Article Text

Download PDFPDF
Can diaphragmatic ultrasound be used to predict extubation failure?
  1. Simon Jackson1,
  2. Julie Richardson2
  1. 1 Paediatric Intensive Care Unit, Royal Belfast Hospital for Sick Children, Belfast, UK
  2. 2 Paediatric Intensive Care Unit, Royal Belfast Children's Hospital, Belfast, UK
  1. Correspondence to Dr Simon Jackson, Paediatric Intensive Care Unit, Royal Belfast Hospital for Sick Children, Belfast BT5 6EA, UK; S.jackson99{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


An 8-year-old boy, with a medical history of congenital myotonic dystrophy (CMD) and scoliosis, was admitted to the paediatric intensive care unit (PICU) with a lower respiratory tract infection. He was intubated and received antibiotics. After a prolonged intubation, he tolerated a spontaneous breathing trial (SBT) and received ‘pre-extubation steroids’. He was extubated to non-invasive ventilation (NIV), but quickly was re-intubated. The reason for extubation failure was determined to be secondary to respiratory muscle insufficiency. Would diaphragmatic ultrasound (DU) assessment of function have predicted extubation failure?

Structured clinical question

In an 8-year-old boy with CMD, can DU (intervention) be used to predict extubation failure (outcome)?


This search was undertaken using PubMed (MeSH) and SumSearch. The descriptive terms of ‘diaphragm’, ‘ultrasound’, ‘mechanical ventilation’ and ‘paediatric’ were combined and generated 56 ‘hits’, of which 6 were relevant. The excluded hits occurred for the following reasons:

  • Duplicates: 4

  • Preterm: 2

  • Not relevant: 38

  • No abstract: 3

  • Not English language: 2

Summary of included studies

A summary of the selected articles is provided in table 1.

Table 1

Articles included in assessment16–21


The decision to extubate should be cautiously and judiciously undertaken. Extubation from positive pressure mechanical ventilation (MV) to negative pressure patient ventilation results in physiological changes for the patient. A recent paediatric study demonstrated 8.3% of mechanically ventilated patients had extubation failure within 48 hours.1 Desirable criteria for extubation include intact airway reflexes, manageable airway secretions, spontaneous ventilation and haemodynamic stability.2 The timing of extubation is imperative, as both premature extubation and prolonged intubation are associated with increased morbidity, mortality and prolonged ICU admission.3 4 The patient groups at risk of extubation failure are those with dysmorphic conditions, age less than 24 months, chronic respiratory or neurological disorders, and requiring endotracheal tube replacement for any reason.4 Current strategies for predicting extubation success include SBTs and extubation readiness tests (leak tests and maximum inspiratory pressure).3 5 The routine strategies to prevent extubation failure include steroids and NIV.3 6 Despite this extubation does fail, and having further methods to assess likelihood of extubation, success would be beneficial. There is emerging evidence that DU can be used to predict extubation success or failure.

The diaphragm, as the principal respiratory muscle, has disordered function in multiple pathologies including post-thoracic, abdominal or cardiac surgery, neurological disorders and myopathies.7 Prolonged MV causes ventilator-induced diaphragmatic dysfunction, with diaphragm atrophy being demonstrable after 18–69 hours of MV.8 9 Adults with diaphragmatic dysfunction have longer ventilation and weaning than those without.10

Adult data, including meta-analyses and systematic reviews, conclude DU can predict extubation success or failure with a sensitivity of 89.3% and specificity of 79.6%, and has excellent intraobserver reproducibility.11–15 The most studied DU method to assess diaphragmatic function is the diaphragm thickening fraction (DTF), while alternatives such as diaphragmatic excursion exist but are less recognised. The DTF measures the diaphragm thickness during a tidal breath, with thickness measurements taken at end-inspiration and end-expiration. These values are inputted into the formula, (DTF=thickness at end-inspiration–thickness at end-expiration/thickness at end-expiration)×100. DiNino et al demonstrated, in adult ICU patients, that a DTF greater than 30% had a sensitivity of 88% and specificity of 71% for predicting extubation success, with a positive predictive value (PPV) of 91% and negative predictive value (NPV) of 63%.15 Adult DU data is not directly applicable to paediatric populations as children have differing respiratory mechanics to adults, with proportionally more diaphragm use, more compliant chest wall and horizontal ribs.

In this review, we analysed six observational studies, which partly or completely, aimed to assess DU ability to predict extubation outcomes in paediatric patients. There was a wide geographical spread of studies located between Europe (one study), Africa (one study) and Asia (four studies). There was a variety of methodological differences between trials, and relatively small sample sizes. The wide variation of extubation success and mortality rates suggest varying case mixes and clinical practice between PICUs.

From the studies reviewed in table 1, DTF was the most widely examined method of diaphragm assessment. In the three out of six studies with a statistically significant outcome of predicting extubation success, the DTF values were 20%, 21% and 23.9%.16–18 In two of these studies, the PPVs were 94% and 86.5%, and NPVs were 56% and 100%, respectively.17 18

DU and more specifically DTF have been widely studied in adult and paediatric populations for estimating diaphragmatic function and extubation readiness. The evidence appears supportive of its use in predicting extubation success or failure. However, there are questions that remain unanswered:

  1. Abdel Rahman et al reported that a DTF of ≥23.175 had a sensitivity of 100% and a specificity of 76.2%, for predicting extubation success, with a PPV of 86.5% and NPV of 100%. However, subgroup analysis in different paediatric age populations showed that DTF was statistically significant in predicting extubation success except in adolescents.18 This is unexpected because in all other age groups, including in adult studies, DU has been demonstrated to be able to used.

  2. Lee et al results suggest that DU after extubation can predict extubation success, but not before extubation.19 This raises questions about timing of DU in relation to extubation, and whether DU post-extubation could be used to predict potential reintubation and interventions such as NIV.

  3. DTF appears to have the best evidence base for assessing extubation readiness. Other studies have assessed variables amenable to ultrasound investigation, including expiratory respiratory muscle atrophy and diaphragm excursion, with varying outcomes. Ijland et al, with very small sample sizes, demonstrated a significant decrease in expiratory muscle thickness in those failing extubation.20 Evidence for other variables to assess may be available in the future.

  4. The exclusion criteria of these trials prevent those with chronic respiratory or neurological disorders (high-risk populations for extubation failure) being study participants. This limits the applicability of results to high-risk populations.

  5. Further knowledge is required about the impact of other variables and their inter-relationship to diaphragm function, including airway steroids and sedation.

  6. When the evidence base for predicting extubation failure is further developed, the next step is what to do with this information. Do we proceed to a tracheostomy or provide further time on supportive ventilation?

Clinical bottom line

  • Diaphragm thickening fraction (DTF) measurement, prior to extubation, has the best evidence base to support its usage as an indicator of readiness for extubation (Grade B).

  • Use of DTF in clinical practice remains challenging as cut-off values are yet to be established (Grade C).

Ethics statements

Patient consent for publication



  • Twitter @julesTVA

  • Contributors SJ wrote the article. SJ and JR conceived the article. JR reviewed the article.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles