Impact of vasomotion type on prognosis of coronary artery spasm induced by acetylcholine provocation test of left coronary artery

Accepted Manuscript Impact of vasomotion type on prognosis of coronary artery spasm induced by acetylcholine provocation test of left coronary artery Eun Mi Lee, Man Ho Choi, Hong Seog Seo, Kim Hyun Ki, Nam-Ho Kim, Cheol Ung Choi, Jin Won Kim, Hong Euy Lim, Eung Ju Kim, Seung-Woon Rha, Chang Gyu Park, Dong Joo Oh PII:

S0021-9150(16)31329-6

DOI:

10.1016/j.atherosclerosis.2016.09.015

Reference:

ATH 14793

To appear in:

Atherosclerosis

Received Date: 14 May 2016 Revised Date:

30 July 2016

Accepted Date: 14 September 2016

Please cite this article as: Lee EM, Choi MH, Seo HS, Ki KH, Kim N-H, Choi CU, Kim JW, Lim HE, Kim EJ, Rha S-W, Park CG, Oh DJ, Impact of vasomotion type on prognosis of coronary artery spasm induced by acetylcholine provocation test of left coronary artery, Atherosclerosis (2016), doi: 10.1016/ j.atherosclerosis.2016.09.015. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Impact of vasomotion type on prognosis of coronary artery spasm induced by acetylcholine provocation test of left coronary artery

RI PT

Eun Mi Leea¶, Man Ho Choib¶, Hong Seog Seoc,d*, Kim Hyun Kia, Nam-Ho Kima, Cheol Ung

Dong Joo Ohc

a

Division of Cardiology, Department of Internal Medicine, Wonkwang University Sanbon

b

M AN U

Hospital, Gyeonggi-do 435-040, Republic of Korea

Future Convergence Research Division, Korea Institute of Science and Technology, Seoul

136-791, Republic of Korea c

SC

Choic, Jin Won Kimc, Hong Euy Limc, Eung Ju Kimc, Seung-Woon Rhac, Chang Gyu Parkc,

Cardiovascular Center, Korea University Guro Hospital, Seoul 152-703, Republic of Korea Korea University–Korea Institute of Science and Technology (KU-KIST) Graduate School

TE D

d

¶

EP

Converging Science and Technology, Seoul 02841, Republic of Korea

*

AC C

These authors contributed equally to this work.

Corresponding author. Cardiovascular Center, Korea University Guro Hospital, 80 Guro-

dong, Seoul 152-703, Republic of Korea. Tel.: 82-2-2626-3018; Fax: 82-2-863-1109. E-mail address: [email protected] (H.S. Seo)

Keywords:Coronary artery spasm,Vasomotion type,Acetylcholine provocation test, Longterm clinical outcome. 1

ACCEPTED MANUSCRIPT Abstract Background and aims: The impact of vasomotion types on long-term clinical outcomes in patients with coronary artery spasm (CAS) induced by the acetylcholine provocation test

RI PT

(ACH-test) remains unclear. Methods: We evaluated 4644 consecutive patients with typical resting chest pain (CP), but no angiographically significant coronary artery lesion (<50% stenosis), who underwent an ACH-

SC

test. According to their vasomotor response, patients were categorized into four types: normal vasomotion (no CP, no ischemic electrocardiographic changes, and no vasoconstriction),

M AN U

microvascular spasm (CP with <75% vasoconstriction but with CP relief after nitroglycerin infusion), epicardial spasm (CP with ≥75% vasoconstriction), and ACH-test inconclusive (vasoconstriction and/or electrocardiographic changes, but no CP). We investigated CP recurrence requiring follow-up angiography and major adverse cardiovascular events

TE D

(MACEs) during 5 years.

Results: CP recurred in 7.9% of patients and was more frequent in abnormal vasomotion types (normal vasomotion, microvascular spasm, epicardial spasm, and inconclusive type:

EP

5.4%, 9.8%, 10.9%, and 8.2%, respectively, log-rank p = 0.009). In multivariate analysis adjusted for medication use after the ACH-test, vasomotion subtype was not an independent

AC C

predictor, whereas male sex, fixed lesion on baseline angiography, and medications including calcium channel blockers (CCBs), nitrates, and statins were independent positive predictors for recurrent CP. Alcohol consumption at the initial interview was a negative predictor. MACEs were observed in 1.6%, and the incidence was similar among subtypes (p = 0.421) Conclusions: Recurrent CP and long-term outcomes are independent of vasomotion subtypes, but long-term use of CCBs, nitrates, and statins is a significant predictor for recurrent CP.

2

ACCEPTED MANUSCRIPT 1. Introduction Coronary artery spasm (CAS) is a transient vasoconstriction of the coronary arteries. It plays a significant role in the pathogenesis of variant angina and acute coronary syndrome

RI PT

[1,2]. CAS attacks often occur at rest, particularly during the interval from midnight to early morning. The intracoronary acetylcholine provocation test (ACH-test) is a sensitive and

reliable method for diagnosing CAS [2,3]. Previous studies described two types of CAS

SC

induced by the ACH-test, epicardial and microvascular spasm, and different clinical features have been reported for the individual types [4-6]. However, it is unknown whether clinical

M AN U

outcomes differ between types; although epicardial spasm is associated with an excellent prognosis for survival, persistent angina seems to increase during the follow-up period [7,8]. Ong et al.[6] recently described the clinical features of another vasomotor response type, the ACH-test inconclusive type, which covers cases where there are disparities in the diagnostic

TE D

parameters, including chest pain (CP), electrocardiographic (ECG) changes, and angiographic vasoconstriction, during the ACH-test. Nevertheless, it is unclear whether patients with this type of response have different clinical features and long-term outcomes compared with those

EP

with other types of CAS.

Therefore, in this study, we aimed to compare the long-term clinical outcomes according to

AC C

the vasomotor response type assessed by ACH-test in patients with typical resting CP.

2. Patients and methods 2.1. Study population

From March 2004 to February 2014, 5826 consecutive patients in the CAS registry of the Cardiovascular Center at Korea University, Guro Hospital (Seoul, Korea), who had typical

3

ACCEPTED MANUSCRIPT resting CP without significant coronary artery lesions (<50% luminal stenosis) on coronary artery angiography (CAG), underwent an ACH-test. Typical resting CP was defined as anginal pain occurring almost exclusively during rest that responds promptly to nitrate. In this

RI PT

study, the exclusion criteria were an ergonovine provocation test (n = 73), right coronary provocation (n = 29), a history of revascularization including percutaneous coronary

intervention or coronary artery bypass graft (n = 795), or incomplete data acquisition (n =

SC

285). Finally, 4644 patients with a left coronary provocation test were evaluated. All aspects of the study complied with the Declaration of Helsinki. The study was approved by the

M AN U

Institutional Review Board of Korea University, Guro Hospital (KUGH13017). The risk factors for coronary artery disease (CAD) were defined as hypertension (systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg on at least two consecutive readings in an outpatient clinic or current use of antihypertensive medications),

TE D

hyperlipidemia (total cholesterol level ≥200 mg/dL or current use of lipid-lowering drugs), diabetes mellitus (fasting blood glucose ≥126 mg/dL, hemoglobin A1c level >6.5%, or current use of diabetes medications), and a family history of CAD. We also recorded current

EP

smoking (active smoking within the last 12 months), current alcohol consumption (at least one alcoholic drink per week), and medications used before and after the ACH-test,

AC C

respectively.

2.2. ACH provocation test

All vasoactive drugs were withheld for at least 48 hours before CAG. We injected ACH in incremental doses of 20 µg (A1), 50 µg (A2), and 100 µg (A3) into the left coronary artery at 5-minute intervals up to the maximum tolerated dose. If epicardial spasm was induced by any 4

ACCEPTED MANUSCRIPT dose of ACH, if intolerable CP and/or ECG changes were present even without epicardial spasm, or if hemodynamic instability occurred, 200 µg of intracoronary glyceryl trinitrate was injected into the spastic artery. A temporary pacemaker was activated in case of high-

RI PT

grade atrioventricular (AV) block with hemodynamic instability. We analyzed images using the quantitative coronary angiographic system from the catheterization laboratory (BH-3000; Philips Medical Systems, Best, The Netherlands). Ischemic ECG changes were defined as a

SC

new appearance of ≥0.1 mV ST elevation/depression at 80 ms after the J point in at least two contiguous leads or a negative U wave, a fixed lesion as 30–50% stenosis on the baseline

M AN U

CAG, the maximum tolerable dose of ACH as the dose that induced significant epicardial spasm, diffuse constriction as more than 30 mm narrowing in coronary arteries, and multivessel CAS as epicardial spasm occurring simultaneously in more than one major coronary artery at the same ACH dose; the location of the epicardial spasm was classified as

TE D

proximal, mid, or distal, as defined by the reporting system of the American Heart Association[9] Fatal complications included events such as ventricular tachycardia,

EP

ventricular fibrillation, and myocardial infarction during the provocation test.

2.3. Definition of vasomotion types

AC C

According to the vasomotor response to the ACH-test, the study population was divided into two categories (normal and abnormal) further subdivided into four subtypes, as described by Ong et al.[6] and defined as follows: 1) normal vasomotion (no CP, no vasoconstriction, and no ECG changes), 2) microvascular spasm (CP with <75% coronary vasoconstriction but pain relief after the administration of 200 µg of intracoronary glyceryl trinitrate), 3) epicardial spasm (CP with ≥75% vasoconstriction and/or ischemic ECG changes [10]), and 4) ACH-test

5

ACCEPTED MANUSCRIPT inconclusive (unclassified cases, such as vasoconstriction only, ischemic ECG changes only, and both vasoconstriction and ischemic ECG changes without CP).

