Clinical Features

Cerebral palsy (CP) is a disorder of the development of movement and posture that causes activity limitation; it is attributed to nonprogressive disturbances that occurred in the developing fetal or infant brain (Bax et al., 2007; Rosenbaum et al., 2007). The motor disorders in CP are often accompanied by mental retardation (52%), epilepsy (45%), ophthalmological defects (e.g., vision impairment, motility problems, strabismus) (28%), hearing impairment (especially in those with very low birth weight, kernicterus, neonatal meningitis, or severe hypoxic-ischemic insults) (12%), and speech, oromotor impairments, and language disorders (38%) (Ashwal et al., 2004). Oral-motor problems including feeding difficulties, swallowing dysfunction, and drooling can affect nutrition and growth, oral health, respiration, and self-esteem (Ashwal et al., 2004). The disturbances of movement and posture include spasticity, hypotonia, ataxia, and dyskinesia. The likelihood and severity of the associated deficits generally parallel the severity of the motor impairment.

In birth cohorts from developed countries, the prevalence of CP is 1 to 2 per 100 live births and 100 in 1000 surviving infants less than 28 weeks gestational age (Paneth et al., 2006). This is comparable to figures from the ninth and most recent report from the western Swedish study on the prevalence and origin of CP in a population-based study of 88,371 live births occurring from 1995 to 1998 that reported a prevalence of 1.92 per 1000 live births. (Himmelmann et al., 2005). Gestational age-specific prevalence was 77 per 1000 for children born before 28 weeks. The prevalence of CP was 20% in those born at 24 to 26 weeks of gestation (Ancel et al., 2006), 40 per 1000 for children born at 28 to 31 weeks, 7 per 1000 for children born at 32 to 36 weeks, and 1.1 per 1000 for children born after 36 weeks. Spastic hemiplegic CP occurred in 38%, diplegic CP in 35%, quadriplegic CP in 6%, dyskinetic CP in 15%, and ataxic CP in 6%. For the first time in this series of long-term cohort studies, hemiplegia was the most common CP type, owing to the decline in preterm diplegia. There was also an increase in full-term dyskinetic CP compared with pre-1994 cohorts.

Studies of childhood cognitive and behavioral outcome in children with CP demonstrate an increased frequency of attention deficit hyperactivity disorder (ADHD) (20%) as well as other learning issues, particularly nonverbal learning disabilities (NVLDs), developmental coordination disorders (DCDs), and visuospatial problems. Those born at 23 to 25 weeks gestation have the highest frequency of CP and cognitive difficulties (Johnson et al., 2009; Vohr, 2010). During the 1990s, increased disability in very preterm infants was notable and probably due to increased survival rates (Wilson-Costello et al., 2005). However, since 2000, neurodevelopmental impairment has decreased. Improved outcomes reflect increased antenatal steroid use and cesarean section delivery, as well as decreases in sepsis, severe cerebral ultrasonographic abnormalities, and postnatal steroid use (Wilson-Costello et al., 2007). Preterm infants followed into adulthood show increased numbers of neurological signs in very preterm–born adults compared to term-born controls and are strongly associated with lower intelligence quotient (IQ) (Allin et al., 2006a). Furthermore, young adults who were born very preterm have different personality styles from their term-born peers; they are at an increased risk for psychiatric difficulties (Allin et al., 2006b). Even with these caveats, outcome has significantly improved over the past 2 decades.

Diagnosis and Etiology

A history of delayed motor milestones and the clinical examination are the basis for the diagnosis of CP. Both the Gross Motor Function Classification System (GMFCS), which defines five levels of motor function (Palisano et al., 1997), and a measure of upper-extremity function such as the ABILHAND-Kids or the Bimanual Fine Motor Function Classification are useful in diagnosing CP. The National Perinatal Collaborative Project, monitoring more than 50,000 pregnancies, documented that among the 200 children with CP, the most important predictors of CP were maternal intellectual impairment, birth weight less than 2000 g, a malformation of any organ system, and breech presentation or delivery (Croen et al., 2001). Commonly used indicators of perinatal hypoxia-ischemia such as abnormal fetal heart rate, presence of meconium, and low Apgar scores did not predict CP and were relatively common in healthy infants. Almost 95% of infants with 5-minute Apgar scores of 3 did not have CP. Even when the Apgar scores are low, the CP risk is minimal unless the infant continues to have neurological problems (e.g., lethargy, hypotonia, poor suck, seizures) in the newborn nursery. In addition to perinatal hypoxic ischemic encephalopathy, other documented causes of CP in the term infant include brain malformations, genetic and metabolic disorders, intrauterine (Nelson, 2009) and perinatal infection (congenital infections, maternal infection during pregnancy and labor, and neonatal meningitis), and stroke, generally resulting in hemiplegic CP (Nelson, 2007). Specific risk factors for stroke in both term and preterm infants includes history of infertility, oligohydramnios, preeclampsia, prolonged rupture of membranes, cord abnormality, chorioamnionitis, and primiparity. Twin gestation is an independent risk factor for CP in both preterm and term infants, as is a vanishing twin in the survivor (Blickstein, 2002; Bonellie et al., 2005). Among preterm infants, risk factors for CP (Hagberg et al., 2009) include long-term ritodrine tocolysis, severe respiratory distress syndrome, low umbilical cord pH (Malin et al., 2010), postnatal steroid therapy (per contra Kent et al., 2005), patent ductus arteriosus requiring surgical ligation, peri-intraventricular hemorrhage, moderate to severe ventricular dilatation, periventricular leukomalacia (PVL), and need for home oxygen (Locatelli et al., 2010; Takahashi et al., 2005; Tran et al., 2005; Vohr, 2010). In the Swedish cohort study (Himmelman et al., 2005), the origin of CP in children born at term was considered to be prenatal in 38%, peri/neonatal in 35%, and unclassifiable in 27%. Among preterm-born children, the etiology was prenatal 17%, peri/neonatal 49%, and unclassifiable 33%.

