Assessment of Intra and Extra-Hospital Outcome after Takotsubo Syndrome in a Single-Center Population.

Objectives: To evaluate short- and long-term outcome in a single prospective cohort of Takotsubo syndrome (TTS) patients, trying to early identify those with better prognosis and to assess the prevalence of left ventricular ejection fraction (LVEF) recovery over time.
Methods: Forty-nine patients prospectively enrolled underwent to assessment of demographic, clinical, and echocardiographic characteristics, and later were followed to identify the outcomes during a mean follow-up (FU) of 93 months. At the end of the FU phase, a subgroup of patients underwent to a clinical and echocardiographic re-evaluation. As major adverse cardiac events (MACE) were considered: Intra and extra-hospital death, re-hospitalizations for acute heart failure or atrial fibrillation (AF) or acute myocardial infarction (AMI) and TTS recurrence; minor events were considered minor symptoms of heart failure and angina not requiring hospitalization.
Results: The prevalence of re-hospitalizations (MACE) over time was: 41% (n = 12) for HF; 6.8% (n = 2) for AMI; 3.4% (n = 1) for TTS relapse and 20% (n = 6) for AF. Minor events were the symptoms of chest pain and dyspnea not requiring hospitalization in 6 (20.7%) and 12 (34.4%) patients, respectively. LVEF at the time of admission was predictor for MACE. Stratifying patients on the LVEF admission median value (40%). Patients with LVEF <40% at admission had a significantly lower survival free from adverse cardiac events compared to patients with LVEF ≥40%. Twenty-seven (93%) patients underwent to a clinical reassessment with electrocardiogram and echocardiographic examination. LVEF value showed a statistically significant increase (P = 0.004) at the end of FU. Conclusions: At admission, "high-risk" patients (LVEF <40%) can be easily detected, allowing an appropriate pharmacological and/or mechanical support strategy and a more "careful" FU. Copyright:

INTRODUCTION

Takotsubo syndrome (TTS) is a cardiovascular disease, characterized by acute transient left ventricular (LV) systolic dysfunction and abnormal changes in electrocardiogram (ECG) pattern in the absence of an obstructive coronary artery disease. Considering the major trigger of TTS is a stressful stimulus, this condition also has been defined as broken heart syndrome[1] or stress-induced cardiomyopathy.[2] In 2006, it was classified as a primary acquired cardiomyopathy, and since, then the diagnosis of TTS was based on Modified Mayo Clinic criteria.[2] TTS is responsible for about 1.0%–2.5% of hospitalizations for acute coronary syndrome (ACS).[3]

Although TTS was initially considered as a benign condition, several studies have demonstrated a substantial incidence of life-threatening complications occurring in the acute phase, with mortality ranging from 1% to 8%.[456] Furthermore, data from the Inter-TAK registry revealed that both short-and long-term outcomes of TTS patients are similar to those of patients with ACS.[7] Advanced age, hypotension at hospital admission, severe LV systolic and diastolic dysfunction, and significant mitral regurgitation (MR) are all conditions associated with an increased risk of intra-hospital adverse events.[68] A risk stratification system based on an arbitrary LV ejection fraction (LVEF) cutoff value of 35% has been proposed by the Task Force on TTS of the Heart Failure Association of the European Society of Cardiology to identify patients at risk.[4] Evidences about long-term follow up (FU) are lacking.

The aim of this study was as follows: (1) To evaluate the short- and long-term outcome in a single prospective cohort of TTS patients trying to early identify those with better prognosis and (2) to assess the prevalence of LVEF recovery over time.

METHODS

This study consists of two phases:

Enrollment phase

assessment of demographic, clinical, and echocardiographic characteristics of a prospective TTS population admitted to the Cardiology Department of the University Hospital of Messina (acute phase).

Follow-up phase

analysis of patients’ outcome during a mean FU of 93 months (range 29–148 months) and clinical and echocardiographic re-evaluation in a subgroup of them.

Ethics committee approved the research protocol, and the study was performed in accordance with the ethical standards laid down in the 1964 declaration of Helsinki and its later amendments. All enrolled patients gave their informed consent to participate in this research.

Forty-nine patients with a diagnosis of TTS, admitted to the Cardiology Department of University of Messina from January 2012 to December 2017, were enrolled. On admission, demographic data, clinical history with particular regard to the presence of stressful events, presence of cardiovascular risk factors, and presentation symptoms were collected for all patients. The diagnosis of TTS was made according to the Mayo Clinic criteria modified in 2008.[3]

Data collection

The clinical variables of patients were collected in a standardized way including demographic characteristics (sex, age, and heart rate), signs and symptoms at the disease's onset, medical history, emotional or physical triggers, electrocardiographic changes of ventricular repolarization phase, and intra-hospital course (in particular the occurrence of major complications).

In addition, the following data were collected: Prevalence of dyslipidemia (total cholesterol >190 mg/dl and/or triglyceridemia >150 mg/dl or intake of lipid-lowering drugs), prevalence of diabetes mellitus (fasting blood glucose values >126 mg/dl on two occasions) or routine intake of oral hypoglycemic agents or insulin; systemic arterial hypertension (blood pressure >140/90 mmHg or patients on antihypertensive treatment), history of ischemic heart disease.

Venous samples were taken at approximately 3-h intervals to measure the concentration of high sensitivity Troponin I during the acute phase and up to the peak value.

All patients underwent coronary angiography and ventriculography within 24 h of symptom onset, and relative data were stored.

The 12-lead electrocardiographic examination was performed at the time of admission and daily until discharge.

An echocardiogram was performed on admission and cardiac images and video were digitally stored for the off-line analysis (see below).

Echocardiography

All echocardiographic examinations were performed before coronary angiography, within 6 h of admission. Ultrasound systems (GE E95, Horten, Norway) of the latest generation have been used, with phased-array 2.5–4.5 MHz transducers. All echocardiographic images were digitalized and reviewed by two experienced cardiologists. For the evaluation of LV wall motion and the distribution of the LV ballooning, clips of at least three cardiac cycles recorded in the 4, 2-, 3-apical chamber view and in the parasternal short axis section of the mitral valve and papillary muscles were used. The LVEF was calculated using the Simpson biplane method from 4- and 2-apical chambers sections.[9] The LV diastolic function was evaluated, according to the recommendations of the American Society of Echocardiography,[10] and patients were divided in three groups: Normal diastolic function, indeterminate, and diastolic dysfunction. The presence and degree of severity of MR were assessed by color Doppler, and the MR was classified as absent, mild, moderate, or severe according to the current recommendations.[11]

Systolic pulmonary arterial pressure was calculated from the peak of the tricuspid regurgitation rate measured with continuous Doppler. Finally, the right atrial mean pressure was estimated based on the caliber variations of the inferior vena cava during the respiratory cycle according to the recommendations of the American Society of Echocardiography.[ 12]

The same echocardiographic study protocol was used during FU.

Cardiac catheterization

Cardiac catheterization was performed within 24 h from the admission, using the femoral or radial approach. Left ventriculography confirmed the diagnosis of TTS in all cases and made it possible to confirm the location and extent of the ballooning. Coronary angiography was considered significant for coronary artery disease in the presence of a reduction in lumen diameter of >50% in the left main or >70% elsewhere or between 30% and 70% with a fractional flow reserve ≤0.80.[13]

Patients’ outcome was evaluated from the day of hospitalization until discharge (short-term outcome) and from the day of hospital discharge until the end of the study (long-term outcome).

As major adverse cardiac events (MACE) were considered the followings: Intra and extra-hospital cardiovascular death and re-hospitalizations for one of the following conditions: Acute heart failure, atrial fibrillation (AF), acute myocardial infarction (AMI), or TTS recurrence. Minor events were considered minor symptoms of HF and angina not requiring hospitalization.

As short-term survival, the time interval between the admission and the day of the death during hospitalization was considered. To assess long-term survival telephonic interviews were planned after 6 months from the day of hospital discharge and 5 years later. Long-term survival was then calculated as the time interval between the date of first hospitalization and the date of the death stated by the familiars during the interview.

All alive patients were invited to undergo a new evaluation with ECG and echocardiogram at the end of the study. Twenty-seven patients underwent to a clinical reassessment with ECG and echocardiographic examination, two were lost owing to physical inability to reach our hospital.

The average duration of FU was 93 months (range 29–148 months).

