Regression of left ventricular mass is accompanied by improvement in rapid left ventricular filling following antihypertensive therapy with metoprolol

Regression of left ventricular mass is accompanied by improvement in rapid left ventricular filling following antihypertensive therapy with metoprolol Left ventricular hypertrophy is associated with abnormal left ventricular diastolic filling in patients with hypertension. To assess the effects of antlhypertensive therapy on the heart in nine previously untreated patients with echocardiographically-dotacted left ventricular hypertrophy, left ventricular mass and rapid left ventricular filling rate were compared before and after 6 months of treatment wlth metoprolol monotherapy. Metoprolol was given In doses of 100 to 400 mg/day (average dose, 167 mg/day in two divided doses) and significantly reduced both casual, offtce blood pressure (150/101 to X30/66 mm Hg, p < 0.01) and 24.hour ambulatory blood pressure (13g/gl to 126179 mm Hg, p < 0.05 for systolic, p < 0.01 for diastolic). Following treatment with metoprolol, left ventricular mass index decreased from 135 +- 20 to 120 * 13 gm/m2 (p < O.O!i), while rapid left ventricular filling rate Increased from 1.89 + 0.24 to 2.09 * 0.27 end-diastolic volumes/set @ < 0.01). The reduction In left ventricular mass index was secondary to decreased posterior and septal wall thicknesses (13% and 1 l%, respectively, p < 0.05 for both), as there were no changes in the left ventricular Internal dlmensions. Neither resting nor exercise left ventricular ejection fraction changed on metoprolol therapy compared to the baseline values. These data demonstrate that regression of left ventricular hypertrophy in never-previously-treated hypertensive patlents la accompanied by improved diastolic performance following 8-adrenergic blocker monotherapy. (Am HEART J 1989;117:145.)

William B. White, MD, Peter Schulman, MD, Moxaferredin and Vivienne-Elizabeth Smith, MD. Farmington, Corm.

Left ventricular hypertrophy detected by echocardiography is an independent cardiovascular risk factor in patients with hypertension.’ Abnormalities in left ventricular compliance and diastolic filling have been described in left ventricular hypertrophy before systolic function is impaired?-’ While abnormal diastolic performance in patients with hypertrophy is generally associated with wall thickness and the degree of interstitial fibrosis, other physiologic factors such a3 age, heart rate, left ventricular filling pressure, and adrenergic activity influence left ventricular filling.5 Regression of left ventricular mass in hypertenFrom the Section of Hypertension Medicine and Nuclear Medicine, Medicine. This work was supported Pharmaceutical Corporation, Received

for publication

Reprint requests: Vascular Diseases, CT 06032.

William IJniversity

and Vascular University

Diseases, Departments of Connecticut School

in part by a grant-in-aid Summit, N.J. Apr.

29, 1988,

accepted

B. White, MD, of Connecticut

from Aug.

Section Health

the

of of

Ciba-Geigy

15, 1968.

of Hypertension and Center, Farmington,

K. Karimeddini,

MD,

sive patients occurs following treatment with most classes of antihypertensive drugs excluding direct vasodilatorss and diuretics.’ Less information is known about the effects of reduced left ventricular mass on cardiac function. In an earlier study,8 treatment with the dihydropyridine calcium channel blocker, nitrendipine, alone or in combination with a @-blocking drug or diuretic, caused reduction in left ventricular mass index that was accompanied by improved rapid left ventricular filling. The patients in that study had all been receiving previous antihypertensive drug therapy and just one third of them had a reduction in left ventricular mass. In the study herein, we analyzed the effects of the fi-adrenergic blocking agent, metoprolol, in patients with mildto-moderate hypertension who had not been previously treated with antihypertensive therapy and who had left ventricular hypertrophy as determined by echocardiographic criteria.1*s The purpose of this study was to ascertain the effects of antihypertensive therapy on left ventricular hypertrophy in patients prior to any potential alterations in left 145

146

White

et al.

