- Email: [email protected]
Impaired left ventricular function is associated with increased recovery time dispersion in patients with previous myocardial infarction
Journal of Electrocardiology Vol. 36 No. 1 2003
Impaired Left Ventricular Function Is Associated With Increased Recovery Time Dispersion in Patients With Previous Myocardial Infarction
Takanao Mine, MD, Hiroki Shimizu, MD, Takanori Inazumi, MD, and Tadaaki Iwasaki, MD
Abstract: Left ventricular (LV) dysfunction after myocardial infarction is associated with higher risk of serious ventricular arrhythmias and sudden death. We suspected that heterogeneity in ventricular repolarization contributes to these arrhythmias. To quantify this heterogeneity, we measured the recovery time (the interval between QRS onset and the time of maximum dV/dt in the ST-T segment) using an 87-lead body surface mapping electrocardiogram and estimated recovery time dispersion (the difference between maximum recovery time and minimum recovery time) in each lead. Differences between 110 patients with previous myocardial infarction and 31 healthy controls were compared. Recovery time dispersion [medians (25th, 75th percentiles)] was greatest in patients with a dilated LV {169 ms (154, 201) vs. 155 ms (137, 172), P ⬍ .005}, impaired ejection fraction {173 ms (155, 202) vs. 152 ms (138, 165), P ⬍ .0005} and LV dyskinesis {175 ms (159, 201) vs. 155 ms (137, 161), P ⬍ .0005}. This study suggests that LV dysfunction associated with myocardial infarction leads to heterogeneous ventricular repolarization and may provide the electrical substrate for ventricular arrhythmias. Key words: Repolarization abnormalities, previous myocardial infarction, impaired left ventricular function, recovery time, ventricular arrhythmias.
Left ventricular (LV) dysfunction after myocardial infarction (MI) is associated with higher risk of serious ventricular arrhythmias and sudden death (1– 4). However, the mechanism responsible for
ventricular arrhythmias in patients with LV dysfunction is unclear. Heterogeneity of repolarization is known to increase susceptibility to reentrant arrhythmias (5–7), and QT dispersion has been reported to be a high risk factor for serious ventricular arrhythmias (8). Recently it was found that QT dispersion is reduced by successful thrombolytic therapy following acute MI (9) and by administration of angiotensin converting enzyme inhibitors in patients who have coronary artery disease (10). These observations suggest that impaired LV function increases the dispersion of ventricular repolarization. However, the relationship between QT dis-
From the Department of Internal Medicine, Cardiovascular Division, Hyogo College of Medicine, Nishinomiya, Japan. Reprint requests: Hiroki Shimizu, MD, Department of Internal Medicine, Cardiovascular Division, Hyogo College of Medicine, 1-1, Mukogawa-cho, Nishinomiya, 663-8501, Japan; e-mail: [email protected]. Copyright 2003, Elsevier Science (USA). All rights reserved. 0022-0736/03/3601-0001$35.00/0 doi:10.1054/jelc.2003.50008
1
2 Journal of Electrocardiology Vol. 36 No. 1 January 2003 persion and LV function is unclear. The QT interval can usually be measured only from the standard 12-electrocardiogram (ECG) leads, and it is sometimes difficult to define the QT interval precisely by measuring it manually in patients who have had MI. Conversely, the activation-recovery interval, measured as the interval between the times of the most rapid decrease voltage (minimum dV/dt) of the QRS and the maximum rate of voltage increase (maximum dV/dt) of the T wave, is closely correlated with the action potential duration, defined as the interval between times of the maximum dV/dt of the upstroke and the minimum dV/dt of the downstroke of the transmembrane action potential (11). Several studies have shown that dispersion of recovery time (RT; defined as the interval between QRS onset and the time of maximum dV/dt in the ST-T segment) measured by 87-lead body surface mapping is useful for estimating dispersion of ventricular repolarization (12,13). In the present study, we investigated the relationship between LV function and dispersion of recovery time as measured by body surface mapping in patients with previous MI.
Materials and Methods Study Patients The study population included 110 patients (93 men, 17 women; mean age, 64 years) who had suffered MI at least 6 months prior to enrollment. Patients taking antiarrhythmic agents or -blockers were excluded. Patients with atrial fibrillation, paced rhythm, bundle branch block, or AV block also were excluded. The site of the MI was anterior in 64 patients, inferior in 44 patients, and both sites were affected in 2 patients. Five patients presented with ventricular arrhythmias (4 with sustained ventricular tachycardia and 1 with ventricular fibrillation). We performed a case controlled study in 31 patients with normal coronary angiograms to identify RT characteristics of MI. Demographics are presented in Table 1. Cardiac Catheterization Cardiac catheterization was performed in all patients according to the Judkin’s technique after a written informed consent had been obtained. Left ventricular end-diastolic pressure and systolic pressure were measured with a pigtail angiographic catheter. Biplane left ventriculography was per-
Table 1. Clinical and Angiographic Data
Male/Female Age (y.o.) Infarction location Anterior Inferior Anterior ⫹ Inferior Ventricular arrhythmia Ventricular tachycardia Ventricular fibrillation Ejection Fraction (%) End-diastolic volume (ml) LV dyskinesis
Control
MI
23/8 64 (56, 69)
93/17 64 (58, 71)
68 (63, 73) 123 (107,128)
64 44 2 5 4 1 52 (40, 64) 143 (118,175) 27
Data presented are expressed as medians (25th, 75th).
formed in 30-degree right anterior oblique and 60-° left anterior oblique views. Left ventricular enddiastolic volume and end-systolic volume were calculated by the area-length method from the biplane views. Left ventricular ejection fraction was calculated as the ratio (end-diastolic volume-end-systolic volume)/end-diastolic volume. The same investigator conducted all studies. An end-diastolic volume ⬎ 180 ML, ejection fraction ⬍40%, and enddiastolic pressure ⬎ 15 mm Hg were used to define a dilated LV, impaired ejection fraction, and increased end-diastolic pressure, respectively. Body Surface Mapping Body surface mapping was performed using the VCM 3000 (Fukuda Denshi Co., Tokyo, Japan). Eighty-seven body surface leads were arranged in a lattice-like pattern (13 ⫻ 7 matrix), excluding the leads in the midaxillary line (59 leads located on the anterior chest and 28 leads on the back). An ECG from these 87 unipolar leads with Wilson’s central terminal as reference, a standard 12-lead ECG, and Frank X, Y, and Z lead scalar ECG, were recorded simultaneously. Sampling frequency was 1,000 samples/second/channel recorded with the patient resting in a supine position. These data were stored on a floppy disk in digital format. The mapping data were transferred to a personal computer (PC-9801 FA, NEC, Tokyo, Japan) with an analysis program developed by Shimizu et al. (12,13). A high-resolution color monitor and printer were connected to the computer. The QRS onset, offset, and T offset were determined visually from the superimposed Frank X, Y, and Z leads and the spatial magnitude. RT was defined as the interval between QRS onset and the time of maximum dV/dt in the ST-T segment and was measured automatically by the com-
LV Function and RT Dispersion •
Mine et al. 3
Fig. 1. The measurement of recovery time in some morphologies of T wave. T wave was positive, and maximum dV/dt occurred before the peak of T wave. T wave was negative, and maximum dV/dt occurred after the peak of T wave. T wave was biphasic, and maximum dV/dt occurred on the upstroke of the positive peak. The points of maximum dV/dt in flat T wave were indeterminate, then the lead was excluded from analysis.
puter. (Fig. 1) The maximum dV/dt points were checked visually on the high-resolution color monitor. When the points were incorrect, the cursor was manually moved to the actual maximum dV/ dt. When the points were indeterminate, the lead was excluded from analysis. RT value was corrected (RTc) according to Bazett formula. Maximum RT and RTc (max RT and max RTc) and minimum RT and RTc (min RT and min RTc) were measured in all 87 leads. RT and RTc dispersion were defined as the difference between maximum and minimum RT and RTc. We examined the regional differences in recovery time interval between controls and patients with previous MI, and the relationship between LV function and dispersion of recovery time in patients with previous MI.
