Author + information
- Received November 9, 2007
- Revision received January 23, 2008
- Accepted January 27, 2008
- Published online May 27, 2008.
- Christos Pitsavos, MD, PhD⁎,
- Stavros A. Kavouras, PhD†,
- Demosthenes B. Panagiotakos, PhD†,⁎ (, )
- Sophia Arapi, MD⁎,
- Costas A. Anastasiou, MS†,
- Spyros Zombolos, MD⁎,
- Petros Stravopodis, MD⁎,
- Yannis Mantas, MD⁎,
- Yannis Kogias, MD⁎,
- Antonis Antonoulas, MD⁎,
- Christodoulos Stefanadis, MD, PhD⁎,
- GREECS Study Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Demosthenes B. Panagiotakos, 46 Paleon Polemiston Street, 16674, Glyfada, Greece.
Objectives We sought to evaluate the association between physical activity levels and the clinical outcome at presentation, as well as the 30-day prognosis of hospitalized patients with acute coronary syndromes (ACS).
Background Regular physical activity has been associated with decreased risk of coronary heart disease. However, less is known about the effects of life-long physical activity on ACS prognosis.
Methods From October 2003 to September 2004, a sample of 6 hospitals located in urban and rural Greek regions were selected, and almost all of their ACS patients were enrolled into the study (2,172 patients were included in the study; 76% men and 24% women). Logistic regression models were applied to evaluate the effect of physical activity status (as assessed using the International Physical Activity Questionnaire) on in-hospital mortality and the 30-day outcome of cardiovascular events (death or rehospitalization due to cardiovascular disease).
Results An inverse association was observed between the level of physical activity and troponin I levels at presentation (p = 0.01). Moreover, after taking into account various potential confounders, physical activity was associated with a 0.56-fold (95% confidence interval [CI] 0.32 to 0.90) lower odds of in-hospital mortality and a 0.80-fold (95% CI 0.50 to 0.99) lower odds of cardiovascular events within the first month after discharge.
Conclusion In conclusion, physical activity is associated with a reduced severity of ACS, reduced in-hospital mortality rates, and improved short-term prognosis.
Ischemic heart disease and cerebrovascular diseases are the leading causes of death worldwide (1). Regular physical activity has been associated with a decreased risk of coronary heart disease (CHD) (2–5) as well as decreased overall mortality (6). Regarding the type of exercise that is most potent in reducing the risk of CHD, most cohort studies have revealed that total energy expended in physical activity is the most significant predictor of reduced risk (3–5), and some studies indicate that exercise intensity may be of significant value, irrespective of the duration of the activity (2).
The mechanisms by which physical activity confers cardioprotective effects are not well understood. In animal models, endurance exercise has been found to provide protection against ischemia-reperfusion injury, possibly through improved myocardial antioxidant capacity (7,8), prevention of ischemia induced apoptosis via attenuation of calpain activation (9,10), and reduction in Ca2+-handling proteins degradation (9). In humans, myocardial ischemia pre-conditioning, induced by repeated bouts of exercise, has been shown to induce a decrease in mean maximal ST-segment depression and ischemia duration on subsequent exercise (11).
Acute coronary syndromes (ACS) are characterized by a high case fatality, and a large proportion of survivors depend on nursing and clinical care. The immediate time following hospital discharge may be critical for long-term prognosis. For example, among survivors from acute myocardial infarction (MI), stroke or transient ischemic attack is more likely to happen during the first month after hospitalization (12). To our knowledge, no reports have examined long-term lifestyle factors, such as physical activity, on the prognosis of ACS. Additionally, the existing data on the relationship between physical activity and overall cardiovascular disease (CVD) have failed to provide a quantitative relationship between the severity of these diseases and physical activity status.
The aim of the present study was to examine the actual effects of lifetime physical activity in the incidence and the severity of acute ischemic syndromes, using biochemical markers (cardiac troponin I release and MB fraction of total creatine phosphokinase [CK-MB]) as indexes of myocardial necrosis. We also evaluated whether a physically active lifestyle might affect short-term prognosis (i.e., 30-day recurrent events).
Population of the study
From October 2003 to September 2004 almost all consecutive patients (n = 2,172) with a diagnosis of ACS (acute MI or unstable angina [UA]) that were hospitalized in the cardiology clinics or the emergency units of 6 major general hospitals in Greece were enrolled into the study. Of these patients, 1,649 (76%) were men and 523 (24%) were women. With the exception of Athens—where there are several other hospitals—the other hospitals cover the whole population of their regions, including urban and rural areas. The participation rate was 98%. A posteriori power analysis showed that the number of enrolled participants was adequate to evaluate 2-sided differences between patient groups, as defined by the discharge diagnosis, and the investigated parameters >20% (±5%), achieving statistical power >80% at 5% probability level (p value).
At entry, as well as during hospitalization (at 6 to 9 h and again at 12 to 24 h if earlier samples were negative and the clinical index of suspicion was high), measurements of troponin I and CK-MB were obtained from all patients to detect evidence of myocardial cell death. In addition, a 12-lead electrocardiogram was performed and clinical symptoms were evaluated in all patients by a cardiologist of the study. Acute MI was defined by at least 2 of the following features (13): 1) electrocardiographic changes indicative of ischemia (ST-segment elevation or depression); 2) compatible clinical symptoms; and 3) specific diagnostic biomarker elevations (troponin I >0.4 ng/ml and CK-MB >8.8 ng/ml). Unstable angina was defined by the occurrence of 1 or more angina episodes, at rest, within the preceding 48 h, corresponding to class III of the Braunwald classification (14). All hospitalized patients received the optimal therapy according to their clinical status.
This study was approved by the Medical Research Ethics Committee of our Institution and was carried out in accordance with the Declaration of Helsinki (1989).
With the exception of 15 patients who died during the first 24 h of their admission, detailed information was recorded from all other patients using standard procedures presented in the following text.
Physical activity ascertainment
We used a translated version of the validated “International Physical Activity Questionnaire” (IPAQ), suitable for assessing population levels of self-reported physical activity (15). The short version of IPAQ (9 items) that we used provided information on weekly time spent walking, in vigorous and moderate intensity, and in sedentary activity. Participants were instructed to refer to all domains of physical activity. Both continuous and categoric indicators were assessed from IPAQ. The continuous indicator was expressed as metabolic equivalent of the task (MET)-minutes per week derived from walking and vigorous- and moderate-intensity activities, and the categoric analysis grouped the subjects in the following 3 levels that were developed based on a key concept in current public health guidelines for physical activity (16): inactive, minimally active, and health enhancing physical activity (HEPA) active, a highly active category, as previously described (17). Participants were instructed to report only episodes of activities of at least 10 min, because this is the minimum required to achieve health benefit (16). Values below 10 were recorded to 0. The majority of the participants (85%) reported that they had the same habits during the last 5 years.
Sociodemographic, clinical, and biochemical characteristics
A detailed medical history was recorded, including previous hospitalization for CVD (i.e. CHD, stroke, or other CVD) and presence and treatment of hypertension, hypercholesterolemia, renal failure, and diabetes. Sociodemographic characteristics included age, gender, and years of school. Usual dietary intake over the year preceding enrollment was assessed by a semiquantitative food frequency questionnaire, including several foods and beverages commonly consumed in Greece. We asked all participants to report the daily or weekly average intake of several food items that they consumed (during the last year). The frequency of consumption was quantified approximately in terms of the number of times a month this food was consumed. The level of adherence to the Mediterranean diet was assessed through a special diet score (18) which incorporates the inherent characteristics of this dietary pattern. Higher values of this score indicate better adherence to this traditional diet. Current smokers were defined as those who smoked at least 1 cigarette per day or had stopped cigarette smoking during the past 12 months. Former smokers were defined as those who had stopped smoking more than 1 year before. The rest were defined as never smokers or rare smokers. Height and weight were measured to the nearest 0.5 cm and 100 g, respectively. Body mass index (BMI) was then calculated as weight (in kg) divided by height (in m) squared.
Continuous variables are presented as mean values (±SD), and categoric variables are presented as absolute and relative (%) frequencies. Associations between continuous variables and group of patients by discharge diagnosis were evaluated through analysis of variance, after controlling for normality and equality of variances (homoscedacity). Because of multiple comparisons, the Bonferroni rule was applied to correct for the inflation of type I error. Associations between categoric variables and group of patients by discharge diagnosis were tested by the use of the chi-square test, without the correction of continuity. Correlations between continuous variables were evaluated by the use of the Pearson correlation coefficient for the normally distributed, and by the use of the Spearman rho coefficient for the ordinal or skewed variables. The association between physical activity and: 1) the presence of acute MI compared with unstable angina; 2) in-hospital mortality; and 3) 30-day outcome (death or rehospitalization due to CVD) was evaluated by multiple logistic regression analysis, after controlling for sociodemographic, clinical, and biochemical characteristics of the patients. Appropriate tests for goodness-of-fit (i.e., deviance residuals) were applied in all models. Confounding was tested by entering and removing each variable from the model and evaluating its effect (using the Wald test) on the point-estimate (i.e., odds ratio) of the main factor of interest (i.e., physical activity). First-order interactions between physical activity and various characteristics of the patients were also evaluated in all models. A probability value of 5% was considered to be statistically significant. All statistical calculations were performed with SPSS version 12.0 software (SPSS Inc., Chicago, Illinois).
From October 2003 to September 2004, 2,172 patients with discharge diagnosis of ACS were enrolled into the study (1,649 men and 523 women). According to the discharge diagnosis, 764 patients (35%) were diagnosed as having UA, 699 patients (32%) as having non–Q-wave MI, and 709 patients (33%) as having Q-wave MI. Of the enrolled patients, 1,199 (55%) were defined as physically inactive, 742 (34%) as minimally active, and 73 (3%) as HEPA active; 158 were not classified owing to missing information. The various sociodemographic, lifestyle, and clinical characteristics of the patients by physical activity status are presented in Table 1. More physically active patients were of younger age, male, less likely to have a history of CHD, hypertension, or diabetes, and more educated and had better dietary habits, but they were also more likely to be smokers. Moreover, physically active patients had lower troponin I levels. No association was observed between electrocardiographic findings at presentation and physical activity status (p = 0.09).
The in-hospital mortality rate in those who had complete physical activity data was 3.2%. Crude analysis showed that physical activity status was inversely associated with the in-hospital mortality: The death rates in physical inactive, minimally active, and HEPA active were 4.2%, 2.0%, and 0%, respectively (p = 0.01). Age- and gender-adjusted analysis showed that physically active patients had as much as 0.53-fold (95% confidence interval [CI] 0.30 to 0.93) less likelihood of dying during hospitalization compared with physically inactive patients.
During the first 30 days after hospitalization, precise information about vital status or rehospitalization due to CHD problems were retrieved from 1,683 out of 2,172 patients (75%). The information was retrieved through personal communication by the physicians of the study, the patients, or their relatives. The number of CVD events among those who had complete information about their physical activity status was 147 (9.4%), and 37 of those events were fatal. Of those who were physically inactive, 10.6% had an event during the first 30 days after hospitalization. Of those who were minimally and HEPA active, 7.1% and 6.3%, respectively, had an event during follow-up (p = 0.08). Age- and gender-adjusted analysis showed that physically active patients had 0.80-fold (95% CI 0.63 to 1.19) less likelihood of having a recurrent event compared with physically inactive patients.