RI PT

2.4. Clinical outcomes

We evaluated the cumulative incidence of recurrent CP requiring follow-up CAG and the

SC

occurrence of major adverse cardiovascular events (MACEs) over a period of up to 5 years (mean duration, 1427 days). We defined a MACE as the composite of total death (cardiac and

M AN U

noncardiac), stroke, de novo myocardial infarction (MI) including spasm-induced MI (shown as epicardial spasm during the ACH-test after follow-up angiography) and revascularization. All study patients completed a clinical follow-up through a face-to-face interview, medical

EP

2.5. Statistics

TE D

chart review, and telephone contact.

We performed all statistical analyses using SPSS software (PASW statistics version 20;

AC C

IBM, Armonk, NY, USA). We expressed continuous variables as mean ± standard deviation and analyzed data using one-way analysis of variance to compare the differences between the four subtypes. For categorical variables, we expressed differences between subtypes as counts (percentages) and analyzed data using the χ2 test (or Fisher’s exact test). To determine the predictors of recurrent CP, multivariable Cox proportional hazard regression analysis was adjusted for age, male sex, current smoking, alcohol consumption status at the initial interview, systolic blood pressure, triglyceride and high-density lipoprotein-cholesterol 6

ACCEPTED MANUSCRIPT levels, fixed lesions on the baseline CAG, and types of response to the ACH-test and medication use including β-blockers, calcium channel blockers (CCBs), long-acting nitrates, and statins before and after the ACH-test. We calculated odds ratios, hazard ratios, and 95%

RI PT

confidence intervals.

SC

3. Results

M AN U

3.1. Baseline characteristics

A total of 4644 (100%) patients were categorized into one of four vasomotion subtypes based on their response to the ACH-test (Supplementary Fig.l): normal vasomotion (36.5%), microvascular spasm (29.5%), epicardial spasm (13.9%), and ACH-test inconclusive (20.1%).

TE D

The ACH-test inconclusive type (n = 930) included vasoconstriction only (n = 888), ECG changes only (n = 23), and both ischemic ECG changes and epicardial vasoconstriction but no CP (n = 19). Patient characteristics are summarized in Table 1. The patients’ mean age was

EP

55.2 ± 12.5 years, and 45.1% were men. The microvascular type was more common in

AC C

women (64.6%), whereas the epicardial and inconclusive types were more common in men (56.0% and 58.5%, respectively). The traditional risk factors for CAD, hypertension, diabetes, hyperlipidemia, and a family history of CAD, did not differ significantly between subtypes. However, systolic blood pressure was lower in patients with microvascular and epicardial spasm than in those with normal and ACH-test inconclusive type. Before the ACH-test, medication use was similar among subtypes. However, after the ACH-test, anti-anginal drugs such as CCBs and long-acting nitrates and statins were prescribed more frequently for the 7

ACCEPTED MANUSCRIPT abnormal vasomotion subtypes, whereas β-blockers were more frequently prescribed for normal vasomotion type. On baseline CAG, fixed lesions were more frequently observed in patients with epicardial spasm and inconclusive type than in those with normal vasomotion

RI PT

and microvascular spasm. During the ACH-test, 28.6% of patients developed transient AV block during the ACH-test. But only 99 patients (2.2%) were managed with a temporal

pacemaker. We did not observe any fatal complications, but six patients (0.1%) developed

M AN U

3.2. Angiographic characteristics of epicardial spasm

SC

atrial fibrillation during the ACH-test.

The angiographic characteristics of patients with epicardial spasm are presented in Table 2. The most frequent finding was the presence of diffuse spasm at the mid to distal portion of

TE D

the left anterior descending artery.

EP

3.3. Occurrence of recurrent CP requiring follow-up CAG

AC C

Recurrent CP requiring CAG was recorded in 7.9% of patients (n = 367) and was more frequent in abnormal vasomotion types than in the normal vasomotion type before adjusting for medication use (normal vasomotion, microvascular spasm, epicardial spasm, and inconclusive type: 5.4%, 9.8%, 10.9%, and 8.2%, respectively, p = 0.009, Fig. 1). Furthermore, recurrent CP more frequently developed in patients with fixed lesions than in those without a fixed lesion on the baseline CAG evaluation (20.7% vs. 5.4%, p < 0.001, data not shown). However, in multivariate analysis adjusted for medications used after the ACH8

ACCEPTED MANUSCRIPT test, abnormal vasomotion subtype was not an independent predictor, whereas male sex, a fixed lesion on baseline angiography, and medications used including CCBs, long-acting nitrates, and statins were independent positive predictors for recurrent CP. Also, alcohol

RI PT

consumption status at the initial interview was a negative predictor (Table 3). In patients with epicardial spasm, recurrent CP more frequently developed in patients administered the A1 or A2 dose than in those administered the A3 dose (A1, A2, A3 dose: 11.8%, 11.2%, and 7.0%,

SC

respectively, p < 0.001, data not shown).