Table I. Effects of metoprolol monotherapy on casual and ambulatory blood pressure and heart rates Baseline period

Casual readings Systolic BP (mg Hg) Diastolic BP (mm Hg) Heart rate (beats/min) 24-Hour average readings Systolic BP (mm Hg) Diastolic BP (mm Hg) Heart rate (beats/min) Awake readings Systolic BP (mm Hg) Diastolic BP (mm Hg) Heart rate (beats/min) Sleep readings Systolic BP (mm Hg) Diastolic BP (mm Hg) Heart rate (beats/min)

mass

or filling

Following metoprolol

150 +- 12 74 * 13

130 * 14t 86 2 lot 66 + 14*

13s f. 9 91 * 2 81 t- 16

126 + 8* IS + 4t 70 Ti 13*

14s * 10 f 3 86 * 14

130 + 10* 82 I+ 4t 74 I 11*

122 t 4 80 +I 2 15 * 8

115 i 9 69 i 6* 68 i 6

101 & 6

Table Il. Summary of the effects of metoprolol dynamics, wall thickness, and left ventricular _-

98

deviation.

by other

forms

of treat-

METHODS Patient selection. Sixteen patients with mild to moderate essential hypertension (supine diastolic blood pressure between 90 and 115 mm Hg in the untreated state) were recruited from the Hypertension Unit of the University of Connecticut. Patients were included only if they had not been previously receiving antihypertensive drug therapy and if they met echocardiographically-defined criteria for left ventricular hypertrophy (in men, 2 134 gm/m2 and in women, L 110 gm/m2). None of the patients had a history of angina pectoris, myocardial infarction, or valvular heart disease, and none had evidence of coronary artery disease detected by electrocardiography, exercise treadmill testing, or radionuclide exercise ventriculography. None of the patients met electrocardiographic (ECG) criteria for left ventricular hypertrophy. All patients gave informed consent and this study was approved by the University of Connecticut Institutional Review Board. Blood pressure determlnations and drug titration. Paired, supine casual (office) blood pressure determinations were made prior to placebo on two occasions at least 1 week apart to assess baseline blood pressure levels. At that time, a screening echocardiogram was performed (as described below) and if results were acceptable, the patient was assigned to metoprolol placebo for 4 weeks. At the third and fourth week, if the patient’s supine diastolic blood pressure was < 90 mm Hg, > 115 mm Hg, or varied by more than 10% between the two visits, the patient was excluded from the study. The baseline measurements were taken during the fourth week of placebo therapy.

Baseline period

Parameter

Systolic BP (mm Hg) Diastolic BP (mm Hg) Heart rate (beats/min) Left atria1 size (cm) IVS thickness

Data are expressed as the mean + standard BP, blood pressure. *p < 0.05 versus baseline placebo period. tp < 0.01 versus baseline placebo period.

ventricular ment.

January 1989 Heart Journal

American

(mm)

PWT thickness (mm) LVIDD (cm) LVMI (pm/m’) LVEF, resting (% ) LVEF, exercise (% ) LVFRav (EDV/sec)

147 95 80 4.2 12

2 i + + It

15 12 14 0.5 2 10 c 1 5.2 zk 0.3

135 +- 20 61 i 4 69 t 5

1.89 * 0.24

on hemofunction ____-~ -Following metoprolol

146 95 75 4.2 10.5 9 5.3 120 62 71 2.09

t 17 It 9 i- 12 +- 0.6 + 1.0* ” 1* + 0.3 t 13* t 3 It- 3 I!Z 0.27t

Note: Blood pressures and heart rates following metoprolol therapy are those obtained after discontinuation of drug therapy for 3 to 5 days (at the time of radionuclide studies). BP, blood pressure; EDV, end-diastolic volumes; IVS, intraventricular septal; PWT, posterior wall; LVIDD, left ventricular internal dimension at diastole; LVMI, left ventricular mass index; LVEF, left ventricular ejection fraction; LVFRav, average left ventricular filling rate. ‘p < 0.05 versus baseline. tp < 0.01 versus baseline.

The goal of antihypertensive therapy was a reduction in supine diastolic blood pressure to less than 90 mm Hg if the baseline blood pressure was greater than 100 mm Hg, or a reduction of at least 10 mm Hg if the baseline diastolic blood pressure was between 90 and 100 mm Hg. All patients were started on treatment with metoprolol at a dose of 50 mg twice daily, with titration every 2 weeks to a maximum daily dose of 400 mg to achieve the blood pressure goal. Patients who did not achieve this goal after 2 months were dropped from the study. Heart rate (ECG) and blood pressure (mercury column sphygmomanometer) were measured when the patients were supine for at least 5 minutes. The values reported as the treatment response represent the average data from two visits 2 weeks apart following 6 months of therapy. Ambulatory blood pressure determinations. Ambulatory blood pressure and heart rate determinations were made on a weekday with a portable, noninvasive recorder (Spacelabs, Hillsboro, Ore.). Readings were obtained automatically at 15-minute intervals during the entire 24-hour period. All studies were performed with the display monitor off, to prevent anticipation of the readings by the patients. Readings obtained with the ambulatory blood pressure recorder and a standard mercury column sphygmomanometer connected via a T-tube assembly were correlated as previously describedlo, I’; all readings were highly significant for the beginning and end of the study period. The monitor-derived systolic and diastolic blood pressures averaged 1 2 2 mm Hg 0, = ns) and 4 & 2 mm Hg lower (p < 0.05), respectively, than the mercury column determinations. Readings were deleted if the pulse pressure was less than 12 mm Hg, or showed an inconsistent decrease or

Volume Number

117 1

LV diastolic

increasein systolic or diastolic blood pressuregreater than 30 mm Hg from previous or subsequentreadings.Individual studies were included only if at least 75% of the maximal number of 96 readingsduring the 24-hour period passedthe deletion criteria. Journals of activity, symptoms, and medication times were carefully kept by the subjects for aid in the editing process. Echocardiagraphy and left ventricular mass determinations. A screening M-mode echocardiogram was

obtained at the first week of study entry. These were evaluated by a single observer (VES) for acceptable quality for serial determinations and the presenceof left ventricular hypertrophy. Subsequently, another M-mode echocardiogramwas obtained at the end of the placebo period, and again after 6 months of antihypertensive therapy with metoprolol. Echocardiograms from the end of the placebo run-in period and following metoprolol therapy were coded by a third party not involved with the study, and were read by a single observer (PS) who was unaware of the patient

data. Baseline and treatment studies were obtained during held expiration in the same body and transducer positions. Left atria1 dimension was measuredwith the

leading edge method according to the standards of the American Society of Echocardiography.’ Calculations of left ventricular masswere made from the left ventricular

function

after hypertension

lowing standard injection

therapy

147

of stannous polyphosphate.

Acquisition of data was ECG gated and wasperformed in the l@ft anterior oblique projection for four million counts at rest and for two million counts with peak exercise. The resting data were reframed as previously describedI for determination of left ventricular filling rate, with a framing rate of 20 msecper frame. Left ventricular region of interest selection was performed manually by a single experienced observer (MKK). Both the left ventricular

filling rate and ejection fraction were computer-calculated by time-activity curve analysis,as previously described.3r8 The maximal percent variability of rapid left ventricular filling rate for our laboratory is 4% (0.042 end-diastolic volumes/sec).s Supine bicycle exercise ventriculography was performed in all subjects as previously described.14In all cases, exercise was terminated becauseof exhaustion and not

because of angina pectoris or dyspnea. Left ventricular ejection fraction was calculated by dividing the stroke counts by end-diastolic counts with subtraction of the background activity. Statistical analysis. The effect of metoprolol therapy

on casual and ambulatory blood pressures and heart rates, left ventricular massindex, ejection fractions, and filling rate for the entire group was evaluated by means of a two-tailed paired t test.

dimensions obtained with the Penn measurement conventions according to the method of Devereux and Reichek.13 The relevant dimensions of three cardiac cycles were averaged to form the basisof the masscalculation. The left ventricular mass was indexed by the body surface area. Intraobserver variation in masscalculations were determined in the following fashion. Echocardiograms were read from three separate,unrelated studies (including the present one) over a period of 3 months. Ten echocardiograms were read twice by the observer (PS) after an interval of 2 to 3 months when his recollection was blunted. Variability in measurement of left ventricular

mass was defined as the difference between the first and secondreading of a given echocardiogmrn,divided by the massobtained in the first reading. The percent variation of the 10 “sample” echocardiograms was averaged and the standard deviation wascalculated. Thus a value of percent variability (mean plus two standard deviations of the mean) for the reading of two identical echocardiograms

was obtained. Radionuclide tricular filling.

studies

and determination

of left ven-

Radionuclide ventriculography was performed with a standard large field of view gammacamera (SiemensBasicam,Erlangen, West Germany) and a dedicated nuclear medicine computer (PDP 11-34, Digital Equipment Corp., Marlboro, Mass.). The studies were

performed at the end of the placebo period and within 3 to 5 days of the t:reatment echocardiogram.To avoid immediate hemodynamic effects of the drug, patients had their doseof metoprolol tapered down for 3 to 5 days before the radionuclide studies. Each patient’s red blood cells were labelled with technetium-99m pertechnetate in vivo fol-

RESULTS Patient demographics. Twelve patients met the criteria for entry into the treatment period. However, two patients did not achieve the goal blood pressure and were not included in this analysis. A third patient moved out of state in the middle of the trial and 6-month treatment data were not obtained. The mean age of the nine patients completing the study was 42 f 9 years (range, 25 to 60 years) and there were five men and four women. Blood pressure and heart rate. The effects of antihypertensive therapy with metoprolol on blood pressure and heart rate are shown in Table I. All patients achieved the goal of therapy within 2 months of initiating drug therapy, at a mean dose of 167 -t 78 mg/day. Casual (office) blood pressure was reduced 20/15 mm Hg (p < 0.01) with metoprolol compared to the end of the baseline placebo period, and heart rate was also significantly reduced by 8 beats/min (p < 0.05). Both the mean 24-hour ambulatory systolic and diastolic blood pressures were significantly reduced by metoprolol (from 139/91 f 9/2 mm Hg at baseline to 126/79 +- 814 mm Hg during metoprolol; p < 0.01 for both) (Table I). Metoprolol induced the most marked reduction in blood pressure while patients were awake (19/16 mm Hg, p < 0.05 for systolic and p < 0.01 for diastolic). A statistically

significant

(p < 0.05) reduction

in sleep diastolic

148

White

et al.

2.4 -

2.2 150# ‘a *^ g -1

c 140-

B ”

2.0. s

130-

pm0.04

8

a

120-

P E G

2 E g

110-

f

f

loa-

P 0: 3

i:

901

Placebo

.

p.o.004

l.B-

. 1.6.

Moiopmlol

1. Individual effects of metoprolol therapy on left ventricular massindex (left panel) and average, rapid left ventricular filling rate (right panel).

Fig.

blood pressure occurred with metoprolol (11 mm Hg), whereas the drug induced a lesser reduction in sleep systolic blood pressure (8 mm Hg, p = 0.06). The 24-hour ambulatory heart rate was significantly reduced with metoprolol (11 beats/min, p < 0.05). Again, the reduction in heart rate with metoprolol was more marked while the patients were awake (12 beats/min, p < 0.05) than during sleep (7 beats/min, p = 0.09). Left ventricular wall thickness, chamber dimensions, and mass. The mean of the difference in left ventric-

ular mass index between the first and second readings of the same echocardiogram expressed as a percent of the baseline value for 10 echocardiograms was 3 f 2%. Thus the variability due to the observer was considered to be a maximum of 7% (mean plus two standard deviations of the mean). The left atria1 size and left ventricular internal dimension did not change after metoprolol therapy (Table II). Posterior wall and septal thicknesses were significantly reduced following 6 months of metoprolol therapy (12.0 to 10.5 mm, p < 0.05; and 10.0 to 9.0 mm, p < 0.05, respectively), as shown in Table II. The left ventricular mass index was reduced by 11% following metoprolol therapy (p < 0.05) (Table II). Five of the nine subjects had regression of mass greater than the maximal reader variability of 7% (Fig. 1, left panel). Left

ventricular

filling

rate

and

ejectlon

fraction.

There were no differences in the resting supine blood pressure or heart rate in the nuclear medicine laboratory at the baseline and treatment periods

(metoprolol th erapy had been withdrawn prior to the study). After metoprolol, the rapid left ventricular filling rate improved from 1.89 f 0.24 to 2.09 + 0.27 end-diastolic volumes/set (EDV/S) (Table II). As shown in Fig. 1 (right panel), filling rate improved (> 5 % ) in seven of the nine patients and did not change in the other two. The left ventricular ejection fraction at rest and following peak exercise did not change after metopro101 therapy (Fig. 2). The changes in left ventricular filling rate were not associated with changes in ejection fraction. DISCUSSION

The data in this study indicate that initial antihypertensive therapy with metoprolol causes regression of left ventricular mass, improvement in left ventricular filling, and unaltered left ventricular ejection fraction. Impaired diastolic performance is characteristic of the ventricle that becomes hypertrophic in response to a chronic pressure overload,2s3* l6 with greater loads resulting in delayed relaxation.16*” Thus the finding of improved diastolic filling occurring simultaneously with regression of left ventricular mass implies a reversal of pathophysiologic processes in the hypertensive left ventricle. Our findings are in previously untreated hypertensive patients with left ventricular hypertrophy determined by echocardiographic criteria. This makes the data unique, since most earlier studies evaluating changes i-n left ventricular mass during antihypertensive therapy were performed in either

Volume

117

Number

1

LV diastolic function after hypertension 60

Left venttfcuter #actbn fmctkm, msi

Len venttfcutsr

eject&n

fmctmn,

therapy

149

fmek em98

p=ns p=ns

60 Placebo

Metoprolol

Plecebo

Msioprolol

Fig. 2. Individual effects of metoprolol therapy on left ventricular ejection fraction at rest (left panel) and after supine bicycle exercise (right panel).

patients who had normal left ventricular mass at baseline or who had been recently treated with antihypertensive drugs. In the last several years, a growing number of clinical reports have described regression of left ventricular hypertrophy in response to antihypertensive therapy. However, related changes in left ventricular function are less well described, and improvement in diastolic function after antihypertensive drug therapy has not been consistently demonstrated. In agreement with our findings, Fouad et al;18 found a significant increase in left ventricular filling rate in patients whose blood pressure was reduced with p-blocker therapy and a decrease in filling in patients who had no blood pressure response to the drug. In contrast, Inouye et aLI9 were unable to show a consistent improvement in diastolic filling in hypertensive patients after therapy with a combined therapy of a diuretic, fl-adrenergic blocker, and ‘calcium channel blocker. One explanation of the discrepancy in these studies is that our patients and those of Fouad et al. were previously untreated and perhaps more likely to demonstrate regression in left ventricular mass with initial therapy. In the study of Inouye et al., patients had been receiving previous therapy, and partial regression of mass and/or changes in rapid filling could have occurred, accounting for a lesser response during the ,period of study. Another possibility that may account for the heterogeneity of responses of the left ventricle is that the office blood pressures reported in most studies do not adequately represent the average

daily blood pressure that determines the cardiac response in patients with hypertension.4* lo*l1 In the present study, 24-hour blood pressure and heart rate with metoprolol were significantly decreased, both while patients were awake and asleep (Table I). Antihypertensive therapy that causes a persistent reduction in average daily blood pressure without stimulation of the sympathetic nervous system may be most efficacious for inducing a regression of left ventricular hypertrophy. Our findings of significantly diminished posterior and septal wall thicknesses and left ventricular mass after monotherapy with @-adrenergic blockade confirm those of previous reports.17*20,21 However, we defined “responders” by different criteria than most other studies because we obtained the echocardiogram observer variability and applied it to the definition of a therapeutic effect. This approach is necessary, since reader variability can approach 5 % to 10% a yet a “statistically” significant change is possible at values of < 3 % .22By the use of the value of a decrease in mass of at least 7 % as an indication of a therapeutic response, about 56 % of the patients had a regression of left ventricular mass with metopro101 (Fig. 1, left panel). There are a number of clinical implications that can be made from the findings of our study. First, the data herein imply that regression of left ventricular hypertrophy in hypertensive patients is associated with a significant improvement in the diastolic function of the heart. Previously, there has been speculation that regression of hypertrophy would increase the fibrosis/myocardial muscle fiber ratio,

150

White

et al

leaving the ventricle less compliant and perhaps with worsened function.23 The data in this study as well as that in our prior work,B and that of other laboratories24 do not support this concept, since regression of left ventricular mass was accompanied by an improvement in left ventricular filling in all the reports. Finally, systolic function was unaltered following metoprolol therapy (Fig. 2). This finding is not surprising, since all patients had normal pretreatment left ventricular ejection fractions and normal left ventricular end-diastolic dimensions. Thus it would be unlikely that any improvement or deterioration in systolic function could occur. It is known that coronary vascular reserve can be significantly reduced in hypertrophic hearts, and this is related to the ratio of coronary perfusion pressure to left ventricular mass.25 Thus a potential outcome of reducing blood pressure without a concomitant reduction of left ventricular mass is a reduction in coronary blood flow. Reversal or regression of left ventricular mass during antihypertensive therapy may decrease functional myocardial ischemia and may be important in the reduction in coronary mortality and morbidity recently observed in patients treated with P-adrenergic blocking agents.26 The authors thank Ellen J. McCabe, RN, for excellent assistance as study coordinator, and Jyll Munson for technical assistance in the echocardiography laboratory. REFERENCES

1. Casale PN, Devereux RB, Milner M, et al. Value of echocardiographic measurement of left ventricular mass in predicting cardiovascular morbid events in hypertensive men. Ann I&rn Med 1986;105:173-8. 2. Fouad FM. Slominski JM. Tarazi RC. Left ventricular diastolic function in hypertension: relation to left ventricular mass and systolic function. J Am Co11 Cardiol 1984;4: 1500-6. 3. Smith VE, Schulman P, Karimeddini MK, White WB, Meeran M, Katz AM. Rapid ventricular filling in left ventricular hypertrophg. II. Pathologic hypertrophy. J Am Co11 Cardiol-1985;5:869-74. 4. White WB, Schulman P, McCabe EJ, Dey HM. Average daily blood pressure not office blood pressure determines left ventricular function in patients with hypertension. JAMA 1989 (In press) 5. Hess OM. Schneider J. Koch R. Bamert C. Grimm J, Krayenbuehl HP. Diastolic function and myocardial structure in patients with myocardial hypertrophy. Circulation 1981;63:360-71. 6. Leenen FH, Smith DL, Farkas RM, Reeves RA, MarquezJulio A. Vasodilator regression of left ventricular hypertrophy: hydralazine versus prazosin in hypertensive humans. Am J Med 1987;82:969-78. I. Drayer JIM, Gardin JM, Weber MA, Aronow WS. Changes in ventricular septal thickness during diuretic therapy. Clin Pharmacol Ther 1982;32:283-8.

American

January 1999 Heart Journal

8. Smith VE, White WB, Meeran MK, Karimeddini MK. Improved left ventricular filling accompanies reduced left ventricular mass during therapy of essential hypertension. J Am Co11 Cardiol 1986;8:1449-54. 9. Devereux RB, Lutas EM, Casale PN, et al. Standardization of M-mode echocardiographic left ventricular anatomic measurements. J Am Co11 Cardiol 1984;4:1222-30. 10. White WB, Smith VE, McCabe EJ, Meeran MK. Effects of chronic nitrendipine on casual (office) and 24-hour ambulatory blood pressure. Clin Pharmacol Ther 1985;38:60-4. 11. White WB. Assessment of patients with office hypertension by 24.hour noninvasive ambulatory blood pressure monitoring. Arch Intern Med 1986;146:2196-9. 12. Sahn DJ, Demaria A, Kisslo J, Weyman A. The Committee on M-mode Standardization of the American Society of Echocardiography. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 19’78;58:1072-83. 13. Devereux RB. Reichek N. Echocardiograuhv in detection of left ventricular mass in man. Circulat&-1977;55:613-8. 14. Bonow RO, Bacharach SL, Green MN, et. al. Impaired left ventricular diastolic filling in patients with coronary artery disease: assessment with radionuclide angiography. Circulation 1981;64:315-23. 15. Inouye I, Massie BM, Loje D, et al. Abnormal left ventricular filling: an early finding in mild-to-moderate systemic hypertension. Am J Cardiol 1984;53:120-6. 16. Brutsaert DL, Rademakers FE, Sys SU, et al. Analysis of relaxation in the evaluation of ventricular function of the heart. Prog Cardiovasc Dis 1985;28:143-63. 17. Trimarco B, DeLuca N, Cuocolo A, et al. Beta blockers and left ventricular hypertrophy in hypertension. AM HEART J 1987;114:975-83. 18. Fouad FM, Slominiski MJ, Tarazi RC, Gallagher JH. Alterations in left ventricular filling with beta-adrenergic blockade. Am J Cardiol 1983;1:161-4. 19. Inouve IK. Massie BM. Loie D. Simnson P. Tuhau JF. Failure of’ antihypertensive - therapy with diuretic, betablocking and calcium channel blocking drugs to consistently reverse left ventricular diastolic filling abnormalities. Am J Cardiol 1984;53:1583-7. 20. Dunn FG, Ventura HO, Messerli FH, Kobrin I, Frohlich ED. Time course of regression of left ventricular hypertrophy in hypertensive patients treated with atenolol. Circulation 1987;76:254-8. 21. Corea L, Bentivoglio M, Verdecchia P, Providenza M, Motolese M. Left ver&ular hypertrophy regression in patients treated with metonrolol. Int J Cardiol 1984:22:365-70. 22. Drayer JIM, Hall*WD, Smith VE, Weber MA, Wollam GL, White WB. Effect of the calcium channel blocker, nitrendipine, on left ventricular mass in patients with hypertension. Clin Pharmacol Ther 1986;40:679-85. 23. Agati L, Fedele F, Pence M, Sciomer S, Dagianti A. Left ventricular filling pattern in hypertensive patients after reversal of myocardial hypertrophy. Int J Cardiol 1987; 17:177-86. 24. DeKock M, Melin JA, Nannan ME, et al. Alteration of left ventricular diastolic filling in hypertensive patients: effects of nitrendipine and atenolol. Eur Heart J 198%7:792-g. 25. Wicker P, Tarazi RC, Kobayashi K. Coronary blood flow during the development and regression of left ventricular hypertrophy in renovascular hypertensive rats. Am J Cardiol 1983;51:1744-9. 26. Wikstrand J, Warnold I, Olsson G, et al. Primary prevention with metoprolol in patients with hypertension: mortality results from the MAPHY study. JAMA 1988;259:1976-82.