Diagnosis of mitochondrial diseases: Clinical and histological study of sixty patients with ragged red fibers
Background: Mitochondrial diseases are caused by mutations in mitochondrial or nuclear genes, or both and most patients do not present with easily recognizable disorders. The characteristic morphologic change in muscle biopsy, ragged-red fibers (RRFs) provides an important clue to the diagnosis. Materials and Methods: Demographic data, presenting symptoms, neurological features, and investigative findings in 60 patients with ragged-red fibers (RRFs) on muscle biopsy, seen between January 1990 and December 2002, were analyzed. The authors applied the modified respiratory chain (RC) diagnostic criteria retrospectively to determine the number of cases fulfilling the diagnostic criteria of mitochondrial disease. Results: The most common clinical syndrome associated with RRFs on muscle biopsy was progressive external ophthalmoplegia (PEO) with or without other signs, in 38 (63%) patients. Twenty-six patients (43%) had only external ophthalmoplegia, 5 (8%) patients presented with encephalomyopathy. Specific syndromes were the presenting feature in 8 (13%), Kearns-Sayre syndrome (KSS) in 4 and myoclonus epilepsy with ragged-red fibers (MERRF) in 4. Myopathy was the presenting feature in 5 (8%) and 4 presented with infantile myopathy. Of the 60 patients, 18 (30%) had proximal muscle weakness. Two patients with KSS and one patient with myopathy had complete heart block necessitating pace making. When the modified RC diagnostic criteria were applied, only 26 (43%) patients had one other major criterion in addition to RRFs for the diagnosis of mitochondrial diseases. The remaining 34 (57%) patients with RRFs on muscle biopsy had only some clinical features suggestive of RC disorder but did not fulfill the clinical criteria (of the modified diagnostic criteria) for the diagnosis of mitochondrial diseases. Conclusion: In patients with clinical features suggestive of RC disorder, demonstration of RRFs on muscle biopsy helps in confirming the diagnosis of mitochondrial disease in only a subgroup of patients.
Mitochondrial diseases with ragged-red muscle fibers (RRF) as well as some without RRF, are caused by mutations in mitochondrial or nuclear genes, or both, which are normally involved in the formation and maintenance of a functionally intact oxidative phosphorylation system in the mitochondrial inner membrane. These disorders present, with bewildering array of clinical presentations and are usually dominated by the involvement of the nervous system and the skeletal muscle (hence the term "mitochondrial encephalomyopathies"). In most cases, myopathy is associated with a cluster of central nervous system (CNS) symptoms, suggesting one of the well-defined mitochondrial encephalomyopathies. However, the encephalomyopathies may be less specific, consisting of one of the CNS symptoms typical of mitochondrial dysfunction.,,,,,,,,,,,,, Most of the recent proposed criteria for the diagnosis of mitochondrial diseases include biochemical and molecular genetic testing., Often these tests are expensive and may not be available in most medical centers in the developing world. However mitochondrial diseases associated with RRFs can be diagnosed on muscle biopsy. RRFs are characterized by large proliferations of subsarcolemmal mitochondria and replacement of some of the contractile elements with intermyofibrillar accumulations of mitochondria. These appear red as detected by Gomori-Trichrome staining and can also be seen using a stain for succinate dehydrogenase. They have a moth-eaten appearance. They may be isolated, infrequent or numerous and may involve type 1 fibers often. Only a few small case series have been reported from India and all the case series were based on muscle biopsy.,, This paper presents the clinical data of 60 patients with RRFs on muscle biopsy and discusses the diagnosis of mitochondrial diseases.
Case files of patients with RRF on muscle biopsy seen between January 1990 and December 2002 were retrieved from the neuropathology files, Department of Pathology, of our center. The clinical data noted included age, gender, family history, presenting symptoms and neurological features. Investigative findings included serum creatine phosphokinase (CK), electromyoneurographic (EMNG) findings, and CT and/or MRI features. One patient had muscle analysis for mitochondrial respiratory chain studies. None of the patients had molecular genetic testing. The diagnostic criteria applied for the diagnosis of mitochondrial diseases were those proposed by Bernie et al [Table - 1]a & b. For categorizing patients into various clinical syndromes, the classification proposed by Nordin and Johns, 2001 was adopted.
Muscle biopsy was done by open biopsy from left vastus lateralis in all patients. The biopsy was divided into three parts: one part was snap frozen in isopentane precooled in liquid nitrogen for cryo