NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
National Guideline Alliance (UK). Cerebral palsy in under 25s: assessment and management. London: National Institute for Health and Care Excellence (NICE); 2017 Jan. (NICE Guideline, No. 62.)
Review question: Does MRI undertaken at the following ages: before 1 month (corrected for gestation); 1 month to 2 years; and 2 to 4 years; help to predict the prognosis of children and young people with cerebral palsy?
9.1. Introduction
Current clinical practice varies, with MRI being done in some neonatal units as part of the monitoring of treatment and recovery from neonatal encephalopathy or intracranial haemorrhage. However, only a few units have the capability to do this and transferring a sick, ventilated baby to another unit for an MRI scan is not without risk.
Interpretation of MRI in a sick neonate is difficult as, at that age, the brain contains a lot of water and the images do not show the same clear distinction between different parts of the brain as seen in older brains.
MRI may also be done between 1 month and 2 years, either because the child has been diagnosed as having cerebral palsy or after follow-up of neonatal difficulties. The distinction between the different parts of the brain is becoming clearer by this age.
The argument for delaying MRI until after the age of 2 years is based on brain development. An important part of development of the white matter of the brain – myelination – continues throughout childhood, with the majority occurring by 2 years. White matter growth and development is important in cerebral palsy and associated comorbidities such as impairments to vision, language and learning. Development of the deep grey matter structures and basal ganglia occurs at a similar stage, which is particularly important in considering the prognosis in dystonic forms of cerebral palsy.
The Committee acknowledged the desire of parents to know prognosis for their child early to allow for planning of potential intervention and multidisciplinary management but also recognises that an early scan may not be sufficiently specific to give prognosis. A scan at a later date may not give more information on prognosis than is apparent for the progress that the child has made developmentally in the intervening period. The later scan will involve sedation or general anaesthetic for the child and the small risk and costs of this need to be balanced against additional information on prognosis obtained from the MRI.
The aim of this review is to analyse what is the best age to predict the severity of functional impairment in motor and other developmental skills in children and young people with cerebral palsy by using MRI findings classified according to the type of brain injury. An early and accurate prognosis allows for planning and initiation of therapies that improve prognostic outcomes.
9.2. Description of clinical evidence
One cohort study was included in this review (Van Kooij 2010).
The study cohort consisted of 80 full-term children who had development of:
- mild neonatal encephalopathy (n=34, including 2 children with cerebral palsy), or;
- moderate neonatal encephalopathy (n=46, including 9 with cerebral palsy), on the basis of the highest Sarnat score as assessed during the first week after birth.
Neonatal and childhood MRI were analysed for the 80 participating children with neonatal encephalopathy, and for 51 control subjects during childhood. Neonatal and childhood MRIs were compared with regard to site and pattern of injury. To assess the relationship between neurodevelopment and MRI findings, the MRI findings were categorised in 3 grades: no injury, mild injury and moderate to severe injury.
The following neurodevelopmental outcomes were considered:
- Motor function, assessed with the Movement Assessment Battery for Children – Second edition (MABC-2), band 3. A total impairment score (TIS) ≤15th percentile was classified as ‘abnormal’.
- Intelligence quotient (IQ) ≤85 was classified as ‘abnormal’.
- Other disabilities, classified as no disabilities, cerebral palsy (level I to V according to GMFCS) diagnosed between 3 and 5 years of age, post-neonatal epilepsy, and need for special education.
For full details, see the review protocol in Appendix D. See also the study selection flow chart in Appendix F, study evidence tables in Appendix J and the exclusion list in Appendix K.
9.3. Economic evidence
No economic evaluations of MRI scans in children and young people with cerebral palsy were identified in the literature search conducted for this guideline. Full details of the search and economic article selection flow chart can be found in Appendix E and Appendix F, respectively.
The clinical evidence base to identify the best age to predict the progression of cerebral palsy was limited. Because of the lack of evidence on the effectiveness of MRI scans, economic considerations were restricted to a description of the costs.
Table 33 below presents the cost of MRI scans, taken from NHS Reference Costs 2015. It is important to note that the national average unit cost presented in Table 33 is likely to be underestimated for scans done in children and young people with cerebral palsy. This is because many patients would need a general anaesthetic, and the procedure would take longer than average to perform.
Table 33
Cost of MRI scans.
MRI scans in additional to a clinical assessment would not be considered cost effective if there is not an effective treatment for the condition being diagnosed, or if the patient’s management is not changed by the results of the scan. In other words, if MRI scans do not add any additional information to a clinical assessment and do not change the patient’s management strategy, MRI scans should not be recommended.
In the US it is recommended that all children and young people with cerebral palsy should receive an MRI scan, generating similar expectations in many UK patients. Knowing the likely aetiology of their child’s cerebral palsy may reduce a parent’s anxiety and distress, but if the findings from the scan would not change the patient’s prognosis or management strategy, an MRI in the presence of a clear history and clinical assessment would not necessarily be considered cost effective.
Cost data for MRI scans have little use without associated benefits. Therefore, while the costs of MRI scans could be significant, without knowing the benefits of MRI scans we cannot know if they will be cost effective. Recommendations on the population identified to need MRI scans, and the frequency of scans, will have significant resource implications. Therefore, a research recommendation to consider the effect of MRI scans, in addition to a clinical assessment, preferably at different frequencies, would benefit from health economic input to assess the cost effectiveness of providing an additional intervention to the clinical assessment.
9.4. Evidence statements
9.4.1. Motor function
One study with 80 children showed that all children with moderate/severe lesions on neonatal MRI and 61.5% children with normal/mild lesions on neonatal MRI had a TIS ≤15th percentile (p value=0.021). When looking at childhood MRI results, the study showed that all children with moderate/severe lesions and 47.1% children with normal/mild lesions on neonatal MRI had a TIS ≤15th percentile (p-value<0.001).
9.4.2. Intelligence quotient
One study with 80 children showed that 66.7% children with moderate/severe lesions on neonatal MRI and 23.1% children with normal/mild lesions on neonatal MRI had an IQ ≤85 (p value=0.013). When looking at childhood MRI results, the study showed that 71.4% children with moderate/severe lesions and 21.8% children with normal/mild lesions on neonatal MRI had an IQ ≤85 (p-value<0.001).
9.4.3. Cerebral palsy
One study with 80 children showed that 47.6% children with moderate/severe lesions on neonatal MRI and none of the children with normal/mild lesions on neonatal MRI had cerebral palsy (p value=0.003). When looking at childhood MRI results, the study showed that 36.4% children with moderate/severe lesions and 5.5% children with normal/mild lesions on neonatal MRI had cerebral palsy (p-value<0.001).
9.4.4. Epilepsy
One study with 80 children showed that 33.3% children with moderate/severe lesions on neonatal MRI and none of the children with normal/mild lesions on neonatal MRI had epilepsy (p value=0.019). When looking at childhood MRI results, the study showed that 36.4% children with moderate/severe lesions and none of the children with normal/mild lesions on neonatal MRI had epilepsy (p-value<0.001).
9.4.5. Special education
One study with 80 children showed that 42.9% children with moderate/severe lesions on neonatal MRI and 15.4% children with normal/mild lesions on neonatal MRI needed special education (p value=0.096). When looking at childhood MRI results, the study showed that 50% children with moderate/severe lesions and 9.1% children with normal/mild lesions on neonatal MRI needed special education (p-value<0.001).
9.5. Evidence to recommendations
9.5.1. Relative value placed on the outcomes considered
The aim of this review was to analyse what is the best age to predict the progression of cerebral palsy using MRI findings classified according to the type of brain injury. The Committee’s view was that an early and accurate prognosis allows for planning and initiation of therapies that improve prognostic outcomes. The Committee prioritised the following outcomes for this evidence review:
- proportion of children and young people with epilepsy
- proportion of children and young people with feeding problems
- severity of functional disability using Gross Motor Function System Classification (GMFSC)
- the Manual Ability Classification System (MACS)
- communication problems
- cognitive problems
- changes in health-related QoL (for example, Lifestyle Assessment Questionnaire – Cerebral Palsy [LAQ-CP])
- mortality.
9.5.2. Consideration of clinical benefits and harms
The Committee noted the lack of evidence for this review and was not aware of any other relevant studies that should have been included. However, they acknowledged that there were many studies looking at other aspects of the use of MRI in cerebral palsy, such as comparisons between MRI changes in different types of cerebral palsy, abnormalities that predict cerebral palsy, MRI changes in infants exposed to different risk factors, and follow-up of infants exposed to treatment for brain injury, for example, therapeutic hypothermia-cooling. In the absence of a clear evidence base on prognosis derived from neuroimaging, the recommendations developed from this evidence review were mainly based on expert opinion and the clinical experience of the Committee and were agreed by consensus.
The Committee considered as part of their clinical experience that some of the features on MRI (causation/aetiology) correlate with functional outcome, particularly regarding motor patterns and presence of developmental comorbidity such as sensory, hearing or visual impairment. However, the Committee did not feel confident to recommend the use of MRI solely to guide prognosis in cerebral palsy.
The Committee agreed that prognosis should not be discussed if the aetiology of cerebral palsy in the first instance is not clear. However, it discussed how a good understanding of MRI findings can help to explain to parents the likelihood of severity and of future outcomes. The Committee recognised the importance of involving families and/or carers in the discussion about prognosis, as it can help them to understand and look out for possible signs of associated disorders.
With regard to the best timing for MRI, the Committee agreed that the developmental and maturational processes of the brain means that the radiological signs observed in some individual’s scans can change over time. Therefore, the Committee agreed there is less value in conducting them too early, for example, as the myelination process in the brain is usually mostly complete at 2 years of age.
Based on all the above points, the Committee decided therefore to recommend that healthcare professionals should take into account findings from MRI scans alongside the likely cause of cerebral palsy when discussing prognosis with the child or young person and their parents and/or carers, and to not rely on MRI scans alone but rather to use it as part of a decision pathway also based on history, clinical and developmental assessment They also agreed that many other variables, such as the intervention received and family environment, can impact on the prognosis of the condition.
9.5.3. Consideration of economic benefits and harms
The Committee highlighted that although the causative brain injury is static in cerebral palsy, the findings from MRI scans would not be wholly informative until the brain had developed. For this reason, the Committee agreed doing MRI scans in neonates and infants would not be as cost effective a use of NHS resources as those done after 2 years of age.
The Committee considered they did not have a strong evidence base to recommend MRI in informing prognosis in cerebral palsy as it was unclear if an MRI alone would lead to a change in the person’s management without clear clinical, functional and developmental parameters.
9.5.4. Quality of evidence
One cohort study was included in the review. The quality of the evidence was rated as very low based on the prognostic study methodology checklist (NICE Manual 2012). Main reasons of bias were: the study sample did not fully represent the population of interest with regard to key characteristics, sufficient to limit potential bias to the results; and important potential confounders were not appropriately accounted for, limiting potential bias with respect to the prognostic factor of interest.
9.5.5. Other considerations
The recommendations related to this evidence review were based on the evidence and the Committee’s clinical experience.
9.5.6. Key conclusions
The Committee concluded that MRI alone should not be used for predicting prognosis in infants and children with cerebral palsy.
9.6. Recommendations
- 30.
Do not rely on MRI alone for predicting prognosis in children with cerebral palsy.
- 31.
Take account of the likely cause of cerebral palsy and the findings from MRI (if performed) when discussing prognosis with the child or young person and their parents or carers.
9.7. Research recommendations
None identified for this topic.
- MRI and prognosis of cerebral palsy - Cerebral palsy in under 25s: assessment an...MRI and prognosis of cerebral palsy - Cerebral palsy in under 25s: assessment and management
Your browsing activity is empty.
Activity recording is turned off.
See more...