Abnormal coronary vasoreactivity in transient left ventricular apical ballooning (tako-tsubo) syndrome

International Journal of Cardiology 250 (2018) 4–10

Contents lists available at ScienceDirect

International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard

Abnormal coronary vasoreactivity in transient left ventricular apical ballooning (tako-tsubo) syndrome Edoardo Verna ⁎, Stefano Provasoli, Sergio Ghiringhelli, Fabrizio Morandi, Jorge Salerno-Uriarte Department of Cardiology, Ospedale di Circolo e Fondazione Macchi, University of Insubria, Varese, VA, Italy

a r t i c l e

i n f o

Article history: Received 28 March 2017 Received in revised form 31 May 2017 Accepted 11 July 2017 Keywords: Tako-tsubo syndrome Coronary blood flow Acetylcholine testing Endothelial dysfunction Coronary artery disease Myocardial bridging

a b s t r a c t Background: The exact etiology and pathophysiologic mechanisms of tako-tsubo syndrome (TTS) remain controversial. Objective: To further evaluate the abnormal coronary vasoreactivity and its possible anatomical substrate in TTS. Methods: We studied 47 patients (46 women; age 67 ± 12 years) who underwent diagnostic cardiac catheterization and evaluation of coronary vasoreactivity by sequential acetylcholine (Ach), nitroglycerine and adenosine testing with angiographic and intracoronary pressure-Doppler flow monitoring. Coronary artery wall morphology was also evaluated by intravascular ultrasound (IVUS) imaging in 45 vessels of 43 patients. Results: Abnormal coronary vasoconstriction to Ach stimulation was elicited in 40 patients (85%) involving the LAD artery and its branches in 39 (83%). Abnormal microvascular function was seen in 39 (83%) patients. Overall, hyperemic microvascular resistance index (HMR) was higher and Doppler coronary flow velocity reserve (CFVR) was lower in the LAD artery territory as compared to the reference territories (2.64 ± 1.23 vs 2.05 ± 0.56; p = 0.008 and 1.95 ± 0.7 vs 2.3 ± 0.6; p = 0.018, respectively). IVUS revealed no plaque rupture, dissection or thrombosis but occult plaque formation and myocardial bridging were found as a possible anatomical substrate of endothelial dysfunction in 67% and 48.8% patients respectively. Conclusions: A global failure of coronary vasomotor function was demonstrated in most TTS patients. These findings implicate abnormal vasoconstrictive response to the activation of the sympathetic system as a potential mechanism involved in the pathogenesis of myocardial stunning in TTS. Perspectives: Competency in medical knowledge: Abnormal coronary vasoconstriction secondary to endothelial dysfunction may actively contribute to the clinical manifestation of acute coronary syndromes in patients with non-obstructive coronary disease. Translational outlook 1: TTS patients reveal a global failure of vasomotor function with both vasoconstrictive response to acetylcholine and increased hyperemic microvascular resistances in the territory of myocardial stunning. They may also show occult coronary atherosclerosis and myocardial bridging as the anatomic substrates of endothelial dysfunction. Translational outlook 2: The cardiac phenotype of TTS includes a high prevalence of coronary vasomotor disturbances. These findings implicate abnormal vasoconstrictive response to the activation of the sympathetic system as a potential mechanism involved in the pathogenesis of TTS in post-menopausal women. Thus, a systematic evaluation of coronary vasoreactivity could better characterize the syndrome. © 2017 Elsevier B.V. All rights reserved.

1. Background and purpose of the study In spite of the large number of case reports, clinical studies and review articles published since the first description by Sato et al. [1] in Abbreviation: TTS, tako-tsubo syndrome; Ach, acetylcholine; NTG, nitroglycerine; Ado, adenosine; LAD, left anterior descending; IVUS, intra-vascular ultrasound imaging; QCA, quantitative coronary angiography; APV, average peak velocity; CFVR, coronary flow velocity reserve; HMR, hyperemic microvascular resistance; LA, lumen area; VA, vessel area; PA, plaque area; RPA, relative plaque area; OCT, optical coherence tomography. ⁎ Corresponding author at: Ospedale di Circolo e Fondazione Macchi, University of Insubria, Varese, Italy. E-mail address: [email protected] (E. Verna).

http://dx.doi.org/10.1016/j.ijcard.2017.07.032 0167-5273/© 2017 Elsevier B.V. All rights reserved.

1990, the exact pathophysiologic mechanisms of transient left ventricular apical ballooning, also termed stress cardiomyopathy, tako-tsubo cardiomyopathy or tako-tsubo syndrome (TTS) [1–7], has not been fully elucidated and remains the subject of debate. Disparate possible mechanisms have been proposed, including simultaneous multivessel coronary artery vasoconstriction [1,2,8,9], primary microcirculatory dysfunction [10–13], direct myocyte catecholamine injury, [14–17] or aborted acute myocardial infarction secondary to transient coronary occlusion by a fast-dissolving clot from an ulcerated plaque undetected by coronary angiography [18,19]. Multivessel epicardial coronary artery spasm has been suggested as a possible cause of post-ischemic stunning since the early descriptions

E. Verna et al. / International Journal of Cardiology 250 (2018) 4–10

of the syndrome [1–3]. However, spontaneous spasm has been demonstrated in few reports and vasomotor dysfunction testing has not been actively investigated in most series of patients. The aim of our study was to further evaluate abnormal coronary vasoreactivity and its possible anatomical substrate in TTS. 2. Patients and methods 2.1. Patients Among 5620 patients undergoing cardiac catheterization for acute coronary syndrome between May 2007 and May 2015, 74 patients fulfilled the diagnostic criteria of TTS (1.3%). The clinical diagnosis of TTS was made in the presence of 1) chest pain with new ischemic ECG changes 2) transient and severe left ventricular regional wall-motion abnormalities, 3) modest elevation in cardiac troponin levels as compared to the extent of wall motion abnormalities, 4) absence of obstructive coronary artery disease or angiographic evidence of acute plaque rupture, 5) absence of myocarditis, new electrocardiographic Q-wave or evidence of myocardial scar and 6) recovery of ventricular systolic function within days or weeks [1–7]. Twenty-seven patients were excluded because of unstable conditions, inability to withdraw vasoactive medication, late presentation, or refusal to give informed consent to the study protocol. Thus, the study population consisted of 47 TTS patients (46 women; age 67 ± 12 years) who underwent cardiac catheterization within 3.5 ± 4.7 days (range: 1–15, median: 2.0 days) from symptoms onset. Patient's clinical and angiographic characteristics are summarized in Table 1. 2.2. Study protocol Coronary vasomotor function was evaluated immediately after the diagnostic coronary angiography or a few days later to allow the withdrawal of vasoactive medications (nitrates, calcium antagonists, ACEinhibitors). Antiplatelet therapy, beta-blocking agents and statins were usually allowed. Informed consent to the full study was obtained from all patients according to the clinical protocol of drug evaluation of coronary vasomotor function approved by the local ethical committee and to the ethical guidelines of the 1975 Declaration of Helsinki. After administration of body weight adjusted heparin (60 UI/kg) a 0.014″ floppy intracoronary Doppler Flow-wire™ (33 patients) or a dual pressure and flow sensor Combo-wire™ (14 patients) connected with a dedicated instrument (Flowire System Volcano, Rancho Cordova, CA) were used to record coronary pressure and flow velocities. The wire was advanced through a 6F guiding catheter into the left anterior descending (LAD) artery and manipulated to obtain the best fitting of the Doppler flow spectrum. Epicardial endotheliumdependent coronary vasomotor function was studied first, followed by assessment of microvascular vasodilator capacity as detailed hereafter. Table 1 Patients clinical and angiographic characteristics. Number of patients Age (years) Women (%) Hypertension Diabetes Hyperlipidemia Smoking Emotional or physical stress trigger Mean peak CK (UI/L) Mean peak TnI (μg/dL) Non-significant CAD Wrap-around apical LAD Intramyocardial LAD path LVEF (%) on admission Apical ballooning type Mid-ventricular type

47 67 ± 12 46 (98%) 30 (64%) 5 (11%) 20 (43%) 8 (17%) 36 (77%) 225.3 ± 138.5 2.88 ± 2.5 16 (30%) 34 (72%) 21 (48.8%) 46 ± 8% 45 (95.7%) 2 (4.2%)

5

Then, coronary intravascular ultrasound imaging (IVUS) was performed by manual pullback of the imaging catheter over the same wire. Finally, the flow-wire was withdrawn from the LAD artery and advanced into the left circumflex (35 patients) or the right coronary artery (6 patients). Coronary pressure and flow velocities were recorded again to evaluate microvascular vasodilator capacity and microvascular resistance indexes of a “non-LAD” territory. Recovery of left ventricular function was evaluated by repeated transthoracic echocardiography during index hospitalization and follow-up (30.8 ± 12.1 days). 2.3. Epicardial coronary function Endothelium-dependent coronary vasodilatation was studied by 3minute infusion of increasing doses (10–7 mol/L, 10–6 mol/L and 10–5 mol/L or 0.72 μg/min, 7.2 μg/min and 36 μg/min) of intracoronary acetylcholine (Ach) (Miovisin, Farmigea SpA, Pisa, Italy). Ach was administered through the guiding catheter into the left main using a syringe pump (Alaris GP Volumetric pump Care-Fusion-Switzerland) with continuous monitoring of intracoronary Doppler flow velocity, blood pressure and electrocardiogram. Ach testing was repeated into the right coronary artery only in two patients. Two-view angiography quantitative coronary analysis (QCA) was performed at each step using the Philips QCA analysis system and the percent change in lumen diameter of major epicardial coronary vessels was measured. Drug infusion was stopped at the end of the high dose administration or at any time due to occurrence of severe chest pain, significant coronary vasoconstriction and abrupt drop of coronary blood flow velocity, atrial standstill or atrial-ventricular (A-V) block. We did not routinely insert a temporary right ventricular pacing electrode into the right ventricle during testing. The normal response to Ach is vasodilation of the epicardial vessels. Epicardial endothelium-dependent coronary vasomotor dysfunction was defined as N 50% reduction in coronary diameter at peak Ach infusion [20–22]. At the end of Ach testing, the non-endothelium dependent relaxation of the epicardial vessels was assessed by QCA after intracoronary nitroglycerine (NTG) administration (250 mcg). Abnormal epicardial non-endothelium dependent vasodilatation was defined as b5% increase from the baseline lumen diameter after NTG. 2.4. Microvascular coronary function The microvascular vasodilator function was studied after Ach testing and intracoronary NTG administration to abolish epicardial artery vasomotor tone. The microvascular vasodilator function was evaluated by assessing coronary blood flow response to intracoronary administration of 100–300 μg of adenosine (Ado, Krenosin, Sanofi-Winthrop). The ratio of hyperemic and baseline intracoronary Doppler average peak velocity (APV) was used to measure the coronary flow velocity reserve (CFVR) and relative flow reserve (RCFR) as previously described [23,24]. Microvascular dysfunction was defined as a CFVR b 2.5. The ratio of mean coronary pressure and APV at peak hyperemia was measured to determine the microvascular hyperemic resistance indexes (HMR). Abnormal microvascular resistances were defined as HMR N 2.0 [25]. 2.5. Intravascular ultrasound imaging IVUS imaging was performed using a 30 mHz phase array ultrasound catheters (Aigle Endosonics, Volcano Inc., Rancho Cordova, CA) connected with the integrated Volcano imaging system (Volcano Inc., Rancho Cordova, CA). Cross sectional 2D reconstruction grey-scale and ECG-gated frequency domain images were obtained during continuous manual pullback of the imaging catheter from the distal to the proximal sections of the vessel under fluoroscopic guidance. Atherosclerotic changes of the arterial wall, significant plaque formation and intimal thickening as well as plaque rupture, thrombosis and dissection were defined according to the guidelines for classification and analysis of

6

E. Verna et al. / International Journal of Cardiology 250 (2018) 4–10

intracoronary ultrasound images of the Working Group of Echocardiography of the European Society of Cardiology [26]. Stop-frames were obtained at the plaque site for quantitative analysis. The lumen area (LA) was determined by computer-assisted planimetry of the lumen-vessel wall interface. The vessel area (VA) was measured by planimetry of the area bounded by the middle echo-lucent zone (media) and the inner echo-reflective adventitial interface corresponding to the external elastic membrane. The plaque area (PA) was then calculated as VA-LA. The relative cross sectional plaque area (RPA) was measured at the site of maximal atherosclerotic involvement as RPA = PA / VA × 100 and used as an index of plaque burden. Plaque formation was defined as the presence of focal or diffuse disease by intravascular ultrasound involving N 40% RPA. LAD coronary artery myocardial bridging was defined as the appearance of a typical half-moon phenomenon around the vessel, associated to systolic vessel compression [27].

2.6. Statistical analysis Continuous variables were expressed as mean ± SD and analyzed using a one-way or two-way ANOVA. In case of significant results multiple comparisons were performed using the Student-NewmanKeuls method. Whenever appropriate a paired or unpaired t-test was used instead of the ANOVA approach. All tests were 2-tailed. Correlations between continuous variables were evaluated by regression analysis. Non-normal variables were evaluated by non-parametric Kruskal-Wallis one-way analysis of variance, Mann-Whitney U test and Spearman rank correlation. Categorical data were expressed as percentage and analyzed using Chi-square test. A p value b 0.05 was considered statistically significant. All calculations were performed using GB-Stat v6.5 (Dynamic Microsystems, Inc. Silver Spring, MD).

3.2. Coronary function testing There were no complications related to coronary wiring and drug testing during the catheterization procedure. Transient A-V block or atrial standstill occurred in three patients at high dose Ach administration. No patients required temporary pacing. Two patients had chest pain and showed transient ST-T segment changes that quickly normalized after NTG administration. Forty patients (85%) showed paradoxical coronary vasoconstriction of the LAD artery and its branches in response to Ach stimulation. Eighteen patients (45%) showed multivessel vasoconstriction involving also the left circumflex artery branches. Vasoconstriction was sub-occlusive or occlusive in 26 patients (65%), diffuse in 34 (85%) and focal in the remaining subjects. The frequency distribution of severity of coronary vasoconstriction of LAD and nonLAD vessel territories is shown in Fig. 1. Mean ± SD diameter reduction at peak Ach infusion was 73 ± 30%. The vasodilator response to intracoronary NTG was normal in all patients. Microvascular dysfunction was observed in 39 patients (80%). CFVR was significantly lower and HMR was significantly higher in the territory related to abnormal wall motion (the LAD artery in 47 patients and the left circumflex in 2 patients) compared to the reference adjacent territory (CFVR 1.95 ± 0.7 vs 2.3 ± 0.6; p = 0.018 and HMR: 2.64 ± 1.23 vs 2.05 ± 0.56; p = 0.008 respectively). Mean RCFR was 0.89 ± 0.25. There was a weak but significant inverse relationship between RCFR and the relative HMR (r = 0.55; p = 0.003). A concordant abnormal response to Ach and Ado indicating a combination of epicardial and microvascular dysfunction was detected in 33 patients (70.2%). Conversely, a discordant outcome of Ach and Ado testing was observed in 13 patients (32%). Among these, microvascular dysfunction and epicardial endothelium-dependent dysfunction were seen as in isolation in 6 (12.8%) and 7 (14.9%) patients respectively. Only one patient had fully normal coronary vasomotor function tests (Fig. 1).

3. Results 3.3. IVUS imaging 3.1. Clinical and angiographic findings Forty-two patients presented to hospital with typical chest pain. Emotional or physically stressful conditions were reported as a trigger to symptoms onset by 36 patients (77%). Two patients were resuscitated from an arrhythmic cardiac arrest. Three patients (6%) had an acute thromboembolic event: two had a transient cerebral ischemic attack and one a systemic artery embolism (leg ischemia). One patient suffered for a spontaneous pneumothorax. Mean Troponin-I peak release was 2,88 ± 2,53 μg/L (n.r. 0.02–0.04 μg/L); peak creatine-phosphokinase (CK) was 225,3 ± 138,5 IU/L (n.r. 33–190 IU/L); and peak MB-CK was 40,6 ± 34,0 IU (n.r. 0–25 IU/L) in contrast with the presence of severe regional left ventricular wall motion abnormality and a mean left ventricular ejection fraction of 45.9 ± 7.8%. Transthoracic echocardiography at hospital admission showed typical antero-apical dyskinesia with basal hyper-contractility (apical ballooning type) in 45 patients. Atypical a-dyskinesia of the lateral wall (mid-ventricular type) was found in the remaining two patients. Left ventricular apical thrombosis was detected by transthoracic echocardiography in all three patients with thromboembolic events. Urgent coronary angiography revealed normal coronary arteries in 33 patients (70%) and non-significant (b 50%) coronary plaques or minor lumen irregularity in the remaining 14 patients. Five patients (10%) with non-significant coronary artery disease showed minor calcifications. Two patients had focal coronary artery ectasia. The left anterior descending (LAD) coronary artery showed a dominant course beyond the LV apex with a recurrent segment to the diaphragmatic wall (wrap-around LAD) in 34 patients (72%). Fourteen patients (29.7%) also showed LAD bridging or intra-myocardial path by angiography. A significant systolic milking effect was observed by angiography in only two of them.

IVUS imaging was achieved in 43 patients and revealed significant occult plaque formation of the LAD artery in 29 (67.4%) and of adjacent or remote vessels in 6 (13.9%) patients. Mean ± SD RPA at the site of maximal atherosclerotic involvement was 44 ± 21%. No patient showed obstructive lesions or ulcerated and thrombosed plaques. Myocardial bridging of the mid-distal LAD was detected in 21 patients (48.8%). In all cases, the involved segment exhibited abnormal Ach-induced coronary vasoconstriction. Conversely, there was no relation between the site of plaque formation and the location and type of vasoconstriction. Angiographic, IVUS and pressure-flow findings of a representative TTS patient of the present study are shown in Fig. 2. 3.4. LV function recovery Recovery of left ventricular wall motion abnormalities occurred in all subjects within two weeks (mean recovery time 11 ± 3 days) resulting in significant improvement in left ventricular ejection fraction (from 45.9 ± 7.8% to 63.8 ± 7.2%; p b 0.001). There were no in-hospital deaths. There was no relation between relative improvement of LVEF and time delay from symptom onset and catheterization study. However, improvement in LVEF was inversely related to the admission LVEF (r = −075). 4. Discussion 4.1. Coronary vasomotor dysfunction The major finding of our clinical study is that most TTS patients revealed a global failure of vasomotor function with both vasoconstrictive response to Ach and increased HMR in the territory of myocardial

E. Verna et al. / International Journal of Cardiology 250 (2018) 4–10

7

Fig. 1. (a) Prevalence of vasomotor dysfunction in the study population. (b) Frequency distribution of the degree of epicardial coronary vasoconstriction of LAD and non-LAD vessel territories during acetylcholine testing.

stunning. Epicardial multivessel coronary artery spasm has been suggested as the possible mechanism of TTS since the first clinical description of the syndrome [1,2,8,9]. However, a few studies in a limited number of TTS patients have systematically assessed epicardial coronary vasomotor function, yielding inconsistent results [3,28–33] (Table 2). The low frequency of epicardial coronary spasm reported in these studies may be related to limited use of vasomotor function testing, delayed timing of testing, dosing of vasoactive drugs and concomitant medication. Only one study by Sato et al. [32] Ach provocation testing showed a high rate (77%) of an abnormal (either focal or diffuse) vasoconstrictive response of epicardial vessels. The results of our clinical study are keeping up with such finding. Most patients in the present study underwent evaluation of coronary vasoreactivity within the two weeks from symptoms onset, whereas, of the 126 patients reported in prior studies, only 13 (10.3%) underwent spasm provocation in the first 24 h and 21 (16.6%) within the first 2 weeks [28–32]

(Table 2). Moreover, we found impairment of both epicardial (endothelium-dependent) and microvascular (mainly non endotheliumdependent) vasomotor dysfunction in the territory corresponding to segmental myocardial stunning, suggesting a “global failure of coronary vasomotor function”. Impaired microvascular function with increase of HMR and reduction of CFVR was observed mostly in association with epicardial dysfunction rather than in isolation. A finding that could be expected since cardiomyocyte injury with associated myocardial edema, in the post-ischemic stunned myocardium, could generate a substrate of microvascular dysfunction. This appears to be a consequence rather than the primary cause of stunning. Patel et al. [33] studied coronary vasoreactivity in 10 TTS patients with only minor differences in the dosage and sequence of drug administration as compared to our study protocol. Most patients (90%) had evidence of microvascular dysfunction (abnormal blood flow response to ACH and/or adenosine). In contrast with our findings only 60% of patients

Fig. 2. Angiography, IVUS and intracoronary pressure-flow velocity findings (left panel) in a TTS patient of our cohort. IVUS imaging showed diffuse, uncomplicated occult atherosclerosis and intra-myocardial path (myocardial bridging) of the mid LAD (frame 5). Pressure-flow recordings after adenosine administration showed reduced coronary flow reserve (CFR) and increased hyperemic microvascular resistances (HMR) (lower left panel). Acetylcholine (Ach) administration induced intense multivessel coronary vasoconstriction with significant fall in distal coronary pressure (yellow line) and flow velocity (right panel).

8

E. Verna et al. / International Journal of Cardiology 250 (2018) 4–10

Table 2 Summary of the studies on coronary response to acetylcholine (Ach) in TTS patients. Author (ref)

Journal (year of publication)

Pts #

Spontaneous spasm # pts (%)

Ach testing # pts (%)

Epicardial vasoconstriction # pts (%)

Time of coronary function test (days)

Kawai et al. (35) Tsuchiashi et al. (3) Kurisu et al. (36) Abe et al. (37) Desmet et al. (38) Sato et al. (11) Angelini et al. (39) Patel et al. (33) All studies

Jpn Circ (2000) JACC (2001) Am Heart J (2002) JACC (2003) Heart (2003) J Nucl Cardiol (2008) Cath Cardiovasc Interv (2008) Eur Heart J (2013)

9 88 30 17 13 35 4 10 206

nd nd 3 (10%) nd nd nd nd nd 3

7 (78%) 48 (54%) 14 (46%) 7 (41%) 1 (7.6%) 35 (100%) 4 (100%) 10 126 (61%)

2 (29%) 10 (21%) 10 (71%) 5 (71%) 1 (100%) 27 (77%) 4 (100%) 3 (30%) 62 (49%)

1–8 13–53 1–7 1–7 1 30 3–56 5–896

had paradoxical constriction of epicardial vessels. The Authors concluded that the magnitude of vasomotor dysfunction was greater in the microcirculation than in the epicardial coronary arteries and suggested a potential primary role of coronary microvascular dysfunction in the pathophysiology of TTS. However, in their study patients were tested at approximately 5 months following the acute presentation (compared with a median of 2 days in our study), well beyond the acute phase and after recovery of epicardial coronary vasoreactivity and left ventricular function has occurred. It is conceivable that the demonstration of abnormal epicardial coronary vasoconstriction is more dependent on the timing of functional evaluation. 4.2. Anatomic substrates of vasomotor dysfunction IVUS imaging revealed occult coronary atherosclerotic changes of the LAD artery in more than a half (67.4%) of our patients, indicating that coronary arteries of TTS patients may not be normal by IVUS as they appear by angiography. This finding is not surprising in the typical elderly post-menopausal female population of TTS patients, merely disclosing the presence of the anatomical substrate of endothelial dysfunction. Conversely, we did not find obstructive lesions or ulcerated and thrombosed plaques. Moreover, IVUS imaging revealed intramyocardial LAD pathway (myocardial bridging) in about half of our patients (48.8%) whereas angiography alone showed myocardial bridging in only 29.7% of them. It is known that coronary-bridged segments have an increased sensitivity to vasoconstrictor stimuli and may be another anatomical substrate of coronary endothelial dysfunction [27,34–37]. In addition, the increased heart rate, contractility and myocardial oxygen consumption during catecholamine stimulation may have a detrimental effect on coronary flow distal to bridging segments [27]. Few previous studies investigated this aspect. In these studies, the frequency of myocardial bridging of the LAD artery was higher in TTS patients than in control subjects when assessed by intravascular ultrasound (40% vs 8%; p b 0.001) or cardiac computed tomography (76% vs 31%; p b 0.001) [34]. Conversely, it was significantly lower, and similar to control (11.8% vs 6.8%, p = 0.18), when assessed by angiography alone [35]. 4.3. Concurrent mechanisms and triggers of TTS Coronary vasospasm or flow-mediated paradoxical coronary vasoconstriction are well recognized mechanisms of ischemia that may actively contribute to the clinical manifestation of both stable and acute coronary syndromes even in the absence of obstructive epicardial coronary artery disease [21,22]. However, endothelial dysfunction is a common manifestation of coronary atherosclerosis and may not necessarily cause an impairment of myocardial blood flow severe enough to result in the sustained post-ischemic myocardial stunning, such as observed in TTS [32], without the contribution of important triggering and enhancing factors. It is commonly reported that TTS may be precipitated by the exposure to excess catecholamine concentration or increased sympathetic drive due to “unusual” emotional, physical or

mental stress. This was observed also in our patients. However, the kind of stresses reported, as triggers for TTS (death of a family member, public speaking, financial loss, automobile crash, etc.) are not unusual in the general population and the serum catecholamine concentration is similar to that observed in other acute clinical conditions. In normal healthy subjects, the exposure to high sympathetic and catecholamine stimulation, with the related increase in myocardial work and oxygen consumption, are commonly well balanced by a corresponding adaptive increase in coronary blood flow and do not usually result in TTS. Only in selected subjects with very impaired coronary vasomotor function due to endothelial dysfunction or other unfavorable conditions, the adaptive increase in coronary blood flow could be prevented leading to ischemia (Graphical abstract). The interplay of neuro-sympathetic activation and endothelial dysfunction has been widely demonstrated. Physical or mental stress has been found to induce coronary vasoconstriction and ischemia in subject with endothelial dysfunction and in elderly patients with occult atherosclerosis [38,39]. The degree of vasoconstriction during mental stress correlated with the response to Ach infusions [39]. Postmenopausal women are exposed to the worst effect of sympathetic stimulation and catecholamine release after physical or emotional and mental stress with greater increase in heart rate and blood pressure, owing to the loss of protective effects of oestrogens on vascular function [40,41]. In TTS patients, mental stress has been found to evoke regional and/or global perfusion abnormalities by gated single photon emission computed tomography [42] and impaired peripheral endothelial mediated vascular response by arterial tonometry [43]. Thus, the coronary vasospastic and the neuro-humoral theories, currently proposed to explain the pathogenesis of TTS, rather than being mutually exclusive, identify possible concurrent mechanisms. The multifactorial pathogenesis of TTS seems to include a more extensive failure of coronary vasomotor function with primary abnormal endothelium-dependent vasoconstrictive response of epicardial vessels. We speculate that the concomitant effect of neuro-hormonal activation and of predisposing factors such as coronary endothelial dysfunction, myocardial hypertrophy, myocardial bridging or a combination of the above, may lead to a “perfect storm” that expose the myocardium to severe ischemia due to sustained but transient paradoxical coronary vasoconstriction of epicardial vessels with secondary microvascular and myocardial stunning. The transient and self-limiting nature of this phenomenon is an essential feature of TTS. Critical coronary spastic obstruction would be long enough to cause myocardial stunning and transient microvascular dysfunction but not long enough to cause irreversible damage to the entire involved area. 5. Study limitations A limit of the present study is that we did not evaluate the coronary vasomotor response to direct adrenergic or sympathetic stimulation. However, the coronary reactivity to sympathetic stimulation depends on the functional integrity of the endothelium. The presence of endothelial dysfunction is associated with a vasoconstrictive response to

E. Verna et al. / International Journal of Cardiology 250 (2018) 4–10

sympathetic mediated stimuli such as exercise, cold pressure test (CPT) and mental stress [38,39,44–46] as well as to direct alpha-adrenoceptor activation [47–49]. A correlation between vasomotor response to Ach and to sympathetic activation has not been demonstrated in TTS patients. Nevertheless, previous studies demonstrated that the vasomotor response to Ach correlates with the sympathetic-mediated response to both CPT and mental stress [39,50]. Likewise, the coronary vasoconstrictor response to direct alpha-agonist administration correlates with the response to Ach, demonstrating the opposing influences of alpha-adrenergic coronary constriction and endothelium-mediated coronary vasodilation [51]. The demonstration of a causative relationship between coronary artery spasm and TTS would have required reproducing TTS by administration of the pathogenetic mechanism (principle of Koch). Some authors have succeeded in reproducing apical dysfunction during acetylcholine testing in few TTS patient [8]. We did not assess left ventricular function during Ach testing to demonstrate whether LV ballooning could be reproduced because all patients were studied in the early phase of TTS and showed persistent regional wall motion abnormalities at the time of study. Our patients were studied at an average of 3.5 days from presentation; therefore, the findings may not reflect the state of the coronary vasomotor function, at baseline prior to the event or after recovery. However, the high prevalence and the severity of abnormal epicardial coronary vasoconstriction after acetylcholine in the early recovery phase of TTS, may support a pathogenetic role of coronary vasospasm as much as the high prevalence of vasospasm in patients with angina or acute coronary syndrome and normal coronary angiography [21, 22], may suggest the vasospastic nature of ischemia. In the present study, we focused on the clinical phenotype of TTS patients and did not include a control group of normal subjects or a reference group of patients with acute coronary syndrome and normal coronary and left ventricular angiography. The abnormal coronary response to drug testing was defined according to prior studies [20–22,24–25]. The prevalence of abnormal vasoreaction in patients with stable angina or acute coronary syndrome and non-obstructed coronary arteries has been previously reported [21,22]. We did not repeat the evaluation of coronary vasoreactivity after normalization of left ventricular function. However, prior studies in TTS patients demonstrated the recovery of impaired microcirculatory function with the progressive improvements in regional left ventricle contraction [11,12]. We recognize that IVUS is less sensitive than optical coherence tomography (OCT) in detecting very early atherosclerotic changes, plaque fissuring and ulceration. However, recent preliminary observations with OCT imaging in TTS patients did not differ significantly from previous IVUS findings [52]. 6. Conclusions In spite of the normal angiographic appearance, the coronary artery system on TTS patients is diseased and dysfunctioning. TTS patients reveal a global failure of vasomotor function with both vasoconstrictive response to Ach and increased HMR in the stunned LAD territory. IVUS imaging reveals occult coronary atherosclerosis and myocardial bridging as potential anatomic substrates of endothelial dysfunction. These findings implicate abnormal vasoconstrictive response to activation of the sympathetic system as a potential mechanism involved in the pathogenesis of TTS in post-menopausal women. 7. Clinical implications Diagnostic investigations in TTS presently include assessment of biomarker, electrocardiogram, coronary angiography, ventriculography, echocardiography and cardiac magnetic resonance [7]. As the cardiac phenotype of TTS include the high prevalence of coronary vasomotor disturbances, a systematic evaluation of coronary vasoreactivity at the

9

time of urgent coronary angiography could better characterize the syndrome. Further studies are warranted to delineate the relation between the degree of coronary vasomotor dysfunction, the response to drug therapy and clinical outcome.

Conflict of interest disclosure The authors have no conflicts of interest to disclosure. All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.

