Author + information
- Received March 10, 1997
- Revision received March 11, 1998
- Accepted April 9, 1998
- Published online July 1, 1998.
- Gianni Mobilia, MDa,* (, )
- Pierluigi Zanco, MD∗,
- Alessandro Desideri, MD∗,
- Gianfilippo Neri, MDa,
- Ferdinando Alitto, MDa,
- Gianleone Suzzi, MD∗,
- Franca Chierichetti, MD∗,
- Leopoldo Celegon, MD∗,
- Giorgio Ferlin, MD∗ and
- Riccardo Buchberger, MDa
- ↵*Address for correspondence: Dr. Gianni Mobilia, Via Monte Pallone 3, 31044 Montebelluna (TV), Italy
Objectives. We investigated the sensitivity and specificity of exercise-induced T wave normalization (TWN) in infarct-related electrocardiographic leads (IRLs) for detection of residual viability in the infarct area.
Background. The meaning of exercise-induced TWN on IRLs is not yet well understood. Recent reports suggest that TWN during dobutamine echocardiography could indicate the presence of viable myocardium.
Methods. We evaluated 40 consecutive patients with a recent acute myocardial infarction and negative T waves in at least two IRLs. All patients underwent exercise testing; positron emission tomography (PET) with nitrogen-13 ammonia and fluorine-18 fluorodeoxyglucose; and coronary angiography.
Results. Twenty-four patients showed exercise-induced TWN: 18 at a work load ≤50 W (group 1a) and 6 at a work load ≥75 W (group 1b); 16 patients did not show TWN (group 2). On the PET study, viability in the infarct area was present in 17 patients (94%) from group 1a, in only 1 (16%) from group 1b and in 4 (25%) from group 2 (p < 0.0001). The sensitivity, specificity and diagnostic accuracy of exercise-induced TWN, in comparison with residual viability, were, respectively, 82%, 67%, 75% for TWN at every work load and 77%, 94%, 85% for TWN at a work load ≤50 W. Moreover, the sensitivity and diagnostic accuracy of TWN at the low work load were higher for anterior infarctions (87% and 88%, respectively).
Conclusions. Exercise-induced TWN on IRLs at low work loads is a sensitive and specific index for the presence of residual viability in the infarct area. Sensitivity and diagnostic accuracy of this sign are higher for anterior infarctions.
The meaning of exercise-induced T wave normalization (TWN) is not yet fully understood. Although this electrocardiographic (ECG) sign has been considered (1)(or not ) a marker of ischemia, it is also present in many other nonischemic conditions, such as mitral valve prolapse, left ventricular hypertrophy and metabolic abnormalities. TWN can also occur in healthy young adults.
Exercise-induced TWN is often present in patients with ischemic heart disease. In patients with a previous myocardial infarction, TWN on infarct-related ECG leads (IRLs), during low dose dobutamine echocardiography has been correlated with the presence of viable myocardium and late functional recovery (3). Moreover, some previous reports (4,5)have suggested that some modifications of ECG repolarization induced by exercise could indicate the presence of viability.
Identification of residual viability of areas of the myocardium that could be saved by a revascularization procedure is very important information for the modern cardiologist. Several imaging techniques, such as dobutamine (6–9)or dipyridamole (10)or post-extrasystolic potentiation (11)echocardiography and thallium-201 redistribution or reinjection scintigraphy (12–14), as well as fluorine-18 fluorodeoxyglucose (F-18 FDG) single-photon emission computed tomography (15), have been proposed and are currently used to assess the presence of residual viability. Positron emission tomography (PET) has proved to be the technique with the greatest sensitivity in identifying viable myocardium in the infarct areas and, at present, is considered the “gold” standard for assessing residual myocardial viability (16–18).
The aim of the present study was to verify whether, similar to low dose dobutamine infusion, exercise-induced TWN on IRLs could be a marker of residual viability in infarct areas, assuming PET as the reference standard.
We studied 40 consecutive patients (35 men, 5 women; mean [±SD] age 60 ± 7 years, range 40 to 74) with a recent, uncomplicated first Q wave acute myocardial infarction (26 anterior, 14 inferior) and persistent negative T waves on at least two IRLs. Diagnosis was based on a history of typical chest pain, ECG modification and serum enzyme determinations. All patients were studied before or immediately after discharge. All patients were in New York Heart Association functional class I or II.
