PEDIATRICS Vol. 119 No. 4 April 2007, pp. 722-733 (doi:10.1542/peds.2006-1866)
ARTICLE |
Long-term Outcome and Clinical Spectrum of 73 Pediatric Patients With Mitochondrial Diseases
a Divisions of Medical Genetics
b Pediatrics, Department of Pediatrics
c Departments of Pathology
d Medical Imaging, Centre Hospitalier Universitaire Sainte-Justine, Université de Montreal, Montreal, Quebec, Canada
e Department of Human Genetics, Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada
f Metabolism Research Program, Research Institute, Departments of Pediatrics and Biochemistry, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVES. We sought to determine the clinical spectrum, survival, and long-term functional outcome of a cohort of pediatric patients with mitochondrial diseases and to identify prognostic factors.
METHODS. Medical charts were reviewed for 73 children diagnosed between 1985 and 2005. The functional status of living patients was assessed prospectively by using the standardized Functional Independence Measure scales.
RESULTS. Patients fell into 7 phenotypic categories: neonatal-onset lactic acidosis (10%), Leigh syndrome (18%), nonspecific encephalopathy (32%), mitochondrial (encephalo)myopathy (19%), intermittent neurologic (5%), visceral (11%), and Leber hereditary optic neuropathy (5%). Age at first symptoms ranged from prenatal to 16 years (median: 7 months). Neurologic symptoms were the most common (90%). Visceral involvement was observed in 29% of the patients. A biochemical or molecular diagnosis was identified for 81% of the patients as follows: deficiency of complex IV (27%), of pyruvate dehydrogenase or complex I (25% each), of multiple complexes (13%), and of pyruvate carboxylase (5%) or complexes II+III (5%). A mitochondrial DNA mutation was found in 20% of patients. At present, 46% of patients have died (median age: 13 months), 80% of whom were <3 years of age. Multivariate analysis showed that age at first symptoms was a major independent predictor of mortality: patients with first symptoms before 6 months had a highly increased risk of mortality. Cardiac or visceral involvement and neurologic crises were not independent prognostic factors. Living patients showed a wide range of independence levels that correlated positively with age at first symptoms. Among patients aged >5 years (n = 32), 62% had Functional Independence Measure quotients of >0.75.
CONCLUSIONS. Mitochondrial diseases in children span a wide range of symptoms and severities. Age at first symptoms is the strongest predictor mortality. Despite a high mortality rate in the cohort, 62% of patients aged >5 years have only mild impairment or normal functional outcome.
Key Words: mitochondrial diseases • long-term outcome • functional status
Abbreviations: MD—mitochondrial disease • CSF—cerebrospinal fluid • RRF—ragged red fibers • MELAS—mitochondrial encephalo(myopathy) with lactic acidosis and stroke-like episodes • LHON—Leber hereditary optic neuropathy • PDH—pyruvate dehydrogenase • mtDNA—mitochondrial DNA • NARP— neurogenic weakness–ataxia and retinitis pigmentosa • FIM—Functional Independence Measure • WeeFIM—Functional Independence Measure for Children • CI—confidence interval
Mitochondrial diseases (MDs) represent a vast group of inherited disorders of energy metabolism that encompass a wide range of symptoms and presentations, severity, and outcome.1 Together, they form one of the most prevalent groups of inherited metabolic diseases.2–4 The minimum birth prevalence of MDs for onset at any age has been estimated at 1 of 7634.5 However, given the diversity of clinical presentations, including nonspecific courses, and because of the difficulty in establishing a diagnosis, this must be considered a minimal estimate.5,6 Because oxidative phosphorylation is a fundamental pathway of cellular metabolism, any organ can be involved. Hence, MDs may present in any of numerous subspecialities. Clinical pediatric reports described the spectrum of signs and symptoms associated with MDs, but the long-term clinical course of this heterogeneous group remains imprecise.7–10 In particular, little is known about the long-term functional outcome of pediatric patients with MDs. The aims of this study were to evaluate the clinical course, mortality, and morbidity in a cohort of pediatric patients with MDs. Here, we describe the phenotypic spectrum of these patients based on long-term follow-up and report, to our knowledge, the first standardized evaluation of the functional independence level of pediatric patients with MDs. We tested the ability to predict mortality and morbidity of several clinical and biochemical characteristics.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Case Ascertainment
We reviewed hospital charts of all patients evaluated for congenital lactic acidosis or presumed MD by the Genetics Division of CHU Sainte-Justine (Montreal, Canada) between January 1985 and December 2005. Thirty-five patients with Saguenay-Lac St-Jean cytochrome c oxidase deficiency (Online Mendelian Inheritance in Man No. 220111) were excluded from the study. Because of a French-Canadian founder effect, this condition forms a substantial fraction of our patients.11 Excluding these patients rendered our cohort more representative of the spectrum of MDs encountered elsewhere in the world. Among the remaining patients (n = 99), 26 were excluded because of insufficient clinical data or lack of sufficient criteria for the diagnosis of MD.
Seventy-three patients from 66 families had long-term follow-up data and evidence of MD. Sixty-seven had a score of 3 with the modified Walker criteria scoring system for MD.12 Six patients had a score of 2 but were included because they had strong evidence of MD based on long-term clinical, biological, and neuroradiological follow-up (mean follow-up time of 66 months in this group) and the absence of other potential etiologies. All 6 patients had declined muscle and/or liver biopsy, impeding definite biochemical diagnosis. Three were sibs of patients with enzymatically proven respiratory chain deficiency, which in addition to suggestive clinical signs was considered as a major criterion for the diagnosis of MD.4
Clinical Data and Phenotypic Classification
For these 73 patients, we retrospectively reviewed all available clinical, neuroradiological, pathologic, biochemical, and molecular data. The clinical variables that were specifically noted were: antenatal and neonatal history, developmental course, abnormalities of muscle tone, epilepsy, movement disorders, myopathic or neuropathic weakness, neurologic crisis, gastric tube feeding requirement, growth parameters (including cranial circumference), visceral involvement (cardiomyopathy, liver disease, renal dysfunction, etc), ophthalmologic involvement, and hearing loss. Neurologic crises were defined as acute or subacute neurologic deterioration with loss of previously acquired skills and appearance of new neurologic symptoms and included episodes of Leigh disease and stroke-like episodes. Available biochemical data (blood gases, lactate, and pyruvate concentrations in blood and cerebrospinal fluid (CSF) and plasma amino acids) were averaged for each patient.
Patients were classified into 7 clinical categories (Table 1): (1) neonatal-onset lactic acidosis, characterized by fulminant acidosis; (2) Leigh syndrome, including typical Leigh syndrome and Leigh-like syndrome according to Rhamann et al13; (3) visceral, in which marked involvement of 
1 organ dominates the clinical picture; (4) mitochondrial myopathy, including patients with myopathy or encephalomyopathy with ragged red fibers (RRF) and patients with classical mitochondrial syndromes: Kearns-Sayre syndrome (n = 2), mitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes [MELAS] (n = 4), and mitochondrial neurogastrointestinal encephalomyopathy (n = 1); (5) Leber hereditary optic neuropathy (LHON, n = 4) (because their clinical course was strikingly different from that of other patients with MD, patients with LHON were not included to study frequency of early manifestations of MD); (6) 4 patients had only intermittent neurologic symptoms (recurrent ataxia or episodic weakness) that recurred over several years, and they were classified as an "intermittent neurologic group"; and (7) the remaining patients had nonspecific encephalopathy. In the text, numbers and percentages of patients, when used alone, indicate that data were available for all 73 patients.
|
Etiologic Investigations
Muscle biopsy was performed in 45 patients (62%). The tissue was snap-frozen and submitted to a standard histopathology examination including hematoxylin and eosin, adenosine triphosphatase, reduced nicotinamide adenine dinucleotide (NADH)-tetrazolium reductase, oil red O, periodic acid Schiff, and modified Gomori trichrome stains. After 1999, succinate dehydrogenase and cytochrome oxidase stains were also performed. Electron microscopy was performed in 40 muscle biopsies (89%). Liver histology (n = 19) and/or ultrastructure (n = 14) were studied on biopsies or autopsy specimens from 22 patients (30%). Mitochondrial enzyme activities were measured spectrophotometrically as described.14–16 Pyruvate dehydrogenase (PDH) complex (native and dichloroacetate-activated), pyruvate carboxylase, complex II+III, and complex IV activities were determined in cultured fibroblasts from 51 patients (70%). Complete respiratory chain studies were performed in muscle (24 patients, 33%) and liver (21 patients, 29%). Blue native gel analysis was also performed as described17 on cultured fibroblasts from 20 patients.
Molecular analysis of mitochondrial DNA (mtDNA) included testing for large deletions-duplications and for point mutations related to classical mitochondrial cytopathies: MELAS, A3243G and T3271C; MERRF, A8344G and T8356C; neuropathy, ataxia and retinitis pigmentosa (NARP), T8993G and T8993C; LHON, G11778A, G3460A, T14484C, and G14459A and was performed in patients with symptoms suggesting these conditions and in patients without definite biochemical diagnosis and for whom tissues were available. Complete mtDNA sequencing was performed as described18 in 6 patients in whom muscle biopsy had revealed RRF suggesting possible heteroplasmy, and in whom screening mtDNA analyses had failed to find a mutation.
Informed consent was obtained from parents or legal guardians for all diagnostic procedures and molecular studies.
Outcome: Mortality and Morbidity
Mortality was ascertained by medical charts review. For 72 of 73 patients (97%), complete follow-up was available (defined as until death or December 2005). Mortality rate was analyzed according to several potential prognostic factors: age at first symptoms, phenotypic category, presence of visceral involvement, mean plasma lactate concentration, occurrence of metabolic and neurologic crisis, and presence of an mtDNA mutation. For Kaplan-Meier analysis, the cohort was divided into 2 subgroups: patients presenting their first symptoms before the age of 6 months and those presenting thereafter.
Morbidity was evaluated prospectively for living patients using the Functional Independence Measure (FIM)/Functional Independence Measure for Children (WeeFIM) scoring systems, which provide a standardized way to obtain information about functional status for individuals at least 8 years of age and children aged 6 months to 8 years, respectively.19–21 These scales were administered in 2005 by a single physician as part of clinical follow-up and monitoring of the patients at routine visits at the genetics clinic. In both scales, a score from 1 to 7 (1 = total dependence; 7 = complete autonomy) is assigned for 18 items in 3 main domains: (1) self-care (including eating, grooming, bathing, dressing upper and lower body, toileting, and bladder and bowel management), (2) mobility (including transfer from wheelchair, transfer to toilet, to tub or shower, walking/wheelchair/crawling distance, and moving up and down stairs), and (3) cognition (evaluating comprehension, expression, social interactions, problem solving, and memory). To standardize for patient age, WeeFIM and FIM raw scores were divided by the age-specific mean (±4 months) and expressed as quotients. Quotients <50% are considered to represent severe impairment, between 50% and 75% are moderate, and >75%, mild or no impairment. For additional analysis, WeeFIM and FIM quotients were combined as described.22
Statistical Analysis
A t test for independent groups was used to compare means, and a 
2 test was used to compare proportions between groups. We tested the association between possible predictors and mortality with Cox's proportional hazard ratio model. Variables significant at P < .01 on univariate analysis were included in the multivariate model. We used Kaplan-Meier curves to describe the survival in some subgroups. The mortality rates were compared between subgroups by using the log-rank test. FIM and WeeFIM quotients were compared between the 3 domains (self-care versus mobility, self-care versus cognition, and mobility versus cognition) with a Wilcoxon rank test for matched pairs. We used analysis of variance and linear regression to test the association between potential predictors (age at first symptoms, mean lactate concentration during follow-up, and occurrence of metabolic and neurologic crises) and the global functional quotient. Statistical analyses were performed with SPSS software (SPSS Inc, Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Of the 73 patients selected, 42 boys (58%) and 31 girls (42%) presented from the fetal period to 16 years old. Over the period examined, the number of new cases constantly increased: 1985–1989, 7 new cases; 1990–1994, 12; 1995–1999, 21; 2000–2005, 32. Sixty-eight patients (93%) were of white ancestry. Other ethnic origins were: Haitian black, Indian, Turkish, Pakistani, and Moroccan (1 patient each). Global mortality rate in the cohort was 46%, and the median age at death was 13 months (range: 1 day to 20 years). Among living patients, all but 1 had complete follow-up; mean follow-up time for living patients was 12.4 ± 7.5 years. A family history consistent with MD was found in 17 patients (28%). Consanguinity (7 families) or recurrence within the family (7 sibling pairs) was present for 14 of 66 families (21%). Two of these sibling pairs had PDH deficiency because of mutations in the X-linked E1
gene. Matrilineal transmission was documented in 7 patients (10%) who had mtDNA mutations.
Clinical Presentation
The first manifestations of the disease were recorded at a median age of 7 months (range: 0–16 years) and encompassed a wide range of symptoms including: nonspecific psychomotor delay (38%), metabolic acidosis (14%), failure to thrive (10%), acute or subacute regression (9%), visceral involvement (liver dysfunction, 4%; hypertrophic cardiomyopathy, 3%), and various combinations of focal neurologic symptoms (22%), including seizures, ataxia, extrapyramidal signs, muscle weakness or pain, ptosis, and headache. Excluding patients with LHON, median age at first symptoms was 5 months (range: 0–14 years), and a MD was first suspected at a median age of 10 months (range: 0–203). The median delay between first symptoms and diagnosis was 3 months (range: 0–121). The number of patients in each clinical category is shown in Table 1.
In the neonatal subgroup, patients presented immediately at birth with fulminant lactic acidosis and neurologic distress. All died in the first months (median age at death: 4 days). Interestingly, prenatal evaluation of these patients showed hydrocephalus in a girl with PDH deficiency, and intraventricular hemorrhage, asymmetric ventricular dilatation, and periventricular leucomalacia in a male patient with pyruvate carboxylase deficiency.23 Of note, we found a high rate of pregnancy complications for the whole cohort including gestational diabetes (8 [11%] of 73); hypertension/preeclampsia (5 [7%] of 73); intrauterine growth retardation (4 [5%] of 73; hemolytic anemia, elevated liver enzymes, and low platelets (HELLP) syndrome (1 of 73); and steatohepatitis (1 of 73). Birth weight was below the third percentile in 13 of 66 patients (20%; P < .0001).
Over follow-up, 66 patients (90%) presented clinical signs of cerebral involvement. Developmental delay was common (48 [79%] of 61) in all subgroups except LHON, but it was less frequent in the myopathic category (7 [50%] of 14) than in other subgroups, with the exclusion of LHON (P < .02). Hypotonia in the first year was found in 44 (69%) of 64 patients. Thirty-one (60%) of 51 patients aged >15 months learned to walk at a mean age of 20 ± 11 months. At follow-up, 7 (23%) had lost the ability to walk. Microcephaly was present in 32 patients (44%) and was progressive in 20. Three patients had macrocephaly, 2 with pyruvate carboxylase deficiency and 1 with complex IV deficiency. Seizures were observed in 25 patients (34%). They were the dominant clinical feature in 4 patients. In 5 patients, the seizure type was myoclonic. Nine patients (12%) exhibited extrapyramidal movement disorders ranging from dystonia to choreoathetosis. Ataxia was present in 14 (32%) of 44 patients aged >2 years. In this group, 2 patients with PDH deficiency experienced recurrent episodes of ataxia over the course of several years, separated by symptom-free intervals. Whereas hypertonia and spasticity were found in 20 (29%) of 69 patients, another 10 (14%) had hypo/areflexia and peripheral weakness, suggesting possible neuropathic or myopathic involvement. Of these 10 patients, 6 had electrophysiological or neuropathological evidence of peripheral neuropathy. In the intermittent neurologic group, 2 patients with PDH deficiency experienced recurrent episodes of isolated peripheral weakness, initially mimicking Guillain-Barré syndrome.24
Acute neurologic events were common: 19 patients (26%) experienced neurologic crisis corresponding to stroke-like episodes (6 cases) or Leigh syndrome (13 cases). In addition, 3 other patients had recurrent episodes of severe muscle cramp with or without myoglobinuria, 2 had intermittent ataxia, and 2 others, intermittent peripheral weakness (see above). In stroke-like episodes, combinations of aphasia, dysarthria, hemiparesia, facial diplegia, hemianopsy, and cortical blindness were observed. The first stroke was recorded at a median age of 13 years (range: 1.3–14.8). Neuroradiology revealed focal cortical lesions in all cases with stroke-like episode. By contrast, Leigh crises occurred at a median age of 0.9 years (range: 2 weeks to 6.4 years). In Leigh syndrome, disturbances of respiratory pattern (including ataxic breathing, intermittent tachypnea with respiratory alcalosis alternating with bradypnea, and hypercapnia requiring endotracheal intubation) were observed in 10 patients (77%), whereas such respiratory findings were much less common in the rest of the cohort (12%). Other signs of brainstem dysfunction in Leigh syndrome included swallowing disturbances, hypertension, and nuclear and internuclear ophthalmoplegia.
Twenty-one patients (29%) had visceral involvement: heart (13 patients, 18%), liver (12 patients, 16%), kidney (8 patients, 11%), gastrointestinal tract (6 patients, 8%), and bone marrow (5 patients, 7%). The distribution of organ involvement according clinical category and diagnosis is shown in Table 2. In most of these patients (15 [71%] of 21), >1 organ was involved. All patients with cardiac signs had hypertrophic nonobstructive cardiomyopathy. In 8 patients (11%, Table 1), visceral disease was the predominant clinical manifestation. Paradoxically, cardiac symptoms were infrequent in this group (2 of 8).
|
Failure to thrive was a frequent observation (38 patients, 52%) and a presenting symptom in 10% of the patients. Forty-two patients (58%) required gastric tube feeding for neurologic and/or nutritional reasons. However, in 8 patients, oral feeding could later be reintroduced. Three patients had major gastrointestinal dysmotility requiring continuous enteral feeding; despite having jejunostomies, 2 required prolonged periods of parenteral nutrition. Nine patients developed hypoglycemia, which was recurrent after short fasting in 4 of them. Two patients had diabetes mellitus, and another 2 had impaired glucose tolerance.
Excluding the 4 patients with LHON, all of whom had isolated optic atrophy, 29 (42%) of 69 patients had ophthalmologic involvement. The spectrum of ophthalmological manifestations in these patients included optic atrophy (38%), pigmentary retinopathy (21%), ophthalmoplegia (45%), ptosis (31%), nystagmus (28%), and cortical blindness (10%). Neurosensory deafness was present in 26% of the patients (18 of 69).
Metabolic Status
Long-term metabolic follow-up revealed that 72% of the patients had chronic lactic acidemia, defined as mean plasma lactate level >2.2 mmol/L. Conversely, 61% of the patients had 1 plasma lactate concentration in the reference range, showing the high variability of this biochemical parameter. The median number of measurements per patient was 12. Abnormal plasma amino acids showed raised alanine and/or proline, consistent with lactic acidosis, in 47 (73%) of 64 of the patients tested. Abnormal urinary organic acids, that is, various combinations of elevated lactic and pyruvic acids, ketone bodies, and Krebs cycle intermediates, were found in 42 (70%) of 60 patients. The CSF lactate level was raised in 37 (88%) of 42 patients. Four patients with normal plasma lactate levels had increased CSF lactate levels. Conversely, 2 patients with intermittent elevations of plasma lactate levels were found to have normal CSF lactate levels. Both had PDH deficiency and PDH E1 gene mutation and presented with intermittent ataxia. There was no association between lactate, alanine, and proline concentrations and the type of biochemical and molecular diagnoses. Mean lactate to pyruvate molar ratios were significantly different between patients with respiratory chain deficiencies or mtDNA mutation and those with PDH deficiency (31.7 ± 13.5 vs 19.6 ± 4.5, respectively; P < .001).
Twenty-two patients (30%) experienced acute crisis of metabolic acidosis defined as plasma bicarbonate concentration 
10 mmol/L and/or requirement for continuous intravenous bicarbonate infusion. Sixteen of these patients (73%) subsequently died, 10 (45%) during one of these episodes. Mean of average individual plasma lactate concentration during crises was 11.6 ± 7.0 mmol/L. Metabolic crises recurred in 5 patients. Median age at the first acute acidotic episode was 9.5 months (range: birth to 56 months).
Neuroradiology
Neuroimaging studies were available in 59 patients (81%), including magnetic resonance imaging for fifty patients. A wide range of neuroradiological abnormalities were observed: the most common were basal ganglia hyperintensities (27 [46%] of 59) and cerebral atrophy (28 [47%] of 59). Brainstem lesions were identified in 34% of magnetic resonance imaging studies. Stroke-like cortical infarcts were less frequent (10%). Some neuroradiological features less specific to MD were also commonly observed: white matter abnormalities (21 [40%] of 59), corpus callosum hyperintensities (5 [10%] of 59), and cerebellar atrophy (9 [15%] of 59). Of note, brainstem and basal ganglia lesions were not restricted to patients with Leigh disease but were also observed in other clinical presentations (19 [41%] of 46), including nonspecific encephalopathy and visceral subgroups.
Pathology
Mitochondrial proliferation or ultrastructural morphologic abnormalities were identified in 32 (80%) of 40 of the muscle samples examined with electron microscopy. However, using NADH-TR, modified Gomori or succinate dehydrogenase-cytochrome C oxidase stains, only 21 (47%) of 45 muscle biopsies showed evidence of mitochondrial proliferation. Twelve patients (16%) exhibited RRFs with Gomori trichrome staining. In 19 liver samples, histology was normal in 7 and revealed widespread steatosis in 8, active hepatitis and fibrosis in 2, and micronodular cirrhosis in 2 others. Widespread mitochondrial proliferation and abnormalities were found by electron microscopy in 10 (71%) of 14 liver and 2 of 2 kidney biopsies.
Biochemical and Molecular Studies
A biochemical or molecular diagnosis was identified in 81% of subjects. The distribution of specific diagnoses according to clinical category is shown in Table 1. Evaluation for respiratory chain defects or PDH deficiency was not performed in patients with a known pathogenic mtDNA mutation. An enzyme defect was identified in 44 patients. Complex IV deficiency was the most frequent enzyme defect (27%), followed by PDH and complex I deficiencies (25% each). Relative frequencies of enzyme defects are shown in Fig 1. All PDH deficiencies were found in cultured fibroblasts, whereas diagnosis of respiratory chain deficiencies by spectrophotometry required investigations of several tissues: fibroblasts (6 subjects), muscle (10 subjects), liver (14 subjects), and heart (4 subjects). Blue native gel studies revealed a specific enzyme defect in fibroblasts of 11 subjects, including 6 complex IV and 5 complex I deficiencies. Diagnostic yields of spectrophotometric assays and blue native gel studies for each tissue are shown in Fig 2.
|
|
An mtDNA mutation was found in 15 patients (20%): "MELAS" A3243G in tRNALeu(UUR) (4 patients with classical MELAS phenotype), large heteroplasmic mtDNA deletions (2 patients with Kearns-Sayre syndrome), "NARP" T8993G/C in MTATP6 (3 patients with Leigh syndrome), and "LHON" T14484C in MTND6 (4 patients with LHON). MtDNA whole genome sequencing detected a new homoplasmic mutation in tRNAser(UCN) (7466_7471delC) and another previously undescribed heteroplasmic mutation in the MTCOIII gene (G9984A, G260X) in 2 patients with mitochondrial myopathy. Among the 12 patients with RRF, a mtDNA mutation was found in 7 (58%). A mutation in the E1
PDH subunit gene was found in 5 patients with PDH deficiency. Except for the intermittent neurologic subgroup (4 patients with PDH deficiency), there were no correlations between the biochemical defect and type of presentation. As expected, the finding of mtDNA mutation was associated with later-onset presentation (P = .001) and was more common in the (encephalo)myopathic and LHON subgroups (8 of 14 and 4 of 4, respectively, vs 3 of 55 for other groups; P = .001).
Outcome: Mortality and Morbidity
The overall mortality rate in the cohort was 46%, based on an observation period of 1.5 to 32 years (median follow-up time of 12 years for living patients). Higher mortality rate was observed in patients in the neonatal, visceral, and Leigh subgroups (100%, 85%, and 69%, respectively). Eighty percent of deaths occurred before the age of 3 years. No association was found between the biochemical defect and mortality rate. In univariate analysis, the occurrence of the first signs before the age of 6 months, average plasma lactate concentration, presence of liver involvement, and occurrence of metabolic crisis were all positive predictors of the mortality rate (Table 3). In contrast, the presence of a mtDNA mutation was associated with a lower mortality rate (after exclusion of LHON patients, hazard ratio: 0.09; 95% confidence interval [CI]: 0.01–0.69). Cardiac hypertrophy was frequently asymptomatic and not associated with increased mortality. By using the multivariate Cox regression model (Table 4), after adjustment for liver involvement, occurrence of metabolic crises and mean lactate concentration, the age at first presenting symptoms remained strongly predictive of mortality: patients with first symptoms before the age of 6 months were 10 times more likely to die than patients who had their first signs after the age of 6 months (Table 4). The Kaplan-Meier curves in Fig 3 illustrate the difference in survival between patients presenting their first signs before 6 months compared with others (P < .0001). Conversely, after adjustment for age at first signs, liver involvement and the presence of metabolic crises, the average plasma lactate concentration was only weakly associated with mortality rate.
|
|
|
Morbidity was assessed in 38 patients with long-term survival. Only 1 patient was lost to follow-up at 12 years old. Median age at last evaluation was 12.1 year (range: 1.5–32 years). Prospective assessment of functional status showed a wide spectrum in outcome, ranging from near complete independence to severe encephalopathy requiring total assistance. Median FIM and WeeFIM scores expressed as quotients of age-related norms are shown in Table 5. Among the 38 survivors with long-term follow-up, self-care and cognition were more severely affected than mobility (Table 5; P = .01 and P < .001, respectively). There was no association between specific biochemical diagnosis and the FIM/WeeFIM scores, but because of the small numbers of patients in each category, this does not eliminate the possibility that such an association may exist. After adjustment for occurrence of neurologic and metabolic crises, analysis of variance identified age at first symptoms as the only predictive factor of the global functional outcome (P = .009). In a linear regression model, for each year of delay in appearance of initial symptoms, global quotient increased by 5 U. Chronic metabolic status (mean plasma lactate levels), metabolic and neurologic crises were not identified as independent predictors of the functional outcome of patients. However, the relatively small sample size limits the power of these analyses. Global FIM and WeeFIM quotients against age of the patients and according to clinical category are presented in Fig 4. Interestingly, among patients >5 years old (n = 32), 62% had global FIM/WeeFIM quotients >0.75.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Nearly any pediatric specialist may be confronted with the initial presentation of MD, to which the mnemonic "any age, any symptom, any organ" has been applied.7 Patients of this study, identified in a community-based tertiary care pediatric center, illustrate the need for a high level of suspicion for MD in patients with unexplained organ dysfunction. The wide base of referring symptoms of our patient sample suggests a relatively unbiased selection, which may reflect the spectrum of MDs in childhood more accurately than in series based on laboratory recruitment. Of note, the increasing numbers of patients diagnosed over time suggests that in the past, diagnosis was reserved for patients with severe or classic presentations. By extension, many MDs probably remain undiagnosed, particularly in their early stages.
We identified a high frequency of pregnancy complications. This was previously suggested by only 1 study.25 The prevalence of low birth weight in our cohort (20%) is similar to that reported by von Kleist et al25 Association between fetal mitochondrial disorders and gestational diabetes, preeclampsia, HELLP syndrome, and acute fatty liver of pregnancy has not been previously reported. Although these associations could be fortuitous, the repercussions of energy defects in the fetoplacental unit are essentially unknown. Interestingly, maternal HELLP syndrome is a well-recognized complication of fetal fatty acid oxidation defects.26 Although more data are necessary, we suggest careful monitoring of pregnancies at risk for an affected fetus.
The overall frequency of neurologic manifestations is similar to previous studies.3,9,27 However, we found a high frequency (35%) of acute neurologic presentations. This has not previously been emphasized in pediatric patients with MD. In addition to Leigh crises and stroke-like episodes, the spectrum of episodic neurologic events included intermittent ataxia, episodic peripheral weakness, and recurrent muscle cramps. Inherited disorders of energy metabolism should be considered in any child with unexplained acute and/or recurrent neurologic symptoms.
Although visceral involvement was frequent in our cohort, we did not observe the previously reported frequency and the high mortality of cardiomyopathies in patients with MD.10,28 In our series, cardiomyopathy was often a mild and asymptomatic echocardiographic finding. By contrast, hepatic involvement was frequent and associated with high mortality. Such discrepancies between studies illustrate that the incidence of clinical manifestations in MD remains incompletely characterized and that additional multicentric studies may be necessary to understand the complete spectrum of the MD in childhood. Interestingly, we showed that patients with visceral manifestations often have multiple organ involvement, suggesting that close follow-up and periodic multivisceral evaluation is required in this group.
Measurement of lactic acid in plasma is a useful but nonspecific test. Although 72% of the patients have chronic lactic acidemia, up to 60% had at least 1 normal lactate concentration, illustrating the variability of this biological parameter and emphasizing that a normal lactate concentration does not exclude MD.29 Conversely, imperfect blood sampling conditions involving tissue hypoxia or physical exertion such as struggling or crying in infants can falsely increase lactate levels or the lactate/pyruvate molar ratio.30 Repeated blood lactate measurements can clarify the true metabolic status of the patients. CSF lactate measurement is less variable and is essential in some patients having "cerebral" lactic acidosis,31 as illustrated by the 4 patients with normal lactic acid in blood who had clearly increased CSF lactate levels. In contrast, 2 sibs with recurrent episodes of ataxia, enzymatically proven PDH deficiency, and PDH E1 gene mutation had normal lactate levels in both blood and CSF, showing that even a normal CSF lactate level cannot exclude MD.
A third of our patients developed acute acidotic crisis, usually in association with otherwise benign infectious diseases. This life-threatening complication, little emphasized in the literature, is a major issue in patient management, requiring ICU admission for intravenous bicarbonate administration and other supportive measures, despite which mortality remains nearly 50%. Most acidotic crises occurred during infancy. We did not find any correlation between clinical or biochemical parameters and the occurrence of metabolic crises. In our practice, we have the impression that optimal nutritional status and complete immunization may reduce the frequency of these metabolic crises.
Diagnostic yield of biochemical analyses according to the tissue tested illustrates the complexity of diagnostic process in MD. On one hand, some respiratory chain defects have a tissue specific expression; on the other hand, technical problems in tissue handling and storage can give rise to artifactually low enzyme activities.32 In addition, tissues may be affected by nonmetabolic disorders like disuse muscle atrophy or liver failure of various etiologies. This can also result in nonspecific low respiratory chain activity.33 The overall yield of liver testing in our cohort was higher than those of muscle and fibroblast. Liver biopsy is invasive and associated with a small but not insignificant risk of complications, particularly in fragile patients such as those suffering from MD. Our study shows that fibroblast analyses reveal a substantial proportion of respiratory chain defects, especially when spectrophotometric assays and blue native gel analysis are combined. Moreover, PDH deficiency is reliably identified in fibroblasts. The utility of blue native gel studies in fibroblasts was already reported,34 and we now routinely perform this in patients suspected of MD.
In our series, PDH deficiency is a frequent diagnosis (15%). Similar or lower PDH deficiency frequencies were reported previously in series of patients in whom PDH activity was measured.8,9,35 However, PDH deficiency has not been reported consistently in cohorts with MD.3,4,7,10 PDH deficiency was excluded from some published series because it is not a respiratory chain defect and is not routinely assayed in all diagnostic laboratories. In our experience, the distinction between respiratory chain and PDH deficiencies is often not possible clinically at presentation. Both disorders belong to the broad group of congenital lactic acidosis, justifying the inclusion of PDH deficiency in clinical review of MD in childhood. Importantly, PDH deficiency is one of the few conditions in this group for which specific treatments exist.36–39 Patients with PDH deficiency in our cohort had no visceral involvement, which is in accordance with the most recent literature.40 Although the L/P ratio was significantly lower in patients with PDH deficiency compared with others, overlap between these groups shows that the L/P ratio can not be used alone to discriminate individual patients. We suggest that PDH deficiency should be considered in any patient with congenital lactic acidosis, particularly those with predominant neurologic symptoms. Measurement of PDH activity in fibroblasts should be routinely performed in such cases.
On the basis of these considerations, we propose an algorithm of investigations for patients with congenital lactic acidosis or presumed MD (Fig 5). Because most of these conditions are currently untreatable, we suggest starting with the least-invasive techniques (fibroblast culture, blood DNA testing for specific etiologies as indicated) before performing muscle and/or liver biopsy. Such tests have a substantial yield. In some cases, the strategy of sequential testing eliminates the need for invasive procedures. For example, 3 of 13 patients with Leigh syndrome in our series were found to have mtDNA T8993G/C mutation in blood leukocytes. Moreover, identification of a respiratory chain defect on fibroblasts is sometimes the first step toward a molecular diagnosis, using complementation studies in research protocols.42,43
|
The age at first symptoms was a major prognostic factor for mortality. Infants presenting their first symptoms before 6 months of age have a tenfold increased risk of mortality during the observation period compared with those presenting later. Although previously suggested,9 this has never been evaluated previously by multivariate analysis. Higher plasma lactate concentration was also positively associated with mortality.
Because the clinical course of MD is chronic, progressive, and unpredictable, long-term follow-up of large cohorts is necessary for a better understanding of the spectrum of severity of these conditions. Except for 1 recent study of MD presenting in the neonatal period,44 there is no report in the medical literature on long-term outcome of pediatric patients with MD. Because of the wide range of age and severity in our cohort, and because we were interested in quality of life and disease burden for patients and families, we evaluated outcome in terms of functional status rather than with classical cognitive testing. In our series, age at first symptoms was positively correlated with the global FIM/WeeFIM quotient. Figure 4 suggests that there may be 2 groups of patients, 1 with a stable course and preserved function, and 1 with deterioration and/or low functional level and encephalopathy. Larger numbers of patients and longer follow-up will be required to draw a definitive conclusion, but encouragingly, a substantial proportion of the patients aged >5 years had mild or no impairment.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We report the first systematic prospective evaluation of the functional outcome of pediatric patients with MD. We identified age at first symptoms as a major prognostic factor for mortality and morbidity. Although a high mortality rate was observed in the first years, the functional status of survivors, based on long-term follow-up, was more favorable than previously thought: a substantial proportion of patients have an unimpaired functional independence level.
|
ACKNOWLEDGMENTS |
|---|
We acknowledge the excellent work of Yolande Lefevre, Danielle Regimbalde, and Manon Bouchard in the support and treatment of patients during the course of this study.
|
FOOTNOTES |
|---|
Accepted Nov 28, 2006.
Address correspondence to Grant A. Mitchell, MD, Medical Genetics Division, CHU Sainte-Justine, 3175 Côte Sainte-Catherine, Montreal, Quebec, Canada H3T 1C5. E-mail: grant.mitchell{at}recherche-ste-justine.qc.ca
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
1. DiMauro S, Schon EA. Mitochondrial respiratory-chain diseases. N Engl J Med. 2003;348 :2656 –2658
2. Applegarth DA, Toone JR, Lowry RB. Incidence of inborn errors of metabolism in British Columbia, 1969–1996. Pediatrics. 2000;105(1) . Available at: www.pediatrics.org/cgi/content/full/105/1/e10
3. Zeviani M, Bertagnolio B, Uziel G. Neurological presentations of mitochondrial diseases. J Inherit Metab Dis. 1996;19 :504 –520[CrossRef][Web of Science][Medline]
4. Darin N, Anders O, Ali-Reza M, Holme E, Tulinius M. The incidence of mitochondrial encephalomyopathies in childhood: clinical features and morphological, biochemical and DNA abnormalities. Ann Neurol. 2001;49 :377 –383[CrossRef][Web of Science][Medline]
5. Skladal D, Halliday J, Thornburn DR. Minimum birth prevalence of mitochondrial respiratory chain disorders in children.
Brain. 2003;126
:1905
–1912
6. Thorburn DR. Practical problems in detecting abnormal mitochondrial function and genomes.
Hum Reprod. 2000;15
:57
–67
7. Munnich A, Rötig A, Chretien D, Saudubray JM, Cormier V, Rustin P. Clinical presentation of mitochondrial disorders in childhood. J Inherit Metab Dis. 1996;19 :521 –527[CrossRef][Web of Science][Medline]
8. Nissenkorn A, Zeharia A, Lev D, et al. Multiple presentation of mitochondrial disorders.
Arch Dis Child. 1999;81
:209
–215
9. Skladal D, Sudmeier C, Konstantopoulou V, et al. The clinical spectrum of mitochondrial disease in 75 pediatric patients.
Clin Pediatr (Phila). 2003;42
:703
–710
10. Scaglia F, Towbin JA, Craigen WJ, et al. Clinical spectrum, morbidity and mortality in 113 pediatric patients with mitochondrial disease.
Pediatrics. 2004;114
:925
–931
11. Morin C, Mitchell G, Larochelle J, et al. Clinical, metabolic, and genetic aspects of cytochrome c oxidase deficiency in Saguenay-Lac-Saint-Jean. Am J Hum Genet. 1993;53 :488 –496[Web of Science][Medline]
12. Bernier FP, Boneh A, Dennett X, Chow CW, Cleary MA, Thorburn DR. Diagnostic criteria for respiratory chain disorders in adults and children.
Neurology. 2002;59
:1406
–1411
13. Rhaman S, Blok RB, Dahl HHM, Danks DM. Leigh syndrome: clinical features and biochemical and DNA abnormalities. Ann Neurol. 1996;39 :343 –351[CrossRef][Web of Science][Medline]
14. McKenzie S, Phillips SM, Carter SL, Lowther S, Gibala MJ, Tarnopolsky MA. Endurance exercise training attenuates leucine oxidation and BCOAD activation during exercise in humans.
Am J Physiol Endocrinol Metab. 2000;278
:E580
–E587
15. Carter SL, Rennie CD, Hamilton SJ. Changes in skeletal muscle in males and females following endurance training. Can J Physiol Pharmacol. 2001;79 :386 –389[CrossRef][Web of Science][Medline]
16. Sheu KF, Hu CW, Utter MF. Pyruvate dehydrogenase complex activity in normal and deficient fibroblasts. J Clin Invest. 1981;67 :1463 –1471[Web of Science][Medline]
17. Nijtmans LGJ, Henderson NS, Holt IJ. Blue native electrophoresis to study mitochondrial and other protein complexes. Methods. 2002;26 :327 –334[CrossRef][Web of Science][Medline]
18. Wong LJC, Liang MH, Kwon H, Park J, Baid R, Tan D. Comprehensive scanning of the whole mitochondrial genome for mutations.
Clin Chem. 2002;48
:1901
–1912
19. Guide for the Uniform Data Set for Medical Rehabilitation. Version 5.1. Buffalo, NY: State University of New York at Buffalo; 1997
20. Msall ME, Di Gaudio KM, Roberts BT. The Functional Independence Measure for Children (WeeFIM): conceptual basis and pilot use in children with disabilities.
Clin Pediatr (Phila). 1994;33
:421
–430
21. Ottenbacher KJ, Taylor ET, Msall ME, et al. The stability and equivalence reliability of the Functional Independence Measure for children (WeeFIM). Dev Med Child Neurol. 1996;38 :906 –916
22. Swaine BR, Pless IB, Friedman DS, Montes JL. Effectiveness of a head injury program for children. Am J Phys Med Rehabil. 2000;79 :412 –420[CrossRef][Web of Science][Medline]
23. Brun N, Robitaille Y, Grignon A, Robinson BH, Mitchell GA, Lambert M. Pyruvate carboxylase deficiency: prenatal onset of ischemia-like brain lesions in two sibs with the acute neonatal form. Am J Med Genet. 1999;84 :94 –101[CrossRef][Web of Science][Medline]
24. Debray FG, Lambert M, Vanasse M, et al. Intermittent peripheral weakness as the presenting feature of pyruvate dehydrogenase deficiency. Eur J Pediatr. 2006;165 :462 –466[CrossRef][Web of Science][Medline]
25. von Kleist-Retzow JC, Cormier-Daire V, Viot G, et al. Antenatal manifestations of mitochondrial respiratory chain deficiency. J Pediatr. 2003;143 :208 –212[CrossRef][Web of Science][Medline]
26. Ibdah JA, Bennett MJ, Rinaldo P, et al. A fetal fatty-acid oxidation disorder as a cause of liver disease in pregnant women.
N Engl J Med. 1999;340
:1723
–1731
27. Nissenkorn A, Zeharia A, Lev D, et al. Neurologic presentations of mitochondrial disorders.
J Child Neurol. 2000;15
:44
–48
28. Holmgren D, Wahlander H, Eriksson BO, Oldfors A, Holme E, Tulinius M. Cardiomyopathy in children with mitochondrial disease.
Eur Heart J. 2003;24
:280
–288
29. Touati G, Rigal O, Lombès A, Frachon P, Giraud M, Ogier de Baulny H. In vivo functional investigations of lactic acid in patients with respiratory chain disorders.
Arch Dis Child. 1997;76
:16
–21
30. Chariot P, Ratiney R, Ammi-Said M, Hérigault R, Adnot S, Gherardi R. Optimal handling of blood samples for routine measurement of lactate and pyruvate. Arch Pathol Lab Med. 1994;118 :695 –697[Web of Science][Medline]
31. Brown GK, Haan EA, Kirby DM, et al. "Cerebral" lactic acidosis: defects in pyruvate metabolism with profound brain damage and minimal systemic acidosis. Eur J Pediatr. 1988;147 :10 –14[CrossRef][Web of Science][Medline]
32. Chretien D, Rustin P. Pitfalls and tips in measuring and interpreting mitochondrial enzyme activities. J Inherit Metab Dis. 2003;26 :189 –198[Medline]
33. Marin-Garcia J, Ananthakrishnan R, Goldenthal MJ, Filiano JJ, Sarnat HB. Skeletal muscle mitochondrial defects in nonspecific neurologic disorders. Pediatr Neurol. 1999;21 :538 –542[CrossRef][Web of Science][Medline]
34. Van Coster R, Smet J, George E, et al. Blue native polyacrylamide gel electrophoresis: a powerful tool in diagnosis of oxidative phosphorylation defects. Pediatr Res. 2001;50 :658 –665[Web of Science][Medline]
35. Miyabayashi S, Ito T, Narisawa K, Iinuma K, Tada K. Biochemical study in 28 children with lactic acidosis, in relation to Leigh's encephalomyopathy. Eur J Pediatr. 1985;143 :278 –283[CrossRef][Web of Science][Medline]
36. Wexler ID, Hemalatha SG, McConnell J, et al. Outcome of pyruvate dehydrogenase deficiency treated with ketogenic diets: studies in patients with identical mutations.
Neurology. 1997;49
:1655
–1661
37. Naito E, Ito M, Yokota I, Saijo T, Matsuda J, Kuroda Y. Thiamine-responsive lactic acidemia: role of the pyruvate dehydrogenase complex. Eur J Pediatr. 1998;157 :648 –652[CrossRef][Web of Science][Medline]
38. Fouque F, Brivet M, Boutron A, et al. Differential effect of DCA treatment on the pyruvate dehydrogenase complex in patients with severe PDHC deficiency. Pediatr Res. 2003;53 :793 –799[CrossRef][Web of Science][Medline]
39. Berendzen K, Theriaque DW, Shuster J, Stacpoole PW. Therapeutic potential of dichloroacetate for pyruvate dehydrogenase complex deficiency. Mitochondrion. 2006;6 :126 –135[CrossRef][Web of Science][Medline]
40. Robinson BH. Lactic acidemia. In: Scriver CR, Beaudet AL, Sly W, Valle S. eds. The Metabolic and Molecular Bases of Inherited Disease. 8th ed. New York, NY: McGraw-Hill; 2001:2275–2295
41. Mootha VM, Lepage P, Miller K, et al. Identification of a gene causing human cytochrome c oxidase deficiency by integrative genomics.
Proc Natl Acad Sci U S A. 2003;100
:605
–610
42. Zuh Z, Yao J, Johns T, et al. SURF1, encoding a factor involved in the biogenesis of cytochrome c oxidase, is mutated in Leigh syndrome. Nat Genet. 1998;20 :337 –343[CrossRef][Web of Science][Medline]
43. Antonicka H, Mattman A, Carlson CG, et al. Mutations in COX15 produce a defect in the mitochondrial heme biosynthetic pathway, causing early-onset fatal hypertrophic cardiomyopathy. Am J Hum Genet. 2003;72 :101 –114[CrossRef][Web of Science][Medline]
44. Garcia-Cazorla A, De Lonlay P, Nassogne MC, Rustin P, Touati G, Saudubray JM. Long-term follow-up of neonatal mitochondrial cytopathies: a study of 57 patients.
Pediatrics. 2005;116
:1170
–1177
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  F. D. Testai and P. B. Gorelick Inherited Metabolic Disorders and Stroke Part 1: Fabry Disease and Mitochondrial Myopathy, Encephalopathy, Lactic Acidosis, and Strokelike Episodes Arch Neurol, January 1, 2010; 67(1): 19 - 24. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Horvath, J. P. Kemp, H. A. L. Tuppen, G. Hudson, A. Oldfors, S. K. N. Marie, A.-R. Moslemi, S. Servidei, E. Holme, S. Shanske, et al. Molecular basis of infantile reversible cytochrome c oxidase deficiency myopathy Brain, November 1, 2009; 132(11): 3165 - 3174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Staley, K. B. Sims, P. E. Grant, and E. T. Hedley-Whyte Case 28-2008 -- An 8-Day-Old Infant with Congenital Deafness, Lethargy, and Hypothermia N. Engl. J. Med., September 11, 2008; 359(11): 1156 - 1167. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
