TABLE 2.

Lung ultrasound findings in COVID‐19

Study	Design	N	COVID‐19 severity	B line n/N	Consolidations n/N	Sub‐pleural Lesions n/N	Micro Emboli n/N	Other	Sensitivity compared with other modality	Diagnostic Reference Standard	Comments
Y Lichter et al.	Retrospective study	120	75‐ mild 31‐moderate 14‐severe	NA	93/120	100/120	NA	Pleural effusion‐ 9	LUS cutoff of 18 (Sensitivity = 62%, specificity = 74%)	N/A	Base‐line LUS score strongly correlates with the eventual need for invasive mechanical ventilation and is a strong predictor of mortality
S Ottaviani et al.	Prospective study	21	NA	19	13	NA	NA	Median B score 6, C score 1.	Correlation coefficient of (r = .935) between LUS and HRCT findings	HRCT (correlation of LUS findings with HRCT)	LUS excellent correlation with lung involvement in HRCT, positive correlation with supplemental oxygen therapy.
Rojatti M et al	Retrospective study	41	All ICU cases	NA	NA	NA	NA	Mean LUS score = 11.	LUS and IL‐6 correlation(r = .52) LUS and oxygen correlation R = .3	LUS versus PaO2/FiO2 ratio and PaCO2	LUS positively correlated with IL‐6 and co2 levels, inverse with oxygen levels, and no correlation with respiratory system compliance.
Zhao et al	Prospective study	35	seven refractory ARDS, 28 non‐refractory ARDS	B line score 4 in refractory, 6.5 in non‐refractory	Mean consolidation score of 1 in refractory versus 0 in non‐refractory	NA	NA	NA	LUS cutoff of 32 points for differentiating refractory disease with specificity of 89.4% and a sensitivity of 57%	N/A	LUS score helpful in differentiating refractory group versus non‐refractory with cut‐off of 32
Bonadia N et al	Prospective	41	16/41 patients in ICU.	NA	NA	NA	NA	NA	NA	N/A	Patients who died had Lung score of 1.43 and discharged had score of 1, patients requiring ICU admission had median score of 1.36 compared to non‐requiring score of 1.
Zieleskiewicz L et al	Retrospective study	100	23/100	96/100	32/100	6/100	NA	NA	An LUS score > 23 predicted severe SARS‐CoV‐2 pneumonia diagnosed by chest CT scan with a Sp > 90% and a PPV of 70%	Chest CT	The LUS score was predictive of pneumonia severity as assessed by a chest CT scan and clinical features
Shumilov et al		18	NA	17/18	14/18	16/18	NA	NA	NA	chest X‐ray	LUS was especially useful to detect interstitial syndrome compared to CXR in COVID‐19 patients (17/18 vs 11/18; p < 0.02). LUS also detected lung consolidations very effectively (14/18 for LUS vs 7/18 cases for CXR; p < 0.02).
Gaspardone et al.	Prospective study	70	Group 1: mild (no ventilator support) 27 Group 2: severe (ventilator support) 43	LUS score: Anterior areas: mild 21% versus severe 36% (p = 0.21) Posterior areas: mild 48% versus severe 32% (p = 0.21) Other areas no statistically significant difference seen.	NA	NA	NA	NA	NA	RT‐PCR	Classified as LUS
Youssef et al.	Prospective study	75	PCR + (n = 3) 4% PCR –(n = 72) 96%	NA	NA	NA	NA	Lung Ultrasound normal in all patients.	Ultrasound Sensitive in symptomatic patients (no changes seen in pregnant asymptomatic patients) Not useful as a screening tool.	RT‐PCR	Pregnant women (median age 34, range, 24–48yrs) (median gestational age 38 weeks, range 25–40wks) Followed for median of 7 days (range 3–9 days)
Lu W et al.	Retrospective study	30	Severe : (> 19 points) Moderate (8‐18 points) Mild (1‐7 points)	27/30 (90%): B‐lines. [15/30: coalescent B‐lines 5/30 : widely spaced B‐lines (> 7 mm) 3/30 diffusely coalescent B‐lines]	6/30 (20%) pulmonary consolidation	NA	NA	3/30 (10%) pleural thickening, 1/30 (3.3%) minimal pleural effusion 1/30(3.3%): pneumothorax.	NA	CT chest	Distribution: 22/30 (73.3%): multiple distributions, 5/30 (16.7%): Focal distribution. 22/30(73.3%) bilateral involvement 5/30(1.6%) unilateral involvement Majority distribution: Sub‐pleural and peripheral zones, with the lower & dorsal regions.
Nouvenne A, et al.	Prospective study	26	Stable patients as critically ill patients and requiring ICU were excluded	Distinct B line 7 (27). Confluent B lines 17 (37).	Parenchymal consolidation 13 (50)	Sub‐pleural consolidation 17(73).		Bilateral involve 26(100). LUS 15+_5	NA	HRCT	LUS score was significantly correlated with CT visual scoring (r = .65, p < 0.001) and oxygen saturation in room air (r = –.66, p < 0.001).
Yasukawa K, et al.	Retrospective study	10	Mild to moderate cases. None required ventilator	Glass rocket 10(10) Septal rocket 2(10)	1(10)	5(10)		Birolleau Variant 5(10)	NA	N/A	LUS is more sensitive than CXR in detection of interstitial findings.
Li S et al	Retrospective study	91	Severe and critical	59/91 had scattered B lines 56/91 Confluent B lines	48/91	6/91 pleural thickening 39/91 had pleural effusion	NA	20/91 had pneumothorax	Not compared to other tests	N/A	Findings support the use of LUS for monitoring response to therapy in sever and critical COVID‐19
Pare et al	Retrospective cohort study	43 (27 positive Covid‐19)	Not Specified	All patients were tested for B lines 24/27	10/27	21/27	NA	NA	Compared to CXR, LUS sensitivity: (88.9%, 95% confidence interval (CI), 71.1‐97.0) CXR sensitivity: (51.9%, 95% CI, 34.0‐69.3; p = 0.013). LUS specificity: 56.3% (95% CI, 33.2–76.9) CXR specificity: 75.0% (95% CI, 50.0–90.3	RT‐PCR and CXR	LUS was statistically significant Higher sensitivity: p = 0.013 Lower Specificity: p = 0.453 LUS considered positive if have B‐lines
Mafort et al	Cross‐sectional study	409	All symptomatic without mentioning the severity	297/409 (72.6%) of participants had B‐lines > 2, 148/409 (36.2%) had coalescent B‐lines	33/409 (8.06%)	NA	NA	NA	Ultrasound has a sensitivity and specificity of 89% and 94%, respectively, for the identification of parenchymal consolidation	RT‐PCR	The aeration score differed significantly regarding the presence of cough (p = 0.002), fever (p = 0.001), and dyspnea (p < 0.0001). The finding of sub‐pleural consolidations in the LUS showed significant differences between participants with or without dyspnea (p < 0.0001)
											B‐lines are the most common ultrasound sign, sub‐pleural consolidations are those that most impact the respiratory condition
Narinx et al.	Retrospective study	93	Not specified	Did not detail LUS findings.	Did not detail LUS findings.	Did not detail LUS findings.	Did not detail LUS findings.	NA	Compared with RT‐PCR, POCUS lung demonstrated outstanding sensitivity and NPV (93.3% and 94.1% respectively) while showing poor values for specificity, PPV, and accuracy (21.3%, 19.2%, and 33.3% respectively).	RT‐PCR	NA
Smargiassi A et al.	Prospective study	38	19 in hospital and 19 isolated at home.	NA	NA	NA	NA	NA	NA	Chest CT (Plain CT chest, CTPA and HRCT)	NA
Alharthy A et al.	Prospective study	89	Severe	Separated B‐lines 67.4% Confluent B‐lines 78.6%	Consolidations 61.7% Lung parenchymal hepatization pattern (22.4%). A “starry sky” pattern of consolidation (bright infiltrates) 49.4%	Sub‐pleural consolidations (26.9%)	NA	Pleural line irregularities in > 6 lung areas 78.6% Pleural effusions 22.4% Small pneumothoraxes 3.37% Pericardial effusions 13.4% DVT 16.8%	Did not compare to other imaging modalities	RT‐PCR	NA
Castelao J et al.	Prospective	63	severe ARDS due to active COVID‐19 infection	B7 pattern in 203 (26.8%). B‐lines ≥7 mm apart B3 pattern in 143 (19%). B‐lines, 3 mm or less apart.	C pattern in 159 (21%). Anterior alveolar consolidation(s).	NA	NA	A small unilateral pleural effusion was observed in 3 patients (4.8%).	Did not compare to other imaging modalities	RT‐PCR	NA
Fonsi GB et al	Prospective study	63 patients (44 COVID positive)	46 (73%) patients had moderate and 17 (27%) had severe symptoms.	≤2 nonconfluent or confluent 20% ≥3 nonconfluent or confluent 80%	Consolidation 45%	NA	NA	Air bronchogram 39% Pleural effusion 18% Pericardial effusion 0% Thickened pleural line 86%	The sensitivity, specificity, PPV, and NPV of LUS for COVID‐19 pneumonia were 68%, 79%, 88%, and 52%, respectively. Whereas for chest CT they were 93%, 90%, 85%, and 95%, respectively.	RT‐PCR Chest CT	NA
Bar S et al.	Prospective study	100 adults of whom 31 had a positive SARS‐CoV‐2 RT‐PCR	ARDS n = 9 (29%), Admission to ICU n = 8 (26%), Death n = 6 (19%)	Upper and lower anterior: Confluent B‐lines n = 3 (10%) Posterolateral: Confluent B‐lines n = 10 (32%)	Upper and lower anterior N = 17 (54%) Posterolateral: n = 18 (58%)	NA	NA	Upper and lower anterior: Thickened pleural line n = 24 (77%) Posterolateral: Thickened pleural line n = 24 (77%)	NA	RT‐PCR	NA
Dargent A et al.	Prospective study	10	10 consecutive patients admitted in our ICU with moderate to severe ARDS	Monitored LUS score over ICU course	Did not compare to other imaging modalities	NA	NA	NA	NA	RT‐PCR	NA
Pivetta E et al.	Prospective study	228	Variable degree of severity. Ranging from mild disease discharged home, to patient admitted to ICU.	Not Reported	Not Reported	Not Reported	Not Reported	Not Reported	Higher sensitivity and NPV for an integrated approach combining clinical and Lung Ultrasound findings when compared to a negative initial RT‐PCR result	RT‐PCR
Calvo‐Cebrián A et al.	Prospective study	61	Moderate symptoms	Coalescent B‐lines 54.1% Multiple separated B‐lines 45.9%	Consolidation 31.1%	Not specified	Not specified	Irregular pleural line 27.9% Mild pleural effusion 6.6% Location of LUS findings: Bilateral 65.6%. Multifocal unilateral 6.6%. Unifocal 18%.	There was a significant association between the proposed LUS severity scale and the CXR severity scale: the higher the grade of US involvement, the higher the grade of radiologic involvement.	dichotomous variable that determined whether the hospital referral was “appropriate”	NA
Yassa M et al.	Prospective study	8	Mild, moderate and Critical.	NA	NA	NA	NA		Chest radiographic findings were negative and were not consistent with the LUS findings, chest CT showed similar findings as and was consistent with the LUS.	RT‐PCR	NA

Abbreviations: ARDS, Acute Respiratory Distress Syndrome; COVID‐19, Coronavirus‐2019; CTPA, Computed tomography pulmonary angiogram; CXR, Chest X‐Ray; HRCT, High resolution computed tomography; LUS, Lung ultrasound score; NPV, Negative Predictive Value; POCUS, Point of care ultrasound; PPV, Positive Predictive Value; RT‐PCR, Real time Polymerase Chain reaction.