Takotsubo syndrome is characterized by the abrupt onset of angina-like chest pain, ST elevations and/or T wave inversions, subsequent elevation of cardiac biomarkers, and a marked decrease in left ventricular systolic function. It occurs most commonly in women, aged 62 to 75. Patients generally present with chest pain, but may also complain of dyspnea or rarely, syncope (Bybee et al 2004). Imaging of the ventricular function reveals a hyperdynamic basal segment and apical akinesis of the left ventricle, whereby takotsubo mimics acute myocardial infarction (Dote et al 1991; Bybee et al 2004).There remain, however, important differences between the two diagnoses. For example, in takotsubo cardiomyopathy, peak troponin values are often observed on initial presentation as opposed to the later peak as seen in myocardial infarction. Furthermore, if any coronary artery disease is discovered on angiogram, it is incongruent with the akinesis visualized. Finally, in takotsubo cardiomyopathy, resolution of the impaired left ventricular systolic function occurs in 7 to 30 days following presentation (Bybee et al 2004).

The pathogenesis of takotsubo cardiomyopathy is incompletely understood. Transient myocardial arterial thrombosis, with subsequent dislodgement of clot or plaque seems unlikely because of the absence of a plausible blockage of coronary arterial territory producing ischemia congruent with the area in which the akinesis is seen. Lymphocytic myocarditis is characterized by a reversible deficit of left ventricular output due to hypokinesis in areas extending beyond a single coronary artery. The hypokinesis seen in myocarditis, however, is global and does not have the characteristic apical ballooning seen with takotsubo. Coronary vasospasm without thrombosis is also a consideration; however, past studies have failed to determine clearly if vascular spasm is the direct mechanism of injury or if the vascular spasm is simply a secondary phenomenon in takotsubo cardiomyopathy (Bybee et al 2004). The most commonly accepted theory is that stress induced catecholamine release causes myocardial damage by microvascular spasm or by direct cardiotoxicity. Microvascular ischemia might account for the lack of a single coronary artery territory producing ischemia in the apical area of akinesis (Dote et al 1991; Bybee et al 2004). Adrenergic receptor density appears to be heterogeneous throughout the left ventricle, which might explain the regional variation in myocardial damage that is observed during the high catecholamine state. Parallels have been drawn to the myocardial stunning seen with subarachnoid hemorrhage and high levels of catecholamines found in pheochromocytoma. Genetic variations in catecholamine receptor location could place susceptible individuals at increased risk for this syndrome (Van De Walle et al 2006).

Although takotsubo syndrome is normally found in women, both patients in these cases were men (Dote et al 1991; Bybee et al 2004). Both patients received bevacizumab as chemotherapy for metastatic cancer. Bevacizumab is a vascular endothelial growth factor (VEGF) inhibitor. VEGF inhibition is a novel target for chemotherapy, and the potential utility of this class of drugs is quite large. While the role of VEGF in embryonic development and neoplastic processes is widely accepted, it is not clear what role this molecule might have in the adult vascular system. There is some suggestion that VEGF may perform arterioprotective functions (Zachary et al 2006). In an experiment designed to elucidate the molecular biology of cardiotoxicity secondary to chemotherapy, high levels of inflammatory cytokines and chemoattractant proteins were isolated in the plasma and myocardial tissue of mice that were exposed to bevacizumab. Researchers have proposed that the expression of these inflammatory cytokines and chemoattractant proteins in plasma during anti-cancer therapy could be indicative of cardiotoxicity (Drimal et al 2006). In an animal model of heart failure and pressure overload (with transaortic constriction), mice that were genetically engineered to block signaling of VEGF demonstrated a reduction in cardiac hypertrophy with increased left ventricle dilatation and a loss in contractile function. These findings suggest that inhibition of VEGF is associated with the transition from compensatory hypertrophy to overt failure (Izumiya et al 2006).

VEGF inhibition has also been associated with arterial thrombotic events. A recent meta-analysis, which included five randomized controlled trials, found that the risk of arterial thrombosis was higher in cancer patients receiving bevacizumab as part of their regimen, than in those patients whose regimen did not include bevacizumab (Scappaticci et al 2007). These findings are consistent with the statistics from prescribing information for bevacizumab. Arterial thrombotic events, such as myocardial infarction and angina, occurred in 4.4% of patients whose regimen included bevacizumab as opposed to 1.9% of patients on a regimen without bevacizumab (Genetech, Inc. 2008). Given the similarities between myocardial infarction and takotsubo cardiomyopathy, it is conceivable that some of the arterial thrombotic events recorded as myocardial infarction might actually have been unrecognized incidents of takotsubo cardiomyopathy.

The cases are somewhat atypical in that both occurred in men and neither patient complained of chest pain. However, both patients had prominent, diffuse ST segment elevations on ECG, the most common ECG finding in takotsubo cardiomyopathy. New Q-waves have been observed in up to 31% of patients, and in some instances, the Q-waves have been a transient phenomenon (Bybee et al 2004). As illustrated in Figure 3, the second patient did have Q-waves on presentation. In this case, the Q-waves persisted despite recovery of ventricular function. Figures 4a and 4b show clearly that both patients had significant left ventricular dysfunction. In Figure 4a, there appears to be some degree of asymmetry in the balloon effect. While takotsubo cardiomyopathy is classically described as apical ballooning, cases of midventricular ballooning have been reported. Although the exact mechanism is unclear, it is believed that the pathophysiology underlying transient midventricular ballooning is the same as that of takotsubo cardiomyopathy (Hurst et al 2006). In the second case, Figure 4b, the ventricular ballooning is more classical in distribution, more closely resembling the octopus trap. In both cases, the affected area is not related to a single coronary artery distribution, which supports an etiology other than coronary artery disease. This is a hallmark feature of takotsubo cardiomyopathy. Another prominent feature is the recovery of ventricular function within weeks. Both patients demonstrated recovery of myocardial function within 30 days.