RR Variability and Baroreflex Sensitivity in Patients with Ventricular Tachycardia Associated with Normal Heart and Patients with Ischemic Heart Disease JASWINDER S. GILL, THOMAS FARRELL, A. BASZKO, DAVID E. WARD, and A. JOHN CAMM From the Department of Cardiological Sciences, St. George's Hospital Medical School, London, United Kingdom
GILL, J.S., ET AL.: RR Variability and Baroreflex Sensitivity in Patients with Ventricular Tachycardia Associated with Normal Heart and Patients with Ischemic Heart Disease. Recent studies have suggested that disordered autonomic/unction, particularly the loss of protective vagal reflexes are associated with an increased incidence of arrhythmic deaths foUowing myocardial infarction (MI). Heart rate variability (HRVj and haroreflex sensitivity (BRS) are measures of myocardial autonomic /unction and predict arrhythmic deaths post-MI. Patients with ventricular tachycardia associated with a "normal heart" frequently have exercise-induced arrhythmia suggesting that the autonomic nervous system is important in the genesis of this form of ventricular tachycardia (VTj. This study examines HRV and BRS in patients with VT associated with a "normaJ heart" and compares these values to patients post-MI with and without evidence o/arrhythmia. Twenty patients with VT associated with a "normal heart," 16 patients with MI but without arrhythmia on follow-up, and 11 patients with MI and VT on follow-up were studied. HRV was measured from 24-hour Holter recordings and BRS was measured from plots of change in systolic hlood pressure versus change in heart rate/ollowing an intravenous injection of 0.4-0.6 mg phenylephrine. HRV was significantly higher in the patients with VT associated with a normal heart (34.2 ± 10.8 msec] compared to the patients post-MI, without (23.7 ± 6.7 msec) and with (14.8 ± 3.8 msecj arrhythmia (F = 9.2, P < 0.001) and these differences were unaffected by adjustment for age. Baroreflex sensitivity was also higher in patients with VT associated with a "normal heart" (10.1 ± 6.8 msec/mmHgj compared to patients post-MI, without (6.1 ± 3.2 msec/mmHgj and with 3.2 ± 3.1 msec/mmHgj arrhythmia, (F = 7.2, P < 0.02), though statistical significance was lost after adjustment for age (F = 1.2, P = 0.3). We conclude that patients with VT associated with "normal hearts" have HRV and BRS that is higher than in patients post-MI. Alterations o/autonomic tone are, there/ore, unlikely to be important in VT associated with a "normal heart," whereas these appear to be important in patients with arrhythmic events postMI. (PACE, Vol. 14, November, Part II 1991)
Introduction Recently, there has been growing awareness of the importance of neural mechanisms in arrhythmogenesis and the prognostic value of autonomic function tests such as heart rate variahility (HRV) analysis and haroreflex sensitivity (BRS)
Address for reprints: Dr. J.S. Gill, Cardiological Sciences, St. George's Hospital Medical School, Granmer Terrace, London SW17 ORE, U.K. Fax: 081-767-7141.
2016
following myocardial infarction (MI).^"" Marked disturbances of autonomic function are seen in the acute and convalescent phases of MI^''^ and human and animal studies suggest that the preservation of vagal reflexes post-MI protect against the induction of life-threatening ventricular arrhythmias.^"® HRV and arterial haroreflex testing provide information on the function of the autonomic nervous system and have value in risk stratification postI^j 10.11 Patients with ventricular tachycardia (VT) associated with "normal heart" are generally
November 1991, Part II
PACE, Vol. 14
AUTONOMIC FUNCTION IN VENTRICULAR TACHYCARDIA
young, with only a slight excess of males, unlike ischemic heart disease. Patients in this group generally suffer from frequent attacks of nonsustained VT, though sustained attacks of arrhythmia are present in a significant proportion and in some cases are associated with syncope.^^'^^ Although the noninvasive and invasive studies of the heart reveal no obvious disorder in these patients, histologic examination of the myocardium exposes abnormalities (principally fibrosis) in a significant proportion.^* The question of risk of arrhythmic death in this group remains unresolved. While the majority of reports suggest that patients are not at risk of sudden death,^^'^'' there are several that demonstrate fatalities in subjects with this condition.^^'^'' No study, to date, has addressed the question whether abnormalities of autonomic function are found in patients with VT associated with "normal heart" and whether such abnormalities may help prognostication. The purpose of this study was to examine HRV and in patients with VT associated with a "normal heart" and compare these results to those from patients with MI, with and without ventricular arrhythmia. Patients and Methods Patients with Ventricular Tachycardia in "Normal Heart" All patients had VT documented on multiple ECG leads. Patients all had a normal resting electrocardiograph and cardiothoracic ratio below 50% on chest X ray. All patients had left ventricular and coronary angiography, and the cineangiograms were reviewed by two independent physicians and classified as normal, with the absence of global or regional wall motion abnormality or significant stenoses in the coronary arteries. All patients underwent recording of a signal-averaged ECG, detailed echocardiography, and biopsy of the right ventricle. Patients were studied in a drugfree state in all cases. Patients with Ischemic Heart Disease The patients with ischemic heart disease were randomly selected, within age strata, from those admitted to St. George's Hospital with an acute MI. Patients over the age of 70 years or those having coexisting valvular heart disease, insulin depen-
PACE, Vol. 14
dent diabetes, arterial hypertension (>160/90 mmHg), atrial fibrillation, or evidence of sinoatrial disease were excluded. There was no use of digitalis or angiotensin converting enzyme inhibitors in the study group and beta blockade was stopped at least 48 hours prior to study. Patients were followed in a special clinic and the minimum followup was at least 1 year. The end points in the study were cardiac death and life-threatening arrhythmic events including sudden death (Cardiac Arrhythmia Pilot Study definition).^^ Baroreflex Sensitivity Testing This was assessed using a method previously described.^^ All patients were studied in a fasting and supine state between day 7-10 in the subjects post-MI. A right femoral line (F gauge 8) was used to record arterial pressure with simultaneous recording of a single ECC lead at 25 mm/second. After 10 minutes to allow stabilization, a bolus of 0.2 mg phenylephrine was injected intravenously over 15 seconds via a peripheral vein. Progressively larger doses were administered until a rise in systolic blood pressure of between 15 and 40 mmHg was obtained. At least three recordings were made using the optimum bolus dose. BRS was calculated by plotting the beat-to-beat change in RR interval against the beat-to-beat change in systolic blood pressure. BRS was then estimated as the value of the slope from the regression analysis of these plots. Only regression lines with a correlation coefficient >0.8 were accepted for analysis. At least three such slopes were calculated for each patient and the mean of these was taken as the BRS and expressed in msec/mmHg. During the procedure patients were asked to breathe at a normal rate and to avoid slow or forced respiration. BRS of
GILL, J.S., ET AL.: RR Variability and Baroreflex Sensitivity in Patients with Ventricular Tachycardia Associated with Normal Heart and Patients with Ischemic Heart Disease. Recent studies have suggested that disordered autonomic/unction, particularly the loss of protective vagal reflexes are associated with an increased incidence of arrhythmic deaths foUowing myocardial infarction (MI). Heart rate variability (HRVj and haroreflex sensitivity (BRS) are measures of myocardial autonomic /unction and predict arrhythmic deaths post-MI. Patients with ventricular tachycardia associated with a "normal heart" frequently have exercise-induced arrhythmia suggesting that the autonomic nervous system is important in the genesis of this form of ventricular tachycardia (VTj. This study examines HRV and BRS in patients with VT associated with a "normaJ heart" and compares these values to patients post-MI with and without evidence o/arrhythmia. Twenty patients with VT associated with a "normal heart," 16 patients with MI but without arrhythmia on follow-up, and 11 patients with MI and VT on follow-up were studied. HRV was measured from 24-hour Holter recordings and BRS was measured from plots of change in systolic hlood pressure versus change in heart rate/ollowing an intravenous injection of 0.4-0.6 mg phenylephrine. HRV was significantly higher in the patients with VT associated with a normal heart (34.2 ± 10.8 msec] compared to the patients post-MI, without (23.7 ± 6.7 msec) and with (14.8 ± 3.8 msecj arrhythmia (F = 9.2, P < 0.001) and these differences were unaffected by adjustment for age. Baroreflex sensitivity was also higher in patients with VT associated with a "normal heart" (10.1 ± 6.8 msec/mmHgj compared to patients post-MI, without (6.1 ± 3.2 msec/mmHgj and with 3.2 ± 3.1 msec/mmHgj arrhythmia, (F = 7.2, P < 0.02), though statistical significance was lost after adjustment for age (F = 1.2, P = 0.3). We conclude that patients with VT associated with "normal hearts" have HRV and BRS that is higher than in patients post-MI. Alterations o/autonomic tone are, there/ore, unlikely to be important in VT associated with a "normal heart," whereas these appear to be important in patients with arrhythmic events postMI. (PACE, Vol. 14, November, Part II 1991)
Introduction Recently, there has been growing awareness of the importance of neural mechanisms in arrhythmogenesis and the prognostic value of autonomic function tests such as heart rate variahility (HRV) analysis and haroreflex sensitivity (BRS)
Address for reprints: Dr. J.S. Gill, Cardiological Sciences, St. George's Hospital Medical School, Granmer Terrace, London SW17 ORE, U.K. Fax: 081-767-7141.
