440 E X E R C I S E A N D T H E H E A R T of clinical evidence of coronary disease at the onset, known heart disease.135,136 Historically, however, have been examined regularly since 1949. Coronary the strength of exercise capacity as a predictor of disease mortality was subsequently found to be risk has not been given its appropriate due.137,138 higher in cohorts with indices consistent with a This is largely because clinicians have tended to sedentary lifestyle. However, physical inactivity focus on ECG changes in terms of risk stratification, did not have the predictive power of three other even though many studies, particularly in recent primary risk factors (smoking, hypertension, years, have demonstrated that exercise capacity increased lipids). Kannel et al133 reanalyzed the may be the strongest predictor of risk among clin- Framingham data for the effects of physical activ- ical and exercise test variables. This topic was ity on overall mortality and cardiovascular disease touched on in Chapter 8. In the following sections, mortality. The effect of being sedentary on mor- studies published about men and women referred tality was rather modest compared with the other for exercise testing for clinical reasons and risk factors, but persisted when these other factors followed-up for both cardiac morbidity and all- were taken into account. A low correlation was cause mortality are reviewed. The major studies noted between physical activity level and the major in this area are outlined in Table 13-9. risk factors. A growing number of studies have been pub- The relationship between self-selected leisure- lished in which physical fitness, determined by time physical activity to first major coronary disease standardized exercise testing, was analyzed among events and overall mortality was studied in 12,138 large samples of men and women who have been middle-aged men participating in the Multiple Risk followed for the incidence of CHD morbidity and Factor Intervention Trial.134 Total leisure-time mortality for up to 20 years.50,111,139-145 Each of these physical activity over the preceding year was quan- studies demonstrated that higher levels of fitness tified in mean minutes per day at baseline by ques- were associated with lower rates of CHD or all- cause tionnaire, with subjects classified into tertiles (low, mortality. Importantly, these associations appear to moderate, and high leisure-time physical activity). be independent of other CHD risk factors. Moreover, During 7 years of follow-up, moderate leisure-time the low levels of fitness in these studies did not physical activity was associated with a 37% reduc- appear to be associated with subclinical disease. tion in fatal CHD events and sudden deaths, and a 30% reduction in total deaths compared with low Early studies in addressing this topic measured leisure-time physical activity (P < 0.01). Mortality various risk factors for cardiovascular disease and rates with high leisure-time physical activity were studied their association with exercise tolerance. similar to those for moderate leisure-time physi- Leon et al146 demonstrated that, in healthy men, cal activity; however, combined fatal and nonfatal the combination of resting measurements and major CHD events were 20% lower with high as questionnaire responses related to health corre- compared with low leisure-time physical activity. lated highly with treadmill time using a multivari- Leisure-time physical activity had a modest inverse ate equation. One hundred and seventy-five relation to CHD and overall mortality in middle- apparently healthy men completed questionnaires aged men at high risk for CHD. The relationship of about habitual physical activity, smoking, alcoholic leisure-time physical activity to mortality was inves- beverage consumption, and sleep habits. Body tigated in 3043 American railroad workers who mass index, heart rate and blood pressure at rest were followed-up for 20 years.131 The Minnesota and during submaximal exercise, frequency of pre- Leisure Time Physical Activity Questionnaire was mature ventricular beats, handgrip strength, and used. After adjusting for age and other risk factors, serum cholesterol were measured. These charac- the risk estimate for CHD deaths was 1.39 in men teristics were correlated with the duration of exer- who were sedentary compared with the most active cise using the Bruce protocol. Univariate analysis men who expended more than 2000 kcal per week. indicated that treadmill performance was signifi- cantly and positively correlated with leisure-time Prospective Studies Relating activity and reports of sweating and dyspnea occur- Exercise Capacity to Cardiac ring regularly during such physical activity. Events or Mortality Performance was negatively correlated with age, body mass index, resting heart rate, cigarette It has been recognized for several decades that smoking, and consumption of caffeine-containing exercise capacity determined from an exercise test beverages. A total correlation of 0.75 was found is a strong predictor of mortality in patients with between treadmill performance and 11 of the vari- ables previously mentioned, and it increased to 0.81 by adding heart rate during submaximal exercise.
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 441 TA B L E 1 3 – 9 . Epidemiologic studies of exercise capacity assessed by exercise testing as it relates to mortality Investigator Year Mean Assessment Subjects Conclusions follow-up Peters 1983 4.8 years Bicycle ergometry 2779 healthy Adjusted RR of 2.2 for low Lie 1985 7.0 years Bicycle ergometry middle-aged men exercise capacity if other Bruce TM risk factors also present Ekelund 1988 8.5 years 149 middle-aged 8 years elite athletes; Higher quintiles of 5.1 years 2014 healthy men physical fitness associated 2 years with decreased coronary 4276 males disease and mortality Blair 1989 USAFSAM TM 10,224 males, RR of 2.7 for cardiovascular Blair 1995 USAFSAM TM 3120 females death in men with low 9777 men exercise capacity Snader 1997 Bruce TM 3400 patients Physical fitness inversely referred for related to all-cause mortality clinical reasons Improvement in fitness between two tests (5 years Erikssen 1998 Bicycle ergometry 1756 men apart) reduces mortality 44% Exercise capacity Myers 2002 6.2 years Ramp TM 6213 patients stronger predictor of death referred for than Tl-201 defects; RR 4.4 Mora 2003 20 years Bruce TM clinical reasons for poor versus good Bruce TM exercise capacity Gulati 2003 8 years Bruce TM 2994 Graded, inverse relation Balady 2004 18 years asymptomatic between change in fitness in women fitness on two exercise tests (7 to 10 years apart) and 5721 mortality, irrespective of asymptomatic initial fitness level women Exercise capacity strongest 1431 men, 1612 predictor of mortality among women without clinical, risk factor and CHD at baseline exercise test variables; each 1 MET increase conferred 12% improved survival Exercise capacity stronger predictor of mortality than ischemic criteria; Each 1 MET increase conferred 20% improved survival 17% better survival per MET achieved In high risk men, each 1 MET increase conferred 13% reduction in CHD risk CHD, coronary heart disease; RR, relative risk; Tl-201, Thallium SPECT imaging; TM, treadmill; USAFSAM, United States Air Force School of Aerospace Medicine. A prospective study by Peters et al147 suggested smoking, above-median systolic blood pressure, that poor physical work capacity, as measured or a combination of these. by bicycle ergometry in apparently healthy Los Angeles County workers, was related to subsequent An impressive body of data has been published MI. This was one of the first follow-up studies to from the Aerobics Center in Dallas, using treadmill measure exercise capacity directly, rather than to performance to associate physical fitness with estimate activity level. An adjusted relative risk of health. In a cross-sectional study of 753 men, tread- 2.2 was found only in men with certain other risk mill performance was found to be inversely related factors present, namely above-median cholesterol, to body weight, percentage of body fat, lipids, glu- cose, and systolic blood pressure.148 In a subsequent
442 E X E R C I S E A N D T H E H E A R T longitudinal study, men who were treadmill The Lipid Research Clinics (LRC) Mortality tested both before and after an exercise program Follow-up Study also used a baseline treadmill test were analyzed to determine if their performance as its estimate of physical fitness in its study of 3106 had improved. Those men who reached the upper men followed-up for an average of 8.5 years.140 quartile of improved aerobic fitness exhibited This study reported a relative risk of cardiovascular decreases in lipids, diastolic blood pressure, death of 2.7 (95%; confidence interval 1.4 to 5.1, serum glucose, uric acid, and weight. Regular P = 0.003) for the less physically fit men, none of exercise resulting in increased aerobic capacity whom had evidence of cardiac disease at entry. An was associated with decreased risk factors. In interesting study by Lie et al150 in Norway prospec- addition, they found that, in a study of 420 men tively followed groups of men with no known car- who were divided into groups of former athletes diac disease at entry for 7 years. One group of 149 and nonathletes, prior athleticism had no signifi- middle-aged elite athletes (Nordic skiers) had an cant effect on cardiovascular risk factors, physical extremely low incidence of death (1%) after 7 years. fitness, and exercise habits.149 The other group was composed of 2014 middle- aged healthy men who were further subdivided into This group also subsequently studied 10,244 physical fitness quartiles based on their perfor- men and 3120 women (99% of whom were white) mance on a bicycle ergometry test on entry. This who all were able to achieve 85% of age-predicted quartile grouping was also consistent with the sub- heart rate.139 They divided physical fitness levels ject’s level of leisure activity determined by ques- into quintiles by estimating VO2 max from a base- tionnaire. The most fit quartile in this group had a line treadmill test. During an average follow-up of 7-year mortality rate similar to that of the Nordic 8 years, they demonstrated by multivariate analysis skiers, and a highly significant difference was found adjusted for age, other known cardiac risk factors, in survival between each of the quartiles. Physical and length of follow-up that all-cause mortality fitness and leisure-time physical activity in this was inversely related to physical fitness. The rela- study were found to be significantly inversely related tive risk in men and women in the highest quintile to mortality. of fitness was 0.29 and 0.22, respectively. Simply changing from the lowest quintile of fitness to the The LRC recently addressed this issue in second-lowest quintile cut the mortality in half women.144 Among 2994 asymptomatic women for women (relative risk, 0.52) and by 60% in men followed-up over a period of 20 years, low exercise (relative risk, 0.40). These results are presented in capacity, low heart rate recovery, and the inability Table 13-10. This study was quite important at the to achieve target heart rate were independent pre- time of publication, being one of the only analyses dictors of mortality. After age-adjustment, there to prospectively study fitness in women, to include was a 20% reduction in mortality for every MET such a large sample size, and to document that achieved on the treadmill. Women who were below gradations in physical fitness led to significant the median values for both exercise capacity and reductions in mortality. heart rate recovery had a 3.5-fold increased risk TA B L E 1 3 – 1 0 . Rates and relative risks of death* among 10,244 men and 3120 women in an 8-year follow- up, by gradients of physical fitness Men Women Quintiles of Number of Death per Relative Number of Deaths per Relative fitness† deaths 100,000 risk of death‡ deaths 100,000 risk of man-years woman-years death‡ 1 (low) 1.00 18 2 75 64.0 0.40 11 39.5 1.00 3 40 25.5 0.42 20.5 0.52 4 47 27.1 0.34 6 12.2 0.31 5 (high) 43 21.7 0.29 4 0.15 35 18.6 4 6.5 0.22 8.5 *Age-adjusted. †Quintiles of fitness determined by maximal exercise testing. ‡p for trend <0.05. From Blair SN, Kohl HW, Paffenbarger RS, et al: Physical fitness and all-cause mortality. JAMA 1989;262:2395-2401.
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 443 of cardiovascular death. Importantly, although those with cardiovascular disease, the least fit exercise capacity was a powerful predictor of risk, individuals had more than four times the risk of ST-segment depression was not associated with risk all-cause mortality compared with the most fit. in these women. Similar observations were made Importantly, an individual’s fitness level was a from the St. James Women Take Heart Project, stronger predictor of mortality than established risk in which 5721 asymptomatic women (mean age factors such as smoking, high blood pressure, high 52 years) underwent exercise testing and were fol- cholesterol, and diabetes. Over the last few years, lowed for a mean of 8.4 years.111 After adjustment other cohorts, such as those from the Cleveland for Framingham risk score, there was a decrease in Clinic143 and the Mayo Clinic141,142 have docu- mortality of 17% for every MET increase in exercise mented the importance of exercise capacity as a capacity. These two studies are important in that predictor of mortality among clinically referred pop- they included large cohorts of women, as opposed to ulations. These clinically based studies confirm the the vast majority of previous studies performed in observations of Blair et al,139 Framingham,48 and men. They suggest that exercise capacity may be an the LRC Trial140 among asymptomatic popula- even more powerful predictor of risk (17% to 20% tions, underscoring the fact that fitness level has risk reduction per MET) than that observed in men. a strong influence on the incidence of cardiovas- In addition, they confirm that the prognostic value cular and all-cause morbidity and mortality. of the exercise test in women is associated with fitness-related variables (exercise capacity and heart Studies Assessing Both Physical rate recovery) more so than ischemic responses. Activity Pattern and Fitness Level More recently, this issue has been addressed in Is physical fitness or activity necessary to achieve clinical populations, for example, patients referred protection from CAD? This controversy is still for exercise testing for clinical reasons.50,141-143,145 debated in the literature151-154 and some have In a recent study performed among U.S. Veterans, argued that it is as much a rhetorical question as it 6213 men underwent maximal exercise testing for is a scientific one. VO2 max correlates with the level clinical reasons and were followed for a mean of of physical activity, and for epidemiologic studies, 6.2 years.50 The subjects were classified into five estimated METs have been used as a reasonable categories by gradients of fitness. After adjustment expression of cardiovascular fitness.50,139,141-143,145 for age, the researchers observed that the largest Fitness level achieved on an exercise test is no doubt gains in terms of mortality were achieved between a more objective and reliable method of assessing a the lowest fitness group and the next lowest fitness person’s cardiovascular health than physical group. Figure 13-1 illustrates the age-adjusted activity as reported by questionnaire. There has relative risks associated with the different cate- gories of fitness. Among both normal subjects and 5 (2.95– Normal group Disease group 4.5 6.83) (3.29– 5.16) 4 Relative risk 3.5 (2.40– 1.0–5.9 METs 1.0–4.9 METs3.73) 6.0–7.9 METs 5.0–6.4 METs3 8.0–9.9 METs 6.5–8.2 METs(1.54– 10.0–12.9 METs2.53.75)(1.73– 8.3–10.6 METs2.76) ≥13.0 METs 2 ≥10.7 METs(1.12–(1.35– 2.75) 2.19) ■ FIGURE 13-1 (0.68– Age-adjusted relative risks of mortal- 1.5 2.22) ity by quintiles of exercise capacity among normal subjects and patients 1 with cardiovascular disease. (From Myers et al: Exercise capacity and 0.5 mortality among men referred for exercise testing. New Engl J Med 0 234 5 2002:346;793-801). 1 Quintiles of exercise capacity
444 E X E R C I S E A N D T H E H E A R T been some recent debate as to whether daily phys- Expressing the data by age-adjusted quartiles, exer- ical activity patterns largely determine one’s fit- cise capacity was a stronger predictor of mortality ness level, and therefore health risk, or whether than activity pattern (hazard ratio 0.56, P < fitness level predicts mortality independently from 0.001). In a multivariate analysis, including clini- activity pattern. cal, risk factor, exercise test data, and activity pat- terns, exercise capacity and energy expenditure Despite the abundance of reports related to fit- from adulthood recreational activity were the only ness or activity and health outcomes, few studies significant predictors of mortality (hazard ratio for have addressed both fitness and physical activity exercise capacity = 0.62, P < 0.001 and hazard in the same population multivariately with other ratio for activity = 0.72, P = 0.002 per quartile); clinical and risk factor data. The available studies these two variables were stronger predictors of generally suggest that fitness level more strongly pre- mortality than established risk factors such as dicts outcomes than physical activity patterns.151-153 smoking, hypertension, obesity, and diabetes. Age- There may be several reasons why this is the case. adjusted hazard ratios for exercise capacity per First, the quantification of physical fitness is more quartile were 1.0, 0.59, 0.46, and 0.28 (P < 0.001); objective than physical activity. Fitness is generally age-adjusted hazard ratios for physical activity were determined directly from symptom or sign-limited 1.0, 0.63, 0.42, and 0.38 (P < 0.001). A 1000-kcal per exercise testing, whereas activity level is dependent week increase in activity was approximately similar upon subject recollection, the judiciousness with to a 1-MET increase in fitness; both conferred a which subjects respond, and other limitations asso- 20% mortality benefit. This analysis suggests that ciated with questionnaires. In addition, the strength although exercise capacity determined from exer- of exercise capacity relative to other clinical and cise testing and energy expenditure from weekly exercise data in stratifying risk, while only recently activity are both strong predictors of mortality, appreciated,137,138 is increasingly being recognized exercise capacity outperforms activity pattern in in the literature among both healthy111,139,140,144 predicting risk. Importantly, however, both are and clinically referred subjects.50,141-143,145 stronger predictors of risk than other clinical and exercise test variables. In the Copenhagen Male Study, the joint effects of fitness and leisure-time activity were analyzed Meta-Analyses of Physical Activity in 4999 men aged 40 to 59 years.155 The men were and Health Outcomes classified according to level of physical fitness by estimated maximal oxygen uptake, physical activity The large volume of data published over several was determined by interview, and mortality and decades has provided the opportunity to combine CHD incidence was recorded over a 17-year period. results in order to better define the association In sedentary men, fitness was not a predictor of between physical activity and various outcomes. future risk of CHD. In moderate or highly active In 1987 Powell et al157 performed an extensive men, however, fitness was a strong predictor of risk. review of 43 such studies and concluded that an The least fit (two least fit quintiles) physically active inverse relationship between physical activity and men had a higher CHD mortality rate; after adjust- the incidence of CHD was observed in over two ment for age, social class, and smoking in a multi- thirds. Moreover, the relationship was strongest in ple logistic regression equation, there was a 67% those studies that best measured physical activity higher mortality risk for the lower fitness groups. (Fig. 13-2). These investigators concluded that: (1) being fit provides no protection against CHD—nor all-cause In a recent meta-analysis, Williams152 com- mortality—in sedentary men and (2) unfit, seden- pared the dose-response relationships between tary men have a higher risk of CHD than unfit, leisure-time physical activity and fitness from pub- active men, that is, those performing light physical lished reports and their association with cardio- activity for at least 4 hours per week. vascular disease endpoints. The analysis included a remarkable 1,325,000 person-years follow-up. In an effort to further clarify this issue, we stud- The results of the Williams meta-analysis151 are ied 6213 men referred for exercise testing between summarized in Figure 13-3. Relative risks are 1987 and 2000, and a subgroup of 842 who under- plotted as a function of the cumulative percent- went an assessment of adulthood activity pat- ages of the samples when ranked from least fit or terns.153 The predictive power of exercise capacity active to most fit or active. In combining study and activity patterns, along with clinical and exer- cise test data, were assessed for all-cause mortal- ity during a mean follow-up period of 5.5 ± 2 years.
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 445 ■ FIGURE 13-2 Box plot illustrating the relative risk of physical inactivity according to the scientific quality of the available studies. (From Powell KE, Thompson PD, Caspersen CJ, Kendrick JS: Physical activity and the incidence of coronary heart disease. Am Rev Public Health 1987;8:253–287). results, a weighted average of the relative risks relative risk reduction is greater for fitness than from the physical activity and fitness cohorts was physical activity. Williams151 interpreted these computed at every 5th percentile between 5% and findings to mean that formulating activity recom- 100%. As illustrated in Figure 13-3, the risks of mendations on the basis of fitness studies may CHD decrease linearly with increasing percentiles inappropriately demote the status of cardiorespi- of physical activity. This is contrasted by the ratory fitness as a risk factor, while exaggerating fitness cohorts, in which a sharp drop in risk the public health benefits of moderate amounts of occurs before the 25th percentile of the fitness physical activity. distribution. This suggests that the largest bene- fits in terms of CHD morbidity occur by the most This recent debate aside, it should be noted unfit becoming moderately fit, confirming the that dozens of studies over the last 4 decades have observations in the Blair et al,139 the Veterans reported that both higher physical fitness levels Administration,50 and other106,107 studies. Perhaps and greater amounts of physical activity have more importantly, the precipitous drop in risk an inverse association with the incidence of before the 25th percentile of the fitness distribu- cardiovascular disease and all-cause mortality. tion results in fitness being a more powerful pre- Answering the question regarding whether fitness dictor of CHD risk than physical activity. In other or activity more strongly predicts outcomes is a words, at all percentiles greater than the 25th, the difficult undertaking, complicated by the fact that the two measures are related (with correlation
446 E X E R C I S E A N D T H E H E A R T ■ FIGURE 13-3 Degree of risk reduction in coronary heart disease or other cardiovascular diseases from 8 physical fitness and 30 physical activity cohorts. (From Williams PT: Physical fitness and activity as separate heart disease risk factors: A meta-analysis. Med Sci Sports Exerc 2001:33;754–761). coefficients ranging in the order of 0.30 to 0.60), mortality was determined over a 5-year follow-up. and because fitness also has a strong genetic com- Age-adjusted all-cause mortality was highest in ponent, in addition to the fact that fitness may be men who were physically unfit at both examina- influenced by subclinical disease and other fac- tions (122/10,000 person-years), and the lowest tors. Even with large, prospective analyses death rate was observed in men who were physi- addressing both fitness and activity measures in cally fit at both examinations (39.6/10,000 person- the same population, a complete answer to this years). Men who improved from unfit to fit between issue will remain elusive. Physical activity devel- tests had a reduction in mortality risk of 44% rel- ops physical fitness, although the magnitude of ative to men who were unfit at both examinations. the response to an exercise stimulus is likely For each minute of increase in maximal treadmill genetically determined. Nevertheless, activity is time between tests, there was a corresponding likely required to develop and maintain a fitness 7.9% decrease in risk of all-cause mortality. The level that is consistent with good health. This sci- results were similar for cardiovascular mortality. entific question should not be a distraction from Erikssen et al113 made similar observations in the important public health message that seden- Sweden; a graded inverse relationship was observed tary individuals should become more physically between change in fitness on two exercise tests active. (performed 7 to 10 years apart) and both cardio- vascular and all-cause mortality, irrespective of Changes in Physical Fitness or initial fitness level. Activity and Health Risk In the Corpus Christi Heart Project, Steffen- Several groups of investigators have also studied Batey et al,157 studied the association between changes in physical fitness (as assessed by serial change in level of physical activity and risk of treadmill testing) and its association with death or reinfarction was studied in 406 patients mortality risk. Blair et al112 conducted two exercise who survived a first MI. Patients were inter- evaluations, performed a mean of 4.9 years apart, viewed at baseline and annually thereafter about to evaluate change or lack of change in fitness, and physical activity, medical history, and risk factors for CHD. Change in level of activity after the infarction was categorized as: (1) sedentary, no
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 447 change (referent group); (2) decreased activity; The 30-minutes per day recommendation paral- (3) increased activity; and (4) active, no change. lels an energy expenditure in the order of 1000 kcal Over a 7-year period, the relative risks of death per week, and comes from the observation that were as follows: 0.21 for the active, no change this degree of energy expenditure is achievable by group; 0.11 for the increased activity group; and most individuals, and is associated with 20% to 0.49 for the decreased activity group. The relative 40% reductions in cardiovascular and all-cause risks of reinfarction were as follows: 0.40 for the morbidity and mortality. Despite the fact that this active, no change group; 0.22 for the increased relatively small investment in physical activity activity group; and 0.93 for the decreased activ- yields a major benefit in health outcomes, studies ity group. These findings are consistent with have shown that few physicians discuss exercise those observed among asymptomatic populations with their patients.160 Despite major educational and suggest a beneficial role of increasing physi- efforts made by governmental agencies, the vast cal activity in men and women who survive a majority of adults in Western societies remain first MI. effectively sedentary,106,107 in part due to the fact that physical activity is not currently integrated RECOMMENDATIONS ON into the healthcare paradigm. Physicians need PHYSICAL ACTIVITY to be more proactive in educating and preparing previously sedentary individuals to become more Previous position statements published by the physically active. AHA,115,158 the ACSM,32 and the National Institutes of Health (NIH)159 on the recom- “Health” versus “Fitness” mended quality and quantity of exercise were Benefits of Exercise extended in 1996 with the publication of the Surgeon General’s Report on Physical Activity A noteworthy theme that is consistent in each of and Health.107 This document was the strongest the recent documents described above is that con- policy statement ever made by the U.S. govern- siderable health benefits are derived from moder- ment concerning physical activity. It represented ate levels of activity; it is generally not necessary a historical turning point redefining exercise as a to engage in vigorous activity to derive many of key component for health promotion and disease these benefits. In past decades, an exercise pro- prevention. The federal government mounted a gram was thought to be effective only if an improve- multiyear educational campaign based on this ment in some measure of cardiopulmonary report. In this report, the epidemiologic evidence function was observed. In recent years, however, the supporting physical activity in the prevention of philosophy on recommending exercise as a means CHD morbidity and mortality was reviewed in to this end (“fitness” measured by exercise capacity) detail. The document also outlines how much has changed significantly. It is now appreciated that exercise is necessary to achieve these benefits. It substantial health benefits can be achieved through is suggested that each individual perform a mod- modest amounts of regular exercise, irrespec- erate amount of activity daily, with the amount of tive of whether exercise results in a measurable activity emphasized rather than the intensity, for improvement in exercise capacity. Epidemiologic 30 minutes or more on most, and preferably on all studies have shown that death rates from cardio- days of the week. These activities can take the vascular causes are considerably lower even form of brisk walking, yard work or other house- among individuals who engage in relatively small hold chores, jogging, or a wide variety of recre- amounts of exercise, less than the threshold that ational activities. Repeated intermittent or was generally thought needed to increase exercise shorter bouts of activity (at least 10 minutes), capacity.32,106,107,115,153,158,159 Although this issue has including occupational, nonoccupational, or also been debated, it is difficult to argue with the tasks of daily living, have similar cardiovascular wealth of evidence demonstrating that modest and health benefits if performed at a level of mod- amounts of activity are associated with consider- erate intensity (such as brisk walking, cycling, able health benefits. As health professionals, it is swimming, home repair, and yard work) with an important to be aware of the distinction between accumulated duration of at least 30 minutes per “health” and “fitness” when making activity rec- day. People who already meet these standards will ommendations to patients with cardiovascular dis- receive additional benefits from increasing this to ease, those at high risk for its development, and more vigorous activity. healthy adults.
448 E X E R C I S E A N D T H E H E A R T DANGERS OF EXERCISE valve, hypertrophic cardiomyopathy (HCM), and gastrointestinal hemorrhage (1% each). In about Sudden death has been defined relative to the onset 10% of cases, the cause of death is unknown. of symptoms, for example, instantaneous versus within 1, 6, or 24 hours. Autopsy findings in peo- Siscovik et al162 studied individuals who were ple dying instantaneously and those dying after reported by the paramedical immediate response 24 hours are different. Deaths occurring within system in Seattle to have had sudden death. They 6 hours of onset of symptoms include all electrical were compared with a matched sample that was deaths and are best defined as “sudden,” because randomly chosen by a special telephone-dialing no anatomic change usually can be demonstrated. device. To examine the risk of primary cardiac Most sudden deaths that occur during exertion do arrest during vigorous exercise, they interviewed so within minutes of the onset of symptoms. the wives of 133 men who had died suddenly of Sudden cardiac death is herein defined as “death heart disease that had not been previously diag- occurring unexpectedly within 6 hours of onset of nosed. The deceased were classified according to symptoms in a previously healthy person.” their time of cardiac arrest and the amount of their “Sudden” death, when defined this way, is the habitual vigorous activity. Among men with low most frequent mode of death. The incidence of levels of habitual activity, the relative risk of cardiac sudden death in the general population is high arrest during exercise compared with that at other (15% to 30%) with the majority (80% to 90%) times was 56. The risk during exercise among men due to cardiovascular causes. at the highest activity level was also elevated, but only fivefold, and their overall risk of cardiac arrest The mechanism of sudden cardiac death and at any time was only 40% that of sedentary men. acute MI in previously healthy adults is most often This was the first study to clearly document that, a rupture of an atherosclerotic plaque and acute although the risk of primary cardiac arrest is tran- coronary thrombosis. This has been confirmed by a siently increased during vigorous exercise, regular number of autopsy studies in recent years, and has exercise markedly decreases the risk. engendered the concept of the “vulnerable plaque” syndrome. Studies suggest that roughly 70% of Waller and Roberts163 reported on the autopsies MIs are caused by inflammation leading to “soft” or of five conditioned runners aged 40 years and older, vulnerable plaque in the arterial wall. Exercise all of whom had severe coronary atherosclerosis. could induce disruption of a vulnerable plaque by A series by Thompson et al164 described 18 jog- the dynamic vasodilation and vasoconstriction that gers, with five “exercising regularly” for at least occurs with exercise, a “twisting” motion of the 1 year and nine exercising for 3 or more years. artery, or the abnormal vasoconstriction that Fifteen of 18 died suddenly while jogging, and of occurs in atherosclerotic segments of a coronary those, 13 had CHD. Waller165 described 10 patients artery during exercise.1 In the following, some of over 30 years of age who ran 1 to 55 miles per week the major studies on the incidence and causes of for 1 to 12 years. All had at least one artery severely sudden death during exercise are reviewed. narrowed by an atherosclerotic plaque and six had MIs. As a sequel to their study of jogging deaths in Coronary Atherosclerosis in California, Thompson et al166 assessed the incidence Joggers and Marathon Runners of sudden death during jogging in Rhode Island. From 1975 through 1980, 12 men died during jog- Interest in causes of death in joggers and marathon ging, and the cause of death in 11 was CHD. From runners was stimulated in the 1970s when claims a telephone survey, they found that 7.4% of adult were made that “marathon running provides com- male Rhode Islanders jogged at least twice a week. plete immunity from CAD.”161 Since then, numer- The incidence of death from jogging was one death ous cases of death in joggers and marathon runners per year for every 7620 joggers and only one death have been reported. The most common cause of per 396,000 man-hours of jogging. This rate is death has been coronary atherosclerosis (75%). The seven times the estimated death rate from CHD most common non-cardiovascular death in run- during more sedentary activities in Rhode Island. ners results from automobile accidents. Rare car- Vander et al167 conducted a 5-year retrospective sur- diovascular causes of death have been reported to vey of fatal and nonfatal cardiovascular events that include amyloidosis and tunnel coronary artery occurred in community recreation centers. Fifty- (2% each), myocarditis, congenital hypoplastic eight facilities reported 30 nonfatal and 38 fatal coronary arteries, heat stroke, prolapsed mitral events. There was one nonfatal and one fatal event every 1,124,200 and 887,526 hours of participation, respectively.
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 449 What is the expected level of cardiovascular CAD; while waiting for surgery, he died suddenly. deaths among runners while running if one consid- The last two cases were 36- and 27-year-old ath- ers chance alone? This is an important question letes who had completed multiple marathons and because it is often assumed that exercise is the were killed accidentally. Both had left anterior cause when a person dies of cardiovascular causes descending coronary artery lesions at autopsy; the during recreational running. Koplan168 used data younger had a 50% and the older had a 90% from the National Center of Health Statistics and lesion. found that approximately 100 cardiovascular deaths per year in runners in the United States An interesting review of this topic was later can be predicted on a purely temporal basis. This published by Noakes171 in which the deaths of 36 is certainly higher than the number of deaths marathon runners previously reported in the reported. Morales et al169 described three healthy medical literature were described. This group had individuals—two males aged 34 and 54 years and a mean age of 43.8 years (range, 18 to 70 years) one female 17 years of age—who died suddenly and 75% had a cardiovascular cause of death. during strenuous exercise and were found to have Seventy-one percent of these runners with coro- a triad of pathologic findings. The pathologic triad nary disease had forewarning symptoms that they was muscle bridging of the left anterior descend- tended to ignore and continued training. Of the 26 ing coronary artery, poor circulation to the poste- runners from whom data were available, 50% died rior surface of the heart, and septal fibrosis. The within 24 hours of a competitive running event. angiographic finding of a coronary artery that Thus, unlike what was thought for many years, it passes underneath a band of myocardium is not is clear that high levels of aerobic performance do that unusual and it has been debated whether or not exclude the presence of significant coronary not it has functional significance. Some studies of disease, and symptoms in such patients should be coronary blood flow have suggested that the con- taken seriously. striction of a coronary artery by this myocardial band during systole results in decreased flow; how- Virmani and McAllister172 reported findings in ever, most coronary flow takes place during dias- 30 joggers or marathon runners who died from tole. In regard to the second finding, there is nontraumatic causes. Twenty-two men died with great variability in the coronary artery distribu- severe atherosclerosis; their ages ranged from 18 tion on the posterior surface of the heart around to 54 years (mean 36 years). The history of jog- the crux and the posterior margin of the septum. ging was well documented in 18 patients who ran In the most common situation, the right coronary 7 to 105 miles per week (mean 33 miles) and had artery branches into a posterior descending artery, been running 1 to 28 years (mean, 10 years). Three which passes down the septum, giving off septal were marathon runners and the others had been perforators. Often there are normal variations jogging for at least 6 months. Review of records where the left circumflex provides this branch or revealed a family history of heart disease in nine, there are only small arteries in the area. Lastly, systemic hypertension in nine, and total choles- septal fibrosis could be due to chronic ischemia. terol greater than 200 mg/dl in seven. None were These anatomic findings could also be purely diabetic and smoking history was uncertain. A his- coincidental. tory of CHD was present in eight (27%); of these, five were from a retrospective review of medical Noakes et al170 presented four marathon run- records. Nineteen died suddenly and three had a ners with autopsy-proven coronary atherosclero- history of prolonged chest pain. In six patients, sis. The first individual was a 44-year-old male death occurred soon after jogging, and two were who, after 14 months of training, had completed found dead in bed. At autopsy, the heart weight seven marathons in less than 4 hours each. He sud- ranged from 345 to 600 g (mean, 432 g). In denly dropped dead half-way through a marathon. 16 patients, the heart weight was increased beyond At autopsy he was found to have an old anteroseptal the normal range. Of the four major arteries exam- MI and 90% lesions of his left anterior descending ined for severe atherosclerosis, only one artery and circumflex coronary arteries. The second was was involved in nine patients (41%), two coronary a 41-year-old man who, after 2 years of running, arteries in nine (41%), and three and four coro- had a symptomatic MI. After release from the hos- nary arteries in one each. Thrombi were noted in pital, he returned to training and ran in five six (27%) patients. The most common single artery marathons. He was later hospitalized again with involved was the left anterior descending, and the unstable angina and coronary angiography was per- most common combination was the left anterior formed. He was found to have severe triple vessel descending and right coronary arteries. Of a total of 70 coronary arteries examined in 20 joggers,
450 E X E R C I S E A N D T H E H E A R T 34 (49%) were severely narrowed, and the average immediately after vigorous activity. Their post- number of coronary arteries greater than 75% mortem studies of 22 young athletes (mean age, narrowed was 1.65 per jogger. Those with a his- 23 years) showed arrhythmic cardiac arrest to be tory of CHD had a similar extent of coronary ath- the most common cause of death (17 cases) and erosclerosis as those without such histories (1.7 right ventricular dysplasia and atherosclerotic CAD versus 1.6 coronary arteries narrowed per patient). were the most frequent underlying cardiovascular In six of the 22 with severe coronary atherosclerosis, diseases (six and four cases, respectively). Of note, isolated healed MI was present; acute MI with or many of the athletes had experienced premonitory without healed MI was present in eight. A total of signs or had abnormal baseline ECGs. 14 (64%) of 22 who died of severe coronary athero- sclerosis had MI. A community-based participation screening examination appeared to prevent sudden death due Virmani and Robinowitz173 studied another to HCM in a 17-year prospective study of young 11 male joggers with a mean age of 41 years (range, people in Italy.177 Of the 33,735 young athletes 19 to 59 years). Sudden death occurred while jog- (mean age, 19 years) screened, 3016 were referred ging in 9 of 11 men. Available risk factor history was for echocardiography and 22 of them were as follows: two had hypercholesterolemia, one had excluded from competition due to HCM. Sudden systemic hypertension, and one had a family his- death occurred in 269 of the total 2,009,600 sub- tory of premature CHD. Only two had a history of jects under the age of 35 years in the geographic prior cardiac disease: one had angina, and one had area. Athletes accounted for 49 deaths and undergone left ventricular aneurysmectomy with nonathletes for 220, resulting in rates of 1.6 versus coronary bypass surgery. A 43-year-old man had 0.75 deaths per 100,000 persons, respectively (rel- been jogging 50 miles per week for 5 to 6 years and ative risk for nonathletes versus athletes = 2.1). had participated in several marathons. His heart The most common cause of death was right ven- weighed 600 g, an acute MI was found, and there tricular cardiomyopathy (22.4% of deaths). This was a greater than 75% cross-sectional area luminal study confirms the extremely low incidence of ath- narrowing of the three major coronary arteries, letic sudden deaths, but points out potentially with thrombus in the left circumflex. Seven had important ethnic differences; HCM may be more at least two vessels severely narrowed, one had common in the United States, whereas right ven- one-vessel disease, and the other two had been tricular cardiomyopathy may be more common in described as having severe coronary atherosclero- Italy. sis. Acute or healed MIs were present in six of 10. Maron et al178 reported sudden unexpected Causes of Death During or Soon death in 29 highly conditioned, competitive ath- After Exercise Other Than letes aged 13 to 30 years (mean age, 19 years) Jogging or Marathon Running. drawn from news media reports, the registry of the cardiovascular division of the Armed Forces Opie174 reported the causes of sudden death in Institute of Pathology, and the pathology branch 21 athletes, 13 of whom took part in rugby or soc- of the NIH. All had been active, highly condi- cer. Eighteen were thought to be caused by CHD. tioned members of an organized athletic team for The Squash Rackets Association has estimated at least 2 years. The type of sport varied, but bas- that there may be 2.5 million people in the United ketball and football were most common. In 28 of Kingdom playing squash once or more a month. the 29 athletes, death occurred suddenly without The circumstances surrounding 60 sudden deaths warning and was virtually instantaneous, occur- associated with squash playing were described ring on the playing field in 13. One athlete sur- by Northcote et al175 in Glasgow. The mean age of vived 12 hours after collapse. In 22 athletes, death those who died was 46 years (range, 22 to 66 years). occurred during or soon after severe exertion, in They were able to collate a series of 89 sudden two after mild exertion, and in five during sedentary deaths associated with squash that occurred activities. Structural cardiovascular abnormali- between October 1976 and February 1984 by exam- ties were found in 28 athletes and were the cause of ining press reports and by a prospective mail sur- sudden death in 22. Of these, the most common vey of sports centers and squash clubs throughout anatomic abnormality was HCM, which was pres- the United Kingdom. ent in 14. HCM was defined as asymmetric septal hypertrophy, with marked ventricular septal dis- Corrado et al176 from Italy also found that sud- organization in another two. Four athletes had den death in athletes occurs most often during or anomalous origin of the left coronary artery from the right sinus of Valsalva, including one patient
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 451 with HCM. Four athletes had concentric left ven- symptomatic individuals are excluded from mili- tricular hypertrophy, two with and two without tary service and none was highly trained, whereas disorganization. Three athletes (24 to 28 years of Maron’s population included only trained ath- age) had severe coronary atherosclerosis. Two letes. Virmani’s subjects had a wide age range, with died of aortic rupture; both had evidence of cystic only 25 being 30 years old or younger, whereas medial necrosis, and one had Marfan’s syndrome. Maron’s athletes were 13 to 30 years old, with a mean of 19 years. Prevalence of CHD is directly In six athletes, the cardiovascular abnormality related to age. This has been confirmed in other was considered probable evidence of cardiovascu- studies of runners older than 40 years of age: CHD lar disease: five had hypertrophied hearts (420 to is the most common cause of death and they usu- 530 g), one had mild prolapse of anterior and pos- ally have had symptoms of CHD prior to the terior mitral leaflets, and one had normal heart event. Moreover, all of these events are extremely weight with a hypoplastic right coronary artery. unusual and it would be difficult to screen for Several died of coronary atherosclerosis, one after them. It is known that athletes frequently have running a pass pattern in a professional football abnormal ECGs and even echocardiographic hyper- game. In this individual, they hypothesized that a trophy. In addition, they have a higher prevalence blow to the chest while being tackled caused a hem- of false-positive exercise tests. However, screening orrhage into a plaque in the left anterior descend- for lipid abnormalities would be a wise public ing coronary artery. Several others had congenital health measure regardless of a lack of specificity, anomalies of the coronary arteries. Virmani and and it would be advisable to get an echocardiogram Robinowitz173 reviewed records of 32 individuals on an athlete with symptoms or signs of HCM. who died suddenly while engaging in either mili- tary training (six) or in other sports activity: bas- Van Camp et al180 examined 136 deaths that ketball (six), running (eight), racquetball (two), occurred during or within 1 hour of sports partic- volleyball (two), tennis (two), swimming, (two), ipation over 10 years among high school or col- football (two), and one each in gymnastics and lege athletes. Cardiac conditions were responsible bowling. Their ages ranged from 14 to 60 years, for 100 of the cases, including HCM (56% of the with a mean age of 28 years; 31 were males and one cases), coronary artery anomalies (13%), myocardi- was female. The anatomic abnormalities were var- tis (7%), aortic stenosis (6%), and dilated cardiomy- ied: coronary disease in eight, idiopathic myocardi- opathy (6%). Together, the studies performed tis in four, congenital coronary abnormalities in among younger individuals (i.e., <30 years) three, HCM in two, tunnel coronary artery in two, demonstrate that CAD is a rare cause of exercise- abnormal mitral valve in two, intramural coro- related death in younger individuals. Among high nary thickening in two, and one each with rheu- school and college athletes, only about 30% of matic heart disease and aortic dissection. Four had deaths associated with exercise are also associated left ventricular hypertrophy of unknown cause with CAD. Rather, exercise-related deaths in young (420- to 600-g hearts) and were 17, 20, 21, and subjects are usually caused by either congenital car- 32 years old. All died during exertion. Three who diac abnormalities or nonatherosclerotic acquired died while running had sickle cell trait and were cardiovascular diseases. only 17, 20, and 22 years old. In a study that has been widely cited since the Maron et al179 more recently quantified deaths mid-1990s as evidence that regular exercise offers during competitive sports among high school ath- protection against triggering MI, Mittleman letes over a 12-year period using an insurance et al181 conducted interviews among 1228 patients database in Minnesota. During this period, there who had sustained an MI an average of 4 days ear- were 1,453,280 sports participations in 27 differ- lier. Information was gathered regarding their ent sports, and there were 12 deaths due to car- usual annual frequency and intensity of physical diovascular disease. The risk of sudden death was exertion, along with their physical activity in the 1 per 217,400 participants per year (or 0.46/100,000 26 hours prior to their MI. The estimated relative annually); the risk was slightly higher in male ath- risk of MI in the hour after heavy physical exertion, letes. The rare incidence of sudden death in this as compared with less strenuous physical exertion and other studies points out the practical and cost or no physical exertion whatsoever, was 5.9. limitations associated with broad screening pro- Comparing subjects who regularly participated in grams for young athletes. physical activity (equaling or more than five times per week) to those who participated in physical The causes of death in Virmani and activity less than once a week the reactive risk for McAllister’s172 subjects were markedly different the less active was 107. Stated differently, while from those of Maron et al179, probably because
452 E X E R C I S E A N D T H E H E A R T a bout of exercise transiently increases the risk of and abnormal muscle enzymes. Treatment MI by six-fold, being habitually sedentary includes submersion in ice water and intravenous increased the risk of MI during exertion by 107 heparin to stop fibrinolysis. At autopsy, the findings times. In a similar study from Germany, interviews usually are nonspecific and consist of petechial were conducted among 1194 acute MI patients.182 hemorrhages in the skin, mucous membranes, The adjusted relative risk of the infarction being brain, lung, and heart. The hemorrhages in the associated with strenuous physical activity was 2.1. heart are most pronounced in the epicardial and However, the risk of infarction among subjects endocardial region, especially on the left side of who were regularly active (equal to or more than the ventricular septum. Damage to myocardial fil- versus less than four times per week) was 5.3 aments and intercalated discs have been described times lower than those who were less active. by electron microscopy in patients with malignant hyperthermia induced by anesthetic agents. Effect of Environment Summary of Exercise-Related An important factor in sudden death among ath- Death in the Athlete letes and joggers is the climate in which exercise is being performed. Serious thermal injuries are Although exercise contributes to sudden death preventable, and the ACSM recommends that long- in susceptible persons, numerous studies have distance races should not be conducted in tem- demonstrated that the risk of exercise is extremely peratures that exceed 28°C (82.4°F). The amount small and suggests that routine screening is not of heat generated is directly related to the intensity justified. Cardiovascular diseases responsible for of exercise. The body is only 25% efficient in con- sudden unexpected death in highly conditioned verting calories generated into external work, and athletes are largely related to the age of the patient. the remaining 75% of energy is converted into heat. In most young, competitive athletes (<35 years of Therefore, a large amount of heat must be lost by age), sudden death is due to congenital cardiovas- the body to prevent raising the core temperature. cular disease. HCM appears to be the most common cause of such deaths, accounting for about half of If the body does not lose any heat, the core tem- the sudden deaths in young athletes. Other cardio- perature would increase by 1°C every 5 minutes. It vascular abnormalities that appear to be less fre- is the efficient mechanisms of thermoregulation quent in young athletes include congenital of the body that prevent hyperthermia. These coronary artery anomalies, ruptured aorta (due to mechanisms include sweating and heat loss by radi- cystic medial necrosis), idiopathic left ventricular ation and by conversion. The factors that prevent hypertrophy, and coronary atherosclerosis. Very heat loss are high ambient temperature, high uncommon causes of sudden death include myo- humidity, dehydration (which prevents cutaneous carditis, mitral valve prolapse, aortic valve stenosis, vasodilation), extremes of age, debilitation, exces- and sarcoidosis. The recent data from Italy impli- sive clothing, and drugs that may impair ther- cating right ventricular dysplasia and hypertrophy moregulation. The spectrum of heat injury as a major cause of sudden death in young athletes includes three well-recognized syndromes: (1) heat requires further study, because these findings have cramps; (2) heat exhaustion; and (3) heat stroke. not been confirmed in other populations. Heat cramps are painful spasms in the muscles, whereas heat exhaustion is characterized by fatigue, Cardiovascular disease in young athletes usually hyperventilation, headache, lightheadedness, nau- is unsuspected during life, and most athletes who sea, and muscle cramps. Patients with heat exhaus- die suddenly have experienced no cardiac symp- tion sweat and have chills despite the core toms. In only about 25% of the competitive ath- temperature being high. Heat stroke, the most letes who die suddenly is underlying cardiovascular serious of thermal injuries, is characterized by an disease detected or suspected before participation, altered state of consciousness that may progress and rarely is the correct clinical diagnosis made. rapidly to unconsciousness and seizure activity. In contrast, in older athletes (≥35 years of age), The heat stroke patient is hot, flushed, and has dry sudden death is usually due to CAD. Noninvasive skin because sweating has stopped. Dehydration screening procedures currently are available that and circulatory collapse soon follow. Body tem- can detect many subjects at risk of sudden death, perature is usually above 41°C (106°F) and labo- but with an uncertain specificity. However, ratory tests show hemoconcentration, leukocytosis, although some potentially lethal diseases can be azotemia, acidosis, abnormal liver function tests, excluded by a relatively simple screening program,
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 453 other diseases require expensive procedures, such Transient ischemia is a plausible cause for as echocardiography, exercise testing, and cardiac most episodes of exertion-related cardiac arrest in catheterization. This means that the sensitivity of patients with coronary disease. In assessing resus- detecting diseases leading to sudden death citated patients who collapsed during or after increases in proportion to the financial resources exertion, Cobb and Weaver183 found that, com- that can be applied to the screening program. Thus, pared with persons with non-exertion-related car- when a screening program designed to identify all diac arrest, these patients had fewer limitations and cardiac diseases that have the potential to cause more often had no recognized preceding heart dis- sudden death is planned by a community, school, ease. In addition, warning symptoms were noted in or nonprofessional athletic team, the costs will be only about 25%, and less than one third had new prohibitive. The practicality of applying a commu- Q waves. These patients have few episodes of ven- nity or school screening program can be ques- tricular arrhythmia during ambulatory monitoring. tioned because of the very low incidence of sudden Although there has been no large, prospective unexpected death in young healthy individuals. assessment of the role of exertion in precipitating Comprehensive screening programs are confined to cardiac arrest, some relevant information is avail- individuals or organizations with adequate financial able. In patients treated by the paramedic system resources. Less expensive, limited screening can be in Seattle, 36 (11%) of 316 consecutive victims undertaken by individuals or groups to identify had collapsed during or immediately after exer- some subjects at risk of sudden death during ath- tion or stress. This incidence is similar to that of letic competition. An important consideration is the 26 of 150 (17%) patients reported in Miami. In the education of the team physician. Symptoms autopsy registries, the incidence of exertion- and family history of sudden death or syncope related cardiac arrest was reported to be 10% to should not be overlooked. However, due to high 30% of all sudden deaths. In studies of unexpected vagal tone, fainting can occur in young athletes. sudden death in younger persons, cardiac arrest In addition, ECG abnormalities, S3 waves, and commonly was associated with physical activity. systolic murmurs are common. In a prospective 5-year survey by Hinkle et al184 The normal heart, even when subjected to vig- involving approximately 270,000 men, 42% of the orous forms of stress, is protected from lethal sudden coronary deaths occurred in persons with- arrhythmias except in unusual conditions such as out previously recognized coronary disease. About profound electrolyte derangement, thermal stress, one third of these deaths occurred within minutes or adverse drug reactions. Victims of sudden death of engaging in activities known to be associated almost always have underlying heart disease. CAD with myocardial ischemia or in the setting of sus- is found in about 80% of victims of sudden car- pected sympathetic nervous system stimulation. diac death, whereas other abnormalities, such as As reported by the Cooper Clinic, in a predomi- cardiomyopathy, valvular heart disease, or primary nantly normal population of middle-aged persons, arrhythmic disorders, also may cause unexpected one cardiac arrest occurred per 375,000 person cardiac arrest. Although exertion-related death hours of exercise. In the Framingham Study, there appears to be confined to patients with structural was a significant association between the mode of heart disease, a third of these individuals may be death and activity; sudden death occurred more asymptomatic. Mechanisms underlying sudden often in the setting of physical activity. Cobb and death in cardiac patients include ventricular fib- Weaver183 reported that in 133 men who experi- rillation and myocardial ischemia. Ventricular fib- enced cardiac arrest in Seattle, the incidence of rillation is the arrhythmia usually underlying the cardiac arrest was 5 to 56 times greater during sudden cardiac death syndrome, particularly in high-intensity exercise than at other times. The exertion-related events. In the follow-up of persons considered in that study were aged 25 to patients resuscitated from out-of-hospital ventric- 75 years and were without previously recognized ular fibrillation, Cobb and Weaver183 recognized cardiovascular disease. The estimated incidence of three major clinical settings in which ventricular cardiac arrest during vigorous activity ranged from fibrillation occurs: (1) as a complication of typical one case per 137,000 hours to one per 4.7 million acute MI; (2) as a manifestation of transient myo- hours at risk. cardial ischemia, especially during or after exer- tion; and (3) as an event unassociated with ischemia These studies serve to point out that physical and occurring while sedentary. In the latter setting, exertion may precipitate cardiac arrest in the ventricular fibrillation most often occurs in patients normal population and that prior recognition of with prior MI and left ventricular dysfunction. susceptible individuals has not been possible. Exercise-induced cardiac arrest is a rare but real
454 E X E R C I S E A N D T H E H E A R T phenomenon, particularly in patients with known running frequently cause such problems. The heart disease. However, the majority of sudden remaining injuries were acute ones—incidents deaths are temporally associated with routine activ- that happened instantly—such as a sprained ankle. ities of daily life and not with exercise. Therefore, These tend to occur in skiing, football, basketball, the number of deaths due to strenuous physical and soccer. Injuries to the knee caused the most exertion is relatively modest. Exertion-related visits and skiers have the most knee problems. cardiac arrest usually is due to ventricular fibrilla- Aerobic dance causes more fractures than any other tion or tachycardia, and studies show that exercise recreational activity. Many problems stem from transiently increases this risk. Importantly, how- acute injuries that occurred in the past. Unlike ever, sudden death or MI during exercise is far less football injuries, most of the basketball injuries common among individuals who are regularly occur in participants older than 25 years of age. active versus those who are not, by a factor of 5 to 100 times. Is the recommended physical activity safe? De Loes and Goldie185 reported the incidence of Complications Other than Death injury from physical exercise as recorded by physi- cian visits over the course of an entire year in a There are numerous risks for amateur and profes- Swedish town with 31,620 inhabitants. They found sional athletes. Heat stroke can be avoided by taking that injuries from sports or physical exercise com- precautions for humid, hot environments, includ- prised 17% of all clinic visits for accidents. This ing adequate oral replacement of dilute electrolyte compared with 26% home-related injuries and solutions. There is no place for fluid restriction in 19% work-related injuries. It should be noted that, order to limit sweating. Runners can have heat whereas the Swedes play a great deal of ice hockey stroke and still be actively sweating, although it (the sport they found causing the greatest incidence was once taught that heat stroke was always pre- of injury), they do not play football. ceded by cessation of sweating. Hematuria after a run can be due to bladder trauma and proteinuria Education regarding how recreational injuries can even be normal. Diarrhea and other gastroin- happen and how to treat them is an important testinal complaints are fairly common in runners step in prevention. Treatment may include weight during and after events. Numerous episodes of lifting for rehabilitation, shoe inserts to correct anaphylaxis thought to be exercise-induced have irregularities in stride or foot strike, ultrasound been reported. Diagnosis by the findings of bron- and electrical stimulation for muscle tears and chospasm and urticaria is important because stiffness, and compression and icing to control treatment with epinephrine and antihistamines swelling. Rest, ice, compression, and elevation are can be lifesaving. still the best treatment for most acute injuries. There has been a trend toward active rehabilita- Orthopedic Injuries. The greater public aware- tion. For instance, to treat a sprained ankle, a pro- ness and participation in physical activity is gram that focuses on muscle strengthening is responsible for both general practitioners and used because if ligaments do not heal well, a tear sports medicine specialists noticing an increase in may become a persistent problem. Prolonged rest sports-related injuries among weekend and after- can cause a decrease in muscle mass around the work athletes. Basketball and soccer leagues, ski ankle, resulting in a loss of strength. The muscles vacations, evening runs, dance classes, and tennis lose their ability to move quickly and stabilize the cause injuries once found chiefly among profes- ankle. Instead, strengthening surrounding mus- sional and college athletes. cles will avoid atrophy. In many cases, the strengthened muscles will compensate for the The Center for Sports Medicine in San Francisco deficient ligaments, making the joint stable for compiled statistics on over 10,000 injuries treated further activities. at the center. They found that nine activities— basketball, dance, football, gymnastics, running, Knochel186 reported that “white collar rhab- skiing, tennis, soccer, and figure skating— domyolysis” (weekend competitive running in accounted for nearly three fourths of the injuries. middle-aged, moderately conditioned people lead- More than two thirds of the injuries were caused ing to rhabdomyolysis) is much more common by overuse—problems such as shin splints and than currently thought. Rhabdomyolysis refers tendonitis that develop from a repetitive trauma to to breakdown of muscle fibers and the release of muscle and bone. Tennis, aerobic dance, and myoglobin and other intracellular contents into the plasma. This syndrome can be avoided by ensuring gradual conditioning and reasonable competition.
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 455 SUMMARY Although regression of atherosclerosis has been repeatedly demonstrated in the 1990s with the use Animal studies have provided substantial evidence of statins, only a few of these studies have incorpo- of the cardiovascular benefits of regular physical rated physical activity; thus, it is not possible at activity. Improved coronary circulation has been present to determine the independent contribu- demonstrated in exercise-trained animals through tion of physical activity to retarding or regressing increases in coronary artery size, capillary density, the atherosclerotic process. Nevertheless, although and collateral development in response to hypoxia. physical inactivity does not necessarily precede the Studies utilizing various animal models have atherosclerotic process, its relationship to cardiac reported improvements in cardiac function second- events is certainly strong. The level of exercise nec- ary to exercise training. Improved intrinsic con- essary to lessen the risk of cardiovascular death tractility, faster relaxation, enzymatic alterations, differs from that required to obtain the hemody- calcium availability, and enhanced autonomic and namic and morphologic benefits of more rigorous hormonal control of function all have been sug- training. The latter requires careful attention to gested as reasons for these findings. training intensity, duration, frequency, and mode. The prescription for good health, however, can be These animal studies demonstrate that there are less demanding. Moderate or vigorous walking for morphologic and metabolic changes that make the a minimum of 30 minutes, five to seven times per cardiovascular system better able to withstand week, is sufficient to obtain many of the health stress, possibly even that imposed by atherosclero- benefits from exercise. sis. The study by Kramsch et al31 provides the strongest evidence for the favorable impact of exer- Interestingly, activity surveys have demon- cise and diet on the primary prevention of coro- strated that more than 50% of the American pop- nary disease. However, although exercise lessened ulation exercises less than 20 minutes three times ischemic manifestations, only diet stopped the a week. This makes inactivity a very prevalent risk progression of coronary atherosclerosis. Although factor and increases the population-attributable myocardial ischemia seems to be a necessary risk of inactivity in modern society far above that of stimulus for the development of collateral vessels, other risk factors. Recent studies of primary preven- exercise appears to enhance their development. tion support a lifestyle of regular physical activity Exercise probably does not affect the atheroscle- to decrease one’s risk for CHD. Such physical rotic process, but instead enlarges coronary arteries activity helps to decrease other risk factors as to provide protection via increased flow. Precisely well.47,115,187,188 Regular, moderate exercise can how these observations made among animals improve one’s quality of life by lessening fatigue relate to the human heart is unknown, particu- and by increasing physical performance. larly because many of the changes are age-related. Although athletic deaths generate a great deal Echocardiographic studies have shown en- of interest in the public and the press, they are durance training in young subjects to result in extremely rare. The guidelines on screening ath- increased ventricular mass, wall thickness, volume, letes suggest that all of the available screening tech- and function, but not all results have been con- niques can cause more harm than good because of clusive, probably because of problems with mea- their high false-positive rates. Knowledge of the surement reproducibility. These increases in left causes of such deaths, however, can help focus our ventricular mass may not occur in younger subjects attention appropriately. Unfortunately, even an unless higher levels of exercise are used and may exercise test is not effective for predicting exercise- never occur in older subjects. In cardiac patients, related deaths in asymptomatic populations. The an exercise program may not lessen exercise- public health prescription of physical activity rather induced ischemia (as assessed by ST-segment than the higher levels of exercise needed for physi- depression), but multidisciplinary secondary pre- cal fitness carries minimal risk. vention programs have recently been shown to retard the progression of, and even regress, angio- Thus, both animal and human studies have graphic CAD. shown beneficial effects on the heart from regular, sustained exercise. Large epidemiologic studies It is difficult to separate the association between have shown significant benefits from improvements physical inactivity, the atherosclerotic process, and in physical fitness/activity, especially with regard other factors such as abnormal lipids, cigarette to decreasing cardiac mortality. Specific recom- smoking, and hypertension. An inverse association mendations have been put forth by various medical between the level of activity and regression of ath- societies; these detail the precise extent and dura- erosclerosis was reported by Hambrecht et al.56 tion of exercise needed to improve both quantity
456 E X E R C I S E A N D T H E H E A R T and quality of life. The challenge ahead is to con- 22. Posel D, Noakes T, Kantor P, et al: Exercise training after experi- vince two thirds of the American public to step mental myocardial infarction increases the ventricular fibrillation out of the ranks of the sedentary. They need not threshold before and after the onset of reinfarction in the isolated become Olympic athletes, but they must engage rat heart. Circulation 1989;80:138–145. in at least moderate activity. 23. Hull SS Jr, Vanoli E, Adamson PB, et al: Exercise training confers REFERENCES anticipatory protection from sudden death during acute myocardial ischemia. Circulation 1994;89:548–552. 1. Gordon JB, Ganz P, Nabel EG, et al: Atherosclerosis influences the vasomotor response of epicardial coronary arteries to exercise. 24. O’Conner GT, Buring JE, Yusaf S, et al: An overview of randomized J Clin Invest 1989;83:1946–1952. trials of rehabilitation with exercise after myocardial infarction. Circulation 1989;80:234–244. 2. Hambrecht R, Wolf A, Gielen S, et al: Effect of exercise on coronary endothelial function in patients with coronary artery disease. 25. Oldridge NB, Guyatt GH, Fischer ME, et al: Cardiac rehabilitation N Engl J Med 2000;342:454–460. with exercise after myocardial infarction. JAMA 1988;260:945–950. 3. Moyna NM, Thompson PD: The effect of physical activity on 26. Smart N, Marwick TH: Exercise training for patients with heart endothelial function in man. Acta Physiol Scand 2004;180: failure: A systematic review of factors that improve mortality and 113–123. morbidity. Am J Med 2004;116:693–706. 4. Dunn AL, Trivedi MH, O’Neal HA: Physical activity dose-response 27. Piepoli MF, Davos C, Francis DP, et al: Exercise training meta- effects on outcomes of depression and anxiety. Med Sci Sports analysis of trials in patients with chronic heart failure. BMJ Exerc 2001;33(suppl):S587–S597. 2004;328:189–195. 5. Bouchard C, Shephard RJ, Stephens T: Exercise, Fitness and 28. Holloszy JO: Minireview. Exercise and longevity: Studies on rats. Health. International Proceedings and Consensus Statement. J Gerontol 1988; 43:149–151. Champaign, Ill., Human Kinetics, 1994. 29. Lundeberg T, Kohler R, Bucinskaite V, et al: Physical exercise 6. Leon AS: Exercise following myocardial infarction. Current recom- increases survival after an experimental myocardial infarction in mendations. Sports Med 2000;29:301–311. rates. Cardiology 1998;90:28–31. 7. Taylor RS, Brown A, Ebrahim S, et al. Exercise-based rehabilitation 30. Powers SK, Quindry J, Hamilton K: Aging, exercise, and cardiopro- for patients with coronary heart disease: Systematic review and tection. Ann NY Acad Sci 2004;1019:462–470. meta-analysis of randomized controlled trials. Am J Med 2004;116:682–692. 31. Kramsch DM, Aspen AJ, Abramowitz BM, et al: Reduction of coro- nary atherosclerosis by moderate conditioning exercise in monkeys 8. Thompson PC, Buchner D, Pina IL, et al. Exercise and physical on an atherogenic diet. N Engl J Med 1981;305:1483–1489. activity in the prevention and treatment of atherosclerotic cardio- vascular disease: A statement from the Council on Clinical 32. American College of Sports Medicine: Position stand on the recom- Cardiology (Subcommittee on Exercise, Rehabilitation, and mended quantity and quality of exercise for developing and maintain- Prevention) and the Council on Nutrition, Physical Activity, and ing cardiorespiratory and muscular fitness in healthy adults. Med Sci Metabolism (Subcommittee on Physical Activity). Circulation Sports Exerc 1998;30:975–991. 2003;107:3109–3116. 33. Myers J, Froelicher VF: Exercise testing and prescription. Phys Med 9. Agency for Health Care Policy and Research: Guidelines for Cardiac Rehabil Clin N Am 1995;6:117–151. Rehabilitation. US Dept. of Health and Human Services, 1995. 34. American Association of Cardiovascular and Pulmonary 10. Poupa O, Rakusan K, Ostadal B: The effect of physical activity upon Rehabilitation: Guidelines for Cardiac Rehabilitation programs, the heart of vertebrates. Physical activity and aging. Med Sports 4th ed. Champaign, Ill., Human Kinetics, 2003. 1970;4:202–203. 35. Verrill DE, Ribisl PM: Resistive exercise training in cardiac rehabil- 11. Thmanek RJ, Tounton CA, Liskop KS: Relationship between age, itation: A review. Sports Med 1996;21:347–383. chronic exercise, and connective tissue of the heart. J Gerontol 1972;27:33–38. 36. Ehsani AA, Hagberg JM, Hickson RC: Rapid changes in left ventric- ular dimensions and mass in response to physical conditioning and 12. Ljungqvist A, Unge G: Capillary proliferation activity in myocardium deconditioning. Am J Cardiol 1978;42:52–56. and skeletal muscle of exercised rats. J Appl Physiol 1978;43:306–312. 37. DeMaria AN, Neumann A, Lee G, et al: Alterations in ventricular 13. Tepperman J, Pearlman D: Effects of exercise and anemia on coro- mass and performance induced by exercise training in man evalu- nary arteries of small animals as revealed by the corrosion-cast ated by echocardiography. Circulation 1978;57:237–244. technique. Circ Res 1961;9:576–584. 38. Stein RA, Michielli D, Fox EL, et al: Continuous ventricular dimen- 14. Eckstein RW: Effect of exercise and coronary artery narrowing on sions in man during supine exercise and recovery. Am J Cardiol coronary collateral circulation. Circ Res 1957;5:230–235. 1978;41:655–660. 15. Spear KL, Koerner JE, Terjung RL: Coronary blood flow in physically 39. Parrault H, Peronnet F, Cleroux J, et al: Electro– and echocardio- trained rats. Cardiovasc Res 1978;12:135–143. graphic assessment of left ventricle before and after training in man. Can J Appl Sports Sci 1978;3:180–186. 16. Bloor CM, White FC, Sanders TM: Effects of exercise on collateral development in myocardial ischemia in pigs. J Appl Physiol 40. Wolfe LA, Martin RP, Watson DD, et al: Chronic exercise and left 1984;56:656–665. ventricular structure and function in healthy human subjects. J Appl Physiol 1985;58:409–415. 17. Heaton WH, Marr KC, Capurro NL, et al: Beneficial effects of physical training on blood flow to myocardium perfused by chronic collat- 41. Adams TD, Yanowitz FG, Fischer AG, et al: Noninvasive evaluation erals in the exercising dog. Circulation 1978;57:575–581. of exercise training in college-age men. Circulation 1981;64: 958–965. 18. Cohen MV, Yipinstoi T, Scheuer J: Coronary collateral stimulation by exercise in dogs with stenotic coronary arteries. J Appl Physiol 42. Landry F, Bouchard C, Dumesnil J: Cardiac dimension changes 1982;52:664–668. with endurance training. JAMA 1985;254:77–80. 19. Wang J, Wolin MS, Hintze TH: Chronic exercise enhances endothe- 43. Ehsani AA, Heath GW, Haberg JM, et al: Effects of 12 months of lium-mediated dilation of epicardial coronary artery in conscious intense exercise training on ischemic ST-segment depression in dogs. Circ Res 1993;73:829–838. patients with coronary artery disease. Circulation 1981;64:1116–1124. 20. Muller JM, Myers PR, Laughlin MH: Vasodilator responses of coronary 44. Ditchey RV, Watkins J, McKirnan MD, et al: Effects of exercise resistance arteries of exercise-trained pigs. Circulation 1994;89: training on left ventricular mass in patients with ischemic heart 2308–2314. disease. Am Heart J 1981;101:701–706. 21. Billman GE, Schwartz PJ, Stone HL: The effects of daily exercise on 45. Giannuzzi P, Temporelli PL, Corra U, et al: Antiremodeling effect of susceptibility to sudden cardiac death. Circulation 1984;69: long-term exercise training in patients with stable chronic heart 1182–1189. failure. Results of the exercise in left ventricular dysfunction and chronic heart failure trial. Circulation 2003;108:554–559. 46. Myers J, Ahnve S, Froelicher V, et al: A randomized trial of the effects of 1 year of exercise training on computer-measured ST seg- ment displacement in patients with coronary artery disease. J Am Coll Cardiol 1984;4:1094–1102. 47. Haskell WL, Alderman EL, Fair JM, et al: Effects of intensive mul- tiple risk factor reduction on coronary atherosclerosis and clinical
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 457 cardiac events in men and women with coronary artery disease: 71. Despres JP: Physical activity and adipose tissue: In Bouchard C, The Stanford Coronary Risk Intervention Project (SCRIP). Shepard RJ, Stephens T (eds): Physical Activity, Fitness and Health. Circulation 1994;89:975–990. Champaign, Ill., Human Kinetics, 1994, pp 358–368. 48. Kannel WB, Wilson P, Blair SN: Epidemiological assessment of the role of physical activity and fitness in development of cardiovascular 72. Glenny AM, O’Meara S, Melville A, et al: The treatment and preven- disease. Am Heart J 1985;109:876–885. tion of obesity: A systematic review of the literature. Int J Obes 49. Paffenbarger RS, Hyde RT, Wing AL, et al: The association of Relat Metab Disord 1997; 21:715–737. changes in physical-activity level and other lifestyle characteristics with mortality among men. N Engl J Med 1993;328:538–545. 73. Atkinson RL, Walberg-Rankin J: Physical activity, fitness, and 50. Myers JN, Prakash M, Froelicher VF, et al: Exercise capacity and severe obesity. In Bouchard C, Shepard RJ, Stephens T (eds): mortality among men referred for exercise testing. N Engl J Med Physical Activity, Fitness and Health. Champaign, Ill., Human 2002;346:793–801. Kinetics, 1994, pp 696–711. 51. Blair SN, Kohl HW, Barlow CE, et al: Changes in physical fitness and all-cause mortality. A prospective study of health and 74. Avenell A, Brown TJ, McGee MA, et al: What interventions should unhealthy men. JAMA 1995;273:1093–1098. we add to weight reducing diets in adults with obesity? A system- 52. Tanasescu M, Leitzmann MF, Rimm EB, et al: Exercise type and atic review of randomized controlled trials of adding drug therapy, intensity in relation to coronary heart disease in men. JAMA exercise, behaviour therapy or combinations of these interven- 2002;288:1994–2000. tions. J Hum Nutr Diet 2004;17:293–316. 53. Vermeulen A, Lie K, Durrer D: Effects of cardiac rehabilitation after myocardial infarction: Changes in coronary risk factors and long- 75. Fagard RH: Physical activity, fitness, and hypertension. In term prognosis. Am Heart J 1983;105:789–801. Bouchard C, Shepard RJ, Stephens T (eds): Physical Activity, Fitness 54. Schuler G, Hambrecht R, Schlierf G, et al: Regular physical exer- and Health. Champaign, Ill, Human Kinetics, 1994, pp 633–655. cise and low-fat diet: Effects of progression of coronary artery dis- ease. Circulation 1992;86:1–11. 76. Criqui MH, Mebane I, Wallace RB, et al: Multivariate correlates of 55. DeBusk RF, Houston Miller N, Superko HR, et al: A case-manage- adult blood pressures in nine North American populations: The ment system for coronary risk factor modification after acute lipid research clinics prevalence study. Prev Med 1982;11:391–402. myocardial infarction. Ann Intern Med 1994;120:721–729. 56. Hambrecht R, Niebauer J, Marburger C, et al: Various intensities of 77. Pescatello LS, Franklin BA, Fagard R, et al: American College of leisure time physical activity in patients with coronary artery dis- Sports Medicine position stand. Exercise and hypertension. Med ease: Effects on cardiorespiratory fitness and progression of coro- Sci Sports Exerc 2004;36:533–553. nary atherosclerotic lesions. J Am Coll Cardiol 1993;22:468–477. 57. Fletcher BJ, Dunbar SB, Felner JM, et al: Exercise testing and 78. Reaven PD, Barret-Connor E, Edelstein S: Relation between training in physically disabled men with clinical evidence of coro- leisure-time physical activity and blood pressure in older women. nary artery disease. Am J Cardiol 1994;73:170–174. Circulation 1991;83:559–565. 58. Nikolaus T, Schlierf G, Vogel G, et al: Treatment of coronary heart disease with diet and exercise-problems of compliance. Ann Nutr 79. Carroll S, Cooke CB, Butterly RJ: Metabolic clustering, physical Metab 1991;85:1–7. activity and fitness in nonsmoking middle-aged men. Med Sci 59. Kinsey GM, Fletcher BJ, Rice CR, et al: Coronary risk factor modi- Sports Exerc 2000;32:2079–2086. fication followed by home-monitored exercise in coronary bypass surgery patients: A four-year follow-up study. J Cardiopulm Rehabil 80. Cooper KH, Pollock ML, Martin RP, et al: Physical fitness levels vs 1989;9:207–212. selected coronary risk factors: A cross-sectional study. JAMA 60. Cannistra LB, Balady GJ, O’Malley CJ, et al: Comparison of the clin- 1976;236:166–169. ical profile and outcome of women and men in cardiac rehabilita- tion. Am J Cardiol 1992;69:1274–1279. 81. Hartung GH, Kohl HW, Blair SN, et al: Exercise tolerance and alco- 61. Milani RV, Lavie CJ, Mehra MR: Reduction in C-reactive protein hol intake: Blood pressure relation. Hypertension 1990;16: through cardiac rehabilitation and exercise training. J Am Coll 501–507. Cardiol 2004;43:1056–1061. 62. Mattusch F, Dufaux B, Heine O, et al: Reduction of the plasma con- 82. Gibbons LW, Blair SN, Cooper KH, Smith M: Association between centration of C-reactive protein following nine months of coronary heart disease risk factors and physical fitness in healthy endurance training. Int J Sports Med 2000;21:21–24. adult women. Circulation 1983;67:977–983. 63. Vincent KR, Braith RW, Bottiglieri T, et al: Homocysteine and lipoprotein levels following resistance training in older adults. Prev 83. Carnethon MR, Gidding SS, Nehgme R, et al: Cardiorespiratory fit- Cardiol 2003;6:197–203. ness in young adulthood and the development of cardiovascular 64. Ali A, Mehra MR, Lavie CJ, et al: Modulatory impact of cardiac reha- disease risk factors. JAMA 2003;290:3092–3100. bilitation on hyperhomocysteinemia in patients with coronary artery disaese and “normal” lipid levels. Am J Cardial 1998;82: 84. Aronson D, Sella R, Sheikh-Ahmad M, et al: The association 1543–1545. between cardiorespiratory fitness and C-reactive protein in sub- 65. Engblom E, Ronnemaa T, Hamaleinen H, et al: Coronary heart disease jects with the metabolic syndrome. J Am Coll Cardiol 2004;44: risk factors before and after bypass surgery: Results of a controlled trial 2003–2007. on multifactorial rehabilitation. Eur Heart J 1992;13:232–237. 66. Kallio V, Hamaleinen H, Hakkila J, Luurila OJ: Reduction in sud- 85. Church TS, Barlow CE, Earnest CP, et al: Associations between car- den deaths by a multifactorial intervention programme after acute diorespiratory fitness and C-reactive protein in men. Arterioscler myocardial infarction. Lancet 1979;2:1091–1094. Thromb Vasc Biol 2002;22:1869–1876. 67. Hamaleinen H, Luurila OJ, Kallio V, et al: Long-term reduction in sudden deaths after a multifactorial intervention programme in 86. Herrmann M, Wilkinson J, Schorr H, et al: Comparison of the patients with myocardial infarction: 10-year results of a controlled influence of volume-oriented training and high-intensity interval investigation. Eur Heart J 1989;10:55–62. training on serum homocysteine and its cofactors in young, 68. Taylor CB, Miller NH: Principles of health behavior change: In healthy swimmers. Clin Chem Lab Med 2003;41:1525–1531. ACSM Resource Manual for Exercise Testing and Prescription, 4th ed. Baltimore, Lippincott Williams & Wilkins, 2001, pp 556–561. 87. Duncan GE, Perri MG, Anton SD, et al: Effects of exercise on 69. Ornish D, Brown SE, Scherwitz LW, et al: Can lifestyle changes emerging and traditional cardiovascular risk factors. Prev Med reverse coronary heart disease? The Lifestyle Heart Trial. Lancet 2004;39:894–902. 1990; 336:129–133. 70. Savage PD, Lee M, Harvey-Berino J, et al: Weight reduction in the car- 88. Hahn RA, Teutsh SM, Rothenberg RB, Marks JS: Excess deaths diac rehabilitation setting. J Cardiopulm Rehabil 2002;22:154–160. from nine chronic diseases in the United States. JAMA 1986;264:2654–2659. 89. McGinnis JM, Foege WH: Actual causes of death in the United States. JAMA 1993;270:2207–2212. 90. Booth FW, Gordon SE, Carlson CJ, et al: Waging war on modern chronic disease: Primary prevention through exercise biology. J Appl Physiol 2000;88:774–787. 91. Haskell WL, Leon AS, Caspersen CJ, et al: Cardiovascular benefits and, assessment of physical activity and physical fitness in adults. Med Sci Sports Exerc 1992;24:5201–5220. 92. Manson JE, Greenland P, LaCroix AZ, et al: Walking compared with vigorous exercise for the prevention of cardiovascular events in women. N Engl J Med 2002;347:716–725. 93. Myers J, Atwood JE, Froelicher VF: Active lifestyle and diabetes. Circulation 2003;107:2392–2394. 94. Tanasescu M, Leitzmann MF, Rimm EB, et al: Physical activity in relation to cardiovascular disease and total mortality among men with type 2 diabetes. Circulation 2003;107;2435–2439.
