Takotsubo Syndrome: Pathophysiology, Emerging Concepts, and Clinical Implications

Trisha Singh, BM; Hilal Khan, MB BCh BAO, MRCP; David T. Gamble, MPharm, MBBS; Caroline Scally, MBChB; David E. Newby, DM, PhD; Dana Dawson, MD, DPhil


Circulation. 2022;145(13):1002-1019. 

In This Article

Pathogenesis and Pathophysiology

Understanding the underlying pathogenesis and pathophysiology (Figure 5) is critical if appropriate treatments are to be developed for the acute episode, and the prevention of subsequent long-term recurrent events, as well. Furthermore, it is unknown whether the underlying mechanism varies according to the anatomic variation of the left myocardial ballooning.

Figure 5.

Mechanisms involved in takotsubo syndrome.

Adrenergic Hypothesis

Endogenous adrenergic surge is the most established theory for the pathogenesis of takotsubo syndrome and is intuitive given the strong association with sudden unexpected stress and major physical illness or trauma.

Two principal aspects to this mechanism need to be considered: the release of catecholamines and the cardiac response to catecholamines. The locus coeruleus, located in the posterior part of the rostral pons, is the primary origin of norepinephrine in the brain. Receiving multiple inputs from the limbic system, it regulates the homeostatic response to emotions.[71] The locus coeruleus is triggered by emotional stimuli and leads to the production of norepinephrine, which in turn activates the hypothalamic-pituitary-adrenal axis.[71] In response to this activation, the adrenal medulla releases epinephrine and norepinephrine into the circulation, thereby increasing plasma catecholamine concentrations.

Iga and colleagues[72] were the first to describe takotsubo syndrome in a patient with pheochromocytoma. This particular case report was important because it first established the relationship between takotsubo syndrome and elevated catecholamine concentrations. Subsequent observational studies have demonstrated elevated blood catecholamine concentrations in the acute phase, although this has not been a universal finding.[73] Preclinical studies have shown that intravenous or intraperitoneal administration of norepinephrine, epinephrine, or isoprenaline can reproduce the characteristic reversible apical left ventricular ballooning coupled with basal hypercontractility.[74,75]

Independent of a systemic increase in catecholamine concentrations through the hypothalamic-pituitary-adrenal axis, a local neurally mediated increase in catecholamine release at myocardial level may also occur.[76] Apart from the locus coeruleus, neural impulses descend (from the rostral pons) into posterior hypothalamus triggering norepinephrine release from sympathetic nerve terminals supplying the myocardium and coronary circulation. Increase in local production may explain why plasma catecholamine concentrations are not always elevated, and several studies have demonstrated myocardial sympathetic hyperactivity.

The second aspect is the cardiovascular response to a surge in catecholamines and how this results in left ventricular dysfunction. Mammalian hearts demonstrate the highest density of β-adrenergic receptors in the apex, although the distribution of β1- and β2-adrenoreceptors has never been mapped in the human heart.[23] Increased responsiveness of the left ventricular apex to catecholamines could explain the characteristic apical ballooning of takotsubo syndrome.[77]

Some have suggested direct myocardial injury from catecholamine excess. Furthermore, catecholamines released directly into the myocardium through sympathetic nerves may have a greater toxic effect than that reaching the heart through the bloodstream.[77] Norepinephrine spillover can decrease myocyte viability, resulting in contraction band necrosis, which is one of the histological findings reported in takotsubo syndrome (Figure 2).[78,79] Contraction band necrosis is also found in patients with pheochromocytoma and subarachnoid hemorrhage, which have also been attributed to catecholamine excess.[78,79]

Some authors have hypothesized that a surge in adrenoreceptor stimulation dysregulates myocardial calcium handling. Immunohistochemistry studies suggest that calcium-regulating proteins such as phospholamban, sarcoendoplasmic reticulum calcium-ATPase, and sarcolipin are altered, resulting in depressed left ventricular contraction during acute takotsubo syndrome.[80] Studies are presently underway that aim to assess myocardial calcium handling in patients with takotsubo syndrome (MEMORY study [Manganese-Enhanced Magnetic Resonance Imaging {MEMRI} in Ischaemic, Inflammatory and Takotsubo Cardiomyopathy], NCT04623788; Figure 6). Others have suggested that adrenoreceptor stimulation can create an imbalance of oxygen supply and demand, thus creating myocellular hypoxia. Hypoxia is further exacerbated by metabolic changes and electrolyte imbalances from alterations in membrane permeability that may contribute to myocardial toxicity.

