Is it now possible to identify all newborn infants at risk of cerebral palsy?

Abstract

To prevent or modify the course of cerebral palsy (CP) in children is a key focus for many researchers and clinicians working with pregnant females and newborn infants. There is increasing evidence that addressing deviations from typical development early and initiating targeted interventions can improve prognosis in CP.¹ Therefore, follow-up programs for children born preterm have been established, as preterm birth is one of the major risk factors for CP. Neonates born at term are also screened for early signs of CP in cases of hypoxic–ischemic encephalopathy or seizures.

Recently, a Canadian group developed a CP risk calculator for neonates born at term,² based on 12 perinatal variables. It has a sensitivity of 56% and specificity of 82%. However, the risk calculator still did not identify about 25% of children with CP born at term. This might be partially explained by post-neonatally acquired CP, but also to difficulty in early detection of CP in children without perinatal risk factors.

Using well-established CP register data from Canada, Suzuki et al. found the same proportion of children with post-neonatal CP in the subgroup without perinatal risk factors as in neonates with high risk of CP.³ This was unexpected and it would be important if this was confirmed in other registers that collect both perinatal and post-neonatal cases of CP. On the other hand, CP may be caused through different pathways and associated with diverse risk factors, with the post-neonatal event the only detectable one. Thus the child might be classified as having neonatal CP. An example is a child with CP due to a post-neonatal insult who has not had a genetic analysis performed. This child could have COL4A2 variant-related arteriopathy and therefore a prenatal/genetic cause of CP.

There is still much we do not know regarding the fetus. It would, potentially, have been useful to include data regarding growth patterns in utero,⁴ especially in the last half of pregnancy since there are no apparent associations with measurements in early pregnancy. Additionally, placental size at birth (an easy measurable indicator of placental dysfunction) might be included in a subsequent version of the CP risk calculator. Furthermore, data regarding genetic etiology or susceptibility ideally might be included if available in early life, as it is now evident that up to 30% of children with CP have a genetic contribution.⁵

Suzuki et al. found that children without ‘detectable’ risk factors more frequently have a focal brain injury with hemiparesis and communicate with words. This is surprising, as one could speculate that a focal injury would be associated with a preceding symptom/disease in the pregnant female or the child. Regarding strategy for early detection of CP in the group without perinatal risk factors, Suzuki et al. point to early handedness as the most important symptom for these children. Fortunately, the validated Hand Assessment for Infants tool is available and can be used with children as young as 3 to 12 months. Increased awareness among parents and health professionals that handedness during the first year of life may be a symptom of unilateral CP will be the best next step to identifying all children with CP as early as possible.