Torsades de Pointes Associated with Takotsubo Cardiomyopathy: Is It Preventable?

Takotsubo cardiomyopathy is characterized by transient left ventricular systolic dysfunction and apical dyskinesis (ballooning) in the absence of significant obstructive coronary artery disease. Most patients with Takotsubo cardiomyopathy are females. An emotional stress is believed to be the main trigger of Takotsubo cardiomyopathy (Bybee & Prasad, 2008). Electrocardiographic features of Takotsubo cardiomyopathy might include: ST elevation, T wave inversion, and QT interval prolongation (Thakar et al., 2011). QT interval prolongation might precede Torsades de Pointes a potentially deadly polymorphic ventricular tachycardia. Female gender, left ventricular systolic dysfunction, recent conversion of atrial fibrillation to sinus rhythm, administering QT interval prolonging agents, hypokalemia, hypocalcemia, severe hypomagnesemia, and high-degree atrioventricular block, are all risk factors for acquired QT interval prolongation and Torsades de Pointes (Antzelevitch, 2007; Roden, 2004). The incidence of QT interval prolongation among patient with Takotsubo cardiomyopathy is higher than 50% (Abe et al., 2003; Cangella et al., 2007; Fang et al., 2008; Wittstein et al., 2005), but the incidence Takotsubo cardiomyopathy-associated Torsades de Pointes is probably much lower. It is of great importance to study the clinical circumstances leading to Torsades de Pointes in patients with Takotsubo cardiomyopathy-associated QT interval prolongation since Torsades de Pointes might be fatal, while the prognosis of Takotsubo cardiomyopathy is usually good (Bybee & Prasad, 2008). Takotsubo cardiomyopathy-associated Torsades de Pointes has been reported in 2005 for the first time (Denney et al., 2005). We have reviewed this case report and additional 14 reports (Akashi et al., 2003; Boulouffe et al., 2007; Finsterer et al., 2007; Furushima et al., 2008; Ghosh et al., 2009; Hirose et al., 2008; Inoue et al., 2009; Kurisu et al., 2008; Mahida et al., 2009; Nault et al., 2007; Okada et al., 2007; Patel et al., 2007; Sasaki et al., 2006) concerning Takotsubo cardiomyopathy-associated Torsades de Pointes in 2009, and we have concluded that males with Takotsubo cardiomyopathy-associated QT interval prolongation are at risk for Takotsubo cardiomyopathy-associated Torsades de Pointes although most patients with Takotsubo cardiomyopathy are females. There has been a trend in the mean maximal QT interval being longer among patients with Takotsubo


Introduction
Takotsubo cardiomyopathy is characterized by transient left ventricular systolic dysfunction and apical dyskinesis (ballooning) in the absence of significant obstructive coronary artery disease.Most patients with Takotsubo cardiomyopathy are females.An emotional stress is believed to be the main trigger of Takotsubo cardiomyopathy (Bybee & Prasad, 2008).Electrocardiographic features of Takotsubo cardiomyopathy might include: ST elevation, T wave inversion, and QT interval prolongation (Thakar et al., 2011).QT interval prolongation might precede Torsades de Pointes -a potentially deadly polymorphic ventricular tachycardia.Female gender, left ventricular systolic dysfunction, recent conversion of atrial fibrillation to sinus rhythm, administering QT interval prolonging agents, hypokalemia, hypocalcemia, severe hypomagnesemia, and high-degree atrioventricular block, are all risk factors for acquired QT interval prolongation and Torsades de Pointes (Antzelevitch, 2007;Roden, 2004).The incidence of QT interval prolongation among patient with Takotsubo cardiomyopathy is higher than 50% (Abe et al., 2003;Cangella et al., 2007;Fang et al., 2008;Wittstein et al., 2005), but the incidence Takotsubo cardiomyopathy-associated Torsades de Pointes is probably much lower.It is of great importance to study the clinical circumstances leading to Torsades de Pointes in patients with Takotsubo cardiomyopathy-associated QT interval prolongation since Torsades de Pointes might be fatal, while the prognosis of Takotsubo cardiomyopathy is usually good (Bybee & Prasad, 2008).Takotsubo cardiomyopathy-associated Torsades de Pointes has been reported in 2005 for the first time (Denney et al., 2005).We have reviewed this case report and additional 14 reports (Akashi et al., 2003;Boulouffe et al., 2007;Finsterer et al., 2007;Furushima et al., 2008;Ghosh et al., 2009;Hirose et al., 2008;Inoue et al., 2009;Kurisu et al., 2008;Mahida et al., 2009;Nault et al., 2007;Okada et al., 2007;Patel et al., 2007;Sasaki et al., 2006) concerning Takotsubo cardiomyopathy-associated Torsades de Pointes in 2009, and we have concluded that males with Takotsubo cardiomyopathy-associated QT interval prolongation are at risk for Takotsubo cardiomyopathy-associated Torsades de Pointes although most patients with Takotsubo cardiomyopathy are females.There has been a trend in the mean maximal QT interval being longer among patients with Takotsubo cardiomyopathy-associated Torsades de Pointes relative to patients with Takotsubo cardiomyopathy-associated QT interval prolongation.Moreover, most patients with Takotsubo cardiomyopathy-associated Torsades de Pointes have had risk factors for Torsades de Pointes other than female gender and left ventricular systolic dysfunction (Samuelov-Kinori et al., 2009).Additional 12 patients with Takotsubo cardiomyopathyassociated Torsades de Pointes have been reported since then (Ahn et al., 2011;Gotyo et al., 2009;Grilo et al., 2010;Kawano et al., 2010;Micallef et al., 2010;Pacha et al., 2010;Peters & Klein, 2011;Purvis et al., 2009;Rotondi et al., 2010;Wedekind et al., 2009;Yamada et al., 2011).We have currently examined whether our previous conclusions are true in face of these new publications.Moreover, we have studied if Torsades de Pointes has been possibly preventable in patients with Takotsubo cardiomyopathy-associated Torsades de Pointes.

Retrieval of reports
We performed a literature search by using the following keywords: "Apical ballooning", "Arrhythmia", "Stress cardiomyopathy", "Sudden death", "Syncope", "Takotsubo", and "Torsades de Pointes".The references in each report were further reviewed for additional publications.Only full-length reports were reviewed.The study group included all patients with Takotsubo cardiomyopathy-associated Torsades de Pointes that had been reported until July 2011.The control group included patients with Takotsubo cardiomyopathyassociated QT interval prolongation that had been reviewed previously (Samuelov-Kinori et al., 2009).

Risk factors for Torsades de Pointes
Each case report was analyzed for the presence of risk factors for Torsades de Pointes other than female gender and left ventricular systolic dysfunction: recent conversion of atrial fibrillation to sinus rhythm, administering QT interval prolonging agents, hypokalemia (<3.5 mmol/L), hypocalcemia (<8.5 mg/dL), severe hypomagnesaemia (<1 mg/dL), and high-degree atrioventricular block (Antzelevitch, 2007;Roden, 2004).Since genetic analysis was not available for all patients, suspicion of congenital long QT syndrome was also considered a risk factor for Torsades de Pointes, and was defined as QT interval prolongation in the baseline ECG recorded before Takotsubo cardiomyopathy appearance or following its resolution.QT interval prolongation was defined as QTc >430 msec for male patients and QTc >450 msec for female patients according to the Bazett's formula (Bazett, 1920).We used the QTc that was mentioned in the text of each case report by the authors.In few cases we measured the QT interval length in lead II and calculated the QTc according to the ECG strip enclosed.

Preventable Torsades de Pointes
Each case report of Takotsubo cardiomyopathy-associated Torsades de Pointes was analyzed for the presence or the absence of treatment and prevention measures taken once QT interval prolongation had been noticed and prior to Torsades de Pointes appearance in face of the above-mentioned risk factors for Torsades de Pointes -regardless of whether Takotsubo cardiomyopathy diagnosis had already been made.

Statistical analysis
Continuous variables were expressed as mean ± standard error.Student t-test was used to compare between mean values of continuous variables with parametric distributions.Mann-Whitney test was used to compare between mean values of continuous variables with nonparametric distributions.Fisher's exact test was used to compare between incidence and prevalence of categorical variables.Two-tailed p<0.05 was considered statistically significant.Version 17.0 of the SPSS statistical package was used for all statistical analyses (SPSS Inc., Chicago, IL, USA).

Results
Reports concerning 113 patients were reviewed.The study group included 27 patients with Takotsubo cardiomyopathy-associated Torsades de Pointes.The control group included 86 patients with Takotsubo cardiomyopathy-associated QT interval prolongation.Mean age of all patients was 65.1±1.2years.Most patients were females (n=102; 90.3%).All patients with Takotsubo cardiomyopathy-associated Torsades de Pointes survived, although one patient died of other complications more than two months following Torsades de Pointes (Hirose et al., 2008).Although most patients were females, the prevalence of male gender was significantly higher among patients with Takotsubo cardiomyopathy-associated Torsades de Pointes relative to patients with Takotsubo cardiomyopathy-associated QT interval prolongation (odds ratio 4.6; 95%Cofidence Interval 1.3-16.7).Mean maximal QTc interval was significantly higher among patients with Takotsubo cardiomyopathy-associated Torsades de Pointes relative to patients with Takotsubo cardiomyopathy-associated QT interval prolongation even prior to Torsades de Pointes appearance.There were no statistical differences between patients with Takotsubo cardiomyopathy-associated Torsades de Pointes and patients with Takotsubo cardiomyopathy-associated QT interval prolongation in terms of mean age, mean lowest ejection fraction, and mean peak Troponin levels (

Discussion
We have studied reports concerning patients with Takotsubo cardiomyopathy-associated QT interval prolongation and reports concerning patients with Takotsubo cardiomyopathyassociated Torsades de Pointes in order to characterize the clinical circumstances leading to Torsades de Pointes in patients with Takotsubo cardiomyopathy.We have done this before in 2009 (Samuelov-Kinori et al., 2009), but reports concerning patients with Takotsubo cardiomyopathy-associated Torsades de Pointes have almost doubled since then, and it is time to examine if our previous conclusions are still true in face of these new publications.Moreover, we have studied if Torsades de Pointes has been possibly preventable in patients with Takotsubo cardiomyopathy-associated Torsades de Pointes.Consistent with our previous observation (Samuelov-Kinori et al., 2009), male patients with Takotsubo cardiomyopathy-associated QT interval prolongation are at higher risk for Torsades de Pointes compared with female patients with Takotsubo cardiomyopathyassociated QT interval prolongation, although most patients with Takotsubo cardiomyopathy are females (Bybee & Prasad, 2008).This finding is coherent with a recognized paradox which is still unexplained: women have longer QT interval compared with men but lower incidence of sudden death (Larsen & Kadish, 1998).
In our previous review there has been a trend in the mean maximal QT interval being longer among patients with Takotsubo cardiomyopathy-associated Torsades de Pointes relative to patients with Takotsubo cardiomyopathy-associated QT interval prolongation, although most of the times QT interval prolongation has been noticed only following Torsades de Pointes appearance (Samuelov-Kinori et al., 2009).In face of the new reports concerning patients with Takotsubo cardiomyopathy-associated QT interval prolongation, the trend has become statistically significant: mean maximal QTc interval is significantly higher among

Limitations
Our study is based on a small number of published case reports.Accordingly, the abovementioned findings are suggestive rather than conclusive.We assume that there are more incidents of Torsades de Pointes in patients with Takotsubo cardiomyopathy that have not been published; for example, when physicians are reluctant to report their deceased patients.Indeed, in all the above-mentioned reports patients have survived Torsades de Pointes.Accordingly; we believe that our findings are more likely an underestimation of the true prevalence of the clinical circumstances that might lead to Torsades de Pointes in Takotsubo cardiomyopathy patients.

Clinical implications
This study defines four risk factors for Torsades de Pointes in patients with Takotsubo cardiomyopathy whom already have QT interval prolongation to begin with: male gander, marked QT interval prolongation, electrolyte depletion, and high degree atrioventricular block.Electrolyte depletion and high degree atrioventricular block are also last triggers for Torsades de Pointes in patients with Takotsubo cardiomyopathy.Accordingly, we believe electrolyte depletion and high degree atrioventricular block should be addressed as soon as possible in patients with Takotsubo cardiomyopathy in general, and in patients with marked QT interval prolongation in particular.

Conclusion
Men with Takotsubo cardiomyopathy-associated QT interval prolongation are at risk for Torsades de Pointes.Most patients with Takotsubo cardiomyopathy-associated Torsades de Pointes have risk factors for Torsades de Pointes other than female gender and left ventricular systolic dysfunction.Electrolyte depletion and high degree atrioventricular block are the most prevalent risk factors and last triggers for Torsades de Pointes.We wish to raise table1)., were the most prevalent risk factors for Torsades de Pointes and the last triggers for Torsades de Pointes.Four (14.8%) patients were also taking QT interval prolonging agents, although initiation of these agents was not the last trigger for Torsades de Pointes in any case (table2).Among patients with Takotsubo cardiomyopathy-associated Torsades de Pointes, the arrhythmia was diagnosed prior to Takotsubo cardiomyopathy most of the times (n=19; 70.4%), and in seven (25.9%) patients Torsades de Pointes was diagnosed upon admission to the emergency department.Nevertheless, Torsades de Pointes was possibly preventable in seven (25.9%) patients: In four patients with QT interval prolongation and apparent electrolyte depletion, hypokalemia and/or severe hypomagnesemia were corrected only following Torsades de Pointes appearance; in three patients with QT interval prolongation and high-degree atrioventricular block, pacemaker was implanted only following Torsades de Pointes appearance (table 2).In other patients, risk factors for Torsades de Pointes were diagnosed only following Torsades de Pointes appearance, or they were addressed too late.

Table 2 .
(Samuelov-Kinori et al., 2009)tients with Takotsubo cardiomyopathy-associated Torsades de Pointes 742 patients with Takotsubo cardiomyopathy-associated Torsades de Pointes relative to patients with Takotsubo cardiomyopathy-associated QT interval prolongation even prior to Torsades de Pointes appearance.This finding is of great clinical significance beyond statistical significance: mean maximal QTc interval is remarkably high (676.8msec) in patients with Takotsubo cardiomyopathy-associated Torsades de Pointes prior to Torsades de Pointes appearance -more than 120 msec higher relative to mean maximal QTc interval in patients with Takotsubo cardiomyopathy-associated QT interval prolongation without Torsades de Pointes appearance.This large difference in QT interval prolongation should urge physicians to start treatment and to take prevention measures as soon as possible prior to Torsades de Pointes appearance.As with our previous review(Samuelov-Kinori et al., 2009), most patients with Takotsubo cardiomyopathy-associated Torsades de Pointes have one or more risk factors for Torsades de Pointes other than female gender and left ventricular systolic dysfunction.Electrolyte depletion and high degree atrioventricular block are the most prevalent risk factors and last triggers for Torsades de Pointes.But this time we have also shown that Torsades de Pointes is possibly preventable in one quarter of the patients should electrolyte depletion and high degree atrioventricular block have been addressed earlier.