Effect of Nicorandil on Left Ventricular End-Diastolic Pressure During Exercise in Patients with Hypertrophic Cardiomyopathy

Abstract

Aims Impaired coronary microcirculation is thought to contribute to myocardial ischaemia, causing an abnormal increase in left ventricular end-diastolic pressure during exercise in individuals with hypertrophic cardiomyopathy. The effects of nicorandil on left ventricular end-diastolic pressure during exercise were examined in patients with this condition.

Methods and results Left ventricular pressures and dimensions were measured simultaneously during supine bicycle exercise in 23 patients with nonobstructive hypertrophic cardiomyopathy, before and after intravenous injection of either nicorandil (0.1mg/kg) or propranolol (0.15mg/kg). Exercise thallium-201 scintigraphy was also performed. Patients were grouped according to the changes in left ventricular end-diastolic pressure during exercise before treatment. Group I comprised 13 patients in whom left ventricular end-diastolic pressure increased progressively to abnormal values during exercise; group II comprised 10 patients in whom left ventricular end-diastolic pressure changed biphasically. The extents of both left ventricular hypertrophy and ischemic burden during exercise were greater in group I than in group II. Of the eight group I patients who received nicorandil, four individuals exhibited biphasic changes in left ventricular end-diastolic pressure during exercise after its administration whereas four subjects showed no such effect of the drug. Left ventricular end-diastolic pressure increased progressively during exercise after propranolol treatment in all 6 group II patients given this drug.

Conclusion Nicorandil has a salutary effect on the changes in left ventricular end-diastolic pressure during exercise in patients with hypertrophic cardiomyopathy.

Coronary microcirculation, Exercise, Hypertrophic cardiomyopathy, Left ventricular end-diastolic pressure, Nicorandil

1 Introduction

Myocardial ischaemia plays an important role in the pathophysiology and natural history of hypertrophic cardiomyopathy.^1–4Chest pain is a common symptom of individuals with hypertrophic cardiomyopathy even though their epicardial coronary arteries are angiographically normal.^5,6Furthermore, such individuals often exhibit reversible exercise-induced myocardial perfusion defects on thallium-201 imaging or abnormal lactate metabolism during pacing, suggesting that the occurrence of ischaemia is the result of impaired coronary microcirculation.⁷However, the effect of dynamic exercise on coronary microcirculation in patients with hypertrophic cardiomyopathy has not been elucidated. We recently demonstrated that left ventricular end-diastolic pressure increases steadily up to abnormal values during exercise in patients with severe hypertrophic cardiomyopathy, whereas it exhibits biphasic changes, with an initial progressive increase (ascending limb) and subsequent gradual decline up to peak exercise (descending limb), in individuals with mild hypertrophic cardiomyopathy.⁸Furthermore, this biphasic pattern of changes in this parameter was not observed after the administration of propranolol. Because the coronary arteries are richly innervated by adrenergic nerves, dynamic exercise induces coronary vasodilation mediated by adrenergiceffect.⁹Then the biphasic changes in left ventricular end-diastolic pressure may be related to improved coronary microcirculation in response to exercise-induced stimulation of β-adrenergic receptors in patients with mild hypertrophiccardiomyopathy.

Nicorandil exhibits complex pharmacological effects, including effects similar to those of nitrate,¹⁰ischaemic preconditioning,¹¹activation of not only sarcolemmal potassium channels but also mitochondrial ATP-sensitive potassium channels,¹⁰and therefore eliciting cardioprotective effects.¹²We therefore examined the hypothesis that nicorandil may favourably affect exercise-induced changes in left ventricular end-diastolic pressure in patients with hypertrophic cardiomyopathy.

2 Methods

2.1 Patients

The study subjects comprised 23 consecutive patients with newly diagnosed nonfamilial, nonobstructive hypertrophic cardiomyopathy. All patients were male, ranging in age from 37 to 58 years (mean±standard deviation, 49±6) who were different patients enrolled in our previous study⁸and referred for cardiac catheterization. All patients experienced occasional mild breathlessness or atypical chest pain, but none had previously taken any prescription medications. All patients exhibited normal sinus rhythm and a normal ejection fraction (mean±standard deviation, 69±6%) on left ventriculograms. Hypertrophic cardiomyopathy was diagnosed by established clinical, haemodynamic, and echocardiographic criteria.¹³No significant intraventricular pressure gradient was detected in any patient either at rest or during provocative maneuvers (the Valsalva manouver and isoproterenol infusion) after completion of the study protocol. Angiography demonstrated that no patient had a narrowing of the coronary arteries of >50%. Written informed consent was obtained from all subjects, and the study protocol was approved by the appropriate institutional review committee.

2.2 Study protocol

An externally balanced and calibrated 6F pigtail angiographic micromanometer-tipped catheter was advanced into the left ventricle through the right radial artery for measurement of left ventricular pressure. A 7F triple-lumen thermistor Swan-Ganz catheter was advanced into the pulmonary artery through the right brachial vein. Micromanometer pressure signals and standard electrocardiograms were recorded with a multichannel recorder throughout the study protocol. Two-dimensional echocardiographic measurements were performed from recordings of at least five consecutive cardiac cycles by two observers who were unaware ofthe patients’ clinical status. The degree of left ventricular hypertrophy was evaluated semiquantitatively as previously described.¹⁴

After we had collected baseline data, patients performed a symptom-limited supine bicycle ergometer exercise test as described previously.¹⁴The workload was 25W initially and increased by 25W every 3min. We were unable to obtain clear echocardiographic recordings at workloads of >50W, probably because of an increase in the air content of the lungs. We therefore analyzed echocardiographic data during exercise at a workload of 50W. During exercise, no patients developed a new outflow tract pressure gradient or more than a mild level of mitral regurgitation as revealed by Doppler echocardiography.

After completion of the exercise study, patients were randomly selected to receive an intravenous injection of nicorandil (0.1mg/kg, 12 patients) or propranolol (0.15mg/kg, 11 patients). The exercise protocol was repeated 15min after injection of these drugs.

Blood (5ml) was collected from the brachial artery at rest and at peak exercise. Plasma samples were prepared and stored at −70°C until determination of norepinephrine concentration by high-performance liquid chromatography. Exercise thallium-201 scintigraphic studies were also performed 2 days before catheterization.

2.3 Analysis of haemodynamic data

Left ventricular pressure signals were digitized at 3-ms intervals and analyzed with software developed in our laboratory and a personal computer system.¹⁵The left ventricular pressure data at baseline and at 7 to 10 points during exercise were selected for analysis. We calculated the maximum first derivative of left ventricular pressure as an index of contractility. To evaluate left ventricular isovolumic relaxation, we computed the pressure half-time directly as previously described.¹⁶We defined the critical heart rate as the heart rate at which the left ventricular end-diastolic pressure reached a maximal value during the exerciseprotocol.

2.4 Scintigraphic analysis

Perfusion was assessed semiquantitatively on the basis of analysis of the apical, midventricular, and basal short-axis and of vertical long-axis tomograms. The left ventricular myocardium was divided into 20 segments (18 from the short-axis images and 2 from the vertical long-axis images). The defect score was defined according to a five-point scale (0=normal, 1=equivocal, 2=mildly reduced perfusion, 3=severely reduced perfusion, 4=absent perfusion) by two observers without knowledge of the clinical data.¹⁷The summed stress score and summed rest score were calculated as the sum of the scores for the 20 segments for the stress and rest images, respectively. The sum of the differences between the 20 segments from the stress and rest images was defined as the summed difference score.¹⁸

2.5 Statistical analysis

Data are expressed as means±standard deviation. Differences among subgroups of group I (see Results) were compared by one-way factorial analysis of variance together with Scheffe’s multiple comparison test. Other comparisons were performed by paired or unpaired Student’s t-test as appropriate. A P value of <0.05 was considered statistically significant.

3 Results

3.1 Classification of patients

The study subjects could be grouped on the basis of the patterns of changes in left ventricular end-diastolic pressure in response to exercise without any drug. Group I consisted of 13 individuals whose left ventricular end-diastolic pressure increased progressively with exercise intensity (Fig. 1). Group II consisted of 10 patients whose left ventricular end-diastolic pressure changes during exercise were biphasic, with an initial progressive increase (ascending limb) and a subsequent gradual decline (descending limb) (Fig. 2). Both the interventricular septal thickness and the left ventricular hypertrophy score were significantly greater in group I than in group II (Table 1). There were no significant differences in haemodynamic variables at baseline between groups I and II.

Fig. 1

Distribution of patients on the basis of the influence of drugs on the pattern of changes in left ventricular end-diastolic pressure (ΔLVEDP) during exercise in group I. The curves shown are from individual patients representative of each subgroup.

Fig. 2

Effects of nicorandil and propranolol on the pattern of changes in left ventricular end-diastolic pressure (ΔLVEDP) during exercise in group II. The curves shown were obtained from representative individuals.

Table 1

Baseline clinical, echocardiographic, hemodynamic, and ventriculographic characteristics of patient groups

Characteristic	Group I(n=13)	Group II(n=10)
Age (years)	51±6 (37–58)	47±5 (39–56)
Coronary risk factors
Smoking	5 (38%)	4 (40%)
Hypertension	0 (0%)	0 (0%)
Diabetes mellitus	0 (0%)	0 (0%)
Hypercholesterolemia	2 (15%)	1 (10%)
Total cholesterol (mg/dl)	208±21	199±25
Interventricular septal thickness (mm)	22±6	17±3a
Posterior wall thickness (mm)	12±2	11±1
LV end-diastolic dimension (mm)	45±3	46±3
LV end-systolic dimension (mm)	28±5	30±3
LV hypertrophy score	7±1	5±2a
LV ejection fraction (%)	69±7	68±4
Heart rate (bpm)	62±6	63±6
LV peak systolic pressure (mmHg)	125±11	127±13
LV end-diastolic pressure (mmHg)	15±6	15±6
LV dP/dtmax(mmHg/s)	1937±296	2021±427
LV dP/dtmin(mmHg/s)	−1727±412	−1745±538
T1/2(ms)	49±6	44±6
PAWP (mmHg)	12±5	10±4
MRAP (mmHg)	3±2	3±2

a

P<0.05 vs group I. LV, left ventricular; LV dP/dt_maxand LV dP/dt_min, maximum and minimum first derivative of left ventricular pressure, respectively; T_1/2, left ventricular pressure half-time; PAWP, pulmonary artery wedge pressure; MRAP, mean right atrial pressure. Data are presented as the mean value ± standard deviation or number (%) of patients.

The summed difference score for exercise thallium-201 scintigraphy was significantly greater in group I than in group II (9.1±3.6 vs 3.2±2.1, P<0.01), suggesting that the extent of the exercise-induced ischaemic burden was lower in group II. Exercise induced an increase in the plasma concentration of norepinephrine in all patients. The plasma norepinephrine concentrations at baseline (group I, 249±46 pg/ml; group II, 221±51pg/ml)or at peak exercise (group I, 883±158pg/ml; group II, 891±243pg/ml) without drugs did not differ significantly between the two groups.

3.2 Changes in haemodynamic and echocardiographic variables during exercise in group I

Six out of 13 patients in group I complained of chest pain during exercise without drugs, and significant ST segment depression was apparent in two of these six patients. Eight and five of the 13 patients in group I received an intravenous injection of nicorandil or propranolol, respectively (Fig. 1). Nicorandil had no effect on the pattern of changes in left ventricular end-diastolic pressure induced by exercise in four of the eight patients in group I treated with this drug (nicorandil-A group), whereas the remaining four patients exhibited a biphasic pattern of changes in pressure duringexercise after nicorandil injection (nicorandil-B group). Nicorandil slightly reduced left ventricular end-diastolic pressure at baseline from 16±4 to 13±6mmHg in the nicorandil-A group and from 14±8 to 11±6mmHg in the nicorandil-B group. There were no differences in age, plasma norepinephrine concentrations, or the prevalence of risk factors for coronary artery disease, including smoking, hypertension, diabetes mellitus, and hypercholesterolemia between the nicorandil-A and nicorandil-B groups. Haemodynamic and echocardiographic variables at baseline, with or without nicorandil, did not differ significantly between the nicorandil-A and nicorandil-B groups. The both two patients with ST segment depression during exercise without drug belonged to the nicorandil-A group.

Changes in these variables from baseline to peak exercise with or without drugs in the three subgroups of group I are shown in Table 2. Left ventricular end-diastolic pressure increased from13±6mmHg at baseline with nicorandil to 28±6mmHg at peak exercise in the nicorandil-A group. In the nicorandil-B group, left ventricular end-diastolic pressure increased from 11±6mmHg at baseline with nicorandil to 20±6mmHg at the critical heart rate, then it gradually returned to the baseline value (12±9mmHg). The mean critical heart rate in the nicorandil-B group was 93±13bpm. Propranolol had no effect on the pattern of changes in left ventricular end-diastolic pressure during exercise in any of the five patients in group I who received this drug (Fig. 1).

Table 2

Percentage or absolute (Δ) changes in various parameters from baseline to peak exercise with or without drugs in group I

	Nicorandil-A (n=4)			Nicorandil-B (n=4)			Propranolol (n=5)
Parameter	Withoutdrug	Withdrug	Withoutdrug	Withdrug	Withoutdrug	Withdrug
ΔHeart rate (bpm)	43±4	48±4	45±7	49±11	48±9	36±7a,b,c
ΔLV peak-systolic pressure (mmHg)	48±37	35±26	40±24	34±19	39±12	27±7a
ΔLV end-diastolic pressure (mmHg)	17±4	15±4	16±3	2±5a,b	16±3	16±3c
LV dP/dtmax(%)	52±21	54±14	48±14	50±21	48±19	24±11b,c
LV dP/dtmin(%)	−44±28	−41±35	−47±29	−46±26	−46±15	−27±10a
T1/2(%)	−16±3	−7±8	−17±8	−0±5	−17±6	−10±4a,c
ΔPAWP (mmHg)	15±6	14±6	15±5	9±5a,b	16±6	15±5c
ΔMRAP (mmHg)	4±2	4±2	4±2	3±2	4±2	3±2
ΔLV end-diastolic dimension (mm)	−4±2	−4±2	−4±3	−5±3	−3±2	−1±1a,c

a

P<0.05 vs Without drug.

b

P<0.05 vs Nicorandil-A.

c

P<0.05 vs Nicorandil-B. Abbreviations as in Table 1.

3.3 Changes in haemodynamic and echocardiographic variables during exercise in group II

No patient in group II complained of chest pain or showed significant ST segment depression during exercise without drugs. The biphasic changes in left ventricular end-diastolic pressure apparent in patients in this group during exercise without drugs were characterized by an initial increase (from 15±6mmHg at baseline to 26±7mmHg at the critical heart rate) and a subsequent progressive decline back to the baseline value (15±7mmHg). The mean critical heart rate was 86±13bpm. Changes in the maximum first derivative of left ventricular pressure and the pressure half-time in group II were significantly greater (P<0.01) than those in group I. The decrease in left ventricular end-diastolic dimension from baseline to peak exercise without drugs in group II was similar to that in group I. Four and six of the 10 patients in group II received an intravenous injection of nicorandil or propranolol, respectively (Fig. 2). The changes in haemodynamic or echocardiographic variables during exercise without drugs did not differ significantly between these two subgroups of group II (Table 3). Nicorandil had no effect on the pattern of changes in left ventricular end-diastolic pressure during exercise in any of the four patients in group II who received this drug (Fig. 2). In contrast, all six patients in group II who received propranolol no longer exhibited a biphasic pattern of exercise-induced changes in this parameter after drug injection; rather, left ventricular end-diastolic pressure increased progressively up to 29±5mmHg at peak exercise. Propranolol significantly reduced (P<0.05) both left ventricular peak systolic pressure and the maximum first derivative of left ventricular pressure at baseline, but it did not affect either left ventricular end-diastolic pressure or the pressure half-time at baseline.

Table 3

Percentage or absolute (Δ) changes in various parameters from baseline to peak exercise with or without drugs in group II

	Nicorandil (n=4)			Propranolol (n=6)
Parameter	Without drug	With drug	Without drug	With drug
ΔHeart rate (bpm)	47±11	47±11	46±11	37±7a,b
ΔLV peak-systolic pressure (mmHg)	43±20	36±13	40±13	25±8a
ΔLV end-diastolic pressure (mmHg)	0±3	−1±4	0±4	16±4a,b
LV dP/dtmax(%)	80±20	74±22	80±24	46±17a,b
LV dP/dtmin(%)	−59±11	−69±16	−63±10	−31±11a,b
T1/2(%)	−31±14	−30±12	−36±12	−20±9a,b
ΔPAWP (mmHg)	10±6	9±5	10±6	15±5a,b
ΔMRAP (mmHg)	4±2	4±2	4±2	3±4
ΔLV end-diastolic dimension (mm)	−4±2	−5±3	−4±3	0±1a,b

a

P<0.05 vs Without drug.

b

P<0.05 vs Nicorandil. Abbreviations as in Table 1.

4 Discussion

Our results demonstrate that the intravenous administration of nicorandil at a clinically relevant dose exerts a salutary effect on the serial changes in left ventricular end-diastolic pressure duringdynamic exercise in patients with hypertrophiccardiomyopathy.

Patients with hypertrophic cardiomhyopathy could be divided into two groups on the basis of differences in the pattern of changes in left ventricular end-diastolic pressure induced by dynamic exercise described previously.⁸Group I comprised patients in whom left ventricular end-diastolic pressure increased steadily up to abnormal values during exercise, whereas group II consisted of patients in whom this parameter exhibited a biphasic pattern of changes during exercise. The extents of both left ventricular hypertrophy (as revealed by echocardiography) and ischemic burden during exercise (as revealed by thallium-201 scintigraphy) were more severe in the patients in group I than in those in group II. The exercise-induced biphasic pattern of changes in left ventricular end-diastolic pressure in group II patients was no longer apparent after injection of propranolol. Conversely, half of the patients in group I who received nicorandil exhibited a biphasic pattern of exercise-induced changes in left ventricular end-diastolic pressure after drug administration, whereas nicorandil had no effect on the remaining patients injected with this drug in group I.

4.1 Coronary microcirculation as a potential determinant of biphasic changes in left ventricular end-diastolic pressure during exercise

Left ventricular end-diastolic pressure is commonly considered to be an index of left ventricular diastolic dysfunction, given that increases in left ventricular end-diastolic volume are usually accompanied by increases in left ventricular end-diastolic pressure. However, left ventricular end-diastolic pressure may be elevated without an increase in left ventricular end-diastolic volume as a result of diminished ventricular compliance.¹⁹In the present study, our observation that patients in groups I and II exhibited similar changes in left ventricular internal dimensions indicated that exercise induced a substantial deterioration of left ventricular compliance in patients in group I and a substantial improvement in left ventricular compliance in patients in group II. One of the major factors responsible for a diminished left ventricular compliance is myocardial ischemia.¹⁹Various studies have suggested that coronary microcirculation is abnormal in patients with hypertrophic cardiomyopathy.^1–5Autopsy revealed a qualitative abnormality of intramural coronary arteries in most patients examined with hypertrophic cardiomyopathy.¹In addition, coronary flow reserve was shown to be reduced, indicating a serious disturbance in coronary microcirculation, in patients with hypertrophic cardiomyopathy^[20].

The adrenergic stimulation associated with dynamic exercise results in dilation of coronary resistance vessels through both of direct activation of vascular β-adrenergic receptors and of the secondary effect of an increase in myocardial metabolic demand.⁹In addition to studies that have concluded that the direct control of coronary resistance vessels is mediated predominantly through β₂-adrenergic receptors,²¹β₁-adrenergic receptors have also been implicated in such control.²²Both α₁- and α₂-adrenergic receptors contribute to regulation of the degree of vasoconstriction of the smaller resistance vessels.²³However, the vasoconstrictive effect of α-adrenergic stimulation is normally overridden by metabolic and direct β-adrenergic receptor-induced vasodilation. Thus, the net effect of dynamic exercise on coronary microcirculation is coronary vasodilation in individuals with normal coronary arteries. However, in subjects with diseased coronary arteries, dynamic exercise more likely induces vasoconstriction because of decreased endothelium-derived nitric oxide.²⁴

In the present study, the exercise-induced biphasic pattern of changes in left ventricular end-diastolic pressure was no longer apparent in patients of group II after propranolol injection. Taken together with this observation, an amelioration of ischaemia as a result of an improvement in coronary microcirculation induced by β-adrenergic stimulation during exercise may underlie this biphasic pattern. The fact that both left ventricular contraction and relaxation during exercise were greater in patients of group II than in those of group I might be indicative of the amelioration of ischaemia during exercise in patients of group II. However, given that no single mechanism predominates in the control of coronary vascular tone, especially during exercise, with neural, humoral, and local metabolic mechanisms all participating^[25], further studies are required to address this complex issue.

4.2 Effect of nicorandil on left ventricular end-diastolic pressure during exercise in patients with hypertrophic cardiomyopathy

Some patients with hypertrophic cardiomyopathy who normally exhibit a progressive increase in left ventricular end-diastolic pressure during exercise manifest a biphasic pattern of exercise-induced changes after injection with nicorandil. It is noteworthy that exercise with nicorandil increased the maximum first derivative of left ventricular pressure to similar extents in both the nicorandil-A and nicorandil-B groups (54±14 vs 50±21%, respectively). Importantly, the left ventricular end-diastolic pressure in the nicorandil-B group decreased after the critical heart rate was achieved, while the left ventricular end-diastolic pressure increased progressively throughout exercise in the nicorandil-A group. When left ventricular preload is decreased, the maximum first derivative of left ventricular pressure is decreased because of Frank-Staring mechanism. Therefore, this finding strongly suggested the augmented left ventricular contractility during exercise with nicorandil in the nicorandil-B group, indicating the possibility of the improved ischaemia. Nicorandil appears to have two principal mechanisms of its action: First, it has a nitrate-like vasodilatoryeffect,¹⁰which, together with arteriolar and venous vasodilation, is likely responsible for the efficacy of this drug in relieving ischaemia caused by an increased myocardial oxygen demand. We previously showed that the ratio of the pressure-rate product to coronary sinus flow, which is an index of the ratio of myocardial oxygen consumption to myocardial oxygen supply, was decreased significantly by nicorandil administration during exercise in patients with old myocardial infarction.²⁶Second, nicorandil also activates mitochondrial ATP-sensitive potassium channels that are cytoprotective during ischemia.¹¹It is mostly the small and intermediate-size resistance vessels (<100μm in diameter) that are dilated by ATP-sensitive potassium channel openers.²⁷

Although it is difficult to evaluate the structural alterations of coronary resistance vessels in intact humans, it is possible that patients in whom left ventricular end-diastolic pressure continues toincrease steadily during exercise after nicorandil injection possess severe structural alterations of their coronary resistance vessels. On the other hand, in patients in whom nicorandil confers a biphasic pattern of changes in left ventricular end-diastolic pressure during exercise, the structural alterations in these vessels might be subtle and the functional coronary flow reserve targeted by nicorandil may be preserved. Nicorandil infusion in conjunction with ergometer exercise assessment may thus prove to be an effective approach for evaluation of structural alterations of coronary resistance vessels in patients with hypertrophic cardiomyopathy. Further investigations at the cellular and molecular levels are required, however, to characterize the functional and structural alterations in the intramural coronary arteries of patients with hypertrophic cardiomyopathy.

4.3 Study limitations

We should discuss the difference between left end-diastolic pressure and pulmonary artery wedge pressure at peak exercise in group II. This discrepancy between these pressures may be caused by several factors. First, the time delay between left end-diastolic pressure and pulmonary artery wedge pressure results in the difference between these pressures at peak exercise. Second, the pulmonary artery wedge pressure may be contaminated by pulmonary artery pressure, yielding greater mean pulmonary artery wedge pressure at peak exercise. Third, the pulmonary artery wedge pressure isaffected by changes in intrathoracic pressure at peak exercise.

5 Conclusions

We have demonstrated that nicorandil has a salutary effect on the serial changes in left ventricular end-diastolic pressure induced by dynamic exercise in patients with hypertrophic cardiomyopathy.Nicorandil treatment induced biphasic pattern of exercise-induced changes in left ventricular end-diastolic pressure in some patients with hypertrophic cardiomyopathy who showed progressive increase in left ventricular end-diastolic pressure during exercise without this drug. Our results suggest that nicorandil might have a beneficial effect on abnormal coronary microcirculation in patients with hypertrophic cardiomyopathy.