Patients with left ventricular hypertrophy, left or right bundle branch block, diabetes or glucose intolerance, as well as patients who could not perform exercise testing, were excluded.
All patients underwent maximal symptom-limited ergometric testing between the 15th and 25th day after the onset of the symptoms, according to a protocol of 25-W increases every 3 min.
Antianginal medications were stopped at least 72 h before the test. A 12-lead ECG and systolic and diastolic blood pressures (cuff sphygmomanometer) were recorded at rest, during the third minute of each exercise stage, at peak exercise, 1 min after exercise and every 3 min into recovery. Leads II, V1and V5were continuously monitored.
Criteria for interrupting the test were target heart rate (220 minus age), severe chest pain, complex ventricular arrhythmias, hypotension, exhaustion, ≥2-mm ST segment depression, ≥2-mm ST segment elevation on non-IRLs. ST segment elevation and TWN on IRLs were not considered criteria for test interruption.
TWN was defined as negative T waves that became upright in two or more IRLs during exercise testing. All TWN (with or without concomitant ST segment elevation) and isolated TWN (without ST segment elevation) were both considered in the data analysis. Interpretation of TWN and ST segment deviation was performed in blinded manner by two independent observers (G.N., F.A.); disagreement was resolved by consensus.
PET studies were performed within 7 days of exercise testing, using a scanner ECAT EXACT, which allows simultaneous acquisition of 47 contiguous transaxial images, with a total axial field of view of 16.2 cm. The resolution of our scanner was 4.8 ± 0.6 mm in the axial direction and 6.1 ± 0.2 mm in transaxial planes. A transmission scan was first obtained for 15 min for attenuation correction, using retractable germanium-68 ring sources. For the PET studies, the tracers were nitrogen-13 (N-13) ammonia (dose 10 MBq/kg body weight) for perfusion, injected at rest, and F-18 FDG (dose 4 MBq/kg) for metabolism, injected 45 min after an oral glucose load (50 g). The PET scan started 4 min after the injection of N-13 ammonia and 45 min later for F-18 FDG. The acquisition lasted 15 min for both tracers. Short-axis and vertical and horizontal long-axis slices, each 0.8 cm thick, were reconstructed using a Hanning filter (cutoff 1.18 cycle/cm) and corrected for attenuation. Both studies were performed on the same day in all patients, with the N-13 ammonia study first, followed 2 h later by the F-18 FDG study. To verify the position of the patients in the scanner, a cross-shaped, low power laser beam and pen skin markers were used.
For image analysis, the left ventricular wall was divided into six segments: anterior, apical, septal, lateral, posterior and inferior. After normalization at the maximal count in the left ventricular wall, the relative percent uptake of F-18 FDG was then calculated in the regions of interest (8 × 8 mm in size) drawn in the middle region of each segment. A semiquantitative four-point score was applied: 1 = normal uptake(≥75%); 2 = moderate defect(50% to 74%); 3 = severe defect(25% to 49%); 4 = absent uptake(≤24%) A dysfunctional segment was considered viable when its uptake was ≥50% of the maximal uptake in the left ventricular wall. With regard to perfusion, a similar semiquantitative classification was applied. In each patient, a perfusion–metabolism mismatch was considered present if the N-13 ammonia score was superior to the F-18-FDG score by at least one point.
All patients underwent cardiac catheterization with selective coronary angiography and left ventriculography within 20 days of exercise testing. Biplane left ventriculography was performed in the 30° right anterior and 60° left anterior oblique projections. Coronary artery stenosis was considered significant if the lumen diameter was narrowed by ≥50%. Angiograms were evaluated in blinded manner by two observers. The left ventricle was divided into six segments, similar to PET imaging, and each segment was defined as normal, hypokinetic (reduced systolic wall motion), akinetic(no systolic wall motion) and dyskinetic(paradoxic systolic expansion). Left ventricular ejection fraction was calculated by the area–length method (19).
Correlations between ECG, PET imaging and coronary angiography
To correlate the location of ECG abnormalities with PET and angiographic abnormalities, two ECG sites were considered: 1) anterior (leads V1to V4), which was assigned to the anterior wall, septum and apex and to the left anterior descending coronary artery; 2) inferolateral (leads II, III, aVF, I, aVL, V5, V6), which was assigned to the lateral, inferior and posterior wall and to the left circumflex and right coronary arteries.
On PET examination, an infarct area was considered viablewhen at least 50% of the Q wave corresponding segments showed F-18 FDG uptake ≥50% of maximum, as described earlier.
Results are presented as mean value ± SD. Statistical significance of continuous variables was determined with a two-tailed Student ttest. Parametric variables were compared by the chi-square test or Fisher exact test, when appropriate. A p value ≤0.05 was considered statistically significant.
Clinical and exercise testing data
Twenty-four patients (22 men, 2 women; mean age 58 ± 8 years, range 40 to 74) showed exercise-induced TWN on more than one IRL. Of these 24 patients, 18 (17 men, 1 woman; mean age 59 ± 8 years, range 45 to 74), 15 with an anterior (83%) and 3 with an inferior infarction (17%), showed TWN at a work load ≤50 W and constitute group 1a; the other 6 patients (5 men, 1 woman; mean age 56 ± 7, range 40 to 72), 5 with an anterior (83%) and 1 with an inferior infarction (17%), showed TWN at a work load ≥75 W and constitute group 1b. The remaining 16 patients (13 men, 3 women; mean age 62 ± 5, range 46 to 69), 6 with an anterior (37%) and 10 with an inferior infarction (63%), did not show TWN and constitute group 2. During exercise, TWN occurred at 87 ± 9 beats/min in group 1a and 116 ± 11 beats/min in group 1b (p = 0.003).
Intravenous thrombolytic therapy was administered within 6 h of the onset of symptoms in 12 patients (75%) from group 1a, 6 (100%) from group 1b and 11 (68%) from group 2 (p = NS).
Eighteen patients (45%) had exercise-induced ST segment elevation on the IRLs: eight from group 1a, one from group 1b and nine from group 2 (p = NS). The isolated TWN (without ST segment elevation) occurred in 10 patients (55%) from group 1a and 5 (83%) from group 1b (p = NS).
Four patients (22%) from group 1a, two (33%) from group 1b and four (25%) from group 2 had ST segment depression ≥1 mm on non-IRLs. Only one patient from group 2 had ST segment depression ≥1 mm on the IRLs. Four patients (22%) from group 1a, two (33%) from group 1b and five (31%) from group 2 experienced angina during the test.
Total exercise time was 647 ± 157 s in group 1a, 719 ± 173 s in group 1b and 536 ± 135 s in group 2 (p = NS). The rate–pressure product at peak exercise was 21,211 ± 3,545 beats/min × mm Hg in group 1a, 19,036 ± 6,420 beats/min × mm Hg in group 1b and 20,832 ± 4,192 beats/min × mm Hg in group 2 (p = NS).
Coronary angiography and left ventriculography
Left ventricular ejection fraction was 60 ± 3% (range 40% to 70%) in group 1a, 57 ± 5% (range 50% to 65%) in group 1b and 57 ± 8% (range 34% to 70%) in group 2 (p = NS).
Of the 18 patients from group 1a, 4 had three-vessel disease, 5 had two-vessel disease, and 9 had one-vessel disease. Of the six patients from group 1b, three had two-vessel disease, and three had one-vessel disease. Of the 16 patients from group 2, 4 had three-vessel disease, 6 had two-vessel disease, and 6 had one-vessel disease.
The infarct-related coronary artery was occluded in four patients (22%) from group 1a, one (16%) from group 1b and three (18%) from group 2 (p = NS). Collateral flow in the infarct-related coronary artery was present in seven patients (38%) from group 1a, three (50%) from group 1b and seven (43%) from group 2 (p = NS).
Clinical, ECG and coronary angiography data for the study group are presented in Table 1.
Myocardial viability in the infarct area was present in 22 patients: 17 from group 1a (94%), 1 from group 1b (16%) and 4 from group 2 (25%) (p < 0.0001). Moreover, viability was present in 11 (73%) of the 15 patients with isolated exercise-induced TWN (without ST segment elevation) and in 7 (77%) of the 9 patients with concomitant exercise-induced TWN and ST segment elevation (p = NS).
