Ventricular arrhythmias and sudden cardiac arrest in Takotsubo cardiomyopathy: Incidence, predictive factors, and clinical implications

Accepted Manuscript Ventricular arrhythmias and sudden cardiac arrest in Takotsubo cardiomyopathy: Incidence, predictive factors and clinical implications Laurence Jesel, MD, PhD, Charlotte Berthon, MD, Nathan Messas, MD, Han S. Lim, MBBS PhD, Mélanie Girardey, MD, Halim Marzak, MD, Benjamin Marchandot, MD, Annie Trinh, MD, Patrick Ohlmann, MD, PhD, Olivier Morel, MD, PhD PII:

S1547-5271(18)30341-2

DOI:

10.1016/j.hrthm.2018.04.002

Reference:

HRTHM 7546

To appear in:

Heart Rhythm

Received Date: 12 November 2017

Please cite this article as: Jesel L, Berthon C, Messas N, Lim HS, Girardey M, Marzak H, Marchandot B, Trinh A, Ohlmann P, Morel O, Ventricular arrhythmias and sudden cardiac arrest in Takotsubo cardiomyopathy: Incidence, predictive factors and clinical implications, Heart Rhythm (2018), doi: 10.1016/j.hrthm.2018.04.002. 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

Ventricular arrhythmias and sudden cardiac arrest in Takotsubo cardiomyopathy :

RI PT

Incidence, predictive factors and clinical implications

Laurence Jesela MD, PhD, Charlotte Berthona MD, Nathan Messasa MD, Han S. Limb, MBBS PhD, Mélanie Girardeya MD, Halim Marzaka MD, Benjamin Marchandota MD, Annie Trinha

SC

MD, Patrick Ohlmanna MD, PhD, Olivier Morela MD, PhD

a

M AN U

University Hospital of Strasbourg, Pôle d’Activité Médico-Chirurgicale Cardio-Vasculaire,

Strasbourg, France b

Department of Cardiology, Austin and Northern Health, University of Melbourne,

TE D

Melbourne, Australia.

Correspondence:

AC C

Laurence Jesel

EP

SHORT TITLE: Life threatening arrhythmia in Takotsubo cardiomyopathy

Pôle Cardiovasculaire, Nouvel Hôpital Civil 67091 Strasbourg, France

Fax: 00(33)369551788 Tel: 00(33)369550948 [email protected] Total word count: 5000 Conflict of interest: none

1

ACCEPTED MANUSCRIPT

Abstract Background: Takotsubo cardiomyopathy (TTC) is a stress-related transient cardiomyopathy. Life threatening arrhythmias (LTA) can occur and worsen prognosis.

RI PT

Objective: The purpose of this study was to assess the incidence and outcome of LTA in TTC, its predictive factors and clinical implications.

Methods: We studied 214 consecutive cases of TTC over 8 years. The study cohort was

SC

divided into two groups – those with LTA (LTA-group) and those without (non-LTA-group). LTA was defined as ventricular tachycardia (VT), ventricular fibrillation (VF) or cardiac arrest.

M AN U

Results: LTA occurred in 10.7% of patients mainly in the first 24-hours of hospitalization: VT (n=2), VF (n=11), cardiac arrest (n=10: 5 asystole, 3 complete heart block, 2 sinoatrial block). LTA were associated with lower LVEF and high rate of conduction disturbances. In-hospital (39.1% vs 8.9%, p<0.001) and one-year mortality (47.8% vs 14.1%, p<0.001) were

TE D

significantly increased in the LTA-group. LVEF and QRS duration>105 ms were independent predictors of LTA. In cases where a device was implanted, conduction disturbances persisted after the index event despite complete recovery of LVEF. There was no ventricular

EP

arrhythmia recurrence during follow-up.

AC C

Conclusion: LTA occur early in patients presenting with TTC and is associated with significantly worse short and long-term prognosis. Left ventricular impairment and QRS duration>105 ms are independent predictors of LTA. Ventricular arrhythmias occurred in the acute phase without further recurrence recorded among hospital survivors whereas severe conduction disorders persisted during long-term follow-up. These findings may have implications on the choice of device therapy on this specific patient subgroup.

Keywords: ventricular tachycardia; ventricular fibrillation; cardiac arrest; complete atrioventricular block; Takotsubo cardiomyopathy 2

ACCEPTED MANUSCRIPT

Introduction Takotsubo

cardiomyopathy

(TTC),

a

stress-related

reversible

cardiomyopathy

is

characterized by transient left ventricular dysfunction with wall motion abnormalities that

RI PT

extending beyond a single epicardial vascular distribution and occurring in the absence of hemodynamically significant coronary obstruction.1 In most cases, the prognosis is favorable with recovery of left ventricular ejection fraction (LVEF) within several days.2 However,

SC

serious complications can occur in patients with TTC, such as cardiogenic shock, heart failure and fatal arrhythmias, resulting in a reported in-hospital mortality rate of 4.1%.2 Life 4

M AN U

threatening arrhythmias (LTA) have been described in 1.8 to 13.5% of these patients.3,

Electrocardiographic abnormalities such as T-wave modification or QT interval prolongation are thought to be involved in the arrhythmia onset.5 However no predictive factors for LTA in the setting of TTC have been reported yet.

TE D

The purpose of this study was to evaluate the incidence and outcome of patients with TTC complicated by LTA in a large consecutive cohort of patients presenting with TTC, its

AC C

Methods

EP

predictive factors and clinical implications.

Study population

We identified 214 consecutive patients diagnosed with TTC between May 2008 and November 2016 in the Cardiac Catheterization Laboratory database of Strasbourg University Hospital. The diagnosis of TTC was made according to Madias’ criteria.1, 6 The study protocol was approved by the Institutional Review Board of Strasbourg University. Retrospective

3

ACCEPTED MANUSCRIPT consent was obtained from the patients during follow-up visits and from their relatives in deceased patients. LTA were defined as sustained ventricular tachycardia (VT), ventricular fibrillation (VF) or sudden cardiac arrest assessed by electrocardiogram on admission and during

RI PT

hospitalization. The patients were divided into two groups – those presenting with LTA (LTAgroup), and those without (non-LTA-group).

SC

Clinical assessment

M AN U

At the time of hospitalization, a complete medical history, episodes of arrhythmias were recorded.

12-lead ECG and routine laboratory data were obtained serially. Peak levels were defined as maximal values recorded during hospitalization.

TE D

VT, VF or cardiac arrest were assessed on admission or during hospital stay on 12-lead ECG or on telemetry analyzed by two independent cardiologists. An episode of VT≥30 seconds was considered sustained.

EP

Cardiac arrest was defined as loss of consciousness with absence of electrical activity

AC C

identified on ECG or pulseless electrical activity requiring resuscitation maneuvers or defibrillation. The causes of cardiac arrest were then analyzed on 12-lead ECG or on telemetry to identify asystole, complete atrio-ventricular (CAVB) or sino-atrial blocks (SB). PR/QRS/QT intervals were measured on admission and during hospitalization. All corrected QT intervals were measured using the Bazett’s method. A QTc>470 ms was considered prolonged. Transthoracic echocardiographic studies, multiplane coronary angiography were performed in all patients and LVEF was calculated.6 4

ACCEPTED MANUSCRIPT

Clinical follow-up Follow-up was obtained in all patients using standardized telephone interviews and reports

RI PT

by a cardiologist or treating physician. In cases of patient death, the cause was ascertained by thorough review of all available clinical information at the time of death.

For patients implanted with a pacemaker (PM) or an implantable cardioverter defibrillator

SC

(ICD), reports of regular outpatient visits in our center or other hospitals were collected.

M AN U

Statistical analysis

Categorical variables are expressed as counts and percentages. Continuous variables are reported as mean±SD or as median and interquartile range (25–75th) according to their

TE D

distribution. Categorical variables were compared with chi-square test or Fisher’s exact test. Continuous variables were compared with Mann-Whitney test. To determine the predictors of LTA, logistic regression analysis was performed. The optimal cut-off value for QRS width

EP

was determined by using a receiver-operator characteristic curve analysis. All tests were 2sided. P<0.05 was considered significant. Calculations were performed using SPSS17.0 for

AC C

Windows (SPSS Inc, Chicago, IL, USA).

Results

Baseline characteristics Data from 214 consecutive patients with TTC over 8 years were analyzed. 27 patients died during their hospital stay and 26 during follow-up. (Table 1) 5

ACCEPTED MANUSCRIPT LTA occurred in 23 patients, 10.7% of the population, on admission or during in-hospital stay. 13 (6%) patients presented with VT or VF: 2 with sustained VT, 9 with VF and 2 with torsades de pointes (TDP) degenerating into VF, all requiring electrical defibrillation. 10 patients presented with sudden cardiac arrest: 5 with asystole, 3 with CAVB, and 2 with SB.

RI PT

The 2 VT patients also presented in CAVB. (Table 2)

Among the 23 patients from the LTA-group, the arrhythmic event occurred in 52% of cases

SC

as an out-of-hospital event and was the reason for hospitalization. In 74% of patients, the

recurrence during hospitalization.

M AN U

event occurred within the first 24 hours of hospitalization. 22% patients presented with LTA

TE D

Characteristics of TTC patients presenting with and without LTA.

There were no differences between the two groups in terms of age, sex and cardiovascular

EP

risk factors (Table 1).

AC C

The LTA-group had a significant lower LVEF compared to the non-LTA-group on admission. Analysis of the electrocardiograms on admission revealed wider QRS duration and longer PR duration in the LTA-group compared to the non-LTA-group. However there was no difference in QT interval duration between the two groups. In the LTA-group, admission levels of troponin were significantly increased compared to the non-LTA-group. CRP levels were higher but this difference was not significant.