Statistical Analysis All data are expressed as the medians (25th, 75th). Comparisons of recovery time were made by the Mann-Whitney U test or Kruskal-Wallis test. Linear regression analysis was used to assess the relationship between RTd and the index of left ventricular function. A P value ⬍ .05 was considered to be statistically significant.
Results Recovery Time in Patients With MI A representative 87-lead ECG from a control subject and a patient with anterior MI are shown in Figure 2. In the control subject, the max RT (350 ms) was recorded from the C7 lead, and the min RT (224 ms) was recorded from the F4 lead. Thus the RT dispersion was 126 ms. The patient with the anterior MI had sustained ventricular tachycardia and LV dyskinesis (ejection fraction 34%), and the max RT (426 ms) was recorded from the F2 lead, and the min RT (250 ms) was recorded from the E3 lead, so RT dispersion was 176 ms. The max RT, min RT, and RT dispersion in controls and in patients with MI are presented in Table 2. The max RT and RT dispersion were greater in patients with an anterior MI than in control patients or in patients with an inferior MI. The RT interval in each lead was compared with anterior MI, inferior MI and healthy controls. Comparisons of RT interval in patients with an anterior and inferior MI to controls are shown in Figs. 3A and 3B. The recovery time in each lead was compared in patients with an anterior to inferior MI (Fig. 3C). The RT in patients with anterior MI was generally greater over the left side of the chest, and
Fig. 2. The 87-lead ECG obtained by body surface mapping from a control subject and a patient with anterior MI having sustained ventricular tachycardia. (A) The max RT was recorded from the C7 lead, and the min RT was recorded from the F4 lead of a 64 year old woman. (B) The max RT was recorded from the F2 lead, and the min RT was recorded from the E3 lead from a 66-year-old man.
LV Function and RT Dispersion • Table 2. Max RT, min RT, and RT Dispersion
Control Anterior MI Inferior MI
max RT
min RT
RT Dispersion
392 (368, 410) 415 (393,436)* 392 (373, 426)
250 (231,268) 249 (238,260) 250 (232,262)
140 (123, 150) 163 (151,189)† 144 (136, 163)
*P ⬍ .005 vs. control and inferior MI; †P ⬍ .001 vs. control and inferior MI;
smaller over the lower back of the chest compared with controls. RT in patients with inferior MI was generally greater over the lower back of the chest and lower over the upper side of the anterior chest compared with controls. RT in patients with anterior MI was greater over the upper side of the anterior chest and left side of the chest, and smaller
Mine et al. 5
over the lower back of the chest compared with inferior MI. Recovery Time in Patients With Ventricular Arrhythmias and MI A serious ventricular arrhythmia was documented in 5 patients (Table 3). There was sustained ventricular tachycardia in 4 patients and ventricular fibrillation in 1 patient. The max RT, min RT and min RTc were similar in patients with and without ventricular arrhythmia {max RT; 426 ms (413, 442) vs. 403 ms (379, 433), P ⫽ .2, min RT; 250 ms (223, 258) vs. 249, ms (234, 261), P ⫽ .7, min RTc; 261 ms (257, 292) vs. 258 ms (245, 277), P ⫽ .3}. The max RTc, RT dispersion and RTc dispersion were
Fig. 3. (A) Comparison of recovery time (RT) in patients with anterior MI and healthy controls, Plus and minus signs identify the lead locations where the RT interval in patients with anterior MI was greater or less than in controls. The RT in patients with anterior MI was generally greater over the left side of the chest and lower over the right side of the chest. (B) Comparison between the recovery time (RT) in patients with inferior MI and RT in healthy controls. Plus and minus signs identify the lead locations where the RT interval in patients with inferior MI was greater or less than in controls. RT in patients with inferior MI was generally greater over the lower back and the left side of the chest and lower over the upper right side of the anterior chest. (C) Comparison between the recovery time (RT) in patients with anterior MI and RT in inferior MI. Plus and minus signs identify the lead locations where the RT interval in patients with anterior MI was greater or less than with inferior MI. The RT in patients with anterior MI was generally greater over the upper front-left side of the chest and smaller over the lower back-right side of the chest.
6 Journal of Electrocardiology Vol. 36 No. 1 January 2003 Table 3. Characteristics of Patients With Ventricular Arrhythmias No. 1 2 3 4 5 Average
Site of MI Anterior Anterior Inferior Inferior Anterior
Age
Sex
VA
62 66 66 70 78 68
F M F M M
VF VT VT VT VT
LVEF (%) 39 34 42 19 26 32
LVDYS ⫺ ⫹ ⫺ ⫹ ⫹
LVEDV (mL)
LVEDP (mm Hg)
maxRT (ms)
minRT (ms)
RTD (ms)
251 148 110 182 187 175
12 9 9 11 9 10
399 426 428 417 483 431
230 250 251 203 281 243
169 176 177 214 202 188
VA, ventricular arrhythmia; LVEF, left ventricular ejection fraction; DYS, left ventricular dyskinesis; LVEDV, left ventricularenddiastolic volume; LVEDP, left ventricular end-diastolic pressure; maxRT, maximum recovery time; min RT, minimum RT; RTD, recovery time dispersion; MI, myocardial infarction; VF, ventricular fibrillation; VT, ventricular tachycardia.
