12.2.1. Clinical evidence profile

One study (Beer 2014) of 5 children with cerebral palsy was included and reported the accuracy of clinical assessment compared to FEES in detecting aspiration. One study (DeMatteo 2005) with a mixed population of children with various conditions was included as indirect evidence and reported on the accuracy of clinical assessment compared to VF in detecting aspiration. The proportion of children with cerebral palsy was not reported and results for cerebral palsy participants were not stratified.

A modified GRADE approach has been used that allows the includsion of diagnostic outcomes (sensitivity, specificity, predictive values and likelihood ratios) while appraising the evidence for the key GRADE domains (risk of bias, imprecision, indirectness and inconsistency).

For full details, see review protocol in Appendix E. See also the study selection flow chart in Appendix F, modified GRADE profiles in Appendix H, study evidence tables in Appendix J and the exclusion list in Appendix K.

For a summary of the study included, see Table 44.

Table 44

Summary of included studies.

One study (DeMatteo 2005) identified predictors of fluid and solid aspiration and penetration, which are outlined in Table 45, Table 46, Table 47 and Table 48. Confidence and imprecision in the provided relative risks could not be assessed, as confidence intervals were not reported in the study.

Table 45

Predictors of fluid aspiration (DeMatteo 2005).

Table 46

Predictors of fluid penetration (DeMatteo 2005).

Table 47

Predictors of solid aspiration (DeMatteo 2005).

Table 48

Predictors of solid penetration (DeMatteo 2005).

12.4.1. Clinical assessment versus VF for aspiration of fluids

Low-quality evidence from 1 cohort study with 59 participants that used clinical assessment was not accurate for ruling in and moderately accurate (uncertainty unclear) for ruling out aspiration of fluids as defined by VF in a mixed population of children with feeding and swallowing difficulties. Sensitivity was 92% (95% CI:73–99) and specificity was 46% (95% CI:29–63).

12.4.2. Clinical assessment versus VF for aspiration of solids

Very low-quality evidence from 1 cohort study with 32 participants that used clinical assessment was not accurate for ruling in or ruling out aspiration of solids as defined by VF in a mixed population of children with feeding and swallowing difficulties. Sensitivity was 33% (95% CI:4.33–77.7) and specificity was 65% (44.3–82.8).

12.4.3. Clinical assessment versus VF for penetration of fluids

Low-quality evidence from 1 cohort study with 68 participants that used clinical assessment was not accurate in ruling in or ruling out penetration of fluids as defined by VF in a mixed population of children with feeding and swallowing difficulties.

12.4.4. Clinical assessment versus VF for penetration of solids

Very low-quality evidence from 1 cohort study with 68 participants that used clinical assessment was not accurate in ruling in or ruling out penetration of fluids as defined by VF in a mixed population of children with feeding and swallowing difficulties.

12.4.5. Clinical assessment versus FEES for aspiration of saliva

Low-quality evidence from 1 cohort study with 5 participants showed that clinical assessment was not accurate in ruling in or ruling out aspiration of saliva as defined by FEES. Sensitivity was 67% (95% CI:9.4–99.2) and sensitivity was 50% (95% CI:1.7–98.7).

12.4.6. Clinical assessment versus FEES for aspiration of puree

Low-quality evidence from 1 cohort study with 2 participants could not show the usefulness of clinical assessment in ruling in or ruling out of aspiration of puree as there were no false negatives. Sensitivity was 100% (95% CI:15.8–100).

12.4.7. Clinical assessment versus FEES for aspiration of liquids

Low-quality evidence from 1 cohort study with 2 participants could not show the usefulness of clinical assessment in ruling in or ruling out of aspiration of liquids as there were no false negatives. Sensitivity was 100% (95% CI:15.8–100).

12.5.1. Relative value placed on the outcomes considered

The critical outcomes identified for this evidence review were the diagnostic accuracy in identifying the mechanisms underlying eating, drinking and swallowing difficulties and demonstration of aspiration into the airway. No evidence was retrieved for outcomes other than the diagnostic accuracy of presence or absence of aspiration.

12.5.2. Consideration of clinical benefits and harms

An understanding of the underlying mechanisms responsible for eating, drinking or swallowing difficulties may help in devising effective management strategies. Some children and young people are at risk of aspiration of liquids and/or solids and this may lead to significant complications, including apnoea, breathing difficulties and aspiration pneumonia. If there is a serious risk of aspiration, drinking or eating some or all fluids and foods may be unsafe.

The Committee considered and discussed the evidence available and noted that the studies presented did not precisely match the intended evidence review protocol. They had hoped to see evidence on the value of adding either VF or FEES to the normal routine practice of clinical assessment in relation to diagnostic accuracy. The available studies, however, used either VF or FEES as a reference test (the gold standard) and examined the relative risk of penetration (passage of swallowed liquids or solids through the glottis but not beyond the vocal cords) and aspiration (passage beyond the vocal cords) in relation to a range of clinical signs (individually or in combination) used as index tests. The subjects included in both studies were children who had been referred to tertiary centres, having been previously identified at high risk for aspiration, through clinical history and assessment.

Broadly, in keeping with the Committee’s knowledge and experience, cough, altered respiration and colour change were identified as significant clinical events suggesting an increased likelihood of airway penetration of liquid and/or solid food.

The Committee noted that the current practices in the assessment of eating, drinking and swallowing included a clinical assessment based on the history of, and sometimes formal observation during, mealtimes. They recommended that a clinical assessment should be undertaken in every child or young person when there is concern raised about difficulties with eating and drinking. They advised that the history should particularly note any reported coughing, gagging, choking behaviour, alteration in breathing pattern or change in colour (particularly of the face). The risk of ‘silent aspiration’ (where swallow dysfunction is not accompanied by common clinical signs such as coughing) was recognised. Clinical assessment should therefore specifically explore a child or young person’s respiratory history. The Committee considered this clinical assessment should be the routine first-line investigation to identify problems with eating or drinking and to identify possible reasons for concern regarding its safety, and the ability to feed effectively. The Committee discussed various other aspects of an eating, drinking, and swallowing assessment, but decided not to incorporate more detailed advice in the guideline recommendations. The Committee noted that there is a wide variation of what is considered to be a normal time span for eating and drinking. There was general agreement that if mealtimes routinely lasted longer than 30 minutes then further assessment is warranted.

The Committee did recommend that if concerns arose, based on this routine clinical assessment, then the child or young person should undergo regional tertiary specialist assessment based on direct observation by a person with expertise in the assessment of eating or drinking problems, such as a dysphagia-trained speech and language therapist (SLT). They recommended that when concerns existed, this specialist SLT assessment should be undertaken as part of a multidisciplinary review with all members having the necessary expertise in their roles of managing a clinically safe feeding regimen.

The Committee recommended that VF or FEES should not be used as initial assessment. The Committee intended that this recommendation should reduce variation in clinical practice across the UK. Some centres may routinely refer children with suspected difficulties in eating, drinking and swallowing directly for VF and may do so without prior assessment by an expert multidisciplinary feeding team. This approach is supported by the lack of evidence showing that VF provided added value over clinical assessment alone in the wider cerebral palsy population, although evidence did suggest an advantage in a group of children already assessed as high risk.

The Committee and co-opted experts agreed a list of contexts, based on their clinical experience and by consensus, in which the specialist mutidisciplinary team (MDT) should consider undertaking VF. However, it was noted that sufficient training and expertise in the provision and interpretation of VF swallow studies in children with postural and movement difficulties was essential. This strengthened the argument for the involvement of an expert feeding MDT before deciding to use VF in children and young people with cerebral palsy.

The Committee also discussed the usefulness of undertaking VF prior to consideration of enteral tube feeding. The Committee agreed that VF was not always needed in such situations, particularly when there was obvious clinical risk of aspiration, recurring respiratory symptoms, significant nutritional compromise and/or food refusal.

The Committee noted that VF is widely used in UK clinical practice as the investigation of choice for the assessment of eating, drinking and swallowing and, based on their clinical knowledge and experience they were confident in the importance of making recommendations regarding its use. The Committee noted that there was less widespread experience in the use of and, hence, more uncertainty, regarding the clinical usefulness of FEES.

12.5.3. Consideration of economic benefits and harms

The Committee believed that the costs for a VF and FEES taken from NHS Reference Costs were underestimated. Firstly, these procedures would tend to take substantially longer in children and young people with cerebral palsy. Secondly, more healthcare professionals may be present for the procedure.

The Committee noted that FEES is not commonly used in UK clinical practice to assess swallow safety in children and young people with cerebral palsy. Moreover, FEES is an invasive procedure that is not well tolerated in children (with or without cerebral palsy). Combined with the lack of clinical evidence, the Committee felt they were able to justify recommending VF rather than FEES. Consequently, the costs of implementing the Committee’s recommendation in favour of VF are reduced because clinical practice would not be significantly changed.

To prevent unnecessary referrals for VF, the Committee agreed the clinical assessment should be undertaken by healthcare professionals with expertise in eating, drinking and swallowing disorders, including a dysphagia-trained speech and language therapist, to decide if any additional value could be achieved from performing VF, as well as the likelihood of a child or young person being able to comply with the procedure. This may incur training costs as the Committee considered that many referrals for VF come from healthcare professionals who are not trained to assess eating, drinking and swallowing difficulties in children and young people with cerebral palsy. However, they noted that improved training may also reduce costs attached to unnecessary or failed investigations.

Following this, the Committee prioritised a recommendation for VF to be performed in a centre with an MDT that has experience and competence in using VF with children and young people with cerebral palsy. The Committee added that this is not limited to specialist national centres as this would be unachievable with finite resources.

12.5.4. Quality of evidence

Two cohort studies were included in the evidence review. The quality of this evidence ranged from low to very low. One study had a very small sample size, which increased the uncertainty around the comparisons. Both studies included only children referred for investigation because of previously identified risk of aspiration, i.e. referral filter bias and diagnostic suspicion bias.

12.5.5. Other considerations

In clinical practice, VF and FEES provide additional qualitative information to the clinical assessment rather than confirmation or as a pass/fail test for swallow safety. Also, parents and/or carers may reject the results of these investigations as being unrepresentative of the child or young person’s usual eating and drinking. To ensure the results from VF are interpreted accurately, the Committee agreed that VF should be performed by an MDT that has expertise in its use in children and young people with cerebral palsy, rather than merely wherever VF may be available. The Committee believed that VF can be useful in demonstrating to parents the risks attached to oral feeding, and the benefits of certain modifications to their feeding strategy, and especially when enteral tube feeding may be needed. They did not consider, however, that VF was routinely necessary prior to commencing tube feeding and made a recommendation to this effect.

The recommendations related to this evidence review were based on the evidence and the Committee’s clinical experience.

12.5.6. Key conclusions

The Committee concluded that VF is an important adjunct to multidisciplinary, clinical assessment where there is uncertainty about the safety of swallowing or in situations where a child or young person with cerebral palsy is experiencing recurrent chest infections without overt signs of aspiration on eating and drinking. VF should be undertaken by a team with specific expertise in the assessment and management of children and young people with complex neurodisability to ensure appropriate procedures (that match a child or young person’s typical mealtimes, as far as possible), to help manage parent and/or carer anxiety, and to ensure accurate interpretation of results in the context of a detailed history and ongoing monitoring of health-related outcomes, including growth, weight gain and respiratory health. The potential role of FEES in the assessment of swallowing difficulties remains unclear.