The application value of lung ultrasound scoring in assessing disease severity: Evaluation of small-scale outbreaks of COVID-19

PURPOSE

This study explored the use of transthoracic lung ultrasound for evaluating COVID-19 patients, compared it with computed tomography (CT), and examined its effectiveness using 8 and 12 lung regions.

METHODS

A total of 100 patients with COVID-19 and 40 healthy volunteers were assessed using 12 regions (bilateral upper/lower regions of the anterior/lateral/posterior chest) and simplified 8 zones (bilateral upper/lower regions of the anterior/lateral chest) transthoracic lung ultrasound. The relationships between ultrasound, CT, and clinical indicators were analyzed to evaluate the diagnostic value of ultrasound scores in COVID-19.

RESULTS

Increased disease severity correlated with increased 8- and 12-zone ultrasound and CT scores (all p < 0.05). The modified 8-zone method strongly correlated with the 12-zone method (Pearson's r = 0.908, p < 0.05). The 8- and 12-zone methods correlated with CT scoring (correlation = 0.568 and 0.635, respectively; p < 0.05). The intragroup correlation coefficients of the 8-zone, 12-zone, and CT scoring methods were highly consistent (intragroup correlation coefficient = 0.718, p < 0.01). The 8-zone ultrasound score correlated negatively with oxygen saturation (rs  = 0.306, p < 0.05) and Ca (rs  = 0.224, p < 0.05) and positively with IL-6 (rs  = 0.0.335, p < 0.05), erythrocyte sedimentation rate (rs  = 0.327, p < 0.05), alanine aminotransferase (rs  = 0.230, p < 0.05), and aspartate aminotransferase (rs  = 0.251, p < 0.05). The 12-zone scoring method correlated negatively with oxygen saturation (rs  = 0.338, p < 0.05) and Ca (rs  = 0.245, p < 0.05) and positively with IL-6 (rs  = 0.354, p < 0.05) and erythrocyte sedimentation rate (rs  = 0.495, p < 0.05).

CONCLUSION

Lung ultrasound scores represent the clinical severity and have high clinical value for diagnosing COVID-19 pneumonia. The 8-zone scoring method can improve examination efficiency and reduce secondary injuries caused by patient movement.

Review of Machine Learning in Lung Ultrasound in COVID-19 Pandemic

Ultrasound imaging of the lung has played an important role in managing patients with COVID-19–associated pneumonia and acute respiratory distress syndrome (ARDS). During the COVID-19 pandemic, lung ultrasound (LUS) or point-of-care ultrasound (POCUS) has been a popular diagnostic tool due to its unique imaging capability and logistical advantages over chest X-ray and CT. Pneumonia/ARDS is associated with the sonographic appearances of pleural line irregularities and B-line artefacts, which are caused by interstitial thickening and inflammation, and increase in number with severity. Artificial intelligence (AI), particularly machine learning, is increasingly used as a critical tool that assists clinicians in LUS image reading and COVID-19 decision making. We conducted a systematic review from academic databases (PubMed and Google Scholar) and preprints on arXiv or TechRxiv of the state-of-the-art machine learning technologies for LUS images in COVID-19 diagnosis. Openly accessible LUS datasets are listed. Various machine learning architectures have been employed to evaluate LUS and showed high performance. This paper will summarize the current development of AI for COVID-19 management and the outlook for emerging trends of combining AI-based LUS with robotics, telehealth, and other techniques.

Modified lung ultrasound scoring system to evaluate the feasibility of pregnant women with COVID-19 pneumonia

OBJECTIVE

To investigate whether physicians with short-term training can use a modified lung ultrasound scoring system for coronavirus disease 2019 (COVID-19) pneumonia to assess lung damage in pregnant women.

METHODS

Sixteen consecutively hospitalized third-trimester pregnant women with pregnancy-induced hypertension, preeclampsia, rheumatoid arthritis or connective tissue disease were selected as the study subjects for the simulation of COVID-19 pneumonia. Two physicians (imaging and internal medicine) without ultrasonic experience performed lung examinations on pregnant women after six days of lung ultrasound training, and their consistency with examinations by the expert was assessed. In addition, 54 healthy third-trimester pregnant women and 54 healthy nonpregnant women of the same age who were continuously treated in the outpatient clinic of this hospital were selected for comparisons of abnormalities on lung ultrasound.

