Focused echocardiography in COVID-19: An unusual case of acute right ventricular dysfunction

During the novel coronavirus disease 2019 (COVID-19) pandemic, the role of lung ultrasound has been recognised as a rapid and sensitive bedside test to diagnose SARS-CoV-2 pneumonia.¹ Secondary cardiac complications such as myocarditis have been reported in a subset of patients with COVID-19 infection and appear to be associated with poorer outcomes.² Focussed echocardiography is often used in combination with lung ultrasound, particularly when evaluating patients with undifferentiated hypotension or shock. Employing its use at the bedside in patients with COVID-19 pneumonia with haemodynamic instability may be helpful in detecting these cardiovascular manifestations early. Peng et al. have recently described some of the echocardiographic signs seen in patients with COVID-19 infection, which include hyperdynamic cardiac function, acute stress-induced cardiomyopathy, acute right ventricular (RV) dysfunction and diffuse myocardial inhibition.³

Emerging evidence has suggested that COVID-19 infection may promote a prothrombotic physiological state leading to an increase in venous thromboembolic events.⁴ Large pulmonary emboli (PE) frequently cause right heart dysfunction, with one study showing that emergency physicians can accurately detect this on echocardiography.⁵ Some of the key echocardiographic features of RV dysfunction include RV dilatation, flattening of the interventricular septal wall (“D-sign” on parasternal short-axis view) and impaired longitudinal RV systolic function denoted by reduced tricuspid annular plane systolic excursion (TAPSE).⁵

It has been proposed that acute right heart failure in COVID-19 may also be caused by increased pulmonary vascular resistance secondary to hypoxic vasoconstriction.³ Furthermore, microvascular thrombus formation has been reported in some studies and thought to be related to a cytokine storm and release of pro-inflammatory mediators, similar to that seen in acute respiratory distress syndrome.⁴

In the state of a pandemic where the risk of anchoring bias is high, it is particularly crucial to consider other differential diagnoses for acute respiratory symptoms. Point-of-care ultrasound (POCUS) can be effective in helping to discriminate between the important and life-threatening causes of acute dyspnoea,⁶ which is especially relevant in an undifferentiated Emergency Department (ED) population.

We present an atypical case of a patient presenting to the ED with acute respiratory failure and shock, displaying classic radiological appearances of COVID-19 pneumonia and right heart strain.

A 53-year-old female was brought to the ED by ambulance with altered mental status and shortness of breath. She had a one-week history of cough and general malaise prior to her presentation. Her past medical history included chronic obstructive pulmonary disease (COPD), a previous deep vein thrombosis (DVT) and she was PEG-fed following surgery for a head and neck cancer.

Pre-hospital vital signs were notable for:

Oxygen saturation of 70% on room air which improved to 93% on high-flow oxygen, respiratory rate of 30 per minute, heart rate of 110 beats per minute and an unrecordable blood pressure.

On arrival to the ED, the patient was hypoxic on 15 L of oxygen via a non-rebreathing mask. She had a tachycardia of 130 beats per minute and signs of shock, with a persistent hypotension (systolic blood pressure between 40 and 70 mmHg), weak peripheral pulses and prolonged capillary refill time.

After the initial resuscitation, a focused ultrasound scan was performed at the bedside to evaluate the cause of respiratory failure and haemodynamic compromise. Lung ultrasound (LUS) revealed pleural line irregularities, subpleural consolidations and B-lines in multiple lung zones with a patchy distribution bilaterally consistent with a viral pneumonia pattern (Figure 1).

On focussed echocardiography, the left ventricle (LV) was noted to be underfilled and partially collapsed due to pressure overload from enlarged right ventricle (Figure 2). This was particularly apparent on the parasternal short-axis view (Figure 3) demonstrating LV septal wall flattening (“D-sign”). On the apical four-chamber view (Figure 4), the patient was noted to have significant right ventricular (RV) dilatation (RV/LV ratio of >1:1) with a reduced TAPSE of 7.3 mm (Figure 5, left) – a marker of RV systolic impairment. A significant tricuspid regurgitation jet was also seen on colour flow Doppler (Figure 5, right) providing further evidence of increased right heart pressure.

Compression ultrasonography of the femoral veins was performed to assess for the presence of a proximal DVT. No evidence of DVT was seen down to the mid-thigh bilaterally; however, the popliteal and calf veins were not evaluated.

The above POCUS findings were suggestive of COVID-19 pneumonia with acute cor pulmonale, which raised suspicion for massive PE as the cause of shock. Thrombolysis was briefly considered; however, a decision was made to obtain a computed tomography pulmonary angiogram (CTPA) to confirm the suspected diagnosis. This confirmed the RV dilatation (Figure 6(c)) and typical features of COVID-19 pneumonia (Figure 6(a) and (b)); however, surprisingly no central or sub-segmental PE was found (Figure 6(d)). On review of the medical notes, the patient had a recent echocardiogram four months prior, which showed normal LV systolic function and a non-dilated RV with normal function.

Laboratory blood tests were notable for leukocytosis, raised C-reactive protein (218 mg/dL), hsTroponin-T (926 ng/L) and D-dimer (59,279 ng/mL), with acute kidney injury (stage 1). Her electrocardiogram (ECG) demonstrated sinus rhythm with dominant R waves in leads V1 and V2 with corresponding T-wave inversions, consistent with a right heart strain pattern (Figure 7).

The patient was assessed by the Intensive Care team, but due to her frailty and co-morbidities she was deemed unsuitable for escalated care. She was transferred to a medical ward for palliation and died later that day. The final clinical diagnosis given was COVID-19 pneumonia with possible myocarditis. Whilst a single reverse transcription-polymerase chain reaction (RT-PCR) nasopharyngeal swab did not detect SARS-CoV-2 RNA in this case, the clinical presentation along with characteristic CT changes informed the final diagnosis.

In our case of a patient presenting to the ED with hypoxia and shock, typical features of COVID-19 pneumonia were seen on LUS and CT. The addition of focused echocardiography allowed the team to better attribute the patient’s shock to acute RV failure, facilitating early specialty input and aiding decision-making regarding ongoing management strategies. Unusually, in this case, there was no demonstrable PE on CTPA. RT-PCR was also negative for SARS-CoV-2 RNA despite classic CT changes, though this is not surprising given the limited sensitivity of RT-PCR.⁷

It should be recognised that acute RV strain may be a complicating feature of COVID-19 infection and may not always be due to PE. Pulmonary hypertension due to hypoxic vasoconstriction has been proposed as a plausible physiological mechanism for acute RV dysfunction in patients with SARS-CoV-2 infection.³ An Italian prospective cohort study examining CT thorax findings in patients with COVID-19 showed a high prevalence of sub-segmental pulmonary vessel enlargement, which may be caused by pro-inflammatory factors.⁸ This might provide corroborating evidence for the inflammatory cascade and microthrombi leading to sub-segmental vasculature dysfunction and pulmonary hypertension. RV strain has also been associated with increased mortality in a recent observational study of 120 consecutive patients with COVID-19.⁹ Therefore, the consequence and prognostic relevance of this finding is important as such patients are likely to require a higher level of care.

Features of myocarditis on echocardiography are variable and non-specific; these include LV failure, RV failure, pericardial effusions and thickened ventricular walls.¹⁰ Isolated right heart failure in myocarditis has not been commonly described, though RV dysfunction has shown to be an independent predictor of adverse outcome.¹⁰ In this case, apart from RV strain, we did not see other typical features of myocarditis on the ED bedside scan; however, this does not exclude the diagnosis. Unfortunately, the patient did not survive to have comprehensive echocardiography and no post-mortem was conducted. It remains unclear whether she had myocarditis or whether her acute RV failure was due to alternative mechanisms, such as hypoxic vasoconstriction or exacerbated by COPD. Given the significantly raised troponin result, it can be inferred that a degree of myocardial injury had occurred. Acute myocardial injury from COVID-19 may be mediated by angiotensin converting enzyme-2 (ACE-2) or due to extensive cytokine release.¹¹

Whilst there may be a significant proportion of patients who develop thrombotic complications as a result of COVID-19 infection,⁴ our case highlights the pitfalls in making this assumption for all patients. Caution should be exercised when considering therapies such as thrombolysis in this cohort of patients on the basis of echocardiography findings alone. Performing compression sonography to evaluate for a proximal DVT can be helpful to aid the diagnosis.

POCUS may be a valuable and feasible bedside tool in facilitating a rapid diagnosis of COVID-19 pneumonia and some of its cardiac and thrombotic complications. Early detection of these findings may inform decision-making around further investigation, treatment strategies and guide disposition planning. It is also a useful adjunct to the clinical assessment to help distinguish other causes for acute dyspnoea and shock, which need to be considered in an undifferentiated cohort of patients, such as those in the ED. Finally, it should be noted that not all RV dysfunction in a shocked patient with COVID-19 is due to acute PE, as highlighted in our case. Acute RV failure is a poor prognostic marker in COVID-19 and may be caused by pulmonary hypertension due to hypoxic vasoconstriction, microthrombi or a combination of the two.