M AN U

3.4. Long-term clinical outcomes

During the follow-up period, MACEs occurred in 72 patients (1.6%) (Table 4). However, there was no statistically significant difference in the incidence of MACEs between subtypes

TE D

(p = 0.421). Cardiac death occurred in five patients (0.1%) (two patients had MI, one patient had congestive heart failure, one patient sustained ventricular tachycardia with syncope, and one patient had an intractable spasm attack), 17 patients (0.4%) developed MI (10 patients

EP

had spasm-induced MI and seven patients had MI with revascularization), and 32 patients

AC C

(0.7%) underwent revascularization during the follow-up period.

4. Discussion

In this study, we showed that the frequency of CAD risk factors did not differ significantly between the subtypes of CAS. Recurrent CP and long-term outcomes were independent of the

9

ACCEPTED MANUSCRIPT vasomotion subtype, but long-term anti-anginal and statin treatment was a significant predictor for recurrent CP. The ACH-test is a sensitive and reliable method for the detection of CAS. Previous studies

RI PT

described two subtypes of CAS, epicardial spasm [3,11] and microvascular spasm [4] using this test. Ong et al. [6] described the ACH-test inconclusive type, which consisted of

discordant responses to this test with respect to CP, ECG changes, and vasoconstriction. In

SC

our study, microvascular spasm showed a female dominance, consistent with the findings of other studies [4,6], whereas the epicardial spasm and inconclusive types showed a male

M AN U

dominance. This study showed that patients with the inconclusive type showed similar clinical characteristics to those seen in patients with epicardial spasm. In addition, systolic blood pressure was lower in microvascular and epicardial spasm than in the normal and ACH-test inconclusive type.

TE D

Our study showed that recurrent CP requiring follow-up CAG occurred in 7.9% of patients. Several studies reported that recurrent CP is a challenging problem in epicardial spasm, leading to repeated CAG examinations in some cases despite continuous anti-anginal

EP

treatment [8,12,13] One study reported that significant stenosis with repeated CAG was detected in 36.5% of patients [13], whereas another study reported that most patients showed

AC C

findings similar to those of the index CAG [8] Furthermore, a previous study with histologic examination showed intimal hyperplasia with few lipid-rich lesions as a distinct histological feature in coronary arteries experiencing spasm compared with coronary arteries with atheromatous fixed stenosis [14]. In our study, although recurrent CP developed more frequently in patients with fixed lesions than in those without fixed lesions on baseline CAG, only 0.7% of patients developed revascularization during the follow-up period. We do not know the true mechanism of this finding. However, we suspect that recurrent CP in CAS may 10

ACCEPTED MANUSCRIPT be related to transient, functional causes such as an increased vascular tone or the poststenotic reduction of perfusion pressure related to autonomic dysregulation [15] rather than atherothrombotic progression. In addition, this study showed that recurrent CP was more

RI PT

frequent in abnormal vasomotion types than in normal vasomotion. However, after adjusting for medications used after the ACH-test, the frequency of CP was not different among

vasomotion subtypes. In addition, users of CCBs, long-acting nitrates, and statins were

SC

significant predictors of recurrent CP. With these findings, we suggest that those drugs were more frequently prescribed to the patients with frequent CP to relieve CP regardless of

M AN U

subtype. It is difficult to explain why alcohol consumption was a negative predictor in this study. However, this paradoxical finding [16] is probably because most doctors may have strongly recommended abstinence from alcohol after the diagnosis of CAS. The long-term prognosis of the microvascular spasm and ACH-test inconclusive types is unknown although

TE D

it is believed to be good in epicardial spasm [7,8]. We speculated that patients with different vasomotor responses to the ACH-test might show different outcomes. In our study, the overall rate of MACEs was very low rate and thus prognosis is very good. Also, MACEs were

EP

similar between subtypes despite their different clinical characteristics. Moreover, total mortality (0.3%) during 5 years was very low and it was similar to age-matched Korean

AC C

population based on Complete Life Tables for Korean resident population in 2012 [17]. We suspect that this result is related to the lack of difference in CAD risk factors between CAS subtypes because CAD risk factors are important in determining future cardiovascular events. Also, the incidence of MACEs involving epicardial spasm (1.5%) was very low in our study, probably because our study population was composed of patients with low Japanese Coronary Spasm Association risk scores [18]. In our study, the incidence of epicardial coronary spasm after ACH infusion was much lower (13.9%) than that reported by Ong et 11

ACCEPTED MANUSCRIPT al.[6] in a Caucasian population (33.4%). The discrepancy is even more surprising because Asian populations usually show a higher prevalence of CAS than do Caucasian populations [19]. Some studies showed that myocardial ischemia due to CAS often occurs without

RI PT

accompanying symptoms [1,20]. Because the clinical features and outcomes of patients with ACH-test inconclusive type were similar to those with epicardial spasm, this type is probably another form of epicardial spasm; thus, the proportion of epicardial spasm may be

SC

underestimated in our study. In this study of provocation testing of the left coronary artery, no fatal complications developed and only 2.2% of patients were managed with a temporal

M AN U

pacemaker although 28.6% developed transient AV block. Thus, these results show reliable evidence that the ACH-test is a very safe procedure to perform routinely in a cardiac catheterization laboratory.

TE D

This study has several limitations. First, it was an observational study based on a prospective all-comer registry from a single center. However, it included a sufficiently large study population for the assessment of clinical outcomes. Second, our protocol for the

EP

provocation test has not changed significantly over 10 years; however, in 2007, radial artery access with a 4F catheter was tried. This change is unlikely to have affected the clinical

AC C

outcomes of the study. Third, we did not assess lactate production [5] as a surrogate marker of myocardial ischemia during a microvascular spasm.

In conclusion, our study demonstrates that the frequency of recurrent CP requiring

follow-up CAG is not different between subtypes of responses to ACH, but long-term CCB, nitrate, and statin treatment is a significant predictor of recurrent CP. Moreover, the long-term

12

ACCEPTED MANUSCRIPT outcomes were good and did not differ depending on the vasomotion subtype in the ACH provocation test.

RI PT

Conflict of interest The authors declared they do not have anything to disclose regarding conflict of interest with

SC

respect to this manuscript.

M AN U

Financial support

This work was partially supported by a grant from the KIST Institutional Program [2E25360], a grant from the Korea University–Korea Institute of Science and Technology (KU-KIST) Graduate School Converging Science and Technology [R1435292], and grants

Author contributions

TE D

from Korea University and Wonkwang University [2015].

EP

H.S.S. and M.H.C. wrote the main manuscript text, conceived and designed the study,

AC C

collected and interpreted data, and performed literature searches. H.S.S., E.M,L and M.H.C. wrote the manuscript text, collected and interpreted data, and performed literature searches. H.K.K., N.H.K., C.U.C., J.W.K., H.E.L., E.J.K., S.W.R., C.G.P. and D.J.O. revised the manuscript, analysed, collected, and interpreted data, and prepared figures and tables. All authors reviewed the manuscript and approved the final version.

13

ACCEPTED MANUSCRIPT References

1. Yasue H, Nakagawa H, Itoh T, Harada E, Mizuno Y (2008) Coronary artery spasm--clinical

RI PT

features, diagnosis, pathogenesis, and treatment. J Cardiol 51 (1):2-17. doi:10.1016/j.jjcc.2008.01.001

2. Group JCSJW (2010) Guidelines for diagnosis and treatment of patients with vasospastic

SC

angina (coronary spastic angina) (JCS 2008): digest version. Circ J 74 (8):1745-1762

3. Yasue H, Horio Y, Nakamura N, Fujii H, Imoto N, Sonoda R, Kugiyama K, Obata K,

M AN U

Morikami Y, Kimura T (1986) Induction of coronary artery spasm by acetylcholine in patients with variant angina: possible role of the parasympathetic nervous system in the pathogenesis of coronary artery spasm. Circulation 74 (5):955-963 4. Mohri M, Koyanagi M, Egashira K, Tagawa H, Ichiki T, Shimokawa H, Takeshita A (1998)

TE D

Angina pectoris caused by coronary microvascular spasm. Lancet 351 (9110):1165-1169. doi:10.1016/S0140-6736(97)07329-7

5. Sun H, Mohri M, Shimokawa H, Usui M, Urakami L, Takeshita A (2002) Coronary

EP

microvascular spasm causes myocardial ischemia in patients with vasospastic angina. J Am Coll Cardiol 39 (5):847-851

AC C

6. Ong P, Athanasiadis A, Borgulya G, Vokshi I, Bastiaenen R, Kubik S, Hill S, Schaufele T, Mahrholdt H, Kaski JC, Sechtem U (2014) Clinical usefulness, angiographic characteristics, and safety evaluation of intracoronary acetylcholine provocation testing among 921 consecutive white patients with unobstructed coronary arteries. Circulation 129 (17):17231730. doi:10.1161/CIRCULATIONAHA.113.004096

14

ACCEPTED MANUSCRIPT 7. Yasue H, Takizawa A, Nagao M, Nishida S, Horie M, Kubota J, Omote S, Takaoka K, Okumura K (1988) Long-term prognosis for patients with variant angina and influential factors. Circulation 78 (1):1-9

RI PT

8. Choi BG, Park SH, Rha SW, Park JY, Choi SY, Park Y, Xu S, Ngow HA, Ali J, Li H, Kim JB, Lee S, Na JO, Choi CU, Lim HE, Kim JW, Kim EJ, Park CG, Seo HS, Oh DJ (2015) Five-year clinical outcomes in patients with significant coronary artery spasm: A propensity