6

ACCEPTED MANUSCRIPT

In-hospital and long term prognosis During hospital stay, the LTA-group presented with significantly higher rates of cardiogenic shock, need for intra-aortic balloon pump and temporary right ventricular pacing compared

RI PT

to the non-LTA-group (Table 3). Analysis of the electrocardiograms during hospitalization revealed significantly greater conduction disturbances and onset of atrial fibrillation in the LTA-group. There was no

SC

difference in maximum QTc interval between the two groups.

M AN U

During hospitalization, peak troponin and peak leucocyte levels were significantly higher in the LTA-group.

At the end of hospital stay, LVEF remained significantly lower, troponin and leukocyte levels

TE D

higher in the LTA-group compared to the non-LTA-group.

Hospital length of stay was longer in the LTA-group. The in-hospital mortality, cardiovascular death and one year mortality were also significantly higher in the LTA-group.

EP

Of note the patients from the LTA-group who died following hospital discharge within the

AC C

first year died from a non-cardiovascular cause. During long-term follow-up (700±656 days), long-term mortality was 25% (53/214) in the whole cohort and 57% (13/23) in the LTA-group. Two patients died following the first year till the end of the follow-up period. During follow-up, 3 recurrences of TTC were observed in the whole cohort. All of these occurred in the non-LTA-group. No LTA was observed in these 3 cases.

7

ACCEPTED MANUSCRIPT

Predictive factors of LTA The optimal cut-off value of QRS duration>105 ms for predicting the risk of LTA was associated with a sensitivity of 59% and a specificity of 81%. (figure 1)

RI PT

On univariate analysis, LVEF and troponin on admission, atrial fibrillation during hospitalization and QRS duration>105 ms were associated with LTA onset (Table 4).

On multivariate analysis, independent predictors of LTA onset were LVEF and QRS

M AN U

SC

duration>105 ms.

Device implantation and follow-up

In the entire cohort 6 patients underwent PM implantation because of CAVB (n=5) and SB

TE D

(n=1). Of these patients, 4 were in the LTA-group, 2 in the non-LTA-group. (Table 5) Two patients in the LTA-group received an ICD: a 42-year-old woman who presented with an

EP

out-of-hospital VF arrest who was resuscitated. Her LVEF was initially 30%, and later stabilized at 50%. Runs of non-sustained VT were observed during hospitalization. A

AC C

subcutaneous ICD was implanted. The second patient was a 75-year-old woman who presented with VT on admission and required defibrillation. She then developed recurrent VT, CAVB and persistent AF during hospitalization with a LVEF of 35%. A single chamber ICD was implanted. During follow-up, her LVEF increased to 64%. Regular device interrogation reports were obtained in all patients (Table 5). The 2 ICD patients revealed no further ICD-intervention and no recurrence of VT after 43 and 71 months of follow-up.

8

ACCEPTED MANUSCRIPT For the 6 PM patients, mean follow-up was 42±22 months. All patients (n=6) implanted for CAVB demonstrated a high percentage of ventricular pacing (92±12%). The patient with a pacemaker implanted for SB showed 37% atrial pacing and following 6 months of follow-up, developed persistent AF. These conduction disturbances persisted despite complete

RI PT

recovery of LVEF (58.5±10%) in all patients.

SC

Discussion

M AN U

The present study examined 214 consecutive cases of TTC over a period of 8 years. LTA occurred in 10.7% of patients. LTA were associated with lower LVEF and a high rate of conduction disturbances. In TTC cases complicated by LTA, hospital stay was longer and inhospital, cardiovascular and one-year mortality were dramatically increased. LVEF and QRS

TE D

duration>105 ms were independent predictors of LTA onset. In cases of device implantation, conduction disturbances persisted after the index event despite complete recovery of LVEF

EP

whereas no ventricular arrhythmia (VA) recurrence were noted during follow-up. Although LTA have been well described as a complication in TTC, their exact incidence and

AC C

mechanisms have not been well studied. The previously reported incidence is highly variable, from 3% in the cohort by Templin et al.,2 to 8% described by Stiermaier et al.4 In this study we included all patients with documented VA and patients requiring resuscitation for cardiac arrest. This included patients presenting with severe conduction disturbances and hemodynamic compromise as a cause of cardiac arrest. Patients from the non-LTA-group evidenced rare conduction disturbances that were transient or well tolerated. In the LTAgroup, all VT or TDP degenerated into VF requiring external defibrillation. In this group, LVEF

9

ACCEPTED MANUSCRIPT was lower, cardiogenic shock and the use of intra-aortic balloon pumps were more frequent, pointing towards arrhythmias that were hemodynamically unstable. Bonello et al. described haemodynamic instability in these cases of VA in the setting of TTC.7 One new finding from the present study is that the majority of LTA occurred during the first 24 hours of

RI PT

hospitalization. Cardiac monitoring upon admission is therefore mandatory to prevent such complications as 22% of the patients presented with LTA recurrence. Two studies8,

9

also

reported a 3 day-period of increased risk of arrhythmia onset. Catecholamine release occurs

SC

in the acute phase, inducing cardiac edema and repolarization changes,10 providing a

M AN U

ventricular substrate for arrhythmia in these first few days. The duration where patients are at risk of arrhythmias and the optimal duration for monitoring are still unclear. However, the prognostic implication of LTA is severe and in our cohort where patients were carefully monitored during hospitalization in the intensive care unit, LTA onset impacted dramatically

TE D

on short and long-term mortality, previously scarcely described in the literature. Notably patients presenting with LTA who survived to hospital discharge have a favorable prognosis with none of them encountering cardiovascular death or an arrhythmic event during the first

EP

year of follow-up.

AC C

Repolarization changes and QT interval increase have been extensively described in TTC patients. Madias et al. showed dynamic changes in QT interval with a decrease on the fifth day5. He also reported 8% of VF/TDP and observed that QT interval increase was significantly associated with VA onset. Migliore et al. studied 51 patients and showed that 5 patients with VA had a QTc>500 ms although this was purely descriptive.3 In our cohort, 80% of the patients demonstrated long QTc intervals during the acute presentation with TTC with a mean of 507 ms for maximal QT interval. However there was no difference in QTc interval on admission or in maximal QTc interval between the two groups. On univariate analysis, QTc 10

ACCEPTED MANUSCRIPT interval was not associated with LTA onset. Matsuoka et al. reported the absence of arrhythmia in TTC patients despite prolonged QT intervals and that QT dispersion was present several weeks after the index event.11 QTc prolongation did not allow for discrimination between the LTA and non-LTA-group. QT prolongation is part of the clinical

RI PT

presentation in TTC patients but can also increase after cardiac arrest and following resuscitation. QT monitoring is certainly of interest given that amiodarone, beta-blockers

SC

and diuretics are often used in TTC.

In the present study, we observed that patients from LTA-group had lower LVEF and higher

M AN U

troponin levels. We could not establish a predictive role of inflammation and amount of tissue damage for LTA onset probably because these factors could both be causes and consequences of arrhythmias. Schneider et al. in his cohort described that LVEF≤30% was a predictive factor of arrhythmic complications in TTC.9 We observed two predictors of LTA

TE D

onset, decreased LVEF and QRS duration>105 ms. LTA-group demonstrated more pronounced conduction disturbances. Conduction disorders are described in TTC patients but to date have not been linked to the risk of LTA. Dib et al. observed marked PR interval

EP

prolongation in a group of patients presenting with arrythmia but when indexed to heart

AC C

rate, no differences were found.12 In the present study we observed longer PR intervals in the LTA-group, whereas heart rates were also higher. This is the first study to report a link between conduction disturbances and LTA in the setting of TTC. From a pathophysiological point of view, conduction disturbance could be explained by edema and inflammatory cell infiltration in the myocardium,13, 14 affecting the conduction system. The reversible edema could explain why some of the conduction defects may be transient.3, 15 From a clinical point of view, QRS duration>105 ms or any worsening in conduction defect should prompt vigilant rhythm monitoring. 11

ACCEPTED MANUSCRIPT Why some conduction defects are transient and reversible while others require device implantation remains unclear. Very few studies on TTC were designed to examine associated arrhythmias and patient cohorts requiring device implantation are small with limited followup. Based on these studies, conduction defects seem to persist after pacemaker

RI PT

implantation whereas VA seem to occur mainly in the acute phase of the index event3-5 without further recurrence. In his cohort of 8/93 patients presenting with LTA, Madias et al. reported, 2 patients requiring ICD and one patient requiring PM implantation5. Follow-up

SC

was only available in 1 ICD-patient, with no VA events. Migliore et al. reported 2 PM and 2

M AN U

ICD-patients in a cohort of 61 patients with TTC with no VA during follow-up and conduction recovery in the 2 PM-patients3. Stiermaier et al. reported 35/286 TTC patients with LTA requiring 7 PM and 1 ICD but follow-up was available in only 3 patients with no VA in the ICD-patient and persistence of CAVB in 2 PM-patients4. In our cohort, 8 patients received PM

TE D

or ICD implantation with follow-up obtained in all patients. After prolonged follow-up, the 2 ICD implanted patients showed no recurrence of VA whereas the 6 PM-patients evidenced persistent conduction defects with a high percentage of ventricular pacing. Patients with VA

EP

in the acute phase who survived to hospital discharge in the study did not appear to be at

AC C

increased risk for late sustained VA during clinical and device follow-up. The absence of VA recurrence is in keeping with the reversible pattern of the TTC substrate and the absence of scar formation16. In the acute phase, through the beta-receptor pathway17, catecholamine discharge can precipitate cytosolic calcium overload and cardiac stunning which can have a direct effect on ionic channels, inducing VA by an automatic mechanism. Catecholamine release can also induce an inflammatory reaction18 through leukocytosis and proinflammatory cytokine release which could play a role in oxidative stress levels and microvascular dysfunction, inducing an ischaemic phenomenon and VA. MRI studies in TTC16 12

ACCEPTED MANUSCRIPT have confirmed the absence of myocardial scar making reentry a less likely mechanism for arrhythmia and lends support to a catecholaminergic, automatic mechanism as being more probable.

