Echocardiography findings in COVID‑19 patients admitted to ICUs

SARS-CoV-2 infection and COVID-19 disease have been linked to myocardial injury, heart failure, pulmonary artery thrombosis, vasoplegia, shock and may lead to increased mortality. 

The ECHO-COVID study aimed to recognise and describe left and right ventricular dysfunction in the initial echocardiographic evaluation after ICU admission in patients hospitalised for severe COVID-19. This retrospective longitudinal observational study recruited patients from 14 intensive care units of tertiary teaching hospitals in 8 countries. It included patients with confirmed SARS-CoV-2 infection who received at least one transthoracic (TTE) or transesophageal echocardiographic (TEE) examination as routine care during their ICU stay between 1 February 2020 and 30 June 2021, which spanned through the consecutive waves of COVID-19. 

The collected study data were filled under three major domains: 

1. patient characteristics, including co-morbidities
2. clinical data, including ventilation and hemodynamic data, at the time of the echocardiographic exam, and 
3. echocardiography data.

A population of 689 patients were recruited, 12 patients were excluded because of missing information, and the remaining 677 patients (336 during the first wave and 341 during the second/third waves) participated in the analysis. Most cases (81.4%, n = 551) received TTE (under mechanical ventilation for 342 of them), while 126 (18.6%) received TEE (125 under mechanical ventilation). Most subjects were over 60 years old, half of the population presented symptoms of SARS-CoV-2 8 days prior to ICU admission, and 69% (n = 467) were mechanically ventilated.

At the time of the initial echocardiographic exam, a mean tidal volume of 435 ± 80 ml and a mean plateau pressure of 24.7 ± 4.5 cmH2O were measured across 84% of available patient data. In addition, low PaO2/ FiO2 measurements were obtained from most patients. Regarding cardiovascular presentation, 41% of cases were on norepinephrine or in combination with another inotrope or vasopressor. Additionally, 19 patients received extracorporeal membrane oxygenation (ECMO) due to severe or refractory ARDS, with the majority (18 patients) being on venous-venous (VV) ECMO. Almost 92% of the subjects exhibited sinus rhythm. Four hundred and sixty-five patients had one or more echocardiographic identifiable pathologies, 234 (34.5%) presented visual LV and/or RV systolic dysfunction: 67 (9.9%) cases had biventricular dysfunction, 85 (12.5%) had isolated RV dysfunction, 82 (12.1%) had only LV dysfunction, and 441 (65.1%) patients displayed normal bi- ventricular function. 

Moderate or severe LV systolic dysfunction was recorded in 149 (22%) subjects. Among them, 45 (30.2%) had a history of cardiomyopathy, 50 were on previous beta-blocker medication, and 15 were treated with dobutamine. The visual assessment of LV systolic function was consistent with measured LVEF. There were no significant differences in LVEF between the age groups, even though 27.5% (n = 91) in the over-60 age group had LVEF < 50%. A small percentage of cases (10.5%) displayed visually dilated LV. Of 563 subjects with available data, 77 (13.7%) had segmental wall motion abnormality SWMA, and 46 had visually identifiable, impaired LV function. The mean LV E/A ratio was 1 ± 0.37 (n = 567).

Moderate to severe RV systolic dysfunction was found in 152 (22.5%) patients. The visual assessment of RV function was correlated with tricuspid annular plane systolic excursion (TAPSE) measurements. There were no discernible deviations in the means of TAPSE between the age groups, even though 22.1% (n = 62) in the > 60 age group had TAPSE ≤ 16 mm. About 41% (n = 277) exhibited visual moderately or severely dilated RV and higher RVEDA/LVEDA. In 521 assessed cases, the mean RVEDA/LVEDA was 0.67 ± 0.25, and 50% presented a dilated RV (RVEDA/LVEDA > 0.6). RV systolic dysfunction correlated positively to the severity of LV dysfunction (p < 0.001). TAPSE was matched with LVEF [r = 0.28 (95% CI 0.19, 0.36)], while mean LVEF was similar in different RV size groups.

Paradoxical septal motion (PSM) was found in almost one-fifth of patients, with an RVEDA/LVEDA ratio higher than in patients without PSM [difference = 0.28 (95% CI 0.23 to 0.33)]. Ninety out of 517 subjects displayed PSM and RVEDA/LVEDA > 0.6, indicating Acute Cor Pulmonale (ACP). The PaO2/ FiO2 were comparable in patients with and without PSM, 140 ± 72 vs 133 ± 67 mmHg, respectively [difference = 7.5 mmHg (95% CI − 5.8 to 20.7), and also similar in patients with and without ACP (138 ± 68 mmHg vs 131 ± 67 mmHg) [difference = 6.5 mmHg (95% CI – 9 to 21.9)]. There was also no difference in PaO2/ FiO2 regarding different RV size. However, patients with PSM or ACP were associated with higher incidence of pulmonary embolism (OR > 5) and mechanical ventilation (OR > 4). These patients also had higher PaCO2 levels, while plateau pressure did not differ. Pulmonary embolism was totally identified in 55 patients. 

Only a few patients presented severe valvular regurgitations, while aortic and mitral regurgitations were absent in most of them. Patent foramen ovale (PFO) was detected in 14 patients among the 256 in whom it was actively sought. Pericardial effusion (albeit with no case of tamponade) was found in 74/605 (12.2%) patients. The median ICU length of stay was 14 days, in-ICU mortality was 32.1%, and in-hospital mortality was 34.3%. ACP increased mortality twice-fold (OR = 2.02 [1.11, 3.66] for in-hospital, and OR = 2.01 [1.09, 3.69] for in-ICU cases. The probability of death increased by 6% for each year increase in age (OR = 1.06 [1.04, 1.08]) for both hospital and ICU mortalities. Visually identifiable LV systolic dysfunction was linked to in-hospital mortality (OR = 1.52 [1.04, 2.23]), but this correlation dissipated following age adjustments. There was no discernible link between visual RV dysfunction and mortality, irrespective of age. Sensitivity analyses which excluded the ECMO patients (n = 19) produced similar results and conclusions.

STUDY STRENGTHS & LIMITATIONS

Strengths

• Visual and quantitative assessments were unvarying and dependable. 
• Experts in CCE performed the echocardiographic and visual assessments. The main investigators are members of the European Society of Intensive Care Medicine (ESICM) ‘s echocardiography working group, further validating the results.
• This study used echocardiography to estimate cardiac status, which has superior specificity in diagnosing clinically significant heart dysfunction in differentiating the types of dysfunction and other important characteristics that may not be detected by using biomarkers. This contrasts with preceding studies in hospitalized COVID-19 patients that used cardiac biomarkers to prove that myocardial injury (as reflected by increased troponin level) is frequent and correlated with mortality. 
• It is one of the largest echocardiographic studies in COVID-19 ICU patients and provides crucial clinical and echo data in this subpopulation. 
• A large number of recruitment centres in different geographical locations provided a relevant depiction of COVID-19 patients with distinctive characteristics. 

Limitations:

• This is a retrospective observational study and is thus affected by the usual limitations of such studies.
• Echo exams were only attained as dictated by the clinical condition and not as a regular exam on every patient (a third of all COVID-19 cases per centre), producing selection bias and limitations in its deduction to other patients. 
• The echo exams were not standardized for an exclusive echo study; thus, missing data occurred. 
• Despite a statistical association, it failed to produce any conclusive connection between ACP and mortality. Additionally, it could not assess the effect of VV ECMO, as the number of patients was minimal. 
• It was not designed to investigate the pathophysiology or treatment of COVID-19. It did not consider any severity index (namely APACHE, SOFA, SAPS) even though all participants were treated in ICU.

TAKE-HOME MESSAGES

The first echocardiography performed after ICU admission is pathological in one-third of cases. Left ventricular systolic dysfunction is observed in 23% and appears similar to septic cardiomyopathy. Right ventricular systolic dysfunction is diagnosed in 22.5% of cases and appears to be related to pressure overload and not to the severity of hypoxemia, which differentiates COVID- from non-COVID-19 ARDS.

Acute cor pulmonale is diagnosed in 17% of patients and is linked to hypercapnia, pulmonary embolism and mechanical ventilation. Acute cor pulmonale and age are matched with in-ICU mortality. Only a few patients exhibited other significant cardiac abnormalities like valvular disorders or PFO.

This article review was prepared and submitted by Dimitrios Papadopoulos MD, MSc, PhD, Senior Consultant in Intensive Care Medicine, General Hospital of Larisa, Larisa, Greece, on behalf of the ESICM Journal Review Club.