Focused Transesophageal Echocardiography in Critical Care: The COVID-19 Pandemic

Abstract

Background:

The use of transesophageal echocardiography (TEE) is controversial in patients with COVID-19. The aim of this case series was to demonstrate the usefulness of transesophageal echocardiography in acute cardiovascular care settings in patients with COVID-19 infection.

Materials and Methods:

We enrolled 13 patients with confirmed SARS-CoV-2 infection admitted to the critical care unit of our center from April 1, 2020, to July 30, 2020, in which transesophageal echocardiography was performed. TOE was performed by three cardiologists with training in echocardiography.

Results:

The main indication was suspected infective endocarditis in four cases, venovenous extracorporeal membrane oxygenation cannulation in four cases, suspected prosthetic mitral valve dysfunction in two patients, suspected pulmonary embolism in two patients, and acute right ventricular dysfunction and prone position ventilation in one patient. The final diagnosis was confirmed in 11 patients and discarded in 2 patients. None of the operators result infected.

Conclusions:

TOE is safe in the context of COVID-19 infection; it must be performed in well-selected cases and in a targeted manner.

INTRODUCTION

The hallmark of severe COVID-19 infection is the presence of acute respiratory distress syndrome due to a severe cytokine release syndrome leading to respiratory failure and hemodynamic collapse.[1] Underlying cardiovascular diseases (CVD) are some of the major risk factors associated with more severe COVID-19. In this disease, cardiovascular system is commonly affected by direct (acute coronary syndromes, arrhythmias, myocarditis, valvular heart diseases, or endocarditis) and indirect conditions (thrombosis, microvascular damage, right ventricular failure, and sepsis-induced myocardial dysfunction). The diagnosis workflow of cardiovascular complications is complex due to overlap between direct and indirect manifestations and their sudden clinical onset; hence, early recognition of CVD should be a priority in critical care settings.

The use of critical care ultrasound, particularly bedside transthoracic echocardiography (TTE), is the main tool for the approach of a critically ill patient, due to its fast application and safety profile. As a previously unseen clinical scenario in COVID-19, with an increased risk of viral spread, the American Society of Echocardiography (ASE)[2] has published some recommendations about TTE imaging acquiring technique protocol in order to maintain health-care personnel safety. Nonetheless, the performance of TTE is limited in some clinical conditions which can be surmounted by adding transesophageal echocardiography (TOE).[2,3,4] The implementation of TTE in this scenario should constitute a unique framework focused on the search of cardiovascular complications, in which anatomical and hemodynamic information is urgently needed that otherwise will not be available.

The aim of this case series was to demonstrate the usefulness of TOE in acute cardiovascular care in patients with COVID-19 infection.

MATERIALS AND METHODS

Study design and setting: Case series

Selection of participants

We enrolled 13 patients with confirmed SARS-CoV-2 infection admitted to the critical care unit of our center from April 1, 2020, to July 30, 2020, in which TOE was performed.

Interventions

TOE was performed by three cardiologists with training in echocardiography (EGC, DSLM, and ADM). The echocardiographic views and parameters were recorded and measured according to the Guidelines for Performing a Comprehensive Transesophageal Echocardiographic Examination from the ASE and the Society of Cardiovascular Anesthesiologists.[5] The following measures were applied to minimize the exposure of the staff to the infection: (1) all studies were performed during regular patient rounds, (2) the machine and sector probe were sanitized adequately after each evaluation, (3) all providers had adequate personal protective equipment, and (4) analysis and image measurements were performed outside the patient's room.

Proposal of transesophageal echocardiography technique

1. The probe is placed in a bag with an opening in the distal part

2. The probe is inserted into the back of the pharynx, with slight anteflexion

3. Patient's head is covered with an acrylic box or transparent plastic bags

4. At the end of the study, the probe is removed, and once it is left, the oral cavity is covered with plastic to limit exposure to aerosols.

With this technique, none of the operators were infected. The presence of infection was proven with two viral tests: (a) identification of viral nucleic acid by reverse transcription–polymerase chain reaction (nasal swab) and (b) immunoglobulin antibody testing; none of our operators had a positive result.

RESULTS

Characteristics of study subjects

In Table 1, demographic characteristics and clinical indications for the TOE are shown. The echocardiographic measurements of the 13 cases evaluated are described in Table 2. Patients are labeled as C1 to C13 in text and tables.

Table 1.

Clinical characteristics of the cases and indications for transthoracic echocardiography

Case	Age	History	Indication	IMV	PaFiO2	Sedation and analgesia	Complications
C1	72	Acute mitral regurgitation, VVI pacemaker	Suspected infective endocarditis	Si	140	1, 2	None
C2	39	None	VV ECMO cannulation	Si	79	1, 2, 3, 4	None
C3	45	None	Suspected infective endocarditis	No	245	1, 5	None
C4	54	None	Acute right ventricular dysfunction and prone position ventilation	Si	80	1, 2, 3, 4	None
C5	39	None	VV ECMO cannulation	Si	89	1, 2, 3, 4	None
C6	48	Prosthetic mitral valve	Suspected prosthetic mitral valve dysfunction	No	200	1, 5	None
C7	48	None	VV ECMO cannulation	Si	90	1, 2, 3	None
C8	50	Prosthetic mitral valve	Suspected prosthetic mitral valve dysfunction	No	170	1, 5	None
C9	48	None	VV ECMO cannulation	Si	210	1, 2, 3, 4	None
C10	77	End-stage renal disease in hemodialysis	Suspected infective endocarditis	No	210	1, 5	None
C11	55	Stress cardiomyopathy	Suspected infective endocarditis	Si	180	1	None
C12	55	None	Suspected pulmonary embolism	Si	120	1, 2, 3, 4	None
C13	61	None	Suspected pulmonary embolism	Si	150	1, 2, 3, 4	None

VV ECMO=Venovenous extracorporeal membrane oxygenation, IMV=Invasive mechanical ventilation, 1=Propofol, 2=Fentanyl, 3=Muscle relaxant, 4=Midazolam, 5=Lidocaine spray, VVI=Ventricular demand pacing

Table 2.

Echocardiographic findings

Case	LVEF (%)	Presence of thrombi	FS (%)	TPG	POF	VHD	Vegetations	Prosthetic valve	Final diagnosis
C1	60	No	33	30	No	Severe MR Mild TR	MV, TV, and pacemaker lead	No	Infective endocarditis
C2	60	No	45	25	No	Mild PR	No	No	VV ECMO cannulation
C3	70	No	40	30	Si	No	No	No	Low probability endocarditis
C4	60	No	30	49	No	No	No	No	Cor pulmonale
C5	60	No	40	30	No	No	No	No	VV ECMO cannulation
C6	55	LA and LAA	25	89	No	Severe TR Moderate PR Mild AR	No	PMV (biologic)	Prosthetic valve thrombosis
C7	60	No	10	41	Si	Mild TR	No	No	VV ECMO cannulation
C8	50	LA and LAA	33	85	Si	Severe TR Moderate PR Mild AR	No	PMV (mechanical)	Prosthetic valve thrombosis
C9	60	No	40	30	Si	Mild TR	No	No	VV ECMO cannulation
C10	55	No	40	35	Si	Severe MR Mild TR	MV and mitroaortic continuity	No	Infective endocarditis
C11	55	No	40	30	No	Mild MR Mild TR	No	No	Low probability endocarditis
C12	55	PA	30	58	No	Moderate TR	No	No	TEP
C13	61	RA and PA	30	45	SI	Moderate TR	No	No	TEP

LVEF=Left ventricular ejection fraction, FS=Fractional shortening, TPG=Tricuspid pressure gradient, POF=Patent oval foramen, VHD=Valvular heart disease, LA=Left atrium, LAA=Left atrial appendage, PA=Pulmonary artery, RA=Right atrium, MR=Mitral regurgitation, TR=Tricuspid regurgitation, PR=Pulmonary regurgitation, AR=Aortic regurgitation, MV=Mitral valve, TV=Tricuspid valve, PMV=Prosthetic mitral valve, VV ECMO=Venovenous extracorporeal membrane oxygenation, PE=pulmonary embolism.

Endocarditis

C1 [Figure 1a] had positive blood cultures with Escherichia coli with vegetations in native mitral and tricuspid valves, and both heart valves underwent bioprosthetic replacement. Surgical findings were multiple vegetations in both valves. Unfortunately, the patient had unfavorable clinical outcome and died 7 days after surgery. C10 [Figure 1b] had positive blood cultures with Staphylococcus epidermidis with vegetations in the aortic and mitral valves. Two patients (C3 and C11) with persistently positive blood cultures, fever, and limited acoustic window underwent TOE with suspicion of endocarditis, which was ruled out, and they progressed favorably with antibiotic treatment.

Figure 1.

(a) Mid-esophageal four-chamber view showing severe mitral regurgitation (white arrowhead). (b) Mid-esophageal commissural view showing vegetation in segments A2–A3 of the mitral valve (white arrowhead) (c) Mid-esophageal 90° view showing thrombi in the left atrial appendage and in a prosthetic mitral valve. (d) Mid-esophageal 140° view showing thrombi in a prosthetic mitral valve. (e and f) Upper esophageal 15°–30° view showing thrombi at the right pulmonary artery. (g) Mid-esophageal bicaval view showing the return (jugular) cannula at mid-right atrium (white arrowhead) and the flow with color Doppler (red arrowhead). (h) Mid-esophageal bicaval view showing the extraction (femoral) cannula at the inferior vena cava (white arrowhead) and the flow with color Doppler (red arrowhead). LA = Left atrium, LV = Left ventricle, RA = Right atrium, RV = Right ventricle, LAA = Left atrial appendage, MMA = Mid-ascending aorta, RPA = Right pulmonary artery, IVC = Inferior vena cava, T = Thrombi

Prosthetic valve dysfunction

Prosthetic valve thrombosis was suspected in C6 [Figure 1c] and C8 [Figure 1d] with an elevated mean gradient across a prosthetic mitral valve (24 mmHg and 30 mmHg, respectively) with thrombi arising from mechanical prosthetic valve. The patients needed emergent surgical valve replacement; prosthetic valve thrombosis was confirmed with abundant thrombus load in both cases.

Extracorporeal membrane oxygenation cannulation

In C2 [Figure 1g and h], C5, C7, and C9, the indication for TOE was to guide the cannulation of venovenous extracorporeal membrane oxygenation ECMO. Although cases of TTE-guided cannulation have been reported, based on our data and experience, TOE-guided cannulation is preferred in our center. Importantly, it is well known that the right cannulae positioning at the very start of ECMO circulation could limit complications, mainly recirculation, among others.

Pulmonary embolism

In C12 [Figure 1e] and C13 [Figure 1f], the indication for TOE was the presence of acute right ventricular dysfunction with high probability of pulmonary embolism; in both cases, the diagnosis was confirmed.

TEE during prone position ventilation

The indication of TOE during the prone position ventilation was only necessary in C4, in a patient with suspected pulmonary embolism (ruled out), with extreme obesity and poor acoustic window for TTE.

Complications

There were no documented complications associated with bleeding, perforation, or hemodynamic instability due to sedation. The studies carried out in patients without invasive mechanical ventilation (IMV) were performed with the support of an anesthesiologist (C3, C6, C8, and C10); none of the patients required IMV at the end of the study.

DISCUSSION

In this work, all patients had absolute justified indication, and we were able to demonstrate the usefulness of TOE to establish a proper diagnosis as well as to guide a procedure; furthermore, TOE helped in the modification of definite medical therapy including an entire switch of medical interventions previously settled, in order to achieve better outcomes for the patients.

The usefulness of TOE in an acute cardiac care setting has been settled previously.[5] The clinical scenarios in which TOE has shown to be a mandatory imaging technique should be (1) suboptimal or not diagnostic TTE imaging; (2) optimal imaging for some specific structures such as acute valvular regurgitation, acute prosthetic valve dysfunction, acute aortic syndromes, and acute chest trauma; intracardiac thrombi in arrhythmias (atrial fibrillation/flutter); and the identification of valvular vegetations when endocarditis is suspected.[3] This imaging technique can carry up an increased risk of viral transmission; thus, the pros and cons should be weighed in order to strictly justify this medical intervention. At present, no changes have been made for clinical recommendations for the use of TOE in COVID-19 patients; however, two further clinical scenarios have been considered: (1) TOE in prone position and (2) poor image quality attributable to hyperinflated lung during IMV.

In patients with COVID-19 infection, TOE has been considered a high-risk procedure for health-care personnel; hence, it should be performed only in irrefutable clinical grounds.[2] The available evidence, but mainly our results suggest the application of these safety measures, leads to no further increased risk of viral infection during the performance of TOE. The cornerstone recommendations are focused on decreasing the exposure time and conducting studies aimed to rule in or rule out the diagnostic suspicions. To increase safety standards in patients without IMV, we suggest the presence of the anesthesiologist aimed to provide appropriate sedation and analgesia in order to reduce viral spread, achieve a better patient comfort, and to limit the risk of orotracheal intubation. Finally, we must establish that the safety profile is a noteworthy concept, since none of the operators were infected with COVID-19 nor any of the patients had complications related to the procedure.

Does an improvement in overall clinical outcome can be achieved with TOE? This question remains unanswered at this moment, and undoubtedly, this must be an active research item in the short term.

Our limitations are the number of patients in the study, the need for specialized training to obtain the desired projections and images, and the knowledge and application of security measures to prevent contagion by COVID-19. Finally, this is a preliminary report and it should be validated in other centers.

CONCLUSIONS

TOE is a safe procedure that must be performed in well-selected cases and in a targeted manner to reduce the time of exposure to aerosols. In patients without IMV, it is better to be supported by an anesthesiologist in order to reduce the risk of complications.

Ethical clearance

The local institutional research and ethics committee of the three centers waived approval. Written informed consent patient information and images to be published were provided by the patient or a legally authorized representative.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.