458 E X E R C I S E A N D T H E H E A R T 95. Hu FB, Sigal RJ, Rich-Edwards JW, et al: Walking compared with 118. Morris JN: Uses of Epidemiology. London, Churchill Livingstone, vigorous physical activity and risk of type 2 diabetes in women: 1975. A prospective study. JAMA 1999;282:1433–1439. 119. Oliver RM: Physique and serum lipids of young London busmen 96. Hu FB, Stampfer MJ, Solomon C, et al: Physical activity and risk in relation to ischemic heart disease. Br J Ind Med 1967;24: for cardiovascular events in diabetic women. Ann Intern Med 181–187. 2001;134:96–105. 120. Stamler J, Kjelsberg M, Hall Y: Epidemiologic studies on cardio- 97. Helmrich SP, Ragland DR, Leung RW, Paffenbarger RS: Physical vascular-renal diseases: Analysis of mortality by age-race-sex- activity and reduced occurrence of non-insulin dependent dia- occupation. J Chronic Dis 1960;12:440–445. betes mellitus. N Engl J Med 1991;325:147–152. 121. Blackburn H, Taylor HL, Keys A: Coronary heart disease in seven 98. Manson JE, Nathan DM, Kroleski AS, et al: A prospective study of countries. Circulation 1970;41:I154–I161. exercise and incidence of diabetes among U.S. male physicians. JAMA 1992;268:63–67. 122. Epstein L, Miller GJ, Stitt FW, et al: Vigorous exercise in leisure time, coronary risk factors, and resting electrocardiogram in 99. Manson JE, Rimm EB, Stampfer MJ, et al: Physical activity and middle-aged male civil servants. Br Heart J 1976;38:403–409. incidence of non-insulin dependent diabetes mellitus in women. Lancet 1991;338:774–775. 123. Costas R, Garcia-Palmieri MR, Nazario E, et al: Relation of lipids, weight and physical activity to incidence of coronary heart dis- 100. Hagberg JM: Exercise, fitness, and hypertension. In Bouchard TC, ease. Am J Cardiol 1978;42:653–660. Shepard RJ, Stephens T, et al. (eds): Exercise, Fitness, and Health. Champaign, Ill., Human Kinetics, 1994, pp 455–566. 124. Paffenbarger RS, Laughlin ME, Gima AS, et al: Work activity of longshoremen as related to death from coronary heart disease and 101. Krousel-Wood MA, Muntner P, He J, et al: Primary prevention of stroke. N Engl J Med 1970;282:1109–1114. essential hypertension. Med Clin North Am 2004;88:223–238. 125. Paffenbarger RS, Wing AL, Hyde RT: Physical activity as an index 102. Marcus R: Drinkwater B, Dalsky G, et al: Osteoporosis and exer- of heart attack risk in college alumni. Am J Epidemiol 1978;108: cise in women. Med Sci Sports Exerc 1992;24:S301–S307. 161–167. 103. Murphy NM, Carroll P: The effect of physical activity and its inter- 126. Paffenbarger RS, Wing AL, Hyde RT: Chronic disease in former action with nutrition and bone health. Proc Nutr Soc 2003; college students: Physical activity as an index of heart attack risk 62:829–839. in college alumni. Am J Epidemiol 1981;108:161–175. 104. Lee I, Paffenbarger RS, Hsieh C: Physical activity and risk of devel- 127. Lee IM, Sesso HD, Oguma Y, et al: Relative intensity of physical oping colorectal cancer among college alumni. J Natl Cancer Inst activity and risk of coronary heart disease. Circulation 2003;107: 1991,83:1324–1329. 1110–1116. 105. Thune I, Furberg AS: Physical activity and cancer risk; dose- 128. Sesso HD, Paffenbarger RS, Lee IM: Physical activity and coronary response and cancer, all sites and site-specific. Med Sci Sports heart disease in men: The Harvard Alumni Health Study. Exerc 2001;33(suppl):S530–S550. Circulation 2000;102:975–980. 106. Pate RR, Pratt MP, Blair SN, et al: Physical activity and public 129. Lakka TA, Venalainen JM, Rauramaa R, et al: Relation of leisure-time health. A recommendation from the Centers for Disease Control physical activity and cardiorespiratory fitness to the risk of acute and Prevention and the American College of Sports Medicine. myocardial infarction in men. N Engl J Med 1994;330:1549–1954. JAMA 1995;273:402–407. 130. Shaper AG, Wannamethee G: Physical activity and ischaemic 107. U.S. Public Health Service, Office of the Surgeon General: heart disease in middle-aged British men. Br Heart J 1991;66: Physical Activity and Health: A Report of the Surgeon General. 384–394. Atlanta, U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic 131. Slattery ML, Jacobs DR, Nichaman MZ: Leisure time physical Disease Prevention and Health Promotion, 1996. activity and coronary heart disease death. The US Railroad Study. Circulation 1989;79:304–311. 108. Paffenbarger RS, Hyde RT, Wing AL, et al: Some interrelations of physical activity, physiological fitness, health, and longevity. In 132. Lee IM, Sesso HD, Paffenbarger RS: Physical activity and coronary Bouchard C, Shephard RJ, Stephens T (eds): Physical Activity, heart disease risk in men. Does the duration of exercise episodes Fitness, and Health, pp 119–133. Champaign, Ill., Human Kinetics, predict risk? Circulation 2000;102:981–986. 1994. 133. Kannel WB, Belanger A, D’Agostino R, et al: Physical activity and 109. Kohl HW: Physical activity and cardiovascular disease: Evidence physical demand on the job and risk of CV disease and death: The for a dose response. Med Sci Sports Exerc 2001;33(suppl): Framingham study. Am Heart J 1986;112:820–825. S472–S483. 134. Leon AS, Connett J, Jacobs DR, Rauramaa R: Leisure-time physi- 110. Lee IM, Paffenbarger RS: Do physical activity and physical fitness cal activity levels and risk of coronary heart disease and death. avert premature mortality? Exerc Sport Sci Rev 1996;24: JAMA 1987;258:2388–2395. 135–172. 135. Morris CK, Ueshima K, Kawaguchi T, et al: The prognostic value 111. Gulati M, Pandey DK, Arnsdorf MF, et al: Exercise capacity and the of exercise capacity: A review of the literature. Am Heart J risk of death in women. The St James Women Take Heart Project. 1991;122:1423–1431. Circulation 2003;108:1554–1559. 136. Myers J, Gullestad L: The role of exercise testing and gas exchange 112. Blair S, Kohl H III, Barlow C, et al: Changes in physical fitness and measurement in the prognostic assessment of patients with heart all-cause mortality: A prospective study of healthy and unhealthy failure. Curr Opin Cardiol 1998;13:145–155. men. JAMA 1995;273:1093–1098. 137. Mark DB, Lauer MS: The prognostic variable that doesn’t get 113. Erikssen G, Liestol K, Bjornholt J, et al: Changes in physical fit- enough respect. Circulation 2003;108:1534–1536. ness and changes in mortality. Lancet 1998;352:759–762. 138. Myers J: Beyond ST-segment displacement: Newer diagnostic and 114. American College of Sports Medicine Position Stand: The recom- prognostic markers from the exercise test. Am J Med Sports mended quantity and quality of exercise for developing and main- 2003;5:332–336. taining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Med Sci Sports Exerc 1998;30:975–991. 139. Blair SN, Kohl HW III, Paffenbarger RS, et al: Physical fitness and all-cause mortality: A prospective study of healthy men and 115. Thompson PC, Buchner D, Pina IL, et al. Exercise and physical women. JAMA 1989;262:2395–2401. activity in the prevention and treatment of atherosclerotic cardio- vascular disease: A statement from the American Heart 140. Ekelund LG, Haskell WL, Johnson JL, et al: Physical fitness as a Association Council on Clinical Cardiology (Subcommittee on predictor of cardiovascular mortality in asymptomatic North Exercise, Rehabilitation, and Prevention) and the Council American men. The Lipid Research Clinics Mortality Follow-up. on Nutrition, Physical Activity, and Metabolism (Subcommittee N Engl J Med 1988;319:1379–1389. on Physical Activity). Circulation 2003;107:3109–3116. 141. Roger VL, Jacobsen SJ, Pellikka PA, et al: Prognostic value of 116. Centers for Disease Control and Prevention. Prevalence of seden- treadmill exercise testing: A population-based study in Olmsted tary lifestyle behavioral risk factor surveillance system, United County, Minnesota. Circulation 1998;98:2836–2841. States 1991. MMWR Morb Mortal Wkly Rep 1993;42:576–579. 142. Goraya TY, Jacobsen SJ, Pellikka PA, et al: Prognostic value of 117. Morris JN, Crawford MD: Coronary heart disease and physical treadmill exercise testing in elderly persons. Ann Intern Med activity. BMJ 1958;2:1485–1496. 2000;132:862–870. 143. Snader CE, Marwick TH, Pashkow FJ, et al: Importance of esti- mated functional capacity as a predictor of all-cause mortality
C H A P T E R 1 3 Effect of Exercise on the Heart and the Prevention of Coronary Heart Disease 459 among patients referred for exercise thallium single-photon emis- 163. Waller BF, Roberts WC: Sudden death while running in condi- sion computed tomography: Report of 3,400 patients from a sin- tioned runners aged 40 years or over. Am J Cardiol 1980;45: gle center. J Am Coll Cardiol 1997;30:641–648. 1291–1300. 144. Mora S, Redberg RF, Cui Y, et al: Ability of exercise testing to pre- dict cardiovascular and all-cause death in asymptomatic women. 164. Thompson PD, Stern MP, Williams P, et al: Death during jogging A 20-year follow-up of the Lipid Research Clinics Prevalence or running (in California). JAMA 1979;242:1265–1267. Study. JAMA 2003;290:1600–1607. 145. Balady GJ, Larson MG, Vasan RS, et al: Usefulness of exercise testing 165. Waller BF: Exercise-related sudden death in young (≤30 years) in the prediction of coronary disease risk among asymptomatic and old (>30 years) conditioned subjects. Cardiovasc Clin persons as a function of the Framingham risk score. Circulation 1985;15:9–73. 2004;110:1920–1925. 146. Leon AS, Jacobs DR, DeBacker G, et al: Relationship of physical 166. Thompson PD, Funk EJ, Carleton RA, Sturner WQ: Incidence of characteristics of life habits to treadmill exercise capacity. Am J death during jogging in Rhode Island from 1975 through 1980. Epidemiol 1981;113:653–660. JAMA 1982;247:2535–2538. 147. Peters RK, Cady LD, Bischoff DP, et al: Physical fitness and subse- quent myocardial infarction in healthy workers. JAMA 1983;249: 167. Vander L, Franklin B, Rubenfire M: Cardiovascular complications 3052–3056. of recreational physical activity. Physician Sports Med 1982; 148. Blair SN, Cooper KH, Gibbons LW, et al: Changes in coronary 10:89–98. heart disease risk factors associated with increased treadmill time in 753 men. Am J Epidemiol 1983;118:352–359. 168. Koplan JP: Cardiovascular deaths while running. JAMA 149. Brill PB, Burkhalter HE, Kohl HW, et al: The impact of previous 1979;242:2578–2579. athleticism on exercise habits, physical fitness, and coronary heart disease risk factors in middle-aged men. Res Q Exerc Sport 169. Morales AR, Romanelli R, Boucek RJ: The mural left anterior 1989;60:202–215. descending coronary artery, strenuous exercise and sudden death. 150. Lie H, Mundal R, Erikssen J: Coronary risk factors and incidence Circulation 1980;62:230–237. of coronary death in relation to physical fitness. Seven year follow-up study of middle-aged and elderly men. Eur Heart J 170. Noakes TD, Opie LH, Rose AG, et al: Autopsy-proved coronary ath- 1985;6:147–157. erosclerosis in marathon runners. N Engl J Med 1979;301:86–89. 151. Williams PT: Physical fitness and activity as separate heart disease risk factors: A meta-analysis. Med Sci Sports Exerc 2001;33: 171. Noakes TD: Heart disease in marathon runners: A review. Med Sci 754–761. Sports Exerc 1987;19:187–194. 152. Blair SN, Jackson AS: Physical fitness and activity as separate heart disease risk factors: A meta-analysis. Med Sci Sports Exerc 172. Virmani R, McAllister HA: Coronary heart disease at young age: A 2001;33:754–761. report of 187 autopsy patients who died of severe coronary athero- 153. Myers J, Kaykha A, George S, et al: Fitness versus physical activ- sclerosis. Cardiovasc Rev Rep 1984;5:799–809. ity patterns in predicting mortality in men. Am J Med 2005;117:912–918. 173. Virmani R, Robinowitz M: Cardiac pathology and sports medicine. 154. Blackburn H, Jacobs DR: Physical activity and the risk of coronary Hum Pathol 1987;18:493–501. heart disease. N Engl J Med 1988;319:1217–1219. 155. Hein HO, Suadicani P, Gyntelberg F: Physical fitness or physical 174. Opie LH: Sudden death and sport. Lancet 1975;306:263–266. activity as a predictor of ischaemic heart disease? A 17-year 175. Northcote RJ, Evans ADB, Ballantyne D: Sudden death in squash follow-up in the Copenhagen Male Study. J Int Med 1992;232: 471–479. players. Lancet 1984;323:148–151. 156. Powell KE, Thompson PD, Caspersen CJ, Kendrick JS: Physical 176. Corrado D, Thiene G, Nava A, et al: Sudden death in young com- activity and the incidence of coronary heart disease. Annu Rev Public Health 1987;8:253–287. petitive athletes: Clinicopathologic correlations in 22 cases. Am J 157. Steffen-Batey L, Nichaman MZ, Goff DC, et al: Change in level of Med 1990;89:588–595. physical activity and risk of all-cause mortality or reinfarction. 177. Corrado D, Basso C, Schiavon M, Thiene G: Screening for hyper- The Corpus Christi Heart Project. Circulation 2000;102: trophic cardiomyopathy in young athletes. N Engl J Med 2204–2209. 1998;339:364–369. 158. Fletcher GF, Blair SN, Blumenthal J, et al: Statement on exercise. 178. Maron BJ, Epstein SE, Roberts WC: Causes of sudden death in Benefits and recommendations for physical activity programs for competitive athletes. J Am Coll Cardiol 1986;7:204–214. all Americans. A statement for health professionals by the com- 179. Maron BJ, Shirani J, Poliac LC, et al: Sudden death in young com- mittee on exercise and cardiac rehabilitation of the council on petitive athletes. Clinical, demographic, and pathological profiles. clinical cardiology. Circulation 1996;94:857–862. JAMA 1996;276:199–204. 159. National Institutes of Health: Consensus development conference 180. Van Camp SP, Bloor CM, Mueller FO, et al: Nontraumatic sports statement on physical activity and cardiovascular health. death in high school and college athletes. Med Sci Sports Exerc Bethesda, Md, 1995. 1995;27:641–647. 160. Damush TM, Stewart AL, Mills KM, et al: Prevalence and corre- 181. Mittleman MA, Maclure M, Tofler GH, et al: Triggering of acute lates of physician recommendations to exercise among older myocardial infarction by heavy physical exertion. Protection adults. J Gerontol A Biol Sci Med Sci 1999;54:M423–M427. against triggering by regular exertion. N Engl J Med 1993;329: 161. Bassler TJ: Marathon running and immunity to atherosclerosis. 1677–1683. Ann NY Acad Sci 1977;301:579–592. 182. Willich SN, Lewis M, Lowel H, et al: Physical exertion as a trigger 162. Siscovik DS, Weiss NS, Fletcher RH, Lasky T: The incidence of pri- of acute myocardial infarction. N Engl J Med 1993;329: mary cardiac arrest during vigorous exercise. N Engl J Med 1684–1690. 1984;311:874–877. 183. Cobb LA, Weaver D: Exercise: A risk for sudden death in patients with coronary artery disease. J Am Coll Cardiol 1986;7:215–219. 184. Hinkle LE, Whitney LA, Lehman EW, et al: Occupation, educa- tion, and coronary heart disease. Science 1968;161:238–246. 185. De Loes M, Goldie I: Incidence rate of injuries during sport activ- ity and physical exercise in a rural Swedish municipality: Incidence rates in 17 sports. Int J Sports Med 1988;9:461–467. 186. Knochel JP: Catastrophic medical events with exhaustive exercise: “White collar rhabdomyolysis.” Kidney Int 1990;38:709–719. 187. Leon AS: Physiological interactions between diet and exercise in the etiology and prevention of ischaemic heart disease. Ann Clin Res 1988;20:114–120. 188. Myers J: Exercise and cardiovascular health. Circulation 2003;107:e2–e5.
CHAPTER fourteen Cardiac Rehabilitation INTRODUCTION professionals, educational videos, and pamphlets can begin the patient’s education. Iatrogenic Cardiac rehabilitation developed in the 1960s as a deconditioning is no longer a problem, because treatment for patients who had sustained a myocar- there are no medical concerns about beginning dial infarction (MI). Before the 1970s, the patient a walking program very early. Psychological reha- who had an MI was almost completely immobilized bilitation takes place in the doctor’s office, for 6 weeks or more and was even washed, shaved, along with prescribing exercise and education. and fed by others in order to keep the work of the Certainly all patients do not need all rehabilitative heart to a minimum (Table 14-1). It was thought interventions, but exercise programs, educational that this approach provided the heart with the sessions, group therapy, and psychological and opportunity to form a firm scar. The patient was vocational counseling are available in most com- also told not to expect to be able to return to a nor- munities, and nearly all patients benefit in some mal life. These were incorrect beliefs, particularly way from the various components of comprehen- with an uncomplicated MI. Prolonged immobiliza- sive rehabilitation. tion not only did not speed healing but exposed the patient to the additional risks of venous thrombosis, Hospital admission for an acute MI or an inter- pulmonary embolism, muscle atrophy, lung infec- vention to avoid its progression is a stressful expe- tion, and deconditioning. Equally serious was the rience, one with a powerful impact. But it must be psychological result of such an approach, which remembered that hospital discharge, although often led to psychological impairment. less dramatic, can be equally stressful after the patient has relied on the highly protective hospi- Today, the physician’s approach to the patient tal support systems. Discharge into an uncertain with acute MI has changed dramatically.1 The future and to home and work, where one is con- infarct is often interrupted by thrombolysis and/ sidered disabled, can be as harmful to one’s self- or coronary interventions that leave the heart rel- esteem as the acute event itself. The healthcare atively undamaged and able to withstand future system is faced with the difficult task not only of ischemic stresses. When the infarct completes supervising the physical recovery of the patient, because of delay in treatment or confounding ill- but also of maintaining morale, providing educa- nesses, a relatively brief period of time monitored tion, helping the family cope and provide support, by the advanced technology in the coronary care and facilitating the return to a gratifying lifestyle. unit is followed by early mobilization, sitting at the Cardiac rehabilitation can be considered as the bedside, graduated exercise and, in the uncompli- conservation of human life. Its goal is to restore cated case, discharge of the patient from the hospi- the patient to optimal physiologic, psychological, tal in less than a week. This policy has been shown and vocational status. by randomized trials to be safe from the point of view of cardiac complications. In-patient rehabil- Though the death rate for coronary heart dis- itation is brief, but interaction with healthcare ease (CHD) has been decreasing steadily since the mid-1960s, it still remains the leading cause of 461
462 E X E R C I S E A N D T H E H E A R T TA B L E 1 4 – 1 . Previous recommendations for emergency PCI as superior to thrombolysis.4 bed rest following acute myocardial infarction from Temporal comparison studies have suggested a con- major cardiology textbooks temporary reduction in mortality due to modern therapies and prevention.2,5 The impact of the 30% Lewis T (Diseases of the Heart. 8 wk of bed rest reduction in mortality with implantable defibrilla- New York, Macmillan, 1937) 4 wk of bed rest tors in patients with history of MI with LV dys- White PD (Heart Disease, 3rd ed. 3–6 wk of bed rest function has not even been factored in yet. New York, Macmillan, 1945) Wood P (Diseases of the Heart and 2–3 wk of Disability Due to MI Circulation, 2nd ed. London, minimum bed rest Eyre and Spottiswoode, 1960) 2 wk of bed rest Cardiovascular diseases, largely atherosclerotic, Friedberg CK (Diseases of the are also the leading cause of activity limitation Heart, 3rd ed. Philadelphia, and disabled worker benefits in the United States W.B. Saunders, 1996) and are the fourth leading cause of days lost Wood P (Diseases of the Heart from work. In fact, coronary artery disease (CAD) and Circulation, 3rd ed. London, alone is responsible for almost one out of five Eyre and Spottiswoode, 1968) disability allowances paid by the Social Security Administration. However, the total economic death in the United States.2 Four deaths in every impact of disability related to cardiovascular dis- 10 are due to cardiac disorders and of these, 90% eases results from the combination of Social can be attributed to CHD. The four distinct clinical Security benefits, welfare support, disability insur- manifestations of CHD are primary cardiac arrest, ance income, unemployment compensation, loss stable angina pectoris, acute coronary syndromes of taxable revenue, and reduced worker produc- (ACS),1 and acute MI. The resting ECG is critical tivity related to these diseases. From a purely eco- to guiding therapy with ST elevation indicating the nomic standpoint, it is essential that patients with prompt application of thrombolysis or percuta- CAD be rehabilitated as quickly and efficiently as neous coronary interventions (PCI) and ST depres- possible in order to enable their return to remu- sion requiring antiplatelet drugs (see Fig. 9-1). nerative employment. Just as important, however, is the psychosocial impact of heart disease, which Each year 900,000 people in the United States cannot be measured in dollars lost. Clearly, experience an acute MI. Of these, roughly 225,000 improved quality of life, including lessened depres- die, including 125,000 who die before obtaining sion and an expedient return to preillness social medical care. The fatality rate in MI patients is roles within the family and community, should be temporally related to onset. The risk of death is another important goal in the effective rehabilita- highest within the first 24 hours of onset of signs tion of patients with heart disease. and/or symptoms and declines throughout the following year. Following the onset of a first-time With the addition of thrombolysis and acute MI in middle-aged males, 30% to 50% die within catheter interventions to MI treatment, disability 30 days and 85% of these deaths occur within the incurred by an MI has decreased. Today’s standard first 24 hours. Those patients with a first-time MI practice is that 85% or more of MI patients undergo who actually reach a hospital alive have a 10% to cardiac catheterization. Because of the functional 18% risk of dying before discharge. The mortality benefits observed in cardiac rehabilitation, physi- thereafter falls from an annualized rate of 9% over cians have extended rehabilitation services to other months 2 through 6, to 4% for months 7 through groups of patients. These patients include those 30, to 3% over the next 3 years. Other studies have who have undergone PCI, coronary artery bypass suggested a mortality rate of 11% in the first grafting (CABG), pacemakers for cardiac resyn- 3 months after hospital discharge and lower rates chronization, transplantation, and valve surgery. thereafter. In comparison to standard medical Other patients for consideration include those therapy, thrombolytic therapy is associated with a limited by angina or heart failure or whose heart highly significant 20% reduction in 35-day mor- disease is complicated by additional illnesses, par- tality among patients with acute MI and ST ele- ticularly diabetes, the metabolic syndrome, and vation, corresponding to an overall reduction of renal insufficiency. We begin this chapter by review- 21 deaths per 1000 patients treated. All of these ing the pathophysiology of MI as it relates to statistics are less meaningful than in the past due to rehabilitation. two changes in healthcare: (1) the use of troponin to define MI3 and ACS and (2) data supporting
C H A P T E R 14 Cardiac Rehabilitation 463 PATHOPHYSIOLOGY OF AN ACUTE the total size rather than transmural extent of MYOCARDIAL INFARCTION underlying MI. Infarct Severity Risk Prediction MIs basically are divided into two groups: those The pathophysiologic determinates of prognosis that evolve Q waves and result in larger area/ are: (1) the amount of viable myocardium and transmural myocardial cell death, and those that (2) the amount of myocardium in jeopardy. do not evolve Q waves and only result in smaller Inferences can be made regarding these two factors area/subendocardial cell death.6 Subendocardial clinically if a patient has had CHF, or cardiogenic MI cannot be localized, whereas transmural MI shock, and continued chest pain, or ischemia. can be roughly localized by the Q-wave pattern. Cardiac catheterization assesses this by ejection Attempts have been made to judge the size or sever- fraction and the number of vessels occluded. The ity of an MI electrocardiographically by Q-wave and clinical findings manifested by abnormalities R-wave scores, and body surface mapping has even related to these two factors are the basis for several been used, but these methods only provide rough indices that have been used to predict risk. These estimates. In general, the greater the number of apply when the infarction damage has not been areas with Q waves and the greater the R-wave loss, averted by an intervention. Clinical data remain the larger the MI. Non-Q-wave MIs are not often very useful in triaging patients with regard to the associated with complications such as conges- necessary length of stay in the hospital. Patients tive heart failure (CHF) or shock; however, they without these criteria, that is, those with uncom- can be complicated, particularly when a prior MI plicated MIs, can be discharged within 3 to 5 days, has occurred or the patient has diabetes. With an while those with these criteria require longer hos- uncomplicated MI, prognosis is particularly good pitalization and closer observation. If the infarction if the patient does not have a prior MI or a decreased has been averted, the completeness of the revas- ejection fraction. Because more myocardium has cularization of the intervention determines the survived, patients with non-Q-wave MIs are more outcome. likely to suffer ischemic events. Anterior Q-wave MIs usually are larger than inferior infarcts and Healthcare professionals must be able to advise are more likely to be associated with CHF and car- post-MI patients as to what they should or should diogenic shock. Anterior infarcts are also more not do to improve their prognosis. One strategy likely to cause aneurysms and a greater decrease has been to identify high-risk patients by using var- in ejection fraction. Surprisingly, however, in ious clinical markers and test results.9,10 Clinical follow-up studies, anterior Q-wave MIs have a sim- markers that have indicated high risk include: ilar or only slightly poorer prognosis than inferior prior MI, CHF, cardiogenic shock, tachycardia, Q-wave MIs.7 Fifteen percent of patients with continued chest pain, older age, stroke or transient Q-wave MIs lose their Q waves over the following ischemic attack (TIA), and complicating illnesses. year but still have the same prognosis as those Procedures used to determine risk, with some suc- who do not lose their Q waves. cess, have included the chest x-ray, routine ECG, ambulatory monitoring, radionuclide cardiac tests, Using magnetic resonance imaging (MRI) to exercise testing, and most recently, cardiac bio- determine the pathologic basis of Q-wave and markers.11 The assumption has been that patients non-Q-wave MI, a total of 100 consecutive patients at high risk should be considered for intervention; with documented previous MIs were recently these interventions are CABG and PCI. Because of studied.8 Subendocardial damage was associated easy access to these procedures, nearly all patients with Q waves about a third of the time, while trans- with history of MI undergo cardiac catheteriza- mural damage was associated with Q waves only tion prior to discharge, particularly because PCI about two thirds of the time. As MI size and num- is currently superseding thrombolysis.12 Exercise ber of transmural segments increased by MRI, the testing is now less often used to decide who needs probability of Q waves on the ECG increased. These cardiac catheterization, because catheterization is findings did not hold for lateral MI. In a multivari- the clinical norm. In addition, success of PCI is ate model, the transmural extent of MI was not an being promoted as superior to exercise testing independent predictor of Q waves when total size and clinical risk status for providing criteria for of MI was removed. The Q-wave/non-Q-wave early discharge after MI. distinction can be useful, but it is determined by
464 E X E R C I S E A N D T H E H E A R T Morbidity and mortality in postinfarc- Animal Experiments of Training tion patients who have complicated courses are Early After Infarction much higher than in those with uncomplicated MIs. Diabetes doubles the mortality with any Hammerman et al13 designed a study to evaluate type of MI. The criteria for a complicated MI the effects of early exercise on late scar formation are listed in Table 14-2. The progressive am- in a rat model. After occlusion of the proximal bulation program should be delayed until such left coronary artery, infarct extent was assessed individuals reach an uncomplicated status, and 24 hours later by ECG criteria. The rats were even then progressive ambulation should be divided into two groups: eight were subjected to slower. daily graded swimming for up to 45 minutes a day for a week, followed by 2 weeks of no swimming; CARDIAC REHABILITATION seven served as a control group. Twenty-two days after coronary occlusion, the hearts of both groups Early Ambulation were excised and wall thickness determined histo- logically. A ratio for transmural infarcts was Prior to 1960, patients with acute MI were obtained from multiple measurements by dividing thought to require prolonged restriction of phys- scar thickness by noninfarcted septal wall thick- ical activity. Patients were often kept at strict bed ness. In the exercise group, there was marked scar rest for 2 months with all activities performed by thinning. Infarct extent was similar in both nursing personnel. The concern was that physical groups. They concluded that short-term swim- activity could lead to complications such as ven- ming during the first week after an MI had effects tricular aneurysm formation, cardiac rupture, heart on scar formation when assessed 2 weeks later. A failure, dysrhythmias, reinfarction, or sudden death. study by Kloner and Kloner,14 with rats forced to Hospitalization could last for 3 to 4 months with swim 7 days post-MI, reported similar results. limitations of activities for at least 1 year. Table 14-1 Studies performed more recently by Gaudron summarizes the recommendations for bed rest in et al15 and Oh et al16 demonstrated that training some of the major cardiology textbooks in this groups of rats, shortly after experimentally induced century. This approach was based on pathologic infarctions, showed evidence of abnormal ventricu- studies that indicated that at least 6 weeks were lar remodeling, including dilation, shape distortion, required for necrotic myocardium to form a firm and scar thinning. Hochman and Healy17 performed scar, and on the increased prevalence of cardiac similar experiments and found no signs in their rupture reported among patients who infarcted in rats of myocardial thinning or aneurysm formation. mental hospitals, where bed rest could not be However, the relevance of rats forced to swim in enforced. this manner to the clinical situation is uncertain. TA B L E 1 4 – 2 . The presence of any one or more The many animal studies of the 1950s and 1960s of the following criteria classify a myocardial were followed by controlled clinical studies of early infarction as complicated mobilization, and a greater incidence of death or other complications was not observed among Prior myocardial infarction patients mobilized early in the period following an Continued cardiac ischemia (pain, late enzyme rise) MI compared to those who remained longer in Left ventricular failure (congestive heart failure, new bed rest. In fact, it has been widely recognized for murmurs, chest x-ray changes) over 30 years that bed rest is associated with many Shock (blood pressure drop, pallor, oliguria) negative physiologic consequences (see in the fol- Important cardiac dysrhythmias (PVCs greater than lowing sections). The promising results of these 6/min, atrial fibrillation) studies led to recommendations of gradual mobi- Conduction disturbances (bundle branch block, AV lization during the early stages following an MI. In block, hemiblock) many patients, the major cause of decreased exer- Severe pleurisy or pericarditis cise capacity is enforced bed rest. The exercise pre- Complicating illnesses scription for MI patients in the coronary care unit Marked creatine kinase rise without a noncardiac can avoid iatrogenically induced deconditioning. explanation Age >75 yr Chair Treatment Stroke or transient ischemic attack A revolutionary approach to treatment occurred in the 1940s, when Levine and Lown18 recommended
C H A P T E R 1 4 Cardiac Rehabilitation 465 “chair treatment” for the patient that had experi- venous thrombosis through a loss of blood volume, enced an MI. This emphasized the benefits of the in which plasma loss exceeds red blood cell mass sitting versus the supine position for increasing loss. Pulmonary function is decreased, and the peripheral venous pooling and reducing preload patient can be in negative nitrogen and calcium on the myocardium. Levine and Lown theorized balance. correctly that such a reduction would lead to a decrease in resting left ventricular wall tension and The question has been raised as to whether the to a decrease in myocardial oxygen demand, in deleterious hemodynamic effects of bed rest, addition to decreasing the risk of thrombosis and including decreased exercise capacity, are due to pulmonary embolism. inactivity or to the loss of the upright exposure to gravity. There are at least four reasons supporting Physiologic studies performed since then have the concept that much of these alterations are documented the hemodynamic alterations caused caused simply by the latter: (1) supine exercise does by bed rest and deconditioning. After prolonged bed not prevent the deconditioning effects of being in rest, tachycardia and hypotension are common bed; (2) there is both a less dramatic and a slower upon standing. They are most likely caused by decline in maximal oxygen uptake with chair rest alterations in the baromotor reflexes and by the than with bed rest; (3) there is a greater decrease hypovolemia that occurs with bed rest. Clearly, in maximal oxygen uptake after a period of bed rest the disability secondary to most MIs is due both to measured during upright exercise versus supine bed rest and to myocardial dysfunction. The spon- exercise, and (4) a lower body positive-pressure taneous hemodynamic improvement usually seen device decreases the deconditioning effects of bed is due both to improving function (scar formation rest. Clearly, intermittent exposure to gravitational and possibly compensatory hypertrophy) and to a stress during the bed rest stage of hospital conva- return to normal activities. lescence from surgery or MI obviates much of the deterioration in cardiovascular performance that Bed Rest: Lack of Activity versus can follow these events. Previous efforts to limit Effects of Gravity the decrease in functional capacity after an acute MI or surgery have emphasized low-level activities, There are now well-defined hemodynamic alter- but studies suggest that simple exposure to grav- ations due to deconditioning from bed rest. These itational stress substantially accomplishes this.20 are outlined in Table 14-3. The classic 1968 study by Saltin et al19 demonstrated that young men Progressive Activity maintained at bed rest for 3 weeks demonstrated a 20% to 25% decrease in maximal oxygen uptake. A consideration often forgotten when dealing Other than decreased functional capacity, prolonged with older patients or those with complicating ill- bed rest can result in orthostatic hypotension and nesses is the level of activity that they maintained prior to their MI. If a patient was physically limited TA B L E 1 4 – 3 . Physiologic consequences of prior to their event, then the plan for progressive prolonged bed rest ambulation must be modified. It is generally not practical to expect a patient to be more physically Loss of muscle mass, strength, and endurance active after an MI than before—unless the patient Decreased plasma and blood volume was previously limited by angina that disappeared Decreased ventricular volume later. It is important to assess the exercise capacity Increased hematocrit and hemoglobin and activity level that existed prior to the MI. Diuresis and natriuresis Venous stasis In addition to the oxygen cost and the heart Bone demineralization rate achieved during activity, the duration of the Increased heart rate at rest and submaximal levels of activity must be considered. The effect of pro- activity longed exercise on myocardial scar formation in Decreased resting and maximum stroke volume the acute recovery phase has not been carefully Decreased maximum cardiac output studied, but it is known that during prolonged Decreased maximal oxygen uptake steady-state dynamic exercise, heart rate increases, Increased venous compliance myocardial contractility declines, and left ventric- Increased risk of venous thrombosis and ular volume increases. It is apparent then, that even thromboembolism though certain work levels can be achieved by Decreased orthostatic tolerance a patient, prolonged exercise should be avoided Increased risk of atelectasis, pulmonary emboli in the acute recovery phase. Probably the safest
466 E X E R C I S E A N D T H E H E A R T recommendation is to tell patients not to fatigue Hayes et al24 studied 189 patients with an themselves and to limit the duration of exercise uncomplicated MI selected at random for early or by their fatigue level and perceived exertion. late mobilization and discharge from the hospital. Patients were admitted to the study once they had Postdischarge activity recommendations have spent more than 48 hours in a coronary care unit, had little basis for their enforcement. Activities if they were free of pain, and if they showed no evi- involving a return to work, to driving, and/or to sex- dence of heart failure or significant dysrhythmias. ual activity have been based on clinical judgments One group of patients was mobilized immediately rather than physiologic assessments. Because of and discharged home after a total of 9 days in this, physicians have left much of this up to their the hospital, and the second group was mobilized patients—allowing them to see how they respond on the 9th day and discharged on the 16th day. in terms of symptoms—rather than the traditional, Outpatient assessment was carried out 6 weeks very conservative approach, which can foster inva- after admission. No significant differences were lidism. These decisions should be made considering observed between the groups in terms of morbidity the consequence of the coronary event (ischemia or mortality. or symptoms of congestive failure, or dysrhyth- mias) and the nature of the activities (manual labor In a randomized study, Bloch et al25 studied versus deskwork, light driving versus congested the effects of early mobilization after uncompli- freeway driving, and sex with an established partner cated MI. For the study, 154 patients younger versus other relationships). than 70 years of age who were hospitalized for an acute MI and had no complications on day 1 or The impetus for mobilizing patients who had day 2 were randomly assigned to two treatment an MI or underwent surgery evolved from clinical groups. In the early mobilization group, patients studies assessing the physiologic effects of activity were treated by a physical therapist with a pro- in these patients in the 1960s and 1970s. In 1961, gressive activity program that began on day 2 or Cain et al21 reported on the use of a progressive day 3 after infarction. In the control group, the activity program for acute MI patients. They had dif- patients underwent the traditional hospital regi- ficulty getting this report accepted for publication men of strict bed rest for 3 or more weeks. The because the approach was considered dangerous. mean duration of hospitalization was 21 days for They reported 335 patients with an uncompli- active patients and 33 days for the control group. cated MI who were at least 15 days postinfarction. The follow-up period ranged from 6 to 20 months, The patients previously had been restricted to bed, with an average of 11 months. There were no sig- chair, and commode. The ECG was monitored after nificant differences between the two groups with the patient performed activities such as climbing regard to hospital or follow-up mortality; rates of stairs and walking up a grade. In 1964, Torkelson22 reinfarction, dysrhythmias, heart failure, angina reported results in 10 patients with an uncompl- pectoris, or ventricular aneurysm; or on the results icated MI. On the sixth week of the in-hospital of an exercise test. On follow-up examination, there rehabilitation program, a low-level treadmill test was greater disability in the control group than in was performed using 1.7 mph at a 10% grade. He the active group. concluded that the treadmill test was a valuable procedure for the documentation of the specific Sivarajan et al26 have reported on the effects of exercise response of patients recovering from an early initiation of supervised exercises in prevent- acute MI. ing deconditioning after an acute MI. Eighty-four patients were randomized to a control group and Later publications did not include ECG moni- 174 to an exercise group. The exercise program toring as part of progressive ambulation. Instead, began an average of 4.5 days after admission. The generalized statements as to the activities on each mean discharge was 10 days after admission for post-infarct day were made for all patients, rather both groups. There were no differences between than individualized activity progression. Sivarajan the two groups in the clinical, hemodynamic, or et al23 described 12 patients with an acute MI whose ECG responses to a low-level treadmill test per- symptoms, signs, and hemodynamic and ECG formed on the day before hospital discharge. In addi- responses during and after three activities (sitting tion, there were no significant differences between upright, walking to the toilet, and walking on a the two groups in the incidence of cardiovascular treadmill) were assessed. Studies of these activi- complications or death. By the time this well- ties were done at 3, 6, and 10 days after infarction. designed study was funded, the standard of com- They concluded that successful performance of munity medical care in Seattle included early these three activities provided useful criteria for ambulation and discharge. Therefore, the control discharge.
C H A P T E R 1 4 Cardiac Rehabilitation 467 group received treatment that was hardly differ- American Association of Cardiovascular and ent from the exercise group. In addition, for safety Pulmonary Rehabilitation (AACVPR),29 were reasons, the sicker patients who most needed described in Chapter 13, and should be specifically rehabilitation were excluded from this study. Six tailored to each individual patient. patients needed cardiac surgery prior to discharge in the exercise group, but none required it in the When prescribing exercise, two basic physio- control group, which can be explained by chance logic principles should be considered. Myocardial distribution (failure of randomization) rather than oxygen consumption is the amount of oxygen by the mild exercises employed. These three ran- required by the heart to maintain itself and to do domized studies of patients with uncomplicated the work of pumping blood to the other organs. It infarctions have provided strong evidence that the cannot be measured directly without catheters but risks of early ambulation are minimal and that pro- can be estimated by the product of systolic blood gressive mobilization during the early stages of an pressure (SBP) and heart rate (double product). The acute MI should be recommended. higher the double product, the higher the myocar- dial oxygen consumption, and vice versa. Patients Exercise Testing Before usually experience their angina at the same double Hospital Discharge product, unless affected by other factors such as catecholamine level, left ventricular end-diastolic The low-level exercise test early after an acute MI volume, oxyhemoglobin dissociation as affected by (from 3 days to 3 weeks) has been shown to be acid-base balance, and coronary artery spasm. safe. Today, it is a standard part of the treatment for MI patients in many hospitals. This test has many The second principle concerns ventilatory oxy- benefits, including clarification of the response to gen uptake (VO2), which is the amount of oxygen exercise and work capacity, determination of an taken in from inspired air by the body to maintain exercise prescription, and recognition of the need itself and to do the work of muscular activity. for additional medical therapy or surgery. It appears Measuring VO2 requires the collection of expired to have prognostic value, can have a beneficial psy- air, gas analyzers, and a skilled technician (these chological impact on recovery, and is an effective techniques are detailed in Chapter 3). However, part of rehabilitation. The applications and inter- VO2 can be estimated from knowing the work- pretation of exercise testing early after an acute load on a treadmill or cycle ergometer. Because MI are detailed in Chapter 9. the body’s mechanical efficiency is relatively con- stant, estimates of the oxygen cost of various activ- EXERCISE PRESCRIPTION ities without using gas analysis can be applied between individuals. There are many tables giving Exercise training can be an important part of car- the approximate oxygen cost of different activities. diac rehabilitation for returning a patient to his Because oxygen uptake is equal to arteriovenous or her formerly active lifestyle, or as functional of a oxygen difference (a-VO2 difference) multiplied by lifestyle as possible, after an acute cardiac event. An the cardiac output, and a-VO2 difference widens early definition of cardiac rehabilitation was pro- by roughly the same amount between individuals at vided by the World Health Organization (WHO)27 maximal exercise, maximal oxygen uptake is some- as “the sum of activities required to ensure them times considered an approximation of maximal car- the best possible physical, mental, and social condi- diac output. However, patients with diseased hearts tions so that they may, by their own efforts, resume often have a lower cardiac output, a wider a-VO2 as normal a place as possible in the life of the com- difference, and a lower VO2 than normal subjects munity… and that…rehabilitation cannot be performing the same submaximal workload. regarded as an isolated form of therapy, but must be integrated into the whole treatment of which it Another important physiologic concept, which constitutes only one facet.” The explicit principles of is an important consideration for exercise pre- exercise prescription, including frequency, intensity, scription, is the type of work performed. Dynamic duration and exercise mode, absolute and relative work (bicycling, running, jogging) requires the contraindications to exercise, and guidelines for movement of large muscle groups and requires a terminating exercise, as outlined by the American high blood flow and increased cardiac output. This College of Sports Medicine (ACSM)28 and the movement is rhythmic, offers little resistance to flow, and provides a “pumping” action from alter- nating contraction and relaxation that helps return blood to the heart. The converse type of muscular work is isometric, which involves a relatively con- stant muscular contraction and thus limits the
468 E X E R C I S E A N D T H E H E A R T increase in blood flow. Activities such as lifting a blood pressure. As with more traditional aerobic weight or squeezing a ball are relatively isometric. training programs, circuit weight training pro- For this type of exercise, blood pressure must be grams have been shown to have musculoskeletal increased in order to force blood into the active, and metabolic benefits, including improved insulin contracting muscles. Pressure work demands much sensitivity in diabetics.32 Numerous other investi- more oxygen supply to the heart than flow work, gators have recently made similar observations. and because coronary artery blood flow depends The AACVPR guidelines29 have even outlined rec- upon cardiac output, the myocardial oxygen supply ommendations on weight training for low-risk can become inadequate. In addition, dynamic exer- patients, an activity once thought to be far too cise is more easily controlled or graded, so that dangerous for this population. myocardial oxygen demand can be gradually increased, whereas isometric exercise can increase INTERVENTION STUDIES myocardial oxygen demands very quickly. Thus, appropriate contraindications must be considered There have been hundreds of studies published before patients with cardiovascular disease are ini- documenting the benefits of cardiac rehabilitation tiated into a resistance-exercise training program. in patients recovering from an MI. Landmark stud- ies performed in the late 1970s helped to define However, nearly all forms of exercise require a cardiac rehabilitation as a standard of care and combination of dynamic and isometric contractions provided important data not only on the physio- to one extent or another. When performed properly, logic effects of exercise training, but also data on resistance exercise enhances muscular strength the effects of exercise on cardiovascular mortality and endurance, and has been shown to provide not and reinfarction. These studies are summarized in only cardiovascular benefits, but favorable meta- Table 14-4. A few of the major studies are reviewed bolic adaptations and benefits to bone health as in the following. well. Thus, appropriately guided resistance exercise is widely used today as a complement to aerobic Kallio et al33 were part of a WHO-coordinated exercise in rehabilitation programs. project to assess the effects of a comprehensive rehabilitation and secondary prevention program Circuit Training on morbidity, mortality, return to work, and vari- ous clinical, medical, and psychosocial factors after Kelemen et al30 performed a prospective, random- an MI. The study included 375 consecutive patients ized evaluation of the safety and efficacy of 10 weeks younger than 65 years of age treated for acute MI of circuit weight training in coronary disease from two urban areas in Finland between 1973 and patients aged 35 to 70 years. Circuit weight training 1975. General advice on rehabilitation and second- consisted of a series of weight-lifting exercises using ary preventive measures was given to all patients a moderate load with frequent repetitions. Patients who were discharged from the hospital. On dis- had participated in a supervised cardiac rehabilita- charge, the patients were randomly allocated to an tion program for a minimum of 3 months before intervention or to a control group, both of which the study. Control patients (n = 20) continued with were followed-up for 3 years. Patients in the control their regular exercise, consisting of a walk/jog and group were followed by their own doctors and were volleyball program, whereas the experimental seen by the study team only once a year during group (n = 20) substituted circuit weight training the 3-year follow-up. The program for the inter- for volleyball. No sustained arrhythmias or car- vention group was started 2 weeks after hospital diovascular problems occurred. The experimental discharge. An exercise prescription was determined group significantly increased exercise tolerance from a bicycle test, and for most patients the pro- by 12%, whereas there was no change in the gram was supervised. control patients. Circuit weight training was safe and resulted in significant increases in aerobic After the 3-year follow-up, the cumulative coro- endurance and musculoskeletal strength com- nary mortality was significantly smaller in the pared with traditional exercise used in cardiac intervention group than in the controls (18.6% rehabilitation programs. Sparling et al31 have also versus 29.4%). This difference was mainly due to a demonstrated the safety and efficacy of circuit reduction of sudden deaths in the intervention weight training in patients with cardiovascular group (5.8% versus 14.4%). The reduction was disease. In a 6-month study of 16 men, there was greatest during the first 6 months after infarction. a 22% gain in strength, without an increase in Total mortality was 21.8% in the intervention group and 29.9% in the control group. Although this was
TA B L E 1 4 – 4 . Summary of the early randomized trials of cardiac rehabilitation assessing cardiac events, mortality, or both in patients with coronary disease Population randomized Percent mortality Exclusions % Mean no. Mean Yr follow- Dropouts Return to Re-mI Sudden Cardiac Total Mo entry age up Cntrl Ex work Cntrl Ex Cntrl Ex Investigator Total Cntrls Ex (>Yr) women after mI Cntrl Ex Cntrl Ex 3% 6% 12% 10% Cntrl Ex 53 2 5% 8% 14% 6% 18% 16% Kentala 1972 158 81 77 >65 0 1.75 51 4 46% 33% 36% 5% 8% 14% 10% 14% 14% Wilhelmsen 1975 315 157 158 >57 11 3 52 2.5 35% 30% 57% 21% 18% 7% 4% 29% 19% 22% 18% Palatsi 1976 380 200 180 >65 19 2.5 55 3 15% 12% 0% 0% 14% 10% Kallio 1979 357 187 188 >65 19 3 51 1.5 25% 25% 73% 75% 11% 18% 23% 13% 6% 4% 30% 22% Mayou 1981 129 42 44 >60 0 1 52 3 31% 23% 81% 81% NEHDP 1981 651 328 323 14 52 2.1 7% 5% 4% 4% 7% 5% Carson 1982 303 152 151 0 1.5 48 3.3 45% 46% 59% 66% 7% 8% 2% 4% 14% 8% Ontario 1982 761 371 390 6 56 0.50 13% 15% 10% 9% 7% 10% Sivarajan 1982 172 84 88 >70 0 56 1 17% 10% Bengtsson 1983 171 90 81 >54 0 0.13 51 3.5 6% 17% 4% 2% 10% 4% 7% 10% Carson 1983 303 152 151 >70 20 1.5 55 9 4% 4% 7% 7% 14% 8% Roman 1983 193 100 93 >65 0 1.5 49 5 23% 17% 3% 0% 24% 14% Vermeulen 1983 98 51 47 >70 0 2 53 1 14% 17% 18% 9% 8% 8% 10% 4% Froelicher 1984 146 74 76 1.75 55 0.5 7% 9% 1% 1% 5% 6% 0% 1% Hung 1984 53 23 30 10 4 57 1 45% 7% 9% 47% 35% 3% 0% Hedback 1985 297 154 143 0.75 4.5 16% 5% 3% 4% 8% 9% Marra 1985 167 83 84 >55 0 1.5 57 10 11% 6% 6% 7% Hamalainen 1989 375 187 188 >65 0 2 1 12% 15% 19% 26% 52% 44% DeBusk 1994 585 292 293 >70 0 <1 7% 3% 3% 4% >65 15 0.10 >65 >65 20 >70 21 Cntrl, controls; Ex, exercise; MI, myocardial infarction.
470 E X E R C I S E A N D T H E H E A R T a landmark study, two weak points were (1) that exercise group, and then 50 entered the control more patients in the intervention group than in the group. The final total included 180 patients for control group took antihypertensive medications exercise, including 37 women, and 200 controls, and beta-blockers and (2) that exercise capacity including 34 women. Patients with non-Q-wave measured at 1, 2, and 3 years after acute infarction MIs were treated with bed rest for 3 days, were was similar in both groups. allowed to sit for 1 week, were allowed to walk on the 10th day, and were discharged on the 12th Kentala34 studied 298 consecutive men less than day. Q-wave MI patients had bed rest for 7 days, 65 years of age admitted to the University of Helsinki were sitting after 1 week, were allowed to walk Hospital in 1969 with a diagnosis of acute MI. They on the 14th day, and were discharged on the were divided by the year of birth: controls were from 16th day. odd number years (n = 146) and exercisers were from even number years (n = 152). The average age Exercise training began 10 weeks after the MI of patients in both groups was 53 years. Exclusions and included breathing and relaxation exercises, for controls included 10 with uncertain diagnosis, calisthenics, and walking that progressed to run- 24 who died in the hospital, five who refused or ning in place. Heart rate was at least 70% of the were not informed, four who had other severe dis- maximal rate during daily 30-minute sessions per- ease, and 22 who lived too far away. Exclusions formed at home. Once a month, the patients for the exercise group included 12 with uncertain returned for revision of their exercise program. No diagnosis; 21 who died in the hospital; three who effort was made to change smoking habits. The were not informed; three with other severe diseases, authors concluded that home training was not as and 36 who lived too far away. Eighty-one controls effective as continual supervised programs, but nev- and 77 exercisers were accepted for the study. Of ertheless accelerated recovery of aerobic capacity. the 81 controls, four died, three were hospitalized, Rehabilitation was associated with slightly lower and one refused to particpate, leaving 73 at 1-year total mortality versus controls (14% versus 10%) follow-up. Of the 77 randomized to exercise, five (NS). However, there were no between-group differ- died, three were hospitalized, and one refused to ences in symptoms, smoking habits, serum choles- participate, leaving 69 at 1-year follow-up. Unless terol, or return to work. contraindicated, patients were kept on anticoagula- tion; beta-blockers were avoided. Both groups Wilhelmsen et al36 studied patients who were made their own decisions on tobacco smoking, born in 1913 or later and who were hospitalized and diet information was given. The training group for an MI between 1968 and 1970 in Goteborg, also was urged to increase home activities in addi- Sweden. Patients were randomized to a control tion to the exercise program, especially walking. group (n = 157) or an exercise group (n = 158). Fifteen of the controls and 20 of the exercisers were There were two training sessions weekly for women. The only criterion was an age of 60 years the exercise group, for the exercise group, later or older, but 27% of patients were excluded for increasing to three per week, with 20 minutes of cardiac complications. The two groups were com- warm-up, 20 minutes of exertion (bicycle, rowing, parable for prevalence of hypertension, diabetes stairs), followed by a cool-down phase. The exer- mellitus, treatment with digoxin, smoking status, cise heart rate was targeted to 10 beats less than CHF, and previous MI. The exercise group trained the maximal heart rate from exercise testing. There three times a week for 30 minutes per session. was no difference in morbidity or mortality between Calisthenics, cycling, and running were performed the groups after 1 year. Both groups showed clear at 80% of the maximal age-predicted heart rate. decreases in heart rate for given workloads, and All follow-up treatments were the same except for both groups showed improved maximal workload, the exercise program. After 1 year, the exercise especially in those patients with greater than 70% group showed increased work capacity and lower attendance. Return to work was not influenced by blood pressure, but no difference in blood lipids. training; 68% who worked before their MI returned At 1 year, only 39% continued to come to the hos- to work after 1 year. pital to exercise, whereas 21% trained elsewhere. Initially, compliance to the exercise sessions was Palatsi35 performed a nonrandomized trial of 18%, but it decreased to 63% at 1 year. Smoking 380 patients who were younger than 65 years old, after MI was found to be a significant predictor of and recovering from MI. Patients were excluded fatal recurrent MI. There was also an association if they had locomotive limitations, psychological between smoking cessation and attending the problems, or CHF. The first 100 patients were allo- exercise program. No significant differences were cated to an exercise program and the second 100 seen with respect to cause, type, or place of death. were the controls. The next 50 patients entered the
C H A P T E R 1 4 Cardiac Rehabilitation 471 Wilhelmsen et al concluded that antismoking discontinued, especially because the results were advice and treatment with beta-blockers deserve so encouraging: 1400 patients (more than twice higher priority than exercise training in the sec- the actual sample size) would have been required ondary prevention of MI. to demonstrate a statistically significant reduc- tion in mortality rate in the exercise group if the The National Exercise and Heart Disease reported trend persisted. The patients in the Project (NEHDP) included 651 men post-MI exercise group who suffered a reinfarction had a enrolled in five centers in the United States.37 It lower mortality rate, suggesting that an exercise was a randomized 3-year clinical trial reporting program increases an individual’s ability to sur- the effects of a prescribed supervised exercise vive an MI. program starting 2 to 36 months after MI (80% were more than 8 months’ post-infarction). In The Ontario Study included seven Canadian this study, 323 randomly selected patients per- centers that collaborated in a randomized prospec- formed exercise three times a week that was tive trial.38 The study involved 733 men with prior designed to increase their heart rate to 85% of history of MI who underwent random stratified allo- that achieved during treadmill testing, and 328 cation to either a high- or low-intensity exercise patients served as controls. This study was care- group. Patients were excluded for cardiac failure, fully designed by experts who took 2 years to com- insulin-dependent or uncontrolled diabetes, dias- plete the protocol. An initial low-level exercise tolic hypertension, orthopedic problems, or severe session in both groups, to exclude the faint of lung disease. The two groups were comparable for heart who would not comply with an exercise pro- initial MI, angina, hypertension, type A personal- gram, was surprisingly effective in improving ity, smokers, ex-smokers, and cholesterol level. patient performance. Stratifying variables included: (1) the presence or absence of hypertension, (2) blue versus white col- The 3-year mortality rate was 7.3% (24 deaths) lar employment, (3) presence or absence of angina, in the control group versus 4.6% (15 deaths) in and (4) type A and B personality. The high-intensity the exercise group. Deaths from all cardiovascular group trained by walking or jogging at 65% causes (acute MI, sudden death, arrhythmias, CHF, to 85% of their maximal oxygen uptake for 1 hour cardiogenic shock, and stroke) for the 3-year fol- twice a week. This continued for 8 weeks, after low-up were 6.1% (20 deaths) in the control group which they trained four times a week on their own. versus 4.3% (14 deaths) in the exercise group. The low-intensity group trained once a week with Neither difference was statistically significant. relaxation exercises, volleyball, bowling, or swim- However, when deaths due to acute MI were con- ming for 1 hour. They attempted to keep their sidered as a separate category, the exercise group heart rate at less than 50% of maximal oxygen had a significantly lower rate: one acute fatal MI per uptake. Both groups were encouraged to stop 3 years (0.3%) in the exercise group versus eight smoking and control their weight. Less than 5% fatal MIs (2.4%) in the control group (P < 0.05). The of the low-intensity group regularly exercised vig- rate of all recurrent MI per 3 years, fatal and non- orously. The dropout rate was 47%. The rate of fatal, did not differ significantly between groups— reinfarction in the high-intensity group was 14% 23 cases (7.0%) in the control versus 17 cases versus 13% in the low-intensity group. They (5.3%) in the exercise group. The number of rehos- found that the high- intensity exercise program pitalizations for reasons other than MI was also had similar results to one designed to produce a similar in the two groups (27.4% versus 28.5% minimal training effect and did not reduce the over 3 years). The need for bypass surgery was also risk of reinfarction. similar in both groups—16 controls and 17 exercis- ers underwent surgery during the 3-year period. Bengtsson39 reported on 171 MI patients This study suggests a beneficial effect of cardiac younger than 65 years of age who were random- rehabilitation on morbidity and mortality, but a ized to a control and exercise group. Patients were definitive conclusion was not reached as there excluded for CHF, aortic insufficiency, hepatitis, were insufficient participants due to financial lim- polio, diabetes, new MI, thyroid disorders, stroke, itations and dropouts. or psychological problems. The rehabilitation program consisted of an outpatient examination, Despite the fact that the exercise group demon- counseling, and supervised exercise (large muscle strated 37% and 24% reductions in mortality and group interval training by use of bicycles, calis- reinfarction rates, respectively, this study was not thenics, and jogging for 30 minutes twice a week definitive from a statistical standpoint. However, it for 3 months at 90% of maximal heart rate). There did demonstrate the feasibility of resolving this were no reported differences between groups for important issue. It is unfortunate that it was
472 E X E R C I S E A N D T H E H E A R T age, sex, number of infarcts, highest enzyme level, control patients after completion of the study heart size, number of days in the hospital, num- was highly significant. There was no significant ber of admissions, angina, CHF, arrhythmias, or decrease in mortality for the exercise group except depression or hypochondriasis on the Minnesota for those with an inferior wall MI. Multiphasic Personality Inventory. The authors reported 100% compliance to the program. The Vermeulen et al41 described a prospective exercisers showed lower mean systolic blood pres- randomized trial with a 5-year follow-up. Approx- sure (SBP) at rest and lower diastolic blood pres- imately 1 month after MI, patients underwent a sure at high workloads than controls. Equal symptom-limited exercise test. There was no total percentages of the exercise group and the con- population description, no training description, no trols (74%) returned to work. The exercisers per- dropout rate reported, and no return to work formed 31% heavier work at the end of training described. Both the control and exercise groups and 63% at the end of follow-up. It was concluded received the same dietary advice. They found that that at 1 year, all patients were less physically and rehabilitation did not influence smoking habits but socially active than they were before their MI. lowered serum cholesterol. The 6-week rehabilita- They were more dependent on their relatives than tion program was associated with a 50% decrease before and they had a poor understanding of their in progression of CAD compared to the control illness. The rehabilitation program—including group. The incidence of progression of CAD was exercise, education, counseling, and social meas- significantly decreased in patients smoking less ures during the first 5 months after an acute MI— than 20 cigarettes a day. They concluded that cardiac did not change outcome 8 to 19 months after the rehabilitation is safe and greatly benefits patients MI compared to controls when considering physical with history of MI due to direct effects on myocar- fitness, return to work, psychological factors, and dial perfusion and lowering of cholesterol levels. an understanding of their illness. Roman42 studied 139 patients, including 19 Carson et al40 performed their 31/2 year study in women, who entered into a cardiac rehabilitation a population of 1311 male MI patients. Of these, program. The control and exercise groups were 12.5% died in the hospital, 4% died after dis- comparable for age, sex, and MI location. The charge but prior to follow-up, and 4.8% failed exercise group trained 30 minutes three times a to attend follow-up appointments. Thus, 70% of week at 70% of maximum heart rate for an aver- the original admissions remained. Patient exclu- age of 42 months. At the 9-year follow-up, the sions included being older than 70 years of age or control group had 24 cardiac deaths, including having CHF, cardiac enlargement, lung disease, 15 acute MIs, seven sudden deaths, and two with hypertension, insulin, angina, orthopedic or med- CHF. The trained group had 13 deaths, which ical problems, or personality disorders. After these included seven acute MIs, four sudden deaths, and exclusions, 442 patients were considered suitable two patients with CHF. The mortality rate was and 139 of these declined, leaving 303. These 5.2% for the control group and 2.9% for the reha- patients were accepted into the study and were bilitation group. There were 23 recurrent MIs in randomized to either a control or exercise group. the control group (4.9% per year) and 16 recurrent There was no group difference with regard to site MIs in the rehabilitation group (3.6% per year). of MI, number of MIs, highest enzyme level, smok- There was no difference in the incidence of myocar- ing habits, known diabetes, previous angina or MI, dial ischemia, severe arrhythmias, or cerebrovas- cholesterol levels, family history, left ventricular cular events between the two groups. There was a failure, or occupation. The exercise group trained significant decrease in angina in the exercise in a gym twice a week for 12 weeks at 85% of the group. The overall attendance was 76%, and the exercise test-determined maximal heart rate or dropout rate was 4.1% in the exercise group and until symptoms of angina, shortness of breath, or 3.9% among controls. a poor SBP response occurred. Isometric exercise was avoided. The dropout rate was 17% in the exer- Mayou43 studied 129 men, 60 years of age or cise group and 6% in the controls. Mean age at younger, admitted to the hospital with an MI. death was significantly different in the two groups: They were sequentially allocated to normal treat- 50 years in the exercise group and 57 years in the ment, exercise training, or counseling groups. control group. Return to work was 81% in both The control group received standard in-patient groups, and both groups showed a similar decrease care, advice booklets, and one to two regular clinic in smoking after their MI. They concluded that visits as outpatients. They had no other education, the difference in fitness between the exercise and walking program, or instructions for exercise. The exercise group received the normal treatment plus eight sessions (twice a week) of circuit training in
C H A P T E R 1 4 Cardiac Rehabilitation 473 groups, written reminders, and reviews of their One year following the MI, there was no group results. The counseling group received normal difference in mortality, but the exercise group treatment plus discussion groups, they kept a had a significantly lower rate of nonfatal reinfarc- daily-activity diary, had couples therapy, and had tion, fewer uncontrolled hypertensives, and fewer three to four follow-up sessions. The three groups smokers. were comparable socially, medically, and psycho- logically. Patients excluded were 13 who died and Meta-Analysis one with a stroke. Evaluation was performed after 12 weeks using exercise testing and standard tests The overall benefits of cardiac rehabilitation are of psychological state and social adjustment. now widely accepted. Comprehensive reviews con- There were no differences among the groups in firming these benefits are available.45,46 Because of psychological outcome, physical activity, or satis- the time and expense involved in conducting con- faction with leisure or work. The exercise patients trolled studies with large numbers of patients, few were more enthusiastic about their treatment and such trials have been performed. We are left with achieved higher workloads on exercise testing. At numerous studies showing significant benefits in 18 months, the significant findings included a bet- exercise capacity, and often psychosocial benefits, ter outcome in terms of overall satisfaction, hours but frequently only trends toward improved mor- of work, and frequency of sexual intercourse for bidity and mortality. Meta-analysis has gained pop- the counseled group. The dropout rate was 25% ularity in recent years as a method of combining overall. There was no difference in exercise capacity separate but similar studies, and this approach at 6 weeks, but at 12 weeks there was a nonsignif- has recently yielded some very important informa- icant increase in the exercise group. The groups tion on the efficacy of cardiac rehabilitation. May were similar in terms of return to work, activities, et al47 presented the first such summary of the sexual activity, and ratings of quality of life. There long-term trials in secondary prevention after MI were no group differences in compliance to advice (Table 14-5). Trials reported prior to November on smoking, diet, or exercise. They concluded that 1981 were considered in which both intervention exercise training increased confidence during and follow-up were carried out beyond the time of exercise in the early stages of convalescence, but hospital discharge. Random assignment and at least that the exercise program had little value with a total sample size of 100 patients were required. regard to cardiac performance, daily function, or Total mortality was used whenever possible, in emotional state. order to minimize bias. All patients randomized were included in the mortality estimates to reduce Hedback et al44 performed a retrospective study the bias of differential withdrawal. Effectiveness in Sweden with a control group of 154 patients was calculated by considering the percent reduc- and an intervention group of 143 patients; 23 of tion in deaths that would have occurred if the inter- the controls and 22 of the exercisers were women. vention had been applied to the control group. There was no group difference regarding age, sex, Although few of the interventions resulted in a risk factors for MI, rate of employment, income significant difference, all of them, except for anti- level, MI location or size, arrhythmias, medications, arrhythmic therapy, showed a trend toward efficacy. or heart size on discharge chest x-ray. Exclusions The 19% reduction in death rate from exer- for the training group included severe CHF, arthri- cise training suggests that exercise is as safe and tis, and stroke. Thirty-one declined to enter the effective as other available means of secondary program. Of the 84 patients who began the training prevention. program, 79 completed it. Both groups were treated the same during their acute hospitalization. To circumvent the problem of inadequate sample Training began 6 weeks after MI, following a bicy- sizes, O’Connor et al48 performed a meta-analysis cle test. Training was performed on a bicycle to a of 22 randomized trials of cardiac rehabilitation maximal heart rate of 5 beats below maximal heart involving 4554 patients. They found a 20% reduc- rate as determined during the exercise test. If symp- tion of risk for total mortality, a 22% reduction toms or signs occurred, heart rate was limited to for cardiovascular mortality, and a 25% reduction 15 beats below maximal heart rate. Sessions were in the risk for fatal reinfarction. Oldridge et al49 25 to 30 minutes long. This was done for 4 weeks performed a similar meta-analysis with 10 ran- and then replaced by calisthenics and jogging domized trials including 4347 patients and found plus a home program. Patients with a cholesterol a similar reduction for all-cause death and car- level of 8 mmol/L were referred to a dietician. Beta- diovascular death in the patients undergoing blockers were administered to 60% of the patients.
474 E X E R C I S E A N D T H E H E A R T TA B L E 1 4 – 5 . Mortality associated with randomized intervention trials after myocardial infarction considered epidemiologically valid Intervention Number of Number of Length of % Mortality % Mortality Effectiveness (% studies (with patients follow-up controls intervention reduction, in Antidysrhythmics significant randomized (range of deaths) Lipid-lowering agents difference) means) 10.3 10.8 Anticoagulants 1675 23.6 19.4 −4.6 Platelet-active drugs 6 (0) 19,834 4 mo–2 yr 17.7 13.7 17.8 Beta-blockers 9 (1) 2327 21 mo–11 yr 10.5 9.7 22.6 Exercise 5 (0) 13,298 11.5 8.8 7.6 7 (0) 11,325 2–6 yr 14.7 11.9 23.5 11 (4) 2752 1–3 yr 19 6 (1) 9 mo–2 yr 1–4.5 yr From May GS, Eberlein KA, Furberg CD, et al: Secondary prevention after myocardial infarction: A review of long-term trials. Prog Cardiovasc Dis 1982;24:331–352. cardiac rehabilitation. A summary of the Oldridge rehabilitation on mortality was independent of results are presented in Table 14-6, in which the chronic heart failure diagnosis, type of rehabilita- pooled odds ratios for the combined studies sug- tion, dose of exercise intervention, length of follow- gest 24% and 25% reductions in all-cause and car- up, trial quality, or trial publication data. diovascular deaths, respectively, among the exercise groups. Criticisms of these analyses are that each Although the mortality effects of exercise-based of the pooled studies was not uniform in its treat- rehabilitation on outcomes in patients with prior ment of patients, and a non-exercise intervention history of MI have been known for some time (i.e., done in the different trials may have biased the since the 1980s), meta-analyses among patients results. Nevertheless, these two meta-analyses with heart failure (HF) have only recently been have been widely cited and have been highly influ- performed. Until the late 1980s, activity was gen- ential in support of cardiac rehabilitation. erally restricted in patients with HF, due largely to concerns over safety and unknown effects on the Recently, Taylor et al50 performed an updated myocardial remodeling process. During the 1990s, meta-analysis of rehabilitation trials among patients numerous trials demonstrated that exercise train- with CHD. While the aforementioned studies ing is safe in these patients, and several landmark focused on research performed during the 1970s trials were published that employed highly techno- and 1980s, the latter study included trials per- logical imaging techniques that allayed concerns formed through 2003. A total of 48 trials met over the effects of training on left ventricular the inclusion criteria, including 8940 patients. remodeling. This is addressed in more detail later Compared to usual care, cardiac rehabilitation was in this chapter (see Table 14–11). associated with reduced all-cause mortality (odds ratio [OR] = 0.80) and cardiac mortality (OR = A collaborative study of European centers 0.74). In addition, participation in cardiac rehabil- that performed exercise training trials in patients itation was associated with greater reductions in with HF during the 1990s was recently completed cholesterol, triglycerides, and SBP. However, (termed the ExtraMATCH study).51 This meta- there were no differences between rehabilitation analysis involved controlled exercise trials in HF, and usual care groups in nonfatal reinfarctions or and was designed to provide estimates of treatment revascularization rates. Importantly, the effect of benefits on mortality and hospital admission. Nine trials met the study inclusion criteria, TA B L E 1 4 – 6 . Meta-analysis of controlled exercise trials in patients following myocardial infarction No. of events/no. of patients % Treatment Control Pooled odds ratio (95% CI) P All-cause death 236/1823 (12.9) 289/1791 (16.1) 0.76 (0.63–0.92) 0.004 Cardiovascular death 204/2051 (9.9) 252/1993 (12.6) 0.75 (0.62–0.93) 0.006 From Oldridge NB, Guyatt GH, Fischer ME, Rimm AA: Cardiac rehabilitation after myocardial infarction. Combined experience of randomized clinical trials. JAMA 1988;260:945–950.
C H A P T E R 1 4 Cardiac Rehabilitation 475 comprising a total of 395 exercise intervention effect on cardiovascular risk factors (discussed later patients and 406 controls. After a mean follow-up in this chapter), other mechanisms may include a period of 705 days, it was found that exercise train- reduction in the incidence of ventricular fibrilla- ing reduced mortality in the order of 35%. Using tion and its effect on ischemic preconditioning. A death or hospital admission as an endpoint, exer- reduction in ventricular fibrillation might help cise was associated with a 28% reduction in mor- explain the reduction in sudden death demon- tality. Moreover, there was no evidence that any strated in the meta-analyses,47-52 and has been subgroup (elderly, severely reduced exercise capac- suggested by the observations of an increased fib- ity or ventricular function, type of HF, duration of rillation threshold after training in animal stud- training, or gender) would be less likely to benefit ies, which is associated with heightened vagal from training. tone.53-55 The reduction in sudden death in sub- jects randomized to exercise programs may also Smart and Marwick52 performed a similar be partly explained by the availability of defibrilla- but less restrictive analysis, including a search tors and trained personnel to resuscitate victims of the medical literature on exercise training in of exercise-related fibrillation. Alternatively, it has HF between 1966 and 2003. A total of 81 studies been suggested that training represents a form were included in the analysis, which included aer- of ischemic preconditioning, which refers to the obic training studies, strength training, inspira- observation that brief periods of ischemia before tory muscle training, and combinations of these coronary occlusion can reduce subsequent infarct approaches. There were no reports of deaths asso- size.56–58 Ischemic preconditioning has been ciated with more than 60,000 patient-hours of suggested to help lessen the effects of subsequent exercise. No significant differences were observed ischemia, through any of several mechanisms. between the exercise and control groups in total These include enhancing thrombolysis by the adverse events. However, the mortality rate was release of adenosine and its effect on platelet lower among subjects who participated in an exer- aggregation, enhanced collateral blood flow (with cise program versus those randomized to a control chronic ischemia), or through a complex series of group (OR = 0.71, P = 0.06). second messenger pathways in the heart.59 These meta-analyses have provided strong sup- Newer Concepts Regarding port for the application of cardiac rehabilitation Physiologic Benefits of Exercise in clinical practice. They demonstrate the mortal- Training ity benefits associated with exercise training, and have led many to suggest that regular physical A longstanding and attractive hypothesis is the activity can provide a powerful protective effect concept that exercise training can reverse or retard against recurrent MI and death that should be the progression of atherosclerosis. The observation additive to the current standardized interventions that regression of atherosclerosis occurred in ani- (PCI, beta-blockade, statins, bypass surgery). This mal studies dating back to the 1950s continues to evidence has sustained the growth of cardiac reha- stimulate interest in the effects of exercise on the bilitation as an important component of health- coronary vasculature in humans. While this idea care around the world. However, the meta-analyses was largely rejected during the 1970s and 1980s, have limitations, and the issue is often raised that several notable studies were performed during the none of the single-center studies alone have been 1990s indicating that exercise training, when com- sufficiently powered to adequately document bined with multidisciplinary risk management, can changes in mortality. In the absence of such data improve myocardial perfusion.60-62 This has been directly testing the hypothesis that exercise train- demonstrated indirectly using nuclear imaging60 ing reduces subsequent MI and mortality, scien- and directly by angiography.61,62 Because most of tific and popular perceptions regarding the these studies involved multidisciplinary risk reduc- benefits of exercise will support it as a treatment tion (e.g., diet, smoking cessation, stress manage- modality for cardiovascular disease. ment, and pharmacologic management of risk factors, including statin therapy) in addition to Mechanism for Reduced Mortality exercise training, it is not possible to determine the with Exercise Training independent effects of exercise training. The precise mechanism as to why an exercise pro- There is also debate regarding the mechanism gram may reduce mortality is unknown. Aside from by which the apparent improvement in myocardial the physiologic benefits of training and its favorable
476 E X E R C I S E A N D T H E H E A R T perfusion might occur following training. It is in both patients with and without existing cardio- generally considered unlikely that changes in vascular disease. coronary blood flow during exercise in animals would apply to humans. Three mechanisms could RISK OF EXERCISE TESTING AND potentially explain an improvement in perfusion TRAINING IN CARDIAC after training: (1) direct regression of atheroscle- REHABILITATION rotic lesions, (2) formation of collateral vessels, or (3) a change in the dynamics of epicardial flow via Risk of Exercise Testing flow-mediated or endogenous stimuli of the vessel. Although there has been evidence of small but There is a small but definite incidence of cardiac significant improvements in lumen diameter after arrest associated with exercise testing of cardiac intensive exercise and risk reduction programs in patients, particularly in the early minutes of recov- patients with CAD, no evidence exists that collateral ery from the exercise test. The first study to address vessel formation occurs after training in humans. this was a large multicenter survey of complica- Interestingly, although changes in lumen diame- tions by Rochmis and Blackburn,68 which showed a ter following these intervention programs are combined mortality and morbidity rate of four quite small, they are associated with considerable events per 10,000 tests. In a retrospective review reductions in hospital admissions for cardiac rea- by Irving and Bruce69 of 10,751 symptom-limited sons.62 This suggests that patients in the interven- exercise tests, five cardiac arrests were reported. All tion groups may achieve greater plaque stability, occurred in the first 4 minutes of recovery, and all without large changes in the coronary artery five patients survived after defibrillation (one arrest lumen. per 2000 tests). Based on these results, the relative risk of developing cardiac arrest with exercise test- In terms of the third mechanism, that is, ing (lasting 15 minutes) can be estimated to be changes in epicardial flow dynamics after train- one arrest per 538 hours of treadmill exercise, or ing, a significant amount of recent research has 160 times greater than what might be expected to demonstrated that training improves endothelial occur spontaneously (one death per 88,000 hours, dysfunction, thus permitting enhanced peripheral assuming a 10% yearly rate of sudden death). and coronary blood flow in response to exercise. This represents a paradigm shift in the pathophys- Perhaps due to expanded knowledge concern- iology of CAD. The last decade has brought an ing indications, contraindications, and endpoints, awareness that the luminal diameter of epicardial maximal exercise testing appears safer today than vessels changes rapidly in response to mechanical 30 years go. Gibbons et al70 reported the safety of (flow-related) and endogenous or pharmacologi- exercise testing in 71,914 tests conducted over a cal stimuli. Hambrecht et al63 studied the effects 16-year period. The complication rate was 0.8 per of exercise training in patients with reduced ven- 10,000 tests. This is markedly lower than earlier tricular function and reported that leg blood flow studies, but it was conducted in a population that during acetylcholine infusion was enhanced com- was generally healthier. The authors also suggested pared to controls. The improvement after training that the low complication rate might be due to was attributed to an increase in endothelium- the use of a cool-down walk, which may make the dependent vasodilation with an increase in basal recovery period safer. However, we have found that nitric oxide formation. In a subsequent study, out of 3351 tests in our laboratory, no cardiac these investigators demonstrated an improvement arrests occurred, and sustained ventricular tachy- in endothelium-dependent vasodilation in epicar- cardia occurred in only five patients.71 In general, dial vessels as well as resistance vessels in patients our population is a higher risk group, and all with CAD. After 4 weeks of exercise training, there patients are placed supine immediately after the was a 29% increase in coronary artery flow reserve test for diagnostic purposes. In 1997, Franklin in comparison to the nonexercise control group.64 et al72 reported their experience from 8 years of exercise testing (more than 58,000 tests) using These findings have been confirmed by other nonphysician healthcare providers. Their very low groups,65-67 and suggest an important role of morbidity (2.1 per 10,000) and mortality (0.3 per endothelial dysfunction contributing to inadequate 10,000) rates underscore the recent debate as to blood flow in patients with cardiovascular disease. whether physician supervision is always necessary Exercise training appears to have a profound effect for exercise testing. More recently, we conducted a on the vasodilatory properties of the vasculature. survey of 72 exercise laboratories in the VA Health Further exploration into the effects of exercise Care System.73 Among a total of 75,828 exercise training on the dynamic behavior of the endothe- lium is an important target area for future research
C H A P T E R 1 4 Cardiac Rehabilitation 477 tests, an event rate (defined as an event serious Of 50 cardiopulmonary resuscitations (CPR), eight enough to require hospitalization) of 1.2 per 10,000 resulted in death, and of seven MIs, two resulted tests was reported (three MIs, one sustained ven- in death. Exercise programs resulted in four other tricular tachycardia). Thus, the recent studies are fatalities occurring after hospitalization. Thus, consistent in the observation that the rate there was one nonfatal event per 35,000 patient- of events during exercise testing is low (≈1 per hours and one fatal event per 160,000 patient- 10,000); this appears to be true for both clinical hours. The complication rates were lower in and asymptomatic populations. The overall com- ECG-monitored programs. These programs plication rates from exercise testing of studies in reported a 4% annual mortality rate during exer- the literature are summarized in Table 14-7. cise, which is a rate not different from that expected for such patients. Other programs have Risk of Complications during reported rates of CPR ranging from 1 in 6000 to 1 Exercise Training in 25,000 man-hours of exercise. Such events are difficult to predict, can occur in patients with mild as Haskell74 surveyed 30 cardiac rehabilitation well as more severe forms of disease, and can occur programs in North America using a questionnaire at any time after being in a program. to assess major cardiovascular complications. This survey included approximately 14,000 A Seattle cardiac rehabilitation program (CAPRI) patients including 1.6 million exercise-hours. reported a comparatively high rate of one CPR in 6000 exercise-hours.75 Of 15 patients requir- ing defibrillation, the CAPRI group successfully TA B L E 1 4 – 7 . Complication rates of exercise testing from previous studies Investigator No. of Morbidity rate Mortality rate Total complications Physician tests (per 10,000) (per 10,000) (per 10,000) supervised? Rochmis & Blackburn 170,000 (JAMA 1971;217:1061– 2.4 1.0 3.4 Yes* 1065) 518,448 Stuart & Ellestad (Chest 353,638† 8.4 0.5 8.9 Yes* 1980;77:94–97) 712,285‡ 0 0 Scherer & Kaltenbach 1,377§ 1.4 0.2 0 Yes* (Disch Med Wochenscht 50,000 232 0 1.6 Yes* 1979;33:1161–1165) 18,707 5.2 0.4 Young et al (Circulation 3.8 0.9 232 Yes* 1984;70:184–191) >12,000 Atterhog et al (Am Heart — 2.5 5.6 Yes* J 1979;98:572–579) 71,914 Cahalin et al/Blessey 4,050 0.7 0.1 4.7 No (Exercise Standards and 3,351 0.3 0 Malpractice Reporter 28,133 14.9|| 0 — No 1989;3:69–74) 58,047 3.2 0 DeBusk (Exercise 75,828 2.1 0.3 0.8 Yes* Standards and 1.2 0 0.3 No Malpractice Reporter 14.9 Yes* 1988;2:65–70) 0.3 No Gibbons et al (Circulation 2.4 No 1989;80:846–850) 4.0 Yes¶ Lem et al (Heart Lung 1985;14:280–284) Yang et al (Arch Intern Med 1991;151:349–353) Knight et al (Am J Cardiol 1995;75:390–391) Franklin et al (Circulation 1995;92(suppl I):737.) Myers et al (J Cardiopulm Rehabil 2000;20:251–258) *>85% of these tests were directly supervised by physicians; †Athletes; ‡Coronary patients; §Patients with a history of malignant ventricular arrhythmias; ||Sustained ventricular tachycardia only; ¶73% supervised directly by physicians.
478 E X E R C I S E A N D T H E H E A R T resuscitated all of them. Eleven had angiogra- from 1 per 50,000 to 1 per 120,000 patient-hours phy, which showed single-vessel disease in four of exercise. It is noteworthy that studies have patients and multivessel disease in seven. shown that in fact, the majority of sudden deaths Subsequently, the CAPRI record improved and they are temporally associated with routine activities of reported defibrillating two patients simultaneously; daily life and not with exercise. Moreover, cardiac on another occasion, a physician monitoring an events during exercise are more likely to occur exercise class was defibrillated. Of 2464 patients among habitually sedentary individuals, by a rate observed during a 13-year period, 25 cardiac arrests of 20 to 30 times.81 occurred during 375,000 hours of supervised exer- cise, a rate of one arrest per 15,000 hours. Similar SPONTANEOUS IMPROVEMENT incidence rates were reported in Toronto and in POSTMYOCARDIAL INFARCTION Atlanta, where five arrests occurred in 75,000 hours of exercise, and a similar rate of one arrest per Clearly, the natural course of healing after an MI 12,000 hours (total of 36,000 gymnasium-hours) is associated with some improvement in function, was reported in Connecticut. In CAPRI, 12 of irrespective of engaging in a formal exercise pro- the 25 victims had been enrolled for 12 or more gram. Several groups have tried to quantify this. months. Fibrillation was recorded in 23 cases and To document spontaneous improvement in aerobic ventricular tachycardia in two. Prompt defibrilla- capacity, the Stanford group measured VO2 max tion was carried out and all patients survived. Each within the first 3 months after an uncomplicated cardiac arrest was a “primary” arrhythmic event, MI.82 Forty-six men underwent symptom-limited and none was associated with acute MI. Eighteen maximal treadmill tests at 3 and 11 weeks after an of the 25 patients had ST-segment depression and MI. There was a significant increase between the five had developed hypotension with prior exercise two periods in heart rate, rate pressure product, and testing. oxygen uptake during exercise. The mean maximal heart rate increased from 137 to 150; VO2 max Fletcher et al76 reported that five coronary increased from 21 to 27 mL/kg/min; and maximal disease patients were resuscitated after ventricu- SBP, double product, and oxygen pulse also lar fibrillation in an exercise program. Multivessel increased. coronary disease that could be treated with bypass surgery was present in four of the patients. To evaluate hemodynamic changes after MI, Resuscitation was required unexpectedly and at Kelbaek et al83 measured VO2 max and performed unpredictable times, occurring 2 to 48 months invasive studies at rest and during two submaximal after being in the exercise program. exercise levels. Thirty men were studied 2, 5, and 8 months after an uncomplicated MI. Fourteen In the largest of these studies, Van Camp and patients participated in an exercise program during Peterson77 obtained statistics from 167 randomly the first 3 months of the study, whereas the other selected outpatient cardiac rehabilitation programs 16 patients attended the training during the second and found that the incidence rate for cardiac arrest 3-month period. An increase in VO2 max occurred was 8.9 per million patient-hours. Of these cardiac at the fifth month in both groups, 16% and 11%, arrests, 86% were successfully resuscitated, giving respectively, along with an increase in cardiac an incidence rate for death of 1.3 per million index at the same relative submaximal workload. patient-hours. This compares favorably with the Later in the study, only slight increments in estimated fatality rate for unselected joggers at VO2 max and no changes in hemodynamics were 2.5 per million person-hours of jogging.78 There recorded within or between the two groups. They also was no significant difference in the cardiac concluded that poor medical advice and pensions event rate between rehabilitation programs with appeared to be the major factors responsible for or without ECG monitoring. These data have been unnecessary unemployment after an acute MI. widely cited to document that the risk of exercise training is quite small.79 In a comprehensive review, Greenland and Chu84 analyzed eight controlled studies of supervised The incidence of exertion-related cardiac exercise programs and their effects on physical arrest in cardiac rehabilitation programs is low, work capacity. In all the studies reviewed, exercise and because of the availability of rapid defibrilla- capacity improved after the intervention, whether tion, death rarely occurs. The 2005 AHA Scientific the patients were in a control or active intervention Statement on Cardiac Rehabilitation and Secondary group. This suggests that either a patient’s exercise Prevention of Coronary Heart Disease80 lists the capacity is artificially limited by the patient himself occurrence of major cardiac events during super- vised exercise in contemporary programs ranging
C H A P T E R 1 4 Cardiac Rehabilitation 479 or herself or by the physicians caring for them (e.g., that chronic training promotes myocardial capil- a low-level predischarge exercise test), or there is a lary growth and enlargement of extramural vessels. spontaneous improvement in exercise capacity as However, it is unclear if these changes actually time passes following infarction. However, the exer- increase perfusion or protect the heart during cise groups always had a greater exercise capacity ischemia. Controversy remains as to whether or than the control groups after the interventions— not exercise training can promote coronary col- on the order of 20% to 25% better. Studies that laterals in the animal model subjected to chronic failed to show any benefit might have been lim- ischemia, even though the landmark study using ited by exercise programs of inadequate duration, ischemic pigs performed by Bloor et al87 supports as it probably takes longer than 3 months for at this contention. least some of the cardiac adaptations to occur, and also by compliance with the exercise program or There have been a number of attempts to prescription. demonstrate the effects of exercise training on the hearts of patients with CHD. Ferguson et al88 per- To study the extent to which spontaneous formed coronary angiography on 14 patients before improvement may occur in patients with reduced and after 13 months of exercise. Despite a 25% ventricular function after an MI, bypass surgery, or increase in maximal oxygen uptake, collateral ves- both, Goebbels et al85 studied 67 consecutive sels were observed in only two coronary arteries, patients referred to a residential rehabilitation pro- and four of 14 patients demonstrated progression gram in Switzerland. A month after their myocar- of disease. Nolewajka et al89 studied 10 male dial event, 42 patients had normal ventricular patients before and after 7 months of exercise function (ejection fraction >50%) and were ran- training. Neither the exercisers nor the 10 control domized to an exercise or control group. Twenty- patients showed any changes in coronary angio- five other patients had reduced ventricular function grams, myocardial perfusion as assessed by intra- after their myocardial event and were also random- coronary injection of radionuclides, or ejection ized to exercise and control groups. After 8 weeks fraction. Sim and Neill90 also failed to demonstrate of training, peak VO2 increased only in the group cardiac changes in trained angina patients, includ- with reduced ventricular function; the exercise ing assessment of myocardial blood flow and oxy- group with normal ventricular function did not gen consumption. Whether these negative findings change significantly. Conversely, control patients can be explained by limitations in the techniques, with normal ventricular function increased peak patient selection, inadequate intensity, or length VO2 spontaneously (by 19%), whereas control of training is uncertain. In the 1990s, evidence patients with reduced ventricular function did demonstrating angiographic regression of coronary not improve peak VO2. The authors suggested that disease in humans became available for the first patients with depressed ventricular function time. However, it is unclear whether these changes strongly benefit from rehabilitation, whereas most are due strictly to intensive therapy with the newer patients with preserved ventricular function fol- lipid-lowering drugs, or to diet and/or exercise. lowing an MI or CABG tend to improve sponta- This issue is reviewed in more detail in the follow- neously after the event. ing sections. CARDIAC CHANGES IN CORONARY Assessment of Cardiac Changes HEART DISEASE PATIENTS Using Radionuclides Many favorable physiologic changes have been doc- With the advent of radionuclide techniques in the umented in patients with CHD who have under- late 1970s and 1980s, numerous efforts were made taken an aerobic exercise program. These include to assess the effects of exercise training on myo- lower submaximal and resting heart rate, decreased cardial perfusion and function in both normal symptoms, and increased maximal oxygen uptake. subjects and patients with cardiovascular disease. Peripheral adaptations are thought to be mostly Verani et al91 used radionuclide ventriculography responsible for these changes, and controversy and thallium scintigraphy to evaluate 16 coronary has existed for many years as to the effects of patients before and after 12 weeks of exercise chronic exercise on the heart itself. In a review of training. Thirty patients entered the study, but the effects of exercise training on myocardial vas- only 16 completed it. Ten patients had a docu- cularity and perfusion, Scheuer86 concluded that, in mented MI at least 2 months prior, and all but one the normal animal heart, there is strong evidence of the others had angiographic documentation of
480 E X E R C I S E A N D T H E H E A R T coronary disease. Nine patients received pro- body weight, total cholesterol, and triglycerides. pranolol throughout the exercise period. Both Exercise capacity increased by 23%, whereas post-training exercise studies were performed at there were no changes in these variables in the same double product as in the pretraining the control group. Exercise-induced myocardial studies. For the ventriculography, a multicrystal ischemia by thallium201 scintigraphy decreased by camera was used and scintigraphy accomplished 13% (P < 0.05). In addition, significantly more within 10 seconds of completion of exercise. After patients demonstrated angiographic regression of the training program, 15 of the 16 patients had coronary lesions in the exercise group, whereas improved exercise tolerance. Resting mean left significantly more control patients demonstrated ventricular ejection fraction increased from 52% angiographic progression. In a third study by this to 57%, but no change was noted in exercise ejec- group, a reduction in radionuclide evidence of tion fraction or regional wall motion abnormali- myocardial ischemia in patients who exercised ties. The thallium studies also were unchanged. was not limited to those exhibiting regression of coronary arteriosclerotic lesions, suggesting that The Duke group reported the effects of 6 months regular exercise and a low-fat diet may retard pro- of exercise training on treadmill and radionuclide gression of coronary disease, independent from ventriculography performance in 15 patients, all regression of stenotic lesions. less than 6 months post-MI.92 A training effect was demonstrated by a lower heart rate at a submaximal Perfext workload and longer treadmill time in spite of a wide variation in resting ventricular function (ejec- Our group at the University of California, San tion fractions ranging from 17% to 67%). The mean Diego (UCSD) performed a study called PERFEXT ejection fraction, end-diastolic volume, and wall (PERFusion, PERFormance, EXercise Trial).98 motion abnormalities during rest and at matched The San Diego community was informed about the workloads and heart rates were not significantly recruitment of male CHD patients between the different after training. DeBusk and Hung93 ran- ages of 35 and 65 years for a free exercise program. domized 11 CHD patients to a home exercise pro- The responding volunteers were a select group of gram and 10 to a control group 3 weeks post-MI. highly motivated patients who were encouraged There was no significant difference in resting to accept randomization by being promised that if or exercise ejection fraction or thallium perfusion they were originally assigned to the control group, images between the two groups after 8 weeks. they could join the exercise classes after the 1- Todd et al94 from Scotland reported improvement year study was completed. Potential subjects were in thallium scores among 40 male patients with screened to determine if they: (1) had CHD, (2) stable angina after 1 year of following the Canadian were willing to be randomized and comply with Air Force plan for physical fitness. either a low-level home walking program or a medically supervised exercise program at UCSD A group from Heidelberg, Germany, performed Hospital, (3) could discontinue their medications a series of studies on the effects of exercise and a for testing, (4) had no complicating illnesses or low-fat diet on myocardial perfusion in patients locomotive limitations, (5) had not recently been with CAD using radionuclide techniques.95-97 In in an exercise program, and (6) had the approval the first of these studies, 18 patients with stable of their physician. The patients were classified by angina and mild hypercholesterolemia (mean 242 the following criteria: (1) history of MI, (2) stable ± 32 mg/dL) underwent a combined regimen of exertional angina pectoris, or (3) CABS. Disease low-fat/low-cholesterol diet and supervised high- stability was assured by careful history taking and intensity training for one year. After the study by not allowing the patient to enter the study until period, serum cholesterol had decreased to 202 ± at least 4 months after a cardiac event, a change 31 mg/dL and low-density lipoprotein (LDL), very in symptoms, or surgery. low-density lipoprotein were lowered to normal levels, and work capacity increased by 21%. Stress- Of all the men interested in participating, 161 induced myocardial ischemia by thallium201 patients were interviewed, signed consent forms, scintigraphy was reduced by 54% in the exercise and agreed to randomization. The patients were group, whereas no changes were observed in blood then scheduled for three entry exercise tests done lipids or measures of ischemia among controls. In on separate days, usually within a 2-week period. a second, larger study, 56 patients were randomized A thallium treadmill test was done first, for famil- to a similar diet and exercise regimen and com- iarization, followed by a maximal oxygen uptake pared with a control group at 1 year. The interven- tion group demonstrated significant decreases in
2 Did not C H A P T E R 1 4 Cardiac Rehabilitation 481 complete testing Interviewed 161 Signed consent forms Agreed to randomization 159 Tested 11 With normal 2 Became unstable radionuclide test results 1 Alchoholism Trained 146 Randomized Controls 1 Myocardial infarction 72 Medical 74 4 Unstable 66 dropouts 2 Coronary artery bypass surgery 7 Quit exercise, 59 1 Myocardial infarction refused retesting 69 1 Both, 1 Death 69 Repeat testing 1 year later ■ FIGURE 14–1 Patient distribution and flow in the PERFEXT study. From Froelicher VF, Jensen D, Genter F, et al: A randomized trial of exer- cise training in patients with coronary heart disease. (From JAMA 1984; 252:1291–1297. Copyright 1984, American Medical Association). treadmill test and a supine bicycle radionuclide oxygen uptake (Table 14-8). The control group study. Of the 161 patients who enrolled in the showed a significant decrease in exercise capacity; study, 15 were excluded; 2 did not complete base- this was partially because of the lower maximal line testing, 2 became unstable, and 11 had nor- heart rate obtained at 1 year. There was also a mal radionuclide test results. Of 146 patients who small but significant decline in the submaximal were randomized, 72 were in the training group heart rate and rate pressure product in the con- and 74 were in the control group. Patients ran- trol group, probably due to habituation. No domized to the exercise group underwent 1 year of changes were observed in maximal perceived supervised exercise sessions; the exercise intensity exertion, respiratory exchange ratio, or SBP progressed in standard fashion throughout the between the two groups initially or at 1 year, or year. The study design is outlined in Figure 14-1. between the initial and 1-year tests. Analysis of variance confirmed that the training effect, Significant training effects in the intervention including an increase in peak VO2, occurred in group were evidenced by a decrease in resting and subgroups of the exercise intervention patients submaximal heart rates, as well as significant relative to controls. These subgroups included increases in measured and estimated maximal
482 E X E R C I S E A N D T H E H E A R T TA B L E 1 4 – 8 . Initial and 1-year measurements rest, during the three stages of exercise, or the per- from maximal treadmill testing in perfext98 centage of change from rest to exercise between the control and trained group at 1 year in ejection Exercise fraction, end-diastolic volume, stroke volume, or cardiac output. However, the intervention group, Control intervention relative to controls, had significantly lower per- centage changes in end-systolic volume at all three Test (n = 69) group (n = 59) workloads. The data suggested that the magnitude of the intervention effect differed in the MI and Heart rate, beats/min 66 (9) 69 (12) non-MI groups, although this was not statistically Supine 69 (11) 65 (11) significant; the intervention effect appeared consis- 2.2 (10) −3.8 (10) tently stronger in the non-MI than in the MI group. Initial 1 year 125 (15) 126 (16) The exercise intervention group also demon- Mean difference 121 (160) 118 (15) strated a significant improvement in the exercise Submaximal, 3.3 mph/5% −3.1 (11)* −9.3 (12)† thallium images after 1 year, using the Atwood scor- Initial ing system99 as well as computer techniques.100 1 year 154 (19) 156 (22) However, comparing thallium scans side-by-side, Mean difference 149 (23) 154 (22) which has been done effectively to evaluate surgical Maximal −5.2 (13)* −2.2 (11) intervention, was not successful in the clinical Initial assessment of changes in myocardial perfusion fol- 1 year 209 (44) 215 (47) lowing the exercise program. Disappointingly, the Mean difference 199 (49) 196 (42) ST-segment changes did not show an improvement, Rate pressure product −8 (35)† −19 (34)* nor did they agree with the thallium changes.101 Submaximal, 3.3 mph/5% Initial 279 (57) 286 (59) One of the only changes in ventricular function 1 year 273 (60) 289 (67) or volume of a consistent nature was the signifi- Mean difference −6 (46) 3 (50) cantly lower percentage of change in end-systolic Maximal volume in the exercise intervention patients, which Initial 33 (8) 33 (9) could not be explained as being due to a decreased 1 year 32 (8) 37 (9) afterload because there were no significant differ- Mean difference 1.3 (5) 4.7 (6)† ences in blood pressure at any stage of bicycle exer- Maximal oxygen uptake −3 (18) 18 (24)† cise. It would appear that the trained heart calls on Estimated, mL/kg/min the Frank-Starling mechanism to a lesser extent Initial 2.1 (.5) 2.2 (.6) than the untrained heart, probably due to lessened 1 year 2.0 (.5) 2.3 (.6) ischemia or improved contractility. This response Mean difference −0.1 (.3)* 0.1 (.3)† may not have been seen had the patients’ legs % change −4 (17)* 8.5 (17)† been elevated during supine exercise testing Measured, L/min (Table 14-9). Initial 1 year Mean difference % change *Significant change from initial within group; †Significant change between groups. those with and without the following features: a TA B L E 1 4 – 9 . Estimated changes in stroke history of a Q-wave MI, treadmill test-induced volume and cardiac output during supine bicycle angina, ejection fraction less than 0.40, abnormal exercise after 1 year98 exercise test-induced ST-segment depression, beta- blocker administration, or a dropping ejection frac- Exercise-intervention group tion response. Measurement With angina Without P value Radionuclide ventriculography demonstrated (n = 20) angina a baseline increase in both end-systolic and end- Resting supine (n = 39) diastolic volume in response to supine exercise. stroke volume We examined the effect of training relative to con- Stroke volume, –9.9 mL 7 mL 0.06 trols on the following variables: heart rate, ejection stage 2 fraction, end-diastolic and end-systolic volumes, Maximal stroke –14.9 mL 11.9 mL 0.02 stroke volume, and cardiac output. Each of these volume variable was tested at rest and at each of the three Maximal cardiac –10.9 mL 10.3 mL 0.03 stages, and the percentage of change from rest to output each of the three stages was calculated for each –1.0 L/min 1.3 L/min 0.05 variable. There were no significant differences at
C H A P T E R 1 4 Cardiac Rehabilitation 483 The other significant change was the effect of uptake is estimated from treadmill time rather the intervention on stroke volume and maximal than measured, the changes are usually very much cardiac output. Training is known to increase both, exaggerated. but the differential effect due to angina was sur- prising. The decrease in stroke volume and cardiac THE EFFECT OF AN EXERCISE output in the angina patients was accompanied by PROGRAM ON THE VENTILATORY a lessening of ischemia and an increase in end- THRESHOLD systolic volume in response to supine exercise. This suggests that absolute volume changes had to The noninvasive measurement of the ventilatory or occur that could not be detected because of the lactate threshold has been considered in numerous variability of the volume technique. Future studies exercise-training trials to document the benefits need to address the mechanism of this response. of chronic exercise during submaximal levels. Some investigators have suggested that the ventilatory In routine clinical practice, cardiac rehabilita- threshold may even represent a more clinically rel- tion is begun as soon as possible after a cardiac evant point than maximal exercise, because most event. However, given our study design and sample activities of daily living are performed below the size limitation, we chose only to study patients ventilatory threshold. Thus, it has been an impor- with stable CHD. Studying patients more acutely tant index of cardiopulmonary function. It is also post-MI is complicated by the degrees of severity often suggested that the intensity of exercise train- and by the variable rate of spontaneous improve- ing must occur at intensities near or above the ment. Our results may not be applicable to the ventilatory threshold to ensure increases in car- cardiac population immediately postevent. diorespiratory variables such as peak VO2, VO2 at the ventilatory threshold, or improvements in One point of criticism might be that our markers of ventilatory efficiency. Many studies have patients did not exercise hard enough and that if demonstrated that this point can be changed with they had, more definite improvements might have programs of exercise training. Increases in the ven- been possible. However, even if we chose those tilatory threshold, expressed either as an absolute patients who trained the most intensely or who had value (VO2 in mL/min), or as a relative percentage the highest exercise class attendance, we did not of peak VO2, typically parallel the changes observed find greater changes. Surprisingly, there was a poor in peak VO2 after training. Although the reason the correlation between the intensity or attendance and ventilatory or lactate threshold improves after radionuclide or aerobic capacity changes. In fact, training has been the source of some debate, the there was a poor correlation between the change in most likely explanation is a greater rate of lactate aerobic capacity and changes in the radionuclide removal from the blood during exercise.104,105 tests. A paradox related to this developed during the 1980s; Ehsani et al102 reported impressive car- CHANGES IN THE EXERCISE ECG diac changes in a highly selected group of cardiac WITH EXERCISE TRAINING patients with asymptomatic ST-segment depres- sion exercised at very high levels. Hossack and It is attractive to think that myocardial perfusion Hartwick103 also have reported an increased risk could be evaluated noninvasively during exercise for exercise-induced events in similar patients. by the exercise ECG; there have been a number of The question remains whether the usual cardiac efforts to address this. Previous studies in this area patient can be exercised safely at higher levels than have had mixed results; these are reviewed in the usually accepted and, if so, whether more definite preceding chapter (see Table 13-6). As part of PER- cardiac changes can be demonstrated. FEXT, 48 patients who exercised and 59 control patients had computerized exercise ECGs per- Care must be taken in interpreting many of the formed initially and 1 year later.101 ST-segment studies evaluating the effect of chronic exercise in displacement was analyzed 60 msec after the end cardiac patients. Often initial testing is submaxi- of the QRS complex in the three-dimensional X, Y, mal, whereas follow-up tests are on a higher level and Z leads and utilizing the spatial amplitude because of increased patient and technician confi- derived from them. There were no significant dif- dence and enthusiasm. This should be suspected ferences between the groups except for less ST- when there are large increases in maximal heart segment displacement at a matched workload, but rate, blood pressure, respiratory exchange ratio, or perceived exertion. Our study did not show signif- icant changes in these parameters because we took care to encourage patients to perform a maximal effort in their initial test. In addition, if oxygen
484 E X E R C I S E A N D T H E H E A R T this could be explained by a lowered heart rate. were due to inhibited glucose metabolism that It is unlikely that myocardial perfusion could could impede the training effect, but this has not be changed after training to an extent that is great been substantiated. These observations suggest enough to cause differences in ST-segment dis- that repeated, sustained sympathetic stimulation placement. It is also unlikely that the ST segment might be an important factor in exercise training. is sensitive enough to detect such changes. If beta-adrenergic sympathetic stimulation is needed for an exercise effect to occur, then beta- THE EFFECT OF BETA-BLOCKERS blockade might be expected to interfere with this ON EXERCISE TRAINING process. There is evidence that a functioning sympathetic In 1974, Malmborg et al107 first reported that a nervous system may be necessary to achieve the training effect could not be obtained in coronary beneficial hemodynamic alterations of training. patients with angina on beta-blockers. However, In addition, the limitation in cardiac output due their exercise program was only held for 18 min- to beta-blockade may result in fatigue and reduce utes twice a week. Obma et al108 later reported a the intensity of training or compliance to exercise. conflicting result. Their patients were limited by Moreover, if ischemia (theoretically the major stim- angina but demonstrated a significant increase in ulus for collateral development) is lessened by beta- peak estimated oxygen uptake after an 8-week, 30- blockade, this potential benefit of training also to 60-minute, 5 to 7 days per week exercise pro- could be impeded. Beta-adrenergic blockade is now gram. Pratt et al109 retrospectively studied 35 used widely to treat patients both with CHD and patients with CHD who underwent a 3-month walk- heart failure. However, one of the beneficial hemo- jog-cycle training program. Fourteen patients dynamic effects of both regular exercise and beta- had received no beta-blockers, 14 received 30 to blockade is that heart rate at rest and submaximal 80 mg of propranolol per day, and seven patients workloads is decreased. If beta-adrenergic stim- received 120 to 240 mg of propranolol per day at the ulation is needed for the effects of exercise training discretion of their physicians. Training consisted to occur or if beta-blockade lessens the ischemia of three 1-hour periods per week at 70% to 85% necessary to promote collateralization, then beta- of maximal pre-training heart rate. Each group’s blockade might be expected to interfere with the estimated peak oxygen uptake, assessed while on beneficial results of exercise. Beta-blockade also medications, increased after training: by 27% in could increase perceived exertion and fatigue, thus those not taking beta-blockers, by 30% in those lessening the tolerance for higher exercise levels on a low dose, and by 46% in those on a high dose. and adherence to an exercise program. Therefore, a pharmacologically imposed limitation in heart rate Vanhees et al110 compared two groups of and cardiac output during exercise may prohibit patients with past history of MI but without angina obtaining an optimal training effect. pectoris; 15 were receiving beta-blockers and 15 were not receiving them. Propranolol and metopro- The mechanisms by which hemodynamic lol were the beta-blockers most commonly used, changes occur secondary to regular exercise are at daily doses ranging from 30 to 120 and 75 to poorly understood. High levels of sympathetic stim- 200 mg, respectively. Exercise training was per- ulation are present during aerobic exercise. It has formed between 60% and 80% of their maximal been shown that regular intermittent infusions of capacity for 3 months. Both groups showed lower dobutamine in dogs result in cardiovascular heart rates, SBP, and rate pressure products after changes similar to those induced by an exercise training, both at rest and during submaximal program. However, the dogs did not get a true exercise. Testing was done while on beta-blockers, “training” effect. Other support for the importance but surprisingly the maximal heart rate was only of sympathetic stimulation for achieving the about 13 beats per minute higher in the group changes induced by exercise is that prolonged infu- not on beta-blockers. Heart rate decreases were sion of epinephrine has enhanced myocardial significantly less in the group on beta-blockade, contractility and induced hypertrophy in dogs. whereas SBP decreases were less pronounced in Likewise, it has been discovered that sympathec- the other group. Peak measured oxygen uptake tomy abolishes the increase in the heart/body increased an average of about 35% in both groups, weight ratio produced by exercise in rats. Hossack but maximal heart rate and rate pressure product et al106 reported ventilatory changes during exer- were also higher in both groups. cise in response to a single 40-mg oral dose of propranolol. They hypothesized that the changes Pavia et al111 studied 27 patients enrolled in a cardiac rehabilitation program after an uncomp- licated MI. Fourteen patients were taking meto- prolol as prescribed by their referring physician,
C H A P T E R 1 4 Cardiac Rehabilitation 485 and 13 patients were on an individually prescribed training at the same intensities. Maximal exercise medical regimen not including a beta-blocker. tests were performed before starting the drug reg- Both groups underwent a training program lasting imen and training, and were then repeated 3 to 3 months using a training intensity designed to 5 days after completing the exercise program, when approximate the ventilatory threshold. After the beta-blockade was withdrawn. The subjects who rehabilitation program, the groups increased received propranolol had no increase in measured peak VO2 similarly (33% increase in the beta- VO2 max, whereas the placebo group changed from blocker group and 27% increase in the placebo a mean of 44 to 53 mL/kg/min. Maximal heart rate group, P < 0.01). Marked increases were also was unchanged in both groups. High levels of pro- observed in VO2 at the ventilatory threshold (28% pranolol had been maintained by monitoring and 39% increases in the beta-blocker and placebo plasma levels, with daily doses ranging from 160 groups, respectively). to 640 mg. This contradicts the findings reported in other studies, possibly because of the high levels Because both exercise training and beta-block- of beta-blockade achieved. However, when these ade have only become widely accepted therapies investigators repeated the same protocol in the for patients with heart failure as of late, this issue subjects using low doses of beta-blockers, a similar has only recently been addressed in this pop- attenuation of changes in VO2 max was observed. ulation. Two recent studies from France112,113 reported that exercise-training responses were Similarly, Marsh et al117 studied 12 normal similar between heart failure patients receiving individuals before and after a 6-week intensive beta-blockade therapy and those not receiving exercise program. Six subjects were randomized beta-blockade therapy. Although data in this to low-dose propranolol, and six were randomized group are limited, it appears that beta-blockers do to placebo and trained at similar intensities. All not impair functional adaptations to an exercise testing was performed off beta-blockade. Maximal program in heart failure. oxygen uptake increased significantly in the placebo group, but was unchanged in those receiving pro- Controversy even exists among normal subjects pranolol. These authors concluded that high levels and the effects of beta-blockade. Ewy et al114 stud- of sympathetic stimulation during training were ied 27 healthy male adults (mean age, 24 years) necessary for the conditioning process to occur. who first underwent two maximal treadmill tests. They were then randomly assigned to either a The work of Gordon et al118 at Stanford, who placebo group or to sotalol 320 mg per day. A randomized normal subjects to drug or placebo and third maximal treadmill test was performed 1 then trained them, has shown particularly interest- week after the administration of the agents. ing results. Beta-blockade eliminated the echo- Subjects then participated in a 13-week training cardiographic changes in left ventricular posterior program in which they exercised 45 minutes five wall and septal thickening that was found in times a week at a training heart rate equivalent to the placebo group who underwent training. Other 75% of measured maximal oxygen uptake. A fourth investigators who have done studies of cardiac maximal treadmill test was performed at the con- patients have not randomized the patients to beta- clusion of the training program while taking the blockade, but have instead taken patients selected agent; 7 days after cessation of medication, a fifth by their physicians to be on or off beta-blockade. maximal treadmill test was performed. Measured Naturally, this can bias the findings. Other possi- VO2 max was increased following training in both ble explanations for the different results obtained groups; however, in the beta-blocked group this in studies of the effects of beta-blockers on train- was demonstrated only off beta-blockers. These ing include: (1) inadequate total time in training, findings suggest that stroke volume had attained (2) high initial levels of training or fitness, (3) dif- its maximal physiologic capacity during beta-adren- ferences in the suppression of maximal heart rate ergic blockade, and the reduction in maximal heart by beta-blockade, and (4) successful blinding of rate with beta-blockade prevented cardiac output subjects as to drug treatment in some studies. to increase optimally following training. These observations are supported by Tesch and Kaiser,115 We performed an analysis of patients in PER- who observed markedly reduced VO2 max values FEXT who exercised for 1 year versus controls, in highly trained athletes after acute administra- in which patients were placed on beta-blockers at tion of propranolol. the prerogative of their physicians.119 The patients’ medical records were reviewed to see who had Sable et al116 studied normal young men before taken beta-blockers, as prescribed by their physi- and after 5 weeks of aerobic training. In double- cians, during the year of training for the exercise blind fashion, eight received placebo and nine group and the year of observation for the controls. received propranolol throughout the period, while This information was then used to separate them
486 E X E R C I S E A N D T H E H E A R T into four groups: (1) controls on beta-blockers, beta-blockade in normal subjects as compared (2) controls not taking beta-blockers, (3) trained with studies among patients with CHD engaged in subjects on beta-blockers, and (4) trained subjects exercise training. In coronary patients selected not taking beta-blockers. All testing was performed for beta-blockade treatment by their physicians, after beta-blocker withdrawal. More patients in the the answer regarding the beneficial effects of exercise group who were on beta-blockers had exercise is more definitive. From previous studies exercise test-induced angina than those who were it has been demonstrated that expected changes in not on beta-blockers (64% versus 16%, P < 0.01), oxygen uptake, submaximal heart rate, and exercise and they tended to have more ST-segment depres- duration usually occur in patients who engage in sion and higher thallium ischemia scores. There exercise training. Our study supports this, but also was a trend for a higher prevalence of prior bypass demonstrates no preferential difference between surgery in those not on beta-blockers. These differ- those patients trained on or off beta-blockers. In ences are probably due to exercise training making addition, PERFEXT demonstrated an increase in limitations due to angina more obvious and lead- myocardial perfusion, implied by improved thal- ing to beta-blocker administration. The average lium scintigrams in angina patients in an exercise exercise intensity in the beta-blocker group for the program. These findings and those summarized year was 77% ± 14% of measured oxygen uptake, above support the beneficial effects of exercise and average calories expended per session was 323 training in coronary patients taking beta-blocker ± 104 (ranging from 130 to 719). There were no medications. A summary of the studies that have significant differences in these values between addressed this issue is presented in Table 14-10. those on or off beta-blockers. Attendance at exercise sessions was a mean of 76% ± 18% (ranging from COMPLIANCE 23% to 97%) with no difference between those on or off beta-blockers (73% versus 78% for those on The success and benefit of any exercise training and off beta-blockers, respectively). program obviously are directly related to the amount of exercise actually performed by the Two-way analysis of variance revealed highly patient—in other words, their compliance with significant changes in the treadmill parameters due the exercise prescription. Kentala120 reported that to the exercise intervention. No interaction was only 13% of his patients carried out their assigned detected due to beta-blocker status during the year. exercise prescription at least 70% of the time. As There was no correlation between beta-blocker time progressed, compliance fell. At 3 months, dosage and the change in measured oxygen uptake compliance was 80%; 1 year later, compliance was in the exercise group. No other changes in tread- only 45% to 60%; and at 4 years it was only 30% mill parameters, including maximal heart rate, to 55%. These findings are similar to those from blood pressure, perceived exertion, or respiratory the U.S. National Exercise and Heart Disease exchange ratio were detected. The changes in sub- Project,121 in which compliance to exercise partic- maximal heart rate were significant despite the ipation dropped from 80% after 2 months of rebound effect of beta-blocker withdrawal. supervised training to only 13% after 3 years. These and other studies clearly show that adher- Considering the clinical classifications of angina, ence to physical activity is often unsatisfactory in prior MI, and CABS revealed significant (P < 0.01) the absence of some form of continued follow-up improvement only in the thallium scintigrams of or supervision. the patients in the exercise program with exercise test-induced angina. Therefore, three-way analysis Several options are available to improve com- of variance for angina, beta-blockers, and interven- pliance behavior: reduce the waiting time for tion was performed. Although there was a trend for enrollment; expert supervision; tailoring of the this improvement to be concentrated in angina exercise prescription to avoid physical discomfort patients not taking beta-blockers, this did not or frustration; use of variable activities, including reach statistical significance. games; incorporation of social events; recalling absent patients; involving the patient’s family or By design, patients were selected by their spouse in the program; case management; and physicians to be on or off beta-blockers, and involving the patients in monitoring themselves beneficial effects of exercise training were demon- and their progress. strated. This clinical study is different from that studying the effects of beta-blockers on exercise training in normal subjects. Conflicting results exist as to the effects of being randomized to
C H A P T E R 1 4 Cardiac Rehabilitation 487 TA B L E 1 4 – 1 0 . Summary of studies evaluating the effects of beta-blockade on exercise training Investigator Population Duration Drug(s) Findings Lester 1978 Pratt 1981 Normals (n = 6) 6 wk Propranolol Peak VO2 increase 24% Normals (n = 6) 12 wk No beta-blocker Peak VO2 increase 25% Sable 1982 CAD (n = 14) Propranolol (30-80 mg/day) Peak VO2 increase 30% Laslett 1983 Peak VO2 increase 46% Vanhees 1984 CAD (n = 7) 5 wk Propranolol (120–240 mg/day) Peak VO2 increase 27% CAD (n = 14) 12 wk No beta-blocker Ehsani 1985 Normals (n = 9) 12 wk Propranolol No change in peak VO2 Fletcher 1985 Normals (n = 8) No beta-blocker Peak VO2 increase 21% Savin 1985 CAD (n = 11) Propranolol METs increase 17% CAD (n = 25) No beta-blocker METs increase 12% Wilmore 1985 CAD (n = 15) Atenolol, metoprolol, Peak VO2 increase 37% CAD (n = 14) 48 wk propranolol Peak VO2 increase 34% CAD (n = 13) 12 wk No beta-blocker Peak VO2 increase 36% CAD (n = 13) 6 wk Propranolol, nadolol, timolol Peak VO2 increase 35% CAD (n = 50) No beta-blocker Exercise time increase 32% Normals (n = 13) 15 wk Various Peak VO2 increase 17% Normals (n = 11) Atenolol Peak VO2 increase 17% Normals (n = 15) Propranolol Peak VO2 increase 19% Normals (n = 47) No beta-blocker Peak VO2 increase 18% Atenolol Peak VO2 increase 17% Ciske 1986 CAD (n = 24) 4 wk Peak VO2 increase 17% Propranolol Peak VO2 increase 16% Madden 1988 CAD (n = 15) 12 wk No beta-blocker Peak VO2 increase 21% CAD (n = 9) Propranol, atenolol, Ades 1990 CAD (n = 7) 10 wk metoprolol, timolol Peak VO2 increase 11% CAD (n = 8) No beta-blocker Peak VO2 increase 22% Pavia 1995 CAD (n = 10) 12 wk Propranolol Peak VO2 increase 12% Malfatto 1998 CAD (n = 10) 48 wk Atenolol Peak VO2 increase 24% Forissier 2001 CAD (n = 10) 4 wk No beta-blocker Peak VO2 increase 8% CAD (n = 14) Metoprolol No change in peak VO2 CAD (n = 13) Propranolol Peak VO2 increase 33% CAD (n = 20) No beta-blocker Peak VO2 increase 27% CAD (n = 19) Metoprolol Exercise capacity increase 23% CHF (n = 24) No beta-blocker Exercise capacity increase 22% Atenolol, metoprolol Peak VO2 increase 16.6% No beta-blocker Carvedilol CAD, coronary artery disease; CHF, chronic heart failure. PATIENTS WITH LEFT for rehabilitation. The incidence of CHF is cur- VENTRICULAR DYSFUNCTION rently about 500,000 per year in the United States, and these improvements in therapy mean that this Prior to the 1990s, patients with left ventricular number will continue to increase. Recent studies dysfunction were thought to be poor candidates suggest that the major physiologic benefit from for exercise programs. This was out of concern for training in CHF occurs in the skeletal muscle safety and the general thinking that they were rather than in the heart itself.122-126 A summary of unable to benefit from training. Safety concerns the major randomized studies in heart failure is have been dispelled, however, by numerous stud- presented in Table 14-11. Extensive studies have ies that have been published since the late 1980s. been performed on the effects of training on central Today it is recognized that patients with CHF derive hemodynamics, peripheral blood flow, myocardial considerable benefits from cardiac rehabilitation. remodeling after an MI using echocardiographic With improvements in therapy (e.g., thrombolytics, and MRI techniques, and skeletal muscle metabo- ACE inhibitors, beta-blockade), survival among lism.122-134 These studies are nearly universal in patients with CHF has improved considerably, and their demonstration that training has beneficial more of these patients are available as candidates effects on these systems. In addition, the rather
488 E X E R C I S E A N D T H E H E A R T TA B L E 1 4 – 1 1 . Trials of exercise training in humans with impaired left ventricular function Randomized Number Mean Etiology Program Adaptions due to controlled studies subjects lvef(%) CAD duration 8 wk training Coats et al (Lancet 17 20 1990;335:63–66) 8 wk ↑ Peak VO2 2.4 mL/kg/min, 4 wk ↑ exercise time 2 min, Meyer et al (J Intern 12 23 CAD 8 wk ↓ HRrest ↑QoL ↓ Med 1991;230:407–413) 7 versus 8 23 >10 wk post-MI 12 wk Ventilation, ↓ Symp, Jette et al (Circulation 12 23 Stable CAD 3 wk ↑ vagal tone 1991;84:151–157) 7 versus 13 35 >12 wk post-MI 8 wk Adamopoulos et al 24 versus 24 30 DCM-CAD 8 wk ↑ Peak VO2 1.6 mL/kg/min, (Am J Cardiol 5 versus 5 23 CAD 3 mo ↑ exercise time 1.4 Min 1993;21:1101–1106) 36 versus 19 26 DCM-CAD 6 mo ↑ Peak VO2 3.6 mL/kg/min, ↑ peak Kostis et al (Chest 12 versus 8 24 DCM-CAD 24 wk work rate 13 W, ↑ PWP 9 mmHg 1994;106;996–1001) 12 versus 10 26 DCM-CAD 2 mo ↑ Peak VO2 1.9 mL/kg/min, Kayanakis et al 21 versus 19 21 DCM-CAD ↑ exercise time 3 min, (Presse Med 12 versus 13 32 CAD ↓ PCr use, ↑ ADP recovery 1994;23:121–126) ↑ Exercise time 1 min, ↑ QoL Barlow et al 18 (Circulation ↑ Peak VO2 0.1 mL/kg/min, 1994;89:1144–1152) 8 versus 8 ↑ vascular resistance Belardinelli et al (Circulation 10 versus 10 ↑ Peak VO2 1.3 mL/kg/min, ↑ peak 1995;91:2775–2784) 12 versus 13 work rate 16%, ↓ [K+]a, ↓ lactate Kilavouri et al (Eur Heart J 1995;16: ↑ Peak VO2 1.2 mL/kg/min, 490–495) ↓ lactate, ↑ diastolic function, Hambrecht et al ↑ a-VO2 (J Am Coll Cardiol ↑ Peak VO2 3.2 mL/kg/min, 1995;25:1239–1249) ↑ exercise time, ↑ vagal tone, Keteyian et al (Ann ↓ symp Intern Med 1996; ↑ Peak VO2 5.8 mL/kg/min, 124:1051–1057) ↓ NYHA, ↑ mitochondria, ↑ peak Dubach et al (J Am leg VO2 ↑ max cardiac output Coll Cardiol ↑ Peak VO2 2.4 mL/kg/min, 1997;29:1591–1598; ↑ exercise time 2.8 min, Circulation ↑ peak work rate 1997;95:2060–2067) ↑ Peak VO2 4.5 mL/kg/min, Meyer et al (Am J ↑ max cardiac output, Cardiol 1997;80:56– ↑ peak work rate 36% 60; Am Heart J 1997;133:447–453) 21 DCM-CAD 3 wk ↑ Peak VO2 2.4 mL/kg/min, Demopoulos et al 23 DCM-CAD 12 wk ↑ 6-min walk test, ↓ VE/VCO2 (Circulation 1997;95:1764– ↑ Peak VO2 3.1 mL/kg/min, 1767) ↑ calf blood flow Hambrecht et al (Circulation 23 DCM-CAD 26 wk ↑ Peak VO2 26%, marked 1998;98:2709–2715) 32 CAD 8 wk improvement in endothelial Myers et al (Med Sci Sports Exerc 28 DCM-CAD 52 wk dysfunction 1999;31:929–937) 29% ↑ in peak VO2, 39% ↑ in Bellardinelli et al 50 versus 49 VO2 AT, ↓ in lactate at matched (Circulation work rates, ↓ 1999;99:1173–1182) VE/VCO2 slope, improved ventilatory efficiency ↑ Peak VO2 18%, ↑ thallium ischemia scores, ↑ QoL, ↓ mortality (RR = 0.37), ↓ hospital readmission (RR = 0.29)
C H A P T E R 1 4 Cardiac Rehabilitation 489 TA B L E 1 4 – 1 1 . Trials of exercise training in humans with impaired left ventricular function—cont’d Randomized Number Mean Etiology Program Adaptions due to controlled studies subjects lvef(%) DCM-CAD duration training 27 Hambrecht et al 36 versus 37 CAD 26 wk ↑ Peak VO2 4.8 mL/kg/min, (JAMA 2000; 32 DCM-CAD ↑ exercise time 4 min, ↑ LVEF, 283:3095–3101) 25 DCM-CAD 8 wk, ↓ peak exercise TPR, 17 1 year ↑ stroke volume Myers et al (Am 12 versus 13 DCM-CAD follow-up 27% ↑ peak VO2, no change in Heart J 11 versus 11 25 4 wk LVEF, LV volumes, or wall 2000;139:252–261) 36 versus 37 CAD-DCM Linke et al (J Amer 27 CAD-DCM 26 wk thickness by MRI Coll Cardiol 45 versus 45 26 CAD-DCM 2001;37:392–397) na CAD ↑ Peak VO2 21%, ↑ VO2 AT 19%, Erbs et al (Eur J 5 26 improved endothelium- Cardiovasc Prev 13 versus 12 CAD-DCM Rehab 2003;10: 10 21 CAD-DCM dependent vasodilation 336–344) 9 28 Giannuzzi et al CAD-CHF ↑ Peak VO2 32%, ↑ VO2 AT 49%, (Circulation 11 ↑ resting and peak exercise, 2003;108:554–559) 14 versus 7 >6 post-MI stroke volume Single Limb Training >3 mo post-MI Minotti et al (J Clin >12 wk post-MI 26 wk ↑ Peak VO2 17%, ↓ LV volumes, Invest 1990;86: CAD-DCM 4% ↑ in LVEF, ↑ 6-min 751–758) 6–32 wk post-MI Koch et al (Chest >6 mo post-MI walk distance, ↑ QoL 1992;101:231S– 235S) 4 wk ↑ Peak VO2 0.9 mL/kg/min, Stratton et al (J Appl Physiol ↑ exercise time, ↓ PCr use 1994;76:1575–1582) Piepoli et al (Am J 3 mo ↑ Exercise time 3 min, ↑ QoL, Physiol ↑ strength 1995;269:H1428– H1436) 1 mo ↑ Exercise time 10 min, ↑ pH, Magnusson et al (Eur ↑ PCr, ATP resynthesis Heart J 1996;17:1048–1055) 6 wk ↑ Peak VO2 0.7 mL/kg/min, Gordon et al (Clin Cardiol 1996;19: ↑ exercise time 3.2 min, ↓ ergo- 568–574) reflex activity Hambrecht et al 10 versus 10 24 2 mo ↑ Peak VO2 0.8 mL/kg/min, 8 wk ↑ peak work rate, ↑ cross- sectional area 6 mo ↑ Peak VO2 0.1 mL/kg/min, (Circulation ↑ capillary, ↑ oxidative enzyme activity, ↑ peak work rate, 1998;98:2709–2715.) ↑ strength, ↑ muscle citrate synthase ↑ Peak VO2 3.7 mL/kg/min, ↑ endothelium-dependent vasodilation Nonrandomized or Uncontrolled Studies Lee et al (Circulation 18 18 18 mo ↑ Exercise time 1.1 Min (12–42) 1979;60:1519–1526.) 12 mo ↑ Peak work rate 1.5 (4–37) METS Conn et al 1982 10 20 12 wk ↑ Peak VO2 7 mL/kg/min, Arvan et al 1998 25 29 16–24 wk ↑ exercise time 4 min Sullivan et al 12 9–33 12 wk ↑ Peak VO2 3.8 mL/kg/min, (Circulation 7 43 leg a-VO2 diff, ↑ leg flow, 1998;78:506–515.) 6 32 ↓ lactate production Jugdutt et al (J Am ↑ Total work, ↓ LVEF 13% Coll Cardiol 1988;12:367–372.) 5 wk ↑ Peak work rate 9 W, ↓ peak VO2 Scalvini et al mL/kg/min (Cardiology 1992;80:417–423.) Continued
Search
Read the Text Version
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
- 142
- 143
- 144
- 145
- 146
- 147
- 148
- 149
- 150
- 151
- 152
- 153
- 154
- 155
- 156
- 157
- 158
- 159
- 160
- 161
- 162
- 163
- 164
- 165
- 166
- 167
- 168
- 169
- 170
- 171
- 172
- 173
- 174
- 175
- 176
- 177
- 178
- 179
- 180
- 181
- 182
- 183
- 184
- 185
- 186
- 187
- 188
- 189
- 190
- 191
- 192
- 193
- 194
- 195
- 196
- 197
- 198
- 199
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 209
- 210
- 211
- 212
- 213
- 214
- 215
- 216
- 217
- 218
- 219
- 220
- 221
- 222
- 223
- 224
- 225
- 226
- 227
- 228
- 229
- 230
- 231
- 232
- 233
- 234
- 235
- 236
- 237
- 238
- 239
- 240
- 241
- 242
- 243
- 244
- 245
- 246
- 247
- 248
- 249
- 250
- 251
- 252
- 253
- 254
- 255
- 256
- 257
- 258
- 259
- 260
- 261
- 262
- 263
- 264
- 265
- 266
- 267
- 268
- 269
- 270
- 271
- 272
- 273
- 274
- 275
- 276
- 277
- 278
- 279
- 280
- 281
- 282
- 283
- 284
- 285
- 286
- 287
- 288
- 289
- 290
- 291
- 292
- 293
- 294
- 295
- 296
- 297
- 298
- 299
- 300
- 301
- 302
- 303
- 304
- 305
- 306
- 307
- 308
- 309
- 310
- 311
- 312
- 313
- 314
- 315
- 316
- 317
- 318
- 319
- 320
- 321
- 322
- 323
- 324
- 325
- 326
- 327
- 328
- 329
- 330
- 331
- 332
- 333
- 334
- 335
- 336
- 337
- 338
- 339
- 340
- 341
- 342
- 343
- 344
- 345
- 346
- 347
- 348
- 349
- 350
- 351
- 352
- 353
- 354
- 355
- 356
- 357
- 358
- 359
- 360
- 361
- 362
- 363
- 364
- 365
- 366
- 367
- 368
- 369
- 370
- 371
- 372
- 373
- 374
- 375
- 376
- 377
- 378
- 379
- 380
- 381
- 382
- 383
- 384
- 385
- 386
- 387
- 388
- 389
- 390
- 391
- 392
- 393
- 394
- 395
- 396
- 397
- 398
- 399
- 400
- 401
- 402
- 403
- 404
- 405
- 406
- 407
- 408
- 409
- 410
- 411
- 412
- 413
- 414
- 415
- 416
- 417
- 418
- 419
- 420
- 421
- 422
- 423
- 424
- 425
- 426
- 427
- 428
- 429
- 430
- 431
- 432
- 433
- 434
- 435
- 436
- 437
- 438
- 439
- 440
- 441
- 442
- 443
- 444
- 445
- 446
- 447
- 448
- 449
- 450
- 451
- 452
- 453
- 454
- 455
- 456
- 457
- 458
- 459
- 460
- 461
- 462
- 463
- 464
- 465
- 466
- 467
- 468
- 469
- 470
- 471
- 472
- 473
- 474
- 475
- 476
- 477
- 478
- 479
- 480
- 481
- 482
- 483
- 484
- 485
- 486
- 487
- 488
- 489
- 490
- 491
- 492
- 493
- 494
- 495
- 496
- 497
- 498
- 499
- 500
- 501
- 502
- 503
- 504
- 505
- 506
- 507
- 508
- 509
- 510
- 511
- 512
- 513
- 514
- 515
- 516
- 517
- 518
- 519
- 520
- 521
- 522
- 523
- 524
- 525
- 526
- 527
- 528
- 529
- 530
- 531
- 532
- 533
- 534
- 535
- 1 - 50
- 51 - 100
- 101 - 150
- 151 - 200
- 201 - 250
- 251 - 300
- 301 - 350
- 351 - 400
- 401 - 450
- 451 - 500
- 501 - 535
Pages:
