Takotsubo syndrome associated with cardioversion: a systematic review

Direct current cardioversion is a well-established and safe procedure to restore normal sinus rhythm for atrial and ventricular arrhythmias. Takotsubo syndrome has been rarely described with cardioversion. We reviewed the literature for descriptions of Takotsubo syndrome associated with the stress of cardioversion, to better understand its risks. We queried MEDLINE, EMBASE, Google Scholar and Cochrane for cases of Takotsubo syndrome secondary to synchronized cardioversion as defined by Mayo Clinic Diagnostic Criteria. We identified 11 cases of cardioversion-associated Takotsubo syndrome. Average age was 76 years (range: 61–87 years) and most (10 out of 11, 91%) were female. Diagnosis was made soon after cardioversion (median: 10 h, range: 0–48 h). Only 2 of 11 had ST elevations noted, while apical ballooning was noted in all cases. Pulmonary edema developed in 6 and cardiogenic shock developed in 5 patients. The median recovery time was 7 days (range: 3–11). Cardioversion-associated TS has an overall favorable outcome with complete recovery in most cases. A higher risk of this complication may exist for elderly females undergoing synchronized cardioversion.


Introduction
Direct current cardioversion is a well-established, safe, and effective procedure to restore normal sinus rhythm for both atrial and ventricular arrhythmias [1]. Although thromboembolism is the most feared complication, hypotension and Takotsubo syndrome (TS) have also been described in the literature [2].
TS is a severe transient left ventricular dysfunction with regional wall motion abnormalities that may present with electrocardiogram (ECG) changes and/or elevated biomarkers in the absence of significant obstructive coronary artery disease [2][3][4][5][6]. This complex disease has a predilection to occur in females who are postmenopausal with a mean age of 67 years [3]. TS is frequently triggered by emotional or physical stress that usually occurs 1-5 days prior to disease manifestation. Exogenous electrical stimuli (both accidental and therapeutic) have also been noted to precipitate this syndrome [7,8]. Sudden catecholamine surge resulting in direct myocardial toxicity is a postulated mechanism for development of TS, a hypothesis supported by demonstration of twofold to threefold higher levels of catecholamines in TS as compared to patients with ST elevation myocardial infarction [9], and by the reported sensitivity of postmenopausal females to catecholamine surges [10,11].
To better understand the association between direct current cardioversion and TS, we performed a systematic review of the literature for cases describing patients with TS following DC cardioversion and studied the patient demographics, procedural characteristics, and disease outcomes.

Open Access
International Journal of Arrhythmia

Search strategy
We performed a comprehensive database search on MEDLINE (via PubMed), EMBASE, Google Scholar and Cochrane database of systematic reviews for case reports, case series and observational studies describing cases of Takotsubo cardiomyopathy or syndrome associated with synchronized direct current cardioversion from inception until January 1, 2020. The search strategy was performed using a concise vocabulary as shown in Supplementary files 1 and 2.

Study selection and characteristics
Adult cases (> 18 years of age) which met Mayo Clinic Diagnostic Criteria for TS were included [6]. Cases with pharmacological cardioversion, unsynchronized cardioversion (defibrillation) and cases without cardioversion were excluded. Screening was performed by two independent authors (TAW, AA) with discrepancies resolved by a third author (UN). Additional hand search for references was also performed. After removal of duplicates, a total of 11 articles yielded 11 cases which were included in the systematic review. The included studies were all case reports [12][13][14][15][16][17][18][19][20][21][22]. Demographics, risk factors, clinical course, complications, and outcomes were extracted by four authors (TAW, AA, UN and AD). Analysis was performed via Microsoft Excel 2019 with continuous variables expressed as means with standard deviations or medians, where data were not normally distributed. Categorical variables were expressed as percentages. Data were analyzed for association between energy used in cardioversion and the time of onset of TS.
Complications were described in 10 of the 11 patients (91%). These included pulmonary edema in 6 of 11 (54%), with 5 meeting criteria for cardiogenic shock of which 4 required inotropic support. Two patients also were reported to require ventilatory support.
Median time to recovery was 7 days. Seven of 11 made a complete recovery, and mean follow-up ejection fraction was 54% (SD: 8%) at a median of 7 days (3-11 days). One patient died from sudden cardiac death after hospital discharge but before follow-up.

Discussion
In this systematic review of DC cardioversion-associated TS, we identified similar demographics to prior TS reviews, in which female predominance is common ranging from 87 to 89.8% (23,24), supporting the hypothesis of postmenopausal female sensitivity to catecholamine surge [10,11]. The mean age reported in the included studies was 76.4 years (SD: 8.78) which is slightly higher than previously reported registry-based study [23].
ST elevation and T-wave inversions were the most prevalent electrocardiogram abnormalities, similar to previous studies [23,24]. Interestingly we noted a slightly higher rate of left bundle branch block (18%), compared to 9% reported in a previous larger registry [25]. We also found profound systolic myocardial dysfunction in the included studies with an average ejection fraction of 25% (SD: 9.5%, compared to ejection fraction of 33-40% in previously reported cases [23,24,26]). We also found that our cohort had higher rates of both cardiogenic shock (45%) and pulmonary edema (50%) than is reported in larger series of TS, in which cardiogenic shock is seen in 9.3% [2,24] and pulmonary edema in only 25% [24].
Takotsubo in our cohort was identified after a median time of 10 h and median cardioversion energy of 150 (range: 100-200 Joules). We could identify no direct relation with energy administered to the time to development of Takotsubo syndrome. Pad placement could not be evaluated, given inconsistent reporting in the included studies. Excess epinephrine and norepinephrine release from the adrenal medulla and sympathetic nervous system is hypothesized to induce direct myocardial injury in TS [9,27,28]. The excess catecholamines are postulated to increase systemic vascular resistance and increase afterload on an already "stressed" myocardium. However, it remains unknown how electrical cardioversion leads to direct or indirect myocardial toxicity leading to the development of TS. We hypothesize the mechanism may be similar to previously reported cases of TS resulting from electroconvulsive therapy and lightening [7,8,29]. Electrical current induces excess autonomic and catecholamine surge which can result in myocardial stunning and coronary vasospasm via proteolysis of contractile apparatus or desensitization of these filaments due to excess intra-cellular calcium release [30]. In our population, direct current would have served as the electrical stimuli for myocardial contractile protein denaturation and saturation of the adrenergic receptors with catecholamines triggering TS. We postulate that the combination of direct myocardial injury from the electrical stimulus in combination with catecholamine surge may have resulted in hyperacute development of TS post-cardioversion.
Study Limitations: The sample size was a major limitation in analyzing our data. The authors were restricted to the limited information provided in the case descriptions. The review was limited to direct current cardioversion, and therefore, our findings cannot be generalized for other cardiac procedures or electricity-triggered TS. Intentional defibrillation for ventricular arrhythmias was excluded from the study to prevent skewing of data. Therefore, result from this review cannot be expanded to unsynchronized cardioversion. Given the limited followup, we were unable to determine if repeat cardioversions were attempted with the recurrence of Takotsubo syndrome. Moreover, it is possible that the sicker patients reported in our review predisposes to some degree of publication bias as subclinical TS may have not been diagnosed or reported post-cardioversion.
The findings in our review did not reveal any other predisposing factor for the development of TS apart from the age and female gender. The details of other precipitants for TS such as neurological and psychiatric disease elements were not reported in the included studies which may have played a confounding role.
We highlight the importance of closer inpatient and outpatient monitoring of the geriatric population after cardioversion especially during the first 24 h for early detection of TS and appropriate intervention. Patients  may also be counselled about this rare complication when obtaining procedural consent for direct current cardioversion.

Conclusion
Cardioversion-associated TS has an overall favorable outcome with complete recovery in most cases. Nonetheless, chest pain and heart failure symptoms may be a manifestation of post-cardioversion TS, especially in elderly females. A higher risk of this complication may exist for elderly females undergoing synchronized cardioversion.