Statistical analysis

Data were expressed as mean ± standard deviation. Statistical analysis was performed using the dedicated software (IBM SPSS, Chicago, Illinois, USA). A Cox-regression analysis was performed for the identification of predictors of cardiovascular mortality-free survival and from a combined endpoint of mortality, hospitalization for cardiac events (MACE). The Kaplan–Meier curves were processed, with the log-rank test for the comparison of survival curves. P ≤ 0.05 was considered statistically significant.

RESULTS

Population: Demographic and clinical characteristics

All the 49 patients hospitalized (48 female) with a TTS diagnosis were included in the study. The average duration of hospitalization was 4.5 days with a minimum of 3 days and a maximum of 8 days. In Table 1, the main demographic and clinical data of overall population are detailed. The mean age was 70.27 ± 10 years ranging between 51 and 90 years. All female patients at the time of diagnosis were postmenopausal. Among the 49 patients enrolled, three (6%) were TTS relapses.

Table 1

Clinical presentation and complications of Takotsubo syndrome in overall population

Patients	Value (n=49), n (%)
Sex
Male	1 (2.4)
Female	48 (97.6)
Risk factors
Hypertension	39 (69.4)
Diabetes	8 (16.3)
Hypercholesterolemia	13 (26.5)
Smoke	7 (14.3)
Symptoms
Chest pain	45 (91.8)
Dyspnea	2 (4.1)
Syncope	1 (2)
Trigger
Emotional	35 (71.4)
Physical	13 (26.5)
Form
Apical ballooning	44 (89.8)
Apical sparing	4 (8.2)
Inverted	1 (2)
Acute complications
AVB III	1 (2)
AF	3 (6)
Death	2 (4)

AVB=Atrioventricular block, AF=Atrial fibrillation

The most common (69.4%) cardiovascular risk factor was arterial hypertension. Diabetes was present in 8 (16.3%) participants, while hypercholesterolemia in 13 (26.5%). Finally, only seven (14.3%) participants were smokers.

Symptoms at the time of admission were: acute chest pain suggestive of ACS in 45 (91.8%) patients, dyspnea in two patients (4.1%), and syncope in only one (2%) case.

Almost all cases presented a trigger event; in particular, 47 patients (96%) reported a stressor event, while only two (4%) did not report any trigger at all. In 35 cases (71.4%), symptomatology appeared at a short distance from an emotional stress (e.g., mourning, family quarrel, and acute emotion), whereas in the other 13 cases (26.5%), it occurred at a short distance from a physical stress (e.g., physical work efforts) often related to some acute events that required hospitalization (e.g., fractures, concomitant neoplastic pathology, and stroke).

Blood samples results are reported in Table 2. High sensitivity troponin was higher than the upper reference limit (4.09 ± 5.230 ng/l).

Table 2

Demographic, clinical and laboratory features of the population on admission

Patients	Average (n=49)
Age (years)	70.27±10.091
SBP mmHg	133.13±21.883
DBP mmHg	77.42±12.902
HR, b/m’	85.57±15.138
Troponin I, ng/dL	4.09±5.230
CK-MB, mg/dL	32.15±24.741
CPK, mg/dl	226.37±165.354
Myoglobin, mg/dL	154.49±193.680
Creatinine, mg/dL	0.86±0.577
Potassium, mEq/L	4.06±0.592
Sodium, mEq/L	140.47±4.895
RBC, n/mm3	6,150,000±550,000
Hemoglobin, g/dL	12.19±1.722
WBC, n/mm3	8395±3196.778

BP=Blood pressure, SBP=Systolic BP, DBP=Diastolic BP, HR=Heart rate, CK=Creatine-kinase, RBC=Red blood cells, WBC=White blood cells

Electrocardiographic features

The ECG at the admission in 30 patients was suggestive of subepicardial ischemia with ST segment elevation. Detailed electrocardiographic characteristics are reported in Table 3.

