen OP. Validity of diagnoses of major coronary events in national registers of hospital diagnoses and deaths in Finland. Eur J Epidemiol. 1997; 13: 133–138.(Stephanie von Klot, PhD; )
Department of Physical Sciences (P.A., M.K.), University of Helsinki, Helsinki, Finland
Department of Occupational and Environmental Health (T.B., N.B.), Stockholm County Council, Stockholm, Sweden
Institute of Environmental Medicine (T.B., N.B.), Karolinska Institutet, Stockholm, Sweden
Department of Epidemiology (D.D., S.P., F.F.), ASLRME-Local Health Authority, Rome, Italy
IMIM-Municipal Institute for Medical Research (R.E., J.S.), Barcelona, Spain
the Department of Public Health and General Practise (J.P.), University of Kuopio, Kuopio, Finland
Unit of Environmental Epidemiology (T.L., J.P.), KTL-National Public Health Institute, Kuopio, Finland.
Abstract
Background— Ambient air pollution has been associated with increases in acute morbidity and mortality. The objective of this study was to evaluate the short-term effects of urban air pollution on cardiac hospital readmissions in survivors of myocardial infarction, a potentially susceptible subpopulation.
Methods and Results— In this European multicenter cohort study, 22 006 survivors of a first myocardial infarction were recruited in Augsburg, Germany; Barcelona, Spain; Helsinki, Finland; Rome, Italy; and Stockholm, Sweden, from 1992 to 2000. Hospital readmissions were recorded in 1992 to 2001. Ambient nitrogen dioxide, carbon monoxide, ozone, and mass of particles <10 μm (PM10) were measured. Particle number concentrations were estimated as a proxy for ultrafine particles. Short-term effects of air pollution on hospital readmissions for myocardial infarction, angina pectoris, and cardiac causes (myocardial infarction, angina pectoris, dysrhythmia, or heart failure) were studied in city-specific Poisson regression analyses with subsequent pooling. During follow-up, 6655 cardiac readmissions were observed. Cardiac readmissions increased in association with same-day concentrations of PM10 (rate ratio [RR] 1.021, 95% CI 1.004 to 1.039) per 10 μg/m3) and estimated particle number concentrations (RR 1.026 [95% CI 1.005 to 1.048] per 10 000 particles/cm3). Effects of similar strength were observed for carbon monoxide (RR 1.014 [95% CI 1.001 to 1.026] per 200 μg/m3 [0.172 ppm]), nitrogen dioxide (RR 1.032 [95% CI 1.013 to 1.051] per 8 μg/m3 [4.16 ppb]), and ozone (RR 1.026 [95% CI 1.001 to 1.051] per 15 μg/m3 [7.5 ppb]). Pooled effect estimates for angina pectoris and myocardial infarction readmissions were comparable.
Conclusions— The results suggest that ambient air pollution is associated with increased risk of hospital cardiac readmissions of myocardial infarction survivors in 5 European cities.
Key Words: air pollution epidemiology hospital readmission myocardial infarction multicenter study
Introduction
Ambient air pollution has been associated with increases in acute morbidity and mortality.1,2 Studies have suggested that patients with chronic obstructive pulmonary disease,3 congestive heart disease,4,5 or previous myocardial infarction or diabetes6 are at higher risk of death on days with high concentrations of air pollution. Patients who have survived a myocardial infarction are at increased risk for recurrent ischemic events7 and heart failure.8 The association of acute hospitalization with air pollution, to the best of our knowledge, has not been specifically assessed in predefined cohorts of vulnerable myocardial infarction survivors.
The Health Effects of Particles on Susceptible Subpopulations (HEAPSS) Study followed cohorts of first myocardial infarction survivors in 5 European cities: Augsburg (Germany), Barcelona (Spain), Helsinki (Finland), Rome (Italy), and Stockholm (Sweden). For the same period, measurements of combustion-related air pollutant concentrations included mass of particles <10 μm (PM10), nitrogen dioxide (NO2), and carbon monoxide (CO). Moreover, the particle number concentration (PNC) was estimated as a proxy for ultrafine particles. The secondary pollutant ozone was also considered. The objective of the present work was to assess the association of cardiac hospital readmission with urban air pollution in this potentially susceptible subpopulation.
Methods
Study Population
Incident myocardial infarction survivors were recruited in the 5 cities during 1992 to 2000 (Table 1). Data sources were population-based myocardial infarction registries in Augsburg (Cooperative Health Research in the Augsburg Region9) and Barcelona and administrative hospital discharge registers in Helsinki, Rome, and Stockholm. Enrollment was restricted to residents of the above cities aged 35 years or more (Augsburg, 35 to 74 years; Barcelona, 35 to 79 years) who had their first myocardial infarction (index event) during the recruitment period. Diagnostic criteria were established for Augsburg and Barcelona within the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) framework.10 Definite and possible myocardial infarctions and successfully resuscitated cardiac arrests according to the MONICA criteria were included.9 Using the registers’ databases, it was possible to exclude recurrent events. In Helsinki, Rome, and Stockholm, subjects with a primary myocardial infarction diagnosis in the hospital discharge record were included (International Classification of Diseases [ICD], 9th revision [ICD-9] code 410; ICD, 10th revision [ICD-10} codes I21 and I22) if they had no myocardial infarction during the 3 years preceding the index event and no diagnosis that indicated a previous myocardial infarction (ICD-9 412, ICD-10 I25.2). For all centers, the study was approved by the respective institutional review boards.
Follow-Up
Subsequent first cardiac rehospitalizations within the study area were recorded from the 29th day after the index event until the center-specific end of the follow-up period. Readmissions of interest were those with primary diagnoses of acute myocardial infarction (ICD-9: 410; ICD-10: I21, I22), angina pectoris (ICD-9: 411, 413; ICD-10: I20, I24), dysrhythmia (ICD-9: 427; ICD-10 I46.0, I46.9, I47–I49, R00.1, R00.8), and heart failure (ICD-9: 428; ICD-10: I50). Follow-up was performed by record linkage with the hospital admissions databases (Helsinki, Rome, and Stockholm), a telephone interview (Barcelona), or a mailed questionnaire sent to all cohort members or their relatives after the end of follow-up, followed by a search in the Central Clinic and in the Myocardial Infarction Registry (Augsburg). Vital status and place of residence at the end of the follow-up period were ascertained via comprehensive national (Helsinki and Stockholm) or regional (Barcelona) mortality registries or via municipal registries (Augsburg and Rome).
Air Pollution and Meteorology Data
Air pollution data from fixed monitors were collected for each city according to the APHEA (Air Pollution and Health, a European Approach) procedures11 during the center-specific recruitment and follow-up periods.12 Daily average concentrations of CO, NO2, and PM10 and the maximum 8-hour average of ozone were calculated for each monitor, when at least 75% of the observations were available. For each pollutant, daily city means were calculated by averaging over all monitors with a modified APHEA procedure.11
In Augsburg, total suspended particles were measured until 1999. PM10 was assessed thereafter with the same device. PM10 was derived for 1995 to 1999 by scaling down total suspended particles by a factor of 0.83, a locally validated procedure,13 which yielded 84% of PM10 data. For Barcelona, 20% of PM10 data (years 1996 and 1997) were estimated with a linear prediction model that used total suspended particles and black smoke, with adjustment for trend and season, based on periods for which all 3 pollutants were available. The R2 of the model was 0.55, and the correlation (observed versus predicted values) was 0.77.
PNC was assessed retrospectively for the HEAPSS project as an indicator of the number concentration of ultrafine particles (diameter <100 nm). In each center, condensation particle counters (CPC 3022A, TSI) were set up in 2001 to measure PNC for at least 1 year.12 On the basis of these data, PNC was fitted with other air pollutant and meteorological data in a regularized linear prediction model, and applied to the HEAPSS study period.14
Meteorological variables measured included daily mean temperature, dew point temperature, average barometric pressure at sea level, and relative humidity. Dates of temporary decreases in the local populations at risk due to vacations or holidays were recorded for each center.
Data Analysis
Poisson regression models were used to analyze the association of air pollution with hospital readmissions within the cohort. End points considered in the analyses were readmissions within the study area for (1) myocardial infarction, (2) angina pectoris, and (3) cardiac causes, a combined end point defined as the first readmission for acute myocardial infarction, angina pectoris, heart failure, or dysrhythmia. A person was considered to be at risk for each specific end point from the 29th day after the index event until the event of interest, death, migration out of the study area, loss to follow-up, or end of follow-up. Model specification was done for each city and outcome individually. All models included the natural logarithm of the number of persons at risk each day as an offset. The outcome was the daily number of readmissions. Generalized additive models were applied to allow for penalized spline functions of the continuous confounders with R software (the R Foundation for Statistical Computing, version 2.0.1) with the package "mgcv" (version 1.1-8).15 To ensure model selection was independent of the results, potential confounders were selected before the addition of air pollution concentration as an independent variable. The covariates tested were penalized regression splines of trend (to control for long-term trends, seasonality, and changes in the baseline rate) and meteorology (daily temperature, relative humidity, air pressure, and the difference between current day temperature and the mean temperature of the previous 3 days), as well as indicators of weekdays, vacation periods, or holidays. Trend and 1 temperature term had to remain in the model. Criteria for the inclusion of other covariates were probability value (P<0.1), and minimization of generalized cross validation score and the absolute value of the sum of partial autocorrelation coefficients.
Next, the lag structure of the effect of each pollutant was assessed with unconstrained distributed lag models that considered air pollution concentrations for the same day and the 3 previous days.16 City-specific results were pooled considering the variance covariance matrix of the estimates.17 Given the results, same-day air pollution concentrations were considered subsequently. City-specific single pollutant estimates were pooled with random effect models if the 2 test suggested heterogeneous city-specific estimates (P0.1).17 Otherwise, variance-weighted averages of city-specific regression coefficients were calculated.
The association of readmissions with ozone was analyzed taking into account only the warm months (April to September), because levels and correlations with other pollutants depend on season, and health effects of ozone have generally been associated with summertime levels. Pollutants were first considered separately, in single-pollutant models, and then jointly, in 2-pollutant models, with subsequent pooling. Two-pollutant models were not performed for PNC, CO, and NO2 because of high correlation between these pollutants. Sensitivity analyses were performed that compared the results of the final models with the estimates achieved when modifying the smooth functions (natural cubic splines or loess instead of penalized splines), setting more stringent convergence criteria,18 or changing confounders. For instance, temperature was replaced by apparent temperature19 or dew point temperature. The results were robust to these changes. Analyses that used a time-stratified case-crossover design yielded comparable results.
Rate ratios (RRs) and 95% CIs for PM10 are expressed for an increase of 10 μg/m3 to present results comparable to previous studies. Estimates for PNC, CO, NO2, and O3 are shown for an increase of 10 000/cm3, 0.2 mg/m3 (0.172 ppm), 8 μg/m3 (4.16 ppb), and 15 μg/m3 (7.5 ppb), respectively, equivalent to the mean of city-specific interquartile ranges multiplied by 0.5 (ratio of 10 μg/m3 and the study mean of PM10 interquartile ranges: 20 μg/m3).
Results
Study Population and Follow-Up
Overall, during the follow-up periods, 2321 hospital readmissions for myocardial infarction, 3541 for angina pectoris, and 6655 for cardiac events were observed among the 22 006 cohort members (Table 1).
Air Pollution Levels
The highest levels of PNC were estimated for Barcelona and Rome, the 2 most polluted study centers (Table 2). PM10 and CO concentrations were on average also higher in Rome and Barcelona than in Helsinki and Stockholm, with Augsburg showing medium concentrations. NO2 levels were comparably high in Augsburg, Barcelona, and Rome. Ozone in the warm months had a similar range in all 5 cities.
PNC was highly correlated with CO and NO2 in Augsburg and Rome, with CO in Barcelona, and with NO2 in Helsinki and Stockholm (Table 2). NO2 and CO were moderately correlated in Augsburg, Rome, and Stockholm. Other correlations were below 0.5 in all centers other than Augsburg. Ozone concentrations in the summer were negatively correlated with all other pollutants.
Association of Air Pollution and Hospital Readmissions
Distributed lag models suggested immediate effects of all air pollutants on cardiac readmissions (Figure 1). No pattern of a delayed or cumulative effect was seen. Therefore, same-day air pollution levels were considered in the subsequent analyses. Whereas models in Figure 1 included lag day 0 to 3 of the analyzed pollutant, Figure 2 shows single lag 0 pollutant estimates. Same-day levels of air pollutants were significantly associated with cardiac readmissions. Although city-specific results differed, the effects were homogeneous from a statistical point of view. Combined effect estimates for myocardial infarction readmissions were higher for PNC, PM10, and CO but had broader confidence intervals (Table 3). Associations of angina pectoris admissions with PM10 and CO were considerably lower than for cardiac or myocardial infarction admissions.
In 2-pollutant models, the effect of PNC, CO, and NO2 on cardiac readmissions remained stable when controlling for PM10 or ozone (results available on request). The association with PM10 was only slightly attenuated by the traffic-related air pollutants. The effect of ozone remained, or was even stronger, when we adjusted for the other pollutants.
Discussion
In this multicenter cohort study of myocardial infarction survivors, an increased risk of cardiac readmissions during days with elevated concentrations of several markers of urban air pollution (PNC, PM10, CO, NO2, and ozone) was observed. The associations were similar overall for angina pectoris and reinfarction, but most had broader confidence intervals because of smaller numbers of cases in the follow-up.
Previous studies on air pollution health effects on cardiovascular morbidity primarily defined cardiac admissions more broadly (ICD-9 390 to 429) than the present study, and none investigated the association with PNC. There is consistent evidence of an association between PM10 and cardiac hospital admissions in the general population,20–25 the elderly,18,21,24,26,27 and subgroups with comorbidities.26,27 Among these, 2 large multicenter studies, the US NMMAPS (National Morbidity, Mortality, and Air Pollution Study)18 and the European APHEA study,21 reported 1% and 0.5% increases in cardiac hospital admission, respectively, in the general population per 10 μg/m3 PM10 compared with an increase of 2.1% in the present study. For CO, increases in risk of 0.1% to 0.5% per 0.2 mg/m3 were found,22,23,28,29 which were lower than in the present work. For NO2 and cardiac admissions, only 1 study reported a higher effect size than the present study,22 whereas others reported similar or lower effects.20,23,24,29 Associations with ozone in the warm months were inconsistent, with significant negative estimates,20 nonsignificant estimates,23,24 or, as here, significant positive20,22,25 estimates.
Previously reported associations of air pollution (CO, NO2, and ozone) with angina pectoris admissions30 were weaker and statistically nonsignificant compared with the present study. Associations of myocardial infarction readmissions were positive but statistically nonsignificant for all air pollutants except ozone in the present study. The few previously conducted studies that examined associations of myocardial infarction with CO, NO2, O3, and particles have shown positive associations,30–35 positive but not significant associations,32,36–38 and occasionally negative38 associations.
The present study, designed to examine health effects of air pollution in a cohort of subjects potentially vulnerable with respect to recurrent coronary events, showed positive and generally larger effects than studies conducted in the general population. Therefore, myocardial infarction survivors might indeed be a susceptible subpopulation.
Various biological mechanisms have been proposed that associate particulate air pollution with cardiac end points. We hypothesize that these mechanisms are most likely related to changes in plaque vulnerability and blood rheology. It has been observed that rapidly translocated particles or PM components might induce endothelial dysfunction that contributes to plaque instability.2 Ambient particle inhalation may induce oxidative stress and inflammatory responses in the lung. Subsequent systemic inflammation may promote atherosclerosis progression, plaque instability, sudden arterial vasoconstriction, and a thrombosis-favoring blood rheology.2 Furthermore, the cardiac autonomic balance might be disturbed via lung neural reflex arcs or via components of PM that directly pass the pulmonary epithelium. The present results suggest the immediacy of such effects, at least among myocardial infarction survivors.
