Early Detection of Cerebral Palsy

Key Points

1.
Cerebral Palsy (CP) originates from an injury or abnormality in the developing brain, resulting in abnormal muscle tone, and consequently, in altered movement, posture, and motor function.
2.
Tools used to accurately detect CP in the first year of life include a combination of Prechtl General Movement Assessment (GMA), clinical history, standardized neurological exam, brain imaging, and an assessment of motor function.
3.
The diagnosis of CP should be made as early as possible so that families may receive necessary support, infants may receive therapies specific to the diagnosis, and surveillance for co-morbidities can be intiated.
4.
CP is often accompanied by other neurodevelopmental impairments, including orthopedic problems, epilepsy, and impaired cognition, perception, communication, and behavior.

Introduction

The inclusion of a chapter on early detection of Cerebral Palsy (CP) in a volume on evidence-based neonatology demonstrates a shift which started in the mid 2010’s in the practice of neonatal follow-up. For several decades prior, neonatal intensive care units (NICUs) primarily leveraged follow-up programs to document the outcomes of their perinatal interventions. ^, At worst, NICU follow-up was seen as a necessary financial burden in order to have “long-term” outcomes for randomized controlled trials, with a single assessment visit at the age of 2 to administer Bayley Scale(s) of Infant and Toddler Development assessment and a neurological exam. At best, these programs functioned as strong reminders of the consequences of neonatal care and of neonatologists’ ^, responsibilities that extended beyond saving patients’ lives. The late Dr. Maureen Hack was the leader of a movement promoting standardized NICU follow-up at 2 years of age, and influenced a generation of physicians to see detection of CP at this 2-year visit as an essential check-and-balance to the care provided in the NICU. Most studies of outcomes after NICU care, whether clinical trials or not, started at the 2-year mark (18–22 or 22–26 months) and included some type of standardized neurological examination. Examiners, many of them without a strong neurological background, were tasked with attributing a diagnosis of CP based on algorithms for this exam. ^, ^, ^, Sometimes, infants with the worst types of neurological insults had already received a diagnosis by a neurologist. This was especially true if they had associated comorbidities such as seizures or coagulopathies. In this case, the follow-up physicians confirmed their colleagues’ observations. Overall, the reported mean age at CP diagnosis in the 1990s ranged from 1 to 8 years, in the 2000’s from 1 to 8 years and in the 2010’s from 0 to 5 years.

Studies of NICU graduates conducted in the 1990’s and early 2000’s showed a binary presence/absence of CP at 2 years after birth. Some studies measured the severity, and categorized the patients as having mild, moderate, or severe disability. More recently, the Gross Motor Function Classification System (GMFCS) was adopted in many NICU follow-up studies to standardize and refine the description of CP ^, ( Table 94.1 ). When reporting outcomes, children with a GMFCS level of 1 (i.e., those with the least gross motor impairment) were grouped with those without CP, ^, intimating that both populations had no impairment. This disregarded the very poor stability of the GMFCS at 2 years of age, especially when trying to predict outcomes into childhood or adulthood. It also did not account for the numerous co-morbidities of CP that may impact children’s ability to function far more than walking independently by the age of 2. ^, Most of all, it overlooked the well-documented concerns and frustrations of parents whose children had CP—whether with a GMFCS of 1 or 5—at waiting until the age of 2 to obtain a diagnosis. ^, Consequences of parental dissatisfaction range from mistrust of the healthcare system to depression and anxiety, with long-term adverse effects on their well-being and ability to advocate for the best care for their children. ^, ^, Studies of parent perceptions surrounding the diagnosis clearly demonstrate their wishes to have an honest and direct conversation as early as possible. They accept the uncertainties inherent to such an early diagnosis and are also content with a classification of “high-risk for CP”, but find delays in surveillance and counseling unacceptable.

Table 94.1

Gross Motor Function Classification System (GMFCS)

Level I:	Walks without restrictions; limitations in more advanced gross motor skills
Level II:	Walks without assistive devices; limitations in walking outdoors in the community
Level III:	Walks with assistive mobility devices; limitations in walking outdoors and in the community
Level IV:	Self-mobility with limitations; children are transported or use power mobility outdoors and in the community
Level V:	Self-mobility is severely limited even with the use of assistive technology

Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol . 1997;39(4):214–223. doi: 10.1111/j.1469-8749.1997.tb07414.x .

These concerns, as well as the initiatives of an international expert group, prompted a world summit in Vienna Austria, in July, 2014. The working group was composed of medical specialists in all disciplines broadly related to CP, including allied health professionals, researchers, community stakeholders, and parents. After a systematic review of the evidence, recommendations for early detection based on the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system were established and ultimately published in 2017. Two types of patients were identified in whom CP could and should be detected before 2 years of age and preferably by 5 months. Those with attributable risks have known risk factors in the perinatal period that fit with the referral criteria for most NICU follow-up programs. They are preterm or have various types of pre- or postnatal insults, ranging from the macrostuctural (malformations, hemorrhage, ischemia, or thrombosis) to the microstructural (encephalopathy of prematurity, inflammation, infections). In these infants, a set of assessment tools can be used to make a diagnosis with predictive values for instruments ranging from 80%–98%. Overall accuracy increases when several instruments are used concurrently. For instance, increased predictive accuracy for CP at 3–4 months was obtained when absent fidgety movement patterns on the General Movements Assessment (GMA) were combined with term-equivalent age magnetic resonance imaging (MRI) results and Hammersmith Infant Neurological examination scores. However, the authors of this report acknowledge that the precision obtained in prediction was balanced by the fact that the GMA and MRI readers were all expert level researchers in the field. The positive predictive value of assessments performed by trained researchers in high-risk population, in contrast to that of clinicians in an outpatient setting with a more diverse patient base remains an issue. In a study of two different countries’ high-risk follow-up programs that included infants with congenital abnormalities or infections, the specificity of the GMA decreased to the mid-80s. Regardless of the setting, the fact remains that a set of best-evidence guidelines, with GRADE recommendations for moderate-to-high quality exists to allow early detection and diagnosis of CP in high-risk infants.

More recently, as awareness of these guidelines has increased, a separate and confounding issue has occurred in neonatology. As advances in neonatal care, such as systematic use of antenatal steroids or of postnatal caffeine, have improved the developmental outcomes of preterm infants, the overall prevalence of severe forms of CP may be decreasing. Reports indicate that in some developed countries the overall incidence of CP may be decreasing, while severity remains constant. ^, However, reports in North America indicate that overall incidence may be increasing, as severity decreases. Differences may be due to cohort characteristics or reporting and assessment methodologies. In either scenario, early detection may be more difficult than originally anticipated. The systematic application of guidelines for early detection of CP then becomes a necessary first step to best practice, and an essential next step is improvement of these guidelines in the face of a changing clinical picture for the next decade of NICU care.

The goal of this section is to help guide neurodevelopmental practitioners in their clinical practice by describing how the pathophysiological processes associated with development of CP can be identified in the perinatal period, what clinical features and assessments can support our best efforts at surveillance, which outcomes can help guide prognostic conversations with families, and why optimal management early on can lead to best outcomes for young patients in the future.

Pathophysiology

By definition, CP originates from an injury to or abnormality in the developing brain resulting in abnormal muscle tone which leads to impairment of motor function. To understand the pathophysiology of CP, it is helpful to understand the progression of fetal brain development which begins in the 3rd week of gestation. At 2 to 4 months, neuronal prolifieration occurs with rapid growth of subplate neurons which are the main neuronal components of the cerebral white matter. This is followed by neuronal migration from 3 to 5 months during which time there is also axonal and dendritic growth and synaptogenesis. Neuronal organization and myelination begin in the 3rd trimester and continues well after delivery, though the most rapid period of myelination occurs in the first 2 years of life. Injury to or maldevelopment of the developing brain causes specific brain injury patterns reflecting the disturbance of brain development during critical time windows. However, in many cases CP may not derive from a single cause but rather from a complex cascade of events.

Congentital Abnormalities

Congenital anomalies are found in 15%–19% of children with CP. Cerebral anomalies are the most common and may result from interruption of neuronal proliferation, migration, and differentiation during the first and second trimester. Aberrancy of these developmental processes may lead to cortical malformations including polymicrogyria, lissencephaly, and pachygyria. Malformations of grey matter development accounts for approximately 10% of children with CP. Other malformations such as hydranencephaly or schizencephaly may be secondary to infections during pregnancy or a vascular insult during periods of neuronal migration in the second trimester. Microcephaly and hydrocephalus are two of the most frequent cerebral anomalies in children with CP. These may be secondary to abnormal brain development such as in Dandy Walker malformation or aqueductal stenosis, but may also be secondary to intrauterine insults such as a fetal intraventricular hemorrhage resulting in post-hemorrhagic hydrocephalus or intrauterine ischemic injury causing suboptimal brain growth. Congenital abnormalities of brain development are more common in term than preterm infants with CP (16% vs 2.5%) and more frequently associated with spastic quadriparesis or ataxic CP than hemiparetic or dystonic CP. ^, The prevalence of non-cerebral anomalies is also higher in children with CP than in the general population and include cardiac, musculoskeletal, and urinary abnormalities.

Prematurity

Twenty-five thousand extremely-low-birth-weight (ELBW; birth weight less than 1000 grams) infants are born in the U.S. each year. Seventeen thousand are born at less than 28 weeks and considered extremely premature; these infants are at a higher risk of disability as the risk of CP is inversely correlated with birth weight and gestational age. The prevalence of CP in children born extremely preterm is 5%–15%. As survival in the very-low-birth-weight (VLBW; birth weight less than 1500 grams) population has improved in recent years, this has resulted in more children with long-term disabilities. These include deficits in motor skills, cognition, language, behavior, and attention. ^, The pathogenesis of brain injury in this group is a combination of ischemia, hemorrhage, and inflammation damaging the cerebral cortex, white matter, thalamus, basal ganglia, and cerebellum. There are three main patterns of brain injury in preterm infants associated with long-term neuromotor impairments: periventricular leukomalacia (PVL), overall decrease in brain volume, and severe germinal matrix hemorrhage-intraventricular hemorrhage (GMH-IVH). Severe GMH-IVH is particularly likely to be associated with neuromotor impairments when accompanied by periventricular hemorrhagic infarction (PVHI). ^,

PVL is the most common injury of white matter in the premature infant. The pathogenesis of PVL is a combination of ischemia and inflammation during the period of rapid axonal and dendritic growth. Ischemia injures premyelinating oligodendrocytes resulting in hypomyelination and damages subplate neurons, likely via apoptosis. The combination of ischemia and inflammation causes decrease in neuronal growth, axonal development and myelination, which can further lead to an overall decrease in volume of the thalamus, basal ganglia, cortex, corpus callosum, and cerebellum. There is both focal and diffuse necrosis which can be visualized macroscopically on head ultrasound as cystic PVL. The less obvious glial scarring only becomes evident after several weeks and requires MRI to visualize. Up to 50% of VLBW infants demonstrate findings of PVL on MRI but with advances in management of the critically-ill VLBW infant, the rates of cystic PVL have decreased to 4%, accounting for only a small percentage of overall infants with white matter injury. ^, Rates of CP in cystic PVL range from 52%–100% with the pattern of spastic diplegia being most common.

GMH-IVH and periventricular hemorrhagic infarction (PVHI) cause injury via some of the same mechanisms resulting in PVL. The germinal matrix is the primary source of proliferation of oligodendrocyte progenitor cells. Hemorrhage in this area leads to the loss of myelin-producing cells, resulting in decreased myelination and lack of axonal development. There can be direct injury to GABAergic neurons (neurons that secrete gamma-amino butyric acid [GABA], the primary inhibitory neutrotransmitter in the brain), leading to decreased volume of the cerebral cortex and thalamus. These injuries may be exacerbated if post hemorrhagic hydrocephalus develops; this further impairs oligodendrocyte growth and myelination and results in axonal loss. The younger the infant, the greater the risk of severe hemorrhage.

Long-term motor deficits after PVHI correlate with topography of the parenchymal lesions with resultant spastic hemiparesis or asymmetrical spastic quadriparesis being most common. Laterality of PVHI also impacts outcomes, with 50%–67% of infants with unilateral PVHI developing CP, and up to 90% of infants with bilateral PVHI developing CP. ^,

Perinatal Stroke

Perinatal stroke is a group of cerebrovascular disorders which occur in the developing brain between 20 weeks of fetal life and 28 days postnatal life. This is the highest risk period in childhood for arterial ischemic stroke and occurs in 1:4000 live births. Perinatal stroke is the most common cause of hemiplegic cerebral palsy, and 68% of children with perinatal stroke develop cerebral palsy. The days before and after birth are a period of increased stroke risk for both mother and baby. This may be related to activation of coagulation in both mothers and newborns. There are six subtypes of perinatal stroke, three of which are detected due to acute symptomatic presentation in the neonatal period, most often presenting as seizures. These include (a) neonatal arterial ischemic stroke (NAIS); (b) neonatal cerebral sinovenous thrombosis; and (c) neonatal hemorrhagic stroke. The other three subtypes of perinatal stroke are detected during evaluation of infant or child with hemiparetic CP or other neurological exam abnormalities and include (d) arterial presumed perinatal ischemic stroke; (e) periventricular venous infarction; and (f) presumed perinatal hemorrhagic stroke.

NAIS is the most common type of acute neonatal stroke. The majority involve the anterior circulation—specifically the middle cerebral artery (MCA) territory—and are more common on the left side. Injury to the MCA territory can result in upper motor neuron injury in the cerebral cortex and descending corticospinal tracts at the levels of the posterior limb of the internal capsule and cerebral peduncle. These in addition to basal ganglia injury are associated with contralateral hemiparesis.

A specific etiology is detected in less than 20% of cases of perinatal stroke. Numerous risk factors have been identified, including placental disease, fetal heart rate abnormalities, emergency caesarean section, need for resuscitation and 5-minute Apgar score less than 7. Intrauterine growth restriction and small for gestational age are also associated with neonatal arterial ischemic stroke. ^, Though nulliparity, pre-eclampsia and gestational diabetes have been proposed to increase risk of perinatal stroke, these associations have been inconsistent across studies.

Infection

Both maternal and neonatal infections increase the risk of CP, though the mechanism is not well understood. Maternal chorioamnionitis has been documented in several studies to be associated with increased risk of CP in the child ; an increasing body of literature associates inflammatory markers with increased risk. In a large California population-based study, extra-uterine maternal infections also increased the risk of CP in the child. This was true both for infections detected prenatally and perinatally.

Neonatal infection also increases the risk of CP, with higher infection rates noted in preterm children. In children born term, those with neonatal infection are more likely to have white matter injury (odds ratio [OR] 2.2) and develop spastic diplegia (OR 1.6) while children born pre-term were more likely to have white matter and cortical injury (OR 4.1) and develop spastic triplegia or quadriplegia (OR 2.4).

Hypoxic-Ischemic Encephalopathy

Hypoxic-ischemic injury in the perinatal period accounts for up to 10% of cases of CP. Ischemia leads to neuronal necrosis and apoptosis as well as injury to oligodendrocytes and impaired cerebrovascular autoregulation. ^, The two predominant patterns of injury in neonatal hypoxic ischemic encephalopathy (HIE) are watershed injury (involving vascular boundary zone white matter and cortical grey matter) and deep grey matter nuclei which includes the basal ganglia and thalamus. ^, Injuries to the thalamus, basal ganglia, and posterior limb of the internal capsule are associated with worse developmental outcomes. Thirty-six percent to 44% of children with moderate to severe HIE were diagnosed with CP by 2 years of age, ^, with moderate to severe basal ganglia lesions the best predictor of CP.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here