Metabolic or genetic causes for CP occur infrequently (i.e., 0%-4%). In almost all such cases, there are atypical complaints such as features in the history of a progressive rather than a static encephalopathy, findings on neuroimaging representative of certain genetic or metabolic disorders, or a family history of childhood neurological disorder with associated “CP”. Metabolic and genetic studies are not routine in the evaluation of the child with CP. Consider metabolic and genetic testing when atypical features exist in the history or clinical examination. Evaluating CSF may be appropriate in children with dyskinetic CP of unknown etiology. Detection of a brain malformation in a child with CP warrants consideration of an underlying genetic or metabolic etiology. Of all CP types, the ataxic form most often has a genetic basis. Dopamine-responsive dystonia may mimic CP and is easily diagnosed and treated when suspected (Chaila et al., 2006). Other genetic or metabolic disorders that may mimic CP include: urea cycle disorders, sialidosis type 1, Sjögren-Larsson disease (fatty aldehyde dehydrogenase deficiency), metachromatic leukodystrophy, Krabbe disease (spastic CP), Lesch-Nyhan syndrome, glutaric aciduria 1 (dyskinetic CP), ataxia telangiectasia, Leigh syndrome and other mitochondrial disorders, and subacute sclerosing panencephalitis (ataxic CP) (García-Cazorla et al., 2009).

Brain imaging is useful in determining the cause and prognosis of CP (Krägeloh-Mann, 2007). Neonatal imaging predicts the localization and extent of PVL as well as the neurological outcome and extent of CP in later childhood (Ancel et al., 2006; Sie et al., 2005). Very low-birth-weight premature infants with either echolucency in the periventricular cerebrum or moderate to severe ventricular enlargement have a greater than 50% risk of developing CP. (Hintz and O’Shea, 2008). However, one-third of infants with CP have no ultrasound abnormalities (Ancel et al., 2006). Ultrasound abnormalities are present in preterms that develop disabling CP but not in preterms (usually male) who ultimately develop mild CP (Paneth, 2001). This suggests a different etiology for CP in these two groups. Hypothyroidism of prematurity is associated with twice the risk of echolucencies on ultrasound, and this group is at increased risk for CP (Leviton et al., 1999).

Based on available data, proposed practice parameters for the evaluation of children with CP include neuroimaging if an etiology is not already established. In a study of 273 children over 35 weeks gestation who underwent neuroimaging, one-third had normal scans; focal infarction was observed in 22% of the children with CP and 45% of those with hemiplegia, with the next most common finding being brain malformation (e.g., hydrocephalus, polymicrogyria, cerebellar abnormalities); 12% of the children had PVL and 5% a history of hypoxia-ischemia (Wu et al., 2006). Magnetic resonance imaging (MRI) demonstrates that brain malformations are relatively common in term infants with CP (10%). PVL is the most common finding in preterm infants with CP (Barkovich, 2006). Oxidative stress and excitotoxicity resulting from excessive stimulation of ionotropic glutamate receptors on immature oligodendrocytes are the most prominent molecular mechanisms for PVL (Fatemi et al., 2009). PVL in term-born children is often associated with neonatal resuscitation and respiratory distress (Tang-Wai et al., 2006), and the degree of disability is generally greater than in the preterm (Krageloh-Mann et al., 1999). Multiple cerebral infarctions are associated with perinatal hypoxic-ischemic encephalopathy, whereas most unilateral infarctions are prenatal in origin (Nelson, 2007; Nelson and Chang, 2008; Volpe, 2009) and warrant evaluation for an underlying coagulopathy (Ashwal et al., 2004). Severe injury to the cerebellum occurs as a complication of extreme prematurity. The clinical features include striking motor impairment and variable degrees of ataxia and athetosis or dystonia, which represent a distinct clinical type of CP. Most children are severely impaired with cognitive, language, and motor delays, and microcephaly is the rule (Bodensteiner et al., 2005). Larger corpus callosum size has been associated with better motor performance in preterms (Rademaker et al., 2004). Outcome studies of children and adolescents born preterm suggest a correlation between the presence of MRI abnormalities, CP, and greater cognitive difficulties (Mirmiran et al., 2004; Woodward et al., 2006). Reduced cortical and cerebellar volumes may result from enhanced apoptosis or excitotoxic damage to highly susceptible immature neurons (Bhutta and Anand, 2001). Experimental models of hypoxic-ischemic encephalopathy indicate that neurons in the neonatal brain are more likely to die by delayed apoptosis extending over days to weeks than those in the adult brain. Neurons die by glutamate-mediated excitotoxicity involving downstream caspase-dependent and caspase-independent cell death pathways (Fatemi et al., 2009).

Prevention and Management

Prevention of CP has focused on prevention of prematurity, optimal treatment during the perinatal period, and the development of neuroprotective agents (Nelson and Chang, 2008). Magnesium sulfate may prevent CP in the very low-birth-weight preterm infant by altering oxygen free radicals, excitatory amino acids, and even cytokines (American College of Obstetricians and Gynecologists Committee on Obstetric Practice, 2010). Indomethacin may prevent intraventricular hemorrhage. Treating transitory hypothyroidism in the preterm may improve outcome, although those preterms with transient hypothyroidism do less well than those who are euthyroid (van Wassenaer et al., 2005). The effectiveness of inhaled nitric oxide is debated, with some studies showing efficacy and others not (Mestan et al., 2005; Nelson and Chang, 2008). A recent study (Takahashi et al., 2005) suggested that long-term ritodrine tocolysis, prolonged hypocarbia, and postnatal steroid therapy (per contra Kent et al., 2005) are associated with an increased risk for CP in the preterm infant, raising the possibility of the need for further changes in pre- and perinatal care. Infections during pregnancy, as well as chorioamnionitis at delivery, are risk factors for CP (Nelson, 2009); the resultant elevated cytokines are the presumed brain toxic agent (Gilstrap and Ramin, 2000; Girard et al., 2009). Improved management aimed at diminishing cytokine production may eventually prevent some cases of CP (Dammann and Leviton, 2004; Girard et al., 2009). The mode of delivery of early preterm infants may affect outcome. Some studies suggest that a policy of elective cesarean section in cases of premature rupture of membranes and breech presentations of fetuses weighing less than 1000 to 1500 g may be beneficial (Gabriel et al., 2004), but others do not (Nelson and Chang, 2008). Intraventricular hemorrhage, which usually occurs within 6 hours after birth and possibly during labor and delivery, may be preventable. Hypothermia treatment of hypoxic-ischemic encephalopathy, generally using a “cool cap,” appears to be improving outcome (Fatemi et al., 2009).

Data from several studies involving children with CP (n = 1918) found that about half (range 35% to 62%) of these children develop epilepsy. However, no evidence exists that electroencephalography (EEG) is useful in determining the etiology of CP (Ashwal et al., 2004) and plays no role in workup or management unless seizures occur. By contrast, since the frequency of associated cognitive and sensory deficits is high in the child with CP, appropriate screening should be undertaken.

Rehabilitative therapy influences outcome. The main goals of treatment are to improve the child’s motor function and modify the environment to improve mobility. Orthopedic surgery is sometimes required (Steinbok, 2006). Dorsal rhizotomy has been used to decrease spasticity and improve gait or make children who are wheelchair bound more comfortable. Botulinum toxin and baclofen also have a role in the treatment of spasticity (Gough et al., 2005; Lundy et al., 2009). Because of the high frequency of oromotor dysfunction, nutrition, growth, and other aspects of swallowing dysfunction should be monitored (Ashwal et al., 2004).

Assessing long-term outcome, Mesterman et al. (2010) accessed by phone 163 of all 204 patients diagnosed with CP (at age 0 to 5 years) between 1975 and 1994 in Tel Aviv. Of the 95 patients between 18 and 30 years at the time of the survey, one-third to one-half required help with activities of daily living like eating, toileting, and dressing, one-third did not walk independently, but 80% had 12 years of schooling or more in either regular or special education. Only five were married, most lived at home, but notably, the majority reported being quite satisfied with their lives (Miller, 2007; Shikako-Thomas et al., 2009). The factors that decrease life expectancy in CP are immobility, profound retardation, and the need for special feeding (Strauss et al., 2007). Mobile children with CP may live well into adulthood.

Clinical Features

The definition of a significant intellectual disability (ID) is a performance two standard deviations or more below the mean on age-appropriate standardized norm-referenced tests. The term global developmental delay (GDD) is usually reserved for young children (typically < 5 years of age), while the term intellectual disability (which will replace mental retardation in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders [DSM 5]) is usually applied to older children, in whom IQ testing results are more reliable (Shevell et al., 2003). The American Association of Intellectual and Developmental Disabilities (formerly known as the American Association of Mental Retardation) defines ID not as an absolute condition but as a construct that assists in educational planning and determines needs in terms of activities of daily living. The diagnosis therefore requires both low IQ (<70) and difficulties in adaptive behavior as expressed in conceptual, social, and practical adaptive skills (www.aaidd.org). This definition links the severity of ID to the degree of community support required to achieve optimal independence (Katz and Lazcano-Ponce, 2008). Mild ID indicates the need for intermittent support; moderate ID for limited support; severe ID for extensive support, and profound ID for pervasive support. The IQ definition of ID uses 100 as the mean and 15 as the standard deviation. Children with mild ID have an IQ of 55 to 69, those with moderate ID have an IQ of 40 to 54, those with severe ID have an IQ of 25 to 39, and those with profound ID have an IQ under 25. The level of functioning of children with an IQ around 60 to 70 varies significantly. About half require special education, and half attend regular school. Those in special schools are more likely to have a personal or family history of language delay. Similarly, IQ scores are only one factor in the differentiation between borderline intellectual function (BIF; IQ 70-84) and mild ID. An individual’s overall IQ score can be misleading because he or she may have strengths or weaknesses that markedly affect functioning, or there may have been even minor measurement errors on the test administration. An individual’s adaptive behaviors and functioning are of great importance when making this determination; an IQ of 70 is not necessarily a reliable cutoff between mild ID and BIF. Someone with an IQ of 69 may potentially function better overall than someone with a borderline range IQ of 71. An assessment of adaptive behaviors is crucial to the diagnostic process here too. Clinical interview with the individual and a collateral contact who knows them well can help assess adaptive functioning, as can standardized measures of adaptive behaviors (e.g., Vineland Adaptive Behavior Scales).

The prevalence of ID in industrialized countries is 0.3% to 3%. Mild ID represents the majority (85%), but roughly 0.4% of the general population is severely intellectually disabled. As a rule, those with severe ID are more likely to have a definable biological cause, whereas those with mild ID tend to come from socially disadvantaged backgrounds and often have a family history of BIF or mild ID (Online Mendelian Inheritance in Man [OMIM] website; Opitz, 2000; Stromme and Magnus, 2000). Environmental factors also play a role. For example, smoking during pregnancy is associated with more than a 50% increase in the prevalence of ID. Other environmental factors may include maternal alcohol use during pregnancy (fetal alcohol syndrome), lead toxicity, and iron deficiency. The risk for recurrence of severe ID is 3% to 10%, and of mild ID, 5% to 40%, suggesting there are many routes to mild ID. The ratio of boys to girls with ID, especially mild ID, is 1.4 : 1. Male excess is present in autism, syndromic X-linked ID (S-XLID) (associated with a specific phenotype), and nonsyndromic X-linked ID (NS-XLID). About 15% of males with ID have X-linked intellectual disability (XLID) (Stevenson and Schwartz, 2009). About 25% of all males with severe ID have XLID, and almost 50% of all cases of mild ID are due to XLID (Partington et al., 2000; Ropers and Hamel, 2005). The recurrence of ID in families with one previous child with severe ID is reported to be between 3% and 9% (Centers for Disease Control and Prevention [CDC], 2009).

Diagnosis and Etiology

Children with ID often have neurological and psychiatric comorbidities. Epidemiological studies suggest that as many as one-fifth of them have epilepsy by the age of 10 years (Airaksinen et al., 2000). The probability of developing epilepsy is fivefold greater for children with severe ID (35%) than for those with mild ID (7%). CP coexists in 6% to 8% of the mildly ID and as many as 30% of the severely ID. Microcephaly occurs in one-fifth of XLID syndromes. Macrocephaly also occurs secondary to increased brain volume or hydrocephalus. Syndromes without hydrocephaly include fragile X, Opitz FG, and Lujan. Syndromes with hydrocephalus include X-linked hydrocephaly-MASA spectrum and hydrocephaly-basal ganglia calcifications syndrome. Children with GDD and ID are at risk for physical disabilities. Impaired vision occurs in 15% to 50% and impaired hearing in about 20%. Thus a formal ophthalmological examination is appropriate. Audiometric assessment can be done by behavioral audiometry or brainstem auditory evoked response (BAER) testing (Shevell et al., 2003). Notably, cutaneous findings are fairly common in XLID, but cardiac anomalies are uncommon (Stevenson and Schwartz, 2009).

An increased prevalence of psychopathology and maladaptive behavior occurs in children with ID. In one epidemiological study, almost 40% of those with ID also had a psychiatric diagnosis, 42% with severe ID and 33% with mild ID (Stromme and Diseth, 2000). The most common comorbid psychiatric diagnoses were ADHD and autistic spectrum disorder (ASD). Overall, autistic features are reported in 9% to 15% of the mildly ID and 12% to 20% of the severely ID. Table 61.1 details specific cognitive and behavioral problems in several common genetically defined ID syndromes, along with the possible neuropathological basis of these disorders.

Etiology is ultimately determined in anywhere from 10% to 81% of children with GDD/ID. Evaluation of ID should be sequential; key elements include the medical, family, and developmental histories, dysmorphology and neurological examinations, and appropriate laboratory and neuroimaging tests (Moeschler and Shevell, 2006; Shevell et al., 2003). The latter can include careful metabolic evaluation together with neuroimaging studies, EEG, cytogenetic studies, and genetic and ophthalmological consultations as appropriate (Mao and Pevsner, 2005; Shevell et al., 2003). Auditory and visual function must be determined, since these are common comorbidities. If a child was born in a locale without universal newborn screening, consider a screening metabolic evaluation that includes a capillary blood gas, serum lactate and ammonia levels, serum amino acids and urine organic acids, and thyroid function studies. An EEG is appropriate when the history suggests possible seizures, paroxysmal behaviors, or an underlying epilepsy syndrome.

Genetic defects are important causes of ID, and recent years have seen important progress in identifying the genes involved in ID. As of June 2010, the entry mental retardation results in 1647 hits on OMIM. If a family history of consanguinity exists or a close family member (sibling, aunt/uncle, or first cousin) is known to have GDD, testing specific to the known disorder should be performed. When there is a family history of unexplained GDD, cytogenetic testing (e.g., high-resolution chromosome studies, testing for subtelomeric deletions, fluorescent in situ hybridization (FISH) and even microarray-based cytogenetics) is appropriate. In the absence of a familial history of GDD, specific historical or physical findings are useful to direct testing (Van Esch et al., 2005). Observed dysmorphic features may prompt specific testing for such entities as Down syndrome, fragile X syndrome (FXS), Rett syndrome, Prader-Willi/Angelman (by FISH), or congenital hypothyroidism (Jones, 2006). Parental consanguinity, documentation of loss or regression of developmental milestones, or unexplained prior parental loss of a child are likely to be caused by a definable disease process, and thus a comprehensive genetic evaluation should be considered (Shevell et al., 2003). Epigenetic factors (e.g., chromatin remodeling) also play a role (Tsankova et al., 2007).