The present data suggest an association of traffic-related air pollutants (NO2, CO, and PNC) with cardiac readmissions. PNC, dominated by fresh combustion particles that originate primarily from mobile sources but that possibly include some particles generated by photochemical processes from gaseous pollutants, might be the culprit from a pathological point of view. NO2, which also originates mainly from motor vehicle emissions, has no known effect on the cardiovascular system. CO ambient concentrations, which are also mainly traffic-related, were low for causing acute poisoning, but associations have been shown at similar levels.28 Because CO and PNC were highly correlated, which is due to the common source but additionally induced in the present study by the estimation process, future studies with measurements of both parameters might disentangle the effects of these traffic-related pollutants. The fact that NO2, CO, and PNC were associated with the outcome in the present study and that the association was independent of PM10 and ozone, as shown in the 2-pollutant models, lends further support to evidence showing that traffic exposure might be linked to myocardial infarction onset 1 hour later.39 The effect of PM10, which is produced in combustion processes but which also comes from soil, tire fragmentation, and secondary particle formation, was only slightly attenuated in 2-pollutant models when the traffic markers were considered. The significant and independent association of ozone with hospital readmissions for angina pectoris and cardiac causes is a new finding that deserves consideration in future studies.
The retrospective PNC estimation is clearly a limitation of the present study. The first concern is the quality of the models. Models fitted PNC data relatively well, with R2 of 0.77, 0.80, 0.58, 0.84, and 0.81 for Augsburg, Barcelona, Helsinki, Rome, and Stockholm, respectively.14 The most important predictor variables were nitrogen oxides. Models were robust with respect to inclusion of irrelevant explanatory variables and were highly tolerant of missing data. The modeling process was fairly reliable, as seen when the estimated coefficients of half of the measurement period were applied to the other half. In these "split-halves" analyses, correlations of fitted and measured PNC were above 0.8 in all cities except Helsinki (r=0.72).
Second, we could not detect possible diverging time trends of PNC and the predictors used in the model, because the 1-year measurement period after April 2001 was used to estimate PNC levels as far back as 1992 (Barcelona). In Erfurt, Germany, time trends of PNC and other pollutants differed after the unification of East and West Germany. This was explained by changes in sources over time.40 In Western European cities, similar changes might have occurred a decade earlier, before the HEAPSS study period.
Additionally, there are issues on exposure assessment of PNC in time series studies that should be considered.41,42 Because ultrafine particles are mostly produced by local traffic, PNC is more heterogeneous spatially and depends more on distance from the roadway than other PM size fractions. The use of central monitors may limit the ability to detect a real PNC effect when data on average personal exposure for a population are lacking. Parallel measurements often showed a good overall correlation and suggested that background sites might well characterize average population exposure to ultrafine particles if the site is chosen carefully.12,43
For Barcelona, 2 years of PM10 data had been imputed with black smoke and total suspended particle data. Misclassification of PM10 might explain in part the finding of no association of cardiac readmissions with PM10 in this center. Sensitivity analyses that excluded the 2 years of imputed PM10 data showed higher but not statistically significant point estimates.
Although the centers differed with respect to myocardial infarction diagnosis in recruitment and in the follow-up, as described specifically in Methods, these diagnoses have been found to be highly comparable.44–46 Differences in susceptibility due to sex, age, and risk factor prevalence and use of drug therapies in these coronary patients could not be evaluated systematically because of the lack of specific data and the limited power of the study. Despite these differences, pooled estimates suggested an association between urban air pollution and hospital readmissions for cardiac events in myocardial infarction survivors in this multicenter study that used a population-based prospective approach and applied a common protocol. These results might be generalized to Europe, because an overall effect was observed despite different inclusion criteria and pollution levels across centers.
Conclusions
The results suggest that ambient air pollution is associated with increased risk of hospital cardiac readmissions for myocardial infarction survivors in 5 European cities. The study highlights the potential health effects of pollutants that originate from mobile sources in Europe. Likewise, an association with secondary pollutants, including ozone, was indicated.
The HEAPSS Study Group
Italy: F. Forastiere, D. D’Ippoliti, P. Michelozzi, C.A. Perucci, S. Picciotto, M. Stafoggia, G. Accetta, P. Compagnucci, G. Cattani, A. Marconi. Germany: A. Peters, S. von Klot, H. Lwel, A. Hrmann, C. Meisinger, J. Cyrys, C. Harmath, C. Greschik, K. Zetsche, U. Kaup, A. Schuler, D. Lukitsch, C. Winter, G. Zimmermann, P. Heilander, J. Schnelle-Kreis. Spain: J. Sunyer, J. Marrugat, R. Elosua, M. Gil, G. Sanz, L. Molina, V. Valle, J. Julvez, C. Rebato, M. Cabaero, L. Muoz. Finland: J. Pekkanen, T. Lanki, P. Tiittanen, V. Salomaa, T. Koskentalo, M. Kulmala, P. Aalto, P. Paatero, K. Hmeri. Sweden: F. Nyberg, N. Berglind, T Bellander, G. Pershagen, A. Bess, B. Sjvall, T. Jonsson, C. Johansson, M. Kster.
Acknowledgments
The study was funded by the European Union (QLK4-2000-00708). The funding source had no role in design or conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript. Finnish patient data were provided by the National Research and Development Centre for Welfare and Health, and vital status was ascertained using the mortality register at Statistics Finland. German data were provided by the GSF-Research Center of Environmental and Health funded by the German Ministry of Research and Science and the State of Bavaria. Swedish patient data were provided by Statistics Sweden and the Swedish National Board of Health and Welfare. Dr Peters was partially supported by the US Environmental Protection Agency STAR center grant R-827354.
References
Brunekreef B, Holgate ST. Air pollution and health. Lancet. 2002; 360: 1233–1242.
Brook RD, Franklin B, Cascio W, Hong Y, Howard G, Lipsett M, Luepker R, Mittleman M, Samet J, Smith SC Jr, Tager I. Air pollution and cardiovascular disease: a statement for healthcare professionals from the Expert Panel on Population and Prevention Science of the American Heart Association. Circulation. 2004; 109: 2655–2671.
Sunyer J, Schwartz J, Tobias A, Macfarlane D, Garcia J, Anto JM. Patients with chronic obstructive pulmonary disease are at increased risk of death associated with urban particle air pollution: a case-crossover analysis. Am J Epidemiol. 2000; 151: 50–56.
Goldberg MS, Burnett RT, Bailar JC III, Tamblyn R, Ernst P, Flegel K, Brook J, Bonvalot Y, Singh R, Valois MF, Vincent R. Identification of persons with cardiorespiratory conditions who are at risk of dying from the acute effects of ambient air particles. Environ Health Perspect. 2001; 109 (suppl 4): 487–494.
Kwon HJ, Cho SH, Nyberg F, Pershagen G. Effects of ambient air pollution on daily mortality in a cohort of patients with congestive heart failure. Epidemiology. 2001; 12: 413–419.
Bateson TF, Schwartz J. Who is sensitive to the effects of particulate air pollution on mortality A case-crossover analysis of effect modifiers. Epidemiology. 2004; 15: 143–149.
Jokhadar M, Jacobsen SJ, Reeder GS, Weston SA, Roger VL. Sudden death and recurrent ischemic events after myocardial infarction in the community. Am J Epidemiol. 2004; 159: 1040–1046.
Hellermann JP, Goraya TY, Jacobsen SJ, Weston SA, Reeder GS, Gersh BJ, Redfield MM, Rodeheffer RJ, Yawn BP, Roger VL. Incidence of heart failure after myocardial infarction: is it changing over time Am J Epidemiol. 2003; 157: 1101–1107.
Lwel H, Meisinger C, Heier M, Hrmann A, Kuch B, Gostomzyk J, Knig W. Sex specific trends of sudden cardiac death and acute myocardial infarction: results of the population based KORA/MONICA-Augsburg register 1985 to 1998. Dtsch Med Wochenschr. 2002; 127: 2311–2316.
Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, Rajakangas AM, Pajak A. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project. Registration procedures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents. Circulation. 1994; 90: 583–612.
Katsouyanni K, Schwartz J, Spix C, Touloumi G, Zmirou D, Zanobetti A, Wojtyniak B, Vonk JM, Tobias A, Ponka A, Medina S, Bacharova L, Anderson HR. Short term effects of air pollution on health: a European approach using epidemiologic time series data: the APHEA protocol. J Epidemiol Community Health. 1996; 50: S12–S18.
Aalto P, Hmeri K, Paatero P, Kulmala M, Bellander T, Berglind N, Bouso L, Castano-Vinyals G, Cattani G, Cyrys J, von Klot S, Lanki T, Marconi A, Nyberg F, Pekkanen J, Peters A, Sjvall B, Sunyer J, Zetzsche K, Forastiere F. Aerosol particle number concentration measurements in five European cities using TSI-3022 condensation particle counter over a three year period during HEAPSS. J Air Waste Manag Assoc. 2005; 55: 1064–1076.
The Council of the European Union. Council directive 1999/30/EC relating to limit values for sulphur dioxide, nitrogen dioxide and oxides of nitrogen, particulate matter and lead in ambient air. April 22, 1999.
Paatero P, Aalto P, Picciotto S, Bellander T, Castano-Vinyals G, Cattani G, Cyrys J, Kulmala M, Lanki T, Nyberg F, Pekkanen J, Peters A, Sunyer J, Forastiere F. Estimating time series of aerosol particle number concentrations in the five HEAPSS cities on the basis of measured air pollution and meteorological variables. Atmos Environ. 2005; 39: 2261–2273.
Wood SN, Augustin NH. GAMs with integrated model selection using penalized regression splines and applications to environmental modelling. Ecol Modell. 2002; 157: 157–177.
Schwartz J. The distributed lag between air pollution and daily deaths. Epidemiology. 2000; 11: 320–326.
Berkey CS, Laird NM. Nonlinear growth curve analysis: estimating the population parameters. Ann Hum Biol. 1986; 13: 111–128.
Schwartz J, Zanobetti A, Bateson T. Morbidity and mortality among elderly residents of cities with daily PM measurements. In: Revised Analyses of the National Morbidity, Mortality, and Air Pollution Study (NMMAPS), Part II. Boston, Mass: Health Effects Institute; 2003.
Kalkstein LS, Valimont KM. An evaluation of summer discomfort in the United States using a relative climatological index. Bull Am Meteorological Soc. 1986; 67: 842–848.
Wong CM, Atkinson RW, Anderson HR, Hedley AJ, Ma S, Chau PY, Lam TH. A tale of two cities: effects of air pollution on hospital admissions in Hong Kong and London compared. Environ Health Perspect. 2002; 110: 67–77.
Le Tertre A, Medina S, Samoli E, Forsberg B, Michelozzi P, Boumghar A, Vonk JM, Bellini A, Atkinson R, Ayres JG, Sunyer J, Schwartz J, Katsouyanni K. Short-term effects of particulate air pollution on cardiovascular diseases in eight European cities. In: Revised Analyses of Selected Time-Series Studies. Boston, Mass: Health Effects Institute; 2003.
Yang CY, Chen YS, Yang CH, Ho SC. Relationship between ambient air pollution and hospital admissions for cardiovascular diseases in Kaohsiung, Taiwan. J Toxicol Environ Health A. 2004; 67: 483–493.
Biggeri A, Baccini M, Bellini P, Terracini B. Meta-analysis of the Italian studies of short-term effects of air pollution (MISA), 1990–1999. Int J Occup Environ Health. 2005; 11: 107–122.
Morgan G, Corbett S, Wlodarczyk J. Air pollution and hospital admissions in Sydney, Australia, 1990 to 1994. Am J Public Health. 1998; 88: 1761–1766.
Chang CC, Tsai SS, Ho SC, Yang CY. Air pollution and hospital admissions for cardiovascular disease in Taipei, Taiwan. Environ Res. 2005; 98: 114–119.
Zanobetti A, Schwartz J, Gold D. Are there sensitive subgroups for the effects of airborne particles Environ Health Perspect. 2000; 108: 841–845.
Zanobetti A, Schwartz J. Cardiovascular damage by airborne particles: are diabetics more susceptible Epidemiology. 2002; 13: 588–592.
Burnett RT, Cakmak S, Brook JR, Krewski D. The role of particulate size and chemistry in the association between summertime ambient air pollution and hospitalization for cardiorespiratory diseases. Environ Health Perspect. 2000; 105: 614–620.
Moolgavkar SH. Air pollution and hospital admissions for diseases of the circulatory system in three U.S. metropolitan areas. J Air Waste Manag Assoc. 2000; 50: 1199–1206.
Poloniecki JD, Atkinson RW, de Leon AP, Anderson HR. Daily time series for cardiovascular hospital admissions and previous day’s air pollution in London, UK. Occup Environ Med. 1997; 54: 535–540.
Mann JK, Tager IB, Lurmann F, Segal M, Quesenberry CP Jr, Lugg MM, Shan J, Van Den Eeden SK. Air pollution and hospital admissions for ischemic heart disease in persons with congestive heart failure or arrhythmia. Environ Health Perspect. 2002; 110: 1247–1252.
D’Ippoliti D, Forastiere F, Ancona C, Agabiti N, Fusco D, Michelozzi P, Perucci CA. Air pollution and myocardial infarction in Rome: a case-crossover analysis. Epidemiology. 2003; 14: 528–535.
Zanobetti A, Schwartz J. The effect of particulate air pollution on emergency admissions for myocardial infarction: a multi-city case-crossover analysis. Environ Health Perspect. 2005; 113: 978–982.
Linn WS, Szlachcic Y, Gong H Jr, Kinney PL, Berhane KT. Air pollution and daily hospital admissions in metropolitan Los Angeles. Environ Health Perspect. 2000; 108: 427–434.
Peters A, Dockery DW, Muller JE, Mittleman MA. Increased particulate air pollution and the triggering of myocardial infarction. Circulation. 2001; 103: 2810–2815.
Sullivan J, Sheppard L, Schreuder A, Ishikawa N, Siscovick D, Kaufman J. Relation between short-term fine-particulate matter exposure and onset of myocardial infarction. Epidemiology. 2005; 16: 41–48.
Peters A, von Klot S, Heier M, Trentinaglia I, Cyrys J, Hrmann A, Hauptmann M, Wichmann H, Lwel H. Part I: Particulate Air Pollution, Personal Activities, and Onset of Myocardial Infarction in a Case-Crossover Study. Boston, Mass: Health Effects Institute; 2005. Report 124.
Ruidavets JB, Cournot M, Cassadou S, Giroux M, Meybeck M, Ferrieres J. Ozone air pollution is associated with acute myocardial infarction. Circulation. 2005; 111: 563–569.
Peters A, von Klot S, Heier M, Trentinaglia I, Hrmann A, Wichmann H-E, Lwel H. Exposure to traffic and the onset of myocardial infarction. N Engl J Med. 2004; 351: 1721–1730.
Kreyling WG, Tuch T, Peters A, Pitz M, Heinrich J, Stlzel M, Cyrys J, Heyder J, Wichmann H-E. Diverging long-term trends in ambient urban particle mass and number concentrations associated with emission changes caused by the German unification. Atmos Environ. 2003; 37: 3841–3848.
Sioutas C, Delfino RJ, Singh M. Exposure assessment for atmospheric ultrafine particles (UFPs) and implications in epidemiologic research. Environ Health Perspect. 2005; 113: 947–955.
Pekkanen J, Kulmala M. Exposure assessment of ultrafine particles in epidemiologic time-series studies. Scand J Work Environ Health. 2004; 30 (suppl 2): 9–18.
Forastiere F, Stafoggia M, Picciotto S, Bellander T, D’Ippoliti D, Lanki T, von Klot S, Nyberg F, Paatero P, Peters A, Pekkanen J, Sunyer J, Perucci CA. A case-crossover analysis of out-of-hospital coronary deaths and air pollution in Rome, Italy: the HEAPSS study. Am J Respir Crit Care Med. 2005;Jun 30. Epub ahead of print.
Ferrario M, Cesana G, Vanuzzo D, Pilotto L, Sega R, Chiodini P, Giampaoli S. Surveillance of ischaemic heart disease: results from the Italian MONICA populations. Int J Epidemiol. 2001; 30 (suppl 1): S23–S29.
Hammar N, Nerbrand C, Ahlmark G, Tibblin G, Tsipogianni A, Johansson S, Wilhelmsen L, Jacobsson S, Hansen O. Identification of cases of myocardial infarction: hospital discharge data and mortality data compared to myocardial infarction community registers. Int J Epidemiol. 1991; 20: 114–120.
Rapola JM, Virtamo J, Korhonen P, Haapakoski J, Hartman AM, Edwards BK, Heinonen OP. Validity of diagnoses of major coronary events in national registers of hospital diagnoses and deaths in Finland. Eur J Epidemiol. 1997; 13: 133–138.(Stephanie von Klot, PhD; )