Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit

+Model

ARTICLE IN PRESS

ACVD-1236; No. of Pages 13

Archives of Cardiovascular Disease (2020) xxx, xxx—xxx

Available online at

ScienceDirect www.sciencedirect.com

CLINICAL RESEARCH

Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit Incidence, caractéristiques cliniques et pronostic du syndrome de Takotsubo en réanimation Denis Doyen a,b,∗, Sébastien Moschietto c, Fabien Squara b, Pamela Moceri b, Hervé Hyvernat a, Emile Ferrari b, Jean Dellamonica a, Gilles Bernardin a a

Medical Intensive Care Unit, Archet 1 University Hospital, 06200 Nice, France Department of Cardiology, Pasteur University Hospital, 06000 Nice, France c General Intensive Care Unit, Avignon Hospital, 84000 Avignon, France b

Received 1st August 2019; received in revised form 9 October 2019; accepted 13 November 2019

KEYWORDS Takotsubo syndrome; Incidence; Intensive care unit; Shock; Arrhythmia

Summary Background. — Most diseases encountered in the intensive care unit are associated with major stress that can potentially trigger Takotsubo syndrome. Many severe cardiovascular complications are associated with Takotsubo syndrome, yet little is known about Takotsubo syndrome in the intensive care unit. Aims. — We sought to determine the incidence of Takotsubo syndrome, and to describe its clinical features and outcome in an intensive care unit. Methods. — This prospective single-centre study included all patients admitted consecutively over a 12-month period who had transthoracic echocardiography, electrocardiography and a troponin I assay performed on admission, at 24 and 48 hours after admission, and at discharge and in the case of clinical worsening.

Abbreviations: ACS, acute coronary syndrome; BNP, B-type natriuretic peptide; ESC, European Society of Cardiology; ICU, intensive care unit; LVEF, left ventricular ejection fraction; MRI, magnetic resonance imaging; TTE, transthoracic echocardiography; TTS, Takotsubo syndrome. ∗ Corresponding author. Medical Intensive Care Unit, Archet 1 University Hospital, 151, route Saint-Antoine-de-Ginestière, 06200 Nice, France. E-mail address: [email protected] (D. Doyen). https://doi.org/10.1016/j.acvd.2019.11.005 1875-2136/© 2020 Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

+Model ACVD-1236; No. of Pages 13

ARTICLE IN PRESS

2

D. Doyen et al. Results. — The incidence of Takotsubo syndrome was 4.6% (13/280 patients) and female sex predominated (69.2%). The median age of the subgroup with Takotsubo syndrome was 64 (56—72) years. Pulmonary disease and sepsis were the most frequent triggers (46.2% and 38.5%, respectively). Median left ventricular ejection fraction was 29.0% (20.0—37.0). Patients with Takotsubo syndrome presented with shock and arrhythmias and needed ventilation more frequently than patients without Takotsubo syndrome (69.2% vs. 36.3%, P = 0.035; 46.2% vs. 13.5%, P = 0.006; and 92.3% vs. 60.7%, P = 0.021), but mortality rates were similar. The median delay to cardiac index recovery, when impaired, was 2.0 (1.0—2.75) days, and that of left ventricular ejection fraction was 12.5 (7—14.75) days. Conclusion. — Takotsubo syndrome in the intensive care unit is not uncommon and is associated with substantial haemodynamic and respiratory instability. New-onset arrhythmias and respiratory and haemodynamic worsening could arouse suspicion of and prompt screening for Takotsubo syndrome in the intensive care unit. © 2020 Elsevier Masson SAS. All rights reserved.

MOTS CLÉS Syndrome de Takotsubo ; Incidence ; Réanimation ; Choc ; Arythmie

Résumé Contexte. — La plupart des pathologies rencontrées en réanimation est associée à un stress majeur pouvant potentiellement déclencher un syndrome de Takotsubo (STT) responsable de nombreuses complications cardio-vasculaires. Peu de données existent concernant les STT en réanimation. Buts. — Déterminer l’incidence du STT, ses caractéristiques cliniques et son pronostic en réanimation. Méthodes. — Il s’agit une étude prospective monocentrique. Pour tous les patients admis consécutivement pendant 12 mois ont été pratiqués une échocardiographie transthoracique, un électrocardiogramme et un dosage de troponine I à l’admission, 24 et 48 heures après l’admission, à la sortie du patient et en cas de dégradation clinique. Résultats. — L’incidence de STT a été de 4,6 % (13 patients sur 280). Le sexe féminin était prédominant (69,2 %). L’âge médian était de 64 (56,0—71,5) ans. Les pathologies pulmonaires et les sepsis représentaient les facteurs déclenchant les plus fréquents (46,2 % et 38,5 %, respectivement). La fraction d’éjection ventriculaire gauche (FEVG) médiane était de 29,0 % (20,0—37,0 %). Les STT ont présenté plus d’état de choc, d’arythmies et de recours à la ventilation que les autres patients (69,2 % vs 36,3 %, 46,2 % vs 13,5 % et 92,3 % vs 60,7 %; p = 0,035, p = 0,006 et p = 0,021, respectivement) mais avec une mortalité similaire. Le délai de récupération du débit cardiaque était de 2,0 (1—2,75) jours, et celui de la FEVG 12,5 (7—14,75) jours. Conclusions. — Le STT en réanimation n’est pas rare et est associé à une instabilité hémodynamique et respiratoire importante. La survenue d’arythmies, d’une dégradation respiratoire ou hémodynamique peut faire suspecter la survenue d’un STT. © 2020 Elsevier Masson SAS. Tous droits r´ eserv´ es.

Background Takotsubo syndrome (TTS), also called ‘‘apical ballooning syndrome’’ or ‘‘stress cardiomyopathy’’, is a syndrome responsible for transitory but severe left ventricular dysfunction, usually occurring after physical or psychological stress [1—3]. In the past few years, TTS has been diagnosed increasingly in the cardiology setting, where the incidence may be around 1—2% of patients with suspicion of acute coronary syndrome (ACS) [4,5]. Patients admitted to the intensive care unit (ICU) present with a variety of severe diseases that cause major stress, such as respiratory distress, sepsis, convulsions and shock.

These patients are therefore potentially at high risk of presenting with TTS. To date, the incidence of TTS in the ICU has not been studied prospectively. The clinical presentation of TTS mimics ACS [1—3], and initial diagnosis is difficult, especially in an ICU where patients are mostly sedated or confused, and do not have the ability to complain of chest pain. TTS is associated with serious complications, such as cardiogenic shock, pulmonary oedema, supraventricular and ventricular arrhythmias, intracardiac thrombi, left ventricular outflow tract obstruction or mitral regurgitation [2]. Therefore, diagnosis of TTS could be of great importance to patient management. Little is known about the clinical features and outcome of TTS diagnosed in the

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

+Model ACVD-1236; No. of Pages 13

ARTICLE IN PRESS

Takotsubo syndrome in the ICU ICU. Identification of specific clinical characteristics could help clinicians to screen for TTS in the ICU. The aims of our study were to determine the incidence of TTS and its clinical characteristics and complications in a medical ICU.

Methods Design We conducted a prospective single-centre study between November 2012 and November 2013 in a medical ICU. All consecutive patients admitted to the ICU were screened. Patients aged < 18 years or with a poor echocardiographic window were excluded. This study was approved by the local ethics committee (Comité de Protection des Personnes—Sud Méditerranée V). The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki. No change in standard care was adopted for the needs of this study, which was therefore considered as a non-interventional descriptive study by the ethics committee. Data collection and analysis respected the modified French law ‘‘Loi Informatique et Libertés’’ of 06 January 1978. All patients or their guardians gave their written informed consent.

TTS cardiomyopathy screening Screening for TTS was performed by transthoracic echocardiography (TTE), with an electrocardiogram and a troponin I assay, systematically on admission, 24 and 48 hours after admission, at ICU discharge and in case of chest pain, arrhythmias and respiratory or haemodynamic worsening. Btype natriuretic peptide (BNP) concentration was measured in all patients upon admission and in case of left ventricular dysfunction or signs of heart failure. TTS was diagnosed initially according to the Mayo Clinic criteria [6]. After the end of this study (2013), new criteria were proposed by the Heart Failure Association of the European Society of Cardiology (ESC) [2]. All patients included in this study were analysed once again using these new criteria. Data required for the ESC Heart Failure Association criteria (i.e. troponin, BNP, electrocardiography, echocardiography, coronary angiography) were at the time specifically collected prospectively, and thus the prospective nature of our study was respected. The ESC Heart Failure Association criteria for TTS diagnosis are as follows: • transient regional wall motion abnormalities of the left ventricular or right ventricular myocardium, which are frequently, but not always, preceded by a stressful trigger (emotional or physical); • the regional wall motion abnormalities usually extend beyond a single epicardial vascular distribution, and often result in circumferential dysfunction of the ventricular segments involved; • the absence of culprit atherosclerotic coronary artery disease, including acute plaque rupture, thrombus formation and coronary dissection, or other pathological conditions to explain the pattern of temporary left ventricular dysfunction observed (e. g. hypertrophic cardiomyopathy, viral myocarditis); • new and reversible electrocardiography abnormalities (ST-segment elevation, ST depression, left bundle-branch

3 block, T-wave inversion and/or QTc prolongation) during the acute phase (3 months); • significantly elevated serum natriuretic peptide (BNP or N-terminal prohormone of BNP) during the acute phase; positive, but relatively small elevation of cardiac troponin measured with a conventional assay (i. e. disparity between the troponin concentration and the amount of dysfunctional myocardium present); • and recovery of ventricular systolic function on cardiac imaging at follow-up (3—6 months). For each patient identified with TTS, we performed daily TTE to evaluate the delay to cardiac index recovery (when impaired) and to left ventricular ejection fraction (LVEF) recovery. We described the pattern of TTS using the four known TTS presentations: apical, mid-left ventricular, inverted and focal forms. We also distinguished primary TTS triggered by emotional stress or without an identified stressful trigger from secondary TTS triggered by an identified physical stress [2,3]. We made two comparisons: first between patients with TTS (TTS group) and all the other patients (non-TTS group); and then between the TTS group and patients presenting with ACS screened during the same period (ACS group). ACS was diagnosed and treated in accordance with ESC guidelines [7,8]. Patients with ACS were excluded from the non-TTS group in the first comparison. Coronary angiography and left ventriculography were performed for all cases of suspicion of TTS and ACS. Analysis of both left ventriculography and TTE was used to determine the TTS pattern.

Electrocardiography Twelve-lead electrocardiograms were performed using an AT-102 system (Schiller AG, Baar, Switzerland). Each patient was monitored with a continuous electrocardiogram to look for arrhythmias. All electrocardiograms were interpreted by an experienced cardiologist blinded to the clinical or paraclinical data.

TTE TTE was performed at the bedside using a M5S-D transducer (bandwidth 1.5—4.6 MHz) and a VividTM E9 healthcare system (General Electric, Amersham, UK). TTE was always performed and interpreted by an experienced cardiologist blinded to the other clinical or paraclinical data. Two different echocardiographers participated in the study. Poor quality echocardiographic images were defined in the case of no clear visualization of the valves and failure to detect more than two segments according to the segmentation of the American Society of Echocardiography [9]. LVEF was assessed using the biplane Simpson’s method. Left ventricular filling pressures were considered to be elevated when the E/e’ ratio was > 14 for patients in sinus rhythm or when the septal E/e’ ratio was ≥ 11 for patients in atrial fibrillation [10]. Right ventricular systolic dysfunction was assessed with at least one of the following criteria: tricuspid annular plane systolic excursion < 17 mm or a systolic tricuspid annular velocity < 9.5 cm/s [9].

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

+Model ACVD-1236; No. of Pages 13

ARTICLE IN PRESS

4

D. Doyen et al.

Collection of other data and analyses Cardiogenic shock was defined by the following criteria: persistent hypotension (systolic blood pressure < 90 mmHg or mean arterial pressure 30 mmHg lower than baseline), with a reduction in cardiac index (< 1.8 L/min/m2 without support or < 2.2 L/min/m2 with support) and adequate or elevated filling pressures [11]. The cardiac index was estimated by Doppler echocardiography. Septic cardiomyopathy was defined by the presence of left ventricular dysfunction (LVEF <45%) in patients with septic shock and global wall motion abnormalities without other obvious causes [12]. Immunodepression was defined in case of immunosuppressive drug intake, known chronic autoimmune diseases, infection with the human immunodeficiency virus, active haemopathy or recent chemotherapy.

Statistical analyses Continuous variables are presented as medians and interquartile ranges (IQR). Categorical variables are presented as percentages. Fisher’s exact test was used for comparison of categorical variables. The Mann-Whitney U test was used for comparison of continuous variables. Differences were considered statistically significant when the P value was < 0.05. The statistical analyses were performed using IBM SPSS Statistics software, version 19 (IBM Corp., Armonk, NY, USA).

Results

P = 0.044; 23.1% vs. 6.0%, P = 0.049) (Table 1). The most frequent other comorbidities found were cancer (30.8%), psychiatric disorders (30.8%) and chronic respiratory disease (23.1%). One patient with TTS died before potential complete LVEF recovery in the ICU. This patient did not fulfil all the ESC Heart Failure Association criteria for TTS diagnosis. However, cardiac magnetic resonance imaging (MRI) was done, and excluded signs of myocarditis or acute myocardial infarction. The coronary angiogram was normal. Furthermore, cardiac MRI and ventriculography exhibited wall motion abnormalities suggestive of a mid-ventricular TTS pattern. We therefore considered the patient to have TTS. Nine patients with TTS (69.2%) were sedated on admission and could not express any symptoms. Among the four other non-sedated patients, one complained of chest pain and another presented with syncope before admission. During the same period, 15 cases of ACS were identified: three ST-segment elevation myocardial infarctions and 12 non − ST-segment elevation myocardial infarctions. Compared with the ACS group, the TTS group presented fewer cardiovascular risk factors (2.0 [IQR 1.0—2.5] vs. 3.0 [IQR 2.0—4.0]; P = 0.005), especially arterial hypertension (15.4% vs. 73.3%; P = 0.003) (Table 1). A history of cardiac chronic disease was much more frequent in the ACS group than the TTS group (53.3% vs. 0.0%; P = 0.002). The presence of left ventricular hypertrophy was also more frequent in the ACS group than in the TTS group (40.0% vs. 0.0%; P = 0.018) (Table 3).

Intensity of supportive care and mortality

Patient characteristics During a 12-month period, 309 patients were admitted. Twenty-nine patients were excluded from further analysis because some cardiovascular examinations were lacking or were unreliable (12 patients had poor echogenicity; for the remaining 17 patients, TTE, electrocardiography, troponin analysis or coronary angiography was lacking). Finally, data from 280 patients were analysed. The study flow chart is shown Fig. 1. The general characteristics of the population are presented in Table 1. During this 12-month period, 13 cases of TTS (4.6%) were identified using the Mayo Clinic criteria. All these cases of TTS were confirmed with the new ESC Heart Failure Association criteria, and no other TTS was identified; among them, the female sex predominated (69.2%). The incidence of TTS rose to 7.6% when considering only women. The median age was 64 (56—72) years. Pulmonary disease and sepsis were the most frequent triggers for TTS (46.2% and 38.5%, respectively). Half of the patients with pulmonary disease (23.1%) also had sepsis (bacterial or viral) (Table 2). Eleven cases of TTS were secondary TTS, triggered by an acute medical emergency, and two cases of TTS were primary TTS without an identified trigger. For 11 patients with TTS, the TTS diagnosis was made upon admission; for the remaining two patients, TTS occurred a few days after admission (one was diagnosed after respiratory worsening; the other was diagnosed after respiratory and haemodynamic worsening). Immunodepression and hepatopathy were more frequent in the TTS group than in the non-TTS group (53.8% vs. 24.2%,

Patients with TTS needed vasopressors or inotropes more often than the rest of the population (69.2% vs. 36.3%; P = 0.035) (Table 3). Nine patients with TTS presented with shock; seven of them had an impaired cardiac index. Patients with TTS needed more mechanical or non-invasive ventilation than the rest of the population (92.3% vs. 60.7%; P = 0.021) (Table 3). No difference was observed concerning the need for renal replacement (P = 1.000) or intrahospital mortality (P = 0.760) (Table 3). All complications observed in patients presenting with TTS are reported in Table 4.

TTS group electrocardiogram features Initially, two patients presented with ST-segment elevation and two others with inverted T waves. Three other patients had left bundle-branch block, and the last six patients initially had no repolarization abnormality. During the 24 hours following admission, inverted T waves appeared in 10 patients. All these repolarization abnormalities involved anterior leads. Finally, all patients presented repolarization abnormalities. The QTc segment was initially prolonged for all patients. Five patients with TTS presented with supraventricular arrhythmias (three with atrial fibrillation and two with atrial tachycardia), more than in the non-TTS group (38.5% vs. 11.1%; P = 0.014) (Table 3). One patient presented with sustained ventricular tachycardia complicated by sudden death.

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

Age (years) Female/male sex

Immunodepression Malignancies Malignant haemopathies Psychiatric disorders Chronic respiratory insufficiency Hepatopathy Chronic renal disease Chronic cardiac disease Tabagism Diabetes Arterial hypertension Dyslipidaemia Number of cardiovascular risk factors Simplified Acute Physiology Score II Troponin I peak (ng/mL) BNP (pg/L) Creatinine (␮mol/L) C-reactive protein (mg/L) White blood cell count (/mL) Haemoglobin (g/dL)

6 5 9 1 1 1 7 4 3 4 3 3 1 0 3 3 2 1 2.0 55.0 0.84 1357 78 159 17,500 9.8

(46.2) (38.5) (69.2) (7.7) (7.7) (7.7) (53.8) (30.8) (23.1) (30.8) (23.1) (23.1) (7.7) (0.0) (23.1) (23.1) (15.4) (7.7) (1.0—2.5) (43.5—74.5) (0.44—3.88) (461—1651) (56—109) (44—289) (7450—18,700) (8.7—12.8)

(n = 252) 63 (52—74) 106 (42.1)/146 (57.9) 139 119 88 49 47 17 61 71 38 73 57 15 23 98 62 53 89 25 2.0 48.0 0.04 251.5 94 91.6 9750 11.8

(55.2) (47.2) (34.9) (19.4) (18.7) (6.7) (24.2) (28.2) (15.1) (29.0) (22.6) (6.0) (9.1) (38.9) (24.6) (21.0) (35.3) (9.9) (1.0—3.0) (35.0—66.0) (0.01—0.165) (93.75—692.25) (65—174) (20.92—238.5) (6375—14,450) (9.85—13.9)

ACS group

(n = 15) 70 (66—83) 4 (26.7)/11 (73.3) 5 3 8 9 3 5 3 3 1 2 2 0 3 8 5 8 11 2 3.0 49.0 2.52 1155 123 54 11,500 11.0

(33.3) (20.0) (53.3) (60.0) (20.0) (33.3) (20.0) (20.0) (6.7) (13.3) (13.3) (0.0) (20.0) (53.3) (33.3) (53.3) (73.3) (13.3) (2.0—4.0) (43.0—81.0) (0.72—7.2) (483—1721) (79—201) (14—90) (8200—15,500) (10.2—13.0)

P value (TTS group versus non-TTS group) 0.830 0.082 0.577 0.582 0.017 0.473 0.474 0.608 0.044 0.763 0.431 1.000 1.000 0.049 1.000 0.003 1.000 0.741 0.230 1.000 0.393 0.239 < 0.001 0.001 0.228 0.239 0.092 0.188

P value (TTS group versus ACS group)

0.068 0.056 ReasonforICUadmission 0.700 0.410 0.460 0.006 0.600 0.173 Comorbidities 0.114 0.670 0.311 0.372 0.639 0.087 0.600 0.002 0.686 0.137 0.003 1.000 0.005 0.872 0.174 0.771 0.025 0.034 0.222 0.213

ARTICLE IN PRESS

Pulmonary disease Sepsis Shock Renal failure Coma Cardiac arrest

(n = 13) 64 (56—72) 9 (69.2)/4 (30.8)

Non-TTS group

+Model

TTS group

ACVD-1236; No. of Pages 13

General characteristics of the population.

Takotsubo syndrome in the ICU

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

Table 1

Data are expressed as median (interquartile range) or number (%). ACS: acute coronary syndrome; BNP: B-type natriuretic peptide; ICU: intensive care unit; TTS: Takotsubo syndrome.

5

Sex

Stressful trigger

ECG abnormalities on TTS diagnosis

Further ECG abnormalities

Coronary lesion (coronary angiogram)

Patient 1 Patient 2

64 60

Female Male

Viral pneumonia Pulmonary embolism

Diffuse inverted T waves —

None None

Patient 3

66

Male

Diffuse inverted T waves

Non-significant

Patient 4

82

Female

Sigmoid diverticulitis Bacterial pneumonia

AF, prolonged QT Diffuse inverted T waves, prolonged QT Prolonged QT

Diffuse inverted T waves, AF

None

Patient 5

56

Female

Anterior inverted T waves

None

Patient 6 Patient 7 Patient 8

61 73 29

Female Female Female

None None None

64

Male

Prolonged QT AT, prolonged QT Anterior inverted T waves, prolonged QT Prolonged QT

Anterior inverted T waves Diffuse inverted T waves —

Patient 9

Anterior inverted T waves

None

Patient 10

56

Male

Listeria monocytogenes meningitis Pneumonectomy Bacterial pneumonia Intra-alveolar haemorrhage Alcohol and benzodiazepine intoxication None

Diffuse ST-segment elevation, prolonged QT Prolonged QT

—

None

Patient 11

82

Female

None

Anterior inverted T waves, AT

None

Patient 12

70

Female

Diffuse inverted T waves

Non-significant

Patient 13

55

Female

Central venous catheter infection Bacterial arthritis complicated with sepsis

Complete LBBB, prolonged QT Incomplete LBBB, prolonged QT Complete LBBB, AF, prolonged QT Anterior ST-segment elevation, prolonged QT

Anterior inverted T waves

None

ARTICLE IN PRESS

Age(years)

+Model

Characteristics of patients presenting with Takotsubo syndrome.

ACVD-1236; No. of Pages 13

6

AF: atrial fibrillation; AT: atrial tachycardia; ECG: electrocardiogram; LBBB: left bundle-branch block; TTS: Takotsubo syndrome.

D. Doyen et al.

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

Table 2

9 12 8 8 2 29.0 12 5 0 6 5 1 0 4 4

(69.2) (92.3) (61.5) (61.5) (15.4) (20.0—37.0) (92.3) (38.5) (0.0) (46.2) (38.5) (7.7) (0.0) (30.8) (30.8)

61 153 60 115 49 57 34 32 28 34 28 4 1 54 70

(24.2) (60.7) (23.8) (45.6) (19.4) (52—62) (13.5) (13.0) (11.1) (13.5) (11.1) (1.6) (0.4) (21.4) (27.8)

ACS group (n = 15)

8 13 4 10 6 37 11 5 6 4 2 1 1 3 3

(53.3) (86.7) (26.7) (66.7) (40.0) (25—53) (73.3) (33.3) (40.0) (26.7) (13.3) (6.7) (6.7) (20.0) (20.0)

P value (TTS group versus non-TTS group)

P value (TTS group versus ACS group)

0.035 0.021 0.003 0.568 1.000 < 0.001 < 0.001 0.024 0.131 0.006 0.014 0.183 1.000 0.490 0.760

0.460 1.000 0.125 1.000 0.221 0.080 0.333 1.000 0.018 0.433 0.198 1.000 1.000 0.670 0.670

ARTICLE IN PRESS

Vasopressor/inotrope use Ventilation Non-invasive ventilation Mechanical ventilation Renal replacement LVEF (%) Left ventricular dysfunctiona Right ventricular dysfunctionb Left ventricular hypertrophyc Arrhythmias Supraventricular arrhythmias Ventricular arrhythmias Bradyarrhythmias ICU mortality Intrahospital mortality

Non-TTS group (n = 252)

+Model

TTS group (n = 13)

ACVD-1236; No. of Pages 13

Intensity of supportive care, echocardiographic features, arrhythmias and mortality.

Takotsubo syndrome in the ICU

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

Table 3

Data are expressed as number (%) or median (interquartile range). ACS: acute coronary syndrome; ICU: intensive care unit; LVEF: left ventricular ejection fraction; TTS: Takotsubo syndrome. a Defined as LVEF < 45%. b Defined as tricuspid annular plane systolic excursion < 17 mm or systolic tricuspid annular velocity < 9.5 cm/s. c Defined as telediastolic left ventricular septal thickness ≥ 12 mm for women and ≥ 13 mm for men.

7

TTS pattern

RV dysfunction

Cardiogenic pulmonary oedema

MR

LVOT obstruction

Arrhythmias

Shock

Inotrope/ vasopressor infusion

NIV

MV

Renal replacement

Death

Patient 1 Patient 2

29 20

Apical Apical

Yes Yes

No Yes

No No

No No

Yes Yes

No Yes

Yes No

No Yes

No No

No Yes

Patient Patient Patient Patient Patient

3 4 5 6 7

17 38 25 37 23

Midventricular Apical Apical Apical Midventricular

No No No Yes No

No Yes Yes Yes Yes

No No No No Yes

No No No No Yes

No Yes No No Yes

No Yes No Yes Yes

No Yes Yes Yes Yes

No Yes No Yes Yes

No No No No Yes

No No No Yes Yes

Patient 8

35

Apical

No

Yes

No

No

No

No

No

Yes

No

No

Patient 9

20

Apical

Yes

Yes

No

No

No

Yes

Yes

Yes

No

No

Patient 10

32

Inverted

No

Yes

No

No

No

Yes

No

Yes

No

No

Patient 11

18

Apical

Yes

Yes

No

No

Yes

Yes

Yes

No

Yes

Yes

Patient 12

37

Apical

No

Yes

No

No

Yes

Yes

Yes

No

No

No

Patient 13

47

Midventricular

No

No

No

No

No

Yes

No Yes (Ep, Norep, Dob) No Yes (Norep) No Yes (Norep) Yes (Ep, Norep, Dob) Yes (Dob, Norep) Yes (Mil, Norep) Yes (Dob, Norep) Yes (Mil, Norep) Yes (Dob, Norep) Yes (Norep)

No

Yes

No

No

ARTICLE IN PRESS

LVEF (%)

+Model

Complications of patients presenting with Takotsubo syndrome.

ACVD-1236; No. of Pages 13

8 Dob: dobutamine; Ep: epinephrine; LVEF: left ventricular ejection fraction; LVOT: left ventricular outflow tract; Mil: milrinone; MR: mitral regurgitation; MV: mechanical ventilation; NIV: non-invasive ventilation; Norep: norepinephrine; RV: right ventricular; TTS: Takotsubo syndrome.

D. Doyen et al.

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

Table 4

+Model ACVD-1236; No. of Pages 13

ARTICLE IN PRESS

Takotsubo syndrome in the ICU

Figure 1.

9

Study flow chart. ECG: electrocardiogram; ICU: intensive care unit; TTE: transthoracic echocardiography.

The features of the electrocardiograms are presented Table 2.

TTS group coronary angiography and ventriculography Two patients with TTS presented non-significant stenoses. No coronary lesion was observed for the 11 other patients with TTS. Ventriculography examples of the three patterns observed in our series are shown in Fig. 2 and Video A.1 (apical pattern), Fig. 3 and Video A.2 (mid-left ventricular pattern) and Fig. 4 and Video A.3 (inverted pattern).

TTS group echocardiogram features The median LVEF of patients with TTS was 29.0% (20.0—37.0) (Table 3). The apical TTS pattern concerned 69.2% of the patients with TTS, the mid-left ventricular pattern 23.1% and the inverted pattern 7.7% (Table 4). No focal pattern was observed. The median LVEF for apical TTS was similar to the median LVEF for the midventricular pattern (P = 0.735). The cardiac index was impaired in seven patients with TTS (53.8%). Right ventricular systolic dysfunction was found in five patients with TTS (38.5%). One patient with TTS presented with left ventricular outflow tract obstruction (< 40 mmHg) complicated by moderate mitral regurgitation. One patient with TTS had moderate pericardial effusion. No apical thrombus was seen. All patients with TTS recovered normal cardiac function (both left and right ventricular functions) except for one patient who died in the ICU before potential recovery. The median delay to cardiac index recovery, when impaired, was 2 (1—2.75) days, and that of LVEF was 12.5 (7—14.75) days.

TTS group biological features Troponin I was elevated for all patients with TTS, with a median peak of 0.84 (IQR 0.44—3.88) ng/mL (Table 1).

Troponin I was initially negative for two patients with TTS on admission, but always rose in the first 24 hours after ICU admission. The median BNP concentration was 1357 (IQR 461—1651) pg/L (Table 1).

Discussion This study is the first to prospectively screen for TTS in consecutive patients in an ICU. We found a relatively significant incidence of TTS. In our series, TTS affected predominantly postmenopausal women admitted for pulmonary disease and sepsis. A simple analysis of electrocardiography, BNP, troponin and TTE can point to this pathology. TTS was associated with a high rate of morbidity. Some clinical characteristics can help to distinguish TTS from ACS. In 2005, when TTS was first described, Park et al. [13] reported a TTS incidence of 28% in a medical ICU. However, in our opinion, their TTS definition was probably incomplete; they identified TTS in the case of left ventricular kinetic abnormalities sparing basal segments, without integrating a description of the electrocardiogram, troponin concentration or a coronary angiogram. Since 2008, the most accepted TTS definition is based on the Mayo Clinic criteria and, more recently, on the ESC Heart Failure Association diagnostic criteria, which require these data, particularly the absence of culprit atherosclerotic coronary artery disease [2,6]. Therefore, Parker et al. may have overestimated TTS incidence by including potential ischaemic cardiomyopathy. More recently, Champion et al. and De Backer et al. reported TTS incidences of 1.5% and 0.9%, respectively, but their analyses were retrospective and used the old Mayo Clinic criteria [6]. Our study is the first to screen for TTS prospectively in all patients admitted to an ICU using the ESC Heart Failure Association diagnostic criteria. In our series, TTS was associated with a high rate of morbidity. Haemodynamics were more unstable in the TTS group than in the rest of the population. Cardiogenic

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

+Model ACVD-1236; No. of Pages 13

ARTICLE IN PRESS

10

D. Doyen et al.

Figure 2.

Ventriculography showing an example of apical Takotsubo syndrome pattern. Panel A End-diastole. Panel B End-systole.

Figure 3. systole.

Ventriculography showing an example of mid-left ventricular Takotsubo syndrome pattern. Panel A End-diastole. Panel B End-

Figure 4.

Ventriculography showing an example of inverted Takotsubo syndrome pattern. Panel A End-diastole. Panel B End-systole.

shock concerned 53.8% of all cases of TTS. The impaired left ventricular function, right ventricular dysfunction and new-onset arrhythmias may explain these haemodynamic conditions. In our study, TTS was also associated with more respiratory failure than the rest of the population: almost all

patients with TTS (92.3%) needed non-invasive or mechanical ventilation. In particular, non-invasive ventilation was used significantly more frequently in TTS, probably because cardiogenic pulmonary oedema was frequent (found in 76.9% of patients with TTS), and non-invasive ventilation is

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

+Model ACVD-1236; No. of Pages 13

ARTICLE IN PRESS

Takotsubo syndrome in the ICU particularly used and efficient in this indication. The haemodynamic and respiratory instability observed in patients with TTS from our series seemed to be higher than that reported for patients with TTS described in the cardiology setting: Templin et al. reported from the International Takotsubo Registry of patients with TTS admitted to cardiology units a lower use of ventilation (17.3%) and catecholamine (12.2%) and a lower incidence of cardiogenic shock (9.9%) [3]. In our series, haemodynamic and respiratory instability can also be explained by the sepsis and pulmonary disease often associated with TTS. The ICU mortality rate (30.8%) of patients with TTS was high, but without significant difference compared with the rest of the population. A relatively fast recovery in cardiac function may explain this result. Haghi et al. reported a lower mortality rate (16.7%), but did not screen all ICU patients systematically for TTS [14]. The mortality rate of patients with TTS in our study was much higher than the mortality of patients with TTS admitted to the cardiology unit (30.8% in our study vs. 4.1% [3]). In fact, patients admitted to a general ICU are much more unstable and present more multiple organ failure than patients with TTS admitted to a cardiology unit. The mortality rate in the ICU is estimated to be around 23% [15]. In general, patients admitted to the cardiology unit only present with cardiac failure, and the mortality rate is much lower. The 69.2% proportion of women and the age of the patients in our cohort were lower than is typically seen in patients with TTS [3]. We suggest that this was related to the higher proportion of cases of TTS triggered by physical stress (86.4% secondary TTS, triggered by a physical stress, and only 13.6% primary TTS without physical stress). Indeed, TTS associated with physical stress is more common in men than in women [2]; this explains the relatively high proportion of men in our series. TTS triggered by physical stress may also affect younger people; Schneider et al. [16] showed in their series that the age of men with TTS was lower, but without significant difference. The high proportion of TTS triggered by physical stress in our series can be explained by the fact that the cases of TTS triggered by physical stress were associated with more complications [3]. Patients admitted to an ICU have, by definition, multiple organ failure. Therefore, patients with TTS triggered by physical stress are more likely to be admitted to an ICU than those with TTS triggered by emotional factors. The main stressful trigger events for TTS identified in our study were pulmonary disease and sepsis; this has been described in the literature [2,17]. Consistently, the main comorbidities associated with TTS in our series were chronic respiratory failure [17,18], psychiatric disorders [2,19], cancer [20] and hepatopathy [2,21]. Immunodeficiency was more frequent in the TTS group than in the rest of the population, which has not been described previously. This finding was probably observed because most of the patients with TTS presenting with immunodeficiency were admitted for sepsis, which was one of the most frequent TTS triggers. TTS diagnosis in the ICU is challenging. Angina is the main symptom of clinical TTS presentation, but patients with TTS in an ICU are mostly sedated and cannot complain of chest pain. Yet, 69.2% of our TTS group presented with shock, 46.2% with arrhythmias and 92.3% with acute respiratory failure. Therefore, our results suggest that haemodynamic

11 instability, arrhythmias or respiratory distress in the ICU may help to arouse suspicion of TTS onset. Haghi et al. also showed that sudden haemodynamic deterioration may be the presenting symptom of TTS [14]. Interestingly, all patients with TTS in the present study showed electrocardiogram abnormalities (prolonged QTc interval, complete left bundle-branch block, supraventricular arrhythmias, ST-segment elevation or inverted T waves). The presence of new and reversible electrocardiography abnormalities has been required for TTS diagnosis since the last statement from the Taskforce on TTS of the Heart Failure Association of the ESC [2]. Electrocardiograms may have very good sensitivity for TTS screening, but they have poor specificity, because these abnormalities are seen in a lot of other cardiomyopathies. We probably miss many cases of TTS and underestimate its incidence. Electrocardiography is not expensive and is easy to perform. TTS screening with a daily electrocardiogram therefore represents an interesting strategy; troponin dosage and echocardiography could be proposed in case of electrocardiogram abnormality (or for other obvious clinically-justified reasons). This should be confirmed in future studies. The comparison of patients with TTS and ACS revealed some interesting and basic clinical data that distinguished between these two pathologies. TTS mimics ACS and distinguishing them is difficult. Patients with TTS presented fewer cardiovascular risk factors, especially arterial hypertension. Consistently, TTS was associated less often with left ventricular hypertrophy, which is often the consequence of arterial hypertension. Patients with TTS also had, in general, less history of chronic cardiac diseases. Nevertheless, these results do not permit a definitive distinction between these two pathologies, and should not lead to avoidance of performing coronary angiography.

Study limitations Our study has some limitations. Firstly, it was a single-centre study, but the study protocol (up to four troponin I measurements, four TTEs and four electrocardiograms for all patients admitted consecutively to the ICU over 1 year) was difficult to conduct in more than one centre. Despite 1 year of screening, the number of patients with TTS was low, probably because of the rarity of the disease. Moreover, we only included patients generally admitted for medical conditions. Surgery can also trigger TTS. Other studies are warranted to evaluate its incidence in this setting. Cardiac MRI could not be performed systematically whenever TTS was suspected because it is difficult to perform. Cardiac MRI is of value in an emergency to help with the diagnosis of TTS, especially to differentiate TTS from ACS or myocarditis [2]. Some patients diagnosed with TTS could have ACS with complete thrombus resolution at the time of coronary angiography. However, the cardiac MRI features of TTS remain controversial, as some patients with TTS can present with minor late gadolinium enhancement during the acute phase [2]. The median LVEF of midventricular TTS was similar to the median LVEF of apical TTS. In general, LVEF is preserved more in the case of midventricular TTS compared with apical TTS [2]. We can explain our findings by the lack of statistical power of the analysis of our results. We identified only three

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

+Model ACVD-1236; No. of Pages 13

ARTICLE IN PRESS

12 cases of midventricular TTS; this small number might not be sufficient to obtain a difference compared with apical TTS. In patients with associated sepsis and presenting with left ventricular dysfunction, it can be difficult to identify the exact cardiomyopathy observed: it can be TTS, septic cardiomyopathy or myocarditis. However, we can say that we were able to make the distinction. In septic cardiomyopathy, left ventricular wall motion abnormalities are global [22,23], whereas in TTS, wall motion abnormalities are segmental. In myocarditis, cardiac MRI is the best diagnostic strategy. We could not perform cardiac MRI for all patients with TTS. However, some other clinical and paraclinical data helped us to make this distinction: myocarditis usually affects young people without sex prevalence, whereas the cases of TTS screened in our cohort were preferentially postmenopausal women [2]; an increase in troponin in myocarditis was significant, proportional to the hypokinetic area, which was exactly the opposite for the patients with TTS from our series (a minimal rise in troponin and a significant hypokinetic area) [2]; wall motion abnormalities can be global in myocarditis [2,24,25], and some patients did not have complete left ventricular dysfunction recovery, whereas in TTS wall motion abnormalities are segmental and complete recovery was observed for all patients who survived [2]; and wall thickness can be increased in myocarditis [2,24], which was not observed in any of the cases of TTS from our series. All patients with TTS with a history of cancer received chemotherapy, which could have been responsible for cardiomyopathy. However, TTS has a very different presentation from cardiomyopathy induced by chemotherapy, particularly concerning wall motion abnormalities [26]. In addition, any chemotherapy administered to patients with TTS in our study was not recent. The choice of TTE in our study can be discussed. Transoesophageal echocardiography can offer better echogenicity than TTE in the ICU. However, transoesophageal echocardiography is difficult to perform for all patients admitted to the ICU, especially those who are not intubated and sedated, present with hypoxia or have severe coagulation disorders. Furthermore, we excluded TTE with poor echogenicity, and the echocardiography was performed by two experienced cardiologists.

Conclusions Takotsubo syndrome is not uncommon in the ICU. Patients with TTS in the ICU share same clinical features as those with TTS admitted to the cardiology setting, but have more haemodynamic and respiratory instability. TTS diagnosis in the ICU is difficult, but our results suggest that new-onset arrhythmias and respiratory and haemodynamic worsening could help physicians to screen for TTS in the ICU.

Sources of funding This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

D. Doyen et al.

Acknowledgements We thank Dr Marie Christiane Brahimi-Horn and Dr Nihal Martis for English language revision.

Disclosure of interest The authors declare that they have no competing interest.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.acvd. 2019.11.005.

References [1] Akashi YJ, Goldstein DS, Barbaro G, Ueyama T. Takotsubo cardiomyopathy: a new form of acute, reversible heart failure. Circulation 2008;118:2754—62. [2] Lyon AR, Bossone E, Schneider B, et al. Current state of knowledge on Tako-Tsubo syndrome: a position statement from the taskforce on Tako-Tsubo Syndrome of the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail 2016;18:8—27. [3] Templin C, Ghadri JR, Diekmann J, et al. Clinical features and outcomes of Tako-Tsubo (stress) cardiomyopathy. N Engl J Med 2015;373:929—38. [4] Gianni M, Dentali F, Grandi AM, Sumner G, Hiralal R, Lonn E. Apical ballooning syndrome or Tako-Tsubo cardiomyopathy: a systematic review. Eur Heart J 2006;27:1523—9. [5] Kurowski V, Kaiser A, von Hof K, et al. Apical and midventricular transient left ventricular dysfunction syndrome (Tako-Tsubo cardiomyopathy): frequency, mechanisms, and prognosis. Chest 2007;132:809—16. [6] Madhavan M, Prasad A. Proposed Mayo Clinic criteria for the diagnosis of Tako-Tsubo cardiomyopathy and long-term prognosis. Herz 2010;35:240—3. [7] Hamm CW, Bassand JP, Agewall S, et al. ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J 2011;32:2999—3054. [8] Steg PG, James SK, Atar D, et al. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J 2012;33:2569—619. [9] Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging 2015;16:233—70. [10] Nagueh SF. Left ventricular diastolic function: understanding pathophysiology. Diagnosis, and prognosis with echocardiography. JACC Cardiovasc Imaging 2019. [11] Reynolds HR, Hochman JS. Cardiogenic shock: current concepts and improving outcomes. Circulation 2008;117:686—97. [12] Vieillard-Baron A. Septic cardiomyopathy. Ann Intensive Care 2011;1:6. [13] Park JH, Kang SJ, Song JK, et al. Left ventricular apical ballooning due to severe physical stress in patients admitted to the medical ICU. Chest 2005;128:296—302.

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005

+Model ACVD-1236; No. of Pages 13

ARTICLE IN PRESS

Takotsubo syndrome in the ICU [14] Haghi D, Fluechter S, Suselbeck T, et al. Takotsubo cardiomyopathy (acute left ventricular apical ballooning syndrome) occurring in the intensive care unit. Intensive Care Med 2006;32:1069—74. [15] Vincent JL, Marshall JC, Namendys-Silva SA, et al. Assessment of the worldwide burden of critical illness: the intensive care over nations (ICON) audit. Lancet Respir Med 2014;2:380—6. [16] Schneider B, Athanasiadis A, Stollberger C, et al. Gender differences in the manifestation of tako-tsubo cardiomyopathy. Int J Cardiol 2013;166:584—8. [17] Sharkey SW, Windenburg DC, Lesser JR, et al. Natural history and expansive clinical profile of stress (tako-tsubo) cardiomyopathy. J Am Coll Cardiol 2010;55:333—41. [18] Madhavan M, Rihal CS, Lerman A, Prasad A. Acute heart failure in apical ballooning syndrome (TakoTsubo/stress cardiomyopathy): clinical correlates and Mayo Clinic risk score. J Am Coll Cardiol 2011;57:1400—1. [19] Delmas C, Lairez O, Mulin E, et al. Anxiodepressive disorders and chronic psychological stress are associated with Tako-Tsubo cardiomyopathy — New Physiopathological Hypothesis. Circ J 2013;77:175—80.

13 [20] Burgdorf C, Nef HM, Haghi D, Kurowski V, Radke PW. Tako-tsubo (stress-induced) cardiomyopathy and cancer. Ann Intern Med 2010;152:830—1. [21] Saner FH, Plicht B, Treckmann J, et al. Tako-Tsubo syndrome as a rare cause of cardiac failure in liver transplantation. Liver Int 2010;30:159—60. [22] Chockalingam A, Mehra A, Dorairajan S, Dellsperger KC. Acute left ventricular dysfunction in the critically ill. Chest 2010;138:198—207. [23] Romero-Bermejo FJ, Ruiz-Bailen M, Gil-Cebrian J, Huertos-Ranchal MJ. Sepsis-induced cardiomyopathy. Curr Cardiol Rev 2011;7:163—83. [24] Fung G, Luo H, Qiu Y, Yang D, McManus B. Myocarditis. Circ Res 2016;118:496—514. [25] Kindermann I, Barth C, Mahfoud F, et al. Update on myocarditis. J Am Coll Cardiol 2012;59:779—92. [26] Carver JR, Shapiro CL, Ng A, et al. American Society of Clinical Oncology clinical evidence review on the ongoing care of adult cancer survivors: cardiac and pulmonary late effects. J Clin Oncol 2007;25:3991—4008.

Please cite this article in press as: Doyen D, et al. Incidence, clinical features and outcome of Takotsubo syndrome in the intensive care unit. Arch Cardiovasc Dis (2020), https://doi.org/10.1016/j.acvd.2019.11.005