Exercise may cause myocardial ischemia at the anaerobic threshold in cardiac rehabilitation programs

Myocardial ischemia may occur during an exercise session in cardiac rehabilitation programs. However, it has not been established whether it is elicited when exercise prescription is based on heart rate corresponding to the anaerobic threshold as measured by cardiopulmonary exercise testing. Our objective was to determine the incidence of myocardial ischemia in cardiac rehabilitation programs according to myocardial perfusion SPECT in exercise programs based on the anaerobic threshold. Thirty-nine patients (35 men and 4 women) diagnosed with coronary artery disease by coronary angiography and stress technetium-99m-sestamibi gated SPECT associated with a baseline cardiopulmonary exercise test were assessed. Ages ranged from 45 to 75 years. A second cardiopulmonary exercise test determined training intensity at the anaerobic threshold. Repeat gated-SPECT was obtained after a third cardiopulmonary exercise test at the prescribed workload and heart rate. Myocardial perfusion images were analyzed using a score system of 6.4 at rest, 13.9 at peak stress, and 10.7 during the prescribed exercise (P < 0.05). The presence of myocardial ischemia during exercise was defined as a difference ≥2 between the summed stress score and summed rest score. Accordingly, 25 (64%) patients were classified as ischemic and 14 (36%) as nonischemic. MIBI-SPECT showed myocardial ischemia during exercise within the anaerobic threshold. The 64% prevalence of ischemia observed in the study should not be looked on as representative of the whole population of patients undergoing exercise programs. Changes in patient care and exercise programs were implemented as a result of our finding of ischemia during the prescribed exercise.


Introduction
According to current guidelines, exercise intensity in cardiac rehabilitation programs should not precipitate myocardial ischemia during a training session (1)(2)(3)(4)(5)(6).However, it has been shown by SPECT myocardial perfusion imaging that physical exercise within the functional capacity, based on standard exercise stress testing, may cause ischemia, characterized by a reversible perfusion defect, even without angina symptoms or ST-segment depression (7).
With the current widespread use of cardiopulmonary exercise testing, exercise intensity can be defined on the basis of the anaerobic threshold (8)(9)(10)(11).At this level of physical exertion, exercise training improves the overall functional capacity of the patients and may be applied even to those with coronary artery disease and ventricular dysfunction (12, , , , ,13).However, it has not yet been established whether exercise intensity determined by anaerobic threshold elicits myocardial ischemia.
Myocardial perfusion studies are more effective in identifying exercise-induced ischemia than other methods used for the diagnosis and risk stratification of obstructive atherosclerotic coronary artery disease (14)(15)(16).The objective of this study was to assess the incidence of myocardial ischemia in exercise prescription based on the anaerobic threshold in cardiac rehabilitation programs, using the myocardial perfusion SPECT test.

Material and Methods
From June 2005 to September 2006, 153 consecutive patients referred to the Nuclear Medicine Department of Instituto Dante Pazzanese de Cardiologia were assessed.All of them showed reversible defect on myocardial perfusion SPECT images and ≥70% diameter stenosis on coronary angiography in at least one of the three main coronary arteries (left anterior descending artery, right coronary artery, left circumflex artery, and their branches).At the initial analysis, 95 patients with at least one of the following characteristics were excluded: age over 75 years, presence of atrial fibrillation, pacemaker, left bundle branch block, preexcitation syndrome, hemodynamically significant valvular heart disease, difficulty in undergoing study procedures appropriately, impossibility to participate in an exercise program after the study, and patients already attending a cardiac rehabilitation program.The remaining 58 patients underwent history taking, clinical examination, and standard 12-lead electrocardiography, in addition to repeat SPECT myocardial perfusion following a cardiopulmonary exercise test using a Medical Graphics CardiO 2 system (USA).This was performed without discontinuation of any medication taken regularly.An individualized ramp protocol with a cycle ergometer was used.Criteria for exercise termination were those established by the American College of Cardiology and American Heart Association guidelines (17). . . . .A 740 Mbq (or 20 mCi) amount of technetium-99m-sestamibi ( 99m Tc-sestamibi) was administered to patients weighing 85 kg or less, and an amount of 0.31 mCi/kg was administered to those weighing more than 85 kg.The radiotracer was injected at near-maximum heart rate (HR; defined as 220-age) or in the presence of impending exhaustion, exertional ST-segment depression equal to or greater than 0.3 mV, or angina pectoris.After the injection, exercise was continued for 30 to 60 s, and myocardial perfusion SPECT was performed 60 min later.
Rest myocardial perfusion SPECT was performed within a week of the first cardiopulmonary exercise test.Under regular medication, only 39 patients showed a persistent reversible defect on myocardial perfusion SPECT images and were eligible for the next phase of the study.Thirty-five of them were men (89%) and 4 were women (10.3%).Ages ranged from 45 to 73 years (mean 59.76), and mean body mass index was 28.4.
The second cardiopulmonary exercise test was similar to the first, but was not associated with myocardial perfusion imaging, and was used to determine the anaerobic threshold using the V-slope method (18).Training level was set at the workload and HR corresponding to the anaerobic threshold.The third cardiopulmonary exercise test was performed at the workload prescribed for tests on mechanically braked bicycles, the same as used in the cardiac rehabilitation program, during over a period of 20 min, corresponding to the continuous aerobic phase of a cardiac rehabilitation program session.The 99m Tc-sestamibi dose was injected approximately 19 min into the test, and images were acquired 60 min later.
Both stress and rest SPECT images were processed using the QGS (CEDAR) dedicated software (19).Qualitative and semiquantitative analyses were performed.Radiotracer uptake was compared in 17 segments of the myocardium.Myocardial perfusion images were analyzed by 3 observers blind to patients' data using a score system for each of the 17 segments based on five tomographic views: basal short-axis slice, midventricular short-axis slice, apical short-axis slice, vertical long-axis slice, and horizontal long-axis slice.For each slice, the segment with the highest uptake was graded zero.The remaining segments were graded according to the following criteria: one, mildly reduced uptake; two, moderately reduced uptake; three, severely reduced uptake, and four, absent uptake.For each SPECT image, a summed score was obtained by adding the 17 segmental scores.No difference between the summed stress score and summed rest score indicated absence of stress-induced hypoperfusion, that is to say, absence of ischemia.Myocardial ischemia is defined as a difference ≥2 between stress and rest scores, according to the study by Berman et al. (20), a criterion that has been used in recent landmark studies (21).
Statistical analysis was performed using repeatedmeasures analysis of variance (ANOVA).The Bonferroni multiple comparison test was used to detect differences between cardiopulmonary exercise tests and the scores derived from SPECT images obtained at rest, at peak stress, and during the prescribed exercise.Summed difference score (peak -rest and prescription exercise -rest) were compared by the Wilcoxon test.P values ≤0.05 were A.R. Fuchs et al.
www.bjournal.com.brconsidered to be statistically significant.All scores for the three situations mentioned above are presented as means and 95% confidence intervals (95%CI).Mean HR was compared using the paired Student t-test.ST-segment changes between rest, 99m Tc-sestamibi injection at peak stress and prescribed exercise were analyzed by the McNemar test.The Pearson chi-square test or the Fisher exact test was used to determine association among groups with summed difference score (≥2 and <2) in relation to clinical, angiographic and electrocardiographic data.Age was compared by the Student t-test.
The study protocol was approved by the Research Ethics Committees of UNIFESP (Universidade Federal de São Paulo, Escola Paulista de Medicina) and Instituto Dante Pazzanese de Cardiologia, and all patients signed a consent form after being fully informed about the procedures involved in writing.

Results
The clinical and angiographic characteristics of the patients are shown in Table 1.Coronary angiography revealed significant stenosis involving the left anterior descending artery in 27 patients, the right coronary artery in 25 patients, and the left circumflex artery in 28 patients The cardiopulmonary variables evaluated in all three cardiopulmonary exercise tests are shown in Table 2.

Second cardiopulmonary exercise test for the determination of the level of exercise prescription
The anaerobic threshold was reached at a mean workload of 33.57± 2.70 watts and mean HR of 89.89 ± 2.47 bpm, corresponding to 56.12 ± 1.62% of age-predicted maximal HR.

Third cardiopulmonary exercise test during prescribed exercise based on the anaerobic threshold in order to obtain the 2nd myocardial perfusion SPECT
Mean prescribed workload was 32.87 ± 2.56 watts.Mean HR during exercise training was 90.36 ± 2.37 bpm, corresponding to 55.92 ± 1.46% of age-predicted maximal heart rate and 40.89 ± 1.46% of predicted VO 2 max.HR at the time of 99m Tc-sestamibi injection was 91.41 ± 2.38 bpm, corresponding to 57.01 ± 1.58% of age-predicted maximal HR and 40.89 ± 1 46% of predicted VO 2 max.Only 7 patients (18%) had ST-segment depression ≥1 mm, while 32 patients (82%) showed no ECG changes.Only 2 patients (5.12%) experienced angina pectoris.

Heart rate comparison
There was a significant difference between HR at peak stress (114 ± 2.70 bpm) and prescribed exercise (91 ± 2.38 bpm; P = 0.001) regarding 99m Tc-sestamibi injection, corresponding to 71 ± 1.67 and 57 ± 1.52% of age-predicted maximal HR, respectively.No differences were found between heart rate at the time of 99m Tc-sestamibi injection during prescribed exercise (91 ± 2.38 bpm) and the prescribed HR (90 ± 2.37 bpm).Peak HR was significantly higher during prescribed exercise than at rest (67 ± 2.49 bpm; P < 0.0001) and was significantly higher 119 ± 2.84 bpm than prescribed HR (P < 0.0001).

Myocardial perfusion SPECT
Significant differences (P < 0.0001) were found between mean scores of SPECT images at rest (6.44, 95%CI = 4.09-8.79),peak stress (13.90, 95%CI = 10.48-17.32),and prescribed exercise (10.77, 95%CI = 7.69-13.85)(Table 3).When myocardial ischemia was defined as a difference ≥2 between the summed stress score and summed rest score, the summed difference score was 10.09 ± 1.40 at peak exercise and 6.83 ± 0.99 during prescription exercise (Table 3).Individual analysis showed that during prescription exercise the summed difference score was ≥2 in 25 (64%) patients and <2 in 14 (36%).Among the clinical characteristics considered, namely, prior history of myocardial infarction, prior revascularization, ST-segment depression, and angiographic features, only the latter was significantly different between groups.The incidence of three-vessel disease was greater in the group with a summed difference score ≥2 (Table 4).

Discussion
Among the several methods used to prescribe exercise intensity for coronary artery disease patients, cardiopulmonary exercise testing to determine the anaerobic threshold has come to the forefront, not only because of its widespread availability but also because it is an extremely valuable tool for risk stratification of patients with congestive heart failure at tertiary cardiac centers.In post-myocardial infarction patients receiving beta-blockers, the anaerobic threshold has been considered to be a "gold standard" for exercise prescription (22).The use of the V-slope method to determine training intensity is individualized, since it depends on patient age and fitness level.It is A.R. Fuchs et al.
www.bjournal.com.brsubmaximal and predominantly aerobic, allowing exercises to be performed for prolonged periods (23).The anaerobic threshold has the advantage of being independent of patient motivation and thus more objective and reproducible than maximal oxygen uptake (23,24).However, it has not been established whether this type of prescription is sufficiently safe according to current guidelines, which recommend that exercise should be prescribed for patients with coronary artery disease at a level that does not elicit myocardial ischemia, characterized by angina pectoris and/or ST-segment depression (2).
The present study is justified because exercise intensity immediately below the ischemic threshold has been proven to produce myocardial ischemia, as measured by SPECT scanning (7,25).This may be accomplished by clinical and electrocardiographic monitoring of the rehabilitation session.Yet, in some conditions such as post-acute myocardial infarction, ventricular hypertrophy, and previous ST-segment changes, among others, myocardial perfusion SPECT permits a more accurate identification of the presence of myocardial ischemia during a cardiac rehabilitation program.This method was used in the present study because of its diagnostic superiority.
Although they did not constitute a homogeneous group, the patients included in the present study had coronary artery disease previously documented by coronary angiography.Thirty-one (79.5%) had at least two-vessel disease and 8 (20.5%) had single-vessel disease, but with at least 70% stenosis, justifying the presence of reduced uptake on stress images.
All 39 patients had stable angina or were asymptomatic and were being medically treated by their cardiologists based on clinical and angiographic criteria.These characteristics make these patients similar to those referred for supervised rehabilitation.It can be noted that the group of 39 patients studied was significantly smaller than the initial sample of 153 selected among patients routinely treated at the Department of Nuclear Cardiology of our institution.This may be explained by the need to exclude all patients in whom ST-segment interpretation would be impaired by their baseline conditions.In a Tertiary Center of Nuclear Cardiology, an elevated number of patients are referred for myocardial perfusion SPECT exactly because they cannot be properly assessed by standard exercise stress testing.Another reason for exclusion was patient difficulty in attending a rehabilitation program.
By performing qualitative and semiquantitative SPECT analyses, we could minimize potential biases associated with qualitative interpretation alone.The use of a summed difference score equal to or greater than 2 followed the recommendations of the studies that introduced this technique (21).
Recently, a summed difference score of 2 was considered to be indicative of myocardial ischemia in the COUR-AGE trial (21), a large multicenter study highly influential among cardiologists.Therefore, the summed difference score ≥2 found here in 25 patients (64%) suggests the presence of exercise-induced myocardial ischemia.Although the accuracy of SPECT imaging was not specifically evaluated in the present study, this had already been done in a previous study performed at our institution, showing a high degree of agreement among interpretations (k = 0.810) (26).Based on the difference between the summed stress score and summed rest score, myocardial ischemia, as measured by SPECT perfusion studies, has been quantified as follows: 0-1 = no ischemia; 2-6 = mild-to-moderate ischemia, and >6 = severe ischemia (COURAGE).According to these criteria, 14 patients (35.89%) had no ischemia, 15 (38.47%) had mild-to-moderate ischemia, and 10 (25.64%) had severe ischemia.One-fourth of our patients, therefore, had severe ischemia, suggesting that these patients should have their clinical and electrocardiographic parameters monitored during the first sessions of a rehabilitation program.No significant differences were found regarding age or clinical, angiographic, and electrocardiographic data of patients with a summed difference score ≥2 and <2, except for the presence of three-vessel disease (P = 0.048; Table 4).Perhaps these patients would be candidates for a more thorough clinical evaluation.Patients with myocardial ischemia documented by myocardial SPECT were prescribed workloads below their anaerobic threshold and referred to a special program consisting of interval training during all aerobic phases of rehabilitation.Alternatively, these patients were prescribed muscular resistance exercise at about 40% of maximal contraction, for complementation.The objective of this study was not to recommend the routine use of myocardial perfusion SPECT to evaluate the accuracy of exercise prescription in rehabilitation programs, but rather to open the possibility of using nuclear medicine to assess the degree of ischemia induced by regular rehabilitation exercises in selected patients.
Limitations: this study was conducted at a single center on a small number of patients, and the female gender was underrepresented.The study sample was a convenience sample, i.e., it was composed of patients with ischemic heart disease due to diffuse, extensive atherosclerotic disease recruited from a tertiary center, and should not be considered to be representative of the larger universe of patients with ischemic heart disease for whom supervised exercise programs are indicated.
Clinical implications: our results suggest that, in supervised exercise training programs, exercise prescription based on the anaerobic threshold is still recommended for coronary artery disease with stress-induced myocardial ischemia.Nevertheless, because myocardial ischemia may be present, the presence of ST-segment depression and angina pectoris should determine a reduction in exercise intensity.These recommendations are valid, even after publication of the study by Noel et al. (27), which did not find significant changes in the course of only 11 patients in the presence of ST-segment depression of 1 mm.In patients with diffuse and severe coronary artery disease, myocardial perfusion SPECT may be a useful adjunct for a more appropriate and accurate detection of ischemia, since it enables exercise to be prescribed more safely below the ischemic zone, regardless of the anaerobic threshold.
Based on our results, the following conclusions can be drawn: in patients with ischemic heart disease due to severe, diffuse atherosclerosis, myocardial ischemia was detected by 99m Tc-sestamibi myocardial perfusion SPECT during the continuous aerobic phase prescribed based on the anaerobic threshold, even in clinically stable patients.Angina pectoris and/or ischemic ECG changes have low sensitivity to identify myocardial ischemia during the aerobic phase of a cardiac rehabilitation session.The 64% prevalence of ischemia observed in the study should not be considered representative of the entire population of patients undergoing an exercise program.The changes in patient care and exercise program were implemented as a result of our findings (ischemia during the prescribed exercise).

Table 1 .
. Clinical and angiographic characteristics of the 39 patients who participated in this study.
CABG = coronary artery bypass graft surgery; PCI = percutaneous coronary intervention; ACE = angiotensin-converting enzyme.*Nine patients underwent revascularization procedures.Of these, 7 were reevaluated later, with failures of at least one graft in 5 and patency identified in 2.
wData are reported as means ± SEM and median.P < 0.05 (statistically significant difference between groups).t= t-test; w = Wilcoxon test.Baseline

Table 4 .Table 4 .Table 4 .Table 4 .Table 4 .
Clinical, angiographic and electrocardiographic data according to summed difference score in exercise prescription for 39 patients.Myocardial ischemia is defined as a difference ≥2 between the summed stress score and the summed rest score.*P < 0.05.t = t-test; f = Fisher exact test; p = Pearson chi-square test.