Increasingly sophisticated genetic testing is becoming available. Reviewing the literature on diagnostic studies in ID, Van Karnebeek et al. (2005) found that chromosome abnormalities detectable by classical cytogenetics were diagnostic in about 10% of patients. The median additional benefit of subtelomeric studies was 4.4%, although the rate of subtelomeric abnormalities in ID may be as high as 10% (Lamm et al., 2006; Shao et al., 2008). The yield for FXS screening is higher in those with more severe ID (4%) versus those in the borderline–mild ID group (1.0%) (Mastergeorge et al., 2010). Overall, the etiology of ID can be detected by classical molecular karyotyping and molecular testing for specific disorders in about 20%. Multiplex ligation-dependent probe amplification may be a rapid, accurate, reliable, and cost-effective alternative to FISH for the screening of subtelomeric rearrangements in patients with idiopathic ID (Palomares et al., 2006). Genomic microarrays, also called array comparative genomic hybridization, may also represent one of several novel technologies that allow rapid detection of chromosomal abnormalities such as microdeletions and microduplications from genomic DNA samples (Engels et al., 2007). More than 5000 chromosome copy number variations (CNVs) are now in a Toronto database (Li and Andersson, 2009).

Progress in the clinical and molecular characterization of XLID has outpaced progress in the delineation of ID due to genes on the other 22 chromosomes. Almost half of the estimated 200 XLID genes are identified, and another 20% have been regionally mapped (Stevenson and Schwartz, 2009). The combination of clinical delineation with gene identification and the development of gene panels for screening nonsyndromic XLID minimizes unproductive laboratory testing (Stevenson and Schwartz, 2009). To date, 87 genes have been associated with nonsyndromic or syndromic XLID or both, including FXS, Duchene muscular dystrophy, and Coffin-Lowry syndrome (Stevenson and Schwartz, 2009). Using imaging technologies and more careful scrutiny, some nonsyndromic XLID families were found to have syndromic findings with time. Eighteen genes have been associated with both nonsyndromic XLID and syndromic XLID. This apparent conundrum may be explained by the type of mutation, the location of the alteration in different domains of a gene, or observer inconsistencies (Stevenson and Schwartz, 2009).

Inherited metabolic diseases are responsible for 1% to 5% of unspecified ID, but the mean yield of metabolic investigations is only about 1% (Van Karnebeek et al., 2005). ID is rarely a unique symptom in inborn errors of metabolism. However, some exceptions should be kept in mind, including creatine transporter defect, some patients with adenylosuccinate lyase deficiency, 4-hydroxybutyric aciduria, and Sanfilippo disease type B (García-Cazorla et al., 2009). Table 61.2 lists some of the metabolic disorders that cause isolated ID. Some are potentially treatable.

The identification of ID-associated genes has provided insights into brain function. Two major types of ID genes exist: those leading to dysfunctional neurodevelopmental programs and brain malformations, and those that lead to alterations in molecular synaptic organization and plasticity mechanisms (Vaillend et al., 2008). Some metabolic disorders appear to influence the expression of one or the other group of genes (Stephen et al., 2003). Affected genes have roles in processes such as neuronal outgrowth, synaptic structure and function, and synaptic plasticity and learning and memory. Polymorphisms that predispose to ID but are insufficient to cause symptoms on their own might be present within the protein-coding regions of genes, within their regulatory regions, or in genes that encode small, regulatory ribonucleic acids (RNAs). Allelic variants of genes that are involved in XLID might be candidates for such polymorphisms (Ropers and Hamel, 2005).

Neuroimaging

Computed tomography (CT) contributes to the etiological diagnosis of GDD in approximately 30% of children, with the yield increasing if physical examination findings are present. MRI is more sensitive than CT, with abnormalities found in 48.6% to 65.5% of children with global delay. The chance of detecting an abnormality increases if physical abnormalities, particularly CP, are present (Shevell et al., 2003). With evolving positron emission tomographic (PET) methodology, further inroads into etiology may occur via assessment of glucose metabolism, serotonin synthesis, and receptor status, such as γ-aminobutyric acid (GABA) (Sundaram et al., 2005). Both structural and functional research imaging are likely to contribute (Gothelf et al., 2005).

Management

Available evidence demonstrates the benefits of early intervention through a variety of programs, at least with respect to short-term outcomes, and suggests that early diagnosis of a child with global delay may improve long-term outcome (Shevell et al., 2003). The management of children with ID focuses on finding the appropriate educational setting for children with mild ID, vocational training for those with moderate ID, and determining home or institutional placement for those with severe and profound ID.

Advances in genetic diagnosis have had immediate benefits for families, allowing for carrier testing, genetic counseling, prenatal diagnosis, and preimplantation genetic diagnosis. Some of the gene discoveries have also pointed to potential strategies for treatment. For example, newborn screening suggests that FXS occurs in 1 in 5000 males (Coffee et al., 2009). The mGluR5 hypothesis for FXS has led to the use of mGluR5 antagonists for the treatment of FXS. For example, R-baclofen, a GABA agonist that downregulates mGluR5 activity, has been anecdotally helpful in FXS; controlled trials are in progress. Minocycline lowers matrix metalloproteinase 9 (MMP9) levels at the synapse, which are upregulated in FXS. Minocycline treatment in patients with FXS and a recent survey of 50 patients treated with minocycline revealed improvements in behavior and language based on parent questionnaire in approximately 70% (Schneider et al., 2009).

Clinical Features

The diagnosis of an autistic spectrum disorder (ASD) currently requires the early onset of a triad of deficits: impaired sociability, impaired verbal and nonverbal communication, and restricted activities and interests (Rapin and Tuchman, 2006). The range of disabilities seen among children on the spectrum cannot be overemphasized. Indeed, the DSM-5 will emphasize the dimensional nature of the disorder, quantifying each feature as mild, moderate, or severe. ASD will become a dyadic disorder (social/communication deficits and fixated interests and repetitive behaviors) (Box 61.1) and will be diagnosed as with or without a language disorder, since not all ASD children have language disorders. Pragmatic language skills, however, will be incorporated into the social domain. The changes are based on research results which thus far have failed to document either pervasive developmental disorder–not otherwise specified (PDD-NOS) or Asperger disorder as separate biological entities. In addition, removing the age criteria allows for inclusion of children, generally those functioning at a higher intellectual level, who present outside of the current age cutoff.

The reported frequency of ASD has increased significantly in the past 2 decades. In the 1990s, the estimated frequency was about 1 per 1000, but the current estimate is about 1 per 100 or even higher. ASDs are 4 times higher in males. The reported increase is attributed to changes in diagnostic practices, with emphasis on the spectrum aspect of the disorder, different referral patterns (more evaluations of language delayed male toddlers), greater availability of services prompting increased utility of diagnosis, loosening of age at diagnosis requirement, and increased public awareness (Frombonne, 2005; Hertz-Picciotto and Delwiche, 2009; Newschaffer et al., 2007). In addition, some children who would have been previously and perhaps correctly diagnosed with a developmental language disorder are receiving ASD diagnoses (Bishop et al., 2008). However, these variables may not completely explain the increase. Environmental factors cannot be discounted (Szpir, 2006), although the theory that the measles, mumps, rubella (MMR) vaccine plays any role in the increase has been completely discredited (Observations, 2010). Both advanced maternal and paternal age may play a role in increasing the frequency of autism (Durkin et al., 2008). The high concordance in monozygotic twins, the increased risk for recurrence in siblings (≈5%-10%), a broader autistic phenotype in families with an autistic proband, which includes anxiety and mood as well as social style and obsessive characteristics (Daniels et al., 2008), and the association with a number of genetic disorders (Box 61.2) support a hereditary basis in many cases (Gillberg, 2010).

The symptoms of ASD, like other developmental disorders, often change with age. Sibship studies provide the basis for investigations of the behavioral manifestation of autism during the first year of life (Elsabbagh and Johnson, 2009). Atypical eye contact, decreased response to name, poor imitation, absent social smile, passive temperament, and delayed language are early markers of the disorder (Zwaigenbaum et al., 2005). Sibling studies suggest there are both early-onset (clear by 12 months) and late-onset (clear by 18 months) forms of autism (Elsabbagh and Johnson, 2009; Zwaigenbaum et al., 2009). Approximately a third of autistic children regress between ages 1 and 3 years, with a peak at 2 years (Lord et al., 2004). Conversely, some toddlers and preschoolers with typical symptoms of autism may no longer appear autistic by school age. They may seem a bit odd and have peculiarities of language prosody and pragmatics. Social skills may be tenuous. NVLDs or ADHD may become the school-age diagnosis (Helt et al., 2008; Lord et al., 2006). Most, however, retain the typical features of autism, particularly those who are intellectually impaired. The natural history of Asperger disorder is less well documented. Often the diagnosis is delayed until late childhood, adolescence, or even adulthood because language is not delayed. NVLDs or ADHD may be the apparent presenting complaint (Martin and Klin, 2004). The characteristic overfocus in a particular or sometimes peculiar interest area may escalate over time, but it may also be the key to a special form of adult success. Outcome studies of children diagnosed with ASD suggest that although 40% show overall improvement during adolescence, as many as a third may deteriorate (Howlin, 2005; Rapin and Tuchman, 2006). The onset of seizures can cause the deterioration. Psychiatric comorbidities commonly develop: depression, bipolar disorder, anxiety, schizophrenia, and even catatonia (Billstedt et al., 2005; Hutton et al., 2008). A minority of patients, usually the higher-functioning group, improve significantly during adolescence. Approximately two-thirds of adults with autism show poor social adjustment (limited independence in social relations), and half require institutionalization. Autistic adults who are not intellectually impaired tend to improve more than those who are intellectually impaired. Prior to the broadening of the diagnostic category, even though higher-functioning individuals with autism and Asperger disorder had the best outcome, only 15% to 30% had fair to good outcomes, and only 5% to 15% became competitively employed, led independent lives, married, and raised families. Psychiatric problems were common in this group. Probably, some “odd” adults go undiagnosed in childhood and adolescence, thus increasing the proportion of those with ASD who ultimately function in the mainstream. Some are highly productive and original in their work (Billstedt et al., 2007; Seltzer et al., 2003). A recent report that assessed outcome in a cohort with normal IQ documented better adult functioning. Half of the participants were rated as “very good” or “good” on a global outcome measure. However, there was considerable variability in measured cognitive ability over time. Over half the sample had large gains or losses in cognitive ability of greater than one standard deviation (Farley et al., 2009).

Specific Clinical Features

Diagnosis and Etiology

Management

Preschool children with ASDs should receive special education in a therapeutic nursery or a home-based behavioral modification program (Handelman and Harris, 2002). Applied behavior analysis (ABA), which uses multiple prompting paradigms, reinforcement schedules, and imitation and modeling, and Treatment and Education of Autistic and Related Communication-Handicapped Children (TEACHH), to a lesser degree, have been validated as effective therapies in children with ASD ranging in age from 1 to 7 years (Callahan et al., 2010; Eikeseth, 2009). Early social intervention (ESI) therapies are under development (Dunn, 2005). Social skills groups and video modeling (video vignettes on TV or computer) appear to be effective therapies (Reichow and Volkmar, 2010). Animated mechanicals may enhance the emotional skills of children with ASD (Golan et al., 2010). Table 61.3 lists medications that have been helpful in some children.

Table 61.3 Medications for Autism

Modified from Soorya, L., Kiarashi, J., Hollander, E., 2008. Psychopharmacologic interventions for repetitive behaviors in autism spectrum disorders. Child Adolesc Psychiatr Clin N Am 17, 753-771.

Dyslexia

Nonverbal Learning Disabilities

Developmental Coordination Disorders

Dyscalculia

Attention Deficit Hyperactivity Disorder

Evaluation and Etiology

History is the primary basis of ADHD diagnosis. Since the disorder must affect functioning in at least two settings, multiple informants are required for diagnosis. In addition to an interview, questionnaires for parents, teachers, and the child or adolescent him/herself are useful for making the diagnosis. Given the high rate of ADHD comorbidities (e.g., anxiety and mood disorders, oppositional defiant disorder, conduct disorders), both narrow-band questionnaires that specifically assess for ADHD and broad-band questionnaires that assess not only for ADHD but for other psychiatric comorbidities as well should be used (Nass, 2006) (Fig. 61.1). The traditional neurological examination is generally normal, although soft signs are common (e.g., synkinesis during repetitive finger tapping and sequencing tasks and choreiform movements with the arms extended). Neuropsychological testing is not necessary unless comorbid learning issues are suspected. The neuropsychological profile reveals a normal IQ, but low scores are common on the Wechsler IQ subtests that demand attention or rapid processing (e.g., digit span, coding, arithmetic, symbol search). Executive frontal lobe functioning—the ability to initiate, inhibit, sustain, and shift attention, working memory, and organizational skills—is often impaired (Crosbie et al., 2007). Measures of spatial working memory, stop task response suppression,* continuous performance tasks (CPT), Stroop naming speed, Reitan trails B score, and mazes score have been found to correlate most highly with a clinical ADHD diagnosis. CPTs like the TOVA (Testing of Variables of Attention Deficit) or the Connors Continuous Performance Test are not useful as isolated diagnostic measures because of the high rate of both false-positive and false-negative results.

Fig. 61.1 ADHD comorbidities.

Overall, neuropsychological deficits are consistent with dysfunction of the basal ganglia–frontal circuits. However, these neuropsychological test abnormalities do not really explain the cognitive neurobiology of ADHD, hence the search for endophenotypes (Castellanos and Tannock, 2002; Konrad et al., 2010; Sonuga-Barke, 2010). Some candidates for core deficits of ADHD include failure of inhibitory control, dysregulation of brain systems mediating reward and response cost, and deficits in arousal, activation, and effortful control. Better delineation of endophenotypes would not only enhance understanding of the disorder but would increase the power of genetic research to identify susceptibility genes (Crosbie et al., 2007).

Several medical and environmental causes of ADHD exist (Box 61.11). The pediatric neurologist should take particular note of its high frequency in infants who were born preterm (about 20%) and its relatively high frequency in patients with epilepsy (10%-20%) even before they receive antiepileptic medications.

Box 61.11

Possible Environmental Causes/Contributors to Attention-Deficit Hyperactivity Disorder

Chronic illness: Allergy, asthma, celiac disease, sleep disorders

Endocrine: Thyrotoxicosis, hypothyroidism, generalized resistance to thyroid hormone

Epilepsy

Genetic disorders: Turner syndrome, fragile X syndrome

Infectious disease: Otitis media, human immunodeficiency virus

Metabolic disorders: Phenylketonuria

Prematurity

Psychosocial adversity

Toxins, pre- and postnatal: Maternal smoking, drugs, alcohol; lead, mercury

Trauma: Concussion, closed head injury

From Mill and Petronis, 2008; Nass, 2006; Nass and Frank, 2010.

ADHD has a substantial genetic component. Reported heritability of ADHD ranges from 0.1 to 0.76. Approximately one-fourth of the first-degree relatives of a child proband with ADHD also has or had ADHD, generally the father or maternal uncle. As many as 10% of ADHD children probably carry the Tourette gene but may not have tics. Girls with the Tourette gene more often have obsessive-compulsive disorder, whereas boys have ADHD and tics.

Dopamine appears to be the primary neurotransmitter involved in ADHD (Cools, 2008; Solanto, 2002). Hence, genetic studies have focused on candidate genes involved in dopaminergic transmission (dopamine transporter and dopamine receptors genes). The dopamine D₄ receptor (DRD4) has been the most well studied and is prevalent in the basal ganglia–frontal networks implicated in the pathophysiology of ADHD (Albayrak et al., 2008). Recent data suggest that the genotype may even influence the physiological response to medication in ADHD (Gilbert et al., 2006).

Most structural imaging studies reveal abnormalities in frontal and striatal regions, but some have implicated the cerebellum and corpus callosum (Durston et al., 2004; Seidman, 2005). One recent longitudinal study shows that plasticity may play a role in outcome. Children with ADHD show relative cortical thinning in regions important for attentional control. Children with a worse outcome have “fixed” thinning of the left medial prefrontal cortex, which may compromise the anterior attentional network and encumber clinical improvement. Right parietal cortex thickness normalization in patients with a better outcome may represent compensatory cortical change (Shaw et al., 2006). Cortical volume, surface area, and folding throughout the cerebral cortex appear to be decreased, suggesting the disorder begins prenatally and affects neurogenesis (Wolosin et al., 2009).

Functional neuroimaging studies have assessed the degree of brain activation associated with neuropsychological tasks tapping attention and disinhibition. In many studies, the ADHD and control subjects do equally well on the experimental task. Thus, activation differences indicate group differences in the neural systems used to accomplish the task. Results of most of these studies are consistent with the structural studies locating abnormalities of brain activation in patients with ADHD in fronto–basal ganglia–cerebellar circuits (Biederman and Faraone, 2005; Biederman et al., 2006; MTA Cooperative Group, 2004).

Management

Medication, in particular psychostimulants, is the mainstay of treatment (Greenhill et al., 2002) (Table 61.7). The National Institutes of Health (NIH)-sponsored Multimodal Treatment Study of ADHD (MTA) has shown both behaviorally and by neuropsychological testing that stimulants work, and that they work better than behavioral modification (Biederman et al., 2006; Epstein et al., 2006). Approximately 75% of children respond to stimulants initially. During the initial 14 months of the MTA study, children receiving intensive behavioral management did not do as well with respect to the core symptoms of ADHD (inattention, hyperactivity, impulsivity) as those receiving stimulants, demonstrating that medication is critical to managing ADHD. Those children who had comorbid issues like anxiety or oppositional behavior needed both medication and behavioral management to do well. Some longitudinal data from MTA study suggest that the effectiveness of medication may decrease over time. At the year 3 outcome assessment about 2 years after the study treatment protocol ended, the different treatment groups no longer differed in terms of ADHD symptoms. Those who had received intensive medication management or combined medication and behavioral management did not fare better than those who had received intensive behavioral therapy or community management (whatever their private doctor prescribed). However, three subgroups that did not correspond to the initial treatment groups were identified at year 3. Group 1 had gradual improvement over time, with increasing and significant benefit from medication use at 3 years out (34% of the sample). Group 2 had a larger initial improvement that was maintained over time (52% of the sample). Group 3 returned to baseline symptoms after initial benefit (14% of the sample) (Fig. 61.2). So although the same treatment benefits were not present at year 3, most of the children were doing well nonetheless. Intensive medication management may only make a persistent long-term difference if continued with the same intensity as during the MTA’s initial 14-month study period. In contrast, starting or adding medication at a less-than-optimal intensity and/or too late in a child’s ADHD clinical course, particularly if a child’s behavior is deteriorating, may not only be ineffective but also (if not carefully examined in data analysis) even make medication appear to be associated with worse outcomes. Some children may eventually be able to stop medication, perhaps just because of early intensive treatment. Medication alone and combined therapy (medication and behavioral management) did decrease the number of ADHD children who developed oppositional defiant disorder at the 3-year follow-up (Molina et al., 2009).

Table 61.7 Treatment of Attention-Deficit Hyperactivity Disorder

Stimulants	Methylphenidate (Ritalin, Focalin, Concerta, Daytrana), dextroamphetamine (Dexedrine, Adderall, Vyvanse)
α-Agonists	Clonidine (Catapres), guanfacine (Tenex), Intuniv
Antidepressants	SSRIs; tricyclic antidepressants; bupropion (Wellbutrin); trazodone; SSNRIs (venlafaxine [Effexor], duloxetine [Cymbalta]); MAOIs (selegiline [Deprenyl])
Antimanic	Lithium carbonate (Lithium)
Mood stabilizers	Carbamazepine (Tegretol), divalproex sodium (Depakote), gabapentin (Neurontin), topiramate (Topamax)
Beta-blockers	Propranolol (Inderal), atenolol (Tenormin)
Anxiolytics	Buspirone (BuSpar); clonazepam (Klonopin), SSRIs
Atypical neuroleptics	Risperidone (Risperdal, Zyprexa) and others
Nonstimulants	Atomoxetine (Strattera), modafinil (Provigil)

MAOI, Monoamine oxidase inhibitor; SSNRI, selective serotonin and norepinephrine reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor.

Data from Biederman and Faraone, 2005; Dopheide and Pliszka, 2009; Swanson et al., 2006.

Fig. 61.2 MTA Study outcome of co-occurring disorders in children.

(Data from MTA Cooperative Group, 1999. Arch Gen Pysch 54, 1088.)

Side effects limit treatment (e.g., weight loss, rebound depression or irritability, flat affect) in 5% to 10%. Guidelines for stimulant use exist (Erenberg, 2006; Pliszka et al., 2006) (Fig. 61.3). Many different formulations of the stimulants exist, with differing duration of action (Fig. 61.4). Neither tics nor Tourette syndrome are contraindications to stimulant use in ADHD, although they may affect medication choices (Pliszka et al., 2006) (Fig. 61.5). Comorbid mood or anxiety disorders also affect their use (Brown, 2000; Pliszka et al., 2006). In addition to stimulants, other medications can be used to treat ADHD. Strattera is an accepted nonstimulant treatment, as are the α-agonists (clonidine, guanfacine, Intuniv). Tricyclics, selective serotonin reuptake inhibitors, serotonin-norepinephrine reuptake inhibitors, bupropion, and modafinil have all been used, generally in those who failed to respond to stimulants (see Table 61.7).

Fig. 61.3 Algorithm for the psychopharmacological treatment of ADHD.

(From Pliszka, S.R., Crismon, M.L., Hughes, C.W., et al., 2006. The Texas Children’s Medication Algorithm Project: revision of the algorithm for pharmacotherapy of attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 45, 642-657.)

Fig. 61.4 Duration in hours of action of stimulants.

Fig. 61.5 Algorithm for the psychopharmacological treatment of ADHD and comorbid tic disorder.

(From Pliszka, S.R., Crismon, M.L., Hughes, C.W., et al., 2006. The Texas Children’s Medication Algorithm Project: revision of the algorithm for pharmacotherapy of attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 45, 642-657.)

In addition to medication, treatment should focus on parent skills training, educational accommodations (extended time, separate testing site), cognitive behavioral therapy, social skills training, and behavioral modification. Behavioral modification requires setting goals, defining progress, and determining the incentives.