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 1: What are the key clinical and developmental manifestations of cerebral palsy at first presentation?
Review question 2: What are the best tools to identify clinical and developmental manifestations of cerebral palsy at first presentation?
6.1. Introduction
The diagnosis of cerebral palsy is often made over a period of time, based on sequential clinical observations and assessments of movement and posture, associated with activity limitation. In clinical practice, the diagnosis of cerebral palsy is typically based on observations and parental reports on the attainment and quality of motor milestones, such as sitting, pulling to stand, walking, feeding and evaluation of posture, deep tendon reflexes and muscle tone.
Infants with risk factors are monitored and watched for developing possible signs of cerebral palsy. Infants without risk factors may present with signs and symptoms noticed by parents or during routine baby surveillance. Some signs are visible in the neonatal period, while others evolve as the infant develops. The time taken between the original suspicion of developmental problems and actual diagnosis can be frustrating for families. Early intervention should be based on the child’s need and not dependant on diagnosis but it is vitally important to give the family an accurate diagnosis and this can take time.
Early signs and symptoms, particularly among preterm children, can be transient and may not result in long-term impairment. Not all signs are visible at birth and may evolve and become more obvious as babies develop. Some of these symptoms are not specific for cerebral palsy.
The Committee hence looked for reliable, objective and valid tools that could be used when an infant first presents to predict those who are likely to develop cerebral palsy and those where the likelihood of developing cerebral palsy is low.
The objectives of this review were to determine the key clinical and developmental manifestations of cerebral palsy and to assess the tools that can assist health professionals (community, primary or secondary) to recognise children with cerebral palsy.
6.2. Description of clinical evidence
A list of clinical and developmental manifestations, including features that are commonly observed in clinical practice, was compiled by the Committee. Two relevant age subgroups were identified: infants below 8 months and infants and children above 8 months. The Committee recognised that routine developmental screening in the UK utilises the lack of independent sitting at 8 months as a sign of abnormal motor development. Therefore, before 8 months, it is more difficult to use delay in motor development as a clue for evolving cerebral palsy and so you need to look for more subtle signs.
In these review questions, the study design prioritised was a prospective cohort. The quality of cohort designs were classed as high quality and downgraded according to the adapted GRADE method.
A total of 18 studies with a total of n=8,239 participants were included in this review. Studies were carried out in Norway (Adde 2007), USA (Allen & Alexander 1992, 1994, Morgan & Aldag 1996), South Africa (Burger 2011), India (Chaudhari 2010), Italy (Brogna 2013, Ferrari 2002), UK (Johnson 1990), Slovenia (Seme-Ciglenecki 2003), Australia (Morgan 2016, Spittle 2013), Zimbabwe (Wolf 1997) and 4 studies were from the Netherlands (Bouwstra 2010, Bruggink 2008, 2009, Groen 2005, Heineman 2011). Seventeen studies included had a prospective cohort study design, in which an index test to measure clinical and/or developmental manifestations was carried out at baseline and a reference test to diagnose cerebral palsy was carried out at follow-up. One study (Allen & Alexander 1992, 1994) was a case control design that used population norms as a control group.
Sixteen studies (Adde 2007, Allen & Alexander 1992, 1994, Boustra 2010, Brogna 2013, Bruggink 2008, 2009, Burger 2011, Ferrari 2002, Heineman 2011, Johnson 1990, Morgan & Aldag 1996, Morgan 2016, Seme-Ciglenecki 2003, Spittle 2013, Wolf 1997) provided diagnostic accuracy measures, including: sensitivity, specificity, positive predictive value, negative predictive value and area under the receiver operating characteristic (ROC) curve, referred to as area under the curve (AUC) throughout this review. The remaining studies provided associations between the manifestation at presentation and diagnosis at follow-up.
There were two studies that looked at using tools to identify clinical and developmental manifestations of cerebral palsy (Morgan & Aldag 1996, Spittle 2013). The tools investigated were the Early Motor Pattern Profile and the Bayley Scales of Infant and Toddler Development.
For full details see review protocol in Appendix D. Evidence are summarised in the clinical GRADE evidence profile in Appendix H. See also the study selection flow chart in Appendix F, study evidence tables in Appendix J and the exclusion list in Appendix K.
6.3. Clinical evidence profile
The following is an overview of the diagnostic accuracy outcomes presented in the modified GRADE tables:
True positive:
The patient has the disease and the test is positive.
Sensitivity:
Probability of being test positive when disease present. Calculated:
= true positive/(true positive + false negative)
Specificity:
Probability of being test negative when disease absent. Calculated:
= true negative/(true negative + false positive)
PPV:
Probability of patient having disease when test is positive. Calculated:
= true positive/(true positive + false positive)
NPV:
Probability of patient not having disease when test is negative. Calculated:
= true negative/(false negative + true negative)
AUC:
A graphical plot of true positive rate (sensitivity) against false positive rate (1 − specificity)
The following criteria was used to define the diagnostic accuracy outcomes:
Sensitivity and Specificity:
- high – 90% and above
- moderate – 75% to 89.9%
- low – 74.9% or below
PPV:
- high – 75% and above
- low – below 75%
NPV:
- high – 70% and above
- low – below 70%
AUC – the classifications of area under the ROC curve (AUC) are as follows (Cook 2008):
- ≥ 0.900 = excellent discriminative ability
- 0.800–0.899 = good discriminative ability
- 0.700–0.799 = fair discriminative ability
- 0.501–0.699 = poor discriminative ability
- 0.000–0.500 = no discriminative ability.
These values have been used in previous NICE guidelines. The Committee was presented with these thresholds and they were comfortable with using them. The specific uses of a diagnostic test and which measures were to be of most interest (for example. for rule in/rule out) were discussed with the Committee and recommendations were made accordingly.
Please see all GRADE tables in Appendix H.
6.4. Economic evidence
This review question is not relevant for economic analysis because it does not involve a decision between alternative courses of action.
No economic evaluations of the key clinical and developmental manifestations that are predictive of 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.
6.5. Evidence statements
6.5.1. Clinical manifestations
6.5.1.1. Abnormality of movement
High-quality evidence was obtained for 1 study with n=187 participants, which used the General Movement Assessment (GMA) to assess the quality of fidgety movements at 12 to 20 weeks post-term in high-risk infants. Forty-eight high-risk infants had absent fidgety (high risk for cerebral palsy), resulting in high diagnostic accuracy of this method in predicting cerebral palsy (above 90%) for sensitivity and specificity.
Moderate-quality evidence was obtained for 1 study with n=74 participants, which used the GMA (Prechtl 1977) to assess the quality of fidgety movements at 10 to 18 weeks post-term in high- and low-risk infants. Ten high-risk infants were diagnosed with cerebral palsy (quadriplegia, right hemiplegia, left hemiplegia) and 1 unspecified type of cerebral palsy. The diagnostic accuracy of this method in predicting cerebral palsy was high (above 90%) for sensitivity, specificity, PPV and NPV.
Moderate-quality evidence from 1 study with n=142 participants used the neonatal neurological examination (NNE) adapted from Prechtl 1977 with several added predictors including variation of movement in term and preterm infants at birth or by 5 days after birth. This method had low sensitivity, but high specificity, PPV and NPV.
Moderate-quality evidence from 1 study with n=52 participants used GMA using the Prechtl 1977 method with age adaptions of the norm according to Touwen 1976. Of the 8 diagnosed with cerebral palsy, 3 were classified by GMA as having definitely abnormal (DA) movements, 4 were DA and 1 was mildly abnormal at fidgety age (8 to 17 weeks post-term). Seven diagnosed with cerebral palsy had cramped, synchronised general movements, which was significantly associated with cerebral palsy development. Four had predominantly jerky movement at fidgety General Movements (GMs) age (2-4 months postterm) and 4 had jerky and stiff movements at writhing age (38 to 47 weeks post-term).
High-quality evidence was obtained for 1 study with n=455 participants, which assessed quality of movements, grouped into ‘definite abnormal general movements’ according to method described by Hadders-Algra 2004 in the primary care setting (‘well-baby’ clinics providing routine assessments). Definite abnormal GMs had high specificity and NPV when predicting cerebral palsy, but low sensitivity and PPV.
Very low-quality evidence was obtained from 1 study with n=89 participants, which used infant motor profile (IMP) to assess motor behaviour in preterm and term infants at 4, 6, 10 and 12 months. IMP had excellent discriminative ability at predicting cerebral palsy, as calculated by AUC, at 6, 10 and 12 months and good discriminative ability at 4 months.
Low-quality evidence from 1 study with n=574 high-risk infants used the GMA at 1 and 3 months. The reference test was carried out at 2 years and consisted of a neurodevelopmental assessment (Touwen’s criteria and Bayley scale). Twenty two infants were diagnosed with cerebral palsy (4%). The sensitivity and specificity were 100% and 86% during the writhing period (1 month) and 100% and 97%, respectively, during the fidgety period. No 95% confidence intervals (CI) were provided.
Low-quality evidence from 1 study with n=115 preterm infants used the GMA (Prechtl 1977) to determine the quality of fidgety movements at 12 weeks. Nine infants were diagnosed with cerebral palsy (quadriplegia (n=1), diplegia (n=5), hemiplegia (n=2 left, n=1 right). Sensitivity analysis was carried out incorporating the ‘suspect’ infants into the ‘normal’, and ‘abnormal’ groups, as well as excluding them from the analysis. When excluded from analysis (n=110) there was high specificity, NPV and PPV with moderate sensitivity (89% [95% CI 51.75%–99.72, calculated from the paper]. ‘Suspect’ infants included in the ‘normal’ group resulted in a moderate sensitivity, specificity and NPV, and high PPV (no CIs were reported). Including the ‘suspect’ infants in the ‘abnormal’ group resulted in a moderate sensitivity and high specificity, PPV and NPV.
Moderate-quality evidence from 1 study with 84 preterm high-risk infants used the quality of GMs, cramped, synchronised movements and neurological examination at preterm (<37 weeks), term (38 to 42 weeks) and post-term to predict cerebral palsy in patients aged 2 to 3 years. Forty-four infants were diagnosed with cerebral palsy (n=22 diplegia, n=14 tetraplegia, n=8 hemiplegia). The area under the ROC for GMs was 97.4 (no 95% CI given). GM assessment had 100% sensitivity and NPV for all ages, the specificity and PPV only became moderate at 47 to 60 weeks postmenstrual age. Cramped synchronised character had high specificity and PPV for all age groups. Sensitivity and NPV was low until >43 weeks, where it was moderate. Neurological performance was low across all measures up to 43 weeks. Sensitivity and NPV were high only at 47 to 60 weeks.
Moderate-quality evidence was obtained from 1 study with 232 infants (randomly selected from 930 eligible infants), of which 120 were classed as high risk and 112 (control group) low risk. The GMA was carried out at 3 months in the high-risk group, when a classical neurological examination was done in both groups. At 2 years all infants had a further neurological examination, according to Illingworth’s method. The high-risk group had 32 (27%) infants with ‘abnormal’ neurological development (13 cerebral palsy without mental retardation, 18 cerebral palsy with mental retardation, 1 mental retardation). The low risk group had 35 (31%) infants with ‘abnormal’ neurological development (11 cerebral palsy without mental retardation, 22 cerebral palsy with mental retardation, 3 mental retardation). The GMA had high sensitivity, specificity and NPV with moderate PPV. The classic neurological examination had high sensitivity and NPV with low specificity and PPV (No 95% CIs were provided).
Under 8 months old
6.5.1.2. Excessive crying/irritability
Moderate-quality evidence from 1 study with n=142 participants used the NNE adapted from Prechtl 1977 with several added predictors, including irritability and consolability in term and preterm infants at birth or by 5 days after birth. This method had low sensitivity, but high specificity, PPV and NPV.
6.5.1.3. Feeding difficulties
Moderate-quality evidence from 1 study with n=142 participants used the NNE adapted from Prechtl 1977 with several added predictors including nasogastric tube feeding in term and preterm infants at birth or by 5 days after birth. This method had low sensitivity, but high specificity, PPV and NPV.
6.5.1.4. Asymmetry of movement
Very low-quality evidence from 1 study with n=89 participants used IMP to assess motor behaviour in preterm and term infants at 4 and 6 months. Total IMP score had excellent discriminative ability at predicting cerebral palsy at 6 months and good discriminative ability at 4 months. However, the subscale of ‘movement symmetry’ had poor discriminative ability at predicting cerebral palsy at both 4 and 6 months (only total IMP score presented in GRADE).
6.5.1.5. Abnormal muscle tone
Low-quality evidence from 1 study with n=239 participants assessed tone abnormalities using the method described by Amiel-Tison 1986 at 3 and 6 months until 12 months in high- and low-risk infants. Ten high-risk infants were diagnosed with cerebral palsy (4 hypertonia, 5 hypotonia) when followed up for 5 years and all of these infants had tone abnormalities.
Moderate-quality evidence from 1 study with n=82 infants reviewed the quantitative aspects of the motor repertoire between 6 and 24 weeks (post-term) and the results of a neurological examination (Touwen’s) at 7 to 11 years of age. Results were given for the presence and absence of an obligatory asymmetric tonic neck posture (ATN) at 11 to 16 weeks and neurological findings at school age, taking in to account the quality of the fidgety movements (FMs) and concurrent motor repertoire (smooth and variable, or abnormal: monotonous, jerky and/or stiff). No children were diagnosed with cerebral palsy who had abnormal FMs or normal FMs with a smooth and variable motor repertoire at 11 to 16 weeks. One infant was diagnosed with cerebral palsy who had normal FMs but abnormal motor repertoire (100% sensitivity, 74% specificity, 12.5% PPV, 100% NPV [large 95% CI for all figures]). The remaining diagnoses of cerebral palsy were children who had absent FMs and abnormal motor repertoire with an equal presence of an obligatory ATN posture (6 and 6 respectively), which had 100% specificity, 50% sensitivity (very large CI).
Over 8 months old
6.5.1.6. Asymmetry of movement
Very low-quality evidence from 1 study with n=89 participants used the IMP to assess motor behaviour in preterm and term infants at 10 and 12 months. Total IMP score had excellent discriminative ability at predicting cerebral palsy at 10 and 12 months. However, the subscale of ‘movement symmetry’ had poor discriminative ability at predicting cerebral palsy at both 10 and 12 months (only total IMP score presented in GRADE).
6.5.1.7. Feeding difficulty
No evidence was retrieved for this clinical manifestation.
6.5.1.8. Persistent toe walking
No evidence was retrieved for this clinical manifestation.
However, it is important to note that very low-quality evidence from 1 study with n=89 participants using IMP has a ‘variability’ subscale (reported as ‘variation’ in the study) that includes ‘variability of toe movements’. This subscale had excellent discriminative ability at predicting cerebral palsy at 10 and 12 months (only total IMP score reported in GRADE).
6.5.2. Developmental manifestations
6.5.2.1. Delayed sitting in under 8 months old
Very low-quality evidence from 1 case control study (Allen & Alexander 1992, 1994) looked at the delay in attaining motor milestones in very preterm infants (n=173). The controls used were term infants (n=381) that were followed to 2 years of age. Analyses were carried out against population and family origin specific norms. Sitting without support and coming to sit for both white and non-white very preterm infants had poor PPV (range 31 to 56%). White very preterm infants had similar sensitivity (range 87 to 94%) and moderate to low specificity compared to the non-white very preterm infants for both milestone measures.
The delay criteria of 12.5%, 25%, 37.5% and 50% were also analysed. It was found that as the delay criteria increased the sensitivity decreased and specificity and PPV increased.
Very low-quality evidence from 1 study with n=89 participants using IMP had a ‘performance’ subscale that included ‘ability to sit’. This subscale had excellent discriminative ability at predicting cerebral palsy at 4 months and good discriminative ability at 6 months (only total IMP score reported in GRADE).
6.5.2.2. Delayed walking in over 8 months old
Very low-quality evidence from 1 case control study (Allen & Alexander 1992, 1994) looked at the delay in attaining motor milestones in very preterm infants (n=173). The controls used were term infants (n=381) that were followed to 2 years of age. Walking independently had high sensitivity in white and non-white preterm infants against population and family origin specific norms (range 94–100%). Specificity was moderate (73–75%), and PPV low (37–44 in non-white, 58% white infants).
Moderate-quality evidence from a prospective cohort study (n=4,275 analysed) assessed the proportion of infants (low birthweight [<2kg] or >24 hours in special care nursery) who were walking at 18 months and its relationship with the diagnosis of cerebral palsy. There were 410 infants walking, of which 66 were diagnosed with definite cerebral palsy and 11 suspected. Including the suspected cases, there was moderate sensitivity and high specificity with low PPV (no 95% CI provided).
Very low-quality evidence from 1 study with n=89 participants using IMP had a ‘performance’ subscale that included ‘walking. This subscale had excellent discriminative ability at predicting cerebral palsy at 10 and 12 months (only total IMP score reported in GRADE).
6.5.3. Use of tools to identify clinical and developmental manifestations of cerebral palsy
6.5.3.1. The Early Motor Pattern Profile (EMPP)
Moderate-quality evidence from a prospective cohort study looked at the use of the EMPP to predict cerebral palsy at 6 and 12 months (corrected age). The study included 1,247 high-risk infants. Both time points yielded moderate or high sensitivity, specificity, PPV and NPV.
6.5.3.2. The Bayley Scales of Infant and Toddler Development – Third edition (Bayley-III)
High-quality evidence from a prospective cohort study used the Bayley-III to assess motor impairment at 2 years of age to predict motor outcome at 4 years. At 4 years, 115 infants completed the Bayley-III assessment and 96 infants completed the Movement Assessment Battery for Children – Second edition (MABC-2). When a cut off of -1SD was used, there was moderate sensitivity (wide 95% CI), high specificity and NPV with a low PPV (wide 95% CI). A cut off of -2SD had low sensitivity (wide 95% CI) and high specificity, NPV and PPV (wide 95% CI).
6.6. Evidence to recommendations
6.6.1. Relative value placed on the outcomes considered
Critical outcomes, as stated by the Committee, were sensitivity and specificity. Important outcomes included: PPV, NPV, AUC, likelihood ratios and proportion diagnosed.
6.6.2. Consideration of clinical benefits and harms
The Committee agreed that the prediction and diagnosis of cerebral palsy are distinct areas. Prediction is as much about recognising risk factors in the history as well as subtle abnormalities on examination. Diagnosing is about hard neurological findings on examination with a history of delay in achieving a developmental milestone or skill. The prediction of cerebral palsy involves the recognition of clinical and developmental manifestations, such as atypical movements, which can allow further assessment and later diagnosis of cerebral palsy. The Committee agreed that these clinical and developmental manifestations allow the early detection of cerebral palsy and are important as they are not widely assessed or recognised at first presentation, which means that children with cerebral palsy can remain undetected until clinical diagnosis at a later age, and therefore not receive beneficial early care.
The Committee agreed that those at high risk should have neonatal follow-up for the first few months of infancy and those at low risk should receive the standard follow-up assessments that are undertaken as part of the Healthy Child Programme. The Committee was aware of the NICE guideline currently in development on developmental follow-up of preterm babies (due for publication in August 2017). Additionally, the Committee noted that signs may not be obvious at first presentation and that there was a need for the continuous record of what children do from presentation so there is a record of change. For example, dyskinetic cerebral palsy may often present as stiffness, irritability and/or low muscle tone in the first year of life. The Committee also pointed out that children with milder forms of cerebral palsy may present to health services for the first time with difficulties of motor function even after age 5 years.
For high-risk infants, the GMA was recommended in the first 3 to 4 months to identify features suggestive of cerebral palsy to supplement routine clinical examination. This was supported by the evidence, as most studies used the GMA as part of their assessments, and was in line with the Committee’s experience. The GMA allows healthcare professionals to identify high-risk infants that need further assessment and follow-up. Therefore, it has not been recommended as a method of diagnosis, rather as a method of identifying children requiring further assessment. If false positives or false negatives arise using this method, children will still receive further assessment and follow-up until the diagnosis of cerebral palsy is ruled in or out. Additionally, the Committee agreed that high-risk infants should continue to receive multidisciplinary assessment undertaken by professionals with specialist training for the first 2 years of life. These assessments are normally carried out in a post-neonatal follow-up service.
In the low-risk infants and children, the Committee agreed that it was reasonable to expect that routine screening assessments would identify infants with delayed and abnormal motor milestones and to help with onward referral to the child development centre for further assessment.
The Committee agreed that, based on the evidence reviewed and their clinical experience, healthcare professionals who are working with young infants either as part of a follow-up of high-risk infants or as part of a developmental surveillance programme should be able to recognise the following clinical features as suggestive of cerebral palsy: unusual fidgety and abnormal movements, asymmetric movements, abnormal tone and abnormal motor development.
In terms of developmental milestones, the Committee considered that the evidence reviewed and their clinical experience supported a recommendation to refer children who showed late sitting and late walking for further assessment. Based on their clinical experience, the Committee agreed that hand preference before the age of 1 year should also be a developmental concern to trigger further assessment as they recognised that hand preference is often not seen until children are 2 to 3 years of age.
Although no significant evidence was found on toe walking, the Committee considered that, based on their clinical experience, children who display obvious and persistent toe walking on its own should be referred for onward assessment. They agreed that no precise definition of ‘persistent’ could be put forward, and that it would depend on clinical judgement.
Finally, the Committee pointed out that in children in which there is a motor delay concern and if a cerebral palsy diagnosis cannot be made, then healthcare professionals should explain to parents the reasons of the increased surveillance. Motor delay may be a sign of muscle disease, peripheral nerve disorders or learning difficulties. It may also be that the child is at the slower end of the normal developmental spectrum. As it may take some time for the abnormal neurological signs to appear that would help confirm a diagnosis of cerebral palsy, it may not be possible to give the child a definite diagnosis at first presentation. Therapy can be started based on the child’s developmental problems while waiting for a diagnosis to be made with time.
The Committee noted the importance of communication between all tiers of service involvement to ensure the best-quality care is provided to all children and young people with cerebral palsy, with the parents and/or carers at the centre of all communications. They agreed that involvement of primary care services in all discussions about ongoing management of the child and young person with cerebral palsy is crucial. They also pointed out that any clinician in primary, secondary or tertiary care can refer to a local specialist multidisciplinary team (MDT).
6.6.3. Consideration of economic benefits and harms
This review question is not relevant for economic analysis because it does not involve a decision between alternative courses of action. Even so, there are considerations for the resources and costs that enhanced surveillance and referrals to child development centres may entail.
Specifically, the Committee highlighted that identifying the clinical and developmental manifestations needs enhanced surveillance for infants and children who have spent time in specialist neonatal care who are at increased risk of developing cerebral palsy. They noted that there was regional variation of the resources available for specialist support for cerebral palsy and, to address geographical variation, the Committee agreed recommendations that identified the levels of surveillance infants and children with cerebral palsy should receive.
The Committee agreed that referrals to child development centres or enhanced clinical and developmental follow-up programmes would not be considered cost effective if they do not add any additional information to routine monitoring and do not lead to an improvement in the infant or child’s management strategy. The Committee noted that recommendations on the population identified to need enhanced surveillance, and the frequency of that surveillance, could have significant resource implications. However, as the Committee advised that enhanced follow-up programmes should only be provided for infants and children who are at increased risk of developing cerebral palsy there should not be a large increase in the demand for enhanced surveillance as those risk factors outlined for cerebral palsy in recommendation 1 already trigger closer surveillance and is already accepted current clinical practice.
The Committee stated that GMAs are not regularly done in UK clinical practice, adding that the resources to do a GMA (2 observers and a 20-minute video clip) were often considered to outweigh the additional value of the GMA to a standard assessment. However, the Committee noted that the clinical evidence review included high-quality evidence on the GMA and the diagnostic accuracy of this method to predict cerebral palsy was high. As a result, the Committee prioritised a recommendation for clinicians to consider using the GMA. A stronger recommendation to use the GMA was not agreed as its value in addition to a standard assessment may not be outweighed in all cases.
The Committee advised that infants with delayed and abnormal motor milestones would be identified during routine screening assessments, at no additional cost, as this is part of the national ‘red book’ screening programme. The Committee also noted that delayed and abnormal motor milestones already result in onward referral to the child development centre for further assessment in current clinical practice. The Committee concluded that the findings from the clinical evidence review, combined with their clinical experience, supported a recommendation to justify current NHS expenditure to refer all infants and children who showed late sitting and late walking for further assessment.
The Committee also added that, although no significant evidence was found on toe walking, children who display obvious and persistent toe walking are often referred for onward assessment in clinical practice, and many are subsequently identified with cerebral palsy. Therefore, referrals initiated from toe walking or delayed and abnormal motor milestones may lead to a timely change in the child’s management, potentially increasing their quality of life and evading downstream costs from complications that could arise from unidentified cases of cerebral palsy.
Overall, knowing the key clinical manifestations of cerebral palsy may lead to better identification (and thus more timely management) and has therefore, indirectly, potentially important resource implications. However, while the costs of referrals or enhanced surveillance could be significant, without knowing the outcomes of those services, we cannot know if they will be cost effective.
6.6.4. Quality of evidence
The QUADAS-2 checklist was used when appraising diagnostic evidence for the best tools to identify clinical and developmental manifestations of cerebral palsy at first presentation. The methodology checklist for prognostic studies (2012) was used instead when appraising evidence for the key clinical and developmental manifestations of cerebral palsy at first presentation. The quality of evidence ranged from very low to high. The main sources of bias in the studies were selection bias and the reference test undertaken with knowledge of index text.
6.6.5. Other considerations
The recommendations related to this evidence review were based on the evidence and the Committee’s clinical experience.
6.6.6. Key conclusions
The Committee concluded that certain manifestations such as abnormality of movement and tone may be suggestive of cerebral palsy and that infants and children with delayed milestones such as late sitting and late walking should be referred for onward assessment.
6.7. Recommendations
- 12.
Provide an enhanced clinical and developmental follow-up programme by a multidisciplinary team for children up to 2 years (corrected for gestational age) who are at increased risk of developing cerebral palsy (see recommendation 1).
- 13.
Consider using the General Movement Assessment (GMA) during routine neonatal follow-up assessments for children between 0 and 3 months who are at increased risk of developing cerebral palsy.
- 14.
Recognise the following as possible early motor features in the presentation of cerebral palsy:
- unusual fidgety movements or other abnormalities of movement, including asymmetry or paucity of movement
- abnormalities of tone, including hypotonia (floppiness), spasticity (stiffness) or dystonia (fluctuating tone)
- abnormal motor development, including late head control, rolling and crawling
- feeding difficulties.
- 15.
Refer children who are at increased risk of developing cerebral palsy and who have any abnormal features listed in recommendation 14 to a child development service for an urgent assessment.
- 16.
Recognise that the most common delayed motor milestones in children with cerebral palsy are:
- not sitting by 8 months (corrected for gestational age)
- not walking by 18 months (corrected for gestational age)
- early asymmetry of hand function (hand preference) before 1 year (corrected for gestational age).
- 17.
Refer all children with delayed motor milestones to a child development service for further assessment.
- 18.
Refer children who have persistent toe walking to a child development service for further assessment.
- 19.
If there are concerns that a child may have cerebral palsy but a definitive diagnosis cannot be made, discuss this with their parents or carers and explain that an enhanced clinical and developmental follow-up programme will be necessary to try to reach a definite conclusion.
- 20.
Refer all children with suspected cerebral palsy to a child development service for an urgent multidisciplinary assessment, in order to facilitate early diagnosis and intervention.
- 21.
Recognise that ongoing communication between all levels of service provision in the care of children and young people with cerebral palsy is crucial, particularly involvement of primary care from diagnosis onwards.
6.8. Research recommendations
None identified for this topic.
- Clinical and developmental manifestations of cerebral palsy - Cerebral palsy in ...Clinical and developmental manifestations of cerebral palsy - Cerebral palsy in under 25s: assessment and management
Your browsing activity is empty.
Activity recording is turned off.
See more...