Clin. Cardiol. 15, 911-915 (1992)
Influence of Atenolol on the Relationship between Heart Rate and QT Interval in Patients with Exercise-Induced Myocardial Ischemia SERGIO FERRARO, M.D., GIOVANNI MADDALENA, M.D., VICENTE D'AGOSTO, M.D., MICHELE D' ALTO,M.D., M.D., MASSIMO ROMANO, M.D.,*MASSIMO CHIARIELLO, M.D. SERAFINO Fkzro, M.D.,**MAURIZIO SANTOMAURO, Department of Cardiology and Cardiovascular Surgery; *I Medical School Cardiopulmonary Exercise Laboratory; **IV Division of Internal Medicine, Second School of Medicine, Federico I1 University, Naples, Italy
Summary: The aim of this study was to analyze the relationship between heart rate and QT interval (HR-QT) during exercise in control subjects (Group A) and in patients with coronary artery disease (CAD) with effort angina and without previous myocardial infarction (MI) (Group B). The diagnosis of CAD was confirmed by coronarographic examination. The correlation HR-QT was significant (p < 0.001) in both groups on effort and at recovery. The analysis of the regression HR-QT was carried out separately, both on effort in upright position and at rest in supine position, to avoid the influence of posture on QT length. During effort, the regression line showed lower slope and intercept values in Group B (p1 mm) frequently occurred, a longer QT interval was present in Group B. Moreover, in Group B, the QT interval in the presence of ECG signs of ischemia was significantly longer (peO.01) than in Group A at comparable heart rates both on effort and at rest, thereby confirming the result obtained by comparing both regression lines. The same effort protocol was repeated in Group B patients after acute administration of atenolol 100 mg per 0s. After
Address for reprints:
Dr. Sergio Ferraro Cattedra di Cardiologia 11"Policlinico Via S. Pansini 5 80131 Napoli, Italy Received: February 29,1992 Accepted with revision: August 26, 1992
atenolol administration, the analysis of the regression HR-QT in Group B clearly showed a shorter QT interval than that obtained in washout period during the baseline test at the highest heart rates where the ECG frequently showed signs of ischemia. This was proven by the presence of a higher slope (p c 0.05) of the regression line HRQT during treatment compared with that obtained in the same group without therapy. In conclusion, transient myocardial ischemia provokes a longer QT interval in patients with CAD in comparison with normal subjects. Aten0101 affected the HR-QT relationship showing a shorter QT interval at the highest heart rates, particularly when ECG showed signs of ischemia, compared with that achieved in the same patients in washout.
Key words: ischemia, beta blockade, HR-QT effort relationship
Introduction The relationship between heart rate (HR) and QT duration at rest can be influenced by several and acquired conditions: Among the acquired conditions, electrolyte abn~rmalities,~ and the use of pharmacologic agents6 including quinidine, procainamide, and amiodarone for intraventricular conduction abnormalities7 can influence ventricular repolarization. On the other hand, in a relatively recent editorial Vaughan Williams8 discussed the prolonged QT interval as a manifestation of prolonged repolarization reflective of a potentially beneficial effect of antiarrhythmic drugs that prolong the duration of ventricular action potential. Finally, myocardial ischemia and particularly myocardial infarction (MI)9 can prolong QT length. Moreover, for the patients with prolonged QT interval and recurrent ventricular arrhythmias, antiarrhythmic agents that prolong the QT interval further should be avoided. On the other hand, beta-blocking agents are effec-
912
Clin. Cardiol. Vol. 15, December 1992
tive in reducing the frequency of syncope and sudden death. In a previous report we evaluated and analyzed HR-QT relationship during effort in normal subjects and in patients with coronary artery disease (CAD) during upright bicycle ergometer stress test.I0The current study was designed to investigatethe HR-QT relationship (Figs. 1,2, Table I) in patients with stable effort angina and exercise ECG signs of ischemia without previous MI and to analyze the effect of atenolol per 0s on this relationship.
