Effect of metoprolol on rest and exercise left ventricular systolic and diastolic function in idiopathic dilated cardiomyopathy☆☆☆

Patients 

Eighteen patients with IDCM were studied. All were in sinus rhythm and had reduced left ventricular ejection fraction, without clinically significant coronary atherosclerosis as demonstrated by coronary angiography. Informed consent to participate in the study protocol was obtained from each patient. The study was approved by the Institutional Review Board of the Mayo Clinic.

The patients were randomized in a double-blind manner to either metoprolol (nine patients) or placebo (nine patients) for 3 months. Metoprolol or corresponding placebo was given in gradually increasing doses as tolerated to a maximum dose of 50 mg orally two times a day. Two patients (one in the metoprolol and one in the placebo group) had deterioration in cardiac status and were referred for cardiac transplantation before completion of the study. The study was not completed in a further three patients—in one patient in the placebo group who declined to continue in the study because of time constraints and in two patients in the metoprolol group for whom the radionuclide data on the second study either were not obtained or were inadequate for analysis. Only patients who completed the study were analyzed. At the end of the study period, of the six patients who remained in the metoprolol group, five were taking 100 mg and one 50 mg of metoprolol per day.

Radionuclide ventriculography 

Resting and exercise radionuclide left ventriculographic-gated blood-pool acquisitions were obtained as previously documented.8 Exercise was performed with a supine bicycle exercise at a workload of 200 kg · m/min for 6 minutes. The initial rest and exercise studies were obtained with the patients receiving no treatment, and the second rest and exercise studies were obtained after 3 months on treatment. The first minute of the exercise level permitted stabilization of the heart rate and blood pressure, and the radionuclide image was acquired during the last 5 minutes of exercise. A 5-minute acquisition was chosen to give sufficient count statistics for analysis of the diastolic measures.

Left ventricular ejection fraction and diastolic filling and cardiac volumes were measured at rest and during exercise from the left ventricular time-activity curve by methods previously validated.9 The diastolic measures were left ventricular peak filling rate (stroke volumes per second), time to peak filling rate (milliseconds), and proportion of stroke volume that filled in the first half of the filling period (ie, the first half filling fraction). Left ventricular volumes were measured by an attenuation-uncorrected, count-based method as previously validated in this laboratory10 and normalized to body surface area. Heart rate and systolic and diastolic blood pressures were recorded at rest and during exercise.

Statistical analysis 

Comparisons between group means were made by use of paired and unpaired t tests, with P ≤ .05 regarded as significant. The relationship between left ventricular ejection fraction, peak filling rate, time to peak filling rate, first half filling fraction, or time to peak filling rate and heart rate was analyzed by simple linear regression.

Before treatment with placebo or metoprolol 

There were no differences in heart rate, systolic and diastolic blood pressure, left ventricular ejection fraction, cardiac volumes, and the three diastolic measures (peak filling rate, time to peak filling rate, and first half filling fraction), at rest and with exercise, between the placebo and the metoprolol group before the initiation of treatment (Tables I to III).

Table III. Diastolic measures in placebo and metoprolol groups (mean ± SD and median values are given)

		Pretreatment			Posttreatment
		Placebo	Metoprolol	Combined*	Placebo	Metoprolol†
	Rest
	Peak filling rate (stroke volume/s)	7.64 ± 2.49	6.16 ± 2.94	6.95 ± 2.73	8.55 ± 2.81	6.04 ± 3.01
		8.45	5.80	7.16	9.52	4.85
	Time to peak filling rate (ms)	161 ± 105	203 ± 70	182 ± 89	178 ± 29	200 ± 67
		180	231	185	175	172
	First half filling fraction	0.56 ± 0.27	0.45 ± 0.18	0.51 ± 0.23	0.28 ± 0.26	0.60 ± 0.30 (2)
		0.64	0.48	0.57	0.19	0.77
	Exercise
	Peak filling rate (stroke volume/s)	8.40 ± 2.48	7.81 ± 3.15	8.12 ± 2.74	8.40 ± 1.85	5.41 ± 1.71 (3)
		8.90	6.60	8.33	7.96	5.02
	Time to peak filling rate (ms)	147 ± 60	165 ± 67	156 ± 62	127 ± 17	164 ± 21 (4)
		162	162	162	132	165
	First half filling fraction	0.29 ± 0.18	0.41 ± 0.26	0.35 ± 0.23 (1)	0.40 ± 0.30	0.48 ± 0.14
		0.22	0.48	0.39	0.34	0.50
	*Pretreatment, combined, rest versus exercise: (1) P = .02. †Posttreatment, metoprolol versus placebo: (2) P = .06, (3) P = .012, and (4) P = .005.

Table I. Hemodynamic data in groups before and after treatment with placebo or metoprolol (mean ± SD and median values are given)

		Pretreatment			Posttreatment
		Placebo	Metoprolol	Combined*	Placebo	Metoprolol†
	Rest
	Heart rate (beats/min)	90 ± 22	83 ± 18	87 ± 20	99 ± 10	61 ± 11 (3)
		97	77	83	104	60
	Systolic blood pressure (mm Hg)	117 ± 16	122 ± 19	119 ± 18	122 ± 12	116 ± 23
		110	112	111	120	115
	Diastolic blood pressure (mm Hg)	76 ± 11	75 ± 9	76 ± 10	78 ± 7	72 ± 11
		76	76	76	80	72
	Exercise
	Heart rate (beats/min)	118 ± 30	114 ± 27	116 ± 28 (1)	126 ± 43	86 ± 18 (4)
		117	102	108	118	80
	Systolic blood pressure (mm Hg)	144 ± 15	142 ± 24	143 ± 20 (1)	151 ± 9	148 ± 28 (5)
		140	132	140	150	144
	Diastolic blood pressure (mm Hg)	86 ± 8	80 ± 10	83 ± 9 (2)	89 ± 8	82 ± 15 (6)
		86	80	80	90	75
	*Pretreatment, combined, rest versus exercise: (1) P < .001 and (2) P = .0008. †Posttreatment, metoprolol versus placebo: (3) P < .0001 and (4) P = .056. Posttreatment, rest versus exercise: (5) P = .0002 and (6) P = .02.

Table II. Left ventricular ejection fraction and cardiac volumes in placebo and metoprolol groups (mean ± SD and median values are given)

		Pretreatment			Posttreatment
		Placebo	Metoprolol	Combined*	Placebo	Metoprolol†
	Rest
	LV ejection fraction	0.21 ± 0.06	0.22 ± 0.09	0.21 ± 0.07	0.17 ± 0.08	0.32 ± 0.10 (2)
		0.19	0.22	0.21	0.15	0.33
	LV end-diastolic volume index (mL/m2)	214 ± 90	207 ± 141	211 ± 113	201 ± 104	190 ± 93
		215	184	207	205	182
	LV end-systolic volume index (mL/m2)	163 ± 72	157 ± 124	160 ± 97	160 ± 94	130 ± 74
		164	130	160	177	129
	LV stroke volume index (mL/m2)	51 ± 24	50 ± 25	50 ± 24	41 ± 14	59 ± 25
		45	42	43	37	61
	Exercise
	LV ejection fraction	0.18 ± 0.09	0.20 ± 0.12	0.19 ± 0.10	0.19 ± 0.07	0.32 ± 0.14 (3)
		0.17	0.18	0.18	0.16	0.31
	LV end-diastolic volume index (mL/m2)	220 ± 94	206 ± 136	213 ± 112	204 ± 85	200 ± 108
		234	204	216	214	190
	LV end-systolic volume index (mL/m2)	171 ± 76	165 ± 132	168 ± 102 (1)	159 ± 77	144 ± 97
		187	139	184	173	133
	LV stroke volume index (mL/m2)	49 ± 28	41 ± 23	45 ± 25	44 ± 12	56 ± 21
		49	40	41	48	62
	*Pretreatment, combined, rest versus exercise: (1) P = .02. †Posttreatment, metoprolol versus placebo: (2) P = .01 and (3) P = .052.

LV, Left ventricular.

After treatment with placebo or metoprolol 

Compared with placebo, metoprolol treatment slowed heart rate at rest (61 ± 11 vs 99 ± 10 beats/min, P < .0001) and with exercise (86 ± 18 vs 126 ± 43 beats/min, P = .056). After metoprolol treatment compared with placebo, left ventricular ejection fraction increased at rest (0.32% ± 0.10% vs 0.17% ± 0.08%, P = .01) and with exercise (0.32% ± 0.14% vs 0.19% ± 0.07%, P = .05). Although cardiac volumes (Table II) did not significantly differ at rest and with exercise between the placebo group and the metoprolol-treated group, there was a tendency for stroke volume index to be greater after metoprolol compared with placebo treatment both at rest (59 ± 25 vs 41 ± 14 mL/m2) and with exercise (56 ± 21 vs 44 ± 12 mL/m2).

After β-blocker treatment, there was no difference at rest between metoprolol and placebo treatment in peak filling rate, time to peak filling rate, and first half filling fraction. With exercise, after metoprolol compared with placebo, the peak filling rate decreased (5.41 ± 1.71 vs 8.40 ± 1.85 stroke volumes/s, P = .012) and the time to peak filling rate was longer (164 ± 21 vs 127 ± 17 ms, P = .005). First half filling fraction was similar after exercise before and after metoprolol and placebo.

Effect of exercise and metoprolol 

When exercise was compared with rest in the metoprolol-treated patients, significant increases in systolic and diastolic blood pressure developed, but there were no significant changes in heart rate, left ventricular ejection fraction, cardiac volumes, or diastolic measures.

Influence of exercise and metoprolol on the interactions of heart rate and systolic and diastolic function 

Before randomization in all 18 patients, heart rate and left ventricular ejection fraction (Table IV and Figure 1) were negatively correlated at rest (r = –0.69, P = .002) and with exercise (r = –0.64, P = .004) and heart rate and peak filling rate (Table IV and Figure 2) were positively correlated at rest (r = 0.82, P < .0001) and with exercise (r = 0.70, P = .002).

View Large Image Download to PowerPoint

Fig. 1. Plots of left ventricular (LV) ejection fraction (y axis) against heart rate (beats/min, x axis) at rest (A) and during exercise (B) . Solid circles, All 18 patients before treatment; open circles, the 6 patients who received metoprolol. Lines of regression are shown (solid line, the 18 patients before treatment; dashed line, the 6 patients who received metoprolol).

View Large Image Download to PowerPoint

Fig. 2. Plots of peak filling rate (stroke volumes/s, y axis) against heart rate (beats/min, x axis) at rest (A) and during exercise (B) . Solid circles, All 18 patients before treatment; open circles, the 6 patients who received metoprolol. Lines of regression are shown (solid line, the 18 patients before treatment; dashed line, the 6 patients who received metoprolol).

Table V. Correlations of heart rate and systolic and diastolic function at rest and with exercise after treatment with metoprolol

	Status	Left ventricular ejection fraction	Peak filling rate	Time to peak filling rate	First half filling fraction
	Rest	–0.68	0.82	–0.40	–0.63
	Statistical significance	NS	P = .05	NS	NS
	Exercise	–0.52	0.98	–0.006	–0.76
	Statistical significance	NS	P = .0007	NS	NS

NS, Not significant.

Table IV. Correlations of heart rate and systolic and diastolic function at rest and with exercise before treatment

	Status	Left ventricular ejection fraction	Peak filling rate	Time to peak filling rate	First half filling fraction
	Rest	–0.69	0.82	–0.56	–0.35
	Statistical significance	P = .002	P < .0001	P = .01	NS
	Exercise	–0.64	0.70	–0.11	–0.19
	Statistical significance	P = .004	P = .002	NS	NS

NS, Not significant.

Heart rate and cardiac function 

Patients with reduced left ventricular dysfunction demonstrate elevated serum catecholamine levels11 and an increased resting heart rate.12 Both these responses can have negative effects on cardiac function in the presence of reduced left ventricular function. Increased catecholamines may cause cardiac remodeling13 because of apoptotic myocardial cellular death14 or increased cardiac afterload.15 These phenomena lead to a progressive deterioration in cardiac function. In addition, an increased heart rate may cause a decrease in left ventricular systolic function because of the presence of a negative force-frequency phenomenon in IDCM.6 It was found that even modest increments in heart rate induced by pacing caused a decrease in left ventricular ejection fraction in patients with IDCM.6 We have previously demonstrated evidence to suggest a negative force-frequency relationship at rest in patients with IDCM.7 In that study, left ventricular ejection fraction was not correlated with heart rate in patients with normal hearts between a resting spontaneous heart rate range of 40 to 120 beats/min, whereas in patients with IDCM left ventricular ejection fraction was moderately and significantly negatively correlated with resting heart rate (r = –0.55, P = .0007).

Diastolic function was influenced by heart rate in patients with IDCM. An increase in heart rate induced by cardiac pacing produced a modest increment in peak filling rate in IDCM but a more marked increase in this measure in patients with normal function.16 Resting heart rate was positively correlated with peak mitral inflow velocity and negatively correlated with mitral inflow deceleration time in IDCM.4 The former of these two Doppler measures can be used to derive an estimate of peak filling rate that correlated well with the radionuclide determination of peak filling rate.17 Previous studies7, 18 in normal hearts showed a modest positive correlation between peak filling rate (in stroke volume per second) and heart rate and a modest negative correlation between time to peak filling rate and heart rate. However, the regression slopes of these relationships were relatively flat.7 In contrast, in IDCM the correlations between these measures, although directionally similar to those in normal hearts, were stronger and the regression lines steeper.7 Thus an increased resting heart rate in IDCM compared with normal hearts produced more restrictive filling in the rapid filling period, with a reduced time to peak filling rate and an increased peak filling rate, and conversely a slower resting heart rate produced a more delayed pattern of relaxation in IDCM. The resting portion of the current study before any intervention was consistent with the findings of our earlier study.7

Hasenfuss et al19 have provided experimental evidence linking the negative force-frequency relationship and the occurrence of frequency-related diastolic dysfunction in patients with failing hearts. These workers found that muscle strips taken from the explanted hearts of patients undergoing heart transplantation for end-stage heart failure demonstrated frequency-related increases in diastolic force and decreases in diastolic force decline with stimulation. These muscle strips also showed a frequency-related decline in developed force with stimulation. These phenomena were related to the balance of sarcoplasmic reticulum Ca++-adenosine triphosphatase and Na+-Ca++ exchanger.

The current study was the first to show that the interaction of heart rate and left ventricular ejection fraction and peak filling rate was not altered by exercise in IDCM. It should be noted that after exercise there was no longer a correlation between heart rate and time to peak filling rate in IDCM. Time to peak filling rate includes the isovolumic relaxation period and a portion of the early rapid filling period and thus may be influenced by deactivation of the contractile process and by passive diastolic properties of the left ventricle. Thus the increase in heart rate associated with exercise may have variable effects on these two portions of the filling period, eliminating the group relationship of heart rate and time to peak filling period. In contrast, left ventricular peak filling rate occurs later in diastole and therefore may be influenced primarily by passive diastolic properties. An earlier study demonstrated that exercise in patients with IDCM was associated with incomplete relaxation and increased left ventricular chamber stiffness.20 That study illustrated very variable left ventricular pressure and volume relationships in 12 patients with IDCM in both early and late diastole.

Metoprolol and cardiac function 

Several studies have observed the effects of metoprolol on cardiac function in patients with IDCM.1, 2, 3, 4, 21, 22 β-Blockade in these studies caused an increase in left ventricular ejection fraction and a decrease in heart rate. The current study also shows that metoprolol but not placebo caused a significant increase in left ventricular ejection fraction in patients with IDCM both at rest and with exercise.

Andersson et al4 showed improvement in resting diastolic function with metoprolol treatment. Metoprolol was associated with a lengthened E-wave deceleration time, measured from the mitral valve inflow signal by Doppler echocardiography. In the current study, with metoprolol during mild exercise, a decrease in peak filling rate and a lengthening in the time to peak filling rate occurred. Similar findings tended to occur at rest but were not statistically significant. The results of this study, like those of Andersson et al,4 suggest that metoprolol has an effect on diastolic function in IDCM. The changes in peak filling rate and in time to peak filling observed in the current study indicate that left ventricular filling became less restrictive. This interpretation was also suggested by the study of Andersson et al,4 in which some patients with IDCM before treatment had a single, early rapid filling wave in the mitral inflow signal consistent with restrictive physiologic mechanisms. Metoprolol treatment in these patients was associated with the appearance of discrete E and A waves in the mitral inflow signal, a pattern consistent with more normal ventricular filling.

A later study by Andersson et al21 failed to show any change in peak filling rate at rest after metoprolol treatment by radionuclide ventriculography. The current study differs from that earlier study in that peak filling rate was measured in stroke volumes per second rather than end-diastolic volumes per second and metoprolol in the current study produced a more marked decline in heart rate. The measurement of peak filling rate in stroke volumes per second may make more apparent the less restrictive pattern of left ventricular filling with metoprolol, particularly because in the current study there was no change in the left ventricular end-diastolic volume index and there was a nonsignificant tendency for the stroke volume index to increase after metoprolol. The greater change in heart rate in the current study may have contributed to the appearance of a less restrictive pattern of left ventricular filling seen in the current study compared with that of the earlier study of Andersson et al.21 The results of the current study are quite consistent with the opinion of Andersson et al21 that the diastolic benefit of metoprolol is related to prolongation of early rapid filling rather than enhancement of maximal early rapid filling. Kim et al22 demonstrated in patients with IDCM that metoprolol produced reduction in left ventricular chamber and myocardial stiffness constants, findings that these workers related to the less restrictive pattern of filling after metoprolol noted by Andersson et al.21

This study shows that after metoprolol treatment heart rate continued to be positively correlated with left ventricular peak filling rate and that there was a continued tendency for heart rate to be negatively correlated with left ventricular ejection fraction. However, the latter finding was not significant, perhaps because of the small number of patients in this analysis. These trends after metoprolol treatment were similar in magnitude and slope to the relationship of resting heart rate and peak filling rate and left ventricular ejection fraction before any treatment. These findings suggest that the beneficial effects of metoprolol on diastolic and systolic function were associated in part with the concomitant heart rate slowing. The possibility of additional mechanisms for the benefit of metoprolol in IDCM was suggested by the study of Andersson et al,21 which indicated that after metoprolol treatment patients with IDCM continued to show an improvement in peak ejection rate even when paced to a higher heart rate; these additional benefits may well be related to blockade of the detrimental effects of sympathetic overactivity.

Time to peak filling rate and heart rate were no longer significantly correlated at rest after treatment with metoprolol. This may be related to the small number of patients studied after metoprolol treatment. Another reason for this finding may be the differential effects of metoprolol on early and late diastolic properties of the left ventricle. Kim et al22 demonstrated that metoprolol shortened the isovolumic relaxation index and decreased left ventricular chamber stiffness. It is hypothesized that the interaction of these two effects may have, in a patient-dependent manner, negated the correlation of time to peak filling rate and heart rate.

Study limitations 

The chief limitation of this study is the small number of patients enrolled. The patient number was limited, in part, because patients with atrial fibrillation were excluded from the study owing to the need to maintain a regular cardiac cycle length to obtain satisfactory gated cardiac blood-pool acquisitions.

The 3-month duration of the study does not allow any projection as to the long-term benefit of metoprolol on cardiac function. In addition, no information can be provided regarding how the effects of metoprolol on systolic or diastolic function affect relief of symptoms, the development of heart failure, or prognosis.