Reduced cardiac function is associated with cardiac injury and mortality risk in hospitalized COVID‐19 Patients

Abstract Background Cardiac injury is common in COVID‐19 patients and is associated with increased mortality. However, it remains unclear if reduced cardiac function is associated with cardiac injury, and additionally if mortality risk is increased among those with reduced cardiac function in COVID‐19 patients. Hypothesis The aim of this study was to assess cardiac function among COVID‐19 patients with and without biomarkers of cardiac injury and to determine the mortality risk associated with reduced cardiac function. Methods/Results This retrospective cohort study analyzed 143 consecutive COVID‐19 patients who had an echocardiogram during hospitalization between March 1, 2020 and May 5, 2020. The mean age was 67 ± 16 years. Cardiac troponin‐I was available in 131 patients and an increased value (>0.03 ng/dL) was found in 59 patients (45%). Reduced cardiac function, which included reduced left or right ventricular systolic function, was found in 40 patients (28%). Reduced cardiac function was found in 18% of patients without troponin‐I elevation, 42% with mild troponin increase (0.04‐5.00 ng/dL) and 67% with significant troponin increase (>5 ng/dL). Reduced cardiac function was also present in more than half of the patients on mechanical ventilation or those deceased. The in‐hospital mortality of this cohort was 28% (N = 40). Using logistic regression analysis, we found that reduced cardiac function was associated with increased mortality with adjusted odds ratio (95% confidence interval) of 2.65 (1.18 to 5.96). Conclusions Reduced cardiac function is highly prevalent among hospitalized COVID‐19 patients with biomarkers of myocardial injury and is independently associated with mortality.

Conclusions: Reduced cardiac function is highly prevalent among hospitalized COVID-19 patients with biomarkers of myocardial injury and is independently associated with mortality.

K E Y W O R D S
cardiac function, COVID-19, mortality, SARS-CoV-2, troponin 1 | INTRODUCTION COVID-19 is caused by infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). As of October 1, 2020, more than 34 million people worldwide became infected and over 1 million have died. The complex pathophysiology of COVID-19, which includes cell death due to intracellular multiplication of virus, hypercoagulability, and cytokine storm [1][2][3][4] contributes to a wide array of complications including acute respiratory distress syndrome, multi-organ failure, sepsis, disseminated intravascular coagulation, as well as complex cardiac injury. [5][6][7][8][9][10][11] Many cardiac features have been reported including high prevalence of cardiac troponin elevation due to myocarditis or acute coronary syndrome, pericardial effusion and cardiac thromboembolism. 9,[12][13][14][15][16][17] Troponin elevation in COVID-19 patients has been linked to adverse clinical outcomes. 12 Additionally, abnormal cardiac function by echocardiography has been reported, [18][19][20] and is associated with biomarkers of cardiac injury. 21,22 It remains largely unknown if reduced cardiac function is associated with biomarkers of cardiac injury and with adverse clinical outcomes. In this retrospective study we aim to examine cardiac function by echocardiography in hospitalized patients with laboratory-confirmed SARS-CoV-2 infection, with and without cardiac injury, and associations with mortality.

| METHODS
We performed a cohort study involving consecutive, laboratory confirmed COVID- 19  troponin-I values were reported the peak value was selected for the analysis. The upper normal limit for cardiac troponin-I was >0.03 ng/ mL in our institution (Abbott Architect system for STAT troponin-I, analytic sensitivity ≤0.01 ng/mL). Troponin-I cutoffs of 0.04 to 5.00, and > 5.00 were also used for evaluation.

| Statistical analysis
We examined associations between a diagnosis of abnormal cardiac function, clinical laboratory results and all-cause mortality in the context of COVID-19. Laboratory values, events, and diagnostic counts are presented as mean (SD), median (interquartile range) or frequency (percent) as appropriate for data type and distribution. Stratified analysis of patients with a past medical history of CAD (defined as history of myocardial infarction, percutaneous intervention with coronary stenting, or coronary artery bypass grafting) was conducted using χ 2 tests. We present odds ratios of reduced cardiac function and significant laboratory findings (with 95% confidence limits) associated with all-cause in-hospital mortality from unadjusted logistic regression models as well as adjusted models (including age, gender, and history with cardiac tamponade physiology seen in 5 cases (3%), which required pericardiocentesis. Examples of cases and echocardiographic findings are shown in the accompanying Figure 1.
We compared the prevalence of having reduced cardiac function in subgroups (Table 2). There was a graded increase in the prevalence of reduced cardiac function of 18%, 42%, and 67% ( Figure 2) for patients with no troponin elevation, mild troponin increase (0.04-5.00 ng/dL) and significant troponin increase (>5 ng/dL), respectively. Reduced LV systolic function was more common in patients with CAD than those without (35% vs 19%, P = .041). The difference was not present for RV dysfunction (P = .23).
The most common cardiovascular outcome included congestive heart failure (16%, N = 23), new onset of arrhythmia including atrial fibrillation and ventricular tachycardia (10%, N = 15), pulmonary embolism (8%, N = 11), acute coronary syndrome (5%, N = 7) and tamponade with subsequent emergent pericardiocentesis (3%, N = 5) ( Myocardial injury marked by the increase in troponin has been observed in hospitalized COVID-19 patients. 5,[9][10][11][12]14 The reported prevalence ranged from 12% to 28%. Those with elevated troponin are more likely to be older and to have cardiovascular comorbidities such as hypertension, diabetes, and coronary artery disease. 12 In a recent series from Wuhan China, elevated troponin was reported to be associated with significant adverse clinical outcomes including arrhythmia, the need of mechanical ventilation and death. 12 In recent reports based on the US experience, abnormal cardiac function was linked to troponin elevation. 21,22 In this study we found that reduced cardiac function is more than twice as prevalent when troponin elevation is present compared to those with normal troponin level. More importantly, we demonstrated that reduced left or right ventricular function is independently associated with mortality in hospitalized COVID-19 patients even after adjusting for existing coronary artery disease. The prevalence of troponin elevation is 45% in our cohort, which is higher than previous reports. 5,[9][10][11][12]14 However, our cohort consisted of a larger number of patients with CAD among whom troponin elevation is present (more than 60%), which is similar to the reported 55% in a cohort from Wuhan, China, 12 suggesting that patients with underlying CAD are susceptible to myocardial injury likely due to ischemia infection, may also mediate cardiac injury. In post mortem biopsies, pathological findings include interstitial inflammatory infiltration and myocyte necrosis as well as micro-thrombosis and vascular inflammation. 26 In severe COVID-19 cases, hypoxia and respiratory failure are common which often result in intubation and mechanical respiratory In conclusion, reduced cardiac function is common among hospitalized COVID-19 patients with biomarkers of cardiac injury, those on mechanical respiratory support, and those deceased. It is also an independent predictor of mortality in addition to older age, elevated troponin and increased D-dimer.

ACKNOWLEDGMENTS
We sincerely thank all the sonographers and cardiac imaging nurses for their dedication and heroism.

DATA AVAILABILITY STATEMENT
Due to the nature of this research and our inability to completely remove personally identifiable information in compliance with HIPAA regulations, data presented here are not available for distribution to the public.