Clinical Characteristics and Outcomes of Southeast Asian Patients with Takotsubo Syndrome in Singapore

Takotsubo syndrome (TTS) is a form of stress cardiomyopathy first described in 1990 by Sato et al.¹ Its name arises from the Japanese word describing an octopus trap, due to the appearance of the left ventricle in patients diagnosed with the condition.

TTS is characterised by ventricular dysfunction in the absence of culprit epicardial coronary artery disease, and is often associated with intense physical or emotional stressors.^2–8 Ventricular dysfunction is typically transient, but can recur in some instances.⁹ The clinical presentation of TTS can mimic an acute coronary syndrome, with chest pain, associated ECG changes and elevated cardiac troponins.^2,4,10 The majority of patients with TTS are post-menopausal women with a mean age of 61–75 years.^3,4,8,11,12 Of all the patients presenting with symptoms suggestive of acute coronary syndrome, 2–3% are found to have TTS, along with 5–6% of female patients presenting with similar symptoms.⁹

The underlying pathophysiology of TTS remains unclear, with proposed mechanisms including coronary spasm and microvascular endothelial dysfunction.^1,13 A catecholaminergic surge also appears to play a central role.¹⁴

Diagnostic tools are available; however, the diagnosis of TTS is challenging and remains a diagnosis of exclusion after ruling out a plaque rupture event.¹⁵

Previously thought to be a benign condition, it is increasingly recognised that patients with TTS can develop complications including cardiogenic shock, acute pulmonary oedema, arrhythmias and dynamic left ventricular outflow tract obstruction (LVOTO).^{10,11,15–24} Patients with TTS have also been found to have increased mortality and morbidity in both the short and long term.^14,25,26

Our study aimed to describe the clinical characteristics, outcomes and factors associated with adverse outcomes in southeast-Asian patients with TTS in a single tertiary centre.

Methods

Study Population

We conducted a retrospective study of consecutive patients admitted with TTS to a single tertiary cardiology referral centre, between January 2010 and December 2017. All patients underwent coronary assessment using either invasive coronary angiogram or CT coronary angiogram. Significant coronary artery disease was defined as >50% stenosis involving an epicardial coronary artery. TTS was diagnosed according to the InterTAK diagnostic criteria as suggested by the European Society of Cardiology.²⁷

Patients were identified using keywords “stress cardiomyopathy,” “takotsubo cardiomyopathy” and “apical ballooning”, as labelled in their ICD-10 code, coronary angiography report or echocardiogram report.³ A total of 246 patients were identified initially. Patients with no anatomical coronary evaluation were excluded, as well as those whose final diagnosis on discharge was not TTS, as depicted in Figure 1.

Baseline demographic data were collected, along with presenting ECG parameters, biochemical results including peak troponin levels and any identifiable triggers. Initial transthoracic echocardiogram (TTE) findings, coronary evaluation, and subsequent follow-up TTE findings were collected, along with data on medications prescribed during the admission and on discharge.

Outcomes

Our study focused on two main adverse outcomes during the inpatient stay: death and cardiogenic shock. On follow-up echocardiography, the absence of full recovery of LV ejection fraction was also analysed. These outcomes were obtained from their discharge documents, inpatient and outpatient clinical notes.

Statistical Analysis

The baseline characteristics of the study patients were obtained from their admission notes and summarised as frequencies and percentages for categorical variables and mean with SD for continuous variables. Associations between the determinants and analysed outcomes were analysed using univariate logistic regression. A p-value of <0.05 was considered statistically significant.

Analyses were performed using IBM SPSS Statistics v23.

Results

Demographic Data and Clinical Characteristics

Table 1 summarises the demographic data and clinical characteristics of our patient population. A total of 68 patients were enrolled, with a mean age of 67.2 years and 82.4% women. The majority of patients had comorbidities of hypertension and hyperlipidaemia. Of note, three patients had a known psychiatric history, and one patient had prior TTS.

A significant proportion of patients (65.7%) had a physical precipitant, of which more than half was due to infection. The large majority (95.6%) was symptomatic and presented with either chest pain or shortness of breath. ECG changes mimicking an acute coronary syndrome were common, of which 54.4% presented with ST-segment elevations. This led to 48.5% of patients undergoing emergency cardiac catheterisation.

The mean LV ejection fraction at presentation was 36.6, and 91.2% of patients presented with the classic apical ballooning pattern of regional wall motion abnormality (RWMA). Evidence of LVOTO was noted in 4.4% of patients, and 8.8% had mitral regurgitation that was at least moderate in severity.

In terms of treatment, 75.0% received a β-blocker, 77.8% received either an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and 69.1% received the loop diuretic furosemide. One patient required mechanical circulatory support in the form of an intra-aortic balloon pump. No patient required extracorporeal membrane oxygenator support.

Clinical Outcomes

The most common complication was cardiogenic shock (23.5%). There were 5.9% of patients who died during the same admission, with infection being the predominant cause of death (Table 2).

On follow-up TTE, 25.0% of patients did not have full recovery of LVEF. The mean LVEF on follow-up was 58%, with a mean time to follow-up TTE of 92 days.

The average duration of follow-up was 5.5 years, and 24 (35%) patients were lost to follow-up.

Variables Associated with Adverse Outcomes

Univariate logistic regression identified age at onset, male sex and physical trigger as statistically significant factors associated with patients developing cardiogenic shock during the inpatient stay (Table 3). Older age was associated with a slightly reduced risk of cardiogenic shock (OR 0.93; 95% CI [0.87–0.99]; p<0.05), while male sex (OR 4.6; 95% CI [1.23–17.26]; p<0.05) and physical triggers (OR 5.13; 95% CI [1.06–24.93]; p<0.05) were associated with increased risk of cardiogenic shock.

All patients who died during the same admission had physical triggers. Other factors analysed were not found to be statistically significant in their association with death during the same admission.

Age at onset (OR 1.09; 95% CI [1.01–1.17]; p<0.05) and concomitant ischaemic heart disease (OR 5.02; 95% CI [1.29–19.49]; p<0.05) were associated with absence of full LVEF recovery on univariate analysis (Table 4).

Discussion

Similar to prevailing demographic data on TTS, our study population consisted predominantly of post-menopausal women in the sixth decade of life.^3,4,8,11,12 The mechanism underlying the large sex disparity is still unknown, and is thought to be centred on post-menopausal oestrogen deprivation and its role in regulating sympathetic drive and microvascular blood flow.⁹

In terms of comorbidities, our study population had a higher baseline cardiovascular risk profile when compared to the local population. Based on the latest National Population Health Survey 2021/2022, the local prevalence of diabetes, hypertension and hyperlipidaemia in Singapore was 8.5%, 37.0% and 31.9%, respectively.^28,29 In contrast, our study population had cardiovascular risk factor prevalence of 42.6%, 63.2% and 61.8%, respectively. This finding was not entirely surprising, given that microvascular dysfunction secondary to atherosclerosis has also been implicated in the pathogenesis of TTS.¹³ Conflicting data exist regarding the prognostic impact of diabetes in TTS. Some studies suggest that diabetes is a protective factor in TTS, while others suggest that it is associated with poor outcomes in patients with TTS.^30–33 In our study population, we did not find a statistically significant association between cardiovascular risk factors and adverse outcomes.

Precipitants of TTS tend to be stressful events; these include psychological stressors, such as bereavement, or conflict or physical stressors, such as severe infections or major surgery. However, up to one-third of patients with TTS do not have a clear precipitant.^11,27 In our study population, the majority of our patients had a physical precipitant, and 13% of patients did not have an identifiable precipitant. Similar findings were reported in two studies involving a southeast Asian cohort and a Japanese cohort, where physical triggers had a higher prevalence compared to emotional triggers.^34,35

The clinical presentation of TTS closely mimics an acute coronary syndrome. Most of our patients presented with either chest pain or shortness of breath and had ECG changes of ST-segment elevation or depression, leading to emergent cardiac catheterisation in almost half (48.5%) of the patients. To date, there have been multiple studies and proposed ECG criteria that sought to distinguish TTS from ST-elevation MI.^36–40 However, due to the significant overlap in clinical features, as well as the time-sensitive nature of diagnosing an acute coronary syndrome, TTS remains a challenging diagnosis to make in the absence of formal coronary evaluation.⁴¹

The typical pattern of RWMA in TTS is an anteroseptal-apical dyskinetic ballooning of the left ventricle with hyperkinetic basal segments, found in the majority (50–80%) of cases.²⁷ In our study, a large majority of patients presented with the typical apical ballooning pattern of RWMA. Patterns of RWMA in TTS are associated with different mechanical complications.^18,42 Mechanical complications of TTS such as LVOTO and mitral regurgitation are not uncommon; a study by Liu et al. in 2015 suggested that the prevalence of such LVOTO was as high as 20%.⁴³ The presence of LVOTO was also associated with increased in-hospital adverse outcomes, as shown in Italian and Spanish TTS registry data.^18,24,42 The effect of LVOTO on adverse outcomes was not demonstrated in our study, likely due to the low event rate.

Previously believed to be benign and self-limiting, TTS is increasingly recognised to be associated with increased mortality and morbidity due to complications from the underlying illness or TTS itself. SWEDEHEART registry data suggest that mortality in TTS is comparable to mortality in MI, and risk of serious adverse events has been described to be as high as 20% during the initial hospitalisation period.^11,14 These adverse outcomes included cardiogenic shock, cardiac arrest due to malignant arrhythmias and stroke.^{9–11,16–21}

Data from the RETAKO, InterTAK, GEIST and more recently established ChiTTS registries showed that patients with physical triggers were more likely to be male and have acute complications.^44–49 Lower LVEF and younger age were also independent predictors of adverse outcomes.^{11,44–46,49,50} Similar findings were noted in our study for age, male sex and physical trigger. Older age was mildly associated with a reduced risk of cardiogenic shock, while a physical trigger and male sex were strongly associated with increased risk of cardiogenic shock. Furthermore, all patients who died had a physical trigger. The association of older age groups with better outcomes in TTS has no clear pathophysiologic explanation to date. However, some studies postulated that because of augmented cardiac sympathetic stimulation in older age, relatively milder stressors are able to induce TTS.^51,52 In contrast, a more intense stressor, such as severe infection or major surgery, is required to induce TTS in a younger patient, and these patients tend to have poorer outcomes due to the severity of the initial stressor. In patients with physical triggers, the increase in adverse events is thought attributable to the underlying disease itself, although the reason still remains unclear.⁴⁷ A study comparing Japanese and European patients in the InterTAK registry found that the presence of physical stressors was the most important prognostic factor in both patient populations, and Japanese patients had worse in-hospital outcomes driven by the higher prevalence of physical triggers.³⁵

Templin et al. described in their 2015 study that angiotensin-converting enzyme inhibitor or angiotensin receptor blocker usage was associated with improved survival, and data from the GEIST registry reported potential survival benefit among TTS patients with admission EF ≤40% or diabetes.^11,53 Chong et al. also reported an association between β-blocker use on discharge and reduced long-term major adverse cardiovascular events based on data from the Chinese registry.⁴⁹ However, to date there are no randomised trials to support specific treatments in TTS.^9,13 In our study as well, there were no medications or interventions that were associated with outcomes during inpatient stay or on longer term follow-up. β-blockers have been thought to be a promising medication class, given the association of TTS and high circulating catecholamine levels; however, trial data have so far been unyielding, apart from animal studies showing improvements in RWMAs.^2,54,55

On follow-up echocardiography, the presence of ischaemic heart disease and increased age may result in incomplete recovery of LVEF and RWMAs due to concomitant ischaemic cardiomyopathy.

Limitations

One limitation of our study was the small sample size, making it challenging to assess individual outcomes with a low incidence. Furthermore, five patients did not attend their follow-up echocardiograms, which would have affected our statistical analysis of factors associated with recovery of LVEF.

LVEF is not the only marker of recovery for patients with TTS. Increasing attention is being paid to the long-term morbidity of TTS, and many patients describe persistent symptoms, including lethargy, palpitations and dyspnoea that can persist despite normalisation of LVEF. Due to the retrospective nature of the study, we were unable to collect data regarding symptoms and quality of life, which impact patients significantly.^25,26 This may potentially be explored in future studies.

Other gaps in evidence include comparison with patients who have acute coronary syndrome in the Asian context, the usage of strain echocardiography and other imaging modalities in the assessment of TTS, as well as their ability to predict outcomes.

Conclusion

TTS is not benign, and much is still unknown about its pathophysiology, predictors of outcomes and effective treatment options. Physical triggers were associated with worse in-hospital outcomes, while older age was associated with a slightly reduced risk of adverse outcomes in our study. Further research efforts are necessary to evaluate treatment strategies that improve short- and long-term outcomes and reduce morbidity in TTS.