RESULTS

(1) Third trimester pregnant women with pregnancy-induced hypertension, preeclampsia, rheumatoid arthritis or connective tissue disease had the same lung ultrasound patterns as those associated with COVID-19 pneumonia. (2) There was no statistically significant difference between the scores of the two trained doctors and the expert when the modified ultrasound scoring system was used (p > .05). (3) The evaluations of the two trained doctors and the expert showed good consistency (kappa value = 0.833-0.957). (4) The incidence of abnormal ultrasound manifestations of the pleura and lung parenchyma was higher among healthy third-trimester pregnant women than among healthy women of the same age (p < .001).

CONCLUSIONS

After receiving short-term training, imaging and internal medicine physicians can use the modified lung ultrasound scoring system to evaluate pregnant women's pulmonary damage, but caution is needed to avoid false-positive results among pregnant women with suspected COVID-19 pneumonia.

The utility of lung ultrasound in COVID-19: A systematic scoping review

Introduction

Lung ultrasound (LUS) has an established evidence base and has proven useful in previous viral epidemics. An understanding of the utility of LUS in COVID-19 is crucial to determine its most suitable role based on local circumstances.

Method

Online databases, specialist websites and social media platforms were searched to identify studies that explore the utility of LUS in COVID-19. Case reports and recommendations were excluded.

Findings

In total, 33 studies were identified which represent a rapidly expanding evidence base for LUS in COVID-19. The quality of the included studies was relatively low; however, LUS certainly appears to be a highly sensitive and fairly specific test for COVID-19 in all ages and in pregnancy.

Discussion

There may be LUS findings and patterns that are relatively specific to COVID-19; however, specificity may also be influenced by factors such as disease severity, pre-existing lung disease, operator experience, disease prevalence and the reference standard.

Conclusion

LUS is almost certainly more sensitive than chest radiograph for COVID-19 and has several advantages over computed tomography and real-time polymerase chain reaction. High-quality research is needed into various aspects of LUS including: diagnostic accuracy in undifferentiated patients; triage and prognostication; monitoring progression and guiding interventions; the persistence of residual LUS findings; inter-observer agreement and the role of contrast-enhanced LUS.

Frequency of Abnormalities Detected by Point-of-Care Lung Ultrasound in Symptomatic COVID-19 Patients: Systematic Review and Meta-Analysis

The COVID-19 pandemic has resulted in significant morbidity, mortality, and strained healthcare systems worldwide. Thus, a search for modalities that can expedite and improve the diagnosis and management of this entity is underway. Recent data suggested the utility of lung ultrasound (LUS) in the diagnosis of COVID-19 by detecting an interstitial pattern (B-pattern). Hence, we aimed to pool the proportion of various reported lung abnormalities detected by LUS in symptomatic COVID-19 patients. We conducted a systematic review (PubMed, MEDLINE, and EMBASE until April 25, 2020) and a proportion meta-analysis. We included seven studies examining the role of LUS in 122 COVID-19 patients. The pooled proportion (PP) of B-pattern detected by lung ultrasound (US) was 0.97 (95% CI: 0.94–1.00 I² 0%, Q 4.6). The PP of finding pleural line abnormalities was 0.70 (95% CI: 0.13–1.00 I² 96%, Q 103.9), of pleural thickening was 0.54 (95% 0.11–0.95 I² 93%, Q 61.1), of subpleural or pulmonary consolidation was 0.39 (95% CI: 0.21–0.58 I² 72%, Q 17.8), and of pleural effusion was 0.14 (95% CI: 0.00–0.37 I² 93%, Q 27.3). Our meta-analysis revealed that almost all SARS-CoV-2–infected patients have abnormal lung US. The most common abnormality is interstitial involvement depicted as B-pattern. The finding from our review highlights the potential role of this modality in the triage, diagnosis, and follow-up of COVID-19 patients. A sizable diagnostic accuracy study comparing LUS, computed tomography scan, and COVID-19–specific tests is warranted to further test this finding and to delineate the diagnostic and prognostic yield of each of these modalities.