Methods Patients
Two groups of patients were studied: a group of 20 patients (13 men and 7 women, mean age 5 l .6 & 5. l years) with stable angina and transient ECG signs of ischemia during a previous effort test (Group B), and a control group of 27 normal subjects (22 men and 5 women, mean age 48.7f7.1 years) (Group A). The diagnosis of CAD in Group B was confirmed by previous coronarographic examination showing stenosis of at least 75% of the diameter in one of three main coronary vessels. The patients with CAD had not suffered from previous MI. The diagnosis of stable effort angina was made using a treadmill stress test at the end of a 10-day period of pharmacological washout; no patient had used amiodarone-only sublingual nitroglycerin was permitted. A second test carried out 24 h after the first was analyzed for the present study. This test showed the presence of stable effort angina and typical STsegment depression of at least 1 mm at a time and at a workload which did not differ more than 20% from the first test. The same 20 patients (Group B) repeated a third exercise test 48-72 h after the second test, 200-240 min after administrationper 0s of atenolol 100 mg. All patients had discontinued drug therapy at least 1 week before the first exercise test. During the study, patients took only sublingual isosorbide dinitrate. Amiodarone was not permitted. The levels of plasma electrolytes (Na, K, Ca) were in the normal range during the study period. Exercise Test
The test was carried out on a treadmill using the Bruce protocol and utilizing the Exer Stress Avionic Mod. 3000 following 30 min of rest at constant temperature in a quiet room. ECG was continuously monitored in three leads (DII, CM2, CM5) and recorded at a paper speed of 25 mm/s for 5 s and at 50 mm/s for 5 s prior to the test in upright position time 0; every 3 min (at the end of every stage) during effort; every min in supine position during the recovery period for the first 5 min; and finally every 5 min until complete recovery of HR, blood pressure (BP), and ST level. During effort, HR was monitored on a digital display, and ST level was automaticallyand continuously displayed. Systolic blood pressure (SBP) was recorded every min by
an automatic device (Metronic). HR, SBP and ST level were measured at time 0; at the end of each stage of exercise (3,6,9 min, etc.); at the end of the test; and every min during the recovery period for the first 5 min and then every 5 min. QT interval was measured on ECG from the beginning of a Q wave to the end of the T wave by two different physicians using the method described by Cinca et aZ.ll and Romano et al.l0 Physicians were blinded to the identity of the patients and their ECGs and to which phase of the study was involved, in order to avoid influence on the results. ST and QT were obtained as the mean of six measurements, three for each physician, and CM5 was used to measure these parameters. To avoid the interference of body position on QT duration, recordings during effort (standing) were considered separately from those during recovery in supine position. According to our previous and actual experiences, the difficulty in determining the end of the T wave can be greater when HR is lower or when the paper speed is faster (50 d s ) . In fact, at the highest HRs, the end of the T wave appears more clearly for the heart cycle immediately following. In our report QT length was read at 25 mm/s paper speed. The test was interrupted according to the standards proposed by American Heart Association.12 Finally, QT duration was measured in correspondence to ECG signs of ischemia (ST > 1 mm) in Group B and compared with the results obtained at similar HR in Group A (Table II). (The HR shown in Table III is different from that in Table I1 because a HR comparable to that during atenolol was always chosen in Group B.) Statistics
We tested the linear regression for HR-QT in normal subjects and in CAD patients in the basal state and after medication with atenolol (in Group B only) during effort and at rest. The regressions obtained in Groups A and B, and also in Group B after atenolol administration, were compared for slopes and intercepts (Figs. 1 and 2). QT dur-
400 *
350
280 300
-
1..
n=125 ,
80
100
.
\..
120 140
Heart rate (beatdmin)
,001;
.
,
.
,
,
70 80 90100110 Heart rate (beatslmin)
FIG.1 Regression lines between heart rate (x) and QT (y) on effort and during recovery achieved in Group A (normal subjects) (m *) and in Group B (patients with CAD) (). The comparison of regression lines is shown in Table I.
S. Ferraro et al.: Effort QT, ischemia, and influence of atenolol
400 ~ ~ 4 6 5 -. 21 . 0 3 . ~ .x
y = 453.5 Y = 523.9
400'
-
0.91 . x 1.92 . X
350-
300 280
1
n=96
'\. 4 / .. 130
, , 70 80 90100 110
Heart rate (beatshin)
Heart rate (beatdmin)
80
100
115
shown in Tables I1 and 111. A p value of at least 5 0.05 was considered significant.
Results
350
L+
913
I
FIG.2 Regression lines between heart rate (x) and QT (y) on effort and during recovery achieved in Group B during washout (-) and after medication with atenolol ( parison of regression lines is shown in Table I.
). The com-
ing ischemia (ST depression > 1 mm) in Group B was compared with the QT length obtained in Group A at comparable HRs using the t-test for unpaired data (Table 11). Nonlinear regression HR-QT was not considered by itself because it was used only to compare different behavior in the group of normal subjects and in ischemic patients. In addition, in Group B, the t-test for paired data was carried out to compare QT length during ischemia with that obtained after medication with atenolol in the same patients both at rest and on effort, always at comparable HRs (Table 111). In our opinion the reliability of the regression lines results either from the extremely high number of points (as shown in Figs. 1 and 2 ) tested for in each group and every situation (at rest, on effort, before, and after beta-
The correlation coefficient of the HR-QT relationship is statistically significant in both groups (pI 10 beatdmin), QT is clearly and progressively longer in Group B. To outline this result better, QT was measured in the presence of ST signs of ischemia in Group B and compared with QT obtained in Group A at comparable HRs. QT was significantly longer in Group B than in Group A (p 1 mm), QT is shorter, even though not significantly, both during effort and at rest (considered separately) in the test taken after atenolol adminis-
TABLE I Regression lines, regression coefficient (r), comparison () of regression lines and statistical significance (p) of the relation of HR (x) and QT (y) in a group of normal subjects (Group A) and in a group of patients with CAD (Group B) in washout and after acute medication with atenolol
Effort Group A Group B Group B + atenolol Comparison of regression lines: Group A Group B: slope (NS); intercept (pQS2 secondary to coronary artery disease. Am J Cardiol50, 1229-1235 (1982) 21. The QT>QS2 syndrome: A new mortality risk indicator in coronary artery disease. Am J Cardiol55,916-920 (1985) 22. Lollgen H,Bonze1 T, Schonrich G, Hust M: Ventricular arrhythmia and QT-prolongation in exercise testing (abstr). J Am Coll Cardiol 1(2), 586-590 (1983) 23. Reinke A, Michel D, Mathes P: Arrythmogenic potential of exercise induced myocardial ischemia. Eur Heart J 8, 119-1 24 ( 1987) 24. Romano M, Cotecchia MR, Di Mar0 T, Caiazzo MR, Golia B, Adinolfi L: Relazione frequenza cardiaca-intervallo QT in coronaropatici. Influenza dei segni elettrocardiografici di ischemia miocardica indotta dallo sforzo. 88"Congr SOC Ital Med Interna, Roma 1987. I1 Policlinico 94, 17-18 (1987) 25. Edvardsson N, Bertil Olsson S: Effects of acute and chronic beta-receptor blockade on ventricular repolarisation in man. Br HeartJ45,628-636 (1981) 26. Warren SG, Bremer DL, Orgain ES: Long-term propranolol therapy for angina pectoris. Am J Cardiol37,420-428 (1976) 27. Ferraro S, Fazio S, Santomauro M, De Magistris L, Cianfrani M, Sac& L: Elettrocardiogramma da sforzo nella valutazione della aterosclerosi coronarica: Correlazione fra segni elettrocardiografici di ischemia ed esame coronarografico. Cardiologia 34,327-33 1 (1989)
Influence of Atenolol on the Relationship between Heart Rate and QT Interval in Patients with Exercise-Induced Myocardial Ischemia SERGIO FERRARO, M.D., GIOVANNI MADDALENA, M.D., VICENTE D'AGOSTO, M.D., MICHELE D' ALTO,M.D., M.D., MASSIMO ROMANO, M.D.,*MASSIMO CHIARIELLO, M.D. SERAFINO Fkzro, M.D.,**MAURIZIO SANTOMAURO, Department of Cardiology and Cardiovascular Surgery; *I Medical School Cardiopulmonary Exercise Laboratory; **IV Division of Internal Medicine, Second School of Medicine, Federico I1 University, Naples, Italy
Summary: The aim of this study was to analyze the relationship between heart rate and QT interval (HR-QT) during exercise in control subjects (Group A) and in patients with coronary artery disease (CAD) with effort angina and without previous myocardial infarction (MI) (Group B). The diagnosis of CAD was confirmed by coronarographic examination. The correlation HR-QT was significant (p < 0.001) in both groups on effort and at recovery. The analysis of the regression HR-QT was carried out separately, both on effort in upright position and at rest in supine position, to avoid the influence of posture on QT length. During effort, the regression line showed lower slope and intercept values in Group B (p1 mm) frequently occurred, a longer QT interval was present in Group B. Moreover, in Group B, the QT interval in the presence of ECG signs of ischemia was significantly longer (peO.01) than in Group A at comparable heart rates both on effort and at rest, thereby confirming the result obtained by comparing both regression lines. The same effort protocol was repeated in Group B patients after acute administration of atenolol 100 mg per 0s. After
Address for reprints:
Dr. Sergio Ferraro Cattedra di Cardiologia 11"Policlinico Via S. Pansini 5 80131 Napoli, Italy Received: February 29,1992 Accepted with revision: August 26, 1992
atenolol administration, the analysis of the regression HR-QT in Group B clearly showed a shorter QT interval than that obtained in washout period during the baseline test at the highest heart rates where the ECG frequently showed signs of ischemia. This was proven by the presence of a higher slope (p c 0.05) of the regression line HRQT during treatment compared with that obtained in the same group without therapy. In conclusion, transient myocardial ischemia provokes a longer QT interval in patients with CAD in comparison with normal subjects. Aten0101 affected the HR-QT relationship showing a shorter QT interval at the highest heart rates, particularly when ECG showed signs of ischemia, compared with that achieved in the same patients in washout.
Key words: ischemia, beta blockade, HR-QT effort relationship
Introduction The relationship between heart rate (HR) and QT duration at rest can be influenced by several and acquired conditions: Among the acquired conditions, electrolyte abn~rmalities,~ and the use of pharmacologic agents6 including quinidine, procainamide, and amiodarone for intraventricular conduction abnormalities7 can influence ventricular repolarization. On the other hand, in a relatively recent editorial Vaughan Williams8 discussed the prolonged QT interval as a manifestation of prolonged repolarization reflective of a potentially beneficial effect of antiarrhythmic drugs that prolong the duration of ventricular action potential. Finally, myocardial ischemia and particularly myocardial infarction (MI)9 can prolong QT length. Moreover, for the patients with prolonged QT interval and recurrent ventricular arrhythmias, antiarrhythmic agents that prolong the QT interval further should be avoided. On the other hand, beta-blocking agents are effec-
912
Clin. Cardiol. Vol. 15, December 1992
tive in reducing the frequency of syncope and sudden death. In a previous report we evaluated and analyzed HR-QT relationship during effort in normal subjects and in patients with coronary artery disease (CAD) during upright bicycle ergometer stress test.I0The current study was designed to investigatethe HR-QT relationship (Figs. 1,2, Table I) in patients with stable effort angina and exercise ECG signs of ischemia without previous MI and to analyze the effect of atenolol per 0s on this relationship.
Methods Patients
Two groups of patients were studied: a group of 20 patients (13 men and 7 women, mean age 5 l .6 & 5. l years) with stable angina and transient ECG signs of ischemia during a previous effort test (Group B), and a control group of 27 normal subjects (22 men and 5 women, mean age 48.7f7.1 years) (Group A). The diagnosis of CAD in Group B was confirmed by previous coronarographic examination showing stenosis of at least 75% of the diameter in one of three main coronary vessels. The patients with CAD had not suffered from previous MI. The diagnosis of stable effort angina was made using a treadmill stress test at the end of a 10-day period of pharmacological washout; no patient had used amiodarone-only sublingual nitroglycerin was permitted. A second test carried out 24 h after the first was analyzed for the present study. This test showed the presence of stable effort angina and typical STsegment depression of at least 1 mm at a time and at a workload which did not differ more than 20% from the first test. The same 20 patients (Group B) repeated a third exercise test 48-72 h after the second test, 200-240 min after administrationper 0s of atenolol 100 mg. All patients had discontinued drug therapy at least 1 week before the first exercise test. During the study, patients took only sublingual isosorbide dinitrate. Amiodarone was not permitted. The levels of plasma electrolytes (Na, K, Ca) were in the normal range during the study period. Exercise Test
The test was carried out on a treadmill using the Bruce protocol and utilizing the Exer Stress Avionic Mod. 3000 following 30 min of rest at constant temperature in a quiet room. ECG was continuously monitored in three leads (DII, CM2, CM5) and recorded at a paper speed of 25 mm/s for 5 s and at 50 mm/s for 5 s prior to the test in upright position time 0; every 3 min (at the end of every stage) during effort; every min in supine position during the recovery period for the first 5 min; and finally every 5 min until complete recovery of HR, blood pressure (BP), and ST level. During effort, HR was monitored on a digital display, and ST level was automaticallyand continuously displayed. Systolic blood pressure (SBP) was recorded every min by
an automatic device (Metronic). HR, SBP and ST level were measured at time 0; at the end of each stage of exercise (3,6,9 min, etc.); at the end of the test; and every min during the recovery period for the first 5 min and then every 5 min. QT interval was measured on ECG from the beginning of a Q wave to the end of the T wave by two different physicians using the method described by Cinca et aZ.ll and Romano et al.l0 Physicians were blinded to the identity of the patients and their ECGs and to which phase of the study was involved, in order to avoid influence on the results. ST and QT were obtained as the mean of six measurements, three for each physician, and CM5 was used to measure these parameters. To avoid the interference of body position on QT duration, recordings during effort (standing) were considered separately from those during recovery in supine position. According to our previous and actual experiences, the difficulty in determining the end of the T wave can be greater when HR is lower or when the paper speed is faster (50 d s ) . In fact, at the highest HRs, the end of the T wave appears more clearly for the heart cycle immediately following. In our report QT length was read at 25 mm/s paper speed. The test was interrupted according to the standards proposed by American Heart Association.12 Finally, QT duration was measured in correspondence to ECG signs of ischemia (ST > 1 mm) in Group B and compared with the results obtained at similar HR in Group A (Table II). (The HR shown in Table III is different from that in Table I1 because a HR comparable to that during atenolol was always chosen in Group B.) Statistics
We tested the linear regression for HR-QT in normal subjects and in CAD patients in the basal state and after medication with atenolol (in Group B only) during effort and at rest. The regressions obtained in Groups A and B, and also in Group B after atenolol administration, were compared for slopes and intercepts (Figs. 1 and 2). QT dur-
400 *
350
280 300
-
1..
n=125 ,
80
100
.
\..
120 140
Heart rate (beatdmin)
,001;
.
,
.
,
,
70 80 90100110 Heart rate (beatslmin)
FIG.1 Regression lines between heart rate (x) and QT (y) on effort and during recovery achieved in Group A (normal subjects) (m *) and in Group B (patients with CAD) (). The comparison of regression lines is shown in Table I.
S. Ferraro et al.: Effort QT, ischemia, and influence of atenolol
400 ~ ~ 4 6 5 -. 21 . 0 3 . ~ .x
y = 453.5 Y = 523.9
400'
-
0.91 . x 1.92 . X
350-
300 280
1
n=96
'\. 4 / .. 130
, , 70 80 90100 110
Heart rate (beatshin)
Heart rate (beatdmin)
80
100
115
shown in Tables I1 and 111. A p value of at least 5 0.05 was considered significant.
Results
350
L+
913
I
FIG.2 Regression lines between heart rate (x) and QT (y) on effort and during recovery achieved in Group B during washout (-) and after medication with atenolol ( parison of regression lines is shown in Table I.
). The com-
ing ischemia (ST depression > 1 mm) in Group B was compared with the QT length obtained in Group A at comparable HRs using the t-test for unpaired data (Table 11). Nonlinear regression HR-QT was not considered by itself because it was used only to compare different behavior in the group of normal subjects and in ischemic patients. In addition, in Group B, the t-test for paired data was carried out to compare QT length during ischemia with that obtained after medication with atenolol in the same patients both at rest and on effort, always at comparable HRs (Table 111). In our opinion the reliability of the regression lines results either from the extremely high number of points (as shown in Figs. 1 and 2 ) tested for in each group and every situation (at rest, on effort, before, and after beta-
The correlation coefficient of the HR-QT relationship is statistically significant in both groups (pI 10 beatdmin), QT is clearly and progressively longer in Group B. To outline this result better, QT was measured in the presence of ST signs of ischemia in Group B and compared with QT obtained in Group A at comparable HRs. QT was significantly longer in Group B than in Group A (p 1 mm), QT is shorter, even though not significantly, both during effort and at rest (considered separately) in the test taken after atenolol adminis-
TABLE I Regression lines, regression coefficient (r), comparison () of regression lines and statistical significance (p) of the relation of HR (x) and QT (y) in a group of normal subjects (Group A) and in a group of patients with CAD (Group B) in washout and after acute medication with atenolol
Effort Group A Group B Group B + atenolol Comparison of regression lines: Group A Group B: slope (NS); intercept (pQS2 secondary to coronary artery disease. Am J Cardiol50, 1229-1235 (1982) 21. The QT>QS2 syndrome: A new mortality risk indicator in coronary artery disease. Am J Cardiol55,916-920 (1985) 22. Lollgen H,Bonze1 T, Schonrich G, Hust M: Ventricular arrhythmia and QT-prolongation in exercise testing (abstr). J Am Coll Cardiol 1(2), 586-590 (1983) 23. Reinke A, Michel D, Mathes P: Arrythmogenic potential of exercise induced myocardial ischemia. Eur Heart J 8, 119-1 24 ( 1987) 24. Romano M, Cotecchia MR, Di Mar0 T, Caiazzo MR, Golia B, Adinolfi L: Relazione frequenza cardiaca-intervallo QT in coronaropatici. Influenza dei segni elettrocardiografici di ischemia miocardica indotta dallo sforzo. 88"Congr SOC Ital Med Interna, Roma 1987. I1 Policlinico 94, 17-18 (1987) 25. Edvardsson N, Bertil Olsson S: Effects of acute and chronic beta-receptor blockade on ventricular repolarisation in man. Br HeartJ45,628-636 (1981) 26. Warren SG, Bremer DL, Orgain ES: Long-term propranolol therapy for angina pectoris. Am J Cardiol37,420-428 (1976) 27. Ferraro S, Fazio S, Santomauro M, De Magistris L, Cianfrani M, Sac& L: Elettrocardiogramma da sforzo nella valutazione della aterosclerosi coronarica: Correlazione fra segni elettrocardiografici di ischemia ed esame coronarografico. Cardiologia 34,327-33 1 (1989)