2016
following myocardial infarction (MI).^"" Marked disturbances of autonomic function are seen in the acute and convalescent phases of MI^''^ and human and animal studies suggest that the preservation of vagal reflexes post-MI protect against the induction of life-threatening ventricular arrhythmias.^"® HRV and arterial haroreflex testing provide information on the function of the autonomic nervous system and have value in risk stratification postI^j 10.11 Patients with ventricular tachycardia (VT) associated with "normal heart" are generally
November 1991, Part II
PACE, Vol. 14
AUTONOMIC FUNCTION IN VENTRICULAR TACHYCARDIA
young, with only a slight excess of males, unlike ischemic heart disease. Patients in this group generally suffer from frequent attacks of nonsustained VT, though sustained attacks of arrhythmia are present in a significant proportion and in some cases are associated with syncope.^^'^^ Although the noninvasive and invasive studies of the heart reveal no obvious disorder in these patients, histologic examination of the myocardium exposes abnormalities (principally fibrosis) in a significant proportion.^* The question of risk of arrhythmic death in this group remains unresolved. While the majority of reports suggest that patients are not at risk of sudden death,^^'^'' there are several that demonstrate fatalities in subjects with this condition.^^'^'' No study, to date, has addressed the question whether abnormalities of autonomic function are found in patients with VT associated with "normal heart" and whether such abnormalities may help prognostication. The purpose of this study was to examine HRV and in patients with VT associated with a "normal heart" and compare these results to those from patients with MI, with and without ventricular arrhythmia. Patients and Methods Patients with Ventricular Tachycardia in "Normal Heart" All patients had VT documented on multiple ECG leads. Patients all had a normal resting electrocardiograph and cardiothoracic ratio below 50% on chest X ray. All patients had left ventricular and coronary angiography, and the cineangiograms were reviewed by two independent physicians and classified as normal, with the absence of global or regional wall motion abnormality or significant stenoses in the coronary arteries. All patients underwent recording of a signal-averaged ECG, detailed echocardiography, and biopsy of the right ventricle. Patients were studied in a drugfree state in all cases. Patients with Ischemic Heart Disease The patients with ischemic heart disease were randomly selected, within age strata, from those admitted to St. George's Hospital with an acute MI. Patients over the age of 70 years or those having coexisting valvular heart disease, insulin depen-
PACE, Vol. 14
dent diabetes, arterial hypertension (>160/90 mmHg), atrial fibrillation, or evidence of sinoatrial disease were excluded. There was no use of digitalis or angiotensin converting enzyme inhibitors in the study group and beta blockade was stopped at least 48 hours prior to study. Patients were followed in a special clinic and the minimum followup was at least 1 year. The end points in the study were cardiac death and life-threatening arrhythmic events including sudden death (Cardiac Arrhythmia Pilot Study definition).^^ Baroreflex Sensitivity Testing This was assessed using a method previously described.^^ All patients were studied in a fasting and supine state between day 7-10 in the subjects post-MI. A right femoral line (F gauge 8) was used to record arterial pressure with simultaneous recording of a single ECC lead at 25 mm/second. After 10 minutes to allow stabilization, a bolus of 0.2 mg phenylephrine was injected intravenously over 15 seconds via a peripheral vein. Progressively larger doses were administered until a rise in systolic blood pressure of between 15 and 40 mmHg was obtained. At least three recordings were made using the optimum bolus dose. BRS was calculated by plotting the beat-to-beat change in RR interval against the beat-to-beat change in systolic blood pressure. BRS was then estimated as the value of the slope from the regression analysis of these plots. Only regression lines with a correlation coefficient >0.8 were accepted for analysis. At least three such slopes were calculated for each patient and the mean of these was taken as the BRS and expressed in msec/mmHg. During the procedure patients were asked to breathe at a normal rate and to avoid slow or forced respiration. BRS of
