Author + information
- Received May 19, 2000
- Revision received October 20, 2000
- Accepted November 22, 2000
- Published online March 1, 2001.
- Andrew D Kelion, MA, MRCPa,*,
- Terence P Webba,
- Maureen A Gardnera,
- Oliver J.M Ormerod, DM, FRCPa and
- Adrian P Banning, MD, MRCPa
- ↵*Reprint requests and correspondence: Dr. A. D. Kelion, Nuclear Cardiology Department, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom
The goal of this study was to investigate whether the “warm-up” effect in angina protects against ischemic left ventricular (LV) dysfunction.
After exercise, patients with coronary disease demonstrate persistent myocardial dysfunction, which may represent stunning, as well as warm-up protection against further angina, which may represent ischemic preconditioning. The effect of warm-up exercise on LV function during subsequent exercise has not been investigated.
Thirty-two patients with multivessel coronary disease and preserved LV function performed two supine bicycle exercise tests 30 min apart. Equilibrium radionuclide angiography was performed before, during and up to 60 min after each test. Global LV ejection fraction and volume changes and regional ejection fraction for nine LV sectors were calculated for each acquisition.
Onset of chest pain or 1 mm ST depression was delayed and occurred at a higher rate-pressure product during the second exercise test. Sectors whose regional ejection fraction fell during the first test showed persistent reduction at 15 min (68 ± 20 vs. 73 ± 20%, p < 0.0001). These sectors demonstrated increased function during the second test (71 ± 20 vs. 63 ± 20%, p = 0.0005). The reduction at 15 min and the increase during the second test were both in proportion to the reduction during the first test. Effects on global function were only apparent when the initial response to exercise was considered.
The warm-up effect is accompanied by protection against ischemic regional LV dysfunction. The degree of stunning and protection after exercise is related to the severity of dysfunction during exercise, consistent with results from experimental models.
There is increasing evidence that transient ischemia in humans can affect myocardial function in similar patterns to those described in experimental animal models. In dogs, demand ischemia triggers both stunning (reversible contractile dysfunction despite the return of normal myocardial perfusion) and ischemic preconditioning (protection against infarction caused by subsequent coronary occlusion) (1,2). In patients with angina, myocardial contractile function may remain depressed for up to 30 min after the completion of exercise, and this may represent stunning (3–5). Patients may also develop less anginal pain and electrocardiographic (ECG) evidence of ischemia during sequential exercise testing: the “warm-up” phenomenon (6). It is hypothesized that this may represent a form of ischemic preconditioning.
There are important differences of definition between the warm-up effect and ischemic preconditioning. The former involves protection against indirect markers of ischemia (chest pain and ECG changes), whereas the latter represents protection against irreversible myocardial necrosis. Although improvement in echocardiographic indices of left ventricular (LV) function after exercise has been demonstrated, previous studies of the warm-up effect have not looked for protection against ischemic myocardial dysfunction during exercise (7). Systolic dysfunction during exercise is a physiologically meaningful marker of ischemia that might bear closer comparison with ischemic preconditioning.
In order to perform a detailed study of the effect of sequential exercise testing on LV function, it is necessary to use a technique that reproducibly assesses it both globally and regionally. While effects on global function may be more significant clinically, the unpredictability of the response to exercise, even in patients with multivessel coronary disease, makes the ability to assess regional function crucial (8). Equilibrium radionuclide angiography (ERNA) allows resting and exercise global ejection fraction (EF) to be derived with an interstudy variability of approximately 2%, far superior to echocardiography (9). Changes in LV volumes can also be accurately calculated (10). Regional function is assessed by deriving regional EFs for a fixed number of LV sectors with a published interstudy variability of less than 6% (11).
We used ERNA to provide an insight into global and regional LV systolic function before, during and for up to 60 min after two exercise stress tests 30 min apart in patients with multivessel coronary disease. We looked for persistent effects after exercise and for protection against dysfunction during and after the second test.
Thirty-two patients (age 64 ± 7 years, 28 men) with stable angina pectoris were studied a median of 13 weeks after day-case coronary angiography. All had three-vessel coronary disease but preserved LV function (global EF by contrast ventriculography ≥50%). Patients were excluded if they had severe resting ECG abnormalities, were in atrial fibrillation or were diabetic and treated with a sulphonylurea. The study protocol was approved by the local research ethics committee, and all patients gave informed consent before participating.
Patients performed two maximal symptom-limited supine bicycle exercise tests 30 min from end to start. Equilibrium radionuclide angiography acquisitions were taken just before, during and 5, 15, 30 (and, for the second test, 60) min after each test. Exercise acquisitions were commenced at the same exercise time for both tests, at least 1 min into a stage to allow stabilization of heart rate, and the workload was allowed to increase during a scan as required by the exercise protocol. As many exercise acquisitions were obtained as possible, usually two or three, and the final acquisition provided the exercise indices described in the “Results” section. Patients remained on the camera table throughout the protocol.
Exercise stress testing
Patients underwent supine bicycle exercise testing, beginning at a workload of 25W and increasing by 25W every 3 min until limited by symptoms. The ECG lead showing the most profound ST depression during the first exercise test was analyzed for both tests. Patients performed two practice tests on separate days at least one week before the study to familiarize them with the technique. All tests were performed at a room temperature of 22°C and at the same time of day to minimize diurnal variation in exercise capacity. All antianginal drugs were discontinued 72 h before the day of the study.
After in vivo blood-pool labeling with 750-800 MBq of 99mTc-pertechnetate, frame-mode (24 frames per cycle) ERNA was performed in the best-septal left anterior oblique projection according to the usual protocol for our laboratory (12). All resting ERNA acquisitions were continued to a count rate of 200,000 per frame, and exercise acquisitions to 150,000 per frame. An acquisition was interrupted during the first test if the interval between R-waves (R-R interval) increased beyond the preset gating window (mean ±10% at rest, ±15% during exercise) or a second peak started to appear in the R-R histogram. During the second test, the equivalent acquisition was then continued only as far as the number of counts obtained during the first test.
Equilibrium radionuclide angiography processing was automated using commercially available software, but each step was carefully checked and, if necessary, corrected by the observer. Global EF was calculated from the background-corrected time-activity curve using a variable region of interest. In our laboratory, reproducibility of this method shows a mean ± standard deviation difference between two studies on different days of 0.3 ± 4.2 EF points at rest, 2.5 ± 4.8 on exercise and 3.8 ± 5.6 for the change from rest to exercise. Changes in LV end-diastolic (EDV) and end-systolic (ESV) volumes from baseline were calculated taking account of the decay of 99mTc (10).
Regional EFs were calculated for nine radial sectors of equal angle (40°) using a fixed centroid method. Sectors that included counts from the aortic root and mitral valve were not used in the analysis. An ERNA acquisition contains counts in three dimensions, so a given sector does not correspond exactly to a specific coronary artery territory. There is considerable heterogeneity of regional EF for different sector positions in normal subjects (lateral 79 ± 16%, apical 73 ± 15%, septal 41 ± 7%; p < 0.0001). In our laboratory, the interstudy reproducibility of regional EF is 0.2 ± 8.0 EF points at rest, 2.2 ± 9.2 on exercise and 2.1 ± 12.0 for the change from rest to exercise. Different sector positions show comparable reproducibility. Sectors are classified as “ischemic” or “nonischemic” depending on whether the regional EF falls by at least 1 EF point or not during the first exercise test. Concordance for the response of sectors to exercise is 78% for two tests performed 1 h apart with patients remaining on the gamma camera table. In middle-aged subjects with a low probability of coronary disease, only 9% of sectors demonstrate a “false positive” “ischemic” response.
A sample size of 30 patients was calculated to have sufficient power to give a 95% chance of detecting a three-point change in global EF or a two-point change in regional EF at the 95% confidence level. Data are expressed as mean ± standard deviation unless otherwise specified. Continuous variables were compared using paired and unpaired Student ttests or Wilcoxon signed rank and Mann-Whitney Utests where there was deviation from the normal distribution. Nonparametric analysis of variance with the Friedman test was used to compare measurements before, during and after a given exercise test. Multiple pairwise comparisons of indices were corrected using the Bonferroni method (p value multiplied by number of comparisons). Regression lines were constructed using the least-squares method, and the intercepts and gradient were calculated together with their 95% confidence intervals. The quality of the fit was given by the R statistic. A p value of <0.05 was taken to denote statistical significance throughout the paper.
Hemodynamic and ECG data
Baseline heart rate, blood pressure and rate-pressure product (RPP) were comparable between the first and second exercise tests, as were total exercise time, peak RPP achieved and peak ST depression (Table 1). Compared with the first exercise test, the onset of chest pain and the development of 1 mm ST depression were delayed during the second test and occurred at higher values of RPP.
Regional LV systolic function
Baseline regional EF of apical (78 ± 16%) and lateral (81 ± 11%) LV sectors was higher than that of septal sectors (41 ± 9%) (p < 0.0001). Seventy (38%) sectors demonstrated an ischemic response during the first exercise test. Apical (42/93, 45%) and lateral (13/31, 42%) sectors were more likely to be ischemic than septal (15/62, 24%) sectors (p = 0.03), so the mean regional EF of ischemic sectors at baseline was higher than that of nonischemic sectors (p = 0.002; Table 2). The following observations were independent of sector location.
After an initial rebound towards baseline at 5 min, ischemic sectors showed a persistent reduction in regional EF 15 and 30 min after the first exercise test (Table 2, Figure 1). The severity of persistent dysfunction 15 min after exercise increased with the severity of dysfunction during exercise for individual ischemic sectors (Fig. 2). During the second test, ischemic sectors demonstrated a significantly higher regional EF with no further reduction afterwards (Table 2, Fig. 1). The degree of this improvement in function between exercise tests increased with the degree of dysfunction during the first test (Fig. 2).
Nonischemic sectors showed a persistent increase in regional EF 5 and 15 min after the first exercise test, which returned to baseline by 30 min (Table 2, Fig. 1). The degree of improvement 15 min after exercise increased with the degree of improvement during exercise (Fig. 2). Mean regional EF during and after exercise was comparable between the two exercise tests (Table 2, Fig. 1).
Global LV function
Both EDV and ESV increased substantially during exercise although mean global EF did not change (Table 3). Five minutes into recovery, EDV had returned to resting levels, while ESV had fallen below baseline; the result was an “overshoot” in global EF to above the baseline level. From 15 min onward, there was no evidence of persistent global LV dysfunction.
The relation between the response of global LV function to the first exercise test and its behavior 15 min after exercise explained the apparent absence of persistent LV dysfunction after exercise when patients were considered together (Fig. 2). Individuals whose global EF fell during exercise demonstrated a proportionate reduction in EF at 15 min, while those whose EF increased with exercise showed a proportionate increase at 15 min. Similarly, the change in ESV between baseline and 15 min after exercise was related to the magnitude of the increase during exercise.
Mean global EF at equivalent time points for each of the two exercise tests was comparable (Table 3). However, the relation between global LV response to initial exercise and the change between exercise tests revealed a more complex pattern (Fig. 2). When global EF fell during the first exercise test, there was a proportionate improvement during the second test. When EF increased during the first exercise test, there was no clear change during the second test. Similarly, the greater the increase in LV volumes during the first exercise test, the greater the reduction from the first to the second exercise test.
During sequential exercise testing, patients with coronary disease exhibit protection not only against angina and ECG changes but also against ischemic LV dysfunction. Left ventricular dysfunction induced by ischemia persists beyond the completion of exercise, and this is consistent with myocardial stunning. Left ventricular exercise responses are heterogeneous, but the development of stunning is predicted by the severity of dysfunction during exercise.
LV function during and after exercise
We have confirmed the findings of Ambrosio and colleagues (5)that regional EF of ischemic LV sectors remains depressed for up to 30 min after exercise. This is consistent with earlier studies using echocardiography and is likely to reflect myocardial stunning (3,4). In addition, we have shown that the severity of the persistent dysfunction is related to the severity of dysfunction during exercise. This is consistent with data from dogs exercising with a fixed coronary stenosis: the severity of postischemic stunning depends on the degree of ischemic myocardial dysfunction during exercise, which itself depends on the severity of subendocardial hypoperfusion (13).
The global LV volume changes observed during the first exercise test in our study are typical for patients with coronary disease undergoing supine bicycle exercise (10). In healthy volunteers, EDV barely changes, while a 30% fall in ESV leads to an increase in global EF. Conversely, in patients with coronary disease and preserved resting LV function, both EDV and ESV increase by approximately 30%. The net effect on global EF is highly variable, even among patients with multivessel coronary disease, but a fall is associated with a high incidence of cardiac events and is likely to reflect more severe global ischemia (8).
Overshoot in global EF in the immediate postexercise period has been previously described (14). Left ventricular volume measurements suggest that this is primarily due to a rebound improvement in inotropic status. Other workers have also failed to demonstrate evidence of persistent global LV dysfunction (stunning) using ERNA after exercise when all patients are considered together (5,14,15). Our data demonstrate that this is related to the heterogeneity of the response of global LV function to exercise, despite coronary disease of uniform angiographic severity. The larger the fall in global EF or rise in LV volumes with exercise, the more likely were these indices to remain abnormal 15 min after exercise. Improved regional function of nonischemic sectors after exercise may represent compensation for the stunned ischemic sectors, and this would prevent global dysfunction in all but the most globally ischemic hearts.
The warm-up effect
In our study, the onset of chest pain and a given severity of ST depression was delayed and occurred at a higher RPP during the second exercise test. Patient-limited exercise time may not have increased during the second test because leg fatigue, rather than cardiac ischemia, is usually the chief determinant during supine bicycle exercise. These results represent the warm-up effect as previously demonstrated using treadmill exercise (16), upright and supine bicycle exercise (6,17)and ambulatory ECG monitoring during daily activity (18).
During the second exercise test, the mean regional EF of ischemic LV sectors increased instead of falling, suggesting that warm-up protection extends to ischemic contractile dysfunction. The degree of protection increased with the severity of ischemia during the first exercise test. This pattern is similar to that observed in canine models of ischemic preconditioning against infarction (2). The infarct size in a given animal is inversely related to the collateral myocardial blood flow to the risk area, as is the reduction in infarct size produced by preconditioning (i.e., the degree of protection increases with the severity of hypoperfusion). The absence of improvement in global LV function during exercise in our study when all patients were considered together was due to the heterogeneity of the initial response to exercise. Protection occurred only in relation to the severity of dysfunction during the first exercise test.
The function of ischemic sectors after exercise was comparable for the two tests, suggesting that cumulative stunning had not occurred. This has been demonstrated echocardiographically in a previous study in which LV function during exercise was not measured and was interpreted as evidence of early preconditioning against stunning (7). This conclusion is at variance with the results of animal studies in which early preconditioning against stunning has not been convincingly demonstrated although delayed protection does develop at approximately 24 h (19,20). Our data suggest that the apparent early protection against stunning, which follows serial exercise testing in patients with coronary disease, is simply due to protection against the preceding ischemic LV dysfunction.
An important criticism of all studies of the warm-up effect is the inability to quantify myocardial perfusion accurately in a clinical setting. Protection may, therefore, be due to improved perfusion via collaterals rather than preconditioning. In patients with a single proximal left anterior descending artery stenosis, the warm-up effect is not accompanied by an increase in blood flow within the great cardiac vein (17). This suggests that there is no increase in overall blood supply although intramural redistribution of myocardial blood flow cannot be excluded.
We have demonstrated that the warm-up effect against angina and ECG changes is accompanied by protection against ischemic regional LV dysfunction. The degree of stunning and protection after exercise is proportional to the severity of dysfunction during exercise, which is consistent with patterns observed in experimental models. Harnessing the processes responsible for the warm-up effect may provide novel therapeutic approaches for patients with coronary disease.
☆ Dr. Kelion is supported by a British Heart Foundation Junior Research Fellowship.
- electrocardiogram or electrocardiographic
- end-diastolic volume
- ejection fraction
- equilibrium radionuclide angiography
- end-systolic volume
- left ventricle or ventricular
- rate-pressure product
- Received May 19, 2000.
- Revision received October 20, 2000.
- Accepted November 22, 2000.
- American College of Cardiology
- Iwamoto T.,
- Bai X.-J.,
- Downey H.F.
- Scognamiglio R.,
- Ponchia A.,
- Fasoli G.,
- Miraglia G.,
- Dalla-Volta S.
- Ambrosio G.,
- Betocchi S.,
- Pace L.,
- et al.
- Rinaldi C.A.,
- Masani N.D.,
- Hall R.J.
- Hecht H.S.,
- Josephson M.A.,
- Hopkins J.M.,
- Singh B.N.,
- Parzen E.,
- Elashoff J.
- Homans D.C.,
- Laxson D.D.,
- Sublett E.,
- Pavek T.,
- Crampton M.
- Tomai F.,
- Crea F.,
- Danesi A.,
- et al.
- Okazaki Y.,
- Kodama K.,
- Sato H.,
- et al.
- Tzivoni D.,
- Maybaum S.
- Ovize M.,
- Przyklenk K.,
- Hale S.L.,
- Kloner R.A.