SC

score-matched analysis. Int J Cardiol 184:533-539. doi:10.1016/j.ijcard.2015.02.021

9. Austen WG, Edwards JE, Frye RL, Gensini GG, Gott VL, Griffith LS, McGoon DC,

M AN U

Murphy ML, Roe BB (1975) A reporting system on patients evaluated for coronary artery disease. Report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation 51 (4 Suppl):5-40 10. Miwa K, Fujita M, Ejiri M, Sasayama S (1991) Usefulness of intracoronary injection of

TE D

acetylcholine as a provocative test for coronary artery spasm in patients with vasospastic angina. Heart Vessels 6 (2):96-101

11. Okumura K, Yasue H, Matsuyama K, Goto K, Miyagi H, Ogawa H, Matsuyama K (1988)

EP

Sensitivity and specificity of intracoronary injection of acetylcholine for the induction of coronary artery spasm. J Am Coll Cardiol 12 (4):883-888

AC C

12. Corinne Bott-Silverman, Frederich A, Heupler Jr. (1983) Natural history of pure coro9nary artery spasm in patients treated medically. J Am Coll Cardiol 2 (2):200-205 13. Bory M, Pierron F, Panagides D, Bonnet JL, Yvorra S, Desfossez L (1996) Coronary artery spasm in patients with normal or near normal coronary arteries. Long-term follow-up of 277 patients. Eur Heart J 17 (7):1015-1021

15

ACCEPTED MANUSCRIPT 14. Yokoyama M, Akita H, Hirata K, Usuki S, Fukuzaki H, Itoh H, Maekawa K, Matsuo T (1990) Supersensitivity of isolated coronary artery to ergonovine in a patient with variant angina. Am J Med 89 (4):507-515

RI PT

15. Lanza GA, Careri G, Crea F (2011) Mechanisms of coronary artery spasm. Circulation 124 (16):1774-1782. doi:10.1161/CIRCULATIONAHA.111.037283

16. Sato A, Taneichi Y, Sekine I, Okabe F, Ueda A, Takahashi M, Ito T, Su KM, Sada T,

SC

Matsumoto S, Ito Y (1981) Prinzmetal's variant angina induced only by alcohol ingestion. Clin Cardiol 4 (4):193-195

M AN U

17. World Health Organization Global health observatory data repository: Life tables by country-Republic of Korea apps.who.int/gho/data/view.main.LT62170?lang=en 18. Takagi Y, Takahashi J, Yasuda S, Miyata S, Tsunoda R, Ogata Y, Seki A, Sumiyoshi T, Matsui M, Goto T, Tanabe Y, Sueda S, Sato T, Ogawa S, Kubo N, Momomura S, Ogawa H,

TE D

Shimokawa H, Japanese Coronary Spasm A (2013) Prognostic stratification of patients with vasospastic angina: a comprehensive clinical risk score developed by the Japanese Coronary Spasm Association. J Am Coll Cardiol 62 (13):1144-1153. doi:10.1016/j.jacc.2013.07.018

EP

19. Beltrame JF, Sasayama S, Maseri A (1999) Racial heterogeneity in coronary artery vasomotor reactivity: differences between Japanese and Caucasian patients. J Am Coll

AC C

Cardiol 33 (6):1442-1452

20. Kim YG, Kim HJ, Choi W-S, Im M-S, Yoon C-H, Suh J-W, Choi D-J. Does a negative ergonovine provocation test truly predictive freedom from variant angina ? Korean Circ J 2013;43:199-203

16

ACCEPTED MANUSCRIPT Figure legends

Fig.1. Cox regression curve for recurrent chest pain requiring follow-up CAG before adjusting for medication use after the ACH-test. The Cox regression curve shows a

RI PT

higher incidence of recurrent chest pain requiring follow-up CAG in patients with abnormal vasomotion types than in those with normal vasomotion type (log-rank p = 0.009); however, there were no significant difference between the abnormal vasomotion subtypes

SC

(microvascular vs. epicardial spasm, p = 0.226; microvascular spasm vs. ACH-test

inconclusive, p = 0.789; epicardial spasm vs. ACH-test inconclusive, p = 0.321). ACH,

AC C

EP

TE D

M AN U

acetylcholine; cum, cumulative; CAG, coronary angiography.

17

ACCEPTED MANUSCRIPT Table 1. Patient characteristics. Total

Normal

Microvascular

Epicardial

ACH-test

subjects,

vasomotion,

spasm,

spasm,

inconclusive

n = 4644

n = 1696

n = 1371

n = 647

n = 930

(100%)

(36.5%)

(29.5%)

(13.9%) a

Sex (male), n (%)

2095 (45.1)

704 (41.5)

485 (35.4)

Age, years

55.2 ± 12.5

54.1 ± 13.4

54.6 ± 12.4

Systolic

134.7 ± 22.1

135.7 ± 22.8

133.4 ± 22.8

Diastolic

78.7 ± 13.2

79.1 ± 13.6

78.2 ± 13.3

Blood pressure, mmHg

39 (0.8)

14 (0.8)

8 (0.6)

561 (12.1)

204 (12.0)

173 (12.6)

a

Risk factors for CAD Diabetes mellitus

(20.1%) 544 (58.5) c,d

<0.001

55.7 ± 10.3b

57.4 ± 11.9d

<0.001

133.5 ± 19.7

135.7 ± 21.1

0.010

78.7 ± 12.2

78.9

0.293

362 (56.0)

b

10 (1.5)

b

12.8

7 (0.8)

0.167

69 (10.7)

1115 (12.4)

0.644

300 (46.4)

444 (47.9)

0.079

SC

History of MI

p-value

RI PT

Clinical parameters

2165 (46.6)

820 (48.3)

601 (43.8)

Hyperlipidemia

798 (17.2)

259 (15.3)

243 (17.7)

120 (18.5)

176 (18.9)

0.058

Family history of CAD

183 (3.9)

55 (3.2)

63 (4.6)

33 (5.1)

32 (3.4)

0.083

Current smoking

945 (20.3)

Current alcohol

1407 (30.3)

drinking Laboratory parameters

M AN U

Hypertension

298 (17.6) 478 (28.2)

233 (17.0)

188 (29.1)

b

372 (27.1)

244 (37.7)b

226 (24.3)

c,d

<0.001

d

<0.001

313 (33.7)

176.9 ± 45.6

177.6 ± 46.4

178.4 ± 43.9

176.8 ± 47.2

173.7 ± 45.6

0.143

LDL-cholesterol, mg/dl

109.6 ± 38.6

110.3 ± 37.6

110.9 ± 37.8

108.7 ± 42.7

107.4 ± 38.6

0.263

Triglyceride, mg/dl

124.6 ± 92.5

119.7 ± 89.8

123.8 ± 85.7

137.6 ± 114.2

125.3 ± 89.0

0.002

51.0 ± 14.6

51.9 ± 14.9

51.4 ± 14.5

47.8 ± 14.2

50.9 ± 14.4

<0.001

5.97 ± 0.88

5.96 ± 0.91

5.99 ± 0.79

6.02 ± 0.91

5.95 ± 0.91

0.696

143 (3.2)

51 (3.2)

40 (2.9)

20 (3.1)

32 (3.5)

0.919

984 (21.2)

442 (26.0)

289 (21.0)

122 (18.9)

131 (14.0)

0.001

361 (7.8)

110 (6.5)

112 (8.2)

56 (8.7)

8.3 (8.9)

0.084

1465 (31.5)

490 (28.9)

429 (31.3)

223 (34.5)

323 (34.4)

0.006

Before

829 (18.8)

262 (17.1)

245 (18.5)

125 (19.3)

197 (21.4)

0.065

After

2427 (52.3)

404 (23.8)

850 (61.9)a

514 (79.4)b

659 (70.8)c,d

<0.001

346 (7.5)

118 (7.0)

100 (7.3)

61 (9.4)

67 (7.2)

0.221

HDL-cholesterol, mg/dl

Medications, n (%) β-blockers

AC C

Before

EP

HbA1c, %

TE D

Total cholesterol, mg/dl

After

CCBs

Before After

Long-acting nitrates

Statins, n(%) Before

1

ACCEPTED MANUSCRIPT After

2250 (48.4)

Fixed lesion on b-CAG

655 (38.6)

698 (50.9)a

388 (60.0)b b

509 (54.7)c

<0.001

d

<0.001

747 (16.6)

248 (15.3)

185 (14.0)

122 (19.15)

192 (21.0)

1328 (28.6)

490 (28.9)

363 (26.5)

183 (28.3)

292 (31.4)

0.082

Managed with pacing

99 (2.2)

35 (2.3)

28 (2.1)

12 (1.9)

24 (2.6)

0.776

Atrial fibrillation

6 (0.1)

4 (0.2)

1 (0.07)

1 (0.15)

0 (0)

0.402

ACH provocation, n (%)

RI PT

AV block

Data are expressed as mean ± standard deviation or number (percent).

Medication use before and after the ACH-test is shown. The p-values represent overall

Statistically significant difference between two types analyzed with paired t-test and χ2

M AN U

a,b,c,d

SC

differences between types, as determined by one-way analysis of variance.

test (anormal vs. microvascular spasm; bmicrovascular spasm vs. epicardial spasm; cepicardial spasm vs. inconclusive type; dmicrovascular spasm vs. inconclusive type).

AC C

EP

TE D

ACH, acetylcholine; MI, myocardial infarction; CAD, coronary artery disease; LDL, lowdensity lipoprotein; HDL, high-density lipoprotein; CCB, calcium channel blocker; b-CAG, baseline coronary angiography.

2

ACCEPTED MANUSCRIPT Table 2. Angiographic characteristics of epicardial spasm (n = 647). Variables

Epicardial spasm 2.49 ± 1.05

Reference diameter after ACH infusion (mm)

0.43 ± 0.21

Diameter narrowing after ACH infusion (%)

82.9 ± 6.6

RI PT

Reference diameter after NTG infusion (mm)

Percent spasm, n(%)

100 (15.5)

75–90% vasoconstriction

547 (84.5)

SC

>90% vasoconstriction

M AN U

Maximum tolerable dose of ACH, n (%) A1 (20 µg/min)

40 (6.2)

A2 (50 µg/min)

289 (44.7)

A3 (100 µg/min)

318 (49.1)

Spasm site, n (%)

TE D

LM LAD LCX

AC C

Multivessel spasm

EP

Both LAD and LCX

1 (0.2) 366 (56.6) 24 (3.7) 255 (39.4) 255 (39.4)

Spasm type, n (%) Focal

68 (10.5)

Diffuse

579 (89.5)

Spasm location, n (%) Proximal

34 (5.3)

Mid

62 (9.6)

3

ACCEPTED MANUSCRIPT Distal

12 (1.8)

Mid to distal

230 (35.5)

Proximal to distal

309 (47.8) 57 (8.8)

RI PT

ECG changes, n (%)

23 (3.5)

STD

25 (3.9)

Negative U wave

9 (1.4)

SC

STE

NTG, nitroglycerin; ACH, acetylcholine; LM, left main; LAD, left anterior descending

M AN U

artery; LCX, left circumflex artery; ECG, electrocardiogram; STE, ST elevation; STD, ST

AC C

EP

TE D

depression.

4

ACCEPTED MANUSCRIPT Table 3. Predictors for recurrent chest pain by multivariate Cox regression analysis adjusted for medication use after the ACH-test. p-value

Age

1.004 (0.992–1.015)

0.536

Sex (male)

1.325 (1.007–1.743)

0.045

Current smoking

0.902 (0.660–1.234)

0.520

Alcohol consumption status in initial interview

0.716 (0.539–0.952)

0.021

Systolic blood pressure

0.998 (0.993–1.003)

0.389

Triglyceride

1.000 (0.999–1.001)

0.410

HDL-C

0.992 (0.984–1.001)

0.074

3.274 (2.559–4.188)

<0.001

RI PT

HR (95% CI)

Fixed lesion on baseline CAG Response types of ACH-test

M AN U

SC

Variable

0.678

0.947 (0.690–1.300)

0.737

Epicardial spasm vs. normal vasomotion

1.026 (0.719–1.466)

0.886

0.846 (0.604–1.186)

0.333

1.186 (0.916–1.536)

0.197

1.800 (1.418–2.285)

0.001

2.919 (2.132–3.996)

0.001

2.346 (1.736–3.173)

0.001

TE D

Microvascular spasm vs. normal vasomotion

ACH-test inconclusive vs. normal vasomotion Post-CAG medications

CCB

EP

β-blockers

AC C

Long-acting nitrates Statins

HR, hazard ratio; CI, confidence interval; HDL-C, high-density lipoprotein-cholesterol; CAG, coronary artery angiography; ACH, acetylcholine; CCB, calcium channel blocker.

5

ACCEPTED MANUSCRIPT Table 4. MACEs during the 5-year follow-up. Normal

Microvascular

Epicardial

ACH-test

subjects, n =

vasomotion, n

spasm, n =

spasm, n =

inconclusive,

4644

= 1696

1371

647

n = 930

MACE, n (%)

72 (1.6)

22 (1.3)

19 (1.4)

10 (1.5)

21 (2.2)

0.421

Total mortality

15 (0.3)

3 (0.2)

5 (0.4)

1 (0.1)

6 (0.6)

0.198

Cardiac death

5 (0.1)

1 (0.1)

1 (0.1)

1 (0.1)

2 (0.2)

0.853

Stroke

21 (0.5)

6 (0.4)

6 (0.4)

4 (0.5)

5 (0.5)

0.847

De novo MI

17 (0.4)

4 (0.2)

5 (0.3)

5 (0.7)

3 (0.4)

0.309

Spasm-induced MI

10 (0.2)

3 (0.2)

3 (0.2)

3 (0.4)

1 (0.1)

0.538

Revascularization

32 (0.7)

10 (0.6)

9 (0.7)

3 (0.6)

10 (1.1)

0.405

SC

Variables

RI PT

Total

p-value

Spasm-induced MI was defined as MI with a positive result on the ACH provocation test.

AC C

EP

TE D

M AN U

ACH, acetylcholine; MACE, major adverse cardiovascular event; MI, myocardial infarction.

6

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Research Highlights Impact of vasomotion types on chest pain and outcomes is evaluated.

RI PT

Chest pain is more frequent in epicardial spasm type. The overall rate of major adverse cardiovascular events (MACEs) is very low, similar

AC C

EP

TE D

M AN U

SC

between subtypes.