Takotsubo Cardiomyopathy

During the 15 years that have passed since the introduction of the term “takotsubo” transient apical cardiomyopathy (TAC), the rich literature on this confusing syndrome has contained little that definitively clarifies any fundamental aspect of the syndrome except for the meaning of the Japanese word tako-tsubo (“octopus trap”). The accompanying article by Lee and colleagues¹ regarding 2 cases of “inverted TAC” proposes more questions than answers, but questions are also part of the learning process, so they are welcome.

Like the better-known and more frequent “midventricular variety” of TAC, inverted TAC is an important alternative manifestation of what is frequently claimed to be a continuous spectrum of TAC, which includes the typical apical variety. Phenotypically, TAC is defined by the following features: a) a transient event; b) the sudden onset of severe segmental left ventricular (LV) dysfunction, which may affect different areas of the LV and is commonly independent of the territory of any single coronary branch; c) the sudden onset and evolution of ST-segment changes (elevation, depression, or both); d) mild elevation of the cardiac enzyme levels with respect to the LV dysfunctional area; and e) an absence of coronary obstructive lesions capable of explaining the LV dysfunctional patterns on early coronary angiography.

The only mechanism known by cardiologists to cause similar transient LV dysfunction clearly is myocardial stunning,² which is usually related to transient coronary occlusion. One suggested explanation for TAC is direct, simple catecholamine overload, also known as “catecholamine heart” or “broken heart” in the English-language medical literature. Actually, catecholamine infusion or endogenous hyperactivity does not usually cause either ST elevation or hypotension but does cause (mild) ST depression, hypertension, and severe tachycardia.

Some patients with an increased catecholamine load may also have vasospasm (although this has never been well demonstrated), but the great majority of patients with a spontaneous or pharmacologic catecholamine overload do not present with TAC.³

The most credible cause of transient (and spontaneously resolving) segmental myocardial stunning is coronary spasm, not hemodynamic overload or metabolic stress. In most of the thousands of single case reports or short series regarding TAC in the current literature, the authors have quickly dismissed—on the basis of only a few early reports of retrospective studies that had inadequate descriptions of the testing protocols—the hypothesis that spasm is the pathophysiologic cause of TAC.⁴

In contrast, our group^5–7 recently resurrected the spasm hypothesis by reporting a pilot series of patients who were prospectively subjected to acetylcholine (ACh) testing during the early recovery period after they had presented with TAC. In this series, such testing of endothelial function^8,9 consistently resulted in a substantially abnormal response characterized by severe (subtotal) spastic narrowing of extensive coronary territories, accompanied by echocardiographic documentation of transient LV dysfunction, as observed in the original, spontaneous presentation of TAC.^4–6 Our experience is admittedly preliminary (we are currently pursuing the goal of performing a larger, controlled study), but it strongly suggests that not only the most typical, apical form, but also the midventricular variety, of TAC is a manifestation of transient endothelial dysfunction, which usually disappears spontaneously a few days after clinical presentation. Also, intracoronary administration of nitroglycerin results in safe, dependable, total resolution of both spasm and LV dysfunction.^4,5

On the basis of preliminary ACh testing, we have proposed that TAC seems to result from a combination of endothelial dysfunction and precipitating factors (mental or physical stress, pain, or some other crisis), which may be mediated in some cases by catecholamine administration or excessive endogenous catecholamine production. Those rare, but well-described, patients (approximately 10% of those with TAC) who have recurrent presentations seem to be the ones with a predominantly endothelial dysfunctional pattern (as reflected by a persistently positive ACh test result over a period of months). The behavior of this condition is similar to that of Prinzmetal angina, a well-known syndrome that involves spontaneous recurrent coronary spasm and that can be reproduced with ACh testing. The most substantial difference between the 2 conditions seems to be that in Prinzmetal angina, only a short coronary segment (typically in the proximal left anterior descending artery) shows enhanced spasticity, which is likely secondary to specific dysfunction of its neural control. A continuous spectrum probably exists between typical TAC and Prinzmetal angina.⁴

In addition, in our early series of ACh tests, we documented that the spastic pattern in typical TAC is different from that in midventricular TAC,⁵ with varying involvement of the different branches of the left coronary arteries. Whereas typical TAC seems to involve critical, diffuse narrowing of all the mid- and distal branches of the left coronary artery, midventricular TAC seems to involve relative sparing of the left anterior descending territory, leading to apical contractile preservation⁵ during echocardiographic monitoring (Fig. 1). We have yet to observe any cases of “inverted TAC,” and it would be interesting to know how this condition would respond to an ACh challenge. A remaining fundamental question concerns whether it is possible for a patient to have “global” TAC (generalized ventricular dysfunction). Most probably, this condition can and does occur; but, if sustained, it would lead to early and sudden death. Not even a necropsy would likely be able to reveal the mechanism of death in such cases (the coronary arteries are usually intact, and the myocardium is “normal” on routine examination). Additional fascinating questions are, “Why is TAC so rarely recurrent?” and “What leads to spontaneous cancellation of endothelial dysfunction and spontaneous recovery?” Surely, the precipitating factors (pain, dyspnea, anxiety, or other stressors) do not improve but, rather, tend to persist. For example, in the 2 cases reported here, Lee and colleagues¹ indicate that the precipitating factors were sepsis in the presence of metastatic cancer in the first case and unclear mild factors in the second case; in both patients, the precipitating factors remained active during spontaneous recovery from TAC.

Fig. 1 Schematic representation of 2 different responses to acetylcholine (ACh) testing, as evaluated by left ventricular angiography. Drawing A shows a baseline, “normal” coronary lumen. Drawing P1 shows that ACh infusion is followed by severe, distal, subtotal spastic coronary occlusion, with simultaneous apical ballooning (typical takotsubo cardiomyopathy). Alternatively, as is shown in drawing P2, ACh infusion is followed by diffuse severe coronary distal narrowing, with simultaneous preservation of the luminal integrity of the left anterior descending coronary artery (midventricular variety of takotsubo cardiomyopathy).

In conclusion, it is essential to better qualify the pathophysiologic mechanisms in individual cases of TAC,⁷ especially in order to document the common baseline characteristics, the effect of treatment, and the prognosis (tendency to recur). Moreover, at experienced centers, the need for routine emergency catheterization is evolving toward elective (pre-discharge) catheterization, including an ACh test. A phenotypic diagnosis of TAC can, in fact, also be established simply on the basis of echocardiography and the clinical presentation; in doubtful cases, computed axial tomographic angiography can also be used to rule out significant coronary occlusive disease. Because of the mechanistic and prognostic information it can provide, endothelial-function testing appears worth the limited discomfort, risk, and cost that it entails.