Rib-indexed quantitative lung ultrasound versus chest X-ray for lung recruitment assessment in neonates with moderate-severe ARDS on surfactant therapy combined with prone position: a prospective observational study

Abstract

Serial lung recruitment assessment in neonates with moderate-to-severe neonatal acute respiratory distress syndrome (NARDS) is crucial. However, current methods involve ionizing radiation or invasiveness, which limits their serial use in neonates. This study evaluated the feasibility of rib-indexed quantitative lung ultrasound (LUS) as a radiation-free alternative for monitoring lung aeration in neonates with moderate-to-severe NARDS on surfactant therapy combined with prone position. A prospective observational study enrolled 35 term neonates with moderate-to-severe NARDS. Lung recruitment was assessed via anterior–posterior approach rib-indexed quantitative LUS and posteroanterior chest X-ray (CXR) before and 6 h after combined surfactant therapy and prone position. Following the intervention, it demonstrated a significant reduction in the LUS aeration score, from a pre-intervention median of 18 points (IQR 16, 22) to a post-intervention median of 15 points (IQR 12, 20) (P < 0.001). In contrast, the decrease in the CXR score (pre-intervention median 3 (IQR 3, 4) vs. post-intervention median 2 (IQR 2, 3)) did not reach statistical significance (P = 0.059). Posterior approach rib-indexed quantitative LUS showed high concordance with posteroanterior CXR in determining the rib level of the pulmonary-diaphragmatic interface (ICC > 0.95, kappa > 0.94, P < 0.001). No adverse events occurred during the LUS assessments.

Conclusion: Posterior approach rib-indexed quantitative LUS is a reliable and non-invasive modality for real-time lung recruitment assessment in neonates with NARDS. It significantly detected improved lung aeration following surfactant therapy combined with prone position, whereas CXR failed to demonstrate a statistically significant improvement. Posterior approach rib-indexed quantitative LUS can also determine the rib level of the pulmonary-diaphragmatic interface, similarly to posteroanterior CXR. The superior sensitivity and safety of rib-indexed quantitative LUS offer a clinically valuable and innovative alternative for dynamic monitoring of lung recruitment in neonatal critical care. Future multi-centre studies should integrate CT validation to confirm broader applicability.

Trial registration: The trial was prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2300074652) on August 11, 2023.

What is Known: • Lung ultrasound (LUS) is a guideline-recommended, radiation-free tool for neonatal lung assessment. • Chest X-ray (CXR) has limitations in detecting real-time lung aeration changes due to radiation risks and projectional artifacts.

What is New: • Posterior-approach rib-indexed quantitative LUS reliably tracks the rib level of the pulmonary-diaphragmatic interface and detects improved lung aeration following surfactant therapy combined with prone position in moderate-severe NARDS. • A novel vascular-anchored first rib indexing protocol that was previously unreported enables reliable neonatal rib counting.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00431-025-06313-3.

Background

Neonatal acute respiratory distress syndrome (NARDS) affects 1.5% of neonates, with persistent mortality of 17 ~ 24% despite advanced care [1–3]. Moderate-to-severe cases (85.5%) significantly increase bronchopulmonary dysplasia (BPD) risk [2]. Etiologies include severe sepsis (37.7%), meconium aspiration (27.2%), and pneumonia (15.9%) [2, 4, 5]. Current management (e.g., mechanical ventilation, surfactant, and prone position) [6, 7] lacks real-time non-invasive monitoring to optimize lung aeration and reduce volutrauma [8].

Conventional assessment methods have limitations: Imaging techniques (e.g., chest X-ray (CXR), computed tomography (CT), and electrical impedance tomography (EIT)) quantify lung volume/aeration [9, 10]. Despite being the diagnostic gold standard for lung recruitment assessment, cumulative radiation exposure restricts neonatal serial use [11, 12]; non-imaging techniques (e.g., ventilator-derived parameters and extravascular lung water (EVLW)) [13, 14] offer real-time monitoring but provide indirect correlations and poor spatial resolution [15].

Lung ultrasound (LUS) is now endorsed by ESPNIC 2020 guidelines for neonatal lung aeration monitoring [16]. It provides reliable, safe, real-time assessment superior to alternatives [17, 18], guiding interventions like prone position [19], positive end-expiratory pressure (PEEP) titration [20], surfactant therapy [21], respiratory support adjustment [22], and predicting BPD [23].

While the validated 4-step LUS aeration score (0 ~ 3 points) as a quantitative LUS scoring system quantifies aeration loss [17, 24–26], we explore a simplified quantitative approach. Radiographically, optimal inflation positions the right hemidiaphragm at/below the 8th posterior rib [27]. Prior studies have validated rib counting via ultrasound in adults and neonates [28–31]. Recent studies confirm high concordance between point-of-care ultrasound (POCUS) and CXR for rib-level diaphragm assessment [32], suggesting POCUS can replicate radiographic logic via rib-level anatomical mapping.

We therefore innovatively employ rib-indexed quantitative LUS to dynamically assess lung recruitment in neonates with moderate-severe NARDS. This replaces serial CXR with a lung aeration score and the craniocaudal displacement of the lung–diaphragm interface relative to rib landmarks following surfactant therapy combined with prone position.

Methods

Study design and setting

This investigator-initiated, prospective, observational, single-center, single-blind, non-inferiority study was conducted between September 1, 2023, and September 1, 2024, in a 60-bed tertiary NICU. The study protocol received ethical approval from the Institutional Review Board of Fujian Provincial Maternity and Children’s Hospital (Approval No. 2023KY020) and was prospectively registered with the Chinese Clinical Trial Registry (Registration ID: ChiCTR2300074652).

Eligibility criteria

This study enrolled neonates admitted to the NICU within 72 h postnatally.

The inclusion criteria were as follows:

1. Term infants diagnosed with moderate-to-severe NARDS;

   The Montreux definition of NARDS [3] comprises four diagnostic criteria: (1) Time frame: acute onset (within 1 week) from a known or suspected clinical insult; (2) Exclusion: respiratory distress syndrome (RDS), transient tachypnea of the newborn (TTN), or congenital anomalies as a primary current acute respiratory condition; (3) Lung imaging: diffuse, bilateral, and irregular opacities or infiltrates, or complete opacification of the lungs, which are not fully explained by local effusions, atelectasis, RDS, TTN, or congenital anomalies; and (4) Origin of edema: absence of congenital heart disease explaining the oedema (this includes patent ductus arteriosus with pulmonary overflow if no acute pulmonary hemorrhage exists) and echocardiography is needed to verify the origin of oedema.

2. Gestational age (GA) > 37 weeks with birth weight > 2500g;

3. Clinical indications for both surfactant therapy and prone position following multidisciplinary evaluation;

   Surfactant therapy was indicated based on these criteria: Eligible neonates had confirmed ARDS with severe hypoxemic respiratory failure and bilateral pulmonary exudates on imaging. Inclusion required an oxygenation index (OI) ≥ 8, calculated as: OI = [Mean Airway Pressure (MAP) × FiO2 × 100]/Arterial Oxygen Pressure (PaO2).

Prone position was considered following the exclusion of absolute contraindications: severe hemodynamic instability, intracranial hemorrhage with hypertension, untreated tension pneumothorax, congenital airway malformations, recent thoracic/abdominal trauma or surgical interventions, and severe spinal/neurological abnormalities.

The exclusion criteria were as follows:

1. Technical contraindications for pulmonary ultrasonography (extensive subcutaneous emphysema or chest/back dressings covering > 50% assessment areas).

2. Multiorgan dysfunction involving ≥ 3 systems (cardiovascular, hematologic, neurologic, renal, or gastrointestinal).

3. Major cardiopulmonary malformations (e.g., congenital diaphragmatic hernia).

4. Genetic predisposition to pulmonary disorders.

Withdrawal protocol activated when:

1. Legal guardians requested discharge against medical advice (DAMA) within the initial 7-day intervention window.

2. Life-threatening clinical deterioration unresponsive to maximal intensive care.

Written informed consent was obtained from legal guardians prior to study enrollment.

Interventions

A standardized sequential protocol is described in Fig. 1:

1. Time 1: Supine positioning with posteroanterior CXR and anterior-approach LUS; (2) Time 2: Prone conversion (Switch 1) with immediate posterior-approach LUS; (3) Time 3: Surfactant therapy after 30-min prone “wash-out period [19]”; (4) Time 4: Posterior-approach LUS at 6-h prone interval; (5) Time 5: Supine conversion (Switch 2) with immediate anterior-approach LUS and posteroanterior CXR.

Fig. 1.

A standardized sequential protocol for intervention. This figure illustrates the step-by-step protocol for intervention: (1) Time 1 (Supine): patient positioned supine; posteroanterior CXR and anterior-approach LUS performed. (2) Switch 1 (Prone conversion): patient turned prone; immediate posterior-approach LUS conducted. (3) Time 3 (Therapy): surfactant administered after a 30-min prone “wash-out period”. (4) Time 4 (Prone maintenance): posterior-approach LUS repeated at 6-h prone interval. (5) Switch 2 (Supine conversion): patient returned to supine; immediate anterior-approach LUS and posteroanterior CXR performed. Note: “supine” and “prone” in the figure denote supine and prone position, respectively. CXR, chest X-ray; LUS, lung ultrasound

Positional changes assisted by two certified nurses; LUS performed by a certified operator.

Standard procedure for anterior–posterior approach rib-indexed quantitative LUS is in the SupplementaryMaterial Section 1. The video for the protocol of anterior–posterior approach rib-indexed quantitative LUS is demonstrated online (Supplementary Material Video S1).

Ultrasonography was conducted using a Philips CX50 system (Bothell, WA) equipped with an L12-3 broadband linear transducer operating on neonatal presets. Examinations were performed by a board-certified neonatologist (10 years of NICU experience, 5 years POCUS specialization, Chinese Critical Care Ultrasound certified). Quality assurance was supervised by a lead sonographer (10 years of neonatal imaging experience), with independent verification by a pediatric radiologist specializing in neonatal thoracic imaging. Continuous physiological monitoring until discharge was managed by two dedicated clinical research coordinators using an electronic data capture (EDC) system.

Routine care

All enrolled neonates received standardized nutritional support and medical interventions that strictly adhered to institutional evidence-based protocols for critical infant care, with therapeutic regimens maintained in accordance with unit-specific clinical pathways approved by the multidisciplinary NICU quality oversight committee.

Baseline demographic and clinical characteristics

Case report forms (CRFs) were used to systematically document two categories of data in the Supplementary Material Section 2.

Outcomes

The primary outcome was the consistency between posteroanterior CXR and anterior–posterior approach rib-indexed quantitative LUS in determining the rib level corresponding to the pulmonary-diaphragmatic interface at two specific time points: before and 6 h after surfactant therapy in combination with prone position. The secondary set of outcomes included LUS aeration score, CXR score, and other clinical parameters between pre-intervention and post-intervention.

LUS aeration score: extended 10-zone LUS scores (eLUS10) Protocol, 5-zone per side, including the scan of upper anterior, lower anterior, lateral, upper posterior, and lower posterior [33, 34]. The eLUS10 system employs a 4-tier scoring scale per zone (0–3 points), with the cumulative score ranging from 0 to 30 points [35]. CXR score: RDS Radiographic Severity Classification (1–4 points) [36].

Bradycardia (defined as a heart rate < 80 beats/min), desaturation (defined as a low saturation of less than 80%), and a decrease in blood pressure (defined as less than 80% of the mean pressure) were recorded as possible side effects during the evaluation of rib-indexed quantitative LUS.

Sample size calculation

PASS (version 2021) software was used to calculate the sample size for this study. The primary objective was to demonstrate that the inferior pulmonary border rib level, as indicated by rib-indexed quantitative LUS, as well as CXR, could be used to assess the effect of lung recruitment precisely. The primary outcome was the consistency of the rib level corresponding to the pulmonary-diaphragmatic interface between the rib-indexed quantitative LUS and CXR. The results of the initial pilot experiment indicated that the pulmonary-diaphragmatic interfaces on the subscapular line measured by the two methods were distributed from the 8th to the 10th rib levels. The consistency between the two methods was evaluated through the kappa value. The significance level α = 0.05 was set by PASS 2021 software, and the marginal classification frequencies of the 8th, 9th, and 10th rib levels were 0.15, 0.70, and 0.15, respectively. With a sample size of 35 subjects, the ability to detect the true Kappa value of 0.8 in the test of H0: Kappa = 0.4 vs. H1: Kappa ≠ 0.4 was 80.34%; that is, the power could reach 80.34%.

In the NICU of Fujian Children’s Hospital, an annual total of 50 ~ 60 term infants are diagnosed with moderate to severe ARDS, necessitating the combination of surfactant therapy and prone position. Assuming that 70% of the term infants met the inclusion criteria (based on retrospective data from 2019 to approximately 2022), the projected enrolment of 35 ~ 40 neonates ensured adequate statistical power within the 12-month recruitment period.

Statistical analysis

All analyses were conducted using SPSS Statistics version 18.0 (IBM Corporation) following a predetermined analytical plan. Continuous variables were subjected to normality assessment through Shapiro–Wilk tests (α = 0.05). The normally distributed parameters are expressed as the means ± standard deviations and were compared using paired t-tests. Nonparametric data are summarized as medians (interquartile ranges [IQRs]) and were compared using the Wilcoxon signed-rank test. Categorical variables are presented as counts (percentages).

Concordance between CXR and rib-indexed quantitative LUS assessments was systematically evaluated using three complementary analytical strategies: (1) intraclass correlation coefficients (ICCs, two-way mixed-effects model) with 95% confidence intervals for continuous measures; (2) weighted kappa statistics (quadratic weights) for ordinal data; and (3) Spearman’s rank correlation coefficients (ρ) to assess monotonic relationships. ICC and kappa analyses were utilized to investigate the level of consistency, whereas Spearman analysis was adopted to explore the level of relevance between posteroanterior CXR and rib-indexed quantitative LUS in determining the rib level corresponding to the pulmonary-diaphragmatic interface at two specific time points, namely, before surfactant therapy in combination with prone position and after 6 h. In the present study, we disregarded systematic errors, and all the data were raw and uncalculated. A two-tailed α level of 0.05 was used to define statistical significance throughout the study.

All the statistical procedures were independently executed by a doctoral-level biostatistician blinded to the clinical groupings and outcome data. No interim analyses or data-driven methodological changes occurred during the study period.

Data collection, management, and monitoring

These were comprehensively documented in the Supplementary Material Section 3.

Quality assurance

These were comprehensively documented in the Supplementary Material Section 4.

Ethical oversight and safety protocols

These were comprehensively documented in the Supplementary Material Section 5.

Results

A total of 41 infants who met the inclusion criteria during the period of recruitment from September 1, 2023, to September 1, 2024, and 4 infants were excluded. Two participants were withdrawn from the study. No intervention-related bradycardia, hemodynamic instability, or oxygen desaturation events occurred. The flow of participants through the stages of the trial is shown in the Supplementary Material Fig. S1.

The triggers of NARDS in this cohort were as follows: seven cases of severe congenital pneumonia, nine cases of severe congenital pneumonia with concurrent pulmonary hemorrhage, five cases of severe congenital pneumonia complicated by tension pneumothorax, one case of severe congenital pneumonia presenting with both tension pneumothorax and pulmonary hemorrhage, 5 five cases of severe perinatal asphyxia, two cases of severe congenital pneumonia associated with patent ductus arteriosus (PDA) > 5 mm, 1 case each of congenital fetal hydrops, viral myocarditis of congenital origin, tricuspid valve malformation, central venous catheter (CVC)-associated pericardial effusion with tamponade following annular pancreas surgery, postanaesthetic NARDS secondary to imperforate anus repair, and early-onset Acinetobacter baumannii sepsis. All enrolled neonates demonstrated concurrent persistent pulmonary hypertension of the newborn (PPHN). The baseline demographic of the study population is shown in Table 1.

Table 1.

Baseline demographic of the study population

Variable	Value
Admission age (hours)	13 (5, 24)
Male sex	80.0% (28/35)
Gestational age (weeks)	38 + 1 (37 + 4, 39 + 3)
Birth weight (g)	3160.57 ± 441.30
Small for gestational age	8.6% (3/35)
Cesarean delivery	62.9% (22/35)
Multiple pregnancies	8.6% (3/35)
Premature rupture of membranes	11.4% (4/35)
Amniotic fluid contamination	25.7% (9/35)
SNAPPE-II	16 (16, 28)
Maternal age (years)	30.80 ± 4.84
Maternal hypertension	8.6% (3/35)
Maternal diabetes	28.6% (10/35)
Antenatal corticosteroids	0% (0/35)
Antenatal antibiotics	5.7% (2/35)
Antenatal infection	5.7% (2/35)

Data were presented as median (IQR), mean ± SD, or n (%). SNAPPE-II: Score for Neonatal Acute Physiology with Perinatal Extension-II

The changes in CXR score and LUS aeration score following intervention (surfactant therapy combined with 6 h prone position) are illustrated in Fig. 2. Both CXR and LUS aeration scores decreased following the intervention. The LUS aeration score demonstrated a statistically significant reduction from pre-intervention (median: 18 points (IQR: 16, 22)) to post-intervention (median: 15 points (IQR: 12, 20); P < 0.001). In contrast, the CXR score decrease (pre-intervention: 3 (3, 4) vs. post-intervention: 2 (2, 3)) did not reach statistical significance (P = 0.059). Other clinical parameters between pre-intervention and post-intervention are shown in Supplementary Material Table S1.

Fig. 2.

Changes in chest X-ray and LUS aeration scores following intervention. This box plot comparing as median (interquartile range) Chest X-ray score and LUS aeration score before (Pre Intervention) and after (Post Intervention) the intervention. Gray box and black box represent Chest X-ray score or LUS aeration score, respectively. Circles and T-bars represent median (interquartile range) and extreme value, respectively. Chest X-ray scores: pre-intervention = 3 (3, 4) , post-intervention = 2 (2, 3). LUS aeration scores: pre-intervention = 18 (16, 22), post-intervention = 15 (12, 20). Statistical significance determined by Wilcoxon signed-rank test (P < 0.001 for LUS; P = 0.059 for CXR)

The consistency between CXR and rib-indexed quantitative LUS in determining the rib level corresponding to the pulmonary-diaphragmatic Interface is shown in Table 2.

Table 2.

The consistency between posteroanterior CXR and posterior/anterior approach rib-indexed quantitative LUS in determining the rib level corresponding to the pulmonary-diaphragmatic interface

Comparison Group	ICC (95% CI)	P value	Kappa value	P value	Spearman's ρ	P value
Pre-intervention: CXR vs. Posterior POCUS	0.957 (0.918, 0.978)	<0.001*	0.942	<0.001*	0.956	<0.001*
Pre-intervention: CXR vs. Anterior POCUS	0.132 (-0.007, 0.452)	<0.001*	-0.029	0.227	0.913	<0.001*
Post-intervention: CXR vs. Posterior POCUS	0.955 (0.913, 0.977)	<0.001*	0.946	<0.001*	0.949	<0.001*
Post-intervention: CXR vs. Anterior POCUS	0.114 (-0.031, 0.399)	<0.001*	-0.047	0.064	0.673	<0.001*

ICC Intraclass Correlation Coefficient, CI Confidence Interval, Spearman’s ρ Spearman correlation values, CXR Chest X-ray, POCUS Point-of-Care Ultrasound. P values marked with * indicate significance (p < 0.05)

The quality control was comprehensively documented in the SupplementaryMaterial Section 6.

Discussion

Summary of key findings

This prospective study evaluated the feasibility of rib-indexed quantitative LUS replacing CXR for lung recruitment assessment in a cohort of 35 neonates with moderate-to-severe NARDS on surfactant therapy combined with prone position. The principal findings revealed a statistically significant improvement in LUS aeration score following the combined intervention, evidenced by a significant reduction from a pre-intervention median of 18 points (IQR 16, 22) to a post-intervention median of 15 points (IQR 12, 20) (P < 0.001). In contrast, the decrease in the CXR score (pre-intervention median 3 (IQR 3, 4) vs. post-intervention median 2 (IQR 2, 3)) did not reach statistical significance (P = 0.059). The study also documented the consistency between posteroanterior CXR and posterior approach rib-indexed quantitative LUS in localizing the rib level of pulmonary-diaphragmatic interface. There were no intervention-related adverse events recorded.

Rib-indexed quantitative LUS for lung recruitment assessment: validation and limitations

The statistically significant reduction in LUS aeration score contrasted with the non-significant CXR score trend highlights LUS’s superior sensitivity in detecting real-time alveolar recruitment. This divergence stems from fundamental technical distinctions: LUS dynamically quantifies resolution of gravity-dependent atelectasis in posterior lung zones through direct visualization of alveolar-interstitial changes (B-line regression, consolidation reduction) during prone positioning. Conversely, CXR’s projectional limitations obscure regional aeration dynamics due to superimposition of structures. The magnitude of LUS improvement further reflects the synergistic physiological action of surfactant and prone positioning, corroborating ventilation-perfusion patterns observed by Loi et al. [19]. Collectively, these findings position LUS as a precision biomarker for guiding recruitment strategies in NARDS, transcending CXR’s static anatomical snapshots to enable real-time management of lung aeration.

The high concordance between rib-indexed quantitative LUS and CXR in determining the rib level corresponding to the pulmonary-diaphragmatic interface underscored the reliability of this POCUS protocol for monitoring lung recruitment. This corresponds with a prior study that validated LUS based on rib count in high-frequency oscillatory ventilation as a sensitive modality for assessing lung aeration [32].

Posterior approach LUS agreed well with posteroanterior CXR for determining the rib level of the pulmonary-diaphragmatic interface, unlike the anterior approach LUS. This differential effect between the posterior (dependent zone in supine position) and anterior approach is consistent with the established impact of gravity on lung aeration [37, 38]. As ultrasound excels at detecting loss of aeration, scanning the dependent lung zones (posterior in supine neonates) more readily reveals atelectasis and thus better captures the potential for recruitment [19, 39]. This principle is well-supported using CT-scan and EIT in adults and has been demonstrated with LUS in neonates [19, 40, 41].

Vascular-anchored first rib indexing protocol

Accurate rib identification constitutes a foundational yet technically demanding prerequisite for rib-indexed quantitative LUS in neonates, owing to unique anatomic constraints including predominantly cartilaginous rib composition, narrow intercostal spacing, and dynamic postnatal thoracic adaptation [42, 43]. While standardized ultrasound-guided rib counting protocols exist for adults [44], neonatal POCUS lacks equivalent standardization [44].

The first rib, a critical landmark, is particularly challenging to delineate. Anteriorly, the acoustic shadowing of the clavicular frequently obscures the first rib, risking misidentification of the second rib. Posteriorly, the acoustic similarity between the C7 transverse process and the first rib complicates differentiation [43]. To address inconsistent osseous landmark reliance [45], we developed a novel vascular-anchored first rib indexing protocol, which was previously unreported in neonatal populations. In the anterior approach, the subclavian vein serves as a key landmark between the clavicle and first rib, while in the posterior approach, the dorsal scapular or deep cervical arteries traversing the first rib’s superior margin provide definitive localization.

Leveraging the inherent vascular stability (which is unaffected by posture or tissue variability), this approach integrates vascular-osseous landmarks, reducing enumeration errors. This represents a significant methodological advance.

Dual thresholds in rib-indexed quantitative LUS implementation

The rib-level protocol demonstrated superior inter-operator reproducibility [46, 47], attributable to protocol robustness and POCUS operator expertise. This aligns with prior evidence that targeted training enables clinicians across specialties to achieve diagnostic-level POCUS competency [48, 49]. While targeted training enables rapid acquisition of basic POCUS competency across specialties, quantitative applications like rib indexing require longitudinal mentorship to maintain precision, highlighting a critical duality in implementation. This protocol serves as an educational foundation for thoracic POCUS. Future work must define minimal training thresholds balancing efficacy and resource constraints, especially in low-resource NICUs.

Limitations

While this study established a novel methodology for POCUS-guided lung recruitment monitoring, several constraints warrant consideration: the single-center design introduces potential institutional bias in surfactant and prone position protocols, limiting generalizability across diverse NICUs. Our sequential intervention protocol, in which surfactant was uniformly combined with prone position, precludes isolation of their individual contributions to the observed changes in LUS or CXR scores. Future studies should employ factorial designs (e.g., surfactant-only vs. prone-only arms) to delineate their distinct physiological impacts on lung recruitment dynamics. Standardized training ensures POCUS operator competency, but inherent sonographic technique variations (probe pressure/angulation) and transitional positioning effects (prone-supine haemodynamic shifts) may introduce measurement variability that is not fully quantified. Crucially, the absence of CT validation precludes definitive volumetric recruitment correlation, whereas the short postintervention window and lack of long-term neurodevelopmental follow-up constrain the assessment of sustained clinical impacts. Future multi-centre trials integrating CT ground-truth verification and extended surveillance periods (corrected age 6 ~ 18 months) are warranted to validate these findings across broader neonatal populations.

Conclusion

This study pioneers a posterior approach rib-indexed quantitative LUS protocol for radiation-free monitoring of lung recruitment in neonates with moderate-to-severe NARDS. It significantly detected improved lung aeration following surfactant therapy combined with prone position. The superior sensitivity and safety in quantifying post-intervention lung aeration improvements makes it ideal for clinical use. By bridging anatomical precision with functional assessment, this methodology empowers clinicians to optimize ventilator strategies dynamically, exemplifying the transformative potential of POCUS in neonatal criticalccare.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

NARDS

Neonatal acute respiratory distress syndrome

CXR

Chest X-ray

CT

Computed tomography

EIT

Electrical impedance tomography

EVLW

Extravascular lung water

POCUS

Point-of-care ultrasound

PEEP

Positive end-expiratory pressure

TTN

Transient tachypnea of the newborn

NICU

Neonatal intensive care unit

BPD

Bronchopulmonary dysplasia

LUS

Lung ultrasound

OI

Oxygenation index

GA

Gestational age

DAMA

Discharge against medical advice

CRFs

Case report forms

SGA

Small for gestational age

SNAPPE-II

Score for Neonatal Acute Physiology with Perinatal Extension-II

FiO₂

Fraction of inspired oxygen

MAP

Mean airway pressure

ECMO

Extracorporeal membrane oxygenation

PPHN

Persistent pulmonary hypertension of the newborn

PDA

Patent ductus arteriosus

Author contributions

X. Ouyang contributes to designing the study, implementing POCUS and drafting the manuscript. L. Fang and H. Yang assist in designing the study and revising the manuscript. W. Ling is a ultrasound physician responsible for POCUS blinded interpretation and POCUS quality control. Q. Ou is a paediatric radiologist responsible for CXR blinded interpretation. H. Zhang and S. Huang collects and register the clinical data. F. Chen analyzes the data statistically and is not involved in the study design or the efficacy evaluation. Y. Fan is the head nurse in charge of managing and training nurses.Y. Lin assists in designing the study and revising the manuscript. All the authors read and approved the final article.

Funding

This work was supported by the Joint Funds for the Innovation of Science and Technology, Fujian province (grant number 2021Y9165), Startup Fund for scientific research, Fujian Medical University (grant number: 2022QH1228), Startup Fund for scientific research, Fujian Medical University (grant number: 2022QH1229), Startup Fund for scientific research, Fujian Medical University (grant number: 2022QH1227), Fujian provincial health technology project (grant number: 2024GGB11), Fujian Provincial Department of Science and Technology–Natural Science Foundation of Fujian Province (grant number 2023J011312), and Clinical Key Specialty Construction Project of Fujian Province (Fujian Medical Policy Letter [2023] No. 1163).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The research was conducted in accordance with the Declaration of Helsinki, and the study protocol received ethical approval from the Institutional Review Board of Fujian Provincial Maternity and Children’s Hospital (approval no. 2023KY020). Informed consent was obtained from the mother and infant’s legal guardian before they participated in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.