Neonatal Respiratory Distress Syndrome: Chest X-Ray or Lung Ultrasound? A Systematic Review
Neonatal Respiratory Distress Syndrome: Chest X-Ray or Lung Ultrasound? A Systematic Review
Neonatal Respiratory Distress Syndrome: Chest X-Ray or Lung Ultrasound? A Systematic Review
Ultrasound
2017, Vol. 25(2) 8091
Abstract
Background and aim: Neonatal respiratory distress syndrome is a leading cause of morbidity in preterm new-
born babies (<37 weeks gestation age). The current diagnostic reference standard includes clinical testing
and chest radiography with associated exposure to ionising radiation. The aim of this review was to compare
the diagnostic accuracy of lung ultrasound against the reference standard in symptomatic neonates of
42 weeks gestation age.
Methods: A systematic search of literature published between 1990 and 2016 identified 803 potentially
relevant studies. Six studies met the review inclusion criteria and were retrieved for analysis. Quality assess-
ment was performed before data extraction and meta-analysis.
Results: Four prospective cohort studies and two case control studies included 480 neonates. All studies
were of moderate methodological quality although heterogeneity was evident across the studies. The pooled
sensitivity and specificity of lung ultrasound were 97% (95% confidence interval [CI] 9499%) and 91%
(CI: 8695%) respectively. False positive diagnoses were made in 16 cases due to pneumonia (n 8), transient
tachypnoea (n 3), pneumothorax (n 1) and meconium aspiration syndrome (n 1); the diagnoses of the
remaining three false positive results were not specified. False negatives diagnoses occurred in nine cases,
only two were specified as air-leak syndromes.
Conclusions: Lung ultrasound was highly sensitive for the detection of neonatal respiratory distress
syndrome although there is potential to miss co-morbid air-leak syndromes. Further research into lung
ultrasound diagnostic accuracy for neonatal air-leak syndrome and economic modelling for service integra-
tion is required before lung ultrasound can replace chest radiography as the imaging component of the
reference standard.
Keywords
Neonatal respiratory distress syndrome, lung ultrasound, chest X-ray, diagnosis
1
Department of Radiology, Hull and East Yorkshire NHS Trust,
Ultrasound Department, UK
2
Introduction Division of Biomedical Imaging, Faculty of Medicine & Health,
University of Leeds, UK
Neonatal respiratory distress syndrome (NRDS) is a 3
Division of Epidemiology and Biostatistics, Leeds Institute of
breathing disorder arising at, or shortly after birth Cardiovascular and Metabolic Medicine, University of Leeds, UK
4
(<24 hours); it increases in severity during the rst Leeds Teaching Hospitals NHS Trust, St Jamess University
Hospital, UK
48 hours of life.1 Although full term new-borns with a
gestational age [GA] between 37 and 42 weeks can be Corresponding author:
Matthew Hiles, Department of Radiology, Ultrasound Department,
aected, approximately four out of ve cases occur in Hull and East Yorkshire NHS Trust, Castle Hill Hospital, Castle Rd,
those born prematurely (<37 weeks).2,3 Severity and Cottingham, East Riding of Yorkshire HU16 5JQ, UK.
incidence of NRDS are inversely related to GA with Email: Matthew.hiles2@hey.nhs.uk
Hiles et al. 81
92% of neonates born at 2425 weeks aected, 88% at and specicity of 91% and 84% respectively when radi-
2627 weeks, 76% at 2829 weeks and 57% at 3031 ologists where blinded to clinical test results. Morris17
weeks.4,5 suggests radiological appearances correlate well with
NRDS is caused by physiological and structural pul- clinical disease severity, atelectasis being represented
monary immaturity insucient levels of pulmonary by a bi-lateral ne granular or ground glass appear-
surfactant compromise alveolar integrity, impeding ance such that extent of disease corresponds to level of
normal gas exchange due to deregulation of acinar sur- lung opacity. Reduced lung expansion, dilated bronchi-
face tension.6,7 Resulting atelectasis causes decreased oles and air bronchograms are also visible depending
lung compliance through an increase of collapsed on disease stage.7
alveoli in the terminal airways.8 NRDS progresses Further to diagnostic use, CXR is used to conrm
through hypoventilation, hypoxemia and respiratory endotracheal tube (ETT) position; premature new-
acidosis.68 It is a leading cause of morbidity in prema- borns with severe NRDS frequently receive continuous
ture new-borns and is a common reason for admission positive airways pressure (CPAP) in order to improve
to the neonatal intensive care unit (NICU).9,10 ventilation and oxygenation as well as facilitating intra-
NRDS is diagnosed by a combination of clinical tracheal administration of surfactant.1,6 Conrmation
signs and symptoms, laboratory analysis and chest of the ETT position minimises lung damage caused by
radiography (CXR).1,6 Early diagnosis is important so malpositioning1.
that interventional therapy, respiratory support and Chest radiography involves exposure to ionising
surfactant replacement, can be instigated.7,8 Follow radiation. Neonates, due to their small size and the
up imaging is required to monitor therapeutic eect close proximity of radiosensitive tissues and organs,
and reduce broncho-pulmonary dysplasia as a result are at greater risk from latent eects of CXR in com-
of unnecessary mechanical ventilation.11 parison to other age groups.18 Although the actual risk
of adverse latent eects from neonatal radiation expos-
ure has not been quantied,19,20 the theoretical risk can
Clinical signs and symptoms be predicted using the linear no-threshold (LNT) model
Clinical presentations of NRDS include non-specic with relative risk increasing as absorbed dose increases.20
tachypnoea, nasal aring, cyanosis, substernal and With neonates undergoing multiple CXR examinations
intercostal retraction and grunting from expiratory air during their stay on the NICU, eorts have been made
colliding with a partially closed glottis.8 The Clinical to identify an alternative diagnostic test.21,22
Risk Index for Babies (CRIB) is a risk assessment tool
scoring birth weight, gestational age, maximum and
minimum fraction of inspired oxygen, maximum base
Lung ultrasound
excess during the rst 12 hours of life and presence of In the past, ultrasound has not been widely used for
congenital malformations.12 In suspected NRDS, the neonatal chest imaging due to the obscuring artefact
CRIB can be used to estimate severity of NRDS and generated by normal air-lled lung.21
trigger administration of assisted ventilation.12 Ultrasound does not involve ionising radiation but is
associated with potential risks due to mechanical (iner-
tial cavitation) and thermal tissue damage.23 The risk of
Laboratory tests these adverse bio-eects is low in routine clinical prac-
Arterial partial oxygen pressure (PaO2) levels below tice, but proportional to duration of ultrasound exam-
50 mmHg with cyanosis in room air, or the need for sup- ination, dependent on the specic tissues under
plementary oxygen to maintain PaO2 > 50 mmHg, is indi- examination and the output of the ultrasound trans-
cative of NRDS.6 A blood sample can determine levels of ducer. Risk is quantied in terms of mechanical and
metabolic and respiratory acidosis which indicate anaer- thermal indices, MI and TI respectively and displayed
obic metabolism and atelectasis, respectively.13 during scanning.24 The as low as reasonably practic-
Swallowed lung uid is a signicant constituent of able (ALARP) principle, along with acoustic safety
neonatal gastric aspirate. The gastric aspirate shake test guidelines are implemented to minimise risk.25
(GAST) identies the presence or a lack of surfactant.14 Lung ultrasound (LUS) has recently emerged as a
GAST is reported to have 100% sensitivity and 92% promising diagnostic tool with studies reporting accur-
specicity for NRDS.15 ate results in the diagnosis of NRDS4,9,11,13,2628 and
other neonatal pulmonary diseases.22,29 The presence
of artefact has been recognised as a useful clinical
Chest radiography
marker to demonstrate normality, its absence being
In a study of 59 neonates with clinically suspected indicative of disease (Table 1 and Figure 1(a) to (c)).21
NRDS, Vergine et al.16 found CXR to have sensitivity Raised uid levels in diseased lung and the absence of
82 Ultrasound 25(2)
Pleural line Smooth echogenic appearance < 0.5 mm Absence or disruption of the line, > 0.5 mm
(lung sliding) thick. Visceral and parietal pleura thickness, no sliding.
visualised sliding with respiration.
B3-lines (Figure 1(c)) Absent B-lines closely merged (within 3 mm) create a
white lung appearance through increased
oedema indicative of alveolar interstitial syn-
drome (AIS).
the normal air-lled gap between the pleura and pul- Ultrasonic verication of ETT position in neonates
monary interstitium provide a propagation medium for has also shown potential. Studies have reported close
ultrasound transmission and demonstration of lung correlation between ultrasound and CXR measure-
tissue.4,9 ments and is comparatively much faster.30,31 Due to a
Hiles et al. 83
Neonate (42 weeks) Ultrasound Chest X-ray Neonatal respiratory distress syndrome
neonata, infanta, ultrasoa, sonoga, X-ray, radiographa, Neonatal respiratory distress syndrome,
pediatrica newborna, lung ultrasound conventional radiographa, infantile respiratory distress syndrome,
preterm, premature, plain film, radiologa, hyaline membrane disease, respiratory
babies, baby. computed radiography, distress syndrome, pulmonary surfactant,
digital radiography, lung disease, respiratory disease, surfac-
radiogram, tant deficiency disorder.
roentgenogram.
a
Truncation command.
lack of high quality supporting evidence CXR remains indicators (presentation, vital signs and auscultation),
the gold standard.32 CXR and/or laboratory blood gas analysis. Limited
resources restricted inclusion to studies published in
English. Although this may introduce language bias,33
Aim there is little evidence to suggest that systematic bias
The aim of this review was to compare the diagnostic occurs with such an approach.35 Articles were not
accuracy of LUS against the reference standard clinical excluded on the basis of geographical location or pub-
test and CXR in symptomatic neonates of 42 weeks lication date to limit bias and maximise retrieval of
gestational age. relevant material.33,34
Studies were excluded where it was not possible to
extract sucient data to populate 2 2 contingency
Method tables, obtain them through the local institutional or
British Library, where requisite permission from par-
Search strategy ents and ethical committees had not been obtained or
Studies were identied during August 2016 using the where studies collected non-human or cadaveric data.
following databases: OVID Embase 19962016, OVID After removing duplicate results, study titles,
Medline (R) 19962016, PUBMED 19962016, Science abstracts or full-papers were reviewed to determine
Direct 19952016, Leeds University Librarys Journals/ inclusion in the review. Dierences of opinion were
Books@OVID (full-text), CINAHL 19902016, The resolved by discussion. The reference lists of included
Cochrane Library 20052016 and Google Scholar. studies were examined to identify further relevant stu-
Initial search terms were identied from a prelimin- dies that had not been retrieved by the database search;
ary literature search and accepted by unanimous agree- forward citation tracking was performed in Google
ment amongst review team members. Medical Subject Scholar. The rigorous search and selection process lim-
Headings (MeSH) were used to generate additional ited selection bias and reduced the chance of random
search terms for ultrasound, neonates, X-ray and error.33,34
NRDS (Table 2). The Boolean operators (AND) and
(OR) were used to minimise irrelevant literature and
maximise the breadth of the search.33 Truncation was
Quality assessment
used to increase the yield of studies that used alternate Since the inclusion of studies other than RCTs can
endings to the search terms.34 increase selection and reporting bias,33 quality assess-
Inclusion and exclusion criteria were designed in ment was performed using the QUADAS-2 (Quality
accordance with the Population, Intervention, Assessment of Diagnostic Accuracy Studies 2) tool.36
Comparator, Outcome (PICO) framework to correlate Risk of bias and applicability were assessed in four key
with the research question. To increase validity and areas relevant to the research question: patient selec-
reproducibility they were dened a priori. Studies tion, index test, reference standard and test ow and
were included if they were randomised control trials timing. Three team members individually scored each
(RCTs), cohort or case-control studies, recruited neo- study awarding one point for each criterion where risk
nates 42 weeks GA in a clinical setting with signs and of bias was considered to be low.36
symptoms of NRDS within 48 hours of birth, and had Patient selection was considered to have low risk of
NRDS diagnosed using a combination of clinical bias if there was a consecutive sample of neonates, they
84 Ultrasound 25(2)
Ahuja et al.14 (5) Gastric aspirate Radiologist Transabdominal HDI 3500 [Advanced Diffuse retrodiaphragmatic <20 h Not blinded
test clinical Technologies Laboratories hyperechogenicity completely
diagnosis CXR (ATL) Ultrasound, Bothell, replacing the normal diaphragm
WA, USA] (512 MHz)
curvilinear probe
Bober and CRIB score CXR Physician Transabdominal Siemens SI 450, unknown origin, Retrophrenic hyperechogenicity <24 h Blinded
S
wietlin
ski13 (6) blood results equipped with a sector 5-MHz with B lines diverging radially
transducer
Copetti et al.42 (4) Clinical diagnosis Paediatrician Transthoracic Megas CVX Esaote, Medical Bi-lateral white lung, absence of <24 h Not blinded
CXR Cardiologist Systems, Florence, Italy spared areas, thickened and
(10 MHz Linear Probe) irregular pleural line
Liu et al.10 (6) Clinical diagnosis 1 expert Transthoracic High resolution line probe Consolidation, Pleural line Immediate Blinded
CXR blood results (1112 MHz) (GE Voluson i or E6, Abnormalities and Bilateral
USA) White Lung
Lovrenski4 (5) Clinical Diagnosis Paediatric Transthoracic 7.5 MHz linear probe (Sonoline Consolidation; air 3.244.96 h Not blinded
CXR blood results Radiologist Transabdominal Adara, Siemens, Erlangen, bronchograms and B-Lines
Germany)
Vergine et al.16 (6) Clinical diagnosis Neonatologist Transthoracic Vivid-I Ge Medical Systems, Bi-lateral white lung, coalescent <24 h Blinded
CXR Milan, Italy using a high res B-lines and thickened and
85
The reference standard, care settings and level of LUS have resulted in over-estimation of diagnostic accuracy
expertise were consistently acceptable across all studies. in this subgroup.36 As such we feel the subgroup ana-
Double blinding between the reference and index tests lysis of prospective cohort studies provides the most
occurred in three (50%) of the six studies10,13,16 with accurate reection of test accuracy (sensitivity 96%,
single blinding of the CXR to LUS results occurring in specicity 86%).
the remaining 50%.4,14,42 All studies conducted CXR The transthoracic technique appeared to be superior
rst followed by LUS. Four studies stated the interval to the transabdominal approach for diagnosing NRDS
between LUS and CXR as less than 24 hours but failed because subgroup analysis demonstrated it to have
to provide more precise timing.13,14,16,42 Two studies marginally better sensitivity (99%, 97% respectively)
reported LUS and CXR examinations were performed and better specicity (98%, 82% respectively).
within 5 hours of each other.4,10 All studies used a com- The increased specicity of the transthoracic technique
bination of ultrasound ndings to formulate the diag- would reduce the number of false positive diagnoses
nostic threshold. The four studies using transthoracic and have the clinical benet of reducing unnecessary
scanning diagnosed NRDS on detection of consolida- additional testing or intervention.
tion, pleural line abnormalities and bilateral white Vergine et al.16 measured the diagnostic accuracy of
lung.4,10,16,42 The two studies adopting a transabdom- CXR without the addition of clinical information and
inal approach dened the presence of retro-diaphrag- found a sensitivity of 91% and a specicity of 84%.
matic hyper-echogenicity with > 3 B-lines as indicative Based on these values, LUS appears to be a comparable
of NRDS.13,14 test.
Ahuja et al.14 5
wietlin
Bober and S ski13 6
Copetti et al.42 4
Liu et al.10 6
Lovrenski4 5
Vergine et al.16 6
Low High
Figure 3. Forest plots describing the sensitivity (a) and specificity (b) of LUS for the diagnosis of NRDS.
LUS: lung ultrasound; NRDS: Neonatal respiratory distress syndrome.
although utilisation of a standard approach helps to to maximise positive health outcomes. This presents eco-
limit operator dependency and can improve diagnostic nomic and administrative challenges as LUS would
accuracy.21 require neonatal clinicians to spend time learning a new
If LUS is to be used as a rst line investigation for skill or alternatively, an LUS practitioner would be
NRDS, it must be carried out soon after birth in order required to service the NICU 24 hours a day 7 days a week.
88 Ultrasound 25(2)
27. Sawires HK, Abdel Ghany EA, Hussein NF, et al. Use of undergoing mechanical ventilation. Pediatr Radiol 1996;
lung ultrasound in detection of complications of respira- 26: 463469.
tory distress syndrome. Ultrasound Med Biol 2015; 41: 41. Abdelsadek A, Khair MDA and Naga OA. Lung ultra-
23192325. sound as early diagnostic tool in neonatal respiratory dis-
28. Wang HW, Cao HY, Liu J, et al. The role of lung tress syndrome (RDS). Egypt J Chest Dis Tubercul 2016;
ultrasound in diagnosis of respiratory distress syn- 65: 377382.
drome in new-born infants. Iran J Paediatr 2015; 25: 42. Copetti R, Cattarossi L, Macagno F, et al. Lung ultra-
147154. sound in respiratory distress syndrome: A useful tool for
29. Cattarossi L. Lung ultrasound: Its role in neonatology early diagnosis. Neonatology 2008; 94: 5259.
and paediatrics. Early Hum Dev 2013; 89: S17S19. 43. Liu B, Liu F, Liu Y, et al. Lung ultrasonography for the
30. Dennington D, Vali P, Finer NN, et al. Ultrasound con- diagnosis of severe neonatal pneumonia. Chest 2014; 146:
firmation of endotracheal tube position in neonates. 383388.
Neonatology 2012; 102(3): 185189. 44. Lichtenstein DA, Lascols N, Mezie`re G, et al. Ultrasound
31. Galicinao J, Bush AJ and Godambe SA. Use of bedside diagnosis of alveolar consolidation in the critically ill.
ultrasonography for endotracheal tube placement in pedi- Intens Care Med 2004; 30: 276281.
atric patients: A feasibility study. Pediatrics 2007; 120(6): 45. Copetti R and Cattarossi L. The double lung point: An
12971303. ultrasound sign diagnostic of transient Tachypnea of the
32. Schmolzer GM and Roehr CC. Techniques to ascertain newborn. Neonatology 2007; 91: 203209.
correct endotracheal tube placement in neonates. 46. Shaw N and Kotecha S. Management of infants with
Cochrane Database Syst Rev 2014; 13(9): CD010221. chronic lung disease of prematurity in the United
33. Higgins J and Green S (eds) Cochrane handbook for sys- Kingdom. Early Hum Dev 2005; 81: 165170.
tematic reviews of interventions: Cochrane book series. 47. Lichtenstein DA, Meziere G, Lascols N, et al. Ultrasound
London: John Wiley & Sons, 2008. diagnosis of occult pneumothorax. Crit Care Med 2005;
34. Centre for Reviews and Dissemination. Systematic 33(6): 12311238.
reviews: CRDs guidance for undertaking systematic 48. Singh BS, Clark RH, Powers RJ, et al. Meconium aspir-
reviews in health care, www.york.ac.uk/ (2009, accessed ation syndrome remains a significant problem in the
28 December 2015). NICU: Outcomes and treatment patterns in term neo-
35. Morrison A, Moulton K, Clark M, et al. English-
nates admitted for intensive care during a ten-year
language restriction when conducting systematic
period. J Perinatol 2009; 7: 497503.
review-based meta-analyses: Systematic review of pub-
49. Shahri MM, Hassan Naghibi S, Mahdavi E, et al.
lished studies. Ottawa: Canadian Agency for Drugs and
Diagnostic utility of chest X-rays in neonatal respiratory
Technologies in Health, 2009.
distress: Determining the sensitivity and specificity. Int J
36. Whiting P, Rutjes A, Westwood M, et al. QUADAS-2: A
Paediatr 2014; 2: 6572.
revised tool for the quality assessment of diagnostic
50. Jung AY, Yang I, Go HS, et al. Imaging neonatal spon-
accuracy studies. Ann Intern Med 2011; 155: 529536.
taneous pneumomediastinum using ultrasound. J Med
37. DerSimonian R and Laird N. Meta-analysis in clinical
Ultrason 2013; 41: 4549.
trials. Control Clin Trials 1986; 7: 177188.
51. Albaum MN, Hill LC, Murphy M, et al. Interobserver
38. Meta-DiSc Version 1.4, Windows [Software]. Madrid:
Hospital Ramon y Cajal, 2006. reliability of the chest radiograph in community-acquired
39. Avni EF, Braude P, Pardou A, et al. Hyaline membrane pneumonia. Chest J 1996; 110: 343350.
disease in the newborn: Diagnosis by ultrasound. 52. Lichtenstein DA. Ultrasound examination of the lungs in
Paediatr Radiol 1990; 20: 143146. the intensive care unit. Paediatr Crit Care Med 2009; 10:
40. Avni EF, Cassart M, de Maertelaer V, et al. Sonographic 693698.
prediction of chronic lung disease in the premature