However, many factors may influence these relationships. Multiadjusted data analysis revealed that minimal or HEPA activity was associated with 0.56-fold (95% CI 0.32 to 0.90) lower odds of in-hospital mortality compared with inactivity, after controlling for age, gender, previous history of CHD, presence of hypertension, diabetes, or hypercholesterolemia and respective medication, previous hospitalization, smoking and dietary habits, body mass index, education status, and clinical status (MI or UA) of the patients. No significant interaction was observed between gender and physical activity status on the likelihood of dying during hospitalization (p = 0.73). Moreover, no differences were observed between physical activity status (i.e., minimally active or HEPA active) on in-hospital mortality. We also found that physical activity was associated with a lower risk of death or recurrent events. Particularly, minimally or HEPA active patients had 0.80-fold (95% CI 0.50 to 0.99) lower odds of having cardiovascular events within the first month after discharge, compared with inactive patients, after controlling for age, gender, previous history of CHD, presence of hypertension, diabetes, hypercholesterolemia, and respective medication, smoking and dietary habits, body mass index, and discharge clinical status (MI or UA) of the patients. As before, no significant interaction was observed between gender and physical activity status on the likelihood of having an event during the 30-day follow-up period (p = 0.39), and no significant differences were observed between levels of physical activity status (i.e., minimally active or HEPA active) and the respective outcome. Furthermore, study sites did not confound the results, because no significant interaction was observed between physical activity status and region on the development of CVD events during the first 30 days after an ACS (p = 0.10). Table 2 illustrates the association between several factors and the likelihood of having a CVD event (fatal or not) during the first days after hospitalization.
Physically active lifestyle has been associated with various health benefits and decreased risk of diseases, such as CVD. In the present study of 2,172 patients with ACS we found that physical activity was negatively associated with the severity of these syndromes. Furthermore, we found that physical activity was also associated with a reduction of in-hospital mortality rates. It appears that a physically active lifestyle may confer protection during the first month after an ACS, in terms of both mortality and rehospitalization due to a recurrent event.
Several extended cohort studies (2–5) have revealed a strong negative association between physical activity levels and the risk of CHD. The present study extends this observation by establishing a quantitative relationship between the severity of ACS, as well as its short-term prognosis, and physical activity. Cardiac troponin and the CK-MB are considered to be sensitive biomarkers of myocardial damage in the clinical setting of acute ischemia (13). In the present study, CK-MB at entry was not actually associated with the physical activity status, whereas troponin I was found to be highly related to physical activity status. The CK-MB level is a less specific marker of myocardial damage than troponin I, which has nearly absolute tissue specificity as well as high sensitivity, thereby reflecting even microscopic zones of myocardial necrosis (13).
Although the cardioprotective effect of regular physical activity is well established, the mechanism by which exercise exerts its cardioprotective effects, especially regarding acute myocardial syndromes, is not quite understood. Endurance exercise training has been associated with a substantial improvement in myocardial contractile performance (7) and infarct size (19) during experimental ischemia-reperfusion in animal models. Exercise may have the potential to act by pre-conditioning the heart to ischemia, because exercise itself is a means of myocardial ischemia (11). Ischemic pre-conditioning has been found to exert its protective effect in a 2-phased manner: the classic or early pre-conditioning protection that lasts for about 3 h after the exercise bout, and the late pre-conditioning, or “second window of protection,” that begins approximately a day later, may last up to 72 h, and is possibly related to increased expression and synthesis of cytoprotective proteins, such as stress proteins and antioxidants (20). The short-lived early pre-conditioning protection cannot easily find any clinical application in ameliorating the degree of myocardial damage during an ACS, or in designing primary and secondary prevention strategies, because there is, as yet, no way to predict the occurrence of an acute ischemic syndrome. However, the second window of protection may at least in part explain the reduced severity of myocardial damage during an ACS that was evident in the physically active subjects of the present study. It is not known whether the cardioprotective effects of physical activity status may have more lasting effects to favorably impact short-term prognosis during recovery from an ACS. Alternatively, the reduced severity of myocardial necrosis observed in the physically active individuals may be the main factor for a better prognosis at the recovery period. It is logical to assume that revascularization therapies applied during the treatment of the episode may be more effective in a less damaged myocardium, such as in the case of physically active individuals. However, most of the ACS do not occur at points of critical coronary narrowing and, therefore, ischemic pre-conditioning might be unlikely to occur with exercise in the absence of critical lesion.
Nevertheless, it should be pointed out that this is a cross-sectional study that cannot provide causal relationships, but only state hypotheses for future research. Misreporting of physical activity status may confound, at least in part, the strength of the observed relationships. Although we took into account dietary and smoking habits, as well as many risk factors (i.e., BMI, hypertension, diabetes, and hypercholesterolemia and use of respective medications), the influence of their confounding effects cannot be entirely excluded. Moreover, another limitation is the over-fitting of the multiadjusted statistical model (i.e., including many variables); however, this was done to take into account as many confounders as possible. Interestingly, the physically active individuals had a more positive profile regarding the aforementioned parameters compared with inactive subjects. Thus, the observed reduced severity of ACS and the better prognosis may be attributed to an overall healthier lifestyle in the physically active group and not only to physical activity levels. Alternatively, physical activity, in combination with other healthy lifestyle factors, may reduce infarct size and myocardial necrosis during an ACS, leading to a more favorable prognosis of the disease. Although habitual physical activity reduces the risk of CHDs, vigorous exercise can increase the risk of sudden cardiac death in susceptible persons (21). For this reason some of the acute cardiac events that occurred during vigorous exercise may have been omitted because those subjects might have died before hospital presentation. However, in the present study we were not interested in the acute effect of vigorous exercise, but in the influence of habitual physical activity. Finally, the nonparticipation rate was 25% (mainly because of wrong contact information); however, no differences were observed regarding physical activity status between those who were found and those who were not found on follow-up (p = 0.09).
The present study revealed that a rather simple, inexpensive lifestyle modification, such as increased levels of physical activity, may result in less severe ACS, with decreased in-hospital mortality and a better short-term prognosis. This may be of clinical relevance in the setting of population-based primary and secondary prevention efforts. In combination with the use of other possible cardioprotective methods, incorporating exercise into one's way of life may be an effective approach for achieving supplementary cardioprotection and reducing both morbidity and mortality in the setting of naturally occurring ACS.
The authors present and thank the field investigators of the GREECS study—George Giannopoulos, Theodoros Gialernios, Constandina Massoura, George Papanagnou, Antonis Karanasios, Lambros Rizos, Michalis Mparmparoussis, George Kassimatis, Skevos Sideris, and Nick Daskalopoulos—for their support in the clinical evaluation and Alexandros Chalamandaris for database management.
- Abbreviations and Acronyms
- acute coronary syndromes
- body mass index
- coronary heart disease
- confidence interval
- MB fraction of total creatine phosphokinase
- cardiovascular disease
- health-enhancing physical activity
- International Physical Activity Questionnaire
- myocardial infarction
- unstable angina
- Received November 9, 2007.
- Revision received January 23, 2008.
- Accepted January 27, 2008.
- American College of Cardiology Foundation
- Sesso H.D.,
- Paffenbarger R.S. Jr..,
- Lee I.M.
- Lee I.M.,
- Sesso H.D.,
- Paffenbarger R.S. Jr.
- Lee I.M.,
- Paffenbarger R.S. Jr.
- Demirel H.A.,
- Powers S.K.,
- Zergeroglu M.A.,
- et al.
- Powers S.K.,
- Demirel H.A.,
- Vincent H.K.,
- et al.
- Tzivoni D.,
- Maybaum S.
- Tanne D.,
- Goldbourt U.,
- Zion M.,
- Reicher-Reiss H.,
- Kaplinsky E.,
- Behar S.,
- SPRINT Study Group
- ↵(2000) Myocardial infarction redefined—a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction. J Am Coll Cardiol 36:959–969.
- Braunwald E.
- Panagiotakos D.B.,
- Pitsavos C.,
- Chrysohoou C.,
- Skoumas J.,
- Stefanadis C.
- Thompson P.D.,
- Franklin B.A.,
- Balady G.J.,
- et al.