greater, however, in patients with ventricular arrhythmia than in patients without ventricular arrhythmia {max RTc; 499 ms (410, 448) vs. 427 ms (406, 448), P ⬍ .005, RT dispersion; 177 ms (174, 205) vs. 156 ms (139, 174), P ⬍ .05, RTc dispersion; 209 ms (189, 223) vs. 160 ms (146, 184), P ⬍ .01}. Recovery Time and Left Ventricular Function in Patients With MI Twenty-three patients had a dilated LV (enddiastolic volume ⬎ 180 mL), 33 had an ejection fraction ⬍40%, 27 had LV dyskinesis, and 29 had elevated end-diastolic pressure (end-diastolic pressure ⬎ 15 mm Hg). The median LV end-diastolic volume in patients with dilated LV was 200 mL and the median LV ejection fraction with an ejection fraction was 34%. The max RT, max RTc, min RT and min RTc were similar in patients without and with a dilated LV {max RT; 403 ms (383, 428) vs. 418 ms (382, 461), P ⫽ .1, max RTc; 428 ms (406, 446) vs. 436 ms (407, 474), P ⫽ .1, min RT; 250 ms (236, 262) vs. 243 ms (229, 258), P ⫽ .2, min RTc; 260 ms (246, 279) vs. 253 ms (238, 266), P ⫽ .1}. However RT dispersion and RTc dispersion were significantly greater in patients with a dilated LV {RT dispersion; 155 ms (137, 172) vs. 169 ms (154, 201), P ⬍ .005, RTc dispersion; 158 ms (146, 178) vs. 181 ms (157, 218), P ⬍ .005}. The max RT and min RTc also were similar in patients without and with impaired ejection fraction [max RT; 403 ms (382, 428) vs. 416 ms (386, 444), P ⫽ .1, min RTc; 259 ms (246, 278) vs. 255 ms (241, 281), P ⫽ .5], but the min RT was lower in patients with impaired ejection fraction than controls {251 ms (239, 262) vs. 244 ms (220, 255), P ⬍ .05}. The max RTc, RT dispersion and RTc dispersion were greater in patients with impaired ejection fraction than controls [The max RTc; 423 ms (402, 443) vs. 440 ms (417, 479), P ⬍ .001, RT dispersion; 152 ms (138, 165) vs. 173 ms (155,
202), P ⬍ .0005, RTc dispersion; 156 ms (145, 175) vs. 188 ms (170, 221), P ⬍ .0001]. There was a positive correlation between the RT dispersion and the left ventricular end-diastolic volume, and a negative correlation between the RT dispersion and the left ventricular ejection fraction (Fig. 4). The max RT, min RT and min RTc were similar in patients without and with LV dyskinesis [max RT; 403 ms (378, 428) vs. 417 ms (391, 441), P ⫽ .1, min RT; 250 ms (238, 264) vs. 243 ms (229, 257), P ⫽ .05, min RTc; 257 ms (245, 274) vs. 260 ms (245, 286), P ⫽ .5]. However, max RTc, RT dispersion and RTc dispersion were significantly greater in patients with LV dyskinesis {max RTc; 417 ms (402, 441) vs. 454 ms (437, 479), P ⬍ .0001, RT dispersion; 155ms (137, 161) vs. 175 ms (159, 201), P ⬍ .0005, RTc dispersion; 155 ms (145, 175) vs. 192 ms (171, 216), P ⬍ .0001}. The max RT, max RTc, min RT, min RTc, RT dispersion and RTc dispersion were similar in patients without and with elevated enddiastolic pressure {max RT; 399 ms (378, 427) vs. 423 ms (394, 453), P ⫽ .05, max RTc; 430 ms (409, 453) vs. 426 ms (390, 445), P ⫽ .5, min RT; 249 ms (231, 260) vs. 249 ms (241, 262), P ⫽ .4, min RTc; 259 ms (246, 283) vs. 253 ms (244, 265), P ⫽ .3, RT dispersion; 156 ms (140, 174) vs. 162 ms (138, 193), P ⫽ .4, RTc dispersion; 162 ms (150, 184) vs. 162 ms (143, 196), P ⫽ .7}.
Discussion The results obtained from measured body surface mapping data suggest that RT is greater in areas of infarction than in healthy myocardium. Additionally, in patients with previous myocardial infarction the RT dispersion was increased in patients with ventricular arrhythmias, dilated left ventricles or impaired LV function.
LV Function and RT Dispersion •
Mine et al. 7
anterior and left chest, while in contrast, the RT in patients with inferior MI was greater in the lower back and the lower left chest compared to healthy controls. RT in patients with anterior MI was greater over the upper side of the anterior chest and left side of the chest, and smaller over the lower back of the chest compared with inferior MI. The RT seems to be increased in the area of infarction; possibly because RT may be affected by areas of viable myocardium existing near the endocardium or at the margin of the infarcted area. The max RT and RT dispersion were greater in patients with an anterior MI than in control patients but there were no difference between RT in patients with an inferior MI and controls. The LV ejection fraction was greater and the LV end-diastolic volume was lower in the patients with inferior MI than anterior MI {ejection fraction: 57% (48, 66) vs. 50 % (39, 61), P ⬍ .05, end-diastolic volume: 131 ml (111, 175) vs. 146 ml (125, 174), P ⬍ .05}. The lack of difference in RT intervals between patients with inferior MI and controls might be related to the fact that fewer patients with inferior MI had LV dysfunction or that body surface mapping did not reflect sufficiently inferior areas.
Repolarization Abnormalities and Left Ventricular Function
Fig. 4. Scattergram showing the relationships between recovery time dispersion (RTd), left ventricular enddiastolic volume, and left ventricular ejection fraction.
Recovery Time in Patients With MI Haws and Lux (11) showed that there is a close correlation between the electrocardiographic activation-recovery intervals (defined as the interval between the times of the minimum derivative of the QRS and maximum derivative of the T wave) and the action potential duration. Franz (14) reported that monophasic action potentials could not be recorded from human endocardium in areas of infarcted, aneurysmal myocardium. In this study, RT in patients with anterior MI was increased in the
The relationship between the heterogeneity of ventricular repolarization and LV function has not been established. Perkio¨ ma¨ ki et al. (15) reported that QTc dispersion is unrelated to the LV ejection fraction after MI. Tamura et al. (16) found no correlation between QT dispersion, LV ejection fraction, and regional wall motion in 94 patients after a first anterior MI (r ⫽ .11, P ⫽ .29; r ⫽ .05, P ⫽ .63, respectively). In contrast, Karagounis et al. (17) reported a significant multivariate association between QT dispersion and LV function in 207 patients (P ⫽ .02) after MI. These discordant conclusions might have arisen because QT dispersion did not account for the full magnitude of ventricular repolarization heterogeneity. The 87-lead body surface mapping technique may be better than the 12-lead QT dispersion technique in evaluating heterogeneity of ventricular repolarization (12,13,17,18). This study shows clearly that LV dilatation and dysfunction in patients with an old myocardial infarction is associated with increased recovery time dispersion suggesting heterogeneous ventricular repolarization.
8 Journal of Electrocardiology Vol. 36 No. 1 January 2003 Repolarization Abnormalities and Ventricular Arrhythmias in Patients With Impaired Left Ventricular Function In patients with impaired left ventricular function, sudden cardiac death is reported to be one of the major causes of death and result from ventricular tachyarrhythmias, such as ventricular tachycardia and fibrillation (19). Packer reviewed cases of sudden death in patients with congestive heart failure and proposed 4 possible mechanisms. First, myocardial fibrosis due to ischemia, infarction, and inflammation could provide the anatomic substrate for re-entrant circuits. Second, changes in myocardial fiber stretch mediated by changes in ventricular pressures and volumes may alter the electrophysiologic properties of the ventricle and predispose the patient to serious ventricular arrhythmias. Third, neurohormonal activation could be arrhythmogenic by way of potassium and magnesium depletion. Fourth, diuretics may enhance electrolyte depletion by promoting the renal excretion (20). After myocardial infarction, and particularly when left ventricular function is impaired, ventricular repolarization is likely to be affected by a number of factors. Myocardial necrosis, disturbed intercellular geometry and dilatation, and reactive hypertrophy that is part of the myocardial remodeling process, may affect the heterogeneity of ventricular repolarization. Moreover, a differential contraction-excitation feedback in the presence of regional wall motion abnormalities can generate local ventricular repolarization heterogeneity (21). Heterogeneous ventricular repolarization may create a condition which facilitates reentry, leading to ventricular arrhythmia (6). In our study, serious ventricular arrhythmias were documented in 5 patients (sustained ventricular tachycardia in 4 patients and ventricular fibrillation in 1 patient), and were associated with increased RT dispersion. Serious ventricular arrhythmias increase the heterogeneity of ventricular repolarization and suggest a causal relationship with the increased risk of sudden death in this population. Re-entrant circuits may be created if premature impulses originating from sites with a short RT are blocked at sites with a long RT, which may be reexcited, retrograde after repolarization. In an animal model of tachycardia-induced heart failure, monophasic action potential duration and spatial dispersion were increased and associated with ventricular arrhythmias and sudden death (22,23). However, the precise mechanism heterogeneous repolarization in heart failure remains to be determined.
Study Limitations There are several limitations to the present study. First, there is a circadian variation in QT dispersion. In a previous study, Ishida showed that QT dispersion is greater during the day than at night (24), and RT measurements may also show circadian variation. However, body surface mapping was recorded at the same time of day in each patient; therefore, circadian variation is not likely to be an important cause of error. Second, Millar showed that activation-recovery intervals are affected by both norepinephrine and cardiac sympathetic activity (25), so those factors could have influenced ventricular repolarization. Third, the end-point of RT was defined as the point of maximum dV/dt on ECG. These points were checked visually on a high-resolution color monitor and the leads were excluded when the points were indeterminate. However, some leads have low-amplitude T waves following MI, and those leads were excluded in this study. This mapping system has many leads and RT dispersion was thought to be adequate to measure the heterogeneity of ventricular repolarization.
Conclusion We estimated the heterogeneity of repolarization by recording recovery time dispersion as measured by body surface mapping in patients with a history of myocardial infarction. Patients with impaired left ventricular function demonstrated increased recovery time dispersion. Heterogeneity of ventricular repolarization may provide the electrical substrate for ventricular arrhythmia.
References 1. Stevenson WG, Brugada P, Waldecker B, et al: Clinical, angiographic, and electrophysiologic findings in patients with aborted sudden death as compared with patients with sustained ventricular tachycardia after myocardial infarction. Circulation 71:1146, 1985 2. Stevenson WG, Stevenson LW, Middlekauff HR, et al: Sudden death prevention in patients with advanced ventricular dysfunction. Circulation 88:2953, 1993 3. Cassidy DM, Vassallo JA, Miller JM, et al: Endocardial catheter mapping in patients in sinus rhythm: relationship to underlying heart disease and ventricular arrhythmias. Circulation 73:645, 1986 4. Gomes JA, Winters SL, Martinson M, et al: The prognostic significance of quantitative signal-aver-
LV Function and RT Dispersion •
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
aged variables relative to clinical variables, site of myocardial infarction, ejection fraction and ventricular premature beats: A prospective study. J Am Coll Cardiol 13:377, 1989 Merx W, Yoon MS, Han J: The role of local disparity in conduction and recovery time on ventricular vulnerability to fibrillation. Am Heart J 94:603, 1977 Kuo CS, Reddy CP, Munakata K, et al: Mechanism of ventricular arrhythmias caused by increased dispersion of repolarization. Eur Heart J 6:63, 1985 Olsson SB, Brorson L, Edvardsson N, et al: Estimation of ventricular repolarization in man by monophasic action potential recording technique. Eur Heart J 6:71, 1985 Higham PD, Furniss SS, Campbell RW: QT dispersion and components of the QT interval in ischaemia and infarction. Br Heart J 73:32, 1995 Moreno FL, Villanueva T, Karagounis LA, et al: Reduction in QT interval dispersion by successful thrombolytic therapy in acute myocardial infarction. TEAM-2 Study Investigators. Circulation 90:94, 1994 Barr CS, Naas AA, Fenwick M, et al: Enalapril reduces QTc dispersion in mild congestive heart failure secondary to coronary artery disease. Am J Cardiol 79:328, 1997 Haws CW, Lux RL: Correlation between in vivo transmembrane action potential durations and activation-recovery intervals from electrograms. Effects of interventions that alter repolarization time. Circulation 81:281, 1990 Shimizu W, Kamakura S, Ohe T, et al: Diagnostic value of recovery time measured by body surface mapping in patients with congenital long QT syndrome. Am J Cardiol 74:780, 1994 Shimizu W, Kamakura S, Kurita T, et al: Influence of epinephrine, propranolol, and atrial pacing on spatial distribution of recovery time measured by body surface mapping in congenital long QT syndrome. J Cardiovasc Electrophysiol 8:1102, 1997 Franz MR: Long-term recording of monophasic action potentials from human endocardium. Am J Cardiol 51:1629, 1983 Perkio¨ ma¨ ki JS, Huikuri HV, Koistinen JM, et al: Heart rate variability and dispersion of QT interval in patients with vulnerability to ventricular tachycardia
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Mine et al. 9
and ventricular fibrillation after previous myocardial infarction. J Am Coll Cardiol 30:1331, 1997 Tamura A, Nagase K, Mikuriya Y, et al: Relation of QT dispersion to infarct size and left ventricular wall motion in anterior wall acute myocardial infarction. Am J Cardiol 83:1423, 1999 Karagounis LA, Anderson JL, Moreno FL, et al: Multivariate associates of QT dispersion in patients with acute myocardial infarction: Primacy of patency status of the infarct-related artery. TEAM-3 Investigators. Third trial of Thrombolysis with Eminase in Acute Myocardial Infarction. Am Heart J 135:1027, 1998 Shimizu W: Concept of activation recovery interval and clinical usefulness of body surface recovery time and recovery time isochrone map. Nippon Rinsho 53:74, 1995 Luu M, Stevenson WG, Stevenson LW, et al: Diverse mechanisms of unexpected cardiac arrest in advanced heart failure. Circulation 80:1675, 1989 Packer M, Lee WH, Kessler PD, et al: Role of neurohormonal mechanisms in determining survival in patients with severe chronic heart failure. Circulation 75:IV80, 1987 Taggart P, Sutton P, John R, et al: Monophasic action potential recordings during acute changes in ventricular loading induced by the Valsalva manoeuvre. Br Heart J 67:221, 1992 Pak PH, Nuss HB, Tunin RS, et al: Repolarization abnormalities, arrhythmia and sudden death in canine tachycardia-induced cardiomyopathy. J Am Coll Cardiol 30:576, 1997 Li HG, Jones DL, Yee R, et al: Electrophysiologic substrate associated with pacing-induced heart failure in dogs: potential value of programmed stimulation in predicting sudden death. J Am Coll Cardiol 19:444, 1992 Ishida S, Nakagawa M, Fujino T, et al: Circadian variation of QT interval dispersion: Correlation with heart rate variability. J Electrocardiol 30:205, 1997 Millar CK, Kralios FA, Lux RL: Correlation between refractory periods and activation-recovery intervals from electrograms: Effects of rate and adrenergic interventions. Circulation 72:1372, 1985
Impaired Left Ventricular Function Is Associated With Increased Recovery Time Dispersion in Patients With Previous Myocardial Infarction
Takanao Mine, MD, Hiroki Shimizu, MD, Takanori Inazumi, MD, and Tadaaki Iwasaki, MD
Abstract: Left ventricular (LV) dysfunction after myocardial infarction is associated with higher risk of serious ventricular arrhythmias and sudden death. We suspected that heterogeneity in ventricular repolarization contributes to these arrhythmias. To quantify this heterogeneity, we measured the recovery time (the interval between QRS onset and the time of maximum dV/dt in the ST-T segment) using an 87-lead body surface mapping electrocardiogram and estimated recovery time dispersion (the difference between maximum recovery time and minimum recovery time) in each lead. Differences between 110 patients with previous myocardial infarction and 31 healthy controls were compared. Recovery time dispersion [medians (25th, 75th percentiles)] was greatest in patients with a dilated LV {169 ms (154, 201) vs. 155 ms (137, 172), P ⬍ .005}, impaired ejection fraction {173 ms (155, 202) vs. 152 ms (138, 165), P ⬍ .0005} and LV dyskinesis {175 ms (159, 201) vs. 155 ms (137, 161), P ⬍ .0005}. This study suggests that LV dysfunction associated with myocardial infarction leads to heterogeneous ventricular repolarization and may provide the electrical substrate for ventricular arrhythmias. Key words: Repolarization abnormalities, previous myocardial infarction, impaired left ventricular function, recovery time, ventricular arrhythmias.
Left ventricular (LV) dysfunction after myocardial infarction (MI) is associated with higher risk of serious ventricular arrhythmias and sudden death (1– 4). However, the mechanism responsible for
ventricular arrhythmias in patients with LV dysfunction is unclear. Heterogeneity of repolarization is known to increase susceptibility to reentrant arrhythmias (5–7), and QT dispersion has been reported to be a high risk factor for serious ventricular arrhythmias (8). Recently it was found that QT dispersion is reduced by successful thrombolytic therapy following acute MI (9) and by administration of angiotensin converting enzyme inhibitors in patients who have coronary artery disease (10). These observations suggest that impaired LV function increases the dispersion of ventricular repolarization. However, the relationship between QT dis-
From the Department of Internal Medicine, Cardiovascular Division, Hyogo College of Medicine, Nishinomiya, Japan. Reprint requests: Hiroki Shimizu, MD, Department of Internal Medicine, Cardiovascular Division, Hyogo College of Medicine, 1-1, Mukogawa-cho, Nishinomiya, 663-8501, Japan; e-mail: [email protected]. Copyright 2003, Elsevier Science (USA). All rights reserved. 0022-0736/03/3601-0001$35.00/0 doi:10.1054/jelc.2003.50008
1
2 Journal of Electrocardiology Vol. 36 No. 1 January 2003 persion and LV function is unclear. The QT interval can usually be measured only from the standard 12-electrocardiogram (ECG) leads, and it is sometimes difficult to define the QT interval precisely by measuring it manually in patients who have had MI. Conversely, the activation-recovery interval, measured as the interval between the times of the most rapid decrease voltage (minimum dV/dt) of the QRS and the maximum rate of voltage increase (maximum dV/dt) of the T wave, is closely correlated with the action potential duration, defined as the interval between times of the maximum dV/dt of the upstroke and the minimum dV/dt of the downstroke of the transmembrane action potential (11). Several studies have shown that dispersion of recovery time (RT; defined as the interval between QRS onset and the time of maximum dV/dt in the ST-T segment) measured by 87-lead body surface mapping is useful for estimating dispersion of ventricular repolarization (12,13). In the present study, we investigated the relationship between LV function and dispersion of recovery time as measured by body surface mapping in patients with previous MI.
Materials and Methods Study Patients The study population included 110 patients (93 men, 17 women; mean age, 64 years) who had suffered MI at least 6 months prior to enrollment. Patients taking antiarrhythmic agents or -blockers were excluded. Patients with atrial fibrillation, paced rhythm, bundle branch block, or AV block also were excluded. The site of the MI was anterior in 64 patients, inferior in 44 patients, and both sites were affected in 2 patients. Five patients presented with ventricular arrhythmias (4 with sustained ventricular tachycardia and 1 with ventricular fibrillation). We performed a case controlled study in 31 patients with normal coronary angiograms to identify RT characteristics of MI. Demographics are presented in Table 1. Cardiac Catheterization Cardiac catheterization was performed in all patients according to the Judkin’s technique after a written informed consent had been obtained. Left ventricular end-diastolic pressure and systolic pressure were measured with a pigtail angiographic catheter. Biplane left ventriculography was per-
Table 1. Clinical and Angiographic Data
Male/Female Age (y.o.) Infarction location Anterior Inferior Anterior ⫹ Inferior Ventricular arrhythmia Ventricular tachycardia Ventricular fibrillation Ejection Fraction (%) End-diastolic volume (ml) LV dyskinesis
Control
MI
23/8 64 (56, 69)
93/17 64 (58, 71)
68 (63, 73) 123 (107,128)
64 44 2 5 4 1 52 (40, 64) 143 (118,175) 27
Data presented are expressed as medians (25th, 75th).
formed in 30-degree right anterior oblique and 60-° left anterior oblique views. Left ventricular enddiastolic volume and end-systolic volume were calculated by the area-length method from the biplane views. Left ventricular ejection fraction was calculated as the ratio (end-diastolic volume-end-systolic volume)/end-diastolic volume. The same investigator conducted all studies. An end-diastolic volume ⬎ 180 ML, ejection fraction ⬍40%, and enddiastolic pressure ⬎ 15 mm Hg were used to define a dilated LV, impaired ejection fraction, and increased end-diastolic pressure, respectively. Body Surface Mapping Body surface mapping was performed using the VCM 3000 (Fukuda Denshi Co., Tokyo, Japan). Eighty-seven body surface leads were arranged in a lattice-like pattern (13 ⫻ 7 matrix), excluding the leads in the midaxillary line (59 leads located on the anterior chest and 28 leads on the back). An ECG from these 87 unipolar leads with Wilson’s central terminal as reference, a standard 12-lead ECG, and Frank X, Y, and Z lead scalar ECG, were recorded simultaneously. Sampling frequency was 1,000 samples/second/channel recorded with the patient resting in a supine position. These data were stored on a floppy disk in digital format. The mapping data were transferred to a personal computer (PC-9801 FA, NEC, Tokyo, Japan) with an analysis program developed by Shimizu et al. (12,13). A high-resolution color monitor and printer were connected to the computer. The QRS onset, offset, and T offset were determined visually from the superimposed Frank X, Y, and Z leads and the spatial magnitude. RT was defined as the interval between QRS onset and the time of maximum dV/dt in the ST-T segment and was measured automatically by the com-
LV Function and RT Dispersion •
Mine et al. 3
Fig. 1. The measurement of recovery time in some morphologies of T wave. T wave was positive, and maximum dV/dt occurred before the peak of T wave. T wave was negative, and maximum dV/dt occurred after the peak of T wave. T wave was biphasic, and maximum dV/dt occurred on the upstroke of the positive peak. The points of maximum dV/dt in flat T wave were indeterminate, then the lead was excluded from analysis.
puter. (Fig. 1) The maximum dV/dt points were checked visually on the high-resolution color monitor. When the points were incorrect, the cursor was manually moved to the actual maximum dV/ dt. When the points were indeterminate, the lead was excluded from analysis. RT value was corrected (RTc) according to Bazett formula. Maximum RT and RTc (max RT and max RTc) and minimum RT and RTc (min RT and min RTc) were measured in all 87 leads. RT and RTc dispersion were defined as the difference between maximum and minimum RT and RTc. We examined the regional differences in recovery time interval between controls and patients with previous MI, and the relationship between LV function and dispersion of recovery time in patients with previous MI.
Statistical Analysis All data are expressed as the medians (25th, 75th). Comparisons of recovery time were made by the Mann-Whitney U test or Kruskal-Wallis test. Linear regression analysis was used to assess the relationship between RTd and the index of left ventricular function. A P value ⬍ .05 was considered to be statistically significant.
Results Recovery Time in Patients With MI A representative 87-lead ECG from a control subject and a patient with anterior MI are shown in Figure 2. In the control subject, the max RT (350 ms) was recorded from the C7 lead, and the min RT (224 ms) was recorded from the F4 lead. Thus the RT dispersion was 126 ms. The patient with the anterior MI had sustained ventricular tachycardia and LV dyskinesis (ejection fraction 34%), and the max RT (426 ms) was recorded from the F2 lead, and the min RT (250 ms) was recorded from the E3 lead, so RT dispersion was 176 ms. The max RT, min RT, and RT dispersion in controls and in patients with MI are presented in Table 2. The max RT and RT dispersion were greater in patients with an anterior MI than in control patients or in patients with an inferior MI. The RT interval in each lead was compared with anterior MI, inferior MI and healthy controls. Comparisons of RT interval in patients with an anterior and inferior MI to controls are shown in Figs. 3A and 3B. The recovery time in each lead was compared in patients with an anterior to inferior MI (Fig. 3C). The RT in patients with anterior MI was generally greater over the left side of the chest, and
Fig. 2. The 87-lead ECG obtained by body surface mapping from a control subject and a patient with anterior MI having sustained ventricular tachycardia. (A) The max RT was recorded from the C7 lead, and the min RT was recorded from the F4 lead of a 64 year old woman. (B) The max RT was recorded from the F2 lead, and the min RT was recorded from the E3 lead from a 66-year-old man.
LV Function and RT Dispersion • Table 2. Max RT, min RT, and RT Dispersion
Control Anterior MI Inferior MI
max RT
min RT
RT Dispersion
392 (368, 410) 415 (393,436)* 392 (373, 426)
250 (231,268) 249 (238,260) 250 (232,262)
140 (123, 150) 163 (151,189)† 144 (136, 163)
*P ⬍ .005 vs. control and inferior MI; †P ⬍ .001 vs. control and inferior MI;
smaller over the lower back of the chest compared with controls. RT in patients with inferior MI was generally greater over the lower back of the chest and lower over the upper side of the anterior chest compared with controls. RT in patients with anterior MI was greater over the upper side of the anterior chest and left side of the chest, and smaller
Mine et al. 5
over the lower back of the chest compared with inferior MI. Recovery Time in Patients With Ventricular Arrhythmias and MI A serious ventricular arrhythmia was documented in 5 patients (Table 3). There was sustained ventricular tachycardia in 4 patients and ventricular fibrillation in 1 patient. The max RT, min RT and min RTc were similar in patients with and without ventricular arrhythmia {max RT; 426 ms (413, 442) vs. 403 ms (379, 433), P ⫽ .2, min RT; 250 ms (223, 258) vs. 249, ms (234, 261), P ⫽ .7, min RTc; 261 ms (257, 292) vs. 258 ms (245, 277), P ⫽ .3}. The max RTc, RT dispersion and RTc dispersion were
Fig. 3. (A) Comparison of recovery time (RT) in patients with anterior MI and healthy controls, Plus and minus signs identify the lead locations where the RT interval in patients with anterior MI was greater or less than in controls. The RT in patients with anterior MI was generally greater over the left side of the chest and lower over the right side of the chest. (B) Comparison between the recovery time (RT) in patients with inferior MI and RT in healthy controls. Plus and minus signs identify the lead locations where the RT interval in patients with inferior MI was greater or less than in controls. RT in patients with inferior MI was generally greater over the lower back and the left side of the chest and lower over the upper right side of the anterior chest. (C) Comparison between the recovery time (RT) in patients with anterior MI and RT in inferior MI. Plus and minus signs identify the lead locations where the RT interval in patients with anterior MI was greater or less than with inferior MI. The RT in patients with anterior MI was generally greater over the upper front-left side of the chest and smaller over the lower back-right side of the chest.
6 Journal of Electrocardiology Vol. 36 No. 1 January 2003 Table 3. Characteristics of Patients With Ventricular Arrhythmias No. 1 2 3 4 5 Average
Site of MI Anterior Anterior Inferior Inferior Anterior
Age
Sex
VA
62 66 66 70 78 68
F M F M M
VF VT VT VT VT
LVEF (%) 39 34 42 19 26 32
LVDYS ⫺ ⫹ ⫺ ⫹ ⫹
LVEDV (mL)
LVEDP (mm Hg)
maxRT (ms)
minRT (ms)
RTD (ms)
251 148 110 182 187 175
12 9 9 11 9 10
399 426 428 417 483 431
230 250 251 203 281 243
169 176 177 214 202 188
VA, ventricular arrhythmia; LVEF, left ventricular ejection fraction; DYS, left ventricular dyskinesis; LVEDV, left ventricularenddiastolic volume; LVEDP, left ventricular end-diastolic pressure; maxRT, maximum recovery time; min RT, minimum RT; RTD, recovery time dispersion; MI, myocardial infarction; VF, ventricular fibrillation; VT, ventricular tachycardia.
greater, however, in patients with ventricular arrhythmia than in patients without ventricular arrhythmia {max RTc; 499 ms (410, 448) vs. 427 ms (406, 448), P ⬍ .005, RT dispersion; 177 ms (174, 205) vs. 156 ms (139, 174), P ⬍ .05, RTc dispersion; 209 ms (189, 223) vs. 160 ms (146, 184), P ⬍ .01}. Recovery Time and Left Ventricular Function in Patients With MI Twenty-three patients had a dilated LV (enddiastolic volume ⬎ 180 mL), 33 had an ejection fraction ⬍40%, 27 had LV dyskinesis, and 29 had elevated end-diastolic pressure (end-diastolic pressure ⬎ 15 mm Hg). The median LV end-diastolic volume in patients with dilated LV was 200 mL and the median LV ejection fraction with an ejection fraction was 34%. The max RT, max RTc, min RT and min RTc were similar in patients without and with a dilated LV {max RT; 403 ms (383, 428) vs. 418 ms (382, 461), P ⫽ .1, max RTc; 428 ms (406, 446) vs. 436 ms (407, 474), P ⫽ .1, min RT; 250 ms (236, 262) vs. 243 ms (229, 258), P ⫽ .2, min RTc; 260 ms (246, 279) vs. 253 ms (238, 266), P ⫽ .1}. However RT dispersion and RTc dispersion were significantly greater in patients with a dilated LV {RT dispersion; 155 ms (137, 172) vs. 169 ms (154, 201), P ⬍ .005, RTc dispersion; 158 ms (146, 178) vs. 181 ms (157, 218), P ⬍ .005}. The max RT and min RTc also were similar in patients without and with impaired ejection fraction [max RT; 403 ms (382, 428) vs. 416 ms (386, 444), P ⫽ .1, min RTc; 259 ms (246, 278) vs. 255 ms (241, 281), P ⫽ .5], but the min RT was lower in patients with impaired ejection fraction than controls {251 ms (239, 262) vs. 244 ms (220, 255), P ⬍ .05}. The max RTc, RT dispersion and RTc dispersion were greater in patients with impaired ejection fraction than controls [The max RTc; 423 ms (402, 443) vs. 440 ms (417, 479), P ⬍ .001, RT dispersion; 152 ms (138, 165) vs. 173 ms (155,
202), P ⬍ .0005, RTc dispersion; 156 ms (145, 175) vs. 188 ms (170, 221), P ⬍ .0001]. There was a positive correlation between the RT dispersion and the left ventricular end-diastolic volume, and a negative correlation between the RT dispersion and the left ventricular ejection fraction (Fig. 4). The max RT, min RT and min RTc were similar in patients without and with LV dyskinesis [max RT; 403 ms (378, 428) vs. 417 ms (391, 441), P ⫽ .1, min RT; 250 ms (238, 264) vs. 243 ms (229, 257), P ⫽ .05, min RTc; 257 ms (245, 274) vs. 260 ms (245, 286), P ⫽ .5]. However, max RTc, RT dispersion and RTc dispersion were significantly greater in patients with LV dyskinesis {max RTc; 417 ms (402, 441) vs. 454 ms (437, 479), P ⬍ .0001, RT dispersion; 155ms (137, 161) vs. 175 ms (159, 201), P ⬍ .0005, RTc dispersion; 155 ms (145, 175) vs. 192 ms (171, 216), P ⬍ .0001}. The max RT, max RTc, min RT, min RTc, RT dispersion and RTc dispersion were similar in patients without and with elevated enddiastolic pressure {max RT; 399 ms (378, 427) vs. 423 ms (394, 453), P ⫽ .05, max RTc; 430 ms (409, 453) vs. 426 ms (390, 445), P ⫽ .5, min RT; 249 ms (231, 260) vs. 249 ms (241, 262), P ⫽ .4, min RTc; 259 ms (246, 283) vs. 253 ms (244, 265), P ⫽ .3, RT dispersion; 156 ms (140, 174) vs. 162 ms (138, 193), P ⫽ .4, RTc dispersion; 162 ms (150, 184) vs. 162 ms (143, 196), P ⫽ .7}.
Discussion The results obtained from measured body surface mapping data suggest that RT is greater in areas of infarction than in healthy myocardium. Additionally, in patients with previous myocardial infarction the RT dispersion was increased in patients with ventricular arrhythmias, dilated left ventricles or impaired LV function.
LV Function and RT Dispersion •
Mine et al. 7
anterior and left chest, while in contrast, the RT in patients with inferior MI was greater in the lower back and the lower left chest compared to healthy controls. RT in patients with anterior MI was greater over the upper side of the anterior chest and left side of the chest, and smaller over the lower back of the chest compared with inferior MI. The RT seems to be increased in the area of infarction; possibly because RT may be affected by areas of viable myocardium existing near the endocardium or at the margin of the infarcted area. The max RT and RT dispersion were greater in patients with an anterior MI than in control patients but there were no difference between RT in patients with an inferior MI and controls. The LV ejection fraction was greater and the LV end-diastolic volume was lower in the patients with inferior MI than anterior MI {ejection fraction: 57% (48, 66) vs. 50 % (39, 61), P ⬍ .05, end-diastolic volume: 131 ml (111, 175) vs. 146 ml (125, 174), P ⬍ .05}. The lack of difference in RT intervals between patients with inferior MI and controls might be related to the fact that fewer patients with inferior MI had LV dysfunction or that body surface mapping did not reflect sufficiently inferior areas.
Repolarization Abnormalities and Left Ventricular Function
Fig. 4. Scattergram showing the relationships between recovery time dispersion (RTd), left ventricular enddiastolic volume, and left ventricular ejection fraction.
Recovery Time in Patients With MI Haws and Lux (11) showed that there is a close correlation between the electrocardiographic activation-recovery intervals (defined as the interval between the times of the minimum derivative of the QRS and maximum derivative of the T wave) and the action potential duration. Franz (14) reported that monophasic action potentials could not be recorded from human endocardium in areas of infarcted, aneurysmal myocardium. In this study, RT in patients with anterior MI was increased in the
The relationship between the heterogeneity of ventricular repolarization and LV function has not been established. Perkio¨ ma¨ ki et al. (15) reported that QTc dispersion is unrelated to the LV ejection fraction after MI. Tamura et al. (16) found no correlation between QT dispersion, LV ejection fraction, and regional wall motion in 94 patients after a first anterior MI (r ⫽ .11, P ⫽ .29; r ⫽ .05, P ⫽ .63, respectively). In contrast, Karagounis et al. (17) reported a significant multivariate association between QT dispersion and LV function in 207 patients (P ⫽ .02) after MI. These discordant conclusions might have arisen because QT dispersion did not account for the full magnitude of ventricular repolarization heterogeneity. The 87-lead body surface mapping technique may be better than the 12-lead QT dispersion technique in evaluating heterogeneity of ventricular repolarization (12,13,17,18). This study shows clearly that LV dilatation and dysfunction in patients with an old myocardial infarction is associated with increased recovery time dispersion suggesting heterogeneous ventricular repolarization.
8 Journal of Electrocardiology Vol. 36 No. 1 January 2003 Repolarization Abnormalities and Ventricular Arrhythmias in Patients With Impaired Left Ventricular Function In patients with impaired left ventricular function, sudden cardiac death is reported to be one of the major causes of death and result from ventricular tachyarrhythmias, such as ventricular tachycardia and fibrillation (19). Packer reviewed cases of sudden death in patients with congestive heart failure and proposed 4 possible mechanisms. First, myocardial fibrosis due to ischemia, infarction, and inflammation could provide the anatomic substrate for re-entrant circuits. Second, changes in myocardial fiber stretch mediated by changes in ventricular pressures and volumes may alter the electrophysiologic properties of the ventricle and predispose the patient to serious ventricular arrhythmias. Third, neurohormonal activation could be arrhythmogenic by way of potassium and magnesium depletion. Fourth, diuretics may enhance electrolyte depletion by promoting the renal excretion (20). After myocardial infarction, and particularly when left ventricular function is impaired, ventricular repolarization is likely to be affected by a number of factors. Myocardial necrosis, disturbed intercellular geometry and dilatation, and reactive hypertrophy that is part of the myocardial remodeling process, may affect the heterogeneity of ventricular repolarization. Moreover, a differential contraction-excitation feedback in the presence of regional wall motion abnormalities can generate local ventricular repolarization heterogeneity (21). Heterogeneous ventricular repolarization may create a condition which facilitates reentry, leading to ventricular arrhythmia (6). In our study, serious ventricular arrhythmias were documented in 5 patients (sustained ventricular tachycardia in 4 patients and ventricular fibrillation in 1 patient), and were associated with increased RT dispersion. Serious ventricular arrhythmias increase the heterogeneity of ventricular repolarization and suggest a causal relationship with the increased risk of sudden death in this population. Re-entrant circuits may be created if premature impulses originating from sites with a short RT are blocked at sites with a long RT, which may be reexcited, retrograde after repolarization. In an animal model of tachycardia-induced heart failure, monophasic action potential duration and spatial dispersion were increased and associated with ventricular arrhythmias and sudden death (22,23). However, the precise mechanism heterogeneous repolarization in heart failure remains to be determined.
Study Limitations There are several limitations to the present study. First, there is a circadian variation in QT dispersion. In a previous study, Ishida showed that QT dispersion is greater during the day than at night (24), and RT measurements may also show circadian variation. However, body surface mapping was recorded at the same time of day in each patient; therefore, circadian variation is not likely to be an important cause of error. Second, Millar showed that activation-recovery intervals are affected by both norepinephrine and cardiac sympathetic activity (25), so those factors could have influenced ventricular repolarization. Third, the end-point of RT was defined as the point of maximum dV/dt on ECG. These points were checked visually on a high-resolution color monitor and the leads were excluded when the points were indeterminate. However, some leads have low-amplitude T waves following MI, and those leads were excluded in this study. This mapping system has many leads and RT dispersion was thought to be adequate to measure the heterogeneity of ventricular repolarization.
Conclusion We estimated the heterogeneity of repolarization by recording recovery time dispersion as measured by body surface mapping in patients with a history of myocardial infarction. Patients with impaired left ventricular function demonstrated increased recovery time dispersion. Heterogeneity of ventricular repolarization may provide the electrical substrate for ventricular arrhythmia.
References 1. Stevenson WG, Brugada P, Waldecker B, et al: Clinical, angiographic, and electrophysiologic findings in patients with aborted sudden death as compared with patients with sustained ventricular tachycardia after myocardial infarction. Circulation 71:1146, 1985 2. Stevenson WG, Stevenson LW, Middlekauff HR, et al: Sudden death prevention in patients with advanced ventricular dysfunction. Circulation 88:2953, 1993 3. Cassidy DM, Vassallo JA, Miller JM, et al: Endocardial catheter mapping in patients in sinus rhythm: relationship to underlying heart disease and ventricular arrhythmias. Circulation 73:645, 1986 4. Gomes JA, Winters SL, Martinson M, et al: The prognostic significance of quantitative signal-aver-
LV Function and RT Dispersion •
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
aged variables relative to clinical variables, site of myocardial infarction, ejection fraction and ventricular premature beats: A prospective study. J Am Coll Cardiol 13:377, 1989 Merx W, Yoon MS, Han J: The role of local disparity in conduction and recovery time on ventricular vulnerability to fibrillation. Am Heart J 94:603, 1977 Kuo CS, Reddy CP, Munakata K, et al: Mechanism of ventricular arrhythmias caused by increased dispersion of repolarization. Eur Heart J 6:63, 1985 Olsson SB, Brorson L, Edvardsson N, et al: Estimation of ventricular repolarization in man by monophasic action potential recording technique. Eur Heart J 6:71, 1985 Higham PD, Furniss SS, Campbell RW: QT dispersion and components of the QT interval in ischaemia and infarction. Br Heart J 73:32, 1995 Moreno FL, Villanueva T, Karagounis LA, et al: Reduction in QT interval dispersion by successful thrombolytic therapy in acute myocardial infarction. TEAM-2 Study Investigators. Circulation 90:94, 1994 Barr CS, Naas AA, Fenwick M, et al: Enalapril reduces QTc dispersion in mild congestive heart failure secondary to coronary artery disease. Am J Cardiol 79:328, 1997 Haws CW, Lux RL: Correlation between in vivo transmembrane action potential durations and activation-recovery intervals from electrograms. Effects of interventions that alter repolarization time. Circulation 81:281, 1990 Shimizu W, Kamakura S, Ohe T, et al: Diagnostic value of recovery time measured by body surface mapping in patients with congenital long QT syndrome. Am J Cardiol 74:780, 1994 Shimizu W, Kamakura S, Kurita T, et al: Influence of epinephrine, propranolol, and atrial pacing on spatial distribution of recovery time measured by body surface mapping in congenital long QT syndrome. J Cardiovasc Electrophysiol 8:1102, 1997 Franz MR: Long-term recording of monophasic action potentials from human endocardium. Am J Cardiol 51:1629, 1983 Perkio¨ ma¨ ki JS, Huikuri HV, Koistinen JM, et al: Heart rate variability and dispersion of QT interval in patients with vulnerability to ventricular tachycardia
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Mine et al. 9
and ventricular fibrillation after previous myocardial infarction. J Am Coll Cardiol 30:1331, 1997 Tamura A, Nagase K, Mikuriya Y, et al: Relation of QT dispersion to infarct size and left ventricular wall motion in anterior wall acute myocardial infarction. Am J Cardiol 83:1423, 1999 Karagounis LA, Anderson JL, Moreno FL, et al: Multivariate associates of QT dispersion in patients with acute myocardial infarction: Primacy of patency status of the infarct-related artery. TEAM-3 Investigators. Third trial of Thrombolysis with Eminase in Acute Myocardial Infarction. Am Heart J 135:1027, 1998 Shimizu W: Concept of activation recovery interval and clinical usefulness of body surface recovery time and recovery time isochrone map. Nippon Rinsho 53:74, 1995 Luu M, Stevenson WG, Stevenson LW, et al: Diverse mechanisms of unexpected cardiac arrest in advanced heart failure. Circulation 80:1675, 1989 Packer M, Lee WH, Kessler PD, et al: Role of neurohormonal mechanisms in determining survival in patients with severe chronic heart failure. Circulation 75:IV80, 1987 Taggart P, Sutton P, John R, et al: Monophasic action potential recordings during acute changes in ventricular loading induced by the Valsalva manoeuvre. Br Heart J 67:221, 1992 Pak PH, Nuss HB, Tunin RS, et al: Repolarization abnormalities, arrhythmia and sudden death in canine tachycardia-induced cardiomyopathy. J Am Coll Cardiol 30:576, 1997 Li HG, Jones DL, Yee R, et al: Electrophysiologic substrate associated with pacing-induced heart failure in dogs: potential value of programmed stimulation in predicting sudden death. J Am Coll Cardiol 19:444, 1992 Ishida S, Nakagawa M, Fujino T, et al: Circadian variation of QT interval dispersion: Correlation with heart rate variability. J Electrocardiol 30:205, 1997 Millar CK, Kralios FA, Lux RL: Correlation between refractory periods and activation-recovery intervals from electrograms: Effects of rate and adrenergic interventions. Circulation 72:1372, 1985