Table 3

Electrocardiographic, echocardiographic and angiographic data of overall population during hospital stay

Patients	Value (n=49)
ST elevation, n (%)
Anterior	25 (54.3)
Inferior	10 (21.7)
Lateral	20 (43.5)
T wave inversion, n (%)	14 (28.5)
QT prolongation, n (%)	8 (17)
Conduction disorders, n (%)
RBBB	2 (4)
LAFB	7 (14.2)
LBBB	1 (2)
Diastolic dysfunction, n (%)
Grade I	31 (63)
Grade III	2 (4)
Indeterminate	16 (33)
MR, n (%)	38 (84.4)
LV EDV	139.8±29.5
LV ESV	83.2±25.4
LVEF	41.80±11.085
Coronary angiography, n (%)
Normal	41 (83.7)
Noncritical stenoses	8 (16.3)

LV=Left ventricle, LVEF=LV ejection function, RBBB=Right bundle branch block, LBBB=Left bundle branch block, LAFB=Left anterior fascicular block, MR=Mitral regurgitation, ESV=End-systolic volume, EDV=End-diastolic volume

Echocardiographic data

At admission, analysis of LV wall motion on transthoracic echocardiogram, showed that in 44 (89.8%) patients, the alterations involved the apical segments (apical ballooning), in 4 (8.2%) only the middle segments without involvement of the apex (apical sparing) whereas one (2%) patient showed basal akinesia (inverted TTS) [Table 1].

LVEF was reduced (41.8% ± 11%).

Grading of diastolic dysfunction was possible in 67% of patients, in particular 63% of all patients showed a Grade I of diastolic dysfunction, while a Grade III was noted in 4%. In the rest of the patients, diastolic dysfunction was not possible, and it was considered indeterminate.

Furthermore, in 38 patients, (84.4%) MR was present in the acute phase, but only in two cases, it was severe [Table 3].

Coronary angiography and ventriculography

All patients underwent coronarography and ventriculography. Forty-one (83.7%) patients had a coronary artery tree free from lesions. In eight (16.3%) participants, coronary angiography showed noncritical stenosis [Table 3].

Complications during hospital stay and survival at short and long-term

Eight patients presented acute complications, in particular: 1 (2%) showed a third-degree AV block, 3 (6%) had episodes of paroxysmal AF, and finally, two (4%) participants died for cardiogenic shock refractory to inotropic agents and intra-aortic balloon pump [Table 1]. Short-term survival in our cohort was, therefore, of 96%.

Out of the 47 patients discharged, 11 left the study (drop-out) due to a lack of compliance with telephonic interviews. Therefore, FU was performed on the remaining 36 (76.6%) patients. Seven (19.4%) died during FU with an estimated long-term survival of 80.5%.

The prevalence of MACE was 41% (n = 12) for HF; 6.8% (n = 2) for AMI; 3.4% (n = 1) for TTS relapse and 20% (n = 6) for AF;

The prevalence of minor events was 20.7% for symptoms of chest pain and 34.4% for dyspnea not requiring hospitalization [Table 4].

Table 4

Prevalence of major adverse cardiac events and minor events reported during follow-up

Patients	Value (n=36), n (%)
Deaths	7 (19.4)
Hospitalizations	21 (58.3)
HF	12 (41)
TTS relapse	1 (3.4)
Atrial fibrillation	6 (20)
AMI	2 (6.8)
Dyspnea not requiring hospitalization	12 (34.4)
Angina not requiring hospitalization	6 (20.7)

TTS=Takotsubo syndrome, HF=Heart failure, AMI=Acute myocardial infarction

From Cox-regression analysis for the identification of predictors of MACE, it appeared that LVEF at the time of admission was an independent prognostic predictor variable with an HR of 0.932 (P = 0.004) [Table 5].

Table 5

Cox regression for major adverse cardiac events predictors

	HR	CI	P
Trigger event	1.179	0.204-6.796	0.854
ST elevation	1.460	0.307-6.953	0.635
More than mild MR	5.095	0.844-30.749	0.076
Age	1.015	0.905-1.139	0.799
HR	0.965	0.838-1.111	0.619
Troponin	0.932	0.764-1.136	0.485
LVEF	0.932	0.889-0.978	0.004

CI=Confidence interval, MR=Mitral regurgitation, LVEF=Left ventricle ejection function, HR=Heart rate

Stratifying patients on LVEF median value at admission (40%), Kaplan–Meyer analysis showed that patients with LVEF <40% had a significantly poorer long-term outcome than patients with LVEF ≥40% [Figure 1].

Figure 1

Kaplan Meier curve assessing major adverse cardiac events in patients with left ventricular ejection fraction <40% (red line) or ≥ of 40% (blue line)

Finally, among the 29 patients alive at the end of FU, 27 (93%) underwent to a new evaluation.

All 27 patients presented in NYHA Class I, sinus rhythm, normal ECG, and normal echocardiographic examination with recovery of contractility.

LVEF value went from 48.5% ± 9.8% in the acute phase to 61% ± 6.5% at the end of follow-showing a statistically significant increase (P = 0.004) [Figure 2].

Figure 2

Differences between left ventricular ejection fraction in patients during acute phase and during follow-up

DISCUSSION

TTS is confirmed as a pathology that mainly affects the female sex in postmenopausal age, it is triggered by a physical or emotional stress and more often characterized by an involvement of the LV apex with reduction of the LVEF. The main findings of this study are the following:

Intra-hospital mortality has a lower incidence than extra-hospital mortality (4% vs. 19.4%)

A full myocardial function recovery occurs over time, expressed by a complete LVEF recovery;

While a complete recovery, LVEF at admission is a powerful predictor of prognosis with a greater MACE-free survival for patients with LVEF ≥40%;

These results refer to a sample of 49 participants (48 women), diagnosed with TTS. In this population, the prevalence of main demographic, clinical, electrocardiographic, echocardiographic, and angiographic data are in line with the literature.[1415161718]

Reduced LVEF may be a common finding in patients with TTS, with even lower values than in acute MI.[7] Patients with LVEF <40% were older than those with LVEF >40%, confirming an association between advanced age and LV systolic dysfunction in TTS patients.[71920] According to the higher rates of acute heart failure and cardiogenic shock in patients with LVEF <40%, it is advisable to add this parameter to other already used variables (such as ST-segment changes) to stratify outcome and guide therapeutic decisions.[6]

We could affirm that TTS is characterized by a transient and reversible myocardial stunning with a relatively benign prognosis at short term (4% of intra-hospital mortality) but a poorer long-term outcome (19.4% of extra-hospital mortality). Indeed, the literature data show that the mortality rate among patients with TTS ranges from 1% to 8%.[21]

Although an in-hospital poor outcome can be expected in patients with severe LV systolic dysfunction at presentation, the peculiarity of this study is the finding that these types of patients had a poor long-term prognosis, as evidenced by higher major cardiovascular events rates, despite a complete recovery of LV systolic function over time.

Despite a complete recovery of LVEF, persistence of subtle LV systolic and diastolic abnormalities, with the development of fibrosis and myocardial metabolic changes, have been reported at a long-term.[2223] It is possible that in patients, who suffer at admission of a more severe acute myocardial dysfunction, can persist structural and metabolic impairment after discharge. In this hypothesis, it is not really surprising that patients with severely reduced LVEF at admission can have worse outcomes at long-term FU.

These results confirm previous ones by Citro et al., who showed higher frequency of fatal events in patients with LVEF less than or equal to 35%,[24] but can also add the information that a complete recovery of LVEF occurs in all patients over the time, not influencing, apparently, risk of development of future major cardiovascular events.

Patients with a marked cardiac dysfunction seem to be a peculiar group in which age, comorbidities, and maybe genetic factors not only play a role in the acute phase and intra-hospital prognosis, but in long-term outcome too. Medications such as beta-blockers might be useful in the long-term treatment of patients with severe cardiac dysfunction during the acute phase.

Finally, the perception of TTS as having a benign prognosis should be ruled out.

Limitations

Our study suffers of some limitations that should be acknowledged. First, this is an observational study; hence, our findings should be considered hypothesis generating rather than conclusive. Second, the number of patients enrolled is not very large. This point is of great importance and future dedicated studies should enrollee a greater number of patients.

CONCLUSIONS

Despite apparent myocardial functional recovery, patients with TTS presenting with severe LV systolic dysfunction remain at high risk during long-term FU, showing increased rehospitalization, acute heart failure, and all-cause mortality rates.

Echocardiography allows a quick evaluation of cardiac function providing a series of predictive parameters of acute hemodynamic impairment and unfavorable long-term outcome. Physicians should keep in mind that patients presenting with a more severe systolic dysfunction have not only a poor short-term prognosis, but a poor long-term one leading a more “careful” FU.

Ethical clearance

The study was approved by the institutional Ethics Committee of the University of Messina, approval no 61/18.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.