Among the patients with maintained F-18 FDG uptake, a metabolism–perfusion mismatch was observed in seven patients (41%) from group 1a, one (100%) from group 1b and three (75%) from group 2 (p = NS).
Left ventriculographic, exercise testing and PET data are presented in Table 2.
The sensitivity, specificity and diagnostic accuracy of exercise-induced TWN at every work load and at a work load ≤50 W, for all patients and for patients with an anterior infarction only, compared with residual viability in the infarct area, are presented in Figure 1. An example of the correlation between exercise-induced ECG modifications and PET findings in a patient from group 1a is shown in Figures 2 and 3. ⇓
Exercise-induced TWN may have different meanings when it occurs in patients without heart disease, in patients with ischemic heart disease but without a previous acute myocardial infarction or in patients with a previous Q or non–Q wave acute myocardial infarction. In healthy adults, often in athletes, exercise-induced TWN could depend on a neurogenic mechanism caused by sympathetic stimulation and is not considered an abnormal sign (20,21).
In patients with ischemic heart disease, exercise-induced TWN was previously considered a possible marker of ischemia (1), but this hypothesis was not confirmed in other studies (2,22,23). Nevertheless, TWN during dobutamine infusion has been identified by echocardiography (24)as an accurate marker of ischemia in patients with a non–Q wave myocardial infarction. In contrast, in patients with a previous Q-wave myocardial infarction, exercise-induced TWN has never been considered a sign of ischemia.
Salustri et al. (3)demonstrated by echocardiography that TWN on the IRLs during low dose dobutamine infusion can be a marker of residual viability in the infarct area, with good accuracy for predicting late mechanical recovery (3).
Our study shows that in patients with a recent Q wave acute myocardial infarction, exercise-induced TWN on the IRLs could be related to the presence of residual viability.
We investigated the meaning of this phenomenon by means of PET F-18 FDG uptake as a marker of viability. In fact, this tracer has been shown (25)to find residual metabolic activity in myocardial areas in which other techniques had revealed complete necrosis. We chose F-18 FDG PET as the reference standard for viability because we wished to determine whether metabolic activity would correspond to TWN. Future studies should investigate the possible correspondence of this ECG sign with late effective functional recovery.
The possibility that exercise-induced modifications of ECG ventricular repolarization could indicate the presence of residual myocardial viability has been previously demonstrated. Margonato et al. (5)showed that exercise-induced ST segment elevation on the IRLs has a high specificity for detection of residual viability in the infarct area. Our study focused instead on the fact that exercise-induced TWN, with or without concomitant ST segment elevation, is a good predictor of myocardial viability.
It is well known that exercise causes catecholamine release (26)and may determine TWN in normal myocardium. Therefore, because it occurs during dobutamine infusion (3), sympathetic stimulation during exercise may induce TWN on the IRLs in the presence of still viable tissue.
In our study, TWN at the lower work load (≤50 W) appeared to be more specific and accurate for the diagnosis of viability. This finding could be dependent on the fact that, in the presence of a high number of viable cells, a low sympathetic stimulation and, consequently, a low load could be sufficient to reveal the presence of viability. In contrast, if only a few viable cells are present, greater sympathetic stimulation and therefore a higher effort level might be necessary to determine TWN. Moreover, at the high work load, the possible presence of homozonal (i.e., in regions supplied by the infarct-related artery) or heterozonal (i.e., in regions supplied by non–infarct-related arteries) ischemia might confuse the meaning of TWN. The finding that some patients had exercise-induced ST segment depression without evidence of viability at PET could reflect the presence of heterozonal or peri-infarct ischemia.
Our results agree with these of Lombardo et al. (27)who found that during dobutamine echocardiography, only TWN occurring at the low dobutamine dose was indicative of the presence of contractile reserve in the infarct-related area, whereas TWN occurring at the high dobutamine dose was specific for homozonal ischemia. Moreover, in the present study, this sign proved to be more sensitive and accurate in patients with an anterior infarction. Similar results were obtained by Margonato et al. (5)in their work on exercise-induced ST segment elevation.
The mean heart rate at which TWN occurred was obviously higher in patients from group 1b than in those from group 1a. Heart rate could have importance in causing TWN in patients with viable myocardium. From our findings, it is not possible to separate the relative contribution of heart rate and work load in causing TWN because they concomitantly increase during exercise; a study with transesophageal stimulation would be useful to resolve this issue.
In our patients, only a minority (20%) had a left ventricular ejection fraction ≤50%. After the study protocol, all patients were evaluated by PET despite their left ventricular ejection fraction, to compare the finding of ergometric testing with the presence or absence of viability. In the routine clinical management of patients with a recent acute myocardial infarction, the PET study should be reserved at present for patients with a low left ventricular ejection fraction because the clinical value of assessing viability in patients with normal ventricular function remains to be determined.
Finally, the results of the present study could also explain how the TWN during exercise testing is, in many studies (2,22,23), a poor predictor of myocardial ischemia. In fact, this ECG sign may show only the presence of viable, but not necessarily ischemic, myocardium.
Limitations of the study
Our group of consecutive patients had a larger incidence of anterior than inferior infarction (65% vs. 35%), most likely because the ECG pattern of negative T waves in at least two IRLs is more frequent in anterior than inferior infarction, where the T waves are often isodiphasic or not clearly negative.
We did not investigate the presence of ischemia with techniques other than exercise testing. However, ischemia, possibly present in the infarct area, should represent a good sign of viability. We cannot totally exclude the possibility that TWN on the IRLs could be a sign of heterozonal ischemia. Nevertheless, this appears unlikely in light of the coronary angiographic findings and the extent TWN that occurs at low load.
Conclusions and clinical implications
The results of our study suggest that in patients with a recent Q wave acute myocardial infarction, exercise-induced TWN on the IRLs is an accurate marker of residual viability in the infarct area. This sign is highly accurate at low work loads and in anterior infarctions. Exercise testing can therefore be used as a screening method for the presence of residual viability, leaving other diagnostic tools, such as stress echocardiography or nuclear medicine techniques, as second-line strategies. Further perspective studies are needed to test the hypothesis that exercise-induced TWN can predict late mechanical recovery.
- electrocardiogram, electrocardiographic
- F-18 FDG
- fluorine-18 fluorodeoxyglucose
- infarct-related ECG leads
- positron emission tomography (tomographic)
- T wave normalization
- Received March 10, 1997.
- Revision received March 11, 1998.
- Accepted April 9, 1998.
- American College of Cardiology
- Yu P.N.G,
- Soffer A
- Wagoner L.F,
- Movahed A,
- Reeves W.C,
- Jolly S.R
- Margonato A,
- Ballarotto C,
- Bonetti F,
- et al.
- Margonato A,
- Chierchia S.L,
- Xuereb R.G,
- et al.
- Pierard L.A,
- De Landsheere C.M,
- Berthe C,
- Rigo P,
- Kulbertus H.E
- Barilla F,
- Gheorghiade M,
- Alam M,
- Khaja F,
- Goldstein S
- Cigarroa C.G,
- deFilippi C.R,
- Brickner M.E,
- Alvarez L.G,
- Wait M.A,
- Grayburn P.A
- Smart S.C,
- Sawada S,
- Ryan T,
- et al.
- Varga A,
- Ostojic M,
- Djordjevic-Dikic A,
- et al.
- Scognamiglio R,
- Fasoli G,
- Casarotto D,
- et al.
- Dilsizian V,
- Perrone Filardi P,
- Arrighi J.A,
- et al.
- Ratib O,
- Phelps M.E,
- Huang S.C,
- Henze E,
- Selin C,
- Schelbert H.R
- Gropler R.J,
- Siegel B.A,
- Sampathkumaran K,
- et al.
- Frais M.A,
- Hoeschen R.J
- Elhendy A,
- Geleijnse A,
- Salustri A,
- et al.
- Brunken R.C,
- Scwwaiger M,
- Grover-McKay M,
- Phelps M.E,
- Tillisch J,
- Schelbert H.R
- Andre F.F,
- Metz K.F,
- Drash A.L
- Lombardo A,
- Loperfido F,
- Pennestrı̀ F,
- et al.