The images do not look like any recognisable structures, but result from the reflection of the beam and reverberation of echoes on tissue interfaces, creating what is known as artefacts. The most important are:

• 1

  Pleural line: straight hyperechoic line formed by the difference in acoustic impedance between subcutaneous cell tissue/muscle (filled with liquid) and the lung (filled with gas).

  • 1.1

    Lung sliding: movement of the pleural line synchronized with respiratory movement produced by the change in the relative positions of the chest wall (parietal pleura) and the lung surface (visceral pleura). Video S1.

• 2

  A-lines: horizontal hyperechoic lines that are a reflection of the pleural line and parallel to it that repeat at intervals equal to the skin-pleural line distance. They appear in variable numbers and, in some cases, there may be A’-lines between 2 A-lines. Their presence is indicative of lung aeration.

• 3

  B-lines: vertical hyperechoic artefacts that arise from the pleural line in varying numbers throughout the entire screen. B-lines move with respiration and erase A-lines. They indicate the presence of liquid in the interstitial space and/or alveoli.

Table S1 describes the main artefacts and signs in lung ultrasound, both in the healthy lung and in specific diseases. Fig. S2.

They appear when the air-to-fluid ratio in the lung decreases and are characterised by the presence of B-lines. If the B-lines are clearly defined and discrete (non-confluent) with a defined pleural line, they are indicative of interstitial syndrome (Fig. S4, Video S3) and interstitial oedema, and if they are confluent, they are indicative of alveolar-interstitial (AI) syndrome and an even lower air-to-liquid ratio either due to alveolar oedema or decreased lung aeration, and the pleural line is usually irregular, thickened or poorly defined. The presence of a homogeneous AI pattern (compact B-lines through the entire lung field) is known as ‘white lung’ (Fig. S5, Video S4).

Pneumothorax

The sonographic features of pneumothorax are absence of lung sliding and of B-lines, exclusive presence of A-lines5 and the ‘lung point’ (LP), a sign that serves to locate the boundary between pneumothorax (absence of lung sliding) and any other lung pattern (with lung sliding) (Fig. 1, Video S5).

Figure 1.

Pneumothorax. (A) Anterior longitudinal plane. Absence of lung sliding, absence of B-lines and clearly defined A-lines. (B) M-mode. Absence of lung sliding. Stratosphere sign: pleural line and clearly defined equidistant A-lines. (C) Transverse plane, 4th–5th intercostal space. Lung point (LP): sonographic sign that marks the boundary between pneumothorax (absence of lung sliding) and any other lung pattern (lung sliding). On ultrasound, pneumothorax is considered mild-moderate if the LP is in the mid-clavicular line or anterior axillary line and severe if it is posterior to the mid-axillary line.

(0.05MB).

Absence of pleural line, alveolar collapse with air bronchograms and lung tissue hepatization. Also characterised by the development of shred signs, irregular hyperechoic artefacts generated by the irregular boundary between consolidated lung tissue and aerated lung tissue (Fig. S6, Video S6).

Pleural effusion

It has an easy to recognise pattern characterised by an anechoic area between the parietal and visceral pleura2,3 with a tendency for the effusion to accumulate in lower-level areas. The ‘quad’ sign may appear in 2D-mode, and the ‘sinusoid’ sign in M-mode. If the lung is compressed by the fluid, it will appear as lung consolidation with static bronchograms or even hepatization, and movement will create a lung flapping effect in the collapsed area known as the ‘jellyfish’ sign6 (Fig. 2, Video S7).

Figure 2.

Pleural effusion. (A) Jellyfish sign, secondary collapse caused by pleural effusion (anechoic contents). (B) Sinusoid sign (M-mode), generated by the movement of the lung toward the parietal pleura with inspiration. (C) Quad sign (2D-mode): rectangular shape in which the anechoic area corresponds to the effusion, demarcated by the visceral pleura (below) and the parietal pleura (above) and on the sides by the shadows of a rib or the diaphragm.

(0.07MB).

Lung ultrasound allows a quick and accurate diagnosis of pneumothorax, even in cases of subclinical leaks undetectable by plain radiography.7 However, lung sliding may also be absent in the context of selective intubation or lung hyperinflation.6 The presence of A-lines behind the sternum in the anterior mid-transverse plane has proven to be an easily identifiable sonographic feature with a high Sen, Spe and reproducibility for diagnosis of pneumothorax15 (Fig. S7, Video S8). The lung point is the most specific sign of pneumothorax and can help assess its severity based on its location relative to the different anatomical lines that define the regions of the thorax5 (Fig. 1C).

Table 2.

Differential diagnosis of neonatal respiratory disease.

Disease	Ultrasound pattern	Figure	Video (link)
Pneumothorax	Anterior mid-transverse plane:		Video S8
	• A-lines behind sternum (Figure S7)
	Longitudinal plane:
	• Absence of lung sliding
	Transverse plane:
	• Lung point
Transient tachypnoea of newborn	Heterogeneous pattern with variable involvement in different areas, with:		Video S9
	• Normal lung pattern
	• Interstitial pattern
	• Alveolar-interstitial pattern
	• Double lung point sign (50%) (Figure S9)
	Mild changes in pleural line
	Mild pleural effusion
Neonatal respiratory distress syndrome	Homogeneous alveolar-interstitial pattern (Fig. 4C)		Video S10
	Irregular or thickened pleural line
	Consolidations of variable size:
	• Small subpleural collapses (Fig. 4D) or atelectasis
	Subclinical pleural effusion
Transient respiratory distress (in term neonates)	Variable pattern:		Video S11
	• Alveolar-interstitial, no consolidations (video S11)
	• Normal pattern
	• Minimal air leaks
Meconium aspiration syndrome	Patchy pattern with areas of:		Video S12
	• Consolidations with air bronchogram (bilateral, variable size). Figure S12 B
	• Alveolar-interstitial pattern
	• Normal pattern
	Pleural effusion
	Dynamic features
Congenital pneumonia	Nonspecific pattern, varies depending on severity:		Video S13
	• Interstitial pattern and/or
	• Alveolar-interstitial with consolidations
	Irregular, thickened pleural line
	Subclinical pleural effusion

The sonographic signs of pneumomediastinum have not been defined formally.1 In the suprasternal plane, depending on the severity, there can be large A-line artefacts (Fig. S8) or small hyperechoic lines in the lateral margins of the thymus or in the thymic parenchyma.16

From a pathophysiological perspective, TTN is characterised by oedema in a structurally mature lung. The sonographic appearance is heterogeneous: zones of normal aeration, interstitial pattern or even fields of white lung.3,13 The presence of an alveolar-interstitial pattern in a gradient, that is, more marked in lower-level areas (lung base or posterior fields) is known as ‘double lung point’ sign (Fig. S9, Video S9), a frequent sign found in nearly 50% of cases, although not specific to TTN.17 There may be abnormalities in the pleural line or pleural effusion, but consolidations are rarely present.3,17,18

Respiratory distress syndrome results from a primary (preterm neonates) or secondary (term neonates) deficiency of surfactant, and manifests with pulmonary oedema and decreased lung volume (collapse or atelectasis). It sonographic appearance is characterised by the loss of aeration with a homogeneous AI pattern and an irregular or thickened pleural line, resulting in a uniformly hyperechoic lung field (white lung),13 with condensations that may range from small subpleural collapses to large consolidations with air bronchogram (atelectasis), possibly associated with subclinical pleural effusion (Fig. 4, Video S10). Depending on the severity and course of disease, there may be different degrees of involvement in each field or hemithorax.1–3,19

Figure 4.

Respiratory distress syndrome. A, B and C: preterm neonate born at 30+2 wk. Urgent caesarean section without prenatal steroids. Lung ultrasound at admission: (A) Anterior plane: homogeneous alveolar-interstitial pattern without areas of aeration (no A-lines) and irregular pleural line, white lung. (B) Lateral plane, similar findings. (C) Posterior plane: alveolar-interstitial pattern with pleural thickening and subpleural consolidations in lung base. (D) Transverse plane: alveolar-interstitial pattern with subpleural consolidations (arrow). (E) Posterior longitudinal plane with alveolar-interstitial pattern with pleural thickening, subpleural consolidations and consolidation with air bronchogram (atelectasis) in 2 intercostal spaces (arrow).

(0.12MB).

There is no characteristic pattern for transient respiratory distress in term neonates, characterised by mild symptoms and quick resolution with a few hours of respiratory support. There is frequently a patchy AI pattern without consolidations and/or a normal lung pattern, sometimes with minimal air leaks3 (Fig. S10, Video S11). Less frequently, there may be an AI pattern on a gradient or small consolidations. This heterogeneity may be due to the different pathophysiological mechanisms that can lead to respiratory distress. Usually, transient respiratory distress resolves quickly within 12–24 h, with normalization of clinical and sonographic features.20

The ultrasound findings in meconium aspiration syndrome are heterogeneous and correlate to the variable severity of this disease. Published descriptions include the presence of consolidations with air bronchograms (bilateral, disperse and of variable size). There may also be a bilateral AI pattern, diffuse or patchy,21,22 with an irregular pleural line with areas of normal aeration and of pleural effusion (Fig. S11, Video S12). These features are dynamic, and their location may change with the course of disease.1

These features can also be seen in other forms of aspiration, such as aspiration of clear fluid, blood or milk, which have a more favourable sonographic presentation.

Congenital pneumonia presents with an interstitial or AI pattern associated with consolidations, an irregular and thickened pleural line and in some cases minimal pleural effusion23 in the context of risk factors for infection, respiratory distress and elevation of acute phase reactants (Fig. S12, Video S13).

There are several semiquantitative tools that assess the sonographic features of RDS and correlate well to parameters of oxygenation in preterm infants (oxygen index, alveolar-arterial oxygen gradient, the ratio of the oxygen saturation and the fraction of inspired oxygen [S/F ratio]) and the need of surfactant in infants delivered before 34 weeks24 (Table S3). The goal is to identify the need for surfactant administration earlier than allowed by traditional clinical criteria to also initiate treatment earlier and reduce the complications associated with delayed treatment.

In this regard, a cohort study in neonates delivered before 32 weeks used a LUS score to determine the need for surfactant and found a significant increase in the proportion of patients that received surfactant before 3 h post birth, with no increase in the total proportion of infants that received surfactant at any time.25 A randomised controlled trial in preterm neonates delivered before 32 weeks that received surfactant in the case of a LUS score greater than 8 or a FiO2 greater than 0.3 found significant differences in the timing of surfactant administration and greater S/F ratio values after surfactant administration in the group given surfactant based on the LUS, without significant differences in the proportion that received surfactant.26 Later studies have developed models for prediction of surfactant administration combining the LUS score and clinical parameters (S/F ratio). These models had an excellent predictive value, and both are now available for application as online calculators.27,28

There are few studies in preterm neonates delivered after 34 weeks are few and include few patients of vastly different characteristics, so their results must be interpreted with caution.13 One of them, conducted by Brat et al.,24 established a cut-off point with a low specificity, which limits its application in clinical practice. Contrary to extremely preterm neonates, respiratory disease in late preterm neonates is more diverse, with the possible coexistence of different diseases with similar clinical and sonographic features.

When it comes to the usefulness of LUS to predict the need of repeated doses of surfactant, only 2 studies in patients born before 32 weeks have been published to date24 that, in addition, used different criteria to determine the need of surfactant. Although the precision was good, the proposed cut-off points had a low sensitivity.

One of the main applications of LUS is the diagnosis of complications associated with ventilation, which justifies its integration in clinical practice for diagnostic purposes.

Lung ultrasound allows early detection of consolidations indicative of endotracheal tube (ETT) malposition (Fig. 5 A–B) or due to the decubitus position in patients with bronchopulmonary dysplasia (BPD). The combination of LUS findings and clinical manifestations increases the PPV of diagnosis of ventilator-associated pneumonia (VAP). A semiquantitative multiparameter score has been proposed for diagnosis of VAP that takes into account the most frequent sonographic findings of the disease: consolidations greater than 0.5 cm, bilateral consolidation, posterior consolidation, non-translobar consolidation and bronchogram (static, present in 100%, and dynamic, present in 70% of cases of VAP), highlighting the inability to completely reopen the consolidated area with the customary ventilation as a specific sign of VAP29 (Fig. 5C–F).

Figure 5.

Detection of complications, forms of consolidation. A and B: Lung collapse/atelectasis due to endotracheal tube malposition. (A) Left anterior field: disappearance of pleural line with lung tissue hepatization and air bronchogram (hyperechoic, solid white arrow) and fluid (anechoic, hollow arrow). (B) Lateral field: absence of pleural line and hepatization, shred sign (hyperechoic border, boundary of collapsed lung/aerated lung) (star). (C) Preterm neonate born at 25+3 wk. Bronchopulmonary dysplasia. Consolidation/atelectasis in posterior field, disorganized bronchogram. (D) Same patient, follow-up ultrasound 2 h after change to prone position: resolution of consolidation. (E) Preterm neonate born at 23+5 wk, 50 h post birth, consolidation in left posterior field in the context of derecruitment. (F) Same patient at 35 days post birth, consolidation of homogeneous appearance in the clinical context compatible with ventilator-associated pneumonia.

(0.28MB).

Lung ultrasound can also be used to monitor oedema or extravascular lung fluid and optimise their management in neonates with poor lung clearance or congenital heart disease (CHD) and pulmonary overflow.30 Based on the severity of oedema, the sonographic appearance ranges from the interstitial to the AI pattern, possibly associated with pleural effusion.

Different approaches have been described to assess ETT placement32,33 and depth ETT34 by means of ultrasound (Fig. S14). However, there is a quick albeit indirect means to verify correct placement by confirming the presence of bilateral lung sliding with LUS.

Lung ultrasound is an adequate tool to predict the development of bronchopulmonary dysplasia (BPD) in infants born preterm before 32 weeks of gestation: several multicentre studies35–37 demonstrate that a LUS score similar to the one used to predict the need of surfactant performs well for diagnosis of BPD from as early as 3 days post birth, although the optimal timing for its use is 1 week post birth.

The usual evolution of pulmonary patterns in neonates delivered before 32 weeks starts with a pattern characteristic of RDS at birth, with high LUS score values in the first days, but then normally improve thereafter as the primary surfactant deficiency resolves, reaching LUS score values near 0 at around 1 week post birth.38 The patients whose LUS scores do not exhibit this decrease or actually increase are more likely to develop BPD (Fig. 6). On the other hand, in preterm infants born before 28 weeks, this decrease in the LUS score delays, even after the initial surfactant deficiency stage has passed, to up to 4 weeks post birth39–41 (Table S4).

Figure 6.

Sonographic evaluation in the first week of life in preterm neonate born at 25 weeks whose mother developed chorioamnionitis and was a carrier of Ureaplasma parvum in cervical fluid. (A) Normal pattern at 4 h post birth. (B) Day 3 post birth: alveolar-interstitial pattern with pleural thickening and fragmentation of the pleural line only in posterior fields. (C) Week 1 post birth: the same abnormal features are visible in every lung field. At this point, the respiratory support needs of the patient did not increase significantly, but starting from 2 weeks post birth she required aggressive scalation of ventilatory parameters and eventually developed severe bronchopulmonary dysplasia.

(0.08MB).

Lung ultrasound is an ideal technique for assessment of lung aeration, lung fluid and alveolar collapse, information that is crucial for the dynamic monitoring of pulmonary status and the response to treatment, including changes in respiratory support, adjustment of fluids, etc.

Studies in preterm neonates have found a correlation between semiquantitative assessment instruments and oxygen saturation,24 the probability of NIV13 as early as the delivery room42 and the need of invasive mechanical ventilation (IMV). However, further research is required to determine whether changes in non-invasive or invasive ventilation parameters are reflected in changes in LUS scores.

Lung ultrasound also plays a key role in decision-making regarding weaning from IMV in combination with the clinical assessment, diaphragmatic ultrasound and respiratory parameters, as it allows ruling out complications and assessing the degree of lung aeration taking into account the lung features associated with the underlying disease and gestational age35,43,44 (Fig. 5C and D).

Lung ultrasound has proven useful for monitoring of pulmonary oedema, diagnosis of atelectasis and ultrasound-guided recruitment manoeuvres and detection of postoperative complications (pleural effusion, pneumothorax or haematomas).50,51

The assessment of the diaphragm by means of ultrasound is important given the prevalence of diaphragmatic paralysis secondary to trauma and its impact on lung function.

From a functional perspective, there is a correlation between different semiquantitative LUS scoring systems and the duration of mechanical ventilation or the length of stay, and their inclusion can increase the predictive power of multiparametric models based on clinical variables.52

Thus, LUS should be recommended in the non-invasive monitoring required for the daily evaluation of cardiac patients, as it offers useful information that may contribute to improved management.