RI PT

It is unanticipated to observe persistent conduction disturbances during long-term followup, despite complete recovery of systolic function. Electrical remodeling seems to not evolve similar to myocardial remodeling, highlighting the sensitivity of the cardiac conduction

SC

system to edema, microvascular ischemia or catecholamine discharge. Despite the small sample size of patients requiring pacemakers, the persistent conduction disorders observed

M AN U

on long-term follow-up support the idea that pacemaker therapy should be pursued in cases of severe bradyarrhythmia. ICD could be considered when the cause of VA is unclear, including cases where it is uncertain whether TTC could be the cause or consequence of the arrhythmia. In case of administration of substantial doses of epinephrine, care must be

TE D

taken to exclude epinephrine as an inducer of arrhythmias. A case-by-case decision would be

AC C

Study limitations

EP

mandatory in this scenario.

This was a retrospective single-center study of patients with a uniformed management approach. Although the cohort is large, the retrospective analysis may have underestimated the prevalence of asymptomatic arrhythmia. Measurement bias could not be formally excluded. Nevertheless the data collection was standardized based on the computerized database of our hospital, with data entry and analysis performed by physicians blinded to the eventual categorization of patient groups. Patients were acutely managed in the intensive or coronary care units, allowing standardized and close monitoring of heart rhythm 13

ACCEPTED MANUSCRIPT and hemodynamics. Despite complete follow-up of device implanted patients, our population sample remains small with limited long-term follow-up and a small risk of recurrent VT/VF among hospital survivors cannot be conclusively excluded. The current data do not allow specific recommendations on the optimal approach for device implantation,

RI PT

which has to be decided on a case-by-case basis. We could not exclude the possibility that some cases of idiopathic VF or CAVB could be the cause of TTC, particularly in the younger patients and that epinephrine during resuscitation could have in some cases precipitated the

SC

stress cardiomyopathy. In table 2 we list the use of epinephrine prior to the diagnosis of TTC.

M AN U

In cases of VT or VF, defibrillation was performed prior to epinephrine administration.

Conclusion

TE D

VA and sudden cardiac arrest are not uncommon in patients presenting with TTC and occur mainly in the first 24 hours. LTA are associated with lower LVEF and a high rate of conduction disturbances. LTA significantly impact on short and long-term prognosis. Left

EP

ventricular impairment and QRS duration>105 ms are independent predictors of LTA.

AC C

Vigilant monitoring of conduction abnormalities could improve the detection of patients at risk of LTA. VA occur in the acute phase without further recurrence among hospital survivors, whereas severe conduction disorders persist during long-term follow-up. These findings may have implications on the choice of device therapy on this specific subgroup of patients.

14

ACCEPTED MANUSCRIPT References 1.

Madias JE. Why the current diagnostic criteria of takotsubo syndrome are outmoded: A proposal for new criteria. Int J Cardiol 2014;174:468-470 Templin C, Ghadri JR, Diekmann J, et al. Clinical features and outcomes of takotsubo (stress) cardiomyopathy. N Engl J Med 2015;373:929-938

3.

RI PT

2.

Migliore F, Zorzi A, Peruzza F, Perazzolo Marra M, Tarantini G, Iliceto S, Corrado D.

SC

Incidence and management of life-threatening arrhythmias in takotsubo syndrome. Int J Cardiol 2013;166:261-263

Stiermaier T, Rommel KP, Eitel C, Moller C, Graf T, Desch S, Thiele H, Eitel I.

M AN U

4.

Management of arrhythmias in patients with takotsubo cardiomyopathy: Is the implantation of permanent devices necessary? Heart Rhythm 2016;13:1979-1986 5.

Madias C, Fitzgibbons TP, Alsheikh-Ali AA, Bouchard JL, Kalsmith B, Garlitski AC, Tighe

TE D

DA, Estes NA, Aurigemma GP, Link MS. Acquired long QT syndrome from stress cardiomyopathy is associated with ventricular arrhythmias and torsades de pointes. Heart Rhythm 2011;8:555-561

Girardey M, Jesel L, Campia U, Messas N, Hess S, Imperiale A, Blondet C, Trinh A,

EP

6.

AC C

Ohlmann P, Morel O. Impact of malignancies in the early and late time course of takotsubo cardiomyopathy. Circ J 2016;80:2192-2198

7.

Bonello L, Com O, Ait-Moktar O, Theron A, Moro PJ, Salem A, Sbragia P, Paganelli F. Ventricular arrhythmias during tako-tsubo syndrome. Int J Cardiol 2008;128:e50-53

8.

Singh K, Carson K, Hibbert B, Le May M. Natural history of cardiac arrest in patients with takotsubo cardiomyopathy. Am J Cardiol 2015;115:1466-1472

15

ACCEPTED MANUSCRIPT 9.

Schneider B, Athanasiadis A, Schwab J, Pistner W, Gottwald U, Schoeller R, Toepel W, Winter KD, Stellbrink C, Muller-Honold T, Wegner C, Sechtem U. Complications in the clinical course of tako-tsubo cardiomyopathy. Int J Cardiol 2014;176:199-205

10.

Perazzolo Marra M, Zorzi A, Corbetti F, De Lazzari M, Migliore F, Tona F, Tarantini G,

RI PT

Iliceto S, Corrado D. Apicobasal gradient of left ventricular myocardial edema underlies transient t-wave inversion and qt interval prolongation (Wellens' ECG pattern) in tako-tsubo cardiomyopathy. Heart Rhythm 2013;10:70-77

Matsuoka K, Okubo S, Fujii E, Uchida F, Kasai A, Aoki T, Makino K, Omichi C, Fujimoto

SC

11.

M AN U

N, Ohta S, Sawai T, Nakano T. Evaluation of the arrhythmogenecity of stress-induced "takotsubo cardiomyopathy" from the time course of the 12-lead surface electrocardiogram. Am J Cardiol 2003;92:230-233 12.

Dib C, Prasad A, Friedman PA, Ahmad E, Rihal CS, Hammill SC, Asirvatham SJ.

TE D

Malignant arrhythmia in apical ballooning syndrome: Risk factors and outcomes. Indian Pacing Electrophysiol J 2008;8:182-192 13.

Kurisu S, Sato H, Kawagoe T, Ishihara M, Shimatani Y, Nishioka K, Kono Y, Umemura T,

EP

Nakamura S. Tako-tsubo-like left ventricular dysfunction with st-segment elevation: A

AC C

novel cardiac syndrome mimicking acute myocardial infarction. Am Heart J 2002;143:448-455

14.

Nef HM, Mollmann H, Kostin S, Troidl C, Voss S, Weber M, Dill T, Rolf A, Brandt R, Hamm CW, Elsasser A. Tako-tsubo cardiomyopathy: Intraindividual structural analysis

in the acute phase and after functional recovery. Eur Heart J 2007;28:2456-2464 15.

Chadha S, Lodha A, Shetty V, Sadiq A, Hollander G, Shani J. Complete heart block in takotsubo cardiomyopathy. Heart Lung 2013;42:48-50

16

ACCEPTED MANUSCRIPT 16.

Syed IS, Prasad A, Oh JK, Martinez MW, Feng D, Motiei A, Glockner JD, Breen JF, Julsrud PR. Apical ballooning syndrome or aborted acute myocardial infarction? Insights from cardiovascular magnetic resonance imaging. Int J Cardiovasc Imaging 2008;24:875-882 Paur H, Wright PT, Sikkel MB, et al. High levels of circulating epinephrine trigger

RI PT

17.

apical cardiodepression in a beta2-adrenergic receptor/gi-dependent manner: A new model of takotsubo cardiomyopathy Circulation. 2012;126:697-706

Zelinka T, Petrak O, Strauch B, Holaj R, Kvasnicka J, Mazoch J, Pacak K, Widimsky J, Jr.

SC

18.

M AN U

Elevated inflammation markers in pheochromocytoma compared to other forms of

EP

TE D

hypertension. Neuroimmunomodulation 2007;14:57-64

Figure 1: The optimal cut-off value of QRS duration for predicting the risk of life-

AC C

threatening arrhythmias

17

ACCEPTED MANUSCRIPT Table 1: Characteristics of Takotsubo patients presenting with and without lifethreatening arrhythmias (LTA) on admission.

Group LTA+ (n=23)

P-value

69.0±12.6 174 (81.3%) 25.2±5.5

69.4±12.6 158 (82.7%) 25.2±5.7

66.4±13.0 16 (69.6%) 25.0±3.8

0.288 0.110 0.849

121 (56.5%) 47 (22%) 82 (38.3%)

108 (56.5%) 41 (21.5%) 76 (39.8%)

13 (56.5%) 6 (26.1%) 6 (26.1%)

0.584 0.391 0.146

54 (25.2%) 28 (13.1%) 113 (52.8%)

47 (24.6%) 26 (14%) 97 (50.8%)

7 (30.4%) 2 (9%) 16 (69.6%)

0.351 0.393 0.068

126±29 71±16 89±18 92±21 40±12

127±29 70±15 90±18 91±20 41±11

115±32 71±22 86±25 99±25 31±10

0.064 0.844 0.380 0.128 <0.001

107 (50%) 46 (21.5%) 60 (28%) 1 (0.5%)

99 (51.8%) 38 (19.9%) 54 (28.3%) 1 (0.5%)

8 (34.8%) 8 (34.8%) 6 (26.1%) 0 (0%)

0.092 0.089 0.522 0.893

3.4±7.9 679±888 12.0±5.8 40.7±67.4

2.8±5.4 699±911 12±5.8 39.4±65.3

8.0±18.2 492±611 12.4±5.9 52.0±83.5

0.003 0.324 0.748 0.396

97±21 164±33 462±45

95±18 162±30 461±44

117±33 185±51 470±51

<0.001 0.002 0.391

EP

AC C

SC

RI PT

Group LTA(n=191)

TE D

Demographics ⋅ Age (years) ⋅ Female gender ⋅ BMI (kg/m²) CVD risk factors ⋅ Arterial hypertension ⋅ Diabetes mellitus ⋅ Dyslipidemia Stress Factor ⋅ History of malignancy ⋅ Emotional stress ⋅ Physical stress Hemodynamic parameters ⋅ SBP (mmHg) ⋅ DBP (mmHg) ⋅ MBP (mmHg) ⋅ Heart rate (bpm) ⋅ LVEF (%) Ballooning pattern ⋅ Apical ⋅ Mid-apical ⋅ Mid-ventricular ⋅ Basal Biology on admission ⋅ Troponin (ng/mL) ⋅ BNP (ng/L) ⋅ Leukocytes (G/L) ⋅ CRP (mg/L) Electrocardiogram on admission ⋅ QRS duration (ms) ⋅ PR (ms) ⋅ QTc (ms)

All patients (n=214)

M AN U

Variables

SBP/DBP/MBP: systolic/diastolic/mean blood pressure, LVEF: left ventricular ejection fraction, BNP: B-type natriuretic peptide, CRP: C-reactive protein

ACCEPTED MANUSCRIPT Table 2: Characteristics of patients presenting with life threatening arrhythmias during the acute Takotsubo Cardiomyopathy episode.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Sex

Age

Type of Arrhythmia during TTC

Dose of Epinephrine Before TTC diagnosis

In-hospital mortality

Post-discharge mortality within the 1st year

1-year mortality

Cause of death

Device

RI PT

P

AC C

EP

TE D

M AN U

SC

F 76 asystole + + ICH 3 F 75 VF + + ICH M 52 VF + + SeptS F 62 TdP+VF + + SeptS F 90 VF + + CardioS 3 CardioS F 63 VF + + SV F 55 VF 4 + F 53 VF + Cancer 2 F 82 VF + + PAE 3 F 75 CAVB+VT 2 F 42 VF 2 M 61 CAVB+VT 1 F 71 VF M 82 TdP+VF F 60 SB M 60 asystole 2 F 62 asystole + + PAE M 42 SB F 70 CAVB 0.5 + M 64 asystole + SeptS 3 M 61 asystole + + CardioS 2 F 82 CAVB F 80 CAVB CAVB: complete AV bloc, SB: sinoatrial block, VF/VT: ventricular fibrillation/tachycardia, ICH: intra-cranial hemorrhage, SeptS/CardioS: septic/refractory cardiogenic shock, PAE: post-anoxic encephalopathy, PM: pacemaker, ICD: implantable cardioverter defibrillator

ICD ICD PM

PM

PM PM

ACCEPTED MANUSCRIPT Table 3: In-hospital outcome and long-term follow-up of Takotsubo patients presenting with and without life-threatening arrhythmias (LTA).

Group LTA+ (n=23)

P-value

32 (15%) 22 (10.3%) 3 (1.4%) 2 (0.9%) 6 (2.8%)

23 (12%) 16 (8.4%) 0 (0%) 0 (0%) 2 (1%)

9 (39.1%) 6 (26.1%) 3 (13%) 2 (8.7%) 4 (17.4%)

0.002 0.019 0.001 0.011 0.002

4 (1.9%) 25 (11.7%) 14 (6.5%) 17 (7.9%) 1 (0.5%) 6 (2.8%) 171 (82.6%) 507±48 57 (26.6%)

2 (1%) 19 (9.9%) 8 (4.2%) 13 (6.8%) 1 (0,5%) 2 (1%) 152 (81.7%) 507±48 46 (24.1%)

2 (8.7%) 6 (26.1%) 6 (26.1%) 4 (17.4%) 0 (0%) 4 (17.4%) 19 (90.5%) 512±50 11 (47.8%)

0.058 0.035 0.001 0.093 0.320 0.001 0.252 0.605 0.018

7.8±17.7 1108±1423 15.5±8.7 74.0±87.0 52.3±12.9

6.4±15.5 1084±1368 14.9±8.3 71.5±85.2 53.0±12.4

19.5±28.4 1330±1900 21.0±9.7 97.0±98.8 46.2±15.8

0.001 0.464 0.001 0.184 0.016

2.3±7.2 560±734 8.6±5.5 25.5±44.4 11.1±14.4

1.4±2.9 542±683 8.3±5.0 23.7±41.6 9.9±10.9

9.3±19.3 728±1118 11.7±8.1 39.7±62.0 20.7±29.5

<0.001 0.283 0.005 0.104 0.001

27 (13%) 38 (17.8%) 16 (7.5%)

17 (8.9%) 27 (14.1%) 9 (4.7%)

9 (39.1%) 11 (47.8%) 7 (30.4%)

<0.001 <0.001 <0.001

EP

AC C

SC

RI PT

Group LTA(n=191)

TE D

In-hospital events ⋅ Cardiogenic shock ⋅ IABP ⋅ Right ventricular pacing ⋅ ICD implantation ⋅ PM implantation Electrocardiogram during hospital stay ⋅ SB ⋅ First degree AVB ⋅ Right bundle branch block ⋅ Left bundle branch block ⋅ Transient complete AVB ⋅ Complete AVB ⋅ Long QT ⋅ Max QTc (ms) ⋅ Atrial Fibrillation Biomarkers peak ⋅ Troponin (ng/mL) ⋅ BNP (ng/L) ⋅ Leukocytes (G/L) ⋅ CRP (mg/L) LVEF at discharge (%) Biomarkers at discharge ⋅ Troponin (ng/mL) ⋅ BNP (ng/L) ⋅ Leukocytes (G/L) ⋅ CRP (mg/L) Hospital length of stay (days) Mortality ⋅ In-hospital mortality ⋅ 1-year ⋅ Cardiovascular

All patients (n=214)

M AN U

Variables

IABP: intra-aortic balloon pump, SB: sinoatrial block, AVB: atrio-ventricular block, LVEF: left ventricular ejection fraction.

ACCEPTED MANUSCRIPT Table 4: Predictive factors of life threatening arrhythmias UNIVARIATE ANALYSIS p

OR

0.982 0.477 0.992

0.949-1.016 0.182-1.252 0.914-1.077

0.288 0.133 0.848

0.999 1.291 0.534

0.417-2.391 0.478-3.485 0.201-1.416

0.998 0.614 0.207

2.215 0.604

0.872-5.627 0.134-2.731

0.095 0.513

1.019 0.983 1.003 0.988 0.922

0.995-1.043 0.966-1.001 0.973-1.034 0.961-1.015 0.879-0.967

0.130 0.066 0.843 0.378 0.001

1.049 1.000 1.012 1.002

1.008-1.092 0.999-1.000 0.941-1.088 0.997-1.008

95 % IC

p

RI PT

95 % IC

SC

OR

0.925

0.877-0.976

0.005

0.019 0.328 0.747 0.398

1.032

0.992-1.074

0.116

2.889 1.195-6.986 0.018 6.139 2.436-15.469 <0.001

1.392 4.473

0.456-4.249 1.514-13.219

0.561 0.007

TE D

Demographics ⋅ Age>65 years ⋅ Female gender ⋅ BMI CVD risk factors ⋅ Arterial hypertension ⋅ Diabetes mellitus ⋅ Dyslipidemia Stress Factor ⋅ Physical ⋅ Emotional Hemodynamic parameters ⋅ Heart rate ⋅ SBP ⋅ DBP ⋅ MBP ⋅ LVEF Biomarkers on admission ⋅ Troponin ⋅ BNP ⋅ Leukocytes ⋅ CRP ECG ⋅ In-hospital AF ⋅ QRS>105 ms

MULTIVARIATE ANALYSIS

M AN U

Characteristics

AC C

EP

SBP/DBP/MBP: systolic/diastolic/mean blood pressure, LVEF: left ventricular ejection fraction, AF: atrial fibrillation

ACCEPTED MANUSCRIPT Table 5: Characteristics of device-implanted patients

LTA12 19 22 LTA23 10 11

F M F F F F F F

63 61 70 82 92 80 75 42

Device indication

Device type

CAVB CAVB CAVB CAVB SB CAVB CAVB+VT VF

SPM DPM DPM DPM DPM DPM SICD s-ICD

Followup (months) 74 44 32 30 28 17 71 43

Patient Pacingdependent No Yes Yes No No Yes No

Device Percentage programming Pacing VVIR65 DDDR65 DDDR60 AAI-DDDR60 DDDR60 DDDR65 VVIR65

VP 71% VP 100% VP 100 % VP 82% AP 37% VP 100% VP 97%

RI PT

Patient Sex Age

AC C

EP

TE D

M AN U

SC

CAVB: complete AV block, SB: sinoatrial block, VF/VT: ventricular fibrillation/tachycardia, SPM/DPM: single/dual chamber pacemaker, SICD: single chamber defibrillator, s-IC: sub-cutaneous defibrillator, VP/AP: percentage of ventricular/atrial pacing

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT