DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY

Transcription

1 DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY ORIGINAL ARTICLE The clinical presentation of mitochondrial diseases in children with progressive intellectual and neurological deterioration: a national, prospective, population-based study CHRISTOPHER M VERITY 1 ANNE MARIE WINSTONE 1 LESLEY STELLITANO 1 DEEPA KRISHNAKUMAR 1 ROBERT WILL 2 ROBERT MCFARLAND 3 1 Progressive Intellectual and Neurological Deterioration (PIND) Surveillance Group, Addenbrooke's Hospital, Cambridge, UK. 2 National Creutzfeldt-Jakob Disease Surveillance Unit, Western General Hospital, Edinburgh, UK. 3 Mitochondrial Research Group, Newcastle University, Newcastle upon Tyne, UK. Correspondence to Dr Christopher M Verity at Child Development Centre, Box 107, Addenbrooke's Hospital, Hills Road, Cambridge CB2 0QQ, UK. christopher.verity@addenbrookes.nhs.uk PUBLICATION DATA Accepted for publication 3rd June Published online 11th September LIST OF ABBREVIATIONS BPSU British Paediatric Surveillance Unit CJD Creutzfeldt-Jakob disease MELAS Mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes MILS Maternally inherited Leigh syndrome MTATP6 ATPase6 gene NARP Neuropathy, ataxia, retinitis pigmentosa PDH def Pyruvate dehydrodenase PIND Progressive intellectual and neurological deterioration POLG Polymerase gamma gene SURF1 Surfeit1 gene vcjd variant Creutzfeldt Jakob disease AIM Our aim was to study the clinical presentation, mode of diagnosis, and epidemiology of mitochondrial disorders in children from the UK who have progressive intellectual and neurological deterioration (PIND). METHOD Since April 1997, we have identified patients aged 16 years or younger with suspected PIND through the monthly notification card sent to all UK consultant paediatricians by the British Paediatric Surveillance Unit. Clinical details obtained from reporting paediatricians are classified by an Expert Group. RESULTS By July 2008, 2493 cases of PIND had been reported, among which there were 112 children (69 males, 43 females) with mitochondrial diseases presenting between birth and 14 years 7 months (median 12mo), divided into 13 subgroups. In some instances, clinical features were characteristic of mitochondrial disease, but many children presented non-specifically with combinations of developmental delay, hypotonia, failure to thrive, and seizures; 16 children had multisystem disease at presentation. Mortality was high: 40 children had died. Blood and or cerebrospinal fluid lactate measurements were abnormal in 87 children, and 47 of 78 brain magnetic resonance images showed increased basal ganglia signal. Definite diagnoses were usually made by muscle enzyme or genetic studies. INTERPRETATION This is a unique population-based study of the mitochondrial disorders that cause childhood neurodegenerative disease. It provides detailed information about the clinical presentation and investigation of these complex cases. Knowledge about the epidemiology of mitochondrial diseases has increased dramatically, 1 5 but there is still a lot to learn. Since 1997, we have conducted a prospective, population-based study in the UK of childhood neurodegenerative diseases 6 that has identified a group of children and young people with mitochondrial diseases, providing unique data about the clinical presentation and mode of diagnosis in this group. The data are of current interest because of reports that mitochondrial diseases are diagnosed surprisingly frequently in children with autism 7 10 and because of a high-profile case that linked vaccination to the onset of an encephalopathy and the development of autistic behaviours in a female later diagnosed as having a mitochondrial disorder. 11,12 Our study was set up in 1997 to perform surveillance for variant Creutzfeldt Jakob disease (vcjd) in children from the UK after Will et al. 13 reported 10 cases of a new variant of CJD. To identify all possible cases of vcjd in the UK we have studied children with progressive intellectual and neurological deterioration (PIND), using the active national surveillance mechanism of the British Paediatric Surveillance Unit (BPSU), established in Our surveillance method has provided a novel insight into many neurodegenerative diseases of childhood. METHOD British Paediatric Surveillance Unit A card listing conditions under surveillance is sent by the BPSU office in the Royal College of Paediatrics and Child Health to all consultant paediatricians in the UK. Between 2800 and 2900 cards are sent monthly, and over 90% are returned, reporting cases seen in the previous month or indicating no cases. The BPSU office informs the surveillance groups about reported cases and they obtain clinical information from the notifying paediatrician. The methodology has been described in detail elsewhere. 14, DOI: /j x ª The Authors. Journal compilation ª Mac Keith Press 2009

2 Progressive intellectual and neurological deterioration study The BPSU card asks paediatricians to report children with PIND (see Table I for the detailed case definition). The surveillance team obtains information from the reporting paediatricians by telephone interview or from the hospital notes. Data that identify patients are kept securely by the surveillance team. Clinical information is extracted, patients names are removed, and these data are held on a password-protected computer database. The PIND Expert Group of paediatric neurologists and a geneticist meet quarterly to study the anonymized clinical information, to identify any PIND cases that fulfil the criteria for definite or probable vcjd, and to assess all the other PIND cases, reviewing diagnoses made by the notifying clinicians and providing feedback about children who have not yet been diagnosed. The Expert Group agreed that 112 of the PIND children had mitochondrial diseases, and these children were separately reviewed by a specialist in mitochondrial disorders (RM). The PIND questionnaire includes questions about the symptoms when the child first presented to the reporting paediatrician, other relevant subsequent symptoms, abnormal physical findings, and about the results of investigations. In this paper, we have concentrated on the clinical presentation and the investigations that led to the diagnosis. Ethical consent for the PIND study was obtained from the Cambridge Local Research Ethics Committee (ref ) and from the Public Health Laboratory Service Ethics Committee. The research has been approved by the Cambridge Research and Development Committee. The PIND Research Group obtained Patient Information Advisory Group (PIAG BPSU 2-10(c) 2005) approval to obtain patient-identifiable data without prior parental consent under Section 60 of the Health and Social Care Act Table I: Progressive intellectual and neurological deterioration: case definition Any child (under 16y of age at onset of symptoms) who fulfils all of the following criteria: Progressive deterioration for more than 3mo with Loss of already attained intellectual or developmental abilities and Development of abnormal neurological signs excluding Static intellectual loss, e.g. after encephalitis, head injury, or near-drowning including Children who meet the case definition even if specific neurological diagnoses have been made Metabolic disorders leading to neurological deterioration Seizure disorders if associated with progressive deterioration Children who have been diagnosed as having neurodegenerative diseases but have not yet developed symptoms Reporting restricted to cases seen in the past month but including those whose condition began earlier (i.e. including old cases of children in follow-up if seen in that month). RESULTS Demographics The PIND Study started in April By July 2008, 2493 children with suspected PIND had been notified. The PIND Expert Group confirmed an underlying diagnosis in 1047 of these children, among whom there were 112 children with mitochondrial diseases, one of the five most common diagnostic groups causing PIND. The others were neuronal ceroid lipofuscinoses (n=133), mucopolysaccharidoses (n=97), gangliosidoses (n=97), and peroxisomal disorders (n=65). The male to female ratio was 69:43, with age at presentation to the reporting paediatrician ranging between birth and 14 years 7 months (median 12mo). Ethnic origin was reported in 87% of the children. Most children were white (n=60) and there were also 29 Asian and four black children, as well as four of other origin; in 15 cases ethnic origin was not stated. Thus, where known, ethnic minorities constituted about a third of the mitochondrial cases in our study, whereas they account for less than 10% of the total UK population. The family history in the mitochondrial cases was as follows: 19 children were born to parents known to be consanguineous; 18 children had siblings with reported mitochondrial diseases; one child with mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (MELAS) had a mother with the MELAS mutation; and one child with Leigh syndrome had a mother, a maternal uncle, and maternal aunt with Leigh syndrome. No child had a father with a diagnosed mitochondrial disease. Of the children with mitochondrial disorders, 40 had died by July 2008, leaving 72 survivors. In 2006, the UK population of children aged 15 years or less was , 16 giving a point prevalence of mitochondrial disease with a PIND presentation, in this paediatric population, of Distribution of mitochondrial diseases As shown in Fig. 1, children were classified into 13 different groups. The largest number of children (n=24) had clinical evidence of a mitochondrial disease but remained in the nonspecific group in spite of extensive investigations. In some children the classification was based on the clinical phenotype (e.g. cases of Alper syndrome and all but two of the Leigh syndrome cases), but in many it was based on abnormalities of muscle or fibroblast enzymes, or the results of genetic studies. Clinical features As is shown in Table II, it was common for children with mitochondrial disease to present rather non-specifically with various combinations of developmental delay (seen in 43 cases), hypotonia (seen in 41 cases), and or failure to thrive (seen in 27 cases); 25 had suffered seizures. Others presented with numerous symptoms and signs so the number of reported clinical features shown in Table II is greater than the number of children. The clinical presentation only partly depended on the age of the child. For instance, 20 children presented at birth or in the first month of life. In 10 of these cases, seizures were a prominent feature, five had a combination of hypotonia, developmental delay, and poor feeding, two had Clinical Presentation of Mitochondrial Diseases in Children with PIND Christopher M Verity et al. 435

3 PDH def, 7 Comp II def, 2 Comp I def, 6 Multi com def, 11 MTATP6, 15 Cyto oxidase def, 7 Leigh, 15 SURF1, 10 POLG, 4 Alper, 5 MELAS, 5 KSS, 1 Non-spec, 24 Figure 1: The classification of mitochondrial diseases in children with progressive intellectual and neurological deterioration. The chart shows how the diagnosis of mitochondrial diseases is evolving, with some cases being diagnosed by clinical phenotype (e.g. Alper syndrome [Alper], most of the Leigh syndrome [Leigh] cases in our series), some by enzyme abnormality (e.g. pyruvate dehydrogenase [PDH def], cytochrome oxidase [Cyto oxidase def]), and some by genetic studies (e.g. mutations in the Surfeit 1 [SURF1], polymerase gamma [POLG], and ATPase 6 [MTATP6] genes). The diagnoses are grouped on the chart to indicate clinical or genetic overlap, for instance the Alper syndrome cases diagnosed by clinical phenotype are placed next to cases with POLG mutations, which may present with an Alper syndrome picture. The ability to document such overlaps on the chart is limited because some recognized clinical phenotypes are genetically heterogeneous, for instance the 15 cases with mutations in the MTATP6 gene include those associated with NARP syndrome and with NARP maternally inherited Leigh syndrome. Multi com def, multiple complex ; Comp I def, complex I ; Comp II def, complex II ; Alper, Alper syndrome; MELAS, mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes; KSS, Kearns Sayre syndrome. metabolic acidosis, two had hypoglycaemia, one had a cardiomyopathy, one had arthrogryposis, one had sideroblastic anaemia plus renal failure, and one was identified because of a previously diagnosed sibling. Thirty-seven children presented between 1 month and 1 year of age. Of these, 23 had a combination of hypotonia, developmental delay, and poor feeding, in 10 the prominent feature was seizures (five who presented in the first year developed infantile spasms), two had cardiomyopathy, two had strabismus, one had tremor and nystagmus, one presented with apnoea, and one was identified via an affected sibling. Among the 55 children presenting after 1 year of age, developmental delay, hypotonia, abnormal gait, and seizures were the most common presenting features. There was a tendency for children to show the clinical features characteristic of particular mitochondrial diseases, but they did not always do so early in the course of the disease. For instance, among the 15 children with Leigh syndrome, only three presented with signs suggesting brainstem dysfunction (nystagmus 2, squint 1); of the 12 children with neuropathy, ataxia, retinitis pigmentosa (NARP) mutations, only three had ataxia clumsiness; and, of the five with a diagnosis of MELAS, just two presented with strokes. Of the five children with Alper syndrome, three presented acutely with prolonged seizures but none had early evidence of the liver dysfunction that is known to develop later. There was evidence of multisystem disease (defined as involvement of an organ outside the nervous system) in 16 of the 112 children at presentation: metabolic acidosis (excluding lactic acidosis) in five, cardiomyopathy in four, hypoglycaemia in three, hypocalcaemia in one, acute hepatic failure in one, renal failure in one, and hepatosplenomegaly in one. In 33 cases there were reports of exacerbation of illness in association with fever or minor illness. However, there were only two reports that the onset of clinical symptoms was temporally associated with vaccination. One child was admitted to hospital at 4 months of age with seizures, having had a triple vaccine (diphtheria tetanus polio) in that week; the child was apyrexial on admission but there was possibly a fever before then. The second child was admitted 2 weeks after a measles, mumps, and rubella (MMR) vaccination, having been unwell for a week; the child was febrile with uncontrolled seizures requiring ventilation and had evidence of a Norwalk virus infection. It was reported that three children had autistic features. Two of these were in the non-specific group. One of these presented at 4 years of age with concern about development; there was loss of speech and autistic features were noted. The other child started school at 4 years of age with dyspraxia and attention-deficit disorder, but speech became less clear and autistic features were noted together with a progressive extrapyramidal disorder. The third child had complex I, presenting initially with learning difficulties and then being suspected of having autistic spectrum disorder, with later deterioration of motor abilities at 10 years of age. In none of these three cases was there any mention of deterioration in association with a vaccination. Method of diagnosis Most of the children with mitochondrial diseases had been extensively investigated. The key diagnostic investigations are shown in Table III. Table IV shows how the results of these investigations contributed to the diagnosis in the 112 cases. Lactate levels were increased in the blood and or the cerebrospinal fluid (CSF) in 87 children, in 19 children the lactate level was raised in the CSF only, and in 17 children the lactate levels were normal in blood and CSF. In 75 children a muscle biopsy had been performed and mitochondrial respiratory chain enzymes were measured: in 31 of these the result was diagnostic. In the seven children with pyruvate dehydrogenase, the enzyme studies were performed on skin fibroblasts. Muscle histochemistry was rarely reported to show changes characteristic of mitochondrial disorders ragged red fibres were described in only three children (of whom one had cytochrome oxidase, one had multiple complex, and the other was in the non-specific group). 436 Developmental Medicine & Child Neurology 2010, 52:

4 Table II: Clinical presentation of 112 patients Subgroup No. Sex (M, F) Age at onset Presenting features reported by clinician Multisystem, n Association with vaccination Exacerbation with fever or minor illness, n Autism, n Non-specific 24 17M, 7F Birth to 14y 7mo Leigh syndrome 15 9M, 6F Birth to 3y 2mo MTATP M 5F Birth to mutations a 7y 7mo Pyruvate dehydrogenase Multiple complex Complex I Complex II Cytochrome oxidase Hypotonia 6, DD 4, FTT 3, gait disturbance ataxia 3, microcephaly poor head growth 3, autistic features 2, cortical visual impairment 2, loss of speech 2, seizures 2, acute encephalopathy + seizures 1, acute hepatic failure 1, cardiomyopathy 1, cataracts 1, cyanotic episodes 1, dyspraxia ADD 1, hypocalcaemia 1, hypoglycaemia 1, hepatosplenomegaly 1, jerky limb movements 1, left convergent squint 1, reduced visual function 1, tubular acidosis 1, renal failure 1, spastic paraparesis and pes cavus 1 DD 5, FTT 4, hypotonia 4, seizures 4 (acute onset, 2), nystagmus 2, respiratory distress 2, cardiomyopathy 1, choreoathetoid movements 1, coarctation 1, myoclonic jerks 1, pigmentary retinopathy 1, squint 1 DD 10, hypotonia 8, ataxia clumsiness 3, FTT 3, acute onset 3, seizures 2, apnoea 3, squint 2, acidosis 1, cardiomyopathy 1, dysarthria 1, encephalopathy 1, pigmentary retinopathy 1, ptosis 1 7 4M, 3F Birth to 12y Neonatal problems 4, acidosis 2, abnormal movements 1, antenatal hydrocephalus 1, cyanotic attacks 1, microcephaly 1, nystagmus 1, seizures M, F5 Birth to 13y DD 6, hypotonia 4, FTT 4, seizures 3, abnormal respiratory pattern 2, increased tone 2, optic atrophy 2, acidosis 1, arthrogryposis 1, cardiomyopathy 1, febrile encephalopathy 1, hearing loss 1, hydrocephalus 1, hypoglycaemia at birth 1, internuclear ophthalmoplegia 1, swallowing problems 1 6 5M, 1F 1mo to 11y Seizures 4, DD 3, FTT 2, autistic spectrum disorder 1, dystonia 1, hypotonia 1, nystagmus 1, swallowing problems 1, weakness 1 7 No No 5 No 1 No 8 No 2 No 2 No 5 No 3 No 0 Triple vaccine 1wk before: 1 Possible meningitis 3wk before: 1 2 1M, 1F Birth Hypotonia 2, seizures 1, hypoglycaemia 1, lactic acidosis 1 2 No No No 7 4M,3F Birthto 2y 5mo SURF1 mutation 10 4M, 6F 6mo to 1y 10mo FTT 5, DD 4, hypotonia 3, abnormal movements ataxia 3, nystagmus 2, acidosis 1, hydrocephalus 1, seizures 1 Hypotonia 9, DD 4, FTT 4, vomiting episodes 3, acute illness (fever + vomiting) 2, nystagmus 1, seizures 1, talipes 1 POLG 4 1M, 3F 9mo to 4y DD 2, hypotonia 1, FTT 1, status epilepticus 1, encephalopathy + fever 1 (2wk post-mmr + unwell 1wk with Norwalk virus) Alper syndrome 5 3M, 2F 8wk to 3y 5mo MELAS 5 4M, 1F 2y 8mo to 11y Kearns Sayre syndrome 0 No No No 0 No 5 No 0 2wk post-mmr: 1 2 No DD 5, acute presentation with prolonged seizures 3, hypotonia 3, FTT 1 0 No No No Stroke-like episodes 2 (monoplegia 1, hemiplegia hemianopia 1), acute encephalopathy 2, ketoacidosis 1, seizures 1, strabismus 1, pigmentary retinopathy 1, sensorineural hearing loss 1 1 M 7y 11mo Progressive external ophthalmoplegia (ptosis, retinitis pigmentosa), myopathic facies, ataxia, narcolepsy, hyperphagia, cognitive decline, behaviourial difficulties, constipation 0 No 2 No 0 No No No autistic spectrum: 1 Children often presented with many symptoms and signs so the number of presenting features is greater than the number of cases. As in Fig. 1, the diagnostic groups are placed together where possible to indicate clinical or genetic overlap (see Fig. 1 legend). a MTATP6 children had mutations in the ATPase gene as follows: 11 with NARP mutations 8993T>G, one with NARP mutation 8993T>C, two with NARP MILS (maternally inherited Leigh syndrome) mutation 9176T>C, and one with 9185T>C mutation. DD, developmental delay; FTT, failure to thrive; MMR, measles, mumps and rubella; SURF1, the Surfeit1 gene; POLG, mutation in the polymerase gamma gene; MELAS, mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes. Clinical Presentation of Mitochondrial Diseases in Children with PIND Christopher M Verity et al. 437

5 Table III: Investigations in the 112 patients Investigation Lactate levels Above normal 87 Above normal in blood and CSF 47 Blood only 21 CSF only 19 Normal in blood and CSF 17 Not done not available 8 Muscle enzymes Measured 75 Diagnostic 31 Not done not available 37 DNA studies Performed 50 Diagnostic 35 Skin fibroblast enzymes Diagnostic: (all: pyruvate dehydrogenase ) 7 Positive histology Muscle: ragged red fibres 3 Liver: consistent with Alper syndrome 1 Neuropathology: consistent with Alper syndrome 3 Imaging findings MRI 78 a Increased signal basal ganglia 47 White matter abnormality 8 Focal changes 7 Normal non-specific 12 Other 9 CT only 8 Atrophy 3 Focal changes 3 Basal ganglia abnormality 2 Not available no scan 17 a The table summarizes the principal abnormality found on each scan. Many of the scans showed more than one abnormality. MRI, magnetic resonance imaging; CT, computed tomography. However, in four of the five children with Alper syndrome, histological study was diagnostic in one child there were characteristic liver abnormalities and in the other three there were characteristic histological changes in brains studied at autopsy. DNA studies were performed in 50 children (usually on muscle biopsy specimens) and yielded a specific diagnosis in 35. There were 15 children with mutations in the ATPase 6 gene (MTATP6); of these, 11 had the NARP m.8993t>g mutation, one had the NARP m.8993 T>C mutation, two had the NARP MILS m.9176t>c mutation, and one had the m.9185t>c mutation. The other diagnostic mutations were SURF1 mutation (10: eight children and two siblings), POLG mutation (four children), Leigh syndrome (m.14453g>a m.14459g>a mutations, two children), and MELAS (m.3243a>g in three cases and m.3252a>g mutations in one child). In those patients with nuclear gene mutations, the family history was interesting. Of the 10 with SURF1 mutations, six had first-cousin parents (five Pakistani and one Indian), three were not born to consanguineous parents, and for one white child family history was not known. The seven children in the pyruvate dehydrogenase group comprised two Pakistani children born to first-cousin parents, one Indian child born to consanguineous parents, two white children whose n Table IV: Summary of the main factors that determined how the 112 cases were classified Factors Raised blood and or CSF lactate plus 33 characteristic MRI abnormalities Raised blood and or CSF lactate plus 29 suggestive a or diagnostic test b Characteristic MRI abnormalities plus 22 suggestive or diagnostic test Suggestive or diagnostic test 9 Raised blood or CSF lactate plus 8 relevant family history c Diagnosis made in sibling d 7 Characteristic MRI abnormalities 4 plus a clinical history suggestive of a mitochondrial disease Nr of cases Characteristic magnetic resonance imaging abnormalities are most commonly high signal in basal ganglia and or the brainstem. a A suggestive test is most commonly an abnormal muscle histology report or a respiratory enzyme abnormality that is not fully validated; b a diagnostic test is usually a validated respiratory chain enzyme abnormality or a positive genetic test; c relevant family history is family history of mitochondrial disease; d diagnosis made in sibling means that the specific mitochondrial disease was diagnosed in the sibling so that full tests were not done in the index case. parents were not consanguineous, and two for whom the family history was unknown. In contrast, of the four children with POLG mutations, three were not the children of consanguineous parents and in one case the parent s status was unknown: all were white. Brain MRI was performed in 78 children, and in 12 was reported as being normal or showing non-specific changes. Table III summarizes the most striking abnormality seen on each scan, although some of the scans showed more than one of these abnormalities. The most consistent abnormality reported on the brain scans was high signal in the basal ganglia seen in 47 MRI and in two of the eight children in whom only a computed tomography (CT) was performed. Some scans showed the features known to be associated with specific diagnoses. For instance, in all five children with MELAS, MRI and or CT showed focal abnormalities. Among the 15 children with Leigh syndrome, scanning revealed high signal in the basal ganglia in six and high signal in the brainstem in five; however, these abnormalities were not restricted to the Leigh syndrome group. DISCUSSION Strengths and weaknesses of the PIND study We have performed prospective surveillance using a well-tried BPSU system. 14,15 Our study gives an overall picture of the distribution of mitochondrial diseases that cause PIND in children in the UK without the selection bias inherent in case series from specialist centres. The BPSU surveillance card goes to all UK consultant paediatricians, and we have obtained high-quality data from specialist centres and laboratories in the UK to assess the diagnoses in children with PIND. We have demonstrated that children with these complex medical problems are usually being very carefully investigated. 438 Developmental Medicine & Child Neurology 2010, 52:

6 The study methodology has limitations. We know that there is underreporting of cases. 14,15 Many children with mitochondrial diseases do not deteriorate and so will not be reported to our study because they do not meet the PIND case definition. The BPSU card gets good response rates, but not all paediatricians report the cases they see. Another limitation is that we are completely dependent on the clinical work being carried out by others. We rely on information obtained from hospital records and laboratories. We make no contact with the patients or their families and we are not actively involved in investigating the children. We rely on the opinion of the local clinicians for instance when classifying children as being developmentally delayed or showing autistic behaviour. Finally, when we have classified cases into diagnostic groups we discontinue follow-up, so we are not able to report long-term outcome. Prevalence Mitochondrial disorders as a group are the second most common cause of PIND in our study, 15 but individually these diseases are rare. The point prevalence that we have calculated is certain to be an underestimate for the reasons given above. There are relatively few other studies of the epidemiology of mitochondrial disorders in children 1 5 and methodology differs, so it is not surprising that there are major differences between the studies. Distribution of mitochondrial diseases and mode of diagnosis Figure 1 demonstrates how the mode of diagnosis of mitochondrial disease is evolving. There were 24 children in the non-specific group, and all of them had a sufficiently characteristic constellation of clinical features and positive investigations to confirm a mitochondrial disease; however, a more definite diagnosis could not be made. Other children were classified on the basis of a recognized clinical phenotype for instance most of the children with Leigh and Alper syndromes. However, in an increasing number of children, the diagnosis is based on the study of muscle or fibroblast enzymes or as a result of genetic studies. Muscle biopsy was performed in 67% of our cases, similar to another case series in the UK. 17 Genetic testing was positive in 31% of our case series, which is much higher than previously reported The findings on brain imaging provided an important clue to the diagnosis in many of our cases, but it should be noted that MRI were reported to be normal or non-specific in 12 of the children with PIND with mitochondrial disease. Scoring systems have been devised to aid the diagnosis of mitochondrial disease, 20,21 but our study shows that when children first present they do not have many of the features thought to be characteristic of mitochondrial disease. The published scoring systems rely partly on the results of invasive investigations such as muscle biopsy. The results of these investigations are often specific enough to obviate the need for a scoring system to make the diagnosis. Clinical features Mitochondrial diseases often present in a non-specific way with combinations of the following: developmental delay, failure to thrive, hypotonia, and seizures. The clinical features typical of particular mitochondrial diseases may only appear with the passage of time, and evidence of multisystem disease was not always present early on. However, despite our relatively limited follow-up, mortality rates were high in our series: 40 of the children with PIND mitochondrial disease had died by July Recently there has been much interest in mitochondrial dysfunction in autistic spectrum disorders 7 10 and a possible association between immunization, mitochondrial disease, and the development of autistic behaviour. 11,12 The PIND questionnaire asked about the presenting clinical features and about the subsequent clinical course. It did not specifically ask about an association between immunization and the onset of clinical symptoms, nor did it ask about autism. However, in two children it was reported that there was a temporal association with immunization in one case onset of symptoms occurred in the same week as administration of a triple vaccine (diphtheria tetanus polio); in the other, symptoms occurred 2 weeks after an MMR vaccination in a child with a Norwalk virus infection. These two cases should be seen in context: 57 of our cases presented between birth and 1 year of age when the onset of symptoms is likely to be close to the time of immunization purely by chance. Also, in 33 cases there were reports of exacerbation of illness in association with fever or minor illness, suggesting a non-specific effect of illness on the course of some mitochondrial disorders. Autistic behaviour was reported in three children with PIND two in the non-specific group and one with complex 1 ; there was no mention of deterioration with vaccination in any of these three children. However, it should be noted that our study was not designed to identify children with autism. Indeed, most would be unlikely to meet the PIND case definition. It has been suggested that children with autism could be evaluated for mitochondrial disorders, 8 but this involves complicated and invasive investigations. CONCLUSIONS Fortunately, childhood neurodegenerative diseases are not very common, but they present a considerable diagnostic challenge to paediatricians. Among the 1047 children in our study who had a definite underlying diagnosis there were 112 (about 11%) with mitochondrial diseases. From this study our findings will help clinicians to diagnose children with mitochondrial disease who present with evidence of progressive intellectual and neurological deterioration. ACKNOWLEDGMENTS The PIND study is funded by the Department of Health in England. Many thanks to the BPSU of the Royal College of Paediatrics and Child Health and to Mr Richard Lynn, the Scientific Coordinator. The authors also wish to thank past and present members of the Expert Group. Our grateful thanks go to all the paediatricians who have reported cases to the study. Clinical Presentation of Mitochondrial Diseases in Children with PIND Christopher M Verity et al. 439

7 REFERENCES 1. Uusimaa J, Remes AM, Rantala H, et al. Childhood encephalopathies and myopathies: a prospective study in a defined population to assess the frequency of mitochondrial disorders. Pediatrics 2000; 105:(3 Pt 1) Chinnery PF, Johnson MA, Wardell TM, et al. The epidemiology of pathogenic mitochondrial DNA mutations. Ann Neurol 2000; 48: Skladal D, Halliday J, Thorburn DR. Minimum birth prevalence of mitochondrial respiratory chain disorders in children. Brain 2003; 126: Castro-Gago M, Blanco-Barca MO, Campos-González Y, Arenas-Barbero J, Pintos-Martínez E, Eirís-Puñal J. Epidemiology of pediatric mitochondrial respiratory chain disorders in northwest Spain. Pediatr Neurol 2006; 34: Schaefer AM, McFarland R, Blakely EL, et al. Prevalence of mitochondrial DNA disease in adults. Ann Neurol 2008; 63: Verity CM, Nicoll A, Will RG, Devereux G, Stellitano L. Variant Creutzfeldt Jakob disease in UK children: a national surveillance study. Lancet 2000; 356: Pons R, Andreu AL, Checcarelli N, et al. Mitochondrial DNA abnormalities and autistic spectrum disorders. J Pediatr 2004; 144: Lerman-Sagie T, Leshinsky-Silver E, Watemberg N, Lev D. Should autistic children be evaluated for mitochondrial disorders? J Child Neurol 2004; 19: Oliveira G, Diogo L, Grazina M, et al. Mitochondrial dysfunction in autism spectrum disorders: a population-based study. Dev Med Child Neurol 2005; 47: Comment on p Oliveira G, Ataíde A, Marques C, et al. Epidemiology of autism spectrum disorder in Portugal: prevalence, clinical characterization, and medical conditions. Dev Med Child Neurol 2007; 49: Poling JS, Frye RE, Shoffner J, Zimmerman AW. Developmental regression and mitochondrial dysfunction in a child with autism. J Child Neurol 2006; 21: Offit PA. Vaccines and autism revisited the Hannah Poling case. N Engl J Med 2008; 358: Will RG, Ironside JW, Zeidler M, et al. A new variant of Creutzfeldt Jakob disease in the UK. Lancet 1996; 347: Verity C, Preece M. Surveillance for rare disorders by the BPSU. Arch Dis Child 2002; 87: British Paediatric Surveillance Unit. How the surveillance system works. In: Lynn RM editor. British Paediatric Surveillance Unit Annual Report London: Royal College of Paediatrics and Child Health, 2008: Office for National Statistics. Population and vital statistics by area of usual residence in the United Kingdom, In: Brown S, Redgwell G, editors. Key Population and Vital Statistics: Local and Health Authority Areas. Basingstoke: Palgrave Macmillan, 2008: Lamont PJ, Surtees R, Woodward CE, et al. Clinical and laboratory findings in referrals for mitochondrial DNA analysis. Arch Dis Child 1998; 79: Liang MH, Wong LJC. Yield of mtdna mutation analysis in 2000 patients. Am J Med Genet 1998; 77: Scagalia F, Towbin JA, Craigen WJ, et al. Clinical spectrum, morbidity and mortality in 113 pediatric patients with mitochondrial disease. Pediatrics 2004; 114: Bernier FP, Boneh A, Dennett X, et al. Diagnostic criteria for respiratory chain disorders in adults and children. Neurology 2002; 59: Wolf NI, Smeitink JA. Mitochondrial disorders: a proposal for consensus diagnostic criteria in infants and children. Neurology 2002; 59: Developmental Medicine & Child Neurology 2010, 52: