Review Article Acta Cardiol Sin 2013;29:1-10

Coronary Vasospastic : Current Understanding and the Role of Inflammation

Ming-Jui Hung and Wen-Jin Cherng

Coronary vasospastic angina (CVsA) plays an important role in myocardial including stable angina, acute coronary syndromes, and sudden cardiac death. Inflammation status from either or adventitia can cause endothelial dysfunction. Thereafter, the endothelial dysfunction further induces vascular smooth muscle hypercontraction through the enhanced rho-kinase with the resultant clinical event. With better understanding of the interactions between inflammation, endothelium, and smooth muscle cells, we and other investigators have provided new insights into the basic pathophysiology of CVsA. Apart from calcium channel blockers, nitrates, and the rho-kinase inhibitor fasudil, anti-inflammatory treatment is helpful in some patients with refractory CVsA. Additional studies are needed to clarify the mechanisms of recurrent CVsA.

Key Words: Angina · Coronary · Inflammation

HISTORY OF CORONARY VASOSPASM (CVsp) angina (CVsA)”.4 Recently, many investigators found that much of the underlying cause of CVsA is associated with In 1959, Prinzmetal and his colleagues described a ST-segment depression rather than ST-segment elevation.5,6 syndrome characterized by angina at rest with transient Therefore, the term “variant angina” is usually denoted ST-segment elevation in patients with diseased coronary as angina with transient ST-segment elevation. arteries, which is different from that seen in classic an- gina.1 In these circumstances where the heart burden is not increased, the angina must be due to a reduction in CLINICAL CHARACTERISTICS OF CVsA coronary blood flow. Prinzmetal proposed the term “vari- ant angina” and suggested that it was caused by spasm of CVsA is different from typical atherosclerotic angina in a major coronary artery, because it was relieved promptly the pathophysiologies. CVsp, however, might be induced by administration of nitroglycerin. With the advent of by exercise, particularly in the morning in some patients coronary angiography, it became apparent that variant with variant angina,7 and might cause exercise-induced angina was caused by CVsp and it may occur at the site of angina with ST-segment depression in some patients with a coronary stenosis2 or in normal coronary arteries, the stable effort angina5 (Figure 1). The authors postulated that so-called “variant of the variant”3 or “coronary vasospastic the spastic arteries are not normal, in that the spastic ar- teries cannot dilate fully in response to exercise as in nor- mal coronary arteries. There are daily, weekly, monthly, Received: June 28, 2012 Accepted: September 19, 2012 and circadian variations in the incidence of CVsA.8 Department of Cardiology and Medical Research Center, Chang Gung Memorial Hospital, Keelung, Chang Gung University College of Medicine, Taiwan. Circadian variation in the incidence of attacks in Address correspondence and reprint requests to: Dr. Ming-Jui Hung, patients with CVsA Department of Cardiology, Chang Gung Memorial Hospital, No. 222, Maijin Road, Keelung 20401, Taiwan. Tel: 886-2-2431-3131 ext. CVsA occurs usually at rest, particularly from mid- 9 10 3168; Fax: 886-2-2433-5342; E-mail: [email protected] night to early morning. Yasue et al. compared coro-

1 Acta Cardiol Sin 2013;29:1-10 Ming-Jui Hung et al.

AB

C D Figure 1. Treadmill exercise electrocardiograms (upper panels) and the coronary angiograms (lower panels) of a patient who had stable angina pectoris. Resting (A) and peak exercise (B) status of 12-lead electrocardiograms show positive for stress-induced ischemia in the V3-6, I, and aVL leads. The coronary angiograms reveal intracoronary methylergonovine-induced diffuse diameter reduction > 70% in the mid-to-distal portion of left ante- rior descending coronary artery (C, arrows), which is relieved after intracoronary nitroglycerin 100 mg administration (D, arrows). nary arteriograms recorded in the early morning with grams and the coronary arteriograms taken after nitro- those recorded in the afternoon in patients with variant glycerin administration. The percentage increase in di- angina. In the early morning, the tone of the major cor- ameter of the major coronary artery after nitroglycerin onary artery was increased and its diameter was small. administration was significantly greater in the early Under such conditions, mild exercise could induce CVsp morning than in the afternoon. This may be one of the resulting in attacks; the administration of nitroglycerin reasons that there is a circadian variation in the exercise dilated the artery markedly. In contrast, in the after- capacity of most patients with variant angina. A pathol- noon, the major coronary artery was usually already di- ogy study illustrated the complexity of the local neural lated, and its tone was low on the control coronary ar- events that modulate the tone of the coronary arter- teriograms. Under such conditions, exercise could in- ies.11 The occurrence of CVsA in the early morning has duce little coronary vasoconstriction and no attacks been noted to be associated with rapid eye move- usually occurred except in patients with severe organic ment.12 Therefore, a rapid elevation of sympathetic ac- stenosis, in whom only a mild spasm could occlude the tivity during augmented parasympathetic activity has artery and result in angina attacks. To record quantita- been suggested to be related to the occurrence of CVsA tively the difference in the tone in the major coronary in the early morning. arteries observed in the early morning and in the after- noon, Yasue et al.10 measured the diameter of the major coronary artery on both the control coronary arterio- There is general consensus in the medical commu-