Figure 6.

Abnormal exercise capacity, energetics, and cardiac performance in patients with takotsubo syndrome.
Cardiopulmonary exercise (treadmill) data from a patient with previous takotsubo syndrome 20 months earlier (A) and an age- and sex-matched healthy control subject (B). The maximal oxygen consumption (Vo2, blue dots) achieved by the patient with takotsubo syndrome is markedly reduced, with an earlier anerobic threshold and a shorter duration of exercise than the healthy control subject. 31Phosporus magnetic resonance spectrum acquisition for a patient with acute takotsubo syndrome (C) and healthy volunteer (D) showing reduced phosphocreatine/adenosine triphosphate ratio. Resonances correspond to phosphocreatine (PCr), γ, β, and α adenosine triphosphate (ATP), and 2,3-diphosphoglycerate (2,3 DPG). Manganese-enhanced magnetic resonance imaging in a patient with acute takotsubo syndrome (E) and an age- and sex-matched volunteer (F). The patient demonstrates abnormal calcium activity (green) throughout the mid ventricle and apex with preserved calcium activity in the basal segments (blue). In comparison, the healthy control subject demonstrates normal calcium activity throughout the myocardium (blue). Twist curves in a patient with takotsubo syndrome at 2-year follow-up (G) and an age- and sex-matched healthy control volunteer (H). The healthy subject demonstrates the characteristic early systolic twist in a clockwise rotation at the apex (blue trace) and in counterclockwise rotation at the base (purple trace), occurring during isovolumic contraction. This is followed by counterclockwise apical rotation (Ar; blue) and clockwise basal rotation (Br; purple), which results in the net systolic twist during left ventricular ejection (twist, white line). In comparison, the patient with takotsubo syndrome demonstrates incomplete recovery of left ventricular twist, predominantly because of the reduced apical rotation. AT indicates anerobic threshold; Vco2 (mL/min), volume of carbon dioxide exhaled; Vo2 (mL/min), volume of oxygen inspired; and Vo2 Max, maximal volume of oxygen inspired.

At high concentrations, epinephrine can act as a negative inotrope through ligand-mediated stimulatory Gs protein. After cAMP-dependent phosphorylation of the β2-adrenoreceptor, receptor coupling is switched from the Gs to Gi protein.[75] The Gi protein prevents excessive activation of the myocyte and reduces contractility, which protects the myocardium from the effects of excess catecholamine stimulation and restricts cardiac damage.[75] The switch back to Gs protein is responsible for the rapid recovery in these patients and could explain resolution of left ventricular dysfunction. However, this theory does not explain all the features of takotsubo syndrome, such as the presence of marked myocardial edema, absence of a stressor in some patients, and why only two-thirds of patients have elevated catecholamine concentrations.

In the final analysis, it should be remembered that patients with takotsubo syndrome are neither tachycardic nor hypertensive at presentation, and this contrasts with all other conditions known to have high catecholamine surges. Thus, takotsubo syndrome might affect the autonomic nervous system in a much more complex way than a simple and pure catecholamine surge.

Brain–Heart Axis

For many patients, psychological stress is the central trigger for takotsubo syndrome, even in the presence of a physical illness, which arguably induces an element of psychological stress. Patients with takotsubo syndrome are more likely to have preexisting psychiatric illness. Patients with depression and anxiety have upregulated microRNA 16 and 26a, and in a rodent model of microRNA 16 and microRNA 26a overexpression, exogenous epinephrine was associated with apical wall motion abnormalities.[81,82] This mechanism is consistent with a predilection of the myocardium to develop takotsubo syndrome in response to stress and could explain the high prevalence of preexisting and acute psychiatric illness in affected individuals.

Neurocardiogenic stunning of the heart is a well-recognized complication after acute neurological injury, and 20% to 30% of patients develop transient left ventricular systolic dysfunction, highlighting the complex brain-heart interaction.[83] Patients with takotsubo syndrome demonstrate altered neuronal connectivity in several stress-associated limbic regions. Altered neuronal activity is predominately seen in the hippocampus, amygdala, cingulate gyrus, and insula and is important in regulating emotional responses and the autonomic nervous system.[84] Furthermore, 18F-fluorodeoxyglucose positron emission tomography imaging has demonstrated heightened amygdala activity years before patients develop takotsubo syndrome.[84] This suggests that, when presented with a potential trigger, such individuals may have a weakened ability to respond appropriately, and an imbalance of the sympathetic and parasympathetic nervous systems results in myocardial injury similar to that of an acute neurological insult.[85]

Several case reports have described takotsubo syndrome in patients after cardiac transplantation. Given denervation of the heart after cardiac transplantation, this does argue against a direct neural stimulus as the trigger for takotsubo syndrome and suggests that humoral mechanisms may have a more prominent role. It has been postulated that parasympathetic denervation may lead to upregulation of catecholamines and β-adrenoreceptors.[86] Combined with impaired neural innervation, this may result in an exaggerated response to catecholamines and susceptibility to takotsubo syndrome. The brain–heart axis is yet to be properly explored and possibly holds the answers to a number of questions that elude us regarding takotsubo syndrome and possibly other cardiac diseases.

Coronary Vasospasm and Microvascular Reactivity

Initial cases from Japan demonstrated multivessel epicardial coronary vasospasm on coronary angiography, raising the possibility that multivessel vasospasm may be a causative factor in the pathogenesis of takotsubo syndrome.[77] However, repeated provocation testing in patients with takotsubo syndrome found such responses to be inconsistent with only ≈20% of patients demonstrating reproducible vasospasm.[87] Furthermore, such provocation testing is not physiological, has poor reproducibility, and is of uncertain relevance to spontaneous epicardial spasm.

Microvascular dysfunction and impaired reactivity do appear to be a feature of takotsubo syndrome with demonstrable reversible abnormalities in both the coronary flow reserve and the index of microvascular resistance. Postmenopausal women have age-related and estrogen deficiency–related coronary vasomotor dysfunction. Under physiological circumstances, estrogen improves coronary blood flow through endothelium-dependent and -independent mechanisms, but its deficiency results in increased sympathetic drive and endothelial dysfunction.[88] These changes may in part explain the preponderance of takotsubo syndrome occurrence in postmenopausal women.

Nuclear myocardial perfusion studies have reported reduced apical perfusion, which gradually recovers at 1 and 6 months.[89] Such microvascular dysfunction and abnormal perfusion may be driven by vasoconstrictor mediators such as endothelin, catecholamines, and the associated reactive oxygen species.[90] Another possible explanation may be myocardial inflammation leading to direct myocyte injury, including vascular endothelial injury causing shedding of the endothelial glycocalyx and consequent myocardial edema. However, there are issues of cause and effect for both myocardial edema and microvascular dysfunction that may be a consequence rather than a cause of the acute episode.

Wittstein[91] postulated the interaction between sympathetic overactivity and microvascular dysfunction and its impact on clinical presentation in patients with takotsubo syndrome. High-risk individuals with elevated sympathetic tone and vasomotor dysfunction (postmenopausal status, depression, and treatment with serotonin reuptake inhibitors) may only require a mild stimulus to precipitate microvascular ischemia and subsequent myocardial stunning.[91] Conversely, low-risk individuals with normal sympathetic and vasomotor tone will likely require a much larger catecholamine surge to precipitate acute takotsubo syndrome. This may explain why some patients present after seemingly mild triggers.