Acknowledgment We are grateful to Chiara Minoia MD and to Erika Colombo MD, for their valuable contribution in data collection and clinical file review. We are indebted to Prof Paolo Angelini (Huston-Texas) for his friendly support and for proof reading of the article. References [1] H. Sato, H. Tateishi, K. Dote, T. Uchida, M. Ishihara, Tako-tsubo-like cardiomyopathy due to multivessel coronary spasm, in: K. Kodama, K. Haze, M. Hori (Eds.), Clinical Aspect of Myocardial Injury: From Ischemia to Heart Failure, Kagakuhyoronsha Publishing Co., Tokyo 1990, pp. 56–64. [2] K. Dote, H. Sato, H. Tateishi, T. Uchida, M. Ishihara, Myocardial stunning due to simultaneous multivessel spasms: a review of five cases, J. Cardiol. 21 (1991) 203–214. [3] K. Tsuchihashi, K. Ueshima, T. Uchida, et al., Transient left ventricular apical ballooning without coronary artery stenosis: a novel heart syndrome mimicking acute myocardial infarction. Angina pectoris-myocardial infarction investigations in Japan, J. Am. Coll. Cardiol. 38 (2001) 11–18. [4] K.A. Bybee, T. Kara, A. Prasad, et al., Systematic review: transient left ventricular apical ballooning: a syndrome that mimics ST-segment elevation myocardial infarction, Ann. Intern. Med. 141 (2004) 858–865. [5] M. Gianni, F. Dentali, A.M. Grandi, G. Sumner, R. Hiralal, E. Lonn, Apical ballooning syndrome or Takotsubo cardiomyopathy: a systematic review, Eur. Heart J. 27 (2006) 1523–1529. [6] A. Prasad, A. Lerman, C.S. Rihal, Apical ballooning syndrome (Tako-Tsubo or stress cardiomyopathy): a mimic of acute myocardial infarction, Am. Heart J. 155 (2008) 408–417. [7] A.R. Lyon, E. Bossone, B. Schneider, et al., Current state of knowledge on Takotsubo syndrome: a position statement from the task force on Takotsubo syndrome of the Heart Failure Association of the European Society of Cardiology, Eur. J. Heart Fail. 18 (2016) 8–27. [8] P. Angelini, Transient left ventricular apical ballooning: a unifying pathophysiologic theory at the edge of Prinzmetal angina, Catheter. Cardiovasc. Interv. 71 (2008) 342–352. [9] K. Ito, H. Sugihara, N. Kinoshita, A. Azuma, H. Matsubara, Assessment of Takotsubo cardiomyopathy (transient left ventricular apical ballooning) using 99mTc-tetrofosmin, 123IBMIPP, 123I-MIBG and 99mTc-PYP myocardial SPECT, Ann. Nucl. Med. 19 (2005) 435–445. [10] T. Kume, T. Akasaka, T. Kawamoto, et al., Assessment of coronary microcirculation in patients with Takotsubo-like left ventricular dysfunction, Circ. J. 69 (2005) 934–939. [11] F. Rigo, R. Sicari, R. Citro, G. Ossena, P. Buia, E. Picano, Diffuse, marked, reversible impairment in coronary microcirculation in stress cardiomyopathy: a Doppler transthoracic echo study, Ann. Med. 41 (2009) 462–470. [12] L. Galiuto, A.R. De Caterina, A. Porfidia, et al., Reversible coronary microvascular dysfunction: a common pathogenetic mechanism in apical ballooning or Takotsubo syndrome, Eur. Heart J. 31 (2010) 1319–1327. [13] H.S. Kim, J.A. Tremmel, C.W. Nam, et al., Quantitative comparison of microcirculatory dysfunction in patients with stress cardiomyopathy and ST-segment elevation myocardial infarction, J. Am. Coll. Cardiol. 58 (2011) 2430–2431. [14] H. Mori, S. Ishikawa, S. Kojima, et al., Increased responsiveness of left ventricular apical myocardium to adrenergic stimuli, Cardiovasc. Res. 27 (1993) 192–198. [15] T. Ueyama, K. Kasamatsu, T. Hano, K. Yamamoto, Y. Tsuruo, I. Nishio, Emotional stress induces transient left ventricular hypocontraction in the rat via activation of cardiac adrenoceptors: a possible animal model of “takotsubo” cardiomyopathy, Circ. J. 66 (2002) 712–713. [16] I.S. Wittstein, D.R. Thiemann, J.A. Lima, et al., Neurohumoral features of myocardial stunning due to sudden emotional stress, N. Engl. J. Med. 352 (2005) 539–548. [17] A.R. Lyon, P.S.C. Rees, S. Prasad, P.A. Poole-Wilson, S.E. Harding, Stress (Takotsubo) cardiomyopathy. A novel pathophysiological hypothesis to explain catecholamine-induced acute myocardial stunning, Nature 5 (2008) 22–29. [18] B. Ibanez, F. Navarro, M. Cordoba, P.M. Alberca, J. Fabre, Tako-tsubo transient left ventricular apical ballooning: is intravascular ultrasound the key to resolve the enigma? Heart 91 (2005) 102–104. [19] T. Pawłowski, G.S. Mintz, T. Kulawik, R.J. Gil, Virtual histology intravascular ultrasound evaluation of the left anterior descending coronary artery in patients with transient left ventricular ballooning syndrome, Kardiol. Pol. 68 (10) (2010) 1093–1098. [20] K. Egashira, T. Inou, Y. Hirooka, A. Yamada, Y. Urabe, A. Takeshita, Evidence of impaired endothelium-dependent coronary vasodilatation in patients with angina pectoris and normal coronary arteriograms, N. Engl. J. Med. 328 (1993) 1659–1664. [21] P. Ong, A. Athanasiadis, S. Hill, H. Vogelsberg, M. Voehringer, U. Sechtem, Coronary artery spasm as a frequent cause of acute coronary syndrome: the CASPAR (Coronary Artery Spasm in Patients With Acute Coronary Syndrome) Study, J. Am. Coll. Cardiol. 52 (2008) 523–527.

