Complexity of the Cerebral Palsy Syndromes: Title and subTitle BreakToward a Developmental Neuroscience Approach

Over the past 2 decades, major advances in obstetrics, genetics, maternal fetal medicine, neonatology, developmental neurosciences, and reproductive epidemiology^{1 - 12} have resulted in unprecedented low rates of infant mortality. In 2004, the overall US infant mortality rate was 7 per 1000, with 90% survival of children born very prematurely at 28 to 32 weeks of gestation and survival as high as 80% for children born extremely prematurely at 24 to 28 weeks.^{13 - 14} In addition, a new consensus definition of cerebral palsy (CP) has been proposed,¹⁵ advances in neuroimaging^{16 - 20} have allowed for the examination of central nervous system structure, and a gross motor function classification system^{21 - 22} has given neurodevelopmental pediatricians, orthopedic surgeons, and physical therapists a common language for interdisciplinary collaboration.

In 2000, Stanley et al²³ proposed that etiologic research on single factors needed a more comprehensive framework of causal pathways in order to understand the complexity of children with CP syndromes. One population at known risk of CP that has not been systematically examined is children who have survived very preterm or extremely preterm birth. Recent data from the 14-center National Institute of Child Health and Human Development Neonatal Network of children born between 1993 and 1998 found rates of CP of 19% in survivors born at 22 to 26 weeks of gestation with birth weights of less than 1000 g, and 12% for children who survived after delivery at 27 to 32 weeks of gestation and weighed less than 1000 g.²⁴ Subsequent analysis from this same group examined outcomes for 1016 infants at the threshold of viability.²⁵ These infants had a mean birth weight of less than 750 g, a gestational age of less than 24 completed weeks, and a 10-minute Apgar score of 3 or less. Furthermore, 75.8% of these infants died. Among the survivors, 30% had a CP syndrome and almost 1 in 2 had cognitive developmental disability as indicated by a Bayley Mental Development Index score of less than 70 in early childhood.

The report by Bax and colleagues²⁶ in this issue of JAMA is a major advance. This multicenter collaboration investigated clinical correlates of CP in a population sample and compared clinical findings with information available from magnetic resonance imaging (MRI). A cross-sectional population involving more than 500 children with CP born between 1996 and 1999 was assembled from 8 major European centers. Four hundred thirty-one children with CP syndromes were clinically assessed using a structured history and a systematic neurodevelopmental evaluation that included topography (diplegia, hemiplegia, and quadriplegia), physiology (spasticity, dyskinesia, dystonia, and ataxia), and neurological comorbidities involving vision, hearing, and epilepsy. Of these children, 351 had a cranial MRI scan at age 18 months or later that was systematically reviewed using a consensus protocol and assessed by a single evaluator.

By obtaining MRI scans at 18 months, the completion of the early developmental stages of central nervous system structural integrity can be linked to the expanded neuromotor classification. Because of the large sample, this study provides some preliminary observations of timing, clinical risks, and functional impact on sitting balance, hand function, and communicative skills. Moreover, this cohort appropriately captures most of the neuromotor-challenging CP syndromes occurring in early childhood (ages 2-5 years). Almost 1 in 3 children had the diplegic pattern of CP, 1 in 4 had hemiplegia, and 1 in 5 had quadriplegia. In addition, 1 in 5 had extrapyramidal CP with either ataxia or dyskinesia.

Important observations in this study include the following: (1) In terms of prenatal risk, approximately 1 in 5 mothers who had a child with CP had a urinary tract infection during pregnancy compared with 2.9% in a regional obstetric database. (2) One in 3 children were born by emergency cesarean delivery. (3) Twelve percent of children with CP were from a multiple pregnancy, compared with 1.5% expected. (4) More than half of the children with CP were of term gestation. (5) Almost 1 in 5 children were small for gestational age (birth weight <10%). (6) More than 40% of the children who were born at term spent more than 5 days in the special care unit and were regarded as significantly ill. (7) Recurrent seizures occurred in 28% of children, hearing impairment in 7.2%, and visual impairment, including strabismus, restricted fields, and refractive errors, in 33%. (8) Only 10.9% of the children with CP were born at less than 28 weeks of gestation. (9) Approximately 1 in 3 children with CP were born at either 28 to 31 weeks or 32 to 36 weeks of gestational age.

Groups in the European Union, considered to be at low risk of neonatal follow-up surveillance, contributed a large number of cases of CP in the study by Bax et al.²⁶ From a population standpoint, reducing the prevalence of CP requires understanding pathways of the CP syndromes in term infants, moderately preterm infants of 32 to 36 weeks of gestation, and multiple births. More than 750 000 children and adults in the United States have a CP syndrome, and the lifetime medical cost per individual has been estimated at $1 million and is currently $1.2 billion in direct medical costs for children born in 2000.²⁷ In this respect, disproportionate attention to both severe perinatal hypoxemic ischemic encephalopathy in term infants and neurodevelopmental complications in extreme prematurity lead to the erroneous perception that these 2 risk groups of children account for the majority of cases of children with CP.

Another major lesson from the study by Bax et al is that the majority of CP is not severe.^{13 ,28} Approximately 3 in 5 children with CP have hemiplegia or diplegia. This is associated with an excellent prognosis for ambulation and, based on the number of limbs involved, can be considered as a less severe disability. In the group of preschool children, 89% of children with hemiplegia and 63% with diplegia were walking. Less than 1 in 5 children with CP had quadriplegia, but only 9% of those with quadriplegia could walk. The majority of these children have sitting challenges, manipulative challenges, and communicative difficulties.^{29 - 31} Children with CP quadriplegia often have comorbidities of dysphagia, seizures, and recurrent pneumonia and may have multiple severe neurodevelopmental functional challenges and medical frailty. In absolute terms, however, these substantially challenged children are only a small minority of individuals with CP.

In the study by Bax et al, neuroimaging data were most informative. White-matter abnormalities were present in 43% overall and in 71% of children with diplegia, 34% of children with hemiplegia, and 35% of children with quadriplegia. This highlights the critical importance of understanding biomarkers and pathways in preterm infants born before 34 weeks of gestation as well as some term infants with CP, suggesting vulnerability during the third trimester of intrauterine life. Other important findings from neuroimaging included basal ganglia abnormalities in 13%, malformation in 9.1%, cortical-subcortical abnormalities in 9.4%, and focal infarcts in 7%. Of children with basal ganglia and thalamic injury, 76% had dystonia. Of children with hemiplegic CP, 27% had focal infarcts. It is in this group that maternal disorders of coagulation need to be comprehensively investigated.³² The frequency of malformations (9.1%) was as prevalent as cortical-subcortical damage (9.4%). Thus, neuroimaging helped in the understanding of timing and extent of lesions. Almost 1 in 8 children with CP had a normal MRI scan, and this group awaits the promise of neurogenetic technologies.

Among the term and near-term children with CP, approximately 1 in 5 had cortical-subcortical or basal ganglia injury. The authors note that since 26% of this group had emergency cesarean deliveries, less than 1 in 10 infants could have had CP because of obstetrical mismanagement. These estimates are similar to estimates from the Western Australia Cerebral Palsy Registry.³³

To understand the complex causal pathways involved in the CP syndromes of hemiplegia, diplegia, quadriplegia, and extrapyramidal disorders, large population studies are required. In contrast with the regionalization of neonatology, there is fragmentation with respect to child neurodevelopmental services in the United States, and both the current early intervention system and preschool systems are far from comprehensive in combining diagnosis, intervention, and medical collaboration. Critical to understanding the relationship among risk, structural difference, motor control, and developmental functioning is having systematic assessments coupled with safe imaging and quality habilitative and family supports.^{27 ,31 ,34 - 35} From a public health perspective, children with the quadriplegic type of CP have significant sensory, feeding, growth, pulmonary, and orthopedic complications.

The key translational question is how to use advances in maternal-fetal medicine, neonatology, and developmental neuroscience to describe the best explanatory mechanisms for children with a CP syndrome. Most important is how to use the best science to understand potential pathways that would decrease the functional severity of CP.

Similar to research findings in coronary artery disease and adult stroke, pediatric investigators have recognized that the clustering of multiple risks affects the developmental vulnerability of the brain. The current task is to understand the pathways of risk and protection for specific groups of children with CP. These groups currently include twins and higher-order multiples, children with intrauterine growth restriction, children with malformations, children undergoing congenital heart surgery, and children with neonatal seizures and encephalopathy. By focusing efforts on white-matter injury in preterm infants and neuroprotection of term infants with the highest risk of encephalopathy, as well as realizing that a substantial number of term and near-term children do not have a simplistic cause for their motor disability, families can be helped and the general public can understand the need for addressing shortcomings in current knowledge. With respect to ongoing management, a long-term whole-child focus that optimizes health, developmental and functional outcomes, community participation, and family well-being is in order. In this way, an informed and compassionate science for children with neurodisability can be developed.