Metabolic and Energetic Alterations

Current data demonstrate metabolic and energetic impairment in acute takotsubo syndrome followed by a protracted and incomplete recovery.[43] Preclinical studies have shown increased myocardial glucose uptake, and although there were appropriate increases in enzymes involved in the glycolytic pathway, there was a reduction in the available metabolites of glycolysis,[90] resulting in a decreased production of Kreb cycle intermediates. It is unknown whether this decrease is attributable to a state of myocardial metabolic enhancement leading to exhaustion and loss of metabolites or if this is attributable to myocardial metabolic stunning leading to enzymatic blockade of the glycolytic, β-oxidative, or pentose phosphate pathways.[90]

The in vivo gold standard for exploring myocardial energetics is 31-phosphorus cardiac magnetic resonance spectroscopy. Resting cardiac energetic status (phosphocreatine to γ-ATP ratio) is reduced in patients with acute takotsubo syndrome (Figure 6). Although there is some recovery by 4 months of follow-up, it has still not completely normalized.[92] Indeed, abnormal long-term myocardial metabolism may explain why patients continue to be symptomatic and have recurrent events despite apparent recovery of left ventricular ejection fraction.[51]

Inflammatory Mechanisms

There is growing evidence to support the presence of myocardial inflammation in the acute phase of takotsubo syndrome. Although this will, in part, be a reaction to the precipitating event, it may be both cause and effect. Furthermore, a maladaptive persistent subacute or chronic inflammation may contribute to long-term cardiac dysfunction.

In a multicenter study (TERRIFIC [Pathogenesis of Acute Stress Induced {Tako-tsubo} Cardiomyopathy: Energy Shut-Down or Intense Inflammation?], NCT02897739),[43] patients with takotsubo syndrome had greater retention of ultrasmall superparamagnetic particles of iron oxide in both ballooning and nonballooning left ventricular segments during the acute phase. Because ultrasmall superparamagnetic particles of iron oxide are predominantly phagocytosed by activated tissue-resident macrophages, the main cellular protagonists of the myocardial cellular inflammation in acute takotsubo syndrome appear to be macrophages, whereas acute myocarditis is principally lymphocyte mediated. In addition, serum interleukin-6 and chemokine (C-X-C motif) ligand 1 concentrations, and classic CD14++CD16 monocytes, as well, are increased, whereas intermediate CD14++CD16+ and nonclassic monocytes are reduced in patients with takotsubo syndrome. At 5 months of follow-up, enhancement with ultrasmall superparamagnetic particles of iron oxide was no longer detectable in the myocardium, although persistent elevations in serum interleukin-6 concentrations and reductions in intermediate CD14++CD16+ monocytes were present. Therefore, takotsubo syndrome is characterized by a myocardial macrophage inflammatory infiltrate, changes in the distribution of monocyte subsets, and an increase in systemic proinflammatory cytokines. Many of these changes persisted for at least 5 months, suggesting a low-grade chronic inflammatory state. Furthermore, postmortem examination of human hearts from patients who died during the acute phase of the condition demonstrated that these macrophages are predominantly of the M1 proinflammatory type as opposed to the reparative M2 type.[93] The presence of M1 macrophages and the persistence of the intermediate (CD14++CD16+) monocyte subset at 5 months follow-up are strongly indicative of a less reparative and more proinflammatory state compared with similar stages of patients with acute myocardial infarction. It remains unclear, however, if this inflammatory activation is causative or consequential to takotsubo syndrome. Nevertheless, these findings offer an explanation for the low-grade chronic inflammatory substrate with subsequent evolution of acute takotsubo into long-term heart failure.

Persistent or Preexisting Syndrome

The heart failure phenotype of takotsubo syndrome has been thoroughly characterized in a cohort of predominantly symptomatic patients (the HEROIC study [Persistent Symptoms and Early Incomplete Recovery After Acute Stress-induced Cardiomyopathy: Is There Ongoing Heart Distress?], NCT02989454)[51] and demonstrates preserved ejection fraction, impaired cardiac energetic status, cardiac limitation on exercise (reduced peak VO2and increased VE/VCO2 slope during cardiopulmonary exercise testing), reduced apical myocardial anticlockwise rotation during systole with altered torsion and twist, and possibly microscopic fibrosis (Figure 6).[43,51] Likewise, cardiac biomarkers, such as BNP, remain mildly elevated long term.[94] This persistence of long-term myocardial abnormalities does beg the question of whether such abnormalities predate the index takotsubo syndrome. This would be consistent with the predisposition for recurrent takotsubo events. Of course, it is difficult to know whether the myocardium in these patients was healthy to begin with or there was a preexisting subtle and undiagnosed cardiomyopathy that is brought to light during an acute stressful event. If this were the case, then it would imply that their cardiac function may never return to normal and has in fact returned to baseline.