10

E. Verna et al. / International Journal of Cardiology 250 (2018) 4–10

[22] P. Ong, A. Athanasiadis, G. Borgulya, H. Mahrtold, J.C. Kaski, U. Sechtem, High prevalence of a pathological response to acetylcholine testing in patients with stable angina pectoris and non-obstructed coronary arteries. The ACOVA Study, J. Am. Coll. Cardiol. 59 (2012) 655–662. [23] J.W. Doucette, P.D. Corl, H.M. Payne, et al., Validation of a Doppler guide wire for intravascular measurement of coronary artery flow velocity, Circulation 85 (1992) 1899–1911. [24] D. Baumgart, M. Haude, G. Goerge, et al., Improved assessment of coronary stenosis severity using the relative flow velocity reserve, Circulation 98 (1998) 40–46. [25] M. Meuwissen, M. Siebes, S.A. Chamuleau, et al., Hyperemic stenosis resistance index for evaluation of functional coronary lesion severity, Circulation 106 (2002) 441–446. [26] C. Di Mario, G. Gorge, R. Peters, et al., Clinical application and image interpretation in intracoronary ultrasound, Eur. J. Cardiol. 19 (1998) 207–229. [27] M.S. Lee, C.H. Chen, Myocardial bridging: an up-to-date review, J. Invasive Cardiol. 27 (2015) 521–528. [28] S. Kawai, H. Suzuki, H. Yamaguchi, et al., Ampulla cardiomyopathy (Takotsubo cardiomyopathy) reversible left ventricular dysfunction with ST segment elevation, Jpn. Circ. J. 64 (2000) 156–159. [29] S. Kurisu, H. Sato, T. Kawagoe, et al., Tako-tsubo-like left ventricular dysfunction with STsegment elevation: a novel cardiac syndrome mimicking acute myocardial infarction, Am. Heart J. 143 (2002) 448–455. [30] Y. Abe, M. Kondo, R. Matsuoka, M. Araki, K. Dohyama, H. Tanio, Assessment of clinical features in transient left ventricular apical ballooning, J. Am. Coll. Cardiol. 41 (2003) 737–742. [31] W.J. Desmet, B.F. Adriaenssens, J.A. Dens, Apical ballooning of the left ventricle: first series in white patients, Heart 89 (2003) 1027–1031. [32] A. Sato, K. Aonuma, T. Nozato, et al., Stunned myocardium in transient left ventricular apical ballooning: a serial study of dual I-123 BMIPP and Tl-201 SPECT, J. Nucl. Cardiol. 15 (2008) 671–679. [33] S.M. Patel, A. Lerman, R.J. Lennon, A. Prasad, Impaired coronary microvascular reactivity in women with apical ballooning syndrome (Takotsubo/stress cardiomyopathy), Eur. Heart J. Acute Cardiovasc. Care 2 (2013) 147–152. [34] F. Migliore, E. Maffei, Perazzolo, et al., LAD coronary artery myocardial bridging and apical ballooning syndrome, JACC Cardiovasc. Imaging 6 (2013) 32–41. [35] T. Stiermaier, S. Desch, S. Blazek, G. Schuler, H. Thiele, I. Eitel, Frequency and significance of myocardial bridging and recurrent segment of the left anterior descending coronary artery in patients with Takotsubo cardiomyopathy, Am. J. Cardiol. 114 (2014) 1204–1209. [36] J.W. Kim, H.S. Seo, J.O. Na, et al., Myocardial bridging is related to endothelial dysfunction but not to plaque as assessed by intracoronary ultrasound, Heart 94 (2008) 765–769. [37] P. Angelini, C. Uribe, P. Lozano, Differential local spasticity in myocardial bridges, Tex. Heart Inst. J. 39 (2012) 384–389.

[38] L. Ghiadoni, A. Donald, M. Cropley, et al., Mental stress induces transient endothelial dysfunction in humans, Circulation 102 (2000) 2473–2478. [39] A.C. Yeung, V.I. Vekshtein, D.S. Krantz, J.A. Vita, T.J. Ryan, P. Ganz, A.P. Selwyn, The effect of atherosclerosis on the vasomotor response of coronary arteries to mental stress, N. Engl. J. Med. 325 (1991) 1551–1556. [40] E.C.J. Hart, N. Charkoudian, Sympathetic neural regulation of blood pressure, Influences Sex Aging Physiol. 29 (2014) 8–15. [41] K.L. Moreau, K.L. Hildreth, A.L. Meditz, Dean KD, W.M. Kohrt, Endothelial function is impaired across the stages of the menopause transition in healthy women, J. Clin. Endocrinol. Metab. 97 (2012) 4692–4700. [42] R. Sciagrà, G. Parodi, S. Del Pace, et al., Abnormal response to mental stress in patients with Takotsubo cardiomyopathy detected by gated single photon emission computed tomography, Eur. J. Nucl. Med. Mol. Imaging 37 (2010) 765–772. [43] E.A. Martin, A. Prasad, C.S. Rihal, L.O. Lerman, A. Lerman, Endothelial function and vascular response to mental stress are impaired in patients with apical ballooning syndrome, J. Am. Coll. Cardiol. 56 (2010) 1840–1846. [44] J.E. Gage, O.M. Hess, T. Murakami, M. Ritter, J. Grimm, H.P. Krayenbuehl, Vasoconstriction of stenotic coronary arteries during dynamic exercise in patients with classic angina pectoris: reversibility by nitroglycerin, Circulation 73 (1986) 865–876. [45] J.B. Gordon, P. Ganz, E.G. Nabel, et al., Atherosclerosis influences the vasomotor response of epicardial coronary arteries to exercise, J. Clin. Invest. 83 (1989) 1946–1952. [46] E.G. Nabel, P. Ganz, J.B. Gordon, R.W. Alexander, A.P. Selwyn, Dilation of normal and constriction of atherosclerotic coronary arteries caused by the cold pressor test, Circulation 77 (1988) 43–52. [47] D. Baumgart, M. Haude, G. Gorge, et al., Augmented α-adrenergic constriction of atherosclerotic human coronary arteries, Circulation 99 (1999) 2090–2097. [48] D. Baumgart, C. Naber, M. Haude, et al., G protein β 3 subunit 825T allele and enhanced coronary vasoconstriction on α2-adrenoceptor activation, Circ. Res. 85 (1999) 965–969. [49] G. Heusch, D. Baumgart, P. Camici, et al., α-Adrenergic coronary vasoconstriction and myocardial ischemia in humans, Circulation 101 (2000) 689–6894. [50] A.M. Zeiher, H. Drexler, H. Wollschlaeger, B. Saurbier, H. Just, Coronary vasomotion in response to sympathetic stimulation in humans: importance of the functional integrity of the endothelium, J. Am. Coll. Cardiol. 14 (1989) 1181–1190. [51] J.A. Vita, C.B. Treasure, A.C. Yeung, et al., Patients with evidence of coronary endothelial dysfunction as assessed by acetylcholine infusion demonstrate marked increase in sensitivity to constrictor effects of catecholamines, Circulation 85 (1992) 1390–1397. [52] M. Fineschi, F. D'Ascenzi, V. Sirbu, S. Mondillo, C. Pierli, The role of optical coherence tomography in clarifying the mechanisms for dobutamine stress echocardiographyinduced Takotsubo cardiomyopathy, Echocardiography 30 (2013) E121–E124.