Sympathetic denervation in association with coronary artery grafting in patients with Prinzmetals' angina

Management of Coronary Artery Spasm

Calcium channel blockers (CCBs) are the first-line treatment for coronary artery spasm (CAS). When CAS-related angina symptoms are not well controlled by CCB therapy, long-acting nitrates or (where available) nicorandil can be added as second-line medications. In the case of CAS refractory to standard treatments, several other alternative drugs and interventions have been proposed, including the Rho-kinase inhibitor fasudil, anti-adrenergic drugs, neural therapies and percutaneous coronary interventions. In patients with syncope or cardiac arrest caused by CAS-related tachyarrhythmias, or even bradyarrhythmias, implantation of an ICD or pacemaker, respectively, should be considered according to the risk of recurrence and efficacy of vasodilator therapy.

Sympathectomy versus conventional treatment for refractory coronary artery spasm

BACKGROUND

There is no clear consensus on the potential efficacy and indications for sympathectomy to prevent recurrence of vasospasm in patients with refractory coronary artery spasm (CAS).

OBJECTIVE

To compare the clinical outcomes of sympathectomy with those of conventional treatment in patients with refractory CAS.

PATIENTS AND METHODS

Patients with refractory CAS were randomly assigned to sympathectomy group (n = 37) or conventional treatment group (n = 42). The primary end point was a composite of major adverse cardiac event (MACE) episodes (including cardiac death, nonfatal myocardial infarction, unstable angina, heart failure, and life-threatening arrhythmia), and the secondary end point was death from any cause within 24 months after randomization.

RESULTS

During the follow-up period of 24 months, the incidence of MACE in the sympathectomy and conventional treatment groups was 16.22 and 61.90%, respectively (P = 0.0001). All-cause death as the secondary end point occurred in zero and six (14.29%) patients, respectively (P = 0.0272). The Kaplan-Meier curve for MACE and all-cause death showed a significant between-group difference (log-rank test, P = 0.0013 and 0.0176, respectively).

CONCLUSION

Compared with conventional treatment, sympathectomy significantly reduced the composite end point of MACE episodes and death from any cause in patients with refractory CAS by effectively preventing recurrence of vasospasm.

Percutaneous Phenol Block of the Upper Thoracic Sympathetic Chain with Computed Tomography Guidance

Twenty-one percutaneous neurolyses of the upper thoracic sympathetic chain were performed in 12 patients with CT guidance by a single injection of 1 to 3 ml of phenol at the level of Th3. Results were assessed after a follow-up period varying from 4 to 33 months. Three patients with hyperhidrosis had immediate and complete disappearance of symptoms, but only one patient remained dry. In 7/14 procedures done for Raynaud's disease symptoms disappeared or diminished. These long term results are competitive with surgery. Three transitory Horner syndromes and one pneumothorax occurred.

Ventral cardiac denervation increased right coronary arterial blood flow

BACKGROUND

Cardiac denervation accompanied with coronary artery bypass surgery has been widely performed for the treatment of vasospastic angina associated with atherosclerotic coronary artery disease. However, the effect of cardiac denervation on phasic coronary blood flow patterns of the left anterior descending coronary artery (LAD), left circumflex coronary artery (LCX) and right coronary artery (RCA) remains unknown. This study aimed to investigate the effect of cardiac denervation on phasic coronary blood flow patterns of the LAD, LCX and RCA.

METHODS

Phasic coronary blood flow patterns were analyzed using three flow probes placed around the LAD, LCX and RCA with and without LAD stenosis. Ventral cardiac denervation (VCD) was performed in 8 pigs, and 16 pigs were used as control subjects. Autonomic activities before and after the VCD were quantified by wavelet analysis of heart rate variability.

RESULTS

The mean LAD flow (34.4+/-9.4 to 32.6+/-7.1 ml/min, p=0.638) and mean LCX flow (26.3+/-10.2 to 27.2+/-6.0 ml/min, p=0.825) showed no significant change after VCD, while the mean RCA flow (31.3+/-9.0 to 38.2+/-11.2 ml/min, p=0.003) significantly increased. The hemodynamic variables in the VCD group were well maintained after creation of LAD stenosis, while they deteriorated in the control group. The low-frequency components, high-frequency components and their ratio did not change after VCD.

CONCLUSIONS

VCD prevented the deterioration of cardiac function after creation of an LAD stenosis and resulted in an increase of the mean RCA flow. VCD did not affect autonomic nervous system activity.

Ventral cardiac denervation reduces the incidence of atrial fibrillation after coronary artery bypass grafting

OBJECTIVES

Because the autonomic nervous system is an important determinant in the appearance of atrial fibrillation, we have assessed the role of ventral cardiac denervation for its prevention.

METHODS

Patients undergoing low-risk coronary artery surgery were enrolled. No routine antiarrhythmic drugs were administered before or after the operation. Ventral cardiac denervation was performed in 207 patients, and 219 patients were used as control subjects. Denervation was performed before cardiopulmonary bypass. The groups were comparable regarding demographic, clinical, and operative variables.

RESULTS

The additional time for the denervation was 5 +/- 2 minutes, and there were no associated complications. Postoperative atrial fibrillation was present in 15 (7%) patients undergoing ventral cardiac denervation (95% confidence interval, 4%-12%) and in 56 (27%) control subjects (95% confidence interval, 18%-35%). Patients submitted to ventral cardiac denervation had fewer and less severe episodes of atrial fibrillation, and no patient had atrial fibrillation after discharge. Ventral cardiac denervation was the most significant predictor of postoperative atrial fibrillation (odds ratio, 0.42; confidence interval, 0.23-0.78; P =.006). Age of greater than 65 years (odds ratio, 1.67; confidence interval, 0.96-2.9; P =.067) was a highly suggestive predictor. The analysis of the effect of ventral cardiac denervation correlated with the patient's age showed a more pronounced effect in patients younger than 70 years (odds ratio, 0.43; confidence interval, 0.22-0.86; P =.022)

CONCLUSIONS

Ventral cardiac denervation is a fast and low-risk procedure. Its use significantly reduces the incidence and severity of atrial fibrillation after routine coronary artery bypass surgery. Patients younger than 70 years of age are expected to have a higher success rate than those older than 70 years.

[The placement of a Wiktor stent for the treatment of vasospastic angina: a case report]

A 46-year-old male with a fixed stenosis in the mid-segment of the left anterior descending artery underwent balloon angioplasty. The procedure included the placement of two Wiktor stents because of severe dissection. Five months later he complained of Prinzmetal angina with ST elevation in the anterior wall. A metilergobasine test during the coronary arteriogram showed a discrete, severe spasm on the proximal segment of the left anterior descending artery. Because of a lack of symptomatic improvement with high-dose nitrates and calcium blockers, a Wiktor coronary stent was successfully implanted in the proximal left anterior descending artery, resulting in complete relief of the angina.

Effects of cardiac denervation on coronary and systemic circulation

The effects of cardiac denervation on coronary circulation and systemic hemodynamics were examined in patients who underwent coronary artery bypass grafting for a totally occluded left anterior descending coronary artery. The results in 14 patients who had undergone a combination of bypass grafting and intrapericardial denervation (group I) were compared with those in 19 patients who had undergone only bypass grafting (group II). The heart rate was 110 +/- 4.9 beats/min in group I and 104 +/- 3.7 beats/min in group II after denervation. Although the cardiac index was not significantly different between the two groups, the calculated systemic vascular resistance index of group I (1,358 +/- 111 dynes.s.cm-5.m2) was significantly lower (p < 0.02) than that in group II (1,713 +/- 75 dynes.s.cm-5.m2). The end-diastolic coronary resistance was significantly lower (p < 0.05) in group I (0.59 +/- 0.05 mm Hg.mL-1.min-1) than that in group II (1.04 +/- 0.14 mm Hg.mL-1.min-1), and the end-diastolic graft flow/mean graft flow ratio was 1.56 +/- 0.05 in group I and 1.28 +/- 0.04 in group II (p < 0.01). The plasma norepinephrine concentration in group I (223 +/- 52 pg/mL) was significantly low (p < 0.05) compared with that in group II (328 +/- 39 pg/mL) after denervation. Thus, cardiac denervation may have a beneficial effect on systemic hemodynamics and coronary circulation.

Long-term prognosis of patients with variant angina

The long-term prognosis of variant angina and the factors influencing it were assessed in 217 consecutive patients hospitalized in our coronary care unit and followed for a mean of 65 months (range 2 to 123). Cardiac death occurred in 30 patients and an additional 54 experienced a nonfatal myocardial infarction. Survival at 1 and 5 years was 95% and 89%, respectively; survival without infarction was 83% and 69%. Coronary disease and the degree of disease activity were strong predictors of survival by Cox analysis. Survival at 1 year was 99%, and that at 5 years was 95% and 94%, respectively, for patients with one-vessel disease (n = 81) and for those without stenoses of 70% or greater (n = 87). Survival at 1 and 5 years was only 87% and 77% for those with multivessel disease (n = 40). The Cox analysis selected left ventricular function, initial treatment, extent score, duration of angina at rest, and disease activity as multivariate predictors of survival without infarction. Coronary disease was a strong predictor (p less than .0001) of survival without infarction by univariate analysis. Treatment with nifedipine, diltiazem, or verapamil improved survival without infarction compared with other medical treatment (p = .002). Myocardial infarction occurred most commonly soon after diagnosis in patients with a short history of angina at rest. Late coronary events were almost never preceded by resting angina.

Sympathectomy in the treatment of angina and arrhythmias

Sympathectomy has been used as treatment for several different cardiac conditions. These include classic angina pectoris, Prinzmetal's angina, paroxysmal atrial tachycardia, ventricular tachycardia, and long QT syndrome. To understand the rationale of such treatment, the innervation of the human heart is reviewed with discussion of the cardiac plexus and coronary innervation. Results in published studies are summarized and discussed.

Anatomy of human extrinsic cardiac nerves and ganglia

The anatomy of the human extrinsic cardiac nerves and ganglia was reinvestigated because descriptions of human cardiac innervation vary, detailed analyses of subhuman mammalian cardiac innervation reveal considerable similarities among species and the anatomic pattern of cardiac innervation observed in subhuman mammals differs significantly from those described for humans. The presence of a consistent pattern of cardiac innervation in subhuman mammals raised the question as to whether a similar pattern exists in humans. To investigate this, the cervical and thoracic autonomic nerves and ganglia were dissected in 13 embalmed and 10 autopsy cadavers. All major sympathetic cardiopulmonary nerves were found to arise from the stellate ganglia and the caudal halves of the cervical sympathetic trunks below the level of the cricoid cartilage. These sympathetic cardiopulmonary nerves usually consisted of 3 nerves on the right side and 4 on the left. In contrast to widely accepted reports, no sympathetic cardiopulmonary nerves were found to arise from the superior cervical ganglia or the thoracic sympathetic trunks inferior to the stellate ganglia. Parasympathetic cardiopulmonary nerves were found to arise from the recurrent laryngeal nerves and the thoracic vagi immediately distal to them. These nerves interconnected with sympathetic cardiopulmonary nerves anterior and posterior to the main pulmonary artery to form the ventral and dorsal cardiopulmonary plexuses. These plexuses contained relatively large discrete nerves as well as smaller interconnections. Emerging from these plexuses to innervate the ventricles were 3 distinct relatively large cardiac nerves, the right and left coronary cardiac nerves and the left lateral cardiac nerve. In addition to these 3 major nerves, small cardiac nerves arose from the plexuses and the thoracic vagi. Histologic examination of representative dissections confirmed the presence of neural tissue and identified the locations of neuronal cell bodies in these structures. Cell bodies were located in the nodose, superior cervical, middle cervical, stellate and thoracic sympathetic ganglia. The middle cervical ganglia varied in size and number. Neuronal cell bodies were found in the cervical and thoracic sympathetic trunks and in small mediastinal ganglia located along the courses of the cardiopulmonary and cardiac nerves. Marked similarities exist between the anatomy of the cardiopulmonary nerves and ganglia of humans and baboons.

Percutaneous transluminal coronary angioplasty for the treatment of variant angina

Among 268 patients undergoing percutaneous transluminal coronary angioplasty between February 1980 and January 1983, a total of 21 patients had variant angina, documented before angioplasty in 14 and after angioplasty in 7. Before angioplasty, all 21 patients had rest angina and 17 also had effort angina; single vessel coronary artery disease with 60 to 95% stenosis was present in all patients and the left anterior descending coronary artery was involved in all but 3 patients. Coronary angioplasty was successful in 19 patients (90%). Eight of the 19 patients remained symptom-free without coronary restenosis after successful angioplasty; in the other 11 patients, angina reappeared within 4 months, usually in association with restenosis. Of the nine patients with coronary restenosis, six had repeat angioplasty (five successful procedures and one failure), two received medical therapy and one underwent coronary bypass surgery. Patients in whom calcium channel antagonists were discontinued immediately after angioplasty had an exceedingly high coronary restenosis rate (8 [80%] of 10 successful attempts), but when calcium antagonists were continued for an average of 6 +/- 4 months after angioplasty, the restenosis rate was low (3 [21%] of 14 successful attempts). After a mean (+/- SD) follow-up period of 33 +/- 13 months, 1 patient had died and the 20 others (95%) were symptom-free; among these 20, 15 patients (75%) had been taking no antianginal drugs for more than 1 year, 2 still received calcium channel antagonists and 3 had had coronary bypass surgery. Repeat coronary arteriography performed 14 +/- 7 months after angioplasty in the 17 patients without angioplasty-related infarction or surgery showed 50% or less coronary stenosis in 13 patients.(ABSTRACT TRUNCATED AT 250 WORDS)

Factors influencing the long-term prognosis of treated patients with variant angina

To determine the prognosis of variant angina and the factors influencing it, 169 consecutive patients hospitalized in our coronary unit were followed for a mean of 15.3 months (range 1 to 68). Survival at 1, 2, and 3 years was 95%, 90%, and 87%, respectively; survival without myocardial infarction was 80%, 78%, and 75%. Twenty of the 22 myocardial infarctions and eight of the 14 deaths occurred within the first 3 months. Mantel-Haenszel log-rank analysis demonstrated that coronary disease, ventricular function, and the degree of disease activity were significant interdependent variables that influenced both survival and survival without infarction. At 1, 2, and 3 years, survival for patients with multivessel disease was 81%, 76%, and 66%; for patients with one-vessel disease, 97%, 92%, and 92%; and for patients without stenoses greater than or equal to 70%, 98% at each year (p = .0003). Survival without infarction at 1 year was 88% in patients with no stenoses greater than or equal to 70% and 82% in patients with single-vessel disease; it did not change thereafter in either group, but was 62%, 58%, and 50% at 1, 2, and 3 years in patients with multivessel disease (p = .001). Treatment did not influence survival in any subgroup (only 14 patients died overall) or survival without infarction in patients with multivessel disease. However, in patients without multivessel disease, treatment with nifedipine, diltiazem, and verapamil improved survival without infarction compared to treatment with perhexiline maleate or long-acting nitrates alone (92% vs 67% at 1, 2, and 3 years; p less than .005). Thus in addition to preventing angina, nifedipine, diltiazem, and verapamil appear to reduce complications in patients with variant angina without multivessel disease.

Intraoperative coronary artery spasm successfully treated with nitroglycerin and nifedipine

Coronary artery spasm is recognized as a major cause of angina and may coexist with fixed coronary artery disease. This report presents the case of a patient who developed recurrent coronary artery spasm in the operating room following effective coronary artery bypass grafting. The coronary artery spasm was manifested by sudden ST segment elevation, hypotension, and decreased cardiac output. Direct injection of nitroglycerin into each coronary artery graft was necessary to reverse the coronary artery spasm. Peripheral intravenous nitroglycerin plus nifedipine administered through a nasogastric tube prevented further recurrence of coronary artery spasm.

Coronary artery spasm in intact dogs induced by potassium and serotonin

Although coronary artery spasm has been implicated as an important cause of myocardial ischemia in humans, an animal model of reversible segmental coronary constriction has not been described. To provoke coronary spasm in open-chest dogs, selected vasoconstricting agents adsorbed to viscous ion exchange gels were applied topically to the surface of epicardial coronary arteries. The procedure provided a sustained localized release of drug, and minimized effects on contiguous myocardium or on the systemic circulation. Segmental arterial constrictor responses were evaluated by sonomicrometry, arteriography, and electromagnetic flow measurements. Potassium evoked sustained constrictions or spasms, and concomitantly reduced flow by -42 +/- 4% (SE; n = 34). Serotonin likewise produced sustained decreases in flow of -22 +/- 6% (SE; n = 5). Other constrictors, including norepinephrine and angiotensin, failed to evoke sustained constrictions. Spasms nearly abolished reactive hyperemic responses elicited by temporary complete occlusion of the artery. Intravenous nitroglycerin and dihydropyridine calcium antagonists promptly relieved the spasms. Scanning electronmicroscopic examination of the intimal surface of arteries undergoing sustained spasm revealed no platelet thrombi. Thus, nonthrombotic, vasodilator-sensitive segmental coronary spasms were elicited by endogenous constrictors which may play a role in regulating flow to ischemic myocardium.

Influence of partial sympathetic denervation on the results of myocardial revascularization in variant angina

Poor results of the aortocoronary bypass graft operation in the treatment of variant angina have been ascribed to recurrent vasospastic activity due to autonomic imbalance. Cardiac sympathetic denervation (plexectomy) may represent a rational approach in the prevention of vasospasm. To test the value of plexectomy in the treatment of variant angina, 31 patients were studied, 17 of whom (Group 1) underwent conventional coronary artery grafting whereas the remaining 14 (Group 2) underwent cardiac sympathetic denervation also. The 2 groups were similar with respect to age (54 +/- 8 versus 50 +/- 7 years), sex distribution (male/female ratio 12/5 versus 9/5), prevalence of coexisting effort angina (10 versus 12 patients), previous myocardial infarction (7 versus 4 patients), and duration of variant angina (3.3 +/- 5.4 versus 2.4 +/- 2.7 months). The left ventricular ejection fraction was comparable in both groups (60 +/- 11 versus 60 +/- 4%) as were left ventricular end-diastolic pressure (15 +/- 4 versus 13 +/- 5 mm Hg) and extent of coronary artery disease (65 versus 71% prevalence of multivessel disease). The average duration of follow-up was 23 +/- 15 months in Group 1 and 22 +/- 18 months in Group 2 (p = not significant [NS]). There were no operative deaths. Four patients, 2 in each group, had a perioperative myocardial infarction. Seven patients in Group 1 and 1 patient in Group 2 had recurrent variant angina. There was sudden death and 2 infarcts in Group 1. Actuarial curves showed the cumulative probability of recurrent variant angina to be significantly lower (p less than 0.05 and p less than 0.001 at 6 and 10 months, respectively) in Group 2. This study suggests that cardiac sympathetic denervation may prevent recurrent vasospastic activity in variant angina.

Prognostic significance of electrocardiographic findings in angina at rest. Therapeutic implications

Ninety-five patients with angina at rest were observed in the coronary care unit. Eighty-one per cent presented concomitantly or had previously presented some other manifestations of coronary artery disease. These patients were divided into two subgroups. In subgroup 1 (40 patients), episodes of non-exertional angina were associated with a pattern of hyperacute subepicardial injury and, frequently, with ventricular arrhythmias. In subgroup 2 (55 patients), the episodes of angina at rest were attended by horizontal ST depression, isolated T wave inversion, or trivial ST-T changes. Coronary angiographic findings were similar in both subgroups. Symptoms regressed in only 9% of patients in subgroup 1 while they were receiving beta-receptor antagonists, whereas amiodarone alone or amiodarone with nifedipine was successful in 58%. Of these patients, 25% developed a myocardial infarction shortly after admission. In subgroup 2 patients, beta-blockers were successful in 61%. Amiodarone isolated or associated with nifedipine was successful in 55% of the patients in whom it was tried. Only 5% of patients in this subgroup developed a myocardial infarction during their hospital stay. It is concluded that: (1) observation of the electrocardiogram during spontaneous angina in patients with known atherosclerotic coronary heart disease may be of prognostic significance and may influence therapeutic decision. (2) Amiodarone by virtue of its anginal and antiarrhythmic properties may be particularly useful in the treatment of non-exertional angina.

Complete denervation of the heart (autotransplantation) for treatment of severe, refractory coronary spasm

A 49 year old man had severe refractory Prinzmetal's variant angina and angiographically documented coronary arterial spasm of a dominant circumflex artery. The spasm was provoked by methergine (an ergot alkaloid) and seemed resistant to various forms of medical therapy including administration of nitrates, nifedipine, verapamil, diltiazem and amiodarone. The attacks of angina at rest persisted at the rate of 7 to 15/day and were frequently associated with atrioventricular (A-V) block. After unsuccessful plexectomy performed in another institution, the patient underwent complete cardiac denervation (produced by autotransplantation). The follow-up data have interesting implications in relation to treatment of refractory variant angina, as well as possible mechanisms of coronary arterial spasm.

Spontaneous remission in variant angina

Four cases of variant angina are reported, in which total remission of anginal pain was documented during a follow-up of seven months, four years, five years, and 15 years, respectively. During this relatively long follow-up, the clinical course of the disease was apparently benign. The possibility of spontaneously and complete recovery may be postulated. The natural history of relatively benign forms a variant angina is poorly known and understood.

Coronary artery spasm in the denervated transplanted human heart: a clue to underlying mechanisms

The mechanism of coronary artery spasm has been poorly understood but there has been some suggestion that cardiac autonomic innervation may play an important role. We report coronary artery spasm in a 43 year old man two years after he had received a transplant. Provocative pharmacologic testing suggested functional denervation of the patient's heart. Thus, coronary artery spasm can occur in the transplanted, denervated human heart. Autonomic innervation of the heart is not essential in all cases of coronary spasm, and circulating catecholamines and/or metabolic of hormonal products may play an important role.

Clinical course of patients following the demonstration of coronary artery spasm by angiography

The clinical course of 25 patients was determined during an average of 2.7 years following the angiographic demonstration of coronary artery spasm (CAS). Seventeen patients received medical treatment after the demonstration of coronary spasm and six patients had cardiac surgery. Twenty-three patients were living and two patients had died at the time of follow-up. Twenty-one of the 23 surviving patients has either no chest pain or markedly reduced symptoms. However, the demonstration of CAS by angiography was associated with a high incidence of subsequent cardiac complications, which included myocardial infarct (four patients), cardiac arrest (four patients), and death (two patients). We concluded from this study that after the demonstration of CAS by angiography: (1) the clinical course was variable, with most patients (21 of 25 patients, 84%) having improvement of symptoms at the time of follow-up; (2) major cardiac complications were frequent (11 out of 25 patients, 44%) and; (3) although clinical and coronary angiographic features were of limited use in predicting major cardiac complications, most of the patients who had an uncomplicated course (11 of 14 patients, 79%) had either less than 50% fixed coronary artery luminal diameter narrowing (CAN) or coronary artery bypass graft operations, the majority of patients with less than 50% CAN (8 of 11 patients, 73%) had no major cardiac complications, and myocardial infarction or death usually occurred during periods of increased angina pectoris.

Treatment of Prinzmetal's variant angina. Role of medical treatment with nifedipine and surgical coronary revascularization combined with plexectomy

This study describes three forms of treatment of Prinzmetal's variant angina. Coronary spasm, frequently found at coronary arteriography in patients with Prinzmetal's variant angina, can be treated with intravenous or intracoronary injection of nitroglycerin as well as of nifedipine. Nifedipine (0.2 mg) was injected directly into the involved artery in 12 patients and suppressed spasm in 9; in 3 patients, nifedipine increased coronary sinus flow, which had been decreased by spasm in one of the branches of the left coronary artery. Patients with Prinzmetal's variant angina who have spasm superimposed on atherosclerotic lesions can benefit from coronary arterial bypass grafting combined with partial denervation of the heart. This combination yielded acceptable results (83.4 percent favorable outcome); recurrence of attacks occurred in only 6.7 percent of this group. In those forms of angina in which spasm occurs in angiographically normal coronary arteries, therapy is essentially medical. In 13 patients treated with oral nifedipine (30 to 40 mg/day), suppression of attacks was achieved in 11 instances. During the period of treatment, the methergine provocative test, which had been consistently positive before treatment, converted to negative in 12 patients. Transient withdrawal of nifedipine caused recurrence of pain in two patients.

Coronary artery spasm: its role in the pathogenesis of myocardial ischemia

The role of coronary artery spasm in the production of myocardial ischemia has recently become the focus of increased attention. This phenomenon is now well established as a causative mechanism underlying the resting chest pain attacks in Prinzmetal's variant angina. There is also evidence that coronary spasm may play a more significant role in the broad spectrum of ischemic heart disease than can be documented by current techniques. The autonomous nervous system constitutes a major element in the pathophysiology of spasm. Coronary arteriography, in spite of important limitations, remains the only technique for final documentation of this phenomenon, but radionuclide scintigraphy appears to be promising. Nitroglycerin is effective for the relief of the acute attack, while long acting nitrates and the calcium antagonists: nifedipine, perhexiline and verapamil are useful in the prevention of recurrences.