Advances in obstetric and neonatal care have been responsible for the improved survival of high-risk neonates. A major concern persists, however, that newer therapies may result in an increased number of permanently disabled infants. The earliest follow-up studies of preterm infants after the introduction of modern methods of neonatal intensive care in the 1960s described a decrease in adverse neurodevelopmental sequelae compared with that of the preceding era.112 During the 1980s and 1990s, there was a continued decrease in mortality, and thus the absolute number of both healthy and neurologically impaired survivors increased.35,109,110 Furthermore, the survival of increasing numbers of extremely immature infants with low birth weight resulted in a relatively high disability rate in the subpopulation of infants born weighing less than 750 g or born at less than 26 weeks’ gestation.49,53,150 Since 2000, mortality rates for infants with very low birth weight have leveled off. Several studies suggest declining rates of neurodevelopmental impairment, including cerebral palsy.111,116,146 Infants at highest risk for later neurodevelopmental problems resulting from perinatal sequelae include those who had severe asphyxia, severe intracranial hemorrhage, infarction or periventricular leukomalacia, meningitis, seizures, respiratory failure resulting from pneumonia, persistent fetal circulation or severe respiratory distress syndrome, and multisystem congenital malformations, as well as children born with extremely low birth weight or at extremely early gestational age (Boxes 68-1 and 68-2). The rates of health problems and neurodevelopmental sequelae are inversely proportional to both birth weight and gestational age (Tables 68-1 and 68-2). TABLE 68-1 Neurodevelopmental Outcomes by Birth Weight at 8 Years Data from Hack M, et al. Long-term developmental outcomes of low birth weight infants. Future Child. 1995;5:176. TABLE 68-2 Health Outcomes by Birth Weight at 8 Years Data from Hack M, et al. Long-term developmental outcomes of low birth weight infants. Future Child. 1995;5:176. In planning neonatal follow-up programs, various models of care are possible, but may be constrained by available resources.79 A minimal requirement for the clinical monitoring of outcomes is a periodic assessment of growth and neurosensory development during the first 2 years of life. The ideal is a comprehensive program involving all aspects of care, including well-baby care, evaluation of outcome, social and educational intervention, and therapy when needed. A home nurse visiting program, especially during the early post-discharge period, and parent support groups for selected high-risk conditions (e.g., children with chronic lung disease) also should be considered. There is evidence that educational enrichment during infancy and early childhood might improve the outcome of high-risk and preterm infants, especially those from socioeconomically deprived groups.70 Outcomes from different centers are heavily influenced by the demographic and socioeconomic profile of the parents, the regional incidence of extreme prematurity, the percentage of inborn patients at a given center, a selective treatment or admission policy, site-specific practice patterns, and the rate of follow-up.140 Intercenter differences in neonatal sequelae and outcome are well described.50,142 Regional results, rather than national or international cohort data, may therefore reflect a more accurate picture of outcome because they include all infants born in an area. This is the ideal situation, but such studies are rarely available in the United States. Individual centers should be aware of their own patients’ social risk factors and rates of neonatal morbidity and, if possible, maintain their own follow-up outcome data. Any evaluation of the outcome studies of high-risk infants must include the population status (inborn, outborn, or regional) and the choice of a comparison group that includes either a normal birth weight group or infants within a similar birth weight or gestational age range who do not have the condition or therapy under study. It also is essential to control for sociodemographic factors such as maternal marital status, ethnicity, and education and to consider possible genetic factors when evaluating cognitive outcome or school performance.90 Consideration of neonatal mortality is important for judging the aggressiveness and level of neonatal care, which might influence the quality of outcome of the survivors.87 Other factors to be considered are the rate of loss of infants to follow-up, the neonatal and postdischarge death rate, the age at follow-up, and the method of follow-up.147 Two years is the earliest age to get a fairly reliable assessment of neurodevelopmental outcome. At age 4 to 5 years, cognitive function and language can be better measured, and follow-up at age 7 to 9 years allows an assessment of subtle neurologic and behavioral dysfunction and school academic performance (Figure 68-1).59,115,137 Because it is impossible to provide ongoing high-risk follow-up care for all infants treated in the neonatal intensive care unit, specific criteria have been proposed to identify children at greatest risk for sequelae.95 Traditionally follow-up programs primarily targeted children with birth weight of less than 1500 g or gestational age of less than 32 weeks. However, therapies such as inhaled nitric oxide and extracorporeal membrane oxygenation have increased the demand for highly specialized follow-up clinics for term infants with persistent pulmonary hypertension, meconium aspiration, and sepsis. In addition, a growing number of infants with major congenital malformations such as congenital diaphragmatic hernia now survive the neonatal period to require intensive ongoing follow-up support. Centers with active research components could select additional candidates for follow-up in the high-risk clinic on the basis of participation in specific research studies. Because of the significant costs associated with evaluating all eligible follow-up patients in the clinic setting, parent and teacher questionnaires have been suggested. These questionnaires typically provide a checklist of various individual measures of health status and disability (Box 68-3).72,74 Intrauterine and neonatal brain growth failure and lack of later brain catch-up growth can affect cognitive functioning.14 Many infants with very low birth weight who are small for gestational age also have subnormal head growth, and brain growth failure can occur during the neonatal period. Catch-up brain growth can occur during infancy in infants with very low birth weight who are either appropriate or small for gestational age; however, as many as 10% of infants with very low birth weight who are appropriate for gestational age and 25% who are small for gestational age still have a subnormal head size at 2 to 3 years of age that persists at school age. Poor growth attainment is especially apparent in the child with an extremely low birth weight or gestational age (Figure 68-2). Growth after discharge is a good measure of physical, neurologic, and environmental well-being. To promote optimal catch-up growth of high-risk infants, neonatal nutrition must be maximized. This is especially important because catch-up of head circumference occurs only during the first 6 to 12 months after the expected date of delivery. In recent years, increased-calorie postdischarge formulas have been introduced. These formulas have been associated with improved growth to 9 months. Reports of osteopenia or rickets of prematurity have increased with the improved survival of extremely premature infants whose birth precedes the period of greatest in utero mineral accretion. Although rickets of prematurity appears to be a self-resolving disease, postdischarge formulas with higher calcium and phosphorus content have enhanced growth and bone mineral accretion among preterm infants.12 However, it is important to balance appropriate catch-up growth and mineral supplementation with an avoidance of excessive rates of growth which are concerning for promotion of later metabolic syndrome and obesity. A high incidence of transient neurologic abnormalities, ranging from 40% to 80%, occurs in high-risk infants. These include abnormalities of muscle tone such as hypotonia or hypertonia (occurring as poor head control at 40 weeks’ postconceptional age), poor back support at 4 to 8 months, or a slight increase in muscle tone of the upper extremities. Because some degree of physiologic hypertonia normally exists during the first 3 months, it may be difficult to diagnose the early developing spasticity related to cerebral palsy.77 Children who will later develop cerebral palsy often initially have hypotonia (poor head control and back support) and only later develop spasticity of the extremities. Spasticity during the first 3 to 4 months is, however, a poor prognostic sign. Persistence of primitive reflexes also might be a sign of early cerebral palsy. Although mild hypotonia or hypertonia persisting at 8 months usually resolves by the second year, it might indicate later subtle neurologic dysfunction.22 Major neurologic sequelae can usually be diagnosed during the latter part of the first year of life or even earlier if they are very severe. Major neurologic disability is usually classified as cerebral palsy (spastic diplegia, spastic quadriplegia, or spastic hemiplegia or paresis), hydrocephalus (with or without accompanying cerebral palsy or sensory deficits), blindness (usually caused by retinopathy of prematurity), seizures, or deafness (Box 68-4). The intellectual outcome can differ greatly according to neurologic diagnosis. For example, children with spastic quadriplegia usually have severe developmental delay, whereas children with spastic diplegia or hemiplegia may have better mental functioning. Cognitive function is not easily measurable until after 2 to 3 years of age, especially among neurologically impaired children. Most neurologic problems either resolve or become permanent during the second year of life. During the second year, the environmental effects of maternal education and social class begin to play a major role in the various cognitive outcome measures. Further problems could emerge during the school-age years. These include subtle motor, visual, and behavioral difficulties even among children with normal intelligence.119 These are best diagnosed and treated in a psychological and educational, rather than a medical, follow-up setting. Cerebral palsy, an umbrella term that refers to “a group of non-progressive, but often changing, motor impairment syndromes secondary to lesions of the developing brain,” occurs about 70 times more frequently among infants with extremely low birth weight than among controls with normal birth weight.100 Risk factors include periventricular echolucencies noted on cranial ultrasound, intrauterine and neonatal infection, hypotension, severe respiratory distress, hypothyroxinemia, postnatal corticosteroid exposure, and multiple gestation.24,39,48,127,128 Despite these identified factors, most cerebral palsy cases, especially among term infants, do not have a readily identifiable cause. Protective factors could include maternal antenatal corticosteroid therapy, preeclampsia, and antenatal magnesium sulfate.31,91,143 Although birth asphyxia has been identified as a frequent cause of cerebral palsy among term infants, low Apgar scores have correlated poorly with cerebral palsy for preterm infants.34 Epidemiologic studies of cerebral palsy rates have been hampered by disagreement over both the specific diagnostic criteria used and the age at which a diagnosis should be made. Furthermore, studies differ on their estimates of the prevalence of cerebral palsy depending on the denominator used for calculation. For example, assessments may be reported relative to the number of live births, the number of NICU admissions, or the number of survivors evaluated. The risk of developing CP is higher in infants of lower gestational age. Typically, term infants have a risk of 1 per 1000 births, versus rates of 10 per 1000 for infants between 32 and 36 weeks’ gestation and 100 per 1000 for those born at extremely low gestational ages.104 Children with global hypotonia are usually not included in the diagnosis of cerebral palsy. Cerebral palsy was previously defined as mild, with no loss of function and independent walking; moderate, with functional disabilities requiring assistance for walking with aids or walkers; and severe, nonambulatory, requiring a wheelchair. Cerebral palsy was alternatively labeled disabling or nondisabling to incorporate a crude measure of functional impairment.107 With the exception of these descriptive terms, there was no reliable measure of the severity of motor disability or consideration of other cognitive or neurosensory problems associated with cerebral palsy. In 2004, an international workshop on the definition and classification of cerebral palsy proposed inclusion not only of motor disorders, but also of other associated deficits that may coexist, including seizures and cognitive, perceptual, sensory (visual and hearing), and behavioral impairments.15 The 2004 classification also includes anatomic and radiologic findings and causation and timing of the lesion. This new system has enhanced the evaluation of functional outcomes for children with cerebral palsy.80 In the future, its use should also improve studies of trends in the rates of cerebral palsy and its correlates. The diagnosis of cerebral palsy is usually delayed until motor development has been established. The minimal age before a definitive diagnosis can be made should be at least 3 years and preferably 5 years of age. This is because in some cases the neurologic findings may decrease or disappear by 5 years of age, and in other mild cases the findings may only become apparent later.13,108,128,130 The longest study of trends in the rates of cerebral palsy has been that of Hagberg and colleagues. They monitored the rates in western Sweden in a series of nine reports from 1954 until the most recent period of 1995 to 1998.61,64 During the 1950s, very few of the children with cerebral palsy in the western Swedish register were born before 28 weeks’ gestation, whereas by 1995 to 1998, 20% to 25% of these children were born at this extremely low gestation, evidence of the increase in survival of these infants. Survival increased progressively during the periods of study. Overall, the prevalence of cerebral palsy among preterm infants decreased between the periods 1954 to 1958 and 1967 to 1970, partly owing to discontinuation of various iatrogenic therapies such as prolonged starvation and discontinuation of limitation of oxygen thought to cause retinopathy of prematurity. After the introduction of methods of neonatal intensive care and the increase in survival of infants of extremely low birth weight and gestation, the rates of cerebral palsy increased by 1987 to 1990, with an increase in cases with severe multiple handicaps. Similar trends were noted by others.28 The prevalence then decreased significantly by 1995 to 1998. More recent data continue to show encouraging trends (Figure 68-3).111,116,146 The advent of modern neuroimaging techniques such as magnetic resonance imaging (MRI) offers improved potential to comprehensively visualize brain lesions associated with cerebral palsy. Cranial ultrasound was judged as normal in 35% of children with cerebral palsy in the entire EPIPAGE study of all preterm children born below 33 weeks’ gestation.4 In a systematic review of MRI studies in children with cerebral palsy, 90% of preterm-born children had periventricular lesions (see Chapter 69).78 Isolated motor disorders, such as developmental coordination disorder (DCD), which affects approximately one third of all preterm children, are more common than cerebral palsy. By definition, these children have no neurosensory impairment and demonstrate intact cognitive function. They display a variety of fine and gross motor delays resulting in difficulties with common motor tasks such as manipulating pencils or silverware, pedaling a bicycle, or performing routine motor tasks of daily living.144 Associations have been made between these “minor” motor disorders and cognition, behavior, and overall decreased function at school age.44 One of the most widely used tools to classify gross motor function for children with cerebral palsy is the Gross Motor Function Classification System (GMFCS) introduced by Palisano.16,105,106 This tool defines motor function on the basis of self-initiated movement with particular emphasis on sitting, walking, and mobility using a five-level classification system in which criteria meaningful to daily living distinguish the levels. Distinctions are based on functional limitations, the need for hand-held mobility devices such as walkers, crutches, or wheeled mobility, and to a much lesser extent, quality of movement. Because classification of motor function is dependent on age, separate descriptions are applied over a variety of age ranges. The focus of GMFCS is on determining which level best represents the child’s present abilities and limitations. Emphasis is placed on usual performance in home, school, and community settings rather than on what the children can do as their best capability. An example of the classification system used for toddlers is presented in Figure 68-4.
Early Childhood Neurodevelopmental Outcomes of High-Risk Neonates
Birth Weight (kg)
<1
1-1.49
1.5-2.49
≥2.5
Neurologic abnormality (%)
20
15
8
<5
Cerebral palsy (%)
>5
4
2
<0.4
Intelligence
Mean IQ
88
96
96
103
IQ <70 (%)
13
5
5
0-3
Behavioral problems (%)
29
28
29
21
Birth Weight (kg)
<1
1-1.49
1.5-2.49
≥2.5
Asthma (%)
17
18
12
11
Rehospitalization, previous year (%)
7
7
5
2
Limitation of >1 activity of daily living because of health (%)
46
34
27
17
Medical Problems
Physical Growth
Neurodevelopmental Outcome
Transient Neurologic Problems
Major Neurologic Sequelae
Assessment of Functional Outcomes
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Early Childhood Neurodevelopmental Outcomes of High-Risk Neonates
