Diagnosis and monitoring of COVID-19 pneumonia in pregnant women: is lung ultrasound appropriate?

Abstract

In their recently published article1, Buonsenso et al. recommend the use of lung ultrasound (LUS) for monitoring pregnant women with confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. In all four cases they describe, the ultrasound features indicative of coronavirus disease 2019 (COVID-19) pneumonia were pleural line irregularities with vertical artifacts (B-lines), and in two of the cases, patchy areas of white lung. However, in our experience from our COVID-19 center, these findings are ‘imaging errors’ that have no diagnostic validity as they are non-specific, not quantifiable and subject to interoperator variability. B-lines are artifacts generated when the ultrasound beam crosses areas with different acoustic impedance, in this case the chest wall and lung air interface, which reduces its propagation speed2. An irregular pleural line with increased number of B-lines may be visible in several conditions, such as acute respiratory distress syndrome, heart failure, nephrotic syndrome, pre-eclampsia, bacterial pneumonia, minimal pleural effusion, hydropneumothorax, fibrosis, pulmonary contusion, exacerbations of chronic obstructive pulmonary disease and neoplastic lymphangitis2, 3 (Figure 1). In contrast, B-lines are absent on intraoperative LUS scans in patients with interstitial lung disease, being imaging errors arising from the difference in acoustic impedance between the superficial and deeper structures4. It should also be considered that this pattern, which the authors indicate to be suggestive of SARS-CoV-2 infection, could instead indicate a different viral pneumonia, such as influenza A5. The number of B-lines and presence of pleural line abnormalities are dependent on the type of probe used, the angle of the probe on the thoracic cage and the physician's experience. The set-up of the device, such as gain, presence or absence of harmonics, and electronic focusing of the image can also modify the appearance of these artifacts2, which is not mentioned in the article of Buonsenso et al.1. Due to lung air content and rib-cage hindrance, only about 70% of the pleural surface can be visualized by LUS, and therefore, pathological conditions can be examined only if they are adherent to this viewable area2. Typical features of COVID-19 on chest computed tomography (CT) include ground-glass opacity and consolidation, which are mainly distributed in the peripheral and posterior part of the lung6. However, these alterations are not always located in areas adherent to the pleural surface and/or accessible to LUS, therefore, there is a risk of missing deeper or hidden lesions and/or underestimating the actual extent of the disease. Due to lack of comparison between CT (which is the gold standard) and LUS findings, it is not possible to know the actual number and type of lung lesions present in the four patients described in the study of Buonsenso et al.1. Moreover, it is incorrect to refer to observation of ‘white lung’ on ultrasound as LUS can assess the pathology of only the part of the lung that is near the pleural surface and not what eventually affects the whole lung parenchyma. White lung appearance of the chest wall and lung interface may occur exclusively due to inappropriate set-up of the ultrasound machine (e.g. excessive total gain, lack of tissue harmonic, use of a too-low frequency)2. Therefore, we believe that LUS should not be considered an alternative to CT scan for assessment of COVID-19 pneumonia in pregnant women, especially considering the higher risk of operator exposure associated with this type of examination. In the presence of quantitative clinical variables indicative of worsening of the condition of a patient, such as low oxygen saturation and/or positive reverse transcription polymerase chain reaction result, chest CT remains the gold standard for assessment of the extent of COVID-19 pneumonia.