Hypertensive heart disease: Relationship of silent ischemia to coronary artery disease and left ventricular hypertrophy

Hypertensive heart disease: Relationship of silent ischemia to coronary artery disease and left ventricular hypertrophy ECG evidence of silent ischemia occurs commonly in patients with systemic hypertension, but its relationship to left ventricular hypertrophy (LVH), large-vessel coronary artery disease (CAD), and neurohumoral factors remains unclear. Accordingly we validated the results of the echocardiographic method used to measure left ventricular (LV) mass in the Soviet Union by comparison with necropsy measurements in 30 patients, and we examined the relationships in 46 men with essential hypertension among ST segment depression during ambulatory monitoring, exercise stress and transesophageal pacing (n = 36), and LV mass, catheterization evidence of CAD (n = 25), and neurohumoral factors (plasma catecholamines and platelet aggregability). Echocardiographic measurements of LV mass by both the Soviet and Penn methods were closely correlated with necropsy values (r = 0.76 and 90, respectively; both p < 0.001). During ambulatory monitoring from 1 to 17 episodes of 21 mm ST depression occurred in 26 of 46 (65%) patients with hypertension; ischemia was also provoked by exercise or pacing stress in most but not all of these patients (65% and 80%, respectively). Neither ST depression nor the occurrence of additional episodes of symptomatic angina was related to the presence of coronary obstruction at catheterization; patients with and without ST depression did not differ in age, blood pressure, or LV mass. No systematic differences between patients with and without CAD or ST depression were noted in catecholamine levels or platelet aggregability, but the subset in whom ischemia developed at a lower heart rate (X) blood pressure product during pacing stress than during exercise stress had significantly lower norepinephrine levels at rest and during pacing stress (88 t- 34 versus 157 f 47 rig/L, p < 0.02; and 86 + 39 versus 254 + 86 rig/L, p < 0.01). Thus asymptomatic ST depression occurs commonly during normal activity and induced stress in patients with systemic hypertension but is not consistently related to the presence of large-vessel CAD, LVH, or neurohumoral abnormalities. (AM HEART J 1990; 120:928.)

Alexei P. Yurenev, MD, Dot Med Sci, Vincent DeQuattro, MD, and Richard B. Devereux, MD. Moscow, USSR, Los Angeles, Calif., and New York, N.Y.

It is now widely recognized that the heart is the major target organ for complications of arterial hypertension. A major question therefore is what makes the heart so vulnerable to these complications, most of which are coronary in nature. One line of evidence, based on epidemiologic data,l suggests that hyperFrom the USSR Cardiology Research Center, Academy of the USSR, the Department of Medicine, Los Angeles of Southern California Medical Center, and The New nell Medical Center. Supported in part by grants HL Heart, Lung, and Blood Institute, Received

for publication

March

31090 and Bethesda, 23, 1990;

4ll22a38

928

HL 18323 Md.

accepted

Reprint requests: Richard B. Devereux, MD, nell Medical Center, 525 East 68th St., Box

of Medical Sciences County-University York Hospital-Cor-

May

from

the National

21, 1990.

The New York Hospital-Cor222, New York, NY 10021.

tension may contribute directly to the development of coronary atherosclerosis. However, data from other research suggest that elevated blood pressure may be only a weak stimulus to coronary atherosclerosis2 but may increase both the likelihood and severity of morbid events being produced by coronary artery disease (CAD) of any given severity.3q 4 Results of several studies performed in the United States have supported the latter interpretation by showing that left ventricular hypertrophy (LVH) is an important risk factor, possibly even more important than blood pressure for morbid events in patients with hypertension.5-7 Results of a recently completed multicenter trial conducted in the USSR, German Democratic Republic, and Czechoslovakia, in which 297 patients treated

Volume Number

120 4

Silent ischemia

in hypertension

929

EDb

Fig.

1. M-mode echocardiogramof left ventricle showingmeasurementsof interventricular septum (IVS),

Ir wall (PW, thickness asroutinely made in USSR.

for hypertension8 were followed for 5 years after baseline echocardiography, supported this interpretation by showing that coronary disease events (angina, myocardial infarction, and sudden death) were predicted both by higher echocardiographic left ventricular (LV) muscle mass at the beginning of the study and by further increases on subsequent annual echocardiograms. However, data from studies in the United States and the USSR have been difficult to compare because of differences in echocardiographic methodology. Furthermore, few data are available relating evidence of myocardial ischemia in patients with hypertension to both LV muscle mass and the presence or absence of CAD. The present studies were undertaken to address both of these deficiencies. METHODS Subjects.

Comparison of echocardiographic methods and assessmentof correlates of myocardial ischemic responseswas performed in separatepopulations. The first consistedof 30 patients (18 womenand 12 men, aged 25 to 82 years) who underwent technically satisfactory echocardiography at The New York Hospital-Cornell Medical Center within 6 months of death and autopsy, without intervening myocardial infarction or major changein hemodynamic status. The secondpatient population consisted of 46 men, aged 35 to 62 years, with arterial hypertension

and a history of either angina or atypical chest pain, who were studied at the USSR Cardiology ResearchCenter. Echocardiographic methods. Echocardiography was performed in New York and Moscow with commercially available echocardiographs by methods previously described from the respective centers.g-ll M-mode tracings recorded on strip-chart paper at 50 mm/set were coded, and blinded measurementsof LV dimensionsweremadeon up to six cardiac cycles according to the Penn convention lo, l2 and by the recommendationsof the American Society of Echocardiography (ASE),13asroutinely implemented in the USSR (Fig. l), by investigators skilled in each method (R. B. D. and A. P. Y., respectively). Penn convention measurementswere usedto calculate LV mass by a regression-correctedcube formula, and ASE measurements were usedto calculate LV massby the volume-corrected formula of Teichholz et al.‘*; results of both methods have been previously validated in separate necropsy comparisonstudies.g,11-12 Necropsy methods. Hearts were studied after formalin fixation; measuredweights wereadjusted to correct for the mean decreasein weight of 5% which we observed after formalin fixation in 38 hearts. No adjustment wasmadefor duration of fixation, becauseno changein heart weight was observedin hearts weighedserially for up to 270 days. The LV myocardium was dissectedfree of the right ventricle and separatedfrom the aorta, mitral valve, and left atrium as described by Bove et a1r5; gross epicardial fat was removedand the resulting coneof LV myocardium weighed.

930

Yurenev et al.

October 1990 Heart Journal

American

A

B Specificity 18121 = 86% sensitivity 9/9 = 100% overall accuracy 27/30 = 90%

Specificity 18/21 = 86% sensitivity 619 = 67% overall accuracy 24/30 = 80%

600

1

/ -Lii 100

200

300

400

500

Necropsy left ventricular mass (g)

1 100

200

300

I

I

400

500

Necropsy left ventricular mass (g)

2. A, Relationship between necropsy left ventricular mass(horizontal axis) and echocardiographic left ventricular massby Penn method (vertical axis). B, Relationship between necropsy left ventricular mass (horizontal axis) and echocardiographic left ventricular mass according to measurementsas illustrated in Fig. 1 in volume-correction formula (vertical axis).

Fig.

Specimensof LV myocardium removed for standard or special studies at the time of autopsy were weighed, and this value was added to that obtained after chamber dissection. Evaluation of myocardial ischemia. Ambulatory monitoring of ECG leadsVs and V5 wasperformed for 24 hours in all subjects in the secondpopulation. All patients also underwent ECG monitoring of responsesto physiologic stressinduced either by transesophagealpacing (beginning at 100beats/min and increasingby incrementsof 15 beats/ min until symptoms or atrioventricular heart block was produced, N = 38), by bicycle ergometry (beginning at 25 W and increasingby increments of 25 W until 75% of the predicted maximal heart rate wasreachedor until the subject wasunable to continue becauseof intolerable fatigue, dyspnea, or pair@; N = 5), or by treadmill exerciseby the Bruce protocol17and with the sameend points (N = 41). In eachtest myocardial ischemiawasconsideredto be present when horizontal or downsloping ST segmentdepression ~1 mm belowthe baselinewasobserved80 msecafter the J point. Coronary arteriography wasperformed for clinical indications in 25 of these 46 patients. Laboratory methods. Becauseof the possiblerole of platelet aggregation in provoking ischemicevents in patients with hypertension, this wasmeasuredin a subsetof 21 patients with hypertension in the present study and in sevennormal adults who served ascontrol subjects. Mea-

surement of platelet aggregation was made by a laser aggregometernewly developed in the USSR Cardiology ResearchCenter, which is 10 times more sensitive than conventional equipment.l8 To assessthe relationship between catecholaminesand myocardial ischemia,plasmafor determination of epinephrine and norepinephrine wasobtained at rest and at the peak of treadmill exercisein 13 of the 25 patients who underwent coronary arteriography. Statistical methods. Data are presented as the mean with one standard deviation as the index of dispersion. Differences between groups in continuous variables were assessed by unpaired Student’s t -tests and in dichotomous variables by the chi-squaretest with the Yates continuity correction when appropriate. Least-squareslinear correlation was used to test the strength of the relationship between variables. RESULTS Relationship between echocardiographic and necropsy findings. LV mass ranged from 90 to 485 gm

(mean = 186 k 79 gm) at necropsy, from 78 to 569 gm (mean = 211 f 99 gm, p = NS versus necropsy) by the Penn method, and from 79 to 357 gm (mean = 193 -t 67 gm, p = NS versus necropsy) by the Teichholz method. Close correlation existed between necropsy LV mass and echocardiographic

Volume Number

120 4

measurements by the Penn method (r = 0.09, SEE = 43 gm, p < 0.001) (Fig 2, A) or the Teichholz method (r = 0.78, (r = 0.78, SEE = 42 gm,p < 0.001) (Fig. 2, B). A highly significant correlation was also observed between the two echocardiographic methods (r = 0.84, p < O.OOl), yielding a regression equation for interconversion of measurements by these techniques: LV mass(p,,,) = 1.25 LV mass(Teichh,,lz) - 30.6 gm. By this formula the partition value of 200 gm used to recognize prognostically disadvantageous levels of LV mass in studies in the USSR8 would correspond to 219 gm by the Penn method or to 122 gm/m2 for an adult with a median body surface area of 1.8 m2, values closely resembling the 215 gm and 125 gm/m2 used in previous anatomic or prognosis-based studies.5* 7l I21I59lg Prevalence and correlates of silent ischemia. Of the 46 patients with hypertension studied by ambulatory ECG, 30 (65 % ) had one or more episodes of ST segment depression (mean = 4.23, range 1 to 17), whereas 16 (35%) had none. Of the episodes of ST segment depression, 19 episodes in six patients were accompanied by angina; 14 patients with clinical angina had episodes of silent ST depression, and 16 patients without angina had from one to nine episodes of silent ST segment depression. Characteristics of patients with and without ST segment depression during Holter monitoring are given in Table I. Among patients with Holter ST segment depression who underwent treadmill testing, 17 of 26 (65%) had ST depression during the latter test; 20 of 25 (80%) who underwent transesophageal pacing had ST depression during the latter test. Coronary arteriography was performed in 21 of 30 who had ST changes on Holter monitoring and in 4 of 16 patients without Holter ischemia. CAD (more than 50% reduction in coronary artery diameter) was seen in 11 patients; 14 had no significant CAD (11 had angiographically normal coronary arteries, and three had nonobstructive luminal irregularities), Patients with hypertension with and without CAD were comparable in age, duration of hypertension, blood pressure level, resting ECG changes, and echocardiographic LV mass. Typical angina was reported more often by patients with CAD (8 of 11) than without CAD (8 of 14), but this difference was not significant. The proportion of patients who not only had silent ST changes but also had ST changes accompanied by pain was similar in groups with and without CAD (8 of 11 versus 11 of 14 and 3 of 11 versus 3 of 14; both p = NS).

Analysis of results of 24-hour Holter monitoring in 12 patients undergoing coronary angiography who had silent ST depressions (80 episodes) showed that

Silent &hernia in hypertension

93 1

Table 1.Characteristics of patients with hypertension with

and without ST segment depression detected by ambulatory echocardiography Present (n = 30)

Characteristics

AgeW Body surfacearea (m) Systolic blood pressure (mm Hd Diastolic blood pressure (mm Hd LV mass index (gm/m2)

p

Absent (n = 16)

49 + 1.0 >0.05 1.92 -I- 0.06 >0.05 168 -e 4.7 >0.05

51 t 1.7 1.90 + 0.05 165 ? 3.1

105 a 2.5

>0.05

110 A 2.8

106.4 ir 4.2

>0.05

97.2 2 5.5

these events were much more common and frequent during normal daily activities: only four (5 % ) occurred between 12 AM and 6 AM (Fig. 3). It is important to note that the heart rate during these events was substantially lower (mean = 112 f. 3.3 beats/ min) than the heart rate during ischemia provoked by transesophageal pacing (mean = 145 -I- 5.1 beats/ min) or treadmill exercise (143 & 4.9 beatslmin). Most ST depressions during Holter monitoring occurred at heart rates of 100 to 120 beats/min but some occurred st heart rates below 70 beats/min. Relationship catecholamine

between ST segment depression levels. Basal levels of epinephrine

and

and norepinephrine were comparable in the five patients with hypertension with obstructive CAD and the eight with angiographically normal coronary arteries at baseline (19 + 7 versus 25 & 24 rig/L and 125 ? 64 versus 150 5 89 rig/L, respectively), as well as during pacing (28 -+ 13 versus 35 * 18 and 185 f 95 versus 199 + 116 rig/L) and during treadmill tests (280 + 475 versus 81 rt 49 and 1027 + 810versus 959 ? 482 rig/L). Among the patients who had ischemic events during both pacing and treadmill tests, we observed two groups. Group 1 included patients with heart rate x blood pressure product (HR*BP) at provocation of ischemia by treadmill comparable to HR*BP at the provocation of ischemia by pacing. These patients required the same hemodynamic load for production of ischemia. In group 2, HR*BP (treadmill) exceeded HR*BP (pacing) by 20% or more. Thus in these patients ischemia was produced during pacing with a lower hemodynamic burden than during exercise. There were four patients (one with and three without CAD) in group 1, and five patients (two with and three without CAD) in group 2. These groups were comparable in age (51 + 7 versus 51 + 3 years) and in baseline (20 & 15 versus 30 & 28 rig/L), pacing (29 + 17 versus 30 + 17 rig/L), and treadmill (70 + 31 versus 88 + 60 rig/L) levels of epinehprine. There were no differences in heart rate

932

Yureneu et al.

American

October 1990 Heart Journal

(75.6 -t 83.7 versus 60.0 -f 85.3 RU) and was not significantly higher than in control subjects (15.7 -+ 17.9 RU).

SILENT ST DEPRESSION "WATCH"

DISCUSSION

It is well established that in the era of treatment of hypertension the majority of complications of arterial hypertension

80 EPISODES

24 IN 19

i23

HYPERTENSIVES

12 13

Fig. 3. Schematic diagram relating prevalence of epi-

sodesof asymptomatic ST segmentdepressiondetected by ambulatory ECG to time of day.

and HR*BP between groups 1 and 2 during treadmill test testing (146 +- 21 versus 148 -t 9 beats/min and 31,130 +- 4200 versus 37,600 f 1720), but during pacing the heart rate was significantly lower in group 2 (160 +- 0 versus 123 + 12 beats/min, p < 0.02), and the ratio of peak HR*BP (treadmill) to peak HR*BP (pacing) was higher (107 t 8 versus 149 t- 17%, p < 0.01). LV mass was insignificantly higher in group 2 (193 +- 58 versus 242 + 51gm), with LVH being diagnosed in all five in group 2 but in only one of four in group 1. Norepinephrine levels were significantly lower in group 2 both under basal conditions (157 +- 47 versus 68 f 34 rig/L, p < 0.02) and during pacing 254 + 86 versus 86 + 39 rig/L; p < 0.01). During peak treadmill exercise it was not significantly lower (930 +- 43 versus 699 f 329 rig/L, p > 0.05). Relationship between ST segment depression and platelet aggregability. In the 21 patients with hyper-

tension in which it was studied, platelet aggregation induced by adenosine diphosphate (*10W7) had a tendency to be higher (40.0 -t 54.8 rat units) than in seven healthy volunteers (15.7 -+ 17.9 RU, 0.05 > p > 0.1). There was no difference in platelet aggregation between 10 patients with CAD and 11 without CAD. Platelet aggregation was the same

are of coronary

origin (angina,

acute myocardial infarction, sudden death). Thus the so-called “coronary factor” plays an important role in hypertensive heart disease. In a recent report asymptomatic ST depression, which was considered to represent silent ischemia, was commonly detected by ambulatory ECG among patients with hypertension .20 Ischemic events in patients with hypertension may be caused by several abnormalities, including obstruction of epicardial coronary arteries, increased microvascular resistance,20t 21 vasospasm, or increased myocardial demand as a result of LVH.22, 23 Further phasic modulation of myocardial oxygen demand or coronary resistance may be induced by fluctuation of circulating factors such as catecholamines24 or by platelet aggregation. In the present study we found no consistent relationship between myocardial ischemic episodes, as reflected by ECG depression during Holter monitoring, transesophageal pacing, or treadmill exercise, and the presence or absence of obstructive CAD, LVH, or indices of hemodynamic load. However, neither our results nor similar findings obtained by other investigators (reviewed recently by Prisant et a1.25) permit us to reject these events as “false positive” in the group without obstructive coronary atherosclerosis, since they may be “true positive” manifestations of disturbed coronary perfusion in both patients with and without large-vessel CAD. Because at present neither clinical symptoms nor noninvasive ECG or radionuclide tests permit accurate separation between patients with hypertension with and without obstructive CAD, invasive coronary arteriography remains justified when patients with hypertension have either typical angina or prominently abnormal physiologic test findings. The lower plasma catecholamine levels found in our patients who had ischemia at a lower HR*BP product during pacing than during treadmill exercise and the strong trend toward higher LV mass in this group are compatible with our previous data showing that higher LV mass correlates with greater density of beta-adrenoreceptors and more sensitive Ca++ channel responsiveness of alpha adrenoceptors.26 These findings could explain equivalent ischemic responses despite lower norepinephrine levels in our group 2 at baseline and during pacing. In conclusion, the results of the present study, in

Volume Number

120 4

accordance with previous observations, indicate that macrovascular coronary atherosclerosis is far from being the sole factor determining impairment of coronary perfusion of the hypertensive heart. Although increased LV muscle mass, readily detectable by echocardiography, is one of the most important risk factors for complications of hypertension, ECG ischemia in our hypertensive patients with or without angiographically demonstrable CAD could not be attributed solely to LVH. Additional possible contributors to myocardial ischemia may include increased platelet aggregation, disregulated sympathetic neural tone, and microvascular CAD. Further research is needed to elucidate the relative importance of myocardial hypertrophy, neurohumoral factors, and coronary microvascular abnormalities in hypertensive heart disease. REFERENCES

1. Kannel WB, Sorlie P. Hypertension in Framingham. In: Paul 0, ed. Epidemiology and control of hypertension. New York: Strolton International, 1975:553-90. 2. Doyle AE. Does hypertension predispose to coronary disease? Conflicting epidemiological and experimental evidence. Am J Hypertension 1988;1:319-24. 3. Koyonagi S, Eastham CI, Harrison DC, Marcus ML. Increased size of myocardial infarction in dogs with chronic hypertension and left ventricular hypertrophy. Circ Res 1982;59:55-62. 4. Koyonagi S, Eastham CL, Marcus ML. Effects of chronic hypertension and left ventricular hypertrophy on the incidence of sudden cardiac death after coronary artery occlusion in conscious dogs. Circulation 1982;65:1192-7. 5. Casale PN, Devereaux RB, Milner M, Zullo G, Harshfield GA, Pickering TG, Laragh JH. Value of echocardiographic measurement of left ventricular mass in predicting cardiovascular morbid events in hypertensive men. Ann Intern Med 1986; 105:173-8. 6. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Left ventricular mass and incidence of coronary heart disease in an elderly cohort: the Framingham study. Ann Intern Med 1989;110:101-8. Koren MJ, Casale PN, Savage DD, Laragh JH, Devereux RB. Relation of left ventricular mass to prognosis in essential hypertension. Circulation 1989;8O(suppl II):II-538. Shkhvatsabaya I, Yurenev AP, Balazs A, Pahl L, Niederle P. First year follow-up results: international multicenter survey on randomized therapy in arterial hypertension. [Abstract] International Conference on Preventive Cardiology. Moscow, June 23-26, 1985. Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol 1986;57:450-8.

Silent &hernia in hypertension 933 10. Devereux RB, Liebson PR, Horan MJ. Recommendations concerning the use of echocardiography in hypertension and general population research. Hypertension 1987;9(suppl2):97104. 11. Belenkov YN. Vyiavlenie rannikh priznakov serdechnoi nedostatochnosti i nekotorykh mekhanizmov ee kompensatsii pri ischemicheskoi bolezni serdtsa. Thesis. Moskva, 1975. 12. Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man. Anatomic validation of the method. Circulation 1977;55:613-8. 13. Sahn DJ, DeMaria A, Kisslo J, Weyman A. The Committee on M-mode Standardization of the American Society of Echocardiography: recommendations regarding quantitation in Mmode echocardiography: results of a survey of echocardiographic methods. Circulation 1978;58:1072-83. 14. Teichholz LE, Kreulen T, Herman MV, Gorline R. Problems in echocardiographic value determinations: echocardiographic-angiographic correlations in the presence or absence of asynergy. Am J Cardiol 1976;37:7-11. 15. Bove KE, Rowland DT, Scott RC. Observations on the assessment of cardiac hypertrophy using a chamber partition technique. Circulation 1966;33:558-68. 16. Aronov DM, Yurenev AP. Usage of ECG test with physical training in cardiology practice. Methodological recommendations. Moscow: USSR Ministry of Health, 197919. 17. Bruce RA, Hornsten TR. Exercise stress testing in evaluation of patients with ischemic heart disease. Prog Cardiovasc Dis 1969;11:371-90. 18. Abbasov LA, Popov EG, Gavrilow IJ, Pozin EG. Platelet aggregation: the use of optical density fluctuations to study microaggregate formation in platelet suspension. Thromb Res 1989;54:215-23. 19. Reichek N, Devereux RB. Left ventricular hypertrophy: relationships of anatomic, echocardiographic and electrocardiographic findings. Circulation 1981;63:1391-8. 20. Motz W, Scheler S, Vogt M, Strauer BE. Transient myocardial ischemia in hypertensive coronary micro-angiopathy [Abstract]. J Am Co11 Cardiol 1989;13:84A. 21. Brush Jr JE, Cannon III RO, Schenke WH, Bonow RO, Leon MB, Maron BJ, Epstein SE. Angina due to coronary microvascular disease in hypertensive patients without left ventricular hypertrophy. N Engl J Med 1989;319:1302-7. 22. Hoffman JIE. Maximal coronary flow and the concept of coronary vascular reserve. Circulation 1984;70:153-9. 23. Carr AA, Prisant LM, Houghton JL. Ischemic heart disease in hypertensives: hypertensive or obstructive epicardial coronary artery disease [Abstract]? Am JHypertension 1989;2:71A. 24. Lee DD-P, Kimura S, DeQuattro V. Noradrenergic activity and silent ischaemia in hypertensive patients with stable angina: effect of metroprolol. Lancet 1989;1:403-6. 25. Prisant LM, Frank MJ, Carr AA, von Dohlen TW, Abdulla AM. How can we diagnose coronary heart disease in hypertensive patients? Hypertension 1987;10:467-72. 26. Yurenev AP, Trachuk VA, Mazau AV, Dubov PB, Dyakonova EG, Patrusheva IF, Gerashchenko YS, Nesterova AZ, Menshikov MY, Atakhanov GE, Tkhostova EB. The hypertensive heart: pathogenesis, variants and prognostic value. Health Psycho1 1988;7(suppl):105-11.