AJH
1990;
3:735-740
ORIGINAL CONTRIBUTIONS
Ventricular Arrhythmias in Hypertensive Left Ventricular Hypertrophy James M. McLenachan and Henry /. Dargie
Ventricular arrhythmias occur with increased fre quency in hypertensive patients with left ventricu lar hypertrophy (LVH). The relationships, however, between ventricular arrhythmias and coexistent cor onary artery disease, left ventricular dysfunction and left ventricular fibrosis have not been examined in hypertensive LVH. We carried out coronary arte riography on fifteen hypertensive patients with LVH and nonsustained ventricular tachycardia (^ 3 consecutive ventricular complexes) of whom nine (60%) were free of significant ( > 50% stenosis) coro nary disease. To identify other possible correlates of left ventricular arrhythmias, 28 patients with LVH, comprising 17 with ventricular tachycardia and 11 without ventricular arrhythmias, underwent quantitative assessment of left ventricular function (angiographic ejection fraction), left ven-
B
oth electrocardiography and echocardiogra phy ' provide evidence that left ventricular hy pertrophy (LVH) is associated with an increased risk of cardiovascular mortality, including sud den death. Both simple and complex ventricular ar rhythmias have recently been shown to occur more commonly in hypertensive patients with LVH than in 1
2 3
4
From the Department of Cardiology, The Western Infirmary, Glas gow, Scotland. Dr. McLenachan was supported by a British Heart Foundation Ju nior Research Fellowship. Address correspondence and reprint requests to James M. McLena chan, Department of Cardiology, The Western Infirmary, Glasgow, Scotland G i l 6NT.
tricular mass (echocardiography), and left ventricu lar fibrosis (endomyocardial biopsy). Ejection frac tion was not significantly different between the two groups (53 ± 8% ν 62 ± 2%, Ρ = NS). However, left ventricular mass was significantly greater (442 ± 28 g ν 339 ± 34 g, Ρ < .05) and percentage fibrosis significantly higher (19 ± 4% ν 3 ± 1%, Ρ < .001) in those patients with ventricular tachycardia. Thus ventricular arrhythmias in hypertensive pa tients with LVH cannot be entirely attributed to coexistent coronary disease, nor to left ventricular dysfunction, but are related to the degree of cardiac hypertrophy and subendocardial fibrosis. Am J Hy pertens 1990; 3:735-740 Ventricular arrhythmia, left ventricu lar hypertrophy, coronary disease, fibrosis.
KEY WORDS:
those without. ' In other forms of cardiac hypertrophy, such arrhythmias are predictive of subsequent sudden death ' and, conceivably, could contribute to the high mortality among hypertensive patients with LVH. However, the relationships between ventricular ar rhythmias in hypertensive LVH and other variables that might influence arrhythmia frequency, such as coexistent coronary artery disease or left ventricular dysfunction, are not known. In addition, no studies have addressed the relationship between structural ab normalities associated with cardiac hypertrophy, in cluding subendocardial fibrosis, and the occurrence of ventricular arrhythmias. Thus the aims of this study were to determine whether ventricular arrhythmias in hypertensive pa tients with LVH are invariably associated with coronary 5 6
7 8
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
Relationship to Coronary Artery Disease, Left Ventricular Dysfunction, and Myocardial Fibrosis
artery disease and to examine the relationships between ventricular arrhythmia, left ventricular function and subendocardial fibrosis, as determined by quantitative analysis of left ventricular endomyocardial biopsies, in a group of patients with hypertension and LVH. STUDY GROUPS
9
2
2
10
1
Clearly, the appropriate control group for this com parison would be hypertensive patients with LVH but without ventricular tachycardia or chest pain; however, it was considered unethical to subject such patients to cardiac catheterization and thus the control population (group Β, η = 17) comprised a group of patients with LVH but without significant arryhthmias and included
METHODS Coronary Arteriography Four views were obtained of the left coronary artery and two of the right. The films were reviewed independently by two observers and classified according to whether or not they demon strated occlusion or stenosis, the latter defined as a re duction in luminal diameter by less than 50%, 50 to 70% or > 70% in one or more of the major coronary vessels. Left Ventricular Mass Mass was estimated in all pa tients undergoing left ventricular biopsy from an Mmode echocardiogram using the cube formula with "Penn" convention as described by Devereux and Rei chek. 11
Left Ventricular Function Left ventricular ejection fraction was also measured in all patients undergoing biopsy; this was estimated from a single-plane cineangiogram (30° right anterior oblique projection), as de scribed by Greene, following injection of 40 to 50 mL nonionic contrast medium. 12
Left Ventricular Biopsy A 7F "pigtail" catheter was used to position a long sheath (CORDIS) in the left ventricular cavity. Three to five biopsies were taken from the free wall of the left ventricle, the direction of the sheath being adjusted slightly before each biopsy. Samples were then stained for light microscopy and coded such that the source of the biopsy was not known at the time of analysis. Sections were stained with Masson's trichrome connective tissue stain, which stains fi brous and other connective tissue blue/green, and Elastin-van Gieson, which stains collagen red/brown. The sections were examined at a magnification of 400 X. For each stain in each patient, 80 fields were examined using an eyepiece graticule with 25 points randomly distrib uted; the eyepiece was rotated four times through 90 ° to give a total of 100 points per field. The number of "hits," ie, .areas where the dot on the graticule overlay an area of fibrous tissue was then expressed as a percent age of the total number of points counted. The error in this type of analysis is a function of both the total number of points counted and the "area frac tion of interest." Thus, if the area fraction of interest is large, the inherent error of the measurement, for a given number of points counted, will be small. We elected to count 8000 points (80 high power fields) per patient because this gave an acceptable relative standard error of approximately 3% when the area of interest (fibrosis etc.) was 10% rising to an error of 11 % when the area of fibrosis was only 1%. 13
Statistical Analysis Values for left ventricular mass, ejection fraction, percentage fibrosis etc, are all ex pressed as mean ± SEM and differences between
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
Fifteen hypertensive patients underwent standard diag nostic coronary arteriography. All patients had electro cardiographic (SV1 + RV5 ^ 35 mm) and echocardio graphic (left ventricular mass index ^ 145 g / m for men and > 110 g / m for women ) evidence of left ventricu lar hypertrophy and all had at least one episode of nonsustained ventricular tachycardia (^ 3 consecutive ven tricular complexes at a heart rate ^ 1 2 0 beats/min) during 48 h ambulatory monitoring. The purpose of cor onary arteriography in this group was to determine whether the high grade ventricular arrhythmias de tected during ambulatory monitoring were attributable to coexistent disease of the major coronary vessels. None had a history of angina pectoris or electrocardio graphic evidence of previous myocardial infarction. Twelve were men and the age range was 30 to 69 years (mean 55 ± 4 years). Left ventricular endomyocardial biopsy was at tempted in 37 patients but sufficient material for analy sis (three to five biopsy samples per patient) was ob tained in only 28. The two most common reasons for biopsy failure were inability to obtain a satisfactory po sition within the ventricle and arrhythmia induction by the biopsy forceps; such arrhythmias were invariably self-terminating on withdrawal of the bioptome. The 28 patients comprised two groups: group A (n = 11), in whom ambulatory electrocardiographic monitoring had demonstrated at least one episode of nonsustained ven tricular tachycardia (VT), and group Β (η = 1 7 ) , in whom no complex ventricular arrhythmias had oc curred during 48 h monitoring. All 28 had both electro cardiographic and echocardiographic evidence of left ventricular hypertrophy but they differed in the indica tion for cardiac catheterization. Thus patients in group A were derived from the group described previously, ie, hypertensive patients undergoing diagnostic cardiac catheterization to identify a cause for their ventricular arrhythmia. The number of episodes of VT varied from 1 to 22 with a mean of 3. The duration of the episodes varied from 3 to 25 beats with a mean of 5. None of the patients showed any symptoms at the time of the arrhythmia.
hypertensive patients with LVH and chest pain (n = 12) and patients with LVH secondary to aortic valve disease (n = 5).
groups were calculated using a two-tailed t test for un paired data. Differences in the prevalence of coronary artery disease were calculated using a χ test. 2
RESULTS Relation of Ventricular Arrhythmias to Coronary Ar tery Disease Of the 15 patients with left ventricular hypertrophy and ventricular tachycardia, nine (60%) were free of significant stenoses of the epicardial coro nary vessels. Of the remainder, two had one or more coronary occlusions, and four had one or more stenoses of greater than 70%; none had triple vessel disease.
TABLE 1. CHARACTERISTICS OF PATIENTS UNDERGOING ENDOMYOCARDIAL BIOPSY (n = 28)
Age (years) Men: women Left ventricular mass (g) Systolic Blood Pressure (mm Hg) Number of diseased vessels None One Two Three VT = ventricular
Patients with VT (n = 11) 58 ± 4 8:3 442 ± 28 166 ± 8
7 0 2 2 tachycardia.
Patients without VT (n = 17) 54 ± 2 13:4 339 ± 34* 168 ± 6
5 3 3 6 * Ρ < .05.
Without VT FIGURE 1. Ejection fraction values (mean ± SEM) for patients with and without ventricular tachycardia.
patients with ventricular tachycardia who also had marked left ventricular dysfunction (ejection fraction values < 30%) (Figure 1), the difference in percentage fibrosis between those patients with and without ven tricular tachycardia remained highly significant (10 ± 2% ν 3 ± 1 %, Ρ < .01). A similar pattern was seen when the Elastin-van Gieson samples were analyzed (13 ± 3% ν 2 ± 1%, Ρ < .001) (Figure 3); concordance be tween the two staining techniques was high (r = 0.835, standard error of estimate 6.49), suggesting that both were detecting the same abnormality. While most patients with advanced fibrosis also had marked hypertrophy, the reverse was not always true, and, overall, there was no significant correlation be tween left ventricular mass and fibrosis (r = 0.12, Ρ = NS). A weak inverse relationship did exist between left ventricular ejection fraction and fibrosis (Figure 4) al though many patients with detectable fibrosis had a normal ejection fraction. DISCUSSION Our results suggest that the increased frequency of ven tricular arrhythmias among hypertensive patients with LVH cannot be attributed solely to coexistent coronary artery disease or left ventricular dysfunction.
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
Relation of Ventricular Arrhythmias to Ventricular Hypertrophy, Ventricular Function and Myocardial Fibrosis There were no significant age or sex differ ences between the groups with and without ventricular tachycardia (Table 1). Although all patients had left ven tricular hypertrophy, mean left ventricular mass was higher among those with ventricular tachycardia than among those without (442 ± 28 ν 339 ± 34 g, Ρ < .05) (Table 1). Those patients with ventricular tachycardia demonstrated a wide range of ejection fraction values (Figure 1), although the mean values for the two groups were not significantly different (53 ± 8% ν 62 ± 2%, Ρ = NS). There was also a trend, although not statisti cally significant, towards increased coronary disease among those patients without ventricular tachycardia, reflecting the inclusion in group Β of hypertensive pa tients being investigated because of chest pain. The most striking difference between the two groups, however, related to quantitative analysis of the myocar dial biopsies; mean percentage fibrosis (Masson's stain) was very significantly higher in those patients with ven tricular tachycardia than in those without (19 ± 4% ν 3 ± 1%, Ρ < .001) (Figure 2). Even after excluding four
The relationships between hypertrophy, coronary disease and ventricular arrhythmia has not previously been studied in hypertensive patients, but have been investigated in aortic stenosis. In a series of 102 patients, Klein reported high grade ventricular arrhythmias in 40 (39%) but, as in the present study, found no relationship between arrhythmias and the presence of coronary artery disease. The second part of our study examined the relationship between endomyocardial fibrosis, left ventricular mass and left ventricular function in patients with and without ventricular arrhythmias. As a result of differences in entry criteria, coronary artery disease was more common among patients without ventricular arrhythmias. Despite this bias, patients with ventricular arrhythmias had significantly more fibrosis than those without and also tended to have higher left ventricular mass values. Increased collagen content has previously been documented in left ventricular hypertrophy ' but the relationship between arrhythmias and fibrosis has not been examined in humans. Arrhythmias, however, have been described in an animal model of left ventricular hypertrophy. Cameron et al induced cardiac hypertrophy by aortic banding and noted that ventricular arrhythmias, including ventricular extrasystoles, couplets 16
19
Without VT FIGURE 2. Percentage fibrous and interstitial tissue volume (Masson's stain) for patients with and without ventricular tachycardia.
The first part of our study demonstrated that of 15 hypertensive patients with LVH and ventricular arrhythmias during ambulatory monitoring, but without a history of chest pain, the majority were free of major stenoses of the epicardial coronary vessels. This does not imply, however, that ventricular arrhythmias in hypertensive patients with LVH are unrelated to myocardial ischemia; it is well established that coronary vascular reserve is reduced in hypertensive LVH and that patients with cardiac hypertrophy develop typical angina pectoris with electrocardiographic evidence of myocardial ischemia even when the major coronary vessels are patient. Furthermore, we have no data relating to the functional importance of minor coronary stenoses in patients with abnormally high myocardial demand. Nevertheless, it is apparent from our data that the high prevalence of ventricular arrhythmias in hypertensive patients with LVH cannot be attributed to coexistent disease of the large coronary arteries, as defined by stenoses of 70% or greater. 14
15
Without VT
FIGURE 3. Percentage fibrous tissue volume (Elastin-van Gieson stain) for patients with and without ventricular tachycardia.
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
17 18
100
more extensive (Figure 4). However, caution is neces sary when interpreting our data, particularly when correlating changes in global left ventricular function with structural changes seen on subendocardial biopsy, in view of the marked regional and transmural ' variations in myocardial fibrosis. In summary, our findings suggest that ventricular ar rhythmias in hypertensive patients with left ventricular hypertrophy cannot be entirely attributed to coexistent coronary artery disease or left ventricular dysfunction but are associated with structural changes in the hyper trophied ventricle including the development of myo cardial fibrosis. 27
28 29
The authors gratefully acknowledge Dr. A. F. Lever of the Medical Research Council Blood Pressure Unit, Western Infir mary, Glasgow, for helpful advice during analysis of the data and Dr. G. B. Lindop, Senior Lecturer in Pathology at the Western Infirmary, Glasgow, for advice regarding the quanti tative analysis of myocardial biopsies. REFERENCES
% FIBROSIS FIGURE 4. Relationship between left ventricular ejection frac tion and percentage fibrosis for all 28 patients. Patients with ven tricular tachycardia are depicted by closed circles and patients without ventricular tachycardia by open circles.
and ventricular fibrillation, occurred more commonly than in control animals. Furthermore, in vitro testing showed that a variety of electrophysiological abnormali ties, including short action potential duration, low am plitude action potentials and electrically silent areas, oc curred at the border zone between hypertrophied myocardium and areas of marked connective tissue in filtration. These electrical abnormalities were not seen in endocardial preparations without endocardial fibro sis. In humans, several studies have shown that ventric ular arrhythmias can be induced by programmed elec trical stimulation in patients with hypertrophic cardiomyopathy ' in whom myofiber disarray and myocardial fibrosis are common. The relationship between fibrosis and cardiac func tion in various forms of left ventricular hypertrophy has been studied by other g r o u p s . In experimental hy pertension, systolic function declines with age, coin ciding with the development of myocardial fibrosis. Studies in humans suggest that systolic function may be preserved in the presence of quite marked fibrosis although diastolic function may be impaired at an ear lier stage. Our data suggest that systolic function may be pre served even when moderate amounts of fibrosis (5 to 15%) are present, but declines when fibrosis becomes
1.
Kannel WB, Gordon T, Offut D: Left ventricular mortal ity by electrocardiogram: prevalence, incidence and mor tality in the Framingham study. Ann Intern Med 1969;71:89-105.
2.
Levy D, Garrison RJ, Savage DD, et al: Left ventricular mass and incidence of coronary heart disease in an el derly cohort. The Framingham Heart Study. Ann Intern Med 1989;110:101-107.
3.
Casale PN, Devereux RB, Milner M, et al: Value of echo cardiographic measurement of left ventricular mass in predicting cardiovascular morbid events in hypertensive men. Ann Intern Med 1986;105:173-178.
4.
Kannel WB, Doyle JT, McNamara PM, et al: Precursors of sudden coronary death: factors related to the inci dence of sudden death. Circulation 1975;51:606-613.
5.
Messerli FH, Ventura HO, Elizardi DJ, et al: Hyperten sion and sudden death: increased ventricular ectopic ac tivity in left ventricular hypertrophy. Am J Med 1984;77:18-22.
6.
McLenachan JM, Henderson E, Morris Kl, Dargie HJ: Ventricular arrhythmias in patients with hypertensive left ventricular hypertrophy. Ν Engl J Med 1987; 317:787-792.
7.
Maron BJ, Savage DD, Wolfson JK, Epstein SE: Prognos tic significance of 24 hour ambulatory electrocardio graphic monitoring in patients with hypertrophic car diomyopathy: a prospective study. Am J Cardiol 1981;48:252-257.
8.
McKenna WJ, England D, Doi YL, et al: Arrhythmia in hypertrophic cardiomyopathy. I. Influence on progno sis. Br Heart J 1981;46:168-172.
9.
Sokolow M, Lyon TP: The ventricular complex in left ventricular hypertrophy as obtained by unipolar p r e cordial and limb leads. Am Heart J 1949;37:161-186.
10.
McLenachan JM, Henderson E, Morris Kl, Dargie HJ: Electrocardiographic diagnosis of left ventricular hyper-
20 21
22-26
22
23-25
26
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
ACKNOWLEDGMENTS
trophy: influence of body build. Clin Sci 1988,75:589592.
21.
Josephson ME: Inducible polymorphic ventricular tachycardia and ventricular fibrillation in a subgroup of patients with hypertrophic cardiomyopathy at high risk for sudden death. J Am Coll Cardiol 1987;10:761-766.
22.
Pfeffer MA, Pfeffer JM: Left ventricular hypertrophy and pressure generating capacity in ageing genetically hypertensive rats. J Cardiovasc Pharmacol 1985;7:S41S47.
23.
Hess OM, Schneider J, Krayenbuehl HL: Left ventricular endomyocardial biopsy in patients with primary and secondary myocardial hypertrophy, in Bolte HD (ed): Myocardial Biopsy. Diagnostic significance. BerlinHeidelberg-New York, Springer-Verlag, 1980, pp 55-64.
Devereux RB, Reichek N: Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation 1977;55:613-618.
12.
Greene DG, Carlisle R, Grant C, Bunnell IL: Estimation of left ventricular volume by one-plane cineangiography. Circulation 1967;35:61-69.
13.
Aherne WA, Dunhill MS: Point counting and the estimation of volume fraction, in Aherne WA, Dunhill MS (eds): Morphometry. London, E. Arnold, 1982, pp 3 3 45.
14.
Pichard AD, Gorlin R, Smith H, et al: Coronary flow studies in patients with left ventricular hypertrophy of the hypertensive type. Evidence for an impaired coronary vascular reserve. Am J Cardiol 1981;47:547-554.
24.
15.
Fallen EL, Elliott WC, Gorlin R: Mechanisms of angina in aortic stenosis. Circulation 1967;36:480-488.
Krayenbuehl HL, Hess OM, Schneider J, Turina M: Physiologic or pathologic hypertrophy? Eur Heart J 1984; 4(suppl A):29-35.
25.
16.
Klein RC: Ventricular arrhythmias in aortic valve disease: analysis of 102 patients. Am J Cardiol 1984;53:1079-1083.
Oldershaw PB, Brooksby IAB, Davies MJ, et al: Correlations of fibrosis in endomyocardial biopsies from patients with aortic valve disease. Br Heart J 1980;44:609611.
17.
Caspari PG, Newcomb M, Gibson K, Harris P: Collagen in the normal and hypertrophied human ventricle. Cardiovasc Res 1977;11:554-558.
26.
Hess OM, Schneider J, Kock R, et al: Diastolic function and myocardial structure in patients with myocardial hypertrophy. Circulation 1981;63:360-371.
18.
Pearlman ES, Weber KT, Janicki JS, et al: Muscle fibre orientation and connective tissue content in the hypertrophied human heart. Lab Invest 1982;46:158-164.
27.
19.
Cameron JS, Myerburg RJ, Wong SS, et al: Electrophysiological consequences of chronic experimentally induced left ventricular pressure overload. J Am Coll Cardiol 1983;2:481-487.
Baantrup U, Florio RA, Olsen EGJ: Do myocardial biopsies represent the morphology of the rest of the myocardium? A quantitative light microscopic study of single v. multiple biopsies with the King's bioptome. Eur Heart J 1982;3:171-178.
28.
20.
Anderson KP, Stinson EB, Derby GC, et al: Vulnerability of patients with obstructive hypertrophic cardiomyopathy to ventricular arrhythmia induction in the operating room: analysis of 17 patients. Am J Cardiol 1983; 51:811-816.
Tanaka M, Fujiwara H, Onodera T, et al: Quantitative analysis of myocardial fibrosis in normals, hypertensive hearts and hypertrophic cardiomyopathy. Br Heart J 1986;55:575-581.
29.
Cheitlin MD, Schneider M, McAllister H, et al: The distribution of fibrosis in the left ventricle in congenital aortic stenosis and coarctation of the aorta. Circulation 1980;62:823-830.
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
11.
1990;
3:735-740
ORIGINAL CONTRIBUTIONS
Ventricular Arrhythmias in Hypertensive Left Ventricular Hypertrophy James M. McLenachan and Henry /. Dargie
Ventricular arrhythmias occur with increased fre quency in hypertensive patients with left ventricu lar hypertrophy (LVH). The relationships, however, between ventricular arrhythmias and coexistent cor onary artery disease, left ventricular dysfunction and left ventricular fibrosis have not been examined in hypertensive LVH. We carried out coronary arte riography on fifteen hypertensive patients with LVH and nonsustained ventricular tachycardia (^ 3 consecutive ventricular complexes) of whom nine (60%) were free of significant ( > 50% stenosis) coro nary disease. To identify other possible correlates of left ventricular arrhythmias, 28 patients with LVH, comprising 17 with ventricular tachycardia and 11 without ventricular arrhythmias, underwent quantitative assessment of left ventricular function (angiographic ejection fraction), left ven-
B
oth electrocardiography and echocardiogra phy ' provide evidence that left ventricular hy pertrophy (LVH) is associated with an increased risk of cardiovascular mortality, including sud den death. Both simple and complex ventricular ar rhythmias have recently been shown to occur more commonly in hypertensive patients with LVH than in 1
2 3
4
From the Department of Cardiology, The Western Infirmary, Glas gow, Scotland. Dr. McLenachan was supported by a British Heart Foundation Ju nior Research Fellowship. Address correspondence and reprint requests to James M. McLena chan, Department of Cardiology, The Western Infirmary, Glasgow, Scotland G i l 6NT.
tricular mass (echocardiography), and left ventricu lar fibrosis (endomyocardial biopsy). Ejection frac tion was not significantly different between the two groups (53 ± 8% ν 62 ± 2%, Ρ = NS). However, left ventricular mass was significantly greater (442 ± 28 g ν 339 ± 34 g, Ρ < .05) and percentage fibrosis significantly higher (19 ± 4% ν 3 ± 1%, Ρ < .001) in those patients with ventricular tachycardia. Thus ventricular arrhythmias in hypertensive pa tients with LVH cannot be entirely attributed to coexistent coronary disease, nor to left ventricular dysfunction, but are related to the degree of cardiac hypertrophy and subendocardial fibrosis. Am J Hy pertens 1990; 3:735-740 Ventricular arrhythmia, left ventricu lar hypertrophy, coronary disease, fibrosis.
KEY WORDS:
those without. ' In other forms of cardiac hypertrophy, such arrhythmias are predictive of subsequent sudden death ' and, conceivably, could contribute to the high mortality among hypertensive patients with LVH. However, the relationships between ventricular ar rhythmias in hypertensive LVH and other variables that might influence arrhythmia frequency, such as coexistent coronary artery disease or left ventricular dysfunction, are not known. In addition, no studies have addressed the relationship between structural ab normalities associated with cardiac hypertrophy, in cluding subendocardial fibrosis, and the occurrence of ventricular arrhythmias. Thus the aims of this study were to determine whether ventricular arrhythmias in hypertensive pa tients with LVH are invariably associated with coronary 5 6
7 8
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
Relationship to Coronary Artery Disease, Left Ventricular Dysfunction, and Myocardial Fibrosis
artery disease and to examine the relationships between ventricular arrhythmia, left ventricular function and subendocardial fibrosis, as determined by quantitative analysis of left ventricular endomyocardial biopsies, in a group of patients with hypertension and LVH. STUDY GROUPS
9
2
2
10
1
Clearly, the appropriate control group for this com parison would be hypertensive patients with LVH but without ventricular tachycardia or chest pain; however, it was considered unethical to subject such patients to cardiac catheterization and thus the control population (group Β, η = 17) comprised a group of patients with LVH but without significant arryhthmias and included
METHODS Coronary Arteriography Four views were obtained of the left coronary artery and two of the right. The films were reviewed independently by two observers and classified according to whether or not they demon strated occlusion or stenosis, the latter defined as a re duction in luminal diameter by less than 50%, 50 to 70% or > 70% in one or more of the major coronary vessels. Left Ventricular Mass Mass was estimated in all pa tients undergoing left ventricular biopsy from an Mmode echocardiogram using the cube formula with "Penn" convention as described by Devereux and Rei chek. 11
Left Ventricular Function Left ventricular ejection fraction was also measured in all patients undergoing biopsy; this was estimated from a single-plane cineangiogram (30° right anterior oblique projection), as de scribed by Greene, following injection of 40 to 50 mL nonionic contrast medium. 12
Left Ventricular Biopsy A 7F "pigtail" catheter was used to position a long sheath (CORDIS) in the left ventricular cavity. Three to five biopsies were taken from the free wall of the left ventricle, the direction of the sheath being adjusted slightly before each biopsy. Samples were then stained for light microscopy and coded such that the source of the biopsy was not known at the time of analysis. Sections were stained with Masson's trichrome connective tissue stain, which stains fi brous and other connective tissue blue/green, and Elastin-van Gieson, which stains collagen red/brown. The sections were examined at a magnification of 400 X. For each stain in each patient, 80 fields were examined using an eyepiece graticule with 25 points randomly distrib uted; the eyepiece was rotated four times through 90 ° to give a total of 100 points per field. The number of "hits," ie, .areas where the dot on the graticule overlay an area of fibrous tissue was then expressed as a percent age of the total number of points counted. The error in this type of analysis is a function of both the total number of points counted and the "area frac tion of interest." Thus, if the area fraction of interest is large, the inherent error of the measurement, for a given number of points counted, will be small. We elected to count 8000 points (80 high power fields) per patient because this gave an acceptable relative standard error of approximately 3% when the area of interest (fibrosis etc.) was 10% rising to an error of 11 % when the area of fibrosis was only 1%. 13
Statistical Analysis Values for left ventricular mass, ejection fraction, percentage fibrosis etc, are all ex pressed as mean ± SEM and differences between
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
Fifteen hypertensive patients underwent standard diag nostic coronary arteriography. All patients had electro cardiographic (SV1 + RV5 ^ 35 mm) and echocardio graphic (left ventricular mass index ^ 145 g / m for men and > 110 g / m for women ) evidence of left ventricu lar hypertrophy and all had at least one episode of nonsustained ventricular tachycardia (^ 3 consecutive ven tricular complexes at a heart rate ^ 1 2 0 beats/min) during 48 h ambulatory monitoring. The purpose of cor onary arteriography in this group was to determine whether the high grade ventricular arrhythmias de tected during ambulatory monitoring were attributable to coexistent disease of the major coronary vessels. None had a history of angina pectoris or electrocardio graphic evidence of previous myocardial infarction. Twelve were men and the age range was 30 to 69 years (mean 55 ± 4 years). Left ventricular endomyocardial biopsy was at tempted in 37 patients but sufficient material for analy sis (three to five biopsy samples per patient) was ob tained in only 28. The two most common reasons for biopsy failure were inability to obtain a satisfactory po sition within the ventricle and arrhythmia induction by the biopsy forceps; such arrhythmias were invariably self-terminating on withdrawal of the bioptome. The 28 patients comprised two groups: group A (n = 11), in whom ambulatory electrocardiographic monitoring had demonstrated at least one episode of nonsustained ven tricular tachycardia (VT), and group Β (η = 1 7 ) , in whom no complex ventricular arrhythmias had oc curred during 48 h monitoring. All 28 had both electro cardiographic and echocardiographic evidence of left ventricular hypertrophy but they differed in the indica tion for cardiac catheterization. Thus patients in group A were derived from the group described previously, ie, hypertensive patients undergoing diagnostic cardiac catheterization to identify a cause for their ventricular arrhythmia. The number of episodes of VT varied from 1 to 22 with a mean of 3. The duration of the episodes varied from 3 to 25 beats with a mean of 5. None of the patients showed any symptoms at the time of the arrhythmia.
hypertensive patients with LVH and chest pain (n = 12) and patients with LVH secondary to aortic valve disease (n = 5).
groups were calculated using a two-tailed t test for un paired data. Differences in the prevalence of coronary artery disease were calculated using a χ test. 2
RESULTS Relation of Ventricular Arrhythmias to Coronary Ar tery Disease Of the 15 patients with left ventricular hypertrophy and ventricular tachycardia, nine (60%) were free of significant stenoses of the epicardial coro nary vessels. Of the remainder, two had one or more coronary occlusions, and four had one or more stenoses of greater than 70%; none had triple vessel disease.
TABLE 1. CHARACTERISTICS OF PATIENTS UNDERGOING ENDOMYOCARDIAL BIOPSY (n = 28)
Age (years) Men: women Left ventricular mass (g) Systolic Blood Pressure (mm Hg) Number of diseased vessels None One Two Three VT = ventricular
Patients with VT (n = 11) 58 ± 4 8:3 442 ± 28 166 ± 8
7 0 2 2 tachycardia.
Patients without VT (n = 17) 54 ± 2 13:4 339 ± 34* 168 ± 6
5 3 3 6 * Ρ < .05.
Without VT FIGURE 1. Ejection fraction values (mean ± SEM) for patients with and without ventricular tachycardia.
patients with ventricular tachycardia who also had marked left ventricular dysfunction (ejection fraction values < 30%) (Figure 1), the difference in percentage fibrosis between those patients with and without ven tricular tachycardia remained highly significant (10 ± 2% ν 3 ± 1 %, Ρ < .01). A similar pattern was seen when the Elastin-van Gieson samples were analyzed (13 ± 3% ν 2 ± 1%, Ρ < .001) (Figure 3); concordance be tween the two staining techniques was high (r = 0.835, standard error of estimate 6.49), suggesting that both were detecting the same abnormality. While most patients with advanced fibrosis also had marked hypertrophy, the reverse was not always true, and, overall, there was no significant correlation be tween left ventricular mass and fibrosis (r = 0.12, Ρ = NS). A weak inverse relationship did exist between left ventricular ejection fraction and fibrosis (Figure 4) al though many patients with detectable fibrosis had a normal ejection fraction. DISCUSSION Our results suggest that the increased frequency of ven tricular arrhythmias among hypertensive patients with LVH cannot be attributed solely to coexistent coronary artery disease or left ventricular dysfunction.
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
Relation of Ventricular Arrhythmias to Ventricular Hypertrophy, Ventricular Function and Myocardial Fibrosis There were no significant age or sex differ ences between the groups with and without ventricular tachycardia (Table 1). Although all patients had left ven tricular hypertrophy, mean left ventricular mass was higher among those with ventricular tachycardia than among those without (442 ± 28 ν 339 ± 34 g, Ρ < .05) (Table 1). Those patients with ventricular tachycardia demonstrated a wide range of ejection fraction values (Figure 1), although the mean values for the two groups were not significantly different (53 ± 8% ν 62 ± 2%, Ρ = NS). There was also a trend, although not statisti cally significant, towards increased coronary disease among those patients without ventricular tachycardia, reflecting the inclusion in group Β of hypertensive pa tients being investigated because of chest pain. The most striking difference between the two groups, however, related to quantitative analysis of the myocar dial biopsies; mean percentage fibrosis (Masson's stain) was very significantly higher in those patients with ven tricular tachycardia than in those without (19 ± 4% ν 3 ± 1%, Ρ < .001) (Figure 2). Even after excluding four
The relationships between hypertrophy, coronary disease and ventricular arrhythmia has not previously been studied in hypertensive patients, but have been investigated in aortic stenosis. In a series of 102 patients, Klein reported high grade ventricular arrhythmias in 40 (39%) but, as in the present study, found no relationship between arrhythmias and the presence of coronary artery disease. The second part of our study examined the relationship between endomyocardial fibrosis, left ventricular mass and left ventricular function in patients with and without ventricular arrhythmias. As a result of differences in entry criteria, coronary artery disease was more common among patients without ventricular arrhythmias. Despite this bias, patients with ventricular arrhythmias had significantly more fibrosis than those without and also tended to have higher left ventricular mass values. Increased collagen content has previously been documented in left ventricular hypertrophy ' but the relationship between arrhythmias and fibrosis has not been examined in humans. Arrhythmias, however, have been described in an animal model of left ventricular hypertrophy. Cameron et al induced cardiac hypertrophy by aortic banding and noted that ventricular arrhythmias, including ventricular extrasystoles, couplets 16
19
Without VT FIGURE 2. Percentage fibrous and interstitial tissue volume (Masson's stain) for patients with and without ventricular tachycardia.
The first part of our study demonstrated that of 15 hypertensive patients with LVH and ventricular arrhythmias during ambulatory monitoring, but without a history of chest pain, the majority were free of major stenoses of the epicardial coronary vessels. This does not imply, however, that ventricular arrhythmias in hypertensive patients with LVH are unrelated to myocardial ischemia; it is well established that coronary vascular reserve is reduced in hypertensive LVH and that patients with cardiac hypertrophy develop typical angina pectoris with electrocardiographic evidence of myocardial ischemia even when the major coronary vessels are patient. Furthermore, we have no data relating to the functional importance of minor coronary stenoses in patients with abnormally high myocardial demand. Nevertheless, it is apparent from our data that the high prevalence of ventricular arrhythmias in hypertensive patients with LVH cannot be attributed to coexistent disease of the large coronary arteries, as defined by stenoses of 70% or greater. 14
15
Without VT
FIGURE 3. Percentage fibrous tissue volume (Elastin-van Gieson stain) for patients with and without ventricular tachycardia.
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
17 18
100
more extensive (Figure 4). However, caution is neces sary when interpreting our data, particularly when correlating changes in global left ventricular function with structural changes seen on subendocardial biopsy, in view of the marked regional and transmural ' variations in myocardial fibrosis. In summary, our findings suggest that ventricular ar rhythmias in hypertensive patients with left ventricular hypertrophy cannot be entirely attributed to coexistent coronary artery disease or left ventricular dysfunction but are associated with structural changes in the hyper trophied ventricle including the development of myo cardial fibrosis. 27
28 29
The authors gratefully acknowledge Dr. A. F. Lever of the Medical Research Council Blood Pressure Unit, Western Infir mary, Glasgow, for helpful advice during analysis of the data and Dr. G. B. Lindop, Senior Lecturer in Pathology at the Western Infirmary, Glasgow, for advice regarding the quanti tative analysis of myocardial biopsies. REFERENCES
% FIBROSIS FIGURE 4. Relationship between left ventricular ejection frac tion and percentage fibrosis for all 28 patients. Patients with ven tricular tachycardia are depicted by closed circles and patients without ventricular tachycardia by open circles.
and ventricular fibrillation, occurred more commonly than in control animals. Furthermore, in vitro testing showed that a variety of electrophysiological abnormali ties, including short action potential duration, low am plitude action potentials and electrically silent areas, oc curred at the border zone between hypertrophied myocardium and areas of marked connective tissue in filtration. These electrical abnormalities were not seen in endocardial preparations without endocardial fibro sis. In humans, several studies have shown that ventric ular arrhythmias can be induced by programmed elec trical stimulation in patients with hypertrophic cardiomyopathy ' in whom myofiber disarray and myocardial fibrosis are common. The relationship between fibrosis and cardiac func tion in various forms of left ventricular hypertrophy has been studied by other g r o u p s . In experimental hy pertension, systolic function declines with age, coin ciding with the development of myocardial fibrosis. Studies in humans suggest that systolic function may be preserved in the presence of quite marked fibrosis although diastolic function may be impaired at an ear lier stage. Our data suggest that systolic function may be pre served even when moderate amounts of fibrosis (5 to 15%) are present, but declines when fibrosis becomes
1.
Kannel WB, Gordon T, Offut D: Left ventricular mortal ity by electrocardiogram: prevalence, incidence and mor tality in the Framingham study. Ann Intern Med 1969;71:89-105.
2.
Levy D, Garrison RJ, Savage DD, et al: Left ventricular mass and incidence of coronary heart disease in an el derly cohort. The Framingham Heart Study. Ann Intern Med 1989;110:101-107.
3.
Casale PN, Devereux RB, Milner M, et al: Value of echo cardiographic measurement of left ventricular mass in predicting cardiovascular morbid events in hypertensive men. Ann Intern Med 1986;105:173-178.
4.
Kannel WB, Doyle JT, McNamara PM, et al: Precursors of sudden coronary death: factors related to the inci dence of sudden death. Circulation 1975;51:606-613.
5.
Messerli FH, Ventura HO, Elizardi DJ, et al: Hyperten sion and sudden death: increased ventricular ectopic ac tivity in left ventricular hypertrophy. Am J Med 1984;77:18-22.
6.
McLenachan JM, Henderson E, Morris Kl, Dargie HJ: Ventricular arrhythmias in patients with hypertensive left ventricular hypertrophy. Ν Engl J Med 1987; 317:787-792.
7.
Maron BJ, Savage DD, Wolfson JK, Epstein SE: Prognos tic significance of 24 hour ambulatory electrocardio graphic monitoring in patients with hypertrophic car diomyopathy: a prospective study. Am J Cardiol 1981;48:252-257.
8.
McKenna WJ, England D, Doi YL, et al: Arrhythmia in hypertrophic cardiomyopathy. I. Influence on progno sis. Br Heart J 1981;46:168-172.
9.
Sokolow M, Lyon TP: The ventricular complex in left ventricular hypertrophy as obtained by unipolar p r e cordial and limb leads. Am Heart J 1949;37:161-186.
10.
McLenachan JM, Henderson E, Morris Kl, Dargie HJ: Electrocardiographic diagnosis of left ventricular hyper-
20 21
22-26
22
23-25
26
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
ACKNOWLEDGMENTS
trophy: influence of body build. Clin Sci 1988,75:589592.
21.
Josephson ME: Inducible polymorphic ventricular tachycardia and ventricular fibrillation in a subgroup of patients with hypertrophic cardiomyopathy at high risk for sudden death. J Am Coll Cardiol 1987;10:761-766.
22.
Pfeffer MA, Pfeffer JM: Left ventricular hypertrophy and pressure generating capacity in ageing genetically hypertensive rats. J Cardiovasc Pharmacol 1985;7:S41S47.
23.
Hess OM, Schneider J, Krayenbuehl HL: Left ventricular endomyocardial biopsy in patients with primary and secondary myocardial hypertrophy, in Bolte HD (ed): Myocardial Biopsy. Diagnostic significance. BerlinHeidelberg-New York, Springer-Verlag, 1980, pp 55-64.
Devereux RB, Reichek N: Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation 1977;55:613-618.
12.
Greene DG, Carlisle R, Grant C, Bunnell IL: Estimation of left ventricular volume by one-plane cineangiography. Circulation 1967;35:61-69.
13.
Aherne WA, Dunhill MS: Point counting and the estimation of volume fraction, in Aherne WA, Dunhill MS (eds): Morphometry. London, E. Arnold, 1982, pp 3 3 45.
14.
Pichard AD, Gorlin R, Smith H, et al: Coronary flow studies in patients with left ventricular hypertrophy of the hypertensive type. Evidence for an impaired coronary vascular reserve. Am J Cardiol 1981;47:547-554.
24.
15.
Fallen EL, Elliott WC, Gorlin R: Mechanisms of angina in aortic stenosis. Circulation 1967;36:480-488.
Krayenbuehl HL, Hess OM, Schneider J, Turina M: Physiologic or pathologic hypertrophy? Eur Heart J 1984; 4(suppl A):29-35.
25.
16.
Klein RC: Ventricular arrhythmias in aortic valve disease: analysis of 102 patients. Am J Cardiol 1984;53:1079-1083.
Oldershaw PB, Brooksby IAB, Davies MJ, et al: Correlations of fibrosis in endomyocardial biopsies from patients with aortic valve disease. Br Heart J 1980;44:609611.
17.
Caspari PG, Newcomb M, Gibson K, Harris P: Collagen in the normal and hypertrophied human ventricle. Cardiovasc Res 1977;11:554-558.
26.
Hess OM, Schneider J, Kock R, et al: Diastolic function and myocardial structure in patients with myocardial hypertrophy. Circulation 1981;63:360-371.
18.
Pearlman ES, Weber KT, Janicki JS, et al: Muscle fibre orientation and connective tissue content in the hypertrophied human heart. Lab Invest 1982;46:158-164.
27.
19.
Cameron JS, Myerburg RJ, Wong SS, et al: Electrophysiological consequences of chronic experimentally induced left ventricular pressure overload. J Am Coll Cardiol 1983;2:481-487.
Baantrup U, Florio RA, Olsen EGJ: Do myocardial biopsies represent the morphology of the rest of the myocardium? A quantitative light microscopic study of single v. multiple biopsies with the King's bioptome. Eur Heart J 1982;3:171-178.
28.
20.
Anderson KP, Stinson EB, Derby GC, et al: Vulnerability of patients with obstructive hypertrophic cardiomyopathy to ventricular arrhythmia induction in the operating room: analysis of 17 patients. Am J Cardiol 1983; 51:811-816.
Tanaka M, Fujiwara H, Onodera T, et al: Quantitative analysis of myocardial fibrosis in normals, hypertensive hearts and hypertrophic cardiomyopathy. Br Heart J 1986;55:575-581.
29.
Cheitlin MD, Schneider M, McAllister H, et al: The distribution of fibrosis in the left ventricle in congenital aortic stenosis and coarctation of the aorta. Circulation 1980;62:823-830.
Downloaded from http://ajh.oxfordjournals.org/ at University of Otago on October 15, 2015
11.
