The role of chest X-ray in the diagnosis of neonatal respiratory distress syndrome: a systematic review concerning low-resource birth scenarios

Figures & data

Figure 1. Flowchart with detailed research data for the identified studies for each phase, according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) [12].

Table 1. General characterisation of eligible articles.

Author/year	Income group	Study design	N	Range of birth weight (grams)	Range of GA (weeks)	Target population	Groups of study
Wood, B.P., et al., [15]	HIC	Clinical trial	60	-	25 to 29	Newborns at risk for NRDS	Endotracheal surfactant vs. placebo (NaCl)
Kurl, S., et al., [16]	HIC	Cross-sectional	99	Median: 3000	24 to 41	Newborns admitted to NICU with early respiratory distress	Clinicians diagnosis vs. radiologists diagnosis
Bober, K., et al., [17]	HIC	Cross- sectional	131	Mean ± DP (min to max): 1898 ± 864 (500 to 4400)	Mean ± DP (min to max): 32 ± 4.4 (24 to 42)	Newborns admitted to NICU with signs of respiratory failure	US vs. CXR for NRDS severity classification
Shahramian, I., et al., [18]	LMIC	Case- control	130	Mean ± DP: 2575 ± 791	Mean ± DP: 35.500 ± 2.213	Newborns admitted to NICU with Apgar score more than 7	Preterm newborn vs. full-term newborns
Raimondi, F., et al., [19]	HIC	Cross- sectional	54	Mean ± DP:1703 ± 583	Mean ± DP: 32.5 ± 2.6	Preterm newborns admitted to NICU with moderate respiratory distress and treated with nCPAP	US vs. CXR for NRDS severity classification
Yin, X., et al., 2014 [20]	UMIC	Case- control	83	Mean ± DP: 2945.3 ± 193.3 (NRDS) and 2969.2 ± 247.3 (control)	Mean ± DP: 38.0 ± 0.7(NRDS) and 38.2 ± 0.8(control)	Full-term newborns with NRDS	NRDS vs. health newborns
Tagliaferro, T., et al., [21]	HIC	Cohort	235	Median: 713 CPAP failure and 805 CPAP success	Median (IQR): 26 (25–26) CPAP failure and 27 CPAP success	All inborn ELBW newborns admitted to NICU	Newborns who succeeded nCPAP vs. newborns who failed nCPAP
El-Malah, H.E. et al., [22]	LMIC	Cohort	100	Mean (min to max): 2400 (2100 to 3000)	≥36 (mean 37.86)	Newborn with clinical and radiographic signs of NRDS.	US vs. CXR for NRDS severity classification
Liu, J., et al., [23]	UMIC	Case- control	100	Mean ± DP: 2320 ± 353 (NRDS) and 2297 ± 411 (control)	Mean ± DP: 34.9 ± 2.7 (NRDS) and 35.1 ± 2.8 (controls)	Newborns admitted to NICU	NRDS vs. health newborns
Sawires, H.K., et al., [24]	LMIC	Case- control	130	Mean ± DP: 1384.22 ± 176.46 (group) and 1580.6 ± 204.44 (control)	Mean ± DP: 29.91± 1.33 (group) and 34.22 ± 1.05 (control)	Preterm newborns admitted to NICU	US vs. CXR for NRDS complications
Abdelsadek, A., et al., [25]	LMIC	Cross- sectional	40	Mean (min to max): 1600 (1000 to 2500)	Mean (min to max) 33 (29 to 35)	Preterm newborns admitted to NICU with respiratory failure within 6h of life and birth weight appropriate for GA	NRDS severity classification
Rachuri, H., et al., [26]	LMIC	Cross- sectional	94	Mean ± DP: 1987 ± 669 (group) and 2349 ± 653 (control)	Mean ± DP: 34.5 ± 3.2 (group) and 35.9 ± 2.7 (control)	Newborns admitted to NICU with respiratory distress who had CXR and US within 4h admission	US vs. Gold Standard (clinic-radiological diagnosis)
Perri, A.; et al., [27]	HIC	Cohort	56	Mean ± DP: 1442 ± 520	Mean ± DP: 31 ± 3	Newborn admitted to NICU with respiratory failure within 2h of life and treated with nCPAP.	US vs. CXR to predict surfactant administration
Li, Y.; Lin, L.; Wang, Q., [28]	UMIC	Cross- sectional	150	Mean ± DP: 1: 2120 ± 430 2: 2050 ± 390 3: 2100 ± 460 4: 2010 ± 470	Mean ± DP: 1: 33.75 ± 1.53 2: 32.78 ± 2.31 3: 33.09 ± 1.67 4: 33.41 ± 1.61	Preterm newborn with LBW and NRDS who underwent alveolar lavage therapy	Expression levels of caspase-3 and Bcl-2 vc. CXR in NRDS severity classification
Grimaldi, C., et al., [29]	HIC	Cohort	52	Mean ± DP: 2065 ± 1174	Median (range): 33 (25–41)	Newborns who needed a CXR because of respiratory conditions during the first 24 h of life	US vs. CXR for NRDS diagnosis
Pasic, I.S., et al., [30]	UMIC	Cross- sectional	51	Median (min to max): 1790 (1570 to 2160)	Median: 31	Newborns ≤35 weeks of GA with positive clinical and laboratory signs of impaired respiratory function	US vs. CXR for NRDS diagnosis
Vardar, G., et al., [31]	UMIC	Cohort	45	Median (IQR): 1290 (975–1720)	Median (IQR): 30 (27–32)	Preterm infants < 34 weeks GA with NRDS	US vs. CXR for NRDS severity classification and to predict surfactant
Kayki, G., et al., [32]	UMIC	Cohort	71	Mean ± DP: 1265 ± 415	Mean ± DP: 29.6 ± 2.3	Preterm newborn admitted to NICU with ≤ 32 weeks GA, treated with nCPAP due to respiratory distress.	US vs. CXR for surfactant administration
Aldecoa-Bilbao, V.; et al., [33]	HIC	Cohort	144	Mean ± DP: 1175 ± 314 (No surfactant) and 1066 ± 397 (Surfactant)	23+0 to 31+6	Preterm newborn admitted to NICU with GA between 23+0 and 31+6 who required noninvasive ventilation for NRDS.	US vs. CXR for surfactant administration
Oktem, A., et al., [34]	UMIC	Cohort	40	Mean ± DP (min to max): 1817.12 ± 211 (600 to 3260)	Mean (min to max): 32 weeks ± 4 days (26 +4 to 37)	Preterm newborns admitted to NICU with NRDS and who need surfactant administration	Before vs. after surfactant administration (US vs CXR)
Reza, M.; et al., [35]	LMIC	Cross- sectional	33	Mean: 1230	Mean (min to max): 29.64 (<28 to 34)	Preterm newborns admitted to NICU	US vs. CXR for NRDS diagnosis
Eldeen, S.M.; et al., [36]	HIC	Case- control	177	Mean ± DP: 2800 ± 580	Mean ± DP: 36.8 ±1.65	Newborns ≥ 35 weeks with NRDS and who required any type of respiratory support	<35 weeks newborns with NRDS vs. ≥ 35 weeks newborns with NRDS
Xiao, Y., et al., [37]	UMIC	Cohort	170	Mean ± DP: 1- mild: 2709 ± 124 2- severe: 1660 ± 750	Mean ± DP: 1- mild NRDS: 34.3 ± 5.14 2- severe NRDS: 30.2 ± 3.39	Newborns admitted to NICU with NRDS	Severe NRDS vs. mild NRDS

HIC: High income country. UMIC: Upper middle-income country. LMIC: Lower middle-income country. nCPAP: Nasal continuous positive airway pressure. NRDS: Neonatal Respiratory distress syndrome. GA: Gestational age. NICU: Neonatal intensive care unit. CXR: Chest x-ray. LBW: Low birth weight. ELBW: Extremely low birth weight. US: Ultrasound.

Table 2. Characterisation of CXR use and clinical features associated with NRDS.

Author/year	Time of the first CXR	CXR features associated to NRDS	Clinical features associated to NRDS	Protocol/guideline for NRDS diagnosis	NRDS severity classification	Other use for CXR	Summary of the topic
Wood, B.P., et al., [15]	Within 90 minutes of NICU admission	Reduced lung expansion, reticular densities, air bronchogram, consolidation of lungs.	Mean airway pressure and FiO2 required for adequate ventilation	No	Class 1 to 3, proposed by the authors	Differential diagnosis and surveillance	The radiological data provided support of the surfactant replacement efficacy evaluation.
Kurl, S., et al., [16]	Within first 3 h of life	Homogeneous, opaque infiltration, air bronchogram.	Retractions, tachypnea, supplemental oxygen, acid-base disturbance in arterial blood gas analysis	No	Mild or Moderate (clinical definition)	Differential diagnosis	There was 95% agreement between clinicians and radiologists for NRDS diagnosis. The first CXR taken had the greatest impact in the care of neonates.
Bober, K., et al., [17]	Within first 24 h of life	Not described but based on Hansen T, Corbet A (1991) criteria	CRIB	Hansen and Corbet; 1991	Grade 1 to 4	None	US examination cannot replace CXR as it overestimates the diagnosis of NRDS
Shahramian, I., et al., [18]	Undefined	Granularity of lungs, air bronchogram and white out lungs with loss of cardiac borders	None	No	Normal, Mild, moderate and severe	None	Increase of serum Brain Natriuretic Peptide (BNP) level correlates to NRDS severity (CXR) in premature infants.
Raimondi, F., et al., 2014 [19]	After 120 minutes from the initiation of CPAP	Ground glass shadowing, air bronchograms, alveolar shadowing, white-out of the lung fields with obscuring of the cardiac border	Retractions, nares dilatation, grunting and acid-base disturbance in arterial blood gas analysis	No	Grade 1 to 4 by Arthur (2001)	Surveillance and prediction of CPAP failure	After a 2-hour nCAP trial, US largely outperformed CXR in predicting the need for intubation.
Yin, X., et al., [20]	Undefined	Bilateral transmittance reduced, small mesh-like particle shadow, air bronchogram; blurred heart and diaphragm contours, white lung	Moaning, dyspnea, acid-base disturbance in arterial blood gas analysis, supplemental oxygen and mechanical ventilation	Shao et al., 2011	Grade I to IV by Shao et al., 2011	None	SP-B expressed in bronchoalveolar lavage fluid is inversely related to the severity of NRDS (CXR) in full-term newborn.
Tagliaferro, T., et al., [21]	Within first 2h of life	Reticulogranular infiltrates, air bronchogram, alveolar opacification making the borders of the heart, thymus and diaphragm unclear, lungs quite airless	Acid-base disturbance in arterial blood gas analysis	Kero and, Mäkinen, 1979	Class 1 to 3 by Kero and Mäkinen (1979)	Prediction of CPAP failure	Early radiologic evidence of severe NRDS is predictive of CPAP failure, especially in infants with GA≤26 weeks
El-Malah, H.E. et al., [22]	Within first 24 h of life	Reticulogranular or ground glass opacification, progressive hypo-aeration and air bronchograms.	CRIB	Hansen and Corbet; 1991	Stage I to IV by Hansen T, Corbet A (1991)	Follow up treatment of NRDS	The US can be an alternative diagnostic imaging modality for CXR in follow up treatment of NRDS and subsequent reduction dose of radiation
Liu, J., et al., [23]	After NICU admission	Hypoexpansion, fine granular densities, air bronchogram, ground-glass opacities, blurred cardiac borders, or white lungs	Tachypnea, grunting, retractions, cyanosis, reduced breath sounds, ventilatory support, acid-base disturbance in arterial blood gas analysis	No	Grade 1 to 4 by Ayachi et al. (2005), Bouziri et al. (2007), Faix et al. (1989) and Liu et al. (2010)	None	The US is accurate and reliable to diagnose NRDS. More research is needed to replace CXR.
Sawires, H.K., et al., [24]	Within first 6h of life	Ground glass veiling, bilateral reticulonodular pattern, air bronchogram, bilateral symmetric parenchymal opaqueness	Tachypnea, dyspnea, retraction, grunting or cyanosis, acid-base disturbance in arterial blood gas analysis	No	Grade 1 to 4	Differential diagnosis, surveillance and detection of NRDS complications	US is superior to CXR in the detection of NRDS complications (except for pneumothorax). It could be value in reducing exposure to unnecessary radiation
Abdelsadek, A., et al., [25]	Within first 6 h of life	Hypovolemic lung reticulogranular mottling or without air bronchograms, bilateral opacification of lungs	Down score and laboratory analysis	No	Mild or Severe NRDS	None	US cannot replace standard CXR in diagnosing potential causes of neonatal respiratory failure because of its tendency to over-diagnose NRDS
Rachuri, H., et al., [26]	Within 4 h of NICU admission	Diffuse atelectasis,‘ground glass’ appearance of the lung fields, low lung volume, air bronchograms	Tachypnea, retractions and/or grunting, Downe’s score	No	No	Differential diagnosis and detection of NRDS complications	US can be used to diagnose different etiologies of respiratory distress in neonates. However, CXR is superior to evaluate complications of NRDS.
Perri, A.; et al., [27]	Within 2h from the initiation of CPAP (before surfactant administration)	Reduced radiolucency, air bronchogram, blurred cardiac and diaphragmatic margins, white lung fields	Shallow breathing, grunting, retractions, acid-base disturbance in arterial blood gas analysis	European Consensus Guidelines, 2016	Class 1 to 4 by Cattarossi et al. (2010)	To guide surfactant administration	US predicts the need for surfactant more reliably than CXR
Li, Y.; Lin, L.; Wang, Q., [28]	Within first 4 h of life	Reduced radiolucency, air bronchogram, unclear heart and diaphragmatic surfaces, white lung syndrome	Dyspnea, grunts, and acid-base disturbance in arterial blood gas analysis	Shao et al., 2011	Grades 1 to 4	None	The severity of NRDS in the CXR is positively related with the concentration of caspase-3 in alveolar lavage fluid, and negatively correlated with the expression level of Bcl-2
Grimaldi, C., et al., [29]	Within first 24 h of life	Ground glass shadowing, air bronchograms, confluent alveolar shadowing, and complete white lungs obscuring the cardiac border	Dyspnea, cyanosis, retraction, respiratory support, surfactant therapy	No	Grades 1 to 4 by Agrons et al. (2005), Lobo (2006)	Differential diagnosis and detection of complications of NRDS	The US is superior to CXR for NRDS diagnosis, for differential diagnosis and complications. However, CXR remains necessary for newborns in mechanical ventilation.
Pasic, I.S., et al., [30]	Undefined	Fine homogenous, ground-glass shadowing, air bronchogram, alveolar shadowing obscuring cardiac border	Not clear: clinical signs and acid-base disturbance in arterial blood gas analysis	No	Stage I to IV by Drorbaugh and Fogg (1956)	None	US can be used as complementary modality to CXR, decreasing the number of ionizing radiations in premature newborns
Vardar, G., et al., [31]	Within first 2 h of life	Reticulogranular or ground-glass pattern, air bronchograms, bilateral opacification of lungs with loss of cardiac borders	Not clear: clinical signs, acid-base disturbance in arterial blood gas analysis, FiO2	European Consensus Guidelines, 2016	Stage I to IV by Wambach and Hamvas (2015)	To guide surfactant administration and predict CPAP failure	US is superior to CXR to predict severity of NRDS, surfactant administration and CPAP failure
Kayki, G., et al., [32]	Within first 2 h of life	Reticular appearance, decreased transparency, air bronchogram, white lung with obscuring cardiac border	Not clear: clinical signs and acid-base disturbance in arterial blood gas analysis	European Consensus Guidelines, 2019	Stage I to IV by Perri et al. (2018), Arthur (2001)	To guide surfactant administration	Earlier US (20–30 min) is superior to CXR to predict surfactant administration
Aldecoa-Bilbao, V.; et al., [33]	After NICU admission	Decreased pulmonary expansion, generalized reticulogranular lung opacities and air bronchograms	Shallow breathing, tachypnea, grunting, nasal flaring, and retractions	No	CXR score (0–8 points) adapted from Perri et al. (2018)	To guide surfactant administration	US showed higher sensitivity and predictive values compared with CXR to predict surfactant administration.
Oktem, A., et al., [34]	Before Surfactant administration	Granular pattern of lung parenchyma, air bronchograms and atelectasis	tachypnea, nasal flaring, retractions, or grunting	No	No	Surveillance, to guide surfactant administration, and differential diagnosis	US is superior to CXR in differential diagnosis, and can be used for surveillance without risks of ionization
Reza, M.; et al., [35]	Within first 24 h of life	Lung consolidation, air bronchogram, white-lung appearance.	None	No	According to grades 1 to 4	None	There was an agreement of 63.3% between US and CXR in NRDS diagnosis.
Eldeen, S.M.; et al., [36]	Undefined	Ground glass, air bronchogram, low lung volume	Tachypnea, apnea, reduced breath sounds, cyanosis, surfactant therapy	Local guideline	Mild to severe by Shashidha (2016), Hansen (1991)	None	NRDS among full-term and near-term newborns present mostly mild-to-moderate courses.
Xiao, Y., et al., [37]	Within first 6 h of life	Reticular shadows, white-out appearance, heart shape blurred	None	No	Grade I to IV by Gómez (2020), Hiroyuki (2018)	To guide surfactant, surveillance	US and CXR had the same diagnostic effect on NRDS

CRIB: clinical risk index for babies Score. US: Ultrasound. CPAP: continuous positive airway pressure. CXR: chest x-ray. NRDS: neonatal respiratory distress syndrome.

Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021; doi: 10.1136/bmj.n71

 Web of Science ®Google Scholar

Wood BP, Sinkin RA, Kendig JW, Notter RH, Shapiro DL. Exogenous lung surfactant: effect on radiographic appearance in premature infants. Radiology. 1987;165:11–13. doi: 10.1148/radiology.165.1.3306780

 PubMed Web of Science ®Google Scholar

Kurl S, Heinonen KM, Kiekara O. The first chest radiograph in neonates exhibiting respiratory distress at birth. Clin Pediatr. 1997;36:285–289. doi: 10.1177/000992289703600506

 PubMed Web of Science ®Google Scholar

Bober K, Swietliński J. Diagnostic utility of ultrasonography for respiratory distress syndrome in neonates. Med Sci Monit Int Med J Exp Clin Res. 2006;12: CR440–446. PMID: 17006405.

 Google Scholar

Shahramian I, Noori NM, Sharafi E, Ramezani AA, Hesaraki M. Brain natriuretic peptide: a predictor for severity respiratory distress syndrome in newborns. J Compr Pediatr. 2013;4:189–193. doi: 10.17795/compreped-11262

 Google Scholar

Raimondi F, Migliaro F, Sodano A, Ferrara T, Lama S, Vallone G, et al. Use of neonatal chest ultrasound to predict noninvasive ventilation failure. Pediatrics. 2014;134:e1089–e1094. doi: 10.1542/peds.2013-3924

 PubMed Web of Science ®Google Scholar

Yin X, Xie L, Chai Y, Fan H, Han X, Feng Z. Surfactant protein b expression in bronchoalveolar lavage fluid of full-term neonates with respiratory distress syndrome. Acta Clin Croat. 2014;53: 161–165. PMID: 25163231.

 PubMed Web of Science ®Google Scholar

Tagliaferro T, Bateman D, Ruzal-Shapiro C, Polin RA. Early radiologic evidence of severe respiratory distress syndrome as a predictor of nasal continuous positive airway pressure failure in extremely low birth weight newborns. J Perinatol. 2015;35:99–103. doi: 10.1038/jp.2014.164

 PubMed Web of Science ®Google Scholar

El-Malah H, Hany S, Koriem M, Ali A. Lung ultrasonography in evaluation of neonatal respiratory distress syndrome. Egypt J Radiol Nucl Med. 2015;46:469–474. doi: 10.1016/j.ejrnm.2015.01.005

 Google Scholar

Liu J, Cao HY, Wang HW, Kong XY. The role of lung ultrasound in diagnosis of respiratory distress syndrome in newborn infants. Iran J Pediatr. 2015;25:e323. doi: 10.5812/ijp.323

 PubMed Web of Science ®Google Scholar

Sawires HK, Abdel Ghany EA, Hussein NF, Seif HM. Use of lung ultrasound in detection of complications of respiratory distress syndrome. Ultrasound Med Biol. 2015;41:2319–2325. doi: 10.1016/j.ultrasmedbio.2015.04.024

 PubMed Web of Science ®Google Scholar

Abdelsadek A, Khair MDA, Naga OA. Lung ultrasound as early diagnostic tool in neonatal Respiratory Distress Syndrome (RDS). Egypt J Chest Dis Tuberc. 2016;65:377–382. doi: 10.1016/j.ejcdt.2015.07.006

 Google Scholar

Rachuri H, Oleti TP, Murki S, Subramanian S, Nethagani J. Diagnostic performance of point of care ultrasonography in identifying the etiology of respiratory distress in neonates. Indian J Pediatr. 2017;84:267–270. doi: 10.1007/s12098-016-2288-7

 PubMed Web of Science ®Google Scholar

Perri A, Riccardi R, Iannotta R, Di Molfetta DV, Arena R, Vento G, et al. Lung ultrasonography score versus chest X‐ray score to predict surfactant administration in newborns with respiratory distress syndrome. Pediatr Pulmonol. 2018;53:1231–1236. doi: 10.1002/ppul.24076

 PubMed Web of Science ®Google Scholar

Li Y, Lin L, Wang Q. Correlation of expression levels of caspase-3 and Bcl-2 in alveolar lavage fluid in neonatal respiratory distress syndrome and prognosis. Exp Ther Med. 2018 [cited 2023 October 22]; [5 p.]. doi: 10.3892/etm.2018.5755

 Web of Science ®Google Scholar

Grimaldi C, Michel F, Brévaut-Malaty V, Hassid S, Nicaise C, Puech B, et al. Thoracic ultrasound accuracy for the investigation of initial neonatal respiratory distress. Arch Pédiatrie. 2019;26:459–465.

 PubMed Web of Science ®Google Scholar

Sefic PI, Terzic S, Nisandzic J, Pokrajac D. Lung ultrasound and neonatal respiratory distress syndrome. J Clin Neonatol. 2020;9:272. doi: 10.4103/jcn.JCN_69_20

 Web of Science ®Google Scholar

Vardar G, Karadag N, Karatekin G. The role of lung ultrasound as an early diagnostic tool for need of surfactant therapy in preterm infants with respiratory distress syndrome. Am J Perinatol. 2021;38:1547–1556. doi: 10.1055/s-0040-1714207

 PubMed Web of Science ®Google Scholar

Kayki G, Yigit S, Tandircioglu UA, Celik HT, Yurdakok M. Lung Ultrasound (LUS) and surfactant treatment: looking for the best predictive moment. J Perinatol. 2021;41:1669–1674. doi: 10.1038/s41372-021-01039-0

 PubMed Web of Science ®Google Scholar

Aldecoa-Bilbao V, Balcells-Esponera C, Herranz Barbero A, Borràs-Novell C, Izquierdo Renau M, Iriondo Sanz M, et al. Lung ultrasound for early surfactant treatment: development and validation of a predictive model. Pediatr Pulmonol. 2021;56:433–441. doi: 10.1002/ppul.25216

 PubMed Web of Science ®Google Scholar

Oktem A, Yigit S, Oğuz B, Celik T, Haliloğlu M, Yurdakok M. Accuracy of lung ultrasonography in the diagnosis of respiratory distress syndrome in newborns. J Matern Fetal Neonatal Med. 2021;34:281–286. doi: 10.1080/14767058.2019.1605350

 PubMed Web of Science ®Google Scholar

Reza M, Utomo M, Aden T. Comparison between lung ultrasonography and chest x-ray in the diagnosis of respiratory distress syndrome in preterm neonates in Dr. Soetomo general hospital Surabaya, Indonesia. Curr Pediatr Res. 2021;25: 570–574. ISSN 0971-9032.

 Google Scholar

Mohy Eldeen S, Ali S, Salama H. Clinical characteristics, diagnosis, and management outcome of surfactant deficiency respiratory distress syndrome in term and near-term neonates. A retrospective observational study. Acta Biomed Atenei Parm. 2022;93:e2022337. doi: 10.23750/abm.v93i6.13794

 PubMedGoogle Scholar

Xiao Y, Guo T. Clinical value of lung ultrasound in the diagnosis and follow-up treatment of nrds. Acta Medica Mediterr. 2022;3:1911–1918. doi: 10.19193/0393-6384_2022_3_293

 Google Scholar