• 
• 

Three Part Question

In [children] presenting with suspected pneumonia), can [bedside lung ultrasound] be used to diagnose [pneumonia]?

Clinical Scenario

A 4 year-old child presents to your local ED with respiratory symptoms and fever. In order to confirm your suspicion of pneumonia, you plan to order a chest radiograph, but a quick look into the child's medical record shows he has already undergone several X-rays in the last few years for the evaluation of upper respiratory tract infections. Being aware of the potential long-term effects of radiation on your patient, you wonder if bedside lung ultrasound could be used to diagnose pneumonia.

Search Strategy

No relevant ongoing trials were found to be registered on the website clinicaltrials.gov.

The BestBETs database was checked but no relevant BETs were found. MEDLINE 1996–4/2016 ((lung) AND (echography OR ultrasonography OR ultrasound) AND (children OR paediatric) AND pneumonia) found 94 papers Thirteen relevant papers were found, one of which was a systematic review. This included eight of the papers found in the search.

D. EMBASE:

1) children OR pediatric: 1 704 799 papers;

2) echography OR ultrasonography OR ultrasound: 576 651 papers;

3) lung: 1 333 641 papers;

4) pneumonia: 229 187 papers;

5) #3 AND #2: 31 844 papers;

6) #1 AND #2 AND #4: 359 papers;

7) Set limits to human and French or English language: 329 papers.

Thirteen relevant papers were found with no new relevant papers.

There were no Cochrane reviews on the subject.

One new relevant paper published after initial BET submission was added to the topic after subsequent literature review.

Search Outcome

Relevant papers about the subject = 14 of which 1 systematic review included 8 of the papers, leaving 6 relevant papers for analysis. Papers available only as meeting or conference abstracts were excluded.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Ianniello et al. 2016 Italy	84 consecutive children presenting to the ED with cough and fever.	Retrospective cohort study	Number of positive LUS and CR.	60 LUS positive, 47 CR positive. Only one negative LUS case explained by the retroscapular location on the consolidation. In the 13 LUS positive and CR negative cases, clinical evolution was consistent with pneumonia.	High pneumonia prevalence. LUS operators were not said to be blinded to the CR results. No info on LUS operators/training.
Urbankowska et al. 2015 Poland	106 children referred to the hospital with suspected CAP. Pneumonia diagnosis was confirmed in 76 patients by at least 2 clinical signs AND radiographic criterion.	Prospective cohort study	Sensitivity	93,4% (5 False-negative LUS were peri-hilar consolidations.)	Only one pediatric sonographer. High prevalence population with patients referred to the hospital. Gold standard used was a single CR/clinical evaluation combo without clinical evolution.
			Specificity	100%
Guerra et al. 2016 Italy	222 children presenting to the ED with fever and respiratory distress, of which 214 had pneumonia detected on either modality. Pneumonia diagnosis by LUS was compared to CR without gold-standard comparison.	Prospective cohort study	Number of positive LUS	207	Very high prevalence population. LUS performed by 3 pediatricians. Also, 101 did not undergo lateral chest radiograph and only a PA view was performed as recommended by the British Thoracic Society. Lack of gold standard confirmation of discordant cases.
			Number of positive CR	197
			Others	All 17 CR negative consolidations were located in retrocardiac and subdiaphragmatic locations which are known to be missed by PA views.
Pereda et al. 2105 USA	765 children (0-17 years old) from 8 studies. Gold standard used were CXR, mostly correlated with clinical evaluation, with chest CT also used in one study to confirm diagnosis in case of discrepancies between CR and LUS.	Meta-analysis of prospective trials	Pooled sensitivity	96%	Bedside LUS was performed by physicians with varying levels of ultrasound training, from residents, to emergency physician and pulmonologists. Results were blinded for 7/8 studies. Also, only three of the studies were conducted on patients presenting to the ED. There was some variation on the scan techniques used for ultrasound with the Copetti protocol used in three studies. Heterogeneity of populations with 2 studies focusing on critically ill neonates, resulting in possibly higher LUS performance. One study showing slightly inferior results used only CR as gold-standard.
			Pooled specificity	93%
			Pooled positive LR	15,3
			Pooled negative LR	0,06
			Ohter	In the Copetti et al. study, 4 cases of false negative CR pneumonias were identified on both LUS and CT.
Iorio et al. 2015 Italy	52 children with respiratory signs and symptoms. LUS and CR were compared to a gold-standard of pneumonia diagnosis (29 total) by pediatricians based on CR and clinical evolution.	Retrospective cohort study	Sensitivity	LUS 96,5% vs CR 86,2%	Only one operator. Admitted patients. PA CR only.
			Specificity	LUS 95,6% vs CR 95,6%
			Positivie LR	LUS 22,2 vs CR 19,8
			Negative LR	LUS 0,04 vs CR 0,14
			Others	The 4 cases of false negative CR identified on LUS were either located in the retrocardiac and diaphragmatic areas or early presentations.
Ho et al. 2014 Taiwan	163 patients admitted to the hospital with pneumonia as defined by pediatricians final diagnosis according to BTS guidelines’ clinical criteria.	Retrospective cohort study	Pneumonia detection by LUS	97,5%	Admitted patients. LUS performed by a single pediatrician pulmonologist. Delays up to 48h between CR and LUS. Gold-standard using only clinical criteria.
			Pneumonia detection by CR	92,6%
Jones et al, 2016 USA	One hundred and ninety-one patients 0–21 years old, with suspected pneumonia in ED. The intervention group underwent LUS with the option of completing the investigation with a CXR in case of diagnosis uncertainty, compared with a control group of CXR followed by LUS.	RCT	Rate of CXR reduction	38.8% reduction of CXR use in the LUS first arm (95% CI 30–48.9%).	Inclusion of older ‘paediatric’ population (up to 21 years). Potential of missed pneumonia cases presenting in another facility on a repeat visit. Significantly higher rate of pneumonia diagnosis could be due to either false-positives or overdiagnosis of viral syndromes, since there was no difference in clinical outcomes. Unclear why patients dropped out after screening. Uneven distribution of patients.
			Missed pneumonia diagnosed on repeat visit.	None in either groups
			Rate of pneumonia diagnosis.	28.2% in intervention group (103 patients) vs 13.6% in standard group (88 patients) (p= 0.02).

Comment(s)

The usefulness of lung ultrasound for the diagnosis of pneumonia in children is one of the most talked about subjects in the ultrasound literature. Two further positive studies were found in the literature but were excluded as they have only been published in abstract form at this time.9 ,10 A consistent area of difficulty in these studies is the lack of a universal gold standard; in the absence of a CT scan, for obvious reasons, the addition of clinical evolution to confirm pneumonia cases was used in most studies as part of the reference standard. However, it remains unclear if more pneumonia diagnoses lead to better clinical outcomes as many cases are probably of viral aetiology. The included studies incorporated different imaging protocols and this may account for some of the variability in the diagnostic utility. The most widely used scanning technique was based on Copetti et al's work5 and recommended a two-point scan in three different zones of the child's chest (anterior, lateral, posterior). Most studies used either microconvex or linear high frequency probes. The minimum level of training adequate for the use of chest ultrasound also remains unknown, and there was variability in the experience of the sonographers in the included studies. The included papers used similar definition for LUS pneumonias with consolidation, dynamic air bronchograms, subpleural irregularities, focal B-lines and pleural effusions being the main signs. These signs have been reported in adults with variable non-infectious lung conditions such as chronic obstructive pulmonary disease, pulmonary oedema and lung contusions, but the rarity of these conditions in children should mean that these findings are more specific for pneumonia in the paediatric population. It should also be noted that false negatives on lung ultrasound were mainly located at the apex, lower left lobe or subscapular areas that are considered harder to reach by ultrasound. Also a minority of consolidations might not reach the pleural, making them undetectable by ultrasound.11

Editor Comment

BTS; British Thoracic Society; CAP, community acquired pneumonia; CXR, chest radiograph LR, likelihood ratio; RCT,randomised controlled trial.

Clinical Bottom Line

Recent evidence suggests that lung ultrasound can be considered as an accurate diagnostic modality and an adequate alternative to chest radiograph for pneumonia evaluation in children.

References

1. Iannello S. et al. First-Line Diagnosis of Pediatric Pneumonia in Emergency: Lung Ultrasound (LUS) in Addiction to Chest-XRay (CXR) and its Role in Follow-Up. Br J Radiol. 2016 Jan 22:20150998. [Epub ahead of print]
2. Urbankowska E. et al. Lung ultrasound in the diagnosis and monitoring of community acquired pneumonia in children. Respiratory Medicine. 2015 109:9 (1207-1212).
3. Guerra M. et al. Ultrasound detection of pneumonia in febrile children with respiratory distress: a prospective study. European Journal of Pediatrics 2016 Feb; ;175(2):163-70.
4. Pereda M.A et al. Lung ultrasound for the diagnosis of pneumonia in children: A meta-analysis. Pediatrics 2015; 135:4 (714-722).
5. Iorio G. et al. Lung ultrasound in the diagnosis of pneumonia in children: Proposal for a new diagnostic algorithm. PeerJ. 2015 :11 Article Number e1374.
6. Ho MC et al. Usefulness of lung ultrasound in the diagnosis of community-acquired pneumonia in children. Pediatr Neonatol. 2015 Feb;56(1):40-5.
7. Jones BP , Tay ET , Elikashvili I , et al . Feasibility and safety of substituting lung ultrasound for chest X-ray when diagnosing pneumonia in children: a randomized controlled trial. Chest 2016:150(1):131-8.
8. Copetti R , Cattarossi L , Macagno F , et al . Ultrasound diagnosis of pneumonia in children. Radiol Med 2008:11;190–8.
9. Berce V , Podgoršek N , Pintaricet A , et al . Sensitivity of chest ultrasound in the detection of pneumonia in children. Eur Respir J 2015:46 (Suppl 59).
10. Ellington LE , Gilman R , Chavez M , et al . Lung ultrasound for the point-of-care diagnosis of pediatric pneumonia in low resource settings. American Journal of Respiratory and Critical Care Medicine. ATS International Conference published in AJRCCM. C95. BEST OF PEDIATRICS, 2015, p. A5130.
11. Lichtenstein DA , Mezier GA . Relevance of lung ultrasound in the diagnosis of acute respiratory failure: the BLUE protocol. Chest 2008;134;117–25.