Takotsubo Cardiomyopathy in COVID-19.

The coronavirus disease-2019 (COVID-19) is a global pandemic caused by the novel severe acute respiratory-syndrome coronavirus-2 that is resulting in substantial morbidity and mortality across the globe (1). Emerging data suggest that patients presenting with COVID-19 requiring hospitalization may have objective evidence of myocardial injury, which has been shown to be associated with increased risk of in-hospital morbidity and mortality (2). The underlying mechanisms of myocardial injury in COVID-19 are unclear, and most published series define myocardial injury only on the basis of elevated myocardial necrosis biomarkers (2). Takotsubo cardiomyopathy (TTC) is a syndrome that is predominantly seen in women (representing over 75% of cases) and is characterized by acute and transient regional left ventricular (LV) systolic dysfunction usually triggered by emotional distress, but has also been associated with respiratory failure and infections (3). Based on our ongoing clinical observations, we report the clinical characteristics and outcomes of patients presenting with confirmed COVID-19 and TTC based on transthoracic echocardiographic (TTE) features at The Mount Sinai Hospital in New York, New York. Myocardial injury was defined as any cardiac troponin I level ≥0.04 ng/ml. This study was approved by the Institutional Review Board of the Icahn School of Medicine at Mount Sinai.

Between March and April 2020, 118 consecutive laboratory-confirmed COVID-19 patients underwent a clinically indicated TTE, of which 5 (4.2%) had features compatible with TTC. The median age of patients with TTC was 66 years (interquartile range: 57 to 68 years) and all were men. At the time of hospital admission, 4 patients had shortness of breath while 1 had chest pain and was found to have an acute ischemic stroke. On the electrocardiogram taken closest to the TTE, 2 patients had diffuse ST-segment elevations, 1 had atrial fibrillation with rapid ventricular response, 1 had new deep diffuse T-wave inversion, and 1 had sinus tachycardia.

Compared with patients with non-TTC myocardial injury and those without myocardial injury, those with TTC had the highest peak levels of cardiac troponin I and creatine-kinase myocardial band (Table 1 ). Conversely, the peak levels of inflammatory and pro-thrombotic biomarkers including interleukin-6, ferritin, and d-dimer were the highest among patients with non-TTC myocardial injury (Table 1). All patients with TTC had LV dysfunction, and the median LV ejection fraction was lower among patients with TTC compared with those with other myocardial injury and those without myocardial injury (36.0% vs. 55.0% vs. 60.0%; p = 0.001). Among the 5 patients with TTC, 4 patients had regional wall abnormalities typical of TTC including circumferential hypokinesis or akinesis of the apical and mid wall segments, whereas 1 patient had circumferential hypokinesis of the basal walls (consistent with reversed TTC). None of the patients underwent cardiac catheterization to rule out coronary artery disease. Two patients had follow-up TTE after 7 days, documenting resolution of the wall motion abnormalities, and were discharged alive. One patient improved clinically and was discharged from the hospital, 1 patient died while on mechanical ventilatory support, and 1 patient died while on venovenous extracorporeal membrane oxygenation. Rates of all-cause death, acute respiratory distress syndrome, and acute kidney injury were significantly higher among those with TTC or other types of myocardial injury compared with those without myocardial injury (Table 1).

Table 1

Clinical and Echocardiographic Characteristics of Patients With COVID-19 and With Versus Without Takotsubo Cardiomyopathy

	Takotsubo Cardiomyopathy (n = 5)	Other Myocardial Injury (n = 69)	No Myocardial Injury (n = 43)	p Value
Age, yrs	57 (39–65)	59 (49–71)	57 (39–65)	0.08
Male	5 (100.0)	50 (71.4)	26 (60.5)	0.14
Time from symptoms onset	6.5 (1.0–8.0)	4.0 (2.0–7.0)	7.0 (3.0–7.0)	0.48
Laboratory characteristics
Troponin I peak, ng/ml	11.40 (0.55–12.55)	0.67 (0.09–2.33)	0.01 (0.00–0.02)	<0.0001
CK-MB peak, ng/ml	26.9 (23.9–101.6)	14.3 (3.5–23.8)	0.3 (0.0–1.2)	0.001
BNP peak, pg/ml	153 (60–599)	197 (33–828)	18 (10–60)	<0.0001
C-reactive protein peak, mg/dl	207 (162–277)	247 (154–326)	181 (39–302)	0.03
Interleukin-6 peak, pg/ml	56 (25–436)	172 (89–446)	56 (29–147)	0.006
Ferritin peak, ng/ml	1,946 (868–18,124)	2,663 (1,186–4,996)	903 (159–2,515)	<0.0001
D-dimer peak, μg/ml	1.8 (1.3–11.5)	3.6 (1.1–13.0)	1.5 (0.5–4.3)	0.04
Echocardiographic data
Ejection fraction, %	36 (35–37)	55 (41–65)	60 (57–65)	0.001
LVEDV, ml	118 (113–153)	109 (83–136)	101 (83–124)	0.13
LVESV, ml	71 (69–74)	46 (30–73)	40 (30–54)	0.01
Wall motion abnormalities	5 (100.0)	10 (14.3)	0 (0.0)	<0.0001
Right ventricular dysfunction	1 (20.0)	27 (38.6)	6 (14.0)	0.02
Pericardial effusion	1 (20.0)	9 (12.9)	1 (2.3)	0.12
In-hospital outcomes
Death	2 (40.0)	21 (30.0)	1 (2.3)	0.001
Intensive care unit admission	3 (60.0)	41 (58.6)	20 (46.5)	0.44
Acute respiratory distress syndrome	4 (80.0)	36 (51.4)	11 (25.6)	0.006
Acute kidney injury	4 (80.0)	45 (64.3)	10 (23.8)	<0.0001

Values are median (interquartile range) or n (%).

BNP = brain natriuretic peptide; CK-MB = creatine kinase-myocardial band; COVID-19 = coronavirus disease-2019; LVEDV = left ventricular end-diastolic volume; LVESV = left ventricular end-systolic volume.

This series suggests that TTC should be considered in the differential diagnosis across the spectrum of myocardial injury in patients presenting with COVID-19. Interestingly, in this series, all patients were men consistent with the sex distribution more frequently associated with complications from COVID-19 (1). The mechanisms of myocardial injury in patients with COVID-19 remain poorly understood, and proposed mechanisms include cytokine-mediated damage, oxygen supply-demand imbalance, microvascular thrombi formation, and direct viral invasion of the myocardium (2). TTC could develop from catecholamine-induced microvascular dysfunction secondary to the metabolic, inflammatory, and emotional distress associated with COVID-19. Patients with TTC exhibited different laboratory characteristics compared with those with non-TTC myocardial injury, including higher levels of myocardial necrosis biomarkers but lower levels of inflammatory and pro-thrombotic biomarkers, suggesting that TTC may constitute a differential phenotype of myocardial injury in COVID-19 as reported in other cohorts of critically ill patients (3). The differential diagnosis of myocardial injury with new LV dysfunction in patients with COVID-19 also includes acute coronary syndrome or myocarditis. However, performing a more extensive work-up in COVID-19 patients is challenging due to their clinical status and the need to limit exposure of health care personnel. Therefore, TTE evaluation may be useful in COVID-19 patients with evidence of myocardial injury to guide management and the need for further invasive work-up.