Acta Cardiol Sin 2013;29:1-10 2 Inflammation and Coronary Vasospasm nity that intracoronary thrombus plays a major role in several agents or procedures, including ergonovine the pathogenesis of acute . Yasue maleate, methylergonovine maleate, acetylcholine or et al.13 found that CVsp was presumed to be responsible hyperventilation reported that induce CVsp in patients for the acute myocardial infarction because the culprit with CVsA. Intracoronary ergonovine administration has artery was patent without delay of visualization in been a popular method to induce CVsp during angio- 17.9% of patients in the early phase of acute myocardial graphic study because of its high sensitivity and specific- infarction. Our report also found that the responsible ity.18 This test was administrated using a step-wise dose coronary arteries were patent in 12% of patients with of ergonovine (1, 5, 10, and 30 mg) every 3 minutes; acute myocardial infarction (Figure 2).14 There was in- farct-related CVsp involvement in 95% in these patients. This may be explained by the spontaneous resolution of either spasm or thrombus, or both. Oshima et al.15 re- ported that CVsp causes intracoronary thrombus forma- tion, supporting the concept that CVsp is one of the pri- mary factors contributing to acute myocardial infarc- A B tion. Therefore, CVsp appears to play a role in the pro- duction of acute myocardial infarction in these patients. Most of the patients with CVsA present as angina with ST-segment depression and/or T-wave inversion on electrocardiogram, which is an acute coronary syn- C D drome. If there is no cardiac enzyme elevation, we de- fine the cardiac event as an because most of the CVsA occurs at rest. Angiographically normal or near-normal coronary arteries occurs in 25% of pa- tients with acute coronary syndrome irrespective of the 16,17 provocative agents. The CVsp can be induced in E F 50-60% of these patients. After initial management (i.e. oxygen, aspirin, nitroglycerin, and/or morphine) for acute coronary syndrome, follow-up electrocardiograms areimportanttoindicatetheroleofCVspfortheacute coronary syndrome. If there is a normalized ST-segment after the initial management, the CVsp may play a role G H in acute coronary syndrome. Figure 2. A serial 12-lead electrocardiograms and the coronary angiograpms of a patient who had coronary vasospasm-induced acute Intracoronary administration of methylergonovine myocardial infarction. The patient had normal coronary angiograms 1 year before acute myocardial infarction. Initial resting 12-lead electro- Provocative testing for CVsp is required to clarify its cardiograms at emergency department showed complete atrioven- role in the pathogenesis of angina pectoris, especially in tricular block with ST-elevation in the II, III, aVF and V1-4 (A) and right patients without significant obstructive coronary artery ventricular infarction (B). The emergency coronary angiogram revealed disease (CAD). To ensure a valid provocative test, vasodi- total occlusion of right coronary artery since the orifice (C, arrow) inititally. After intra-aortic nitroglycerin 200 mg administration, patent lators (calcium antagonists and nitrates) must be with- proximal and mid-portion of right coronary artery with thrombi-oc- drawn for at least 24 hours, except sublingual nitroglyc- cluded (D, arrow) in the distal portion of right coronary artery was erin if necessary. The nitroglycerin solution must be well noted. A thrombuster-catheter (E, arrow) was used to aspirate thrombi. prepared before starting intracoronary methylergo- After thrombi suction and further intracoronary 200 mg nitroglycerin ad- ministration, the right coronary artery was patent (F). The 12-lead elec- novine testing, in part to abolish intracoronary me- trocardiograms recorded at the day after coronary intervention (G) and thylergonovine-induced CVsp immediately through the 6 months later (H) showed evolutional electrocardiographic ST-segment intracoronary route (50-1000 mg). There have been and T-wave changes of inferior myocardial infarction.

3 Acta Cardiol Sin 2013;29:1-10 Ming-Jui Hung et al. the drug was first introduced into the right coronary ar- our prior study.24 They concluded that spasm provoca- tery and subsequently into the left coronary artery. tion tests have an acceptable level of safety; the evalua- CVsp was defined as a decrease of > 50%19 or 70%20 in tion of spasm type may provide useful information for the diameter of an arterial lumen with concurrent chest the risk stratification of CVsA patients. Consistent with pain and/or ischemic ST-T changes during the provoca- the Japanese study, there was no mortality reported in tion testing. Recently, the Japanese Circulation Society our prior study.24 Because ventricular or published a guideline to define a positive indicator on a ventricular is a possible complication follow- provocative test as a decrease of > 90% in the diameter ing intracoronary methylergonovine administration, its of an arterial lumen, with concurrent and/or use outside the laboratory is not ischemic ST-segment changes during the provocation recommended. test.4 However, Yasue et al.8 suggested that there are no In theory, the diagnosis of CVsp must be made on limits on the degree of lumen diameter reduction re- the coronary angiographic findings during the attack. quired to diagnose CVsp since ischemia must accom- However, it is not practical to perform coronary angio- pany the changes of vessel size in a period of time. Al- graphyduringanattackineverypatient,andthisstepis though there are different criteria in vessel diameter, unnecessary. During coronary angiography, adequate the angina and/or ischemic electrocardiographic changes doses (50-1000 mg) of intracoronary nitroglycerin ad- during provocation testing are necessary in defining a ministration help cardiologists differentiate spontane- positive provocation test result. After a CVsp had been ous CVsp from fixed obstructive CAD. Angina pectoris diagnosed, the administration of intracoronary ergo- that is relieved promptly after nitroglycerin may be diag- novine was stopped and 50-1000 mg of intracoronary ni- nosed as CVsA even without angiographic evidence, if troglycerin was administered. The ergonovine maleate one of following characteristics is noticed:8 1) the attack and acetylcholine are not available in Taiwan. Therefore, occurs at rest, particularly from midnight to early morn- only methylergonovine maleate was used in the CVsp ing; 2) there is marked circadian variation of exercise ca- provocative testing in our studies, with the same in- pacity, the attack easily induced by exercise in the tracoronary dose regimen as in ergonovine maleate. Al- morning but not by even vigorous exercise in the after- though it is useful in the diagnosis of CVsA, safety is still noon; 3) the attack is associated with ST-segment eleva- a major concern. The contraindications for intraco- tion on the electrocardiogram; 4) the attack is induced ronary methylergonovine testing included pregnancy, by hyperventilation; or 5) the attack is suppressed by severe hypertension (systolic blood pressure > 180 mm calcium antagonists, but not by beta-blockers. Hg), moderate to severe and uncon- trolled ventricular .19 Some studies have re- Prevalence of CVsp ported complications of intravenous injection of er- In the Japanese population, CVsp frequency appears gonovine in patients who underwent cardiac cathe- to be greater than that in Western populations,25 and terization, including , ventricular the diagnosis of variant angina (angina with transient fibrillation, and death.21,22 Theimportanceofintra- ST-segment elevation) is made in a high percentage coronary nitroglycerin rather than intravenous or sub- (10% to 70%) of patients with anginal symptoms re- lingual route in relieving provocative CVsp is empha- ferred to Japanese medical centers. Our prior investiga- sized.22 Since Hackett et al.18 introduced the intra- tion also found that 27% of angina patients who under- coronary route for provocation testing in 1987, the gen- went coronary angiography and had CVsA (angina with eral consensus has been that the intracoronary rather and without transient ST-segment elevation).26 In pa- than the intravenous route is safer because of negligible tients who had angina and no significant obstructive drug recirculation and avoidance of effects on branches CAD, the prevalence of CVsp was around 50%.27 In pa- with critical stenosis.19 Recently, a Japanese multicenter tients who had acute coronary syndrome and no signifi- study found that intracoronary ergonovine provocation cant obstructive CAD, the prevalence of CVsp was esti- testing induces ventricular tachycardia/ventricular fibril- mated to be 57%,17 which is similar to data from Ger- lation in 0.8% of patients with CVsA,23 which is similar to many.16 The diagnosis of CVsp depends on angiographic

Acta Cardiol Sin 2013;29:1-10 4 Inflammation and Coronary Vasospasm protocols and provocation tests, which varies from labo- model in women provide evidence that more smokers in ratory to laboratory. Premedication with spasmolytic men (life-style) and age (induction time) contribute to drugs such as nitroglycerin or calcium antagonists, the natural history of CVsp development. In addition, avoidance of coronary constrictors, and daily or monthly we also found that hypertension is a negative predictor variation of disease activity may lead to a failure to diag- for CVsp, which suggests that the pathogenesis of CVsp nose CVsp. Therefore, an estimate of the prevalence of differs from that of coronary atherosclerosis. CVsp in different populations remains to be defined. Al- though the racial difference in coronary constrictor re- Treatment and prognosis of CVsA sponse has been strongly suggested,25 the underestima- Table 1 depicts the treatment strategies for CVsA tion of CVsp worldwide is an important issue in the cur- patients. Calcium antagonists play a cornerstone role in rent percutaneous coronary intervention era. the management of CVsA.8 Of importance, the long-act- ing calcium antagonist should be given before going to Risk factors for the development of CVsA bed at night as CVsA occurs usually from midnight to the Cigarette smoking was found to be an unequivocal early morning.8 Usually, a high-dose long-acting calcium risk factor for CVsA in the literature (adjusted odds ratio antagonist (e.g. nifedipine 80 mg/day, amlodipine 20 of 2.41).28 Similarly, our investigation29 has found a mg/day, diltiazem 360 mg/day, or verapamil 480 mg/ multivariate-adjusted odds ratio of 2.58, which is similar day) is started.31 It may require 2 calcium antagonists to other studies. In our recent study, synergistic interac- (dihydropyridine and non-dihydropyridine) to relieve tion was verified between smoking and high-sensitivity CVsA. While long-acting nitrates are also effective in C-reactive protein (hs-CRP) in the development of preventing CVsA attack, the occurrence of nitrate toler- CVsp.29 This interaction was linear and monotonic among ance may limit their use as a first-line approach.8,32 In smokers. In nonsmokers, however, hs-CRP had a th- contrast, b-blockers do not suppress, but rather may ag- resholdeffectonCVspdevelopment.Adecreasedodds gravate CVsp in patient with variant angina.32 Although ratio to 2.12 was observed when hs-CRP was added into platelets may play a role in precipitating or aggravating the multivariate analysis between smoking and CVsp, CVsp, intravenous prostacyclin or cyclooxygenase inhibi- suggesting that hs-CRP was an important covariate of tors have thus far not been shown to have beneficial ef- CVsp. Further analysis showed that the relationship be- fects.32 Recent clinical research shows that magnesium, tween hs-CRP and CVsp differs between men and wo- antioxidants, rho-kinase (ROCK) inhibitor fasudil, and men.30 The non-threshold model in men and a threshold fluvastatin are also beneficial to treat CVsA.8 Addition-

Table 1. Treatment strategies for CVsA Strategy Recommendation Cigarette cessation Obligatory Long-acting calcium antagonist 1. Use before going to bed at night 2. Combination of dihydropyridine and non-dihydropyridine if recurrent CVsA Long-acting nitrates Decreased potency by tolerance Statin Fluvastatin Magnesium Intravenous infusion RhoA/ROCK inhibitor Intravenous or oral Fasudil Vitamin C Intravenous infusion Vitamin E Oral Valsartan A case report Prednisolone Case reports Coronary artery bypass surgery Controversial Coronary stenting Controversial Implantable cardioverter defibrillator For life-threatening ventricular CVsA, coronary vasospastic angina; ROCK, Rho-associated coiled-coil containing protein kinase.

5 Acta Cardiol Sin 2013;29:1-10 Ming-Jui Hung et al. ally, the aggravating factors for CVsA such as cigarette The prognosis among Japanese patients with variant smoking, catecholamines, muscarinic agonists, ergot al- angina is better than that among western patients.40,41 kaloids, prostaglandins, alcohol, emotional stress, and The difference is probably due to the fact that the per- propranolol must be avoided. Coronary intervention centage of patients with multivessel disease or impaired plays a limited role in patients with CVsA and organic left ventricular function or both was smaller, and the stenosis.33 In the current coronary stenting era, there percentage of patients who received a calcium antago- are some reports of coronary stenting in the manage- nist (diltiazem or nifedipine) as the initial treatment was ment of drug-refractory CVsA.34-36 Gaspardone et al.34 higher in the Japanese series.40 By multivariate analysis, reported that coronary stent placement is effective in the intake of calcium antagonists, extent and severity of preventing angina attacks and methylergonovine-in- , and multivessel CVsp were duced CVsp in 6 of 9 patients 6 months after stenting. shown to be significantly independent predictors of sur- They performed coronary stent placement in CVsA pa- vival without myocardial infarction. In Taiwanese pa- tients because of persistent angina attacks despite med- tients who had CVsA and no significant obstructive CAD, ical treatment (up to 960 mg diltiazem or 100 mg the long-term prognosis was found to be good despite nifedipine and nitrates). Other reports also showed suc- 11% of recurrent CVsA.42 The most common factors for cessful coronary stenting in the management of CVsA. recurrent CVsA are continued smoking and self-discon- However, their reports did not use the combination and tinuation of calcium antagonists. The occurrence of an- maximal doses of 2 calcium antagonists in terms of ag- gina attacks is difficult to assess in individuals with CVsA gressive medical therapy for CVsA.35,36 In theory, the cor- because their frequency tends to fluctuate spontane- onary stenting is considered to be effective to abolish ously and the attacks are not necessarily accompanied the CVsA in the absence of concomitant calcium antago- by symptoms. Based on the observation of recurrent an- nist treatment after coronary stenting. Calcium antago- gina pectoris, it appears reasonable to suggest that nists were still used after coronary stenting in these re- treatment should not be discontinued in cases of CVsA, ports indicating that single-vessel or multi-vessel CVsp even when the patients are asymptomatic.42,43 may develop at sites different from the initial stenosis. Therefore, Yuksel et al.37 suggested that before coronary stenting can be advocated as an established treatment INFLAMMATION AND CVsA resistant to maximal medical therapy, a randomized trial should be carried out. Nevertheless, approximately 20% Determinants of CVsp: endothelium and smooth of CVsA patients do not respond to treatment with 2 cal- muscle cium antagonists plus a long-acting nitrates.31 Under the Thereisnosinglemechanismthatcanexplainthe circumstances, coronary stenting in combination with entire process of CVsA. In 1980s, the autonomic nervous adequate coronary vasodilators doses may be helpful in system was found to play an important role in the patients CVsA and severe organic stenosis.4,31,32,38 On pathophysiology of CVsA.8 In the 1990s, oxidative stress, the other hand, coronary interventions are contraindi- deficiency of nitric oxide activity, respiratory alkalosis, cated in patients with CVsA without severe organic ste- magnesium deficiency and insulin deficiency were iden- nosis.4,31,38 As coronary arteries treated with coronary tified as other possible mechanisms for CVsA.8 In the interventions show time-dependent loss of endothe- late 1990s and early 2000s, mutation of the promoter in lial-dependent and -independent vasomotor function, endothelial nitric oxide synthase gene8 and polymor- with imbalanced contraction/dilation capacity,39 the phism of paraoxonase gene44 were found to be associ- suggestion has been that cardiologists must keep CVsA ated with CVsA. In another large Japanese cohort in mind before performing coronary intervention.8,34,37 study,45 The nicotinamide adenine dinucleotide hydro- An implantable cardioverter defibrillator with aggressive gen/nicotinamide adenine dinucleotide phosphate hy- medical therapy for CVsA was reported to be effective in drogen oxidase p22 phox gene is a susceptibility locus patients who had syncope, ventricular tachycardia, or forCVspinmen,andthestromelysin-1andinterleukin- survived outside the hospital.4 6 genes are susceptibility loci in women. In this study,

Acta Cardiol Sin 2013;29:1-10 6 Inflammation and Coronary Vasospasm the inflammation gene was firstly identified to be as- arteries are inflammatory and are supersensitive to sociated with CVsp. These findings indicate that dys- both vasoconstrictor agonists and nitrates. functional endothelium secondary to down-regulation of endothelial nitric oxide synthase and oxidative inacti- Role of inflammation vation of nitric oxide is one of the major mechanisms re- In 1978, Lewis et al.52 first reported a patient who sponsible for CVsA. In addition, inflammation is indi- died of cardiogenic shock due to variant angina and lo- cated as a possible contributor in the development of calized . They postulated that there was a CVsA. However, a deficiency of endothelial nitric oxide link between inflammation and CVsA. In the mid and may not explain the complete mechanism of CVsA be- later 2000s, we showed that chronic inflammation was cause all atherosclerotic coronary arteries are not nec- associated with CVsA, as evidenced by elevated peri- essarily associated with CVsp in spite of the deficiency pheral leukocyte and monocyte counts, hs-CRP, inter- of nitric oxide activity. In 2006, Kakket et al.46 found that leukin-6, and adhesion molecules.27 It was demon- spontaneous CVsp occurs in KATP mutant mice, which strated that the serum level of cortisol, one of the im- arise from a smooth muscle-extrinsic process. They pos- portant stress hormones, causes coronary hyperre- 53 tulated that endothelial dysfunction with loss of KATP activitythroughactivationofROCKinpigsinvivo. channels and decreased nitric oxide production and/or Cigarette smoking, a major risk factor for CVsA, is as- bioavailability promotes smooth hypercontraction. Ano- sociated with low-grade inflammation.54 These findings ther possibility includes the sympathetic neurons, suggest that there is increased inflammatory status in where opening of presynaptic KATP channels decreases patients with CVsA and the inflammation may con- norepinephrine release, enhancing smooth muscle relax- tribute to the occurrence of CVsp. An interaction be- ation to dilate coronary arteries. A defect in these chan- tween smoking and hs-CRP was recently reported by nels decreasing the threshold for norepinephrine re- our group and the relationship between hs-CRP and lease might be associated with CVsp. In 2011, Qipshidz CVsA differed between men and women.30 We also et al.47 found that folic acid treatment attenuates acetyl- have recently demonstrated that ROCK activity in pe- choline-induced coronary vasoconstriction in hyper- ripheral leukocyte independently predicts the presence homocysteinemic cystathionine beta synthase hyter- and severity of CVsA and the level of ROCK activity cor- zygote mice. They suggested that CVsp is related to the relates with plasma interleukin-6.50 These findings sug- regulation of endothelial nitric oxide synthase expres- gest that there is increased inflammatory status in pa- sion, nitric oxide availability, and tissue homocysteine. tients with CVsA and the inflammation may cause endo- In addition to deficient endothelial nitric oxide ac- thelial dysfunction, ROCK activity accentuation and fi- tivity, hyperreactivity of the coronary smooth muscle nally contribute to the vascular smooth muscle hyper- seems to play an important role in the pathogenesis of contraction, the occurrence of CVsp. CVsA.48 Shimokawa and colleagues developed swine Some investigators found that inflammatory clinical models of CVsp and showed that ROCK activity is en- condition is associated with CVsA.8 Several reports have hanced in coronary artery smooth muscle after wrap- suggested a possible link between allergic disease and ping the coronary artery with interleukin-1 beads.48 CVsA.55,56 The pathogenesis of bronchial asthma has They also demonstrated that ROCKs expression and ac- been recently attributed to hyperreactivity of the air- tivity are enhanced at the inflammatory/arteriosclerotic way caused by inflammation, and corticosteroids are coronary lesions.49 Recently, we50 found that ROCK ac- considered to work by alleviating that inflammation.57 tivity in circulating neutrophils is a useful biomarker for There is an analogy with the pathophysiology of CVsA the diagnosis and disease activity assessment in patients and bronchial asthma; that is CVsp may be induced by with CVsA. Some investigators found that decreased en- arterial hyperreactivity caused by local inflammation in dothelial nitric oxide synthase activity increase ROCK ac- the coronary arterial wall and corticosteroids suppress tivity in coronary arteries.51 These findings connect the the hyperreactivity by alleviating the inflammation in activity of ROCK to endothelial nitric oxide and are in the vessel wall. In fact, Forman et al.58 reported a pa- agreement with the clinical observations that spastic tient with CVsA complicated by sudden death in whom

7 Acta Cardiol Sin 2013;29:1-10 Ming-Jui Hung et al. mast cell infiltration was found at the site of angio- dence suggests that inflammation substantially con- graphic documentation of CVsp, which indicates that tributes to the development of CVsA. Clinical studies CVsp can also occur as the result of adventitial inflam- with antiinflammtory therapies further support the role mation in addition to endothelial inflammation. Some of inflammation in CVsA. Release of inflammatory medi- other studies also reported that focal infiltration of in- ators, such as interleukin-6, histamine, or serotonin flammatory cells was seen in the adventitia or plaque of leads to nitric oxide availability reduction, endothelial the coronary artery in patients with CVsA.59 Even where dysfunction, and precursor medial contraction. The final there is no evident narrowing during angiography, dif- common pathway in CVsp is vascular smooth muscle fuse intimal thickening or intimal bump by precursor contraction by increased ROCK activity. With the under- medial contraction is demonstrated from optical coher- standing of interactions between inflammation, endo- ence tomography.60,61 Nitric oxide availability is reduced thelium, and smooth muscle cells, we and other investi- under the circumstances of low-grade inflammation of gators have provided some insights into the basic patho- atherosclerosis or diffuse intimal thickening.62 Endothe- physiology of CVsp. More studies are needed to clarify lial nitric oxide synthase has been reported to be quan- the mechanisms of recurrent CVsA. titatively associated with caveolin-1 in endothelial cells. The vascular relaxation in response to acetylcholine was also much larger in aortic rings collected from caveolin- REFERENCES 1–/– mice than in their wild-type littermates.63 Of note, isolated aortic rings from caveolin-1–/– mice were unable 1. Prinzmetal M, Kennamer R, Merliss R, et al. Angina pectoris I. A to maintain a constant contractile tone, oscillating at 1 variant form of angina pectoris; preliminary report. Am J Med Hz frequency. These studies support the proposition 1959;27:375-88. 2. MacAlpin RN, Kattus AA, Alvaro AB. Angina pectoris at rest with that, in the basal condition, endothelial nitric oxide preservation of exercise capacity: Prinzmetal’s variant angina. synthase becomes hyperactivated in the absence of Circulation 1973;47:946-58. caveolin-1. Based on the above analyses, it appears rea- 3. Cheng TO, Bashour T, Kelser GA Jr, et al. Variant angina of sonable to speculate that CVsp is an early inflammatory Prinzmetal with normal coronary arteriograms. A variant of the coronary artery condition because of the presence of variant. Circulation 1973;47:476-85. low-grade inflammation-related endothelial dysfunction 4. JCS Joint Working Group. Japanese Circulation Society of Guide- with resulting diffuse intimal thickening and impaired ni- lines for Diagnosis and Treatment of Patients with Vasospastic Angina (Coronary Spastic Angina) (JCS 2008): digest version, Cir J tric oxide production.64 Because the lack of normal nitric 2010;74:1745-62. oxide activity does impact the normal vessel, the vessel 5. Cheng CW,Yang NI, Lin KJ, et al. Role of coronary spasm for a pos- tends to constrict in the presence of low-grade inflam- itive noninvasive stress test result in angina pectoris patients mation such as cigarette smoking, stress, airway dis- without hemodynamically significant coronary artery disease. ease, and allergy. Am J Med Sci 2008;335:354-62. 6. Nakagawa H, Morikawa Y, Mizuno Y, et al. Coronary spasm preferentially occurs at branch points: an angiographic com- parison with atherosclerotic plaque. Circ Cardiovasc Interv CONCLUSION 2009;2:97-104. 7. Hung MJ, Hung MY,Cheng CW,et al. Clinical characteristics of pa- Identification of CVsp is important in our daily tients with exercise-induced ST-segment elevation without prior clinical practice because of a wide disparity between myocardial infarction. Circ J 2006;70:254-61. treatment strategies for fixed obstruction versus vaso- 8.YasueH,NakagawaH,ItohT,etal.Coronaryarteryspasm-- spasm of coronary arteries. Adequate doses of intra- clinical features, diagnosis, pathogenesis, and treatment. J coronary nitroglycerin administration help cardiologists Cardiol 2008;51:2-17. 9. Liu PC, Cheng CW, Hung MJ, et al. Variant angina with angio- differentiate spontaneous CVsp from fixed obstructive graphically normal or near-normal coronary arteries: a 10-year CAD. However, cessation of cigarette smoking and experience. J Int Med Taiwan 2010;21:79-89. adequate dose and timing of calcium antagonists are 10. Yasue H, Omote S, Takizawa A, et al. Coronary arterial spasm in the cornerstone for CVsA therapy. Accumulating evi- ischemic heart disease and its pathogenesis. A review. Circ Res

Acta Cardiol Sin 2013;29:1-10 8 Inflammation and Coronary Vasospasm

1983;52:I147-52. 25. Beltrame JF, Sasayama S, Maseri A. Racial heterogeneity in coro- 11. Jougasaki M, Yasue H, Takahashi K. Perivascular nerve lesion of nary artery vasomotor reactivity: differences between Japanese the coronary artery involved in spasm in a patient with variant and Caucasian patients. J Am Coll Cardiol 1999;33:1442-52. angina. Pathology 1989;21:304-7. 26. Hung MJ, Cherng WJ, Yang NI, et al. Relation of high-sensitivity 12. Otsuka K, Yanaga T, Watanabe H. Variant angina and REM sleep. C-reactive protein level with coronary vasospastic angina pec- Am Heart J 1988;115:1343-6. toris in patients without hemodynamically significant coronary 13. Yasue H, Ogawa H, Okumura K. Coronary artery spasm in the gen- artery disease. Am J Cardiol 2005;96:1484-90. esis of myocardial ischemia. Am J Cardiol 1989;63:29E-32E. 27. Hung MJ, Cherng WJ, Cheng CW, et al. Comparison of serum lev- 14. Hung MJ, Cherng WJ. Comparison of white blood cell counts in elsofinflammatorymarkersinpatientswithcoronaryvaso- acute myocardial infarction patients with significant versus in- spasm without significant fixed coronary artery disease versus significant coronary artery disease. Am J Cardiol 2003;91: patients with stable angina pectoris and acute coronary syn- 1339-42. dromes with significant fixed coronary artery disease. Am J 15. Oshima S, Ogawa H, Yasue H, et al. Increased plasma fi- Cardiol 2006;97:1429-34. brinopeptide A levels during attacks induced by hyperventilation 28. Sugiishi M, Takatsu F. Cigarette smoking is a major risk factor for in patients with coronary vasospastic angina. J Am Coll Cardiol coronary spasm. Circulation 1993;87:76-9. 1989;14:150-4. 29. Hung MY, Hsu KH, Hung MJ, et al. Interaction between cigarette 16. Ong P,Athanasiadis A, Hill S, et al. Coronary artery spasm as a fre- smoking and high-sensitivity C-reactive protein in the develop- quent cause of acute coronary syndrome: the CASPAR (Coronary ment of coronary vasospasm in patients without hemodyna- Artery Spasm in Patients with Acute Coronary Syndrome) Study. J mically significant coronary artery disease. Am J Med Sci Am Coll Cardiol 2008;52:523-7. 2009;338:440-6. 17. Hung MJ, Cheng CW, Yang NI, et al. Coronary vasospasm-induced 30. Hung MY, Hsu KH, Hung MJ, et al. Interactions among gender, acute coronary syndrome complicated by life-threatening car- age, hypertension and C-reactive protein in coronary diac arrhythmias in patients without hemodynamically signifi- vasospasm. Eur J Clin Invest 2010;40:1094-103. cant coronary artery disease. Int J Cardiol 2007;117:37-44. 31. de Lemons JA, O’Rourke RA. Unstable angina and non-ST-seg- 18. Hackett D, Larkin S, Chierchia S, et al. Induction of coronary ment elevation myocardial infarction. In: Fuster V, Eds. Hurst’s artery spasm by a direct local action of ergonovine. Circulation the Heart.12th ed. New York: McGraw-Hill; 2008: 1371-2. 1987;75:577-82. 32. Pinto DS, Beltrame JF, Crea F. Variant angina. Available at: http:// 19. Scanlon PJ, Faxon DP, Audet AM, et al. ACC/AHA guidelines for www.uptodate.com/contents/variant-angina?source=search_ coronary angiography. A report of the American College of Cardi- result&search=variant+angina&selectedTitle=1%7E103. ology/American Heart Association Task Force on practice guide- Accessed August 24, 2012. lines (Committee on Coronary Angiography). Developed in col- 33. Tanabe Y, Itoh E, Suzuki K, et al. Limited role of coronary laboration with the Society for Cardiac Angiography and Inter- angioplasty and stenting in coronary spastic angina with organic ventions. J Am Coll Cardiol 1999;33:1756-824. stenosis. J Am Coll Cardiol 2002;39:1120-6. 20. Baim DS. Coronary angiography. In: Baim DS, eds. Grossman’s 34. Gaspardone A, Tomai F, Versaci F, et al. Coronary artery stent Cardiac Catheterization, Angiography, and Intervention.7th ed. placement in patients with variant angina refractory to medical Philadelphia: Lippincott Williams & Wilkins; 2006:215. treatment. Am J Cardiol 1999;84:96-8. 21. Bertrand ME, LaBlanche JM, Tilmant PY, et al. Frequency of 35. Chou HH, Lim KE, Ko YL. Treatment of spontaneous left main cor- provoked coronary arterial spasm in 1089 consecutive patients onary artery spasm with a drug-eluting stent. Acta Cardiol Sin undergoing coronary arteriography. Circulation 1982;65:1299- 2009;25:43-6. 306. 36. Chou MT, Huang TY. Coronary stenting for coronary vasospasm 22. Buxton A, Goldberg S, Hirshfeld JW, et al. Refractory ergonovine- complicated with refractory ventricular tachycardia and fibril- induced coronary vasospasm: importance of intracoronary nitro- lation. Acta Cardiol Sin 2012;28:145-7. glycerin. Am J Cardiol 1980;46:329-34. 37. Yuksel UC, Celik T, Iyisoy A, et al. Polymorphic ventricular tachy- 23. Takagi Y, Yasuda S, Takahashi J, et al. Clinical implications of prov- cardia induced by coronary vasospasm: a malignant case of vari- ocation tests for coronary artery spasm: safety, arrhythmic com- ant angina. Int J Cardiol 2007;121:210-2. plications, and prognostic impact: Multicentre Registry Study of 38. Cannon CP, Braunwald E. Unstable angina and non-ST elevation the Japanese Coronary Spasm Association. Eur Heart J 2012 July myocardial infarction. In: Libby P, Eds. Braunwald’s Heart Dis- 10 [E-pub ahead of print], doi: 10.1093/eurheartj/ ehs199. ease: A Textbook of Cardiovascular Medicine.8th ed. Philadel- 24. Hung MY, Hung MJ, Cheng CW, et al. Safety and predictors of a phia: Saunders, 2008:1339. positive result of intracoronary ergonovine testing in patients 39. Plass CA, Sabdyusheva-Litschauer I, Bernhart A, et al. Time with ischemic heart disease without hemodynamically signifi- course of endothelium-dependent and -independent coronary cant coronary artery stenosis in Taiwan. Acta Cardiol Sin 2007; vasomotor response to coronary balloons and stents: compari- 23:150-9. son of plain and drug-eluting balloons and stents. JAmColl

9 Acta Cardiol Sin 2013;29:1-10 Ming-Jui Hung et al.

Cardiol Intv 2012;5:741-51. 52. Lewis JR, Kisilevsky R, Armstrong PW. Prinzmetal’s angina, nor- 40. Yasue H, Takizawa A, Nagao M, et al. Long-term prognosis for pa- mal coronary arteries and pericarditis. Can Med Assoc J tients with variant angina and influential factors. Circulation 1978;119:36-9. 1988;78:1-9. 53. Hizume T, Morikawa K, Takaki A, et al. Sustained elevation of se- 41. Walling A, Waters DD, Miller DD, et al. Long-term prognosis of rum cortisol level causes sensitization of coronary vasocon- patients with variant angina. Circulation 1987;76:990-7. stricting responses in pigs in vivo: a possible link between stress 42. Hung MJ, Hung MY, Cheng CW, et al. Comparison of clinical char- and coronary vasospasm. Circ Res 2006;99:767-75. acteristics and prognosis in Taiwanese patients with coronary 54. Yasue H, Hirai N, Mizuno Y, et al. Low-grade inflammation, vasospastic angina pectoris without significant fixed coronary ar- thrombogenicity, and atherogenic lipid profile in cigarette tery disease versus patients with significant fixed coronary artery smokers. Circ J 2006;70:8-13. disease and either stable angina pectoris or acute coronary syn- 55. Kounis NG, Zavras GM. Histamine-induced coronary artery dromes. Am J Med Sci 2007;334:160-7. spasm: the concept of allergic angina. Br J Clin Pract 1991; 43. Ueda O, Kohchi K, Kishi Y, et al. Long lasting spasticity in con- 45:121-8. trolled vasospastic angina. Heart 1999;81:528-32. 56. Chang PH, Hung MJ, Yeh KY, et al. Oxaliplatin-induced coronary 44. Ito T, Yasue H, Yoshimura M, et al. Paraoxonase gene Gln192Arg vasospasm manifesting as Kounis syndrome: a case report. J Clin (Q192R) polymorphism is associated with coronary artery Oncol 2011;29:e776-8. spasm. Hum Genet 2002;110:89-94. 57. Busse WW, Lemanske RF Jr. Asthma. N Engl J Med 2001; 45. Murase Y, Yamada Y, Hirashiki A, et al. Genetic risk and gene- 344:350-62. environment interaction in coronary artery spasm in Japanese 58.FormanMB,OatesJA,RobertsonD,etal.Increasedadventitial men and women. Eur Heart J 2004;25:970-7. mast cells in a patient with coronary spasm. N Engl J Med 46. Kakkar R, Ye B, Stoller DA, et al. Spontaneous coronary vaso- 1985;313:1138-41. spasm in KATP mutant mice arises from a smooth muscle- 59. Kohchi K, Takebayashi S, Hiroki T, et al. Significance of adventitial extrinsic process. Circ Res 2006;98:682-9. inflammation of the coronary artery in patients with unstable 47. Qipshidze N, Metreveli N, Lominadze D, et al. Folic acid improves angina: results at autopsy. Circulation 1985;71:709-16. acetylcholine-induced vasoconstriction of coronary vessels iso- 60. Morikawa Y, Uemura S, Ishigami K, et al. Morphological features lated from hyperhomocysteinemic mice: an implication to coro- of coronary arteries in patients with coronary spastic angina: as- nary vasospasm. J Cell Physiol 2011;226:2712-20. sessment with intracoronary optical coherence tomography. Int 48. Kandabashi T, Shimokawa H, Miyata K, et al. Inhibition of myosin J Cardiol 2011;146:334-40. phosphatase by upregulated rho-kinase plays a key role for coro- 61. Tanaka A, Shimada K, Tearney GJ, et al. Conformational change in nary artery spasm in a porcine model with interleukin-1 beta. coronary artery structure assessed by optical coherence tomo- Circulation 2000;101:1319-23. graphy in patients with vasospastic angina. J Am Coll Cardiol 49. Kandabashi T, Shimokawa H, Miyata K, et al. Evidence for protein 2011;58:1608-13. kinase C-mediated activation of Rho-kinase in a porcine model of 62. Hung MJ, Cherng WJ, Hung MY, et al. Interleukin-6 inhibits coronary artery spasm. Arterioscler Thromb Vasc Biol 2003; endothelial nitric oxide synthase activation and increases endo- 23:2209-14. thelial nitric oxide synthase binding to stabilized caveolin-1 in 50. Hung MJ, Cherng WJ, Hung MY,et al. Increased leukocyte Rho-as- human vascular endothelial cells. J Hypertens 2010;28:940-51. sociated coiled-coil containing protein kinase activity predicts 63. Drab M, Verkade P, Elger M, et al. Loss of caveolae, vascular dys- the presence and severity of coronary vasospastic angina. Ath- function, and pulmonary defects in caveolin-1 gene-disrupted erosclerosis 2012;221:521-6. mice. Science 2001;293:2449-52. 51. Nohria A, Grunert ME, Rikitake Y, et al. Rho kinase inhibition im- 64. Noma K, Kihara Y, Higashi Y. Striking crosstalk of ROCK signaling proves endothelial function in human subjects with coronary ar- with endothelial function. J Cardiol 2012;60:1-6. tery disease. Circ Res 2006;99:1426-32.

Acta Cardiol Sin 2013;29:1-10 10