Impact of triggering event in outcomes of stress-induced (Takotsubo) cardiomyopathy

Background: Takotsubo syndrome is also known as stress cardiomyopathy because of the regularity with which it has been associated with physical or emotional stress. Such stress may well be a trigger of the syndrome. Aims: This analysis was undertaken to describe our experience with this disorder and in particular to examine the effects of the underlying trigger on outcomes. Methods: We conducted a retrospective review of the medical records of 345 consecutive patients treated at our institution from 2006 to 2014. All presented with acute cardiac symptoms, a characteristic left ventricular contraction pattern (typical, atypical), and no major obstructive coronary artery disease. Patients were grouped based on their triggering event: (a) medical illness; (b) post-operative period; (c) emotional distress; or (d) no identified trigger. Baseline demographic characteristics, death in hospital, length of stay in hospital, and cardiac complications were abstracted from the patients' medical records. Results: The mean +/- SD age of the population was 72 +/- 12 years and 91% were women. No significant difference in baseline characteristics was noted between the groups except for a higher prevalence of African Americans in the group with a medical illness. ST elevation was noted in 13.3% of patients and the average peak troponin level was 5 +/- 12 ng/dl. An inotropic drug was required in 49 (14.2%) patients, an intra-aortic balloon pump in 37 (10.7%) patients, and mechanical ventilation in 54 (15.7%) patients; 43.5% required treatment in the intensive care unit. Overall, 12 (3.5%) patients died. In only two (16.7%) patients was a there a direct cardiac cause of death. In those patients in whom the cardiac manifestations seemed to be triggered by a medical illness, the death rate was 7.1% and this was significantly higher than in the other groups (p=0.03). Medical illness (odds ratio=6.25, p=0.02) and ST elevation (odds ratio=5.71, p=0.04) were both significantly associated with death. Conclusions: Our study showed that different triggers for Takotsubo syndrome confer different prognoses, with medical illness conferring the worst prognosis. Overall, the in-hospital death rate was low and mostly related to non-cardiac death secondary to the underlying medical illness. Although an unidentified trigger was prevalent in a third of this population, efforts should be made to identify the triggering event to classify the risk group of patients with Takotsubo syndrome.