Introduction

Diseases in which there is a gradual loss of cognitive function can be described as progressive encephalopathies (PE). A number of toxic, infectious, inflammatory, and neoplastic disorders of the central nervous system (CNS) can have subacute to chronic presentations, and must be considered in the differential diagnosis of patients initially presenting with PE, but this chapter is primarily focused on the diagnostic challenge posed by patients with genetic disorders that present with progressive neurological dysfunction.

One challenge posed by these disorders relates to their being individually rare and thus generally unfamiliar to clinicians. A second challenge relates to phenotypic variability, as many of the inborn errors of metabolism (IEMs) that had been known by one or more “classic” phenotypes are now known to vary greatly in presentation, based on the degree of the enzymatic defect. For example, neonatal encephalopathy and multi-organ failure may result from the complete deficiency of an enzyme, whose partial deficit may cause only nonspecific psychiatric symptoms in adulthood.

Pathophysiology

Most genetic causes of PE can be classified as an IEM or neurodegenerative disorder (ND). The IEMs are themselves frequently divided into three groups, based on pathophysiology. In the first group are those disorders in which symptoms of acute or chronic intoxication are caused by the intracellular and extracellular (and thus measurable in blood, urine, and cerebrospinal fluid) accumulation of the compounds proximal to the defective enzyme. This includes errors of amino acid catabolism (e.g., phenylketonuria and maple syrup urine disease), organic acid catabolism (e.g., methylmalonic aciduria and propionic acidemia), urea synthesis (e.g., ornithine transcarbamylase deficiency and argininemia), sugar catabolism (e.g., galactosemia and hereditary fructose intolerance), metal transport (e.g., Wilson’s disease and Menkes’ disease), and porphyrin metabolism. Because the placenta acts to maintain homeostasis, these disorders are unlikely to cause embryonic toxicity. Most patients develop symptoms in infancy and childhood, following a symptom-free period whose length depends in part on the degree of enzyme deficiency. Other circumstances, such as fever, illness, and diet changes, can also influence the timing and severity of symptoms.

The second group of IEMs are those in which symptoms are due, at least in part, to the inability of the brain and other organs to produce or utilize sufficient energy for normal function. Energy deficiency can result from defective function of the mitochondria, including defects of pyruvate transport and modification, the Krebs cycle enzymes, fatty acid oxidation enzymes, and the respiratory chain enzymes that allow for aerobic metabolism. Energy deficiency can also result from defects in cytoplasmic enzymes, such as those responsible for glycogen synthesis, glycolysis and gluconeogenesis, insulin secretion and responsiveness, creatine synthesis and transport, and the pentose phosphate pathway. It is not uncommon for children with IEMs causing energy defects to present with congenital dysmorphism or cerebral dysgenesis.

The third group of IEMs are typically thought of as storage disorders, in which incompletely catabolized complex molecules accumulate within neuronal and extraneuronal tissues and cause progressive neurologic symptoms and somatic changes. This would include the mucopolysaccharidoses, the oligosaccharidoses, and the lysosomal storage disorders. Some authors expand this third group to include disorders of complex molecule synthesis and catabolism that do not result in measurable storage, including the Peroxisomal disorders, congenital disorders of glycosylation, and disorders of cholesterol biosynthesis.

Genetic disorders causing PE, which are not known to have a specific metabolic basis but which result in the progressive loss of neurons, usually demonstrated as progressive atrophy on neuroimaging, are classified as neurodegenerative. While the diagnosis of a ND previously relied solely on clinical features and expert pattern recognition, the past decades have seen elucidation of the genetic basis for most and the pathophysiologic basis for many. Those NDs in which the pathophysiology remains unclear are often subdivided into those affecting the brain homogenously (diffuse encephalopathies) and those tending to affect the cerebral cortex (poliodystrophies), cerebral white matter (leukodystrophies), basal ganglia (CORENCEPHALOPATHIES), cerebellum, or to preferentially affect the brainstem.

Epidemiology

Although the causes of PE are individually rare, the combined incidence of PE has been estimated to be as high as 1 in 2000 live births [Surtees, 2002]. Much of what has been published regarding these disorders has been retrospective and focused on individual conditions, providing little basis for a discussion of their collective epidemiology. A few studies have been notable exceptions.

An early paper examining the experience with PE at two large academic centers in the United States found that, of 1218 admissions to their child neurology services over the course of 10 years, 341 patients were diagnosed with 1 of more than 50 disorders causing neurological dysfunction [Dyken and Krawiecki, 1983]. Table 44-1 shows the results of their analysis of the relative frequency of the various diagnoses. Although 72 percent of the cases studied had a genetic or metabolic disorder causing PE, the study also included a significant number of children with pure lower motor neuron syndromes and acquired injuries due to infection, immunologic disorders, refractory epilepsy, chronic environmental insults, nutritional deficiencies, and iatrogenic factors. A study from the Children’s Hospital of Lahore, Pakistan [Sultan et al., 2006], found that, of the 1273 children admitted to the neurology service from 2004 to 2005, 66 were diagnosed with PE and most received a specific diagnosis. The most common diagnoses, in descending order of frequency, were metachromatic leukodystrophy (14 cases), adrenoleukodystrophy (11), subacute sclerosing panencephalitis (8), Wilson’s disease (6), Friedreich’s ataxia (5), liposis (4), Gaucher’s disease (3), Alexander’s disease (2), and pantothenate kinase-associated neurodegeneration (PKAN) (2). More than half of the patients underwent funduscopic examination, electroencephalography, and cerebrospinal fluid examination as part of their diagnostic work-up.

Table 44-1 Diagnosis in 340 Cases of Developmental Regression

* Poliodystrophies = predominant cortical involvement.

† Leukodystrophies = predominant cerebral white-matter involvement.

‡ CORENCEPHALOPATHIES = predominant basal ganglia involvement.

§ Spinocerebellopathies = predominant spinal cord and cerebellar involvement.

Following the initial description in 1996 of 10 cases of new variant Creutzfeldt–Jakob disease (nvCJD) affecting young adults in the United Kingdom [Will et al., 1996], several countries instituted prospective surveillance programs to collect data on patients with PE to better identify additional cases of nvCJD. Although these studies have relied on reports from pediatricians and have been unable to describe absolute incidence or prevalence figures, they have reported relative prevalences within their areas. The first report from the surveillance done in the UK [Devereux et al., 2004] collected and analyzed pediatric cases of progressive intellectual and neurological deterioration (PIND) over a 5-year span. The cases included children who had:

The study excluded children with intellectual and neurological deterioration after a nonprogressive insult, such as encephalitis, trauma, or global hypoxic-ischemic injury, but did include children with seizure disorders who otherwise met the case definition and children carrying diagnoses that could be expected to lead to progressive deterioration in the future. Of the 798 cases collected, 577 had a confirmed diagnosis, 6 had definite or probable nvCJD, and 211 had no clear etiologic diagnosis at the time of publication but did not have clinical features suggestive of nvCJD. There were nearly 100 different confirmed diagnoses, but more than one-quarter of the cases were explained by the five most common: mucopolysaccharidosis type III (Sanfilippo’s syndrome), adrenoleukodystrophy, late infantile neuronal ceroid-lipofuscinosis, mitochondrial diseases, and Rett’s syndrome. Higher rates of prevalence and of consanguinity were reported in families of South Asian origin. A follow-up of the UK study [Verity et al., 2010a] reported a confirmed etiologic diagnosis in 1047 of the 2493 cases of PIND that had been collected by 2008, with nearly one-quarter of cases again explained by the five most common diagnoses: neuronal ceroid-lipofuscinoses, mitochondrial diseases, mucopolysaccharidoses, GANGLIOSIDOSES, and Peroxisomal disorders. The most recent update of the study [Verity et al., 2010b] reported that, after 12 years, 147 different etiologies were found to explain 1114 of the 2636 cases of PIND collected. In total, only 6 children with confirmed or probable nvCJD had been identified. The 30 most common diagnoses identified in the study are presented in Table 44-2.

Table 44-2 Common Diagnoses in 1114 Cases of Progressive Encephalopathy

A survey-based study conducted in Australia [Nunn et al., 2002] identified 230 cases of childhood PE in a 2-year period, with 134 patients having Rett’s syndrome, 20 having a lysosomal storage disorder, 16 having a leukodystrophy, and 15 having a mitochondrial disease. A study done in Oslo, Norway, gathered cases of pediatric PE over an 18-year period from the area’s one children’s hospital and from the national diagnostic laboratory for metabolic diseases [Strømme et al., 2007]. The authors excluded patients with diseases in which cognitive impairment was either atypical (e.g., spinocerebellar ataxia and spinal muscular atrophy) or typically seen only late in the course (multiple sclerosis). Also, unlike the studies already discussed, this study excluded disorders, such as regressive autism and Rett’s syndrome, in which intellectual deterioration may be seen early in the course but typically stabilizes. They reported a total of 84 cases of PE, of which they classified two-thirds as metabolic, one-third as neurodegenerative, and 2, both due to HIV/AIDS, as infectious. The metabolic and neurodegenerative cases were further subcategorized as shown in Table 44-3.

Table 44-3 Diagnoses in 84 Cases of Progressive Encephalopathy in Oslo, Norway

There were 28 children with disorders of subcellular organelles (23 lysosomal, 3 mitochondrial, and 2 Peroxisomal) and 27 with disorders of intermediary metabolism (11 organic acidurias, 6 fatty acid oxidation disorders, 4 urea cycle disorders, 4 galactosemia, and 2 unspecified). The neurodegenerative cases included 10 children with a specific diagnosis (1 ataxia telangiectasia, 2 Cockayne’s syndrome, 1 megalencephalic leukoencephalopathy with subcortical cysts, 3 microphthalmia and brain atrophy, 1 pontocerebellar hypoplasia and infantile spinal muscular atrophy, and 2 Schinzel–Gideon syndrome) and 17 in which only the portion of the CNS most affected could be specified (8 cerebellum, 3 cerebral cortex, 3 cerebral white matter, 1 basal ganglia, 1 cerebellum and basal ganglia, and 1 cerebellum and brainstem). Analysis of the study data found that there was a 7-fold increase in risk of PE in children of Pakistani origin, due largely to the predominantly autosomal-recessive inheritance pattern for causes of PE and the much higher incidence of reported consanguinity in that community [Strømme et al., 2010]. It was estimated that 30 percent of all cases of PE, and at least 50 percent of the cases in children of Pakistani origin, would have been prevented if the practice of consanguinous marrage were avoided.

The same authors [Strømme et al., 2008] used local population data to calculate an overall incidence rate for PE of 6.43 per 100,000 person years (95 percent CI 5.15–7.97), with the age-specific rates being highest for infants <1 year old (79.9 per 100,000 person years) and lowest for children over 5 years (0.65 per 100,000 person years). They also found that the age at diagnosis averaged 0.5 years for patients with metabolic diseases and 4.5 years for patients with neurodegenerative diseases, and that children with neonatal onset and metabolic etiology had the highest risk of mortality.