Point of care echocardiography and lung ultrasound in critically ill patients with COVID-19

Patients with COVID-19 frequently present with fever and cough but a loss of taste and smell has also been reported [11, 12]. Of COVID-19 positive cases 81% lead to a mild disease, whilst 14% of diagnosed patients become severely ill (requiring hospitalization) and 5% become critically ill [11]. Patients presenting with hypoxia, Glasgow coma scale <15, and a breathing rate >30/min, have a higher mortality rate and should be admitted to hospital [13]. Patients who become critically ill due to acute respiratory failure or acute respiratory distress syndrome (ARDS) need to be treated in an intensive care unit (ICU). Furthermore, bacterial superinfection, cardiac complications, such as arrhythmia or acute myocardial injury (AMI) and iatrogenic complications (pneumothorax), also have to be combatted in an intensive care setting [11, 14‐16].

Diagnosing COVID-19 can be challenging. At the beginning of the pandemic, the real-time polymerase chain reaction (RT-PCR) had not been validated. Therefore, false negatives occurred due to inadequate diagnostic timing [17]. Overall, combining molecular diagnostics with thorough clinical examinations and imaging can reduce the risk of false negatives, and allow efficient management in the context of this pandemic [18].

There are important factors to be considered with respect to the presets and settings of devices used for lung ultrasound (LUS). In order to be able to detect ultrasound artifacts associated with COVID-19 pneumonia, a specific lung preset with a low mechanical index, single-focal point modality and no harmonic imaging or any other cosmetic filters should be chosen [27]. The focal zone should be placed at the pleural line.

In COVID-19 patients, a linear transducer should be used for scanning the anterior and posterior regions of the chest (at a depth setting of 6–8 cm). Alternatively, a convex transducer can be used, and is particularly useful for scanning obese patients and the lateral or posterior regions of supine patients [23].

Typical LUS findings in patients with COVID-19 pneumonia are bilateral areas of fragmentation and irregularities of the pleural line, sometimes with massive reverberation artifacts (recently named light beam appearance), small subpleural consolidations, and a reduction or absence of pleural sliding [24]. Reverberation artifacts arising from a fragmented pleural line are indicative of diffuse alveolar damage, diffuse parenchymatous lung diseases, or inflammatory diseases such as COVID-19 [28]. As the disease progresses, consolidations can be seen (Fig. 1). Small consolidations and subpleural consolidations are typical, yet not specific to COVID-19 [26]. As seen on ultrasound images, consolidations can be described as hypoechoic areas with small hyperechoic structures within (bronchograms), in combination with a tissue-like appearance or hepatization of the lungs [31, 32]. If there is a consolidation of an entire lobe, the borders will be well defined, whereas in smaller consolidations, the deeper borders appear irregular, described as the shred sign [33]. As lung ultrasound may help to predict the clinical course and outcomes of patients with COVID-19, and for follow-up purposes, the authors recommend a quantification of reverberation artifacts (Fig. 2), and suspected areas of consolidation (Figs. 1 and 3; Table 1; [24, 34]).

Dynamic air bronchograms (hyperechoic artifacts moving with the respiratory cycle) are observed within areas of consolidation. Fluid bronchograms (bronchi filled with anechoic fluid within areas of consolidation) present frequently as well. Consolidations can involve entire lobes or whole lungs [24, 35]. Doppler can be used to identify vascularization in consolidations, which is an additional indicator of pneumonia: a tree-like appearance of vessels in color Doppler and a predominately triphasic flow signal in pulsed wave Doppler, can be used to identify pulmonary arteries in consolidations. Monophasic blood flow is found in bronchial arteries. This can be help detect complications such as necrosis, by visualizing hypoechoic areas with no blood flow [36, 37].

Overall, pleural effusions in the context of bacterial superinfections tend to be small and localized in patients with COVID-19 [28]. If large pleural effusions are present, other differential diagnoses, such as heart failure (in particular right ventricular failure), kidney failure, and liver cirrhosis, have to be considered (Fig. 4b; [38]).

A pneumonia diagnosis, however, should never be made solely by imaging, but by interpreting the overall clinical status and the history of the patient, as well as laboratory parameters [35].

As COVID-19 pneumonia is characterized by a patchy distribution of affected lung parenchyma, areas with a normal appearance in LUS are seen next to pathological areas (Figs. 5a, 6 and 7; [24]). Due to the patchy distribution, a thorough examination is recommended, yet there is no validated scanning scheme [23]. We recommend a 12-zone scanning protocol, as it is, in our experience, simple to apply, reproducible, and easy to learn (Fig. 8). Beginning with zone one with a longitudinal view, followed by a transverse view, a scan of all intercostal spaces should be performed. Whenever possible, scanning of the posterior regions should be included for patients in a supine position, although this may not always be feasible. In this case, it is recommended to consider scanning the posterior regions during prone positioning [23].

POCUS has been demonstrated to be a valuable tool to diagnose complications of COVID-19, such as pneumothorax or pleural effusions, or to confirm the suspected complications such as bacterial superinfections (Fig. 4a, c).

Pleural effusions are reportedly rare in COVID-19 pneumonia, and thus indicate a bacterial superinfection or heart failure [26]. To detect pleural effusions in supine patients, e.g. in an ICU, we recommend scanning the posterior regions with a convex or cardiac transducer as posterior as possible (Figs. 9 and 10). In an upright patient, the posterior regions—zone six (Fig. 8)—can be visualized with a convex or cardiac transducer to identify pleural effusions. In large pleural effusions, compression atelectasis can be visualized. Likewise, as previously mentioned, differential diagnoses of pleural effusions have to be considered [39]. The suspicion of heart failure is further supported by the presence of reverberation artifacts, which arise from a non-fragmented pleural line (B-lines). These B‑lines arise from the smooth pleural line in a bilateral homogenous way, not fading until the end of the screen, moving with pleural sliding and erasing other artifacts (A-lines). The presence and quantity of B‑lines correlate with the severity of clinical presentation of patients [24, 35, 40].

Patients who are critically ill often need to undergo procedures, for many of which a pneumothorax is a potential complication [15, 16]. In COVID-19, there are also cases with spontaneous pneumothorax described, making this entity an important differential diagnosis of acute respiratory distress [41]. In ultrasound, an easy protocol for detecting a pneumothorax can be applied [40]. Scanning should start at the highest point of the thorax, identifying whether lung sliding is present. Lung sliding corresponds to the motion of the parietal and visceral pleura. In a pneumothorax, there is gas between the parietal and visceral pleura, and pleural sliding is lost. The presence of pleural sliding rules out a pneumothorax [32]. In cases where there is an absence of lung sliding but a presence of vertical reverberation artifacts (B-lines, comet tails), a pneumothorax can be ruled out as well, as there has to be a connection between the parietal and visceral pleura to create such artifacts [32]. The presence of a lung pulse (a movement of the pleural line which is synchronous with the cardiac rhythm) rules out pneumothorax as well [42]. The most reliable rule in diagnosing a pneumothorax is the so-called lung point sign—the transition zone between the pneumothorax and the aerated lung (sliding on one side of the screen lung can be identified, whilst on the other, no lung sliding is present, video 2—lung point sign) [40]. It is recommended to use a point of care approach, performing a scan of the most anterior region (mostly in between zones 1 and 2) in a supine positioned patient [40].

Patients with COVID-19 pneumonia can develop myocardial injury or even fulminant myocarditis [44, 45]. Echo findings in myocarditis include chamber size alterations, increasing wall thickness, global or regional ventricular dysfunction, and diastolic dysfunction [46]. In pericarditis, a pericardial effusion can be detected [47]. It can mimic ischemic cardiomyopathy [48]. In fulminant myocarditis, a nondilated, thickened and hypocontractile left ventricle is seen, as the inflammatory response results in interstitial edema and loss of ventricular contractility (video 4—myocardial edema of the left ventricle (LV) in myocarditis) [49].

For detecting myocarditis in an intensive care setting, a subcostal 4‑ChV should initially be chosen (Fig. 11). In a subcostal approach, pericardial effusion can also be visualized (Fig. 14). An assessment of the global, left and right ventricular function is also possible. An alternative view to detect pericardial effusion and reduced global ventricular function is the 4‑ChV. Regional wall motion abnormalities (WMA) should be assessed in a comprehensive examination [43].

In our experience, hyperdynamic right ventricular function (RVF) and left ventricular function (LVF) are frequently seen in combination with mildly elevated systolic pulmonary artery pressures (sPAP), remaining after intensive care treatment in patients, with or without prior cardiovascular disease such as hypertension; however, a normalization can be seen after a few months in a majority of patients.

In strain imaging, there can be regional and global changes, which persist over the course of several months. The implications for treatment over a longer period and the clinical relevance in the follow-up have not yet been established [50].

Hypercoagulopathy is described in COVID-19 patients, and pulmonary embolism is a life-threatening complication thereof. Therefore, it is crucial to recognize the echocardiographic features of a hemodynamically relevant pulmonary embolism, to be able to initiate treatment [28, 38, 51].

In an acute, hemodynamically relevant pulmonary embolism (PE), the main features in echocardiography are a dilated right ventricle (RV), with the basal diameter >41 mm (as seen in Fig. 15a), reduced RVF, especially at the basal parts, S’ <9.5 cm/s, tricuspid annular plane systolic excursion (TAPSE) <17 mm, sometimes with a hyperdynamic apical region (McConnell sign). The sPAP is normal, low or mildly elevated [52]. A 60/60 sign which describes an sPAP of more than 30 mm Hg but less than 60 mm Hg, and a right ventricular outflow tract acceleration time of less than 60 ms, can be present [53]. The inferior vena cava (IVC) can be dilated with reduced, or even without, respiratory collapsibility. In a parasternal short axis view, a D-shape of the left ventricle can be visualized, and severe tricuspid regurgitation (TR) can be present (Fig. 15b; [54, 55]). None of these echocardiographic findings are specific for acute PE, but in the setting of COVID-19, in particular in an intensive care setting, it is reported that up to one third of patients do develop PE [56].

On the ICU, the optimal view to visualize the right heart is a subcostal 4‑ChV, permitting visual assessment of RVF and size. Color Doppler of the region of the tricuspid valve can provide information as to whether severe TR is present (Fig. 15b). Additional views, especially in follow up examinations, include apical and parasternal views, as seen in Figs. 12 and 13. SPAP should be measured with continuous wave Doppler across the tricuspid valve, according to current guidelines [57].

Acute heart failure can be the primary symptom of COVID-19 infection with 50% of COVID-19 patients having no prior cardiovascular diseases [44]. Severe inflammation increases the risk of acute myocardial infarction [58].

Detecting regional WMA in ICU patients can be challenging. Several views should be obtained. In an apical 4‑ChV, the apical regions can be visualized, and in the PLAX, the posterolateral and the anteroseptal wall of the LV can be seen. For evaluation of RV dysfunction and larger LV infarcts, a subcostal 4‑ChV is optimal [43, 57]. For an evaluation of regional WMA and RVF, a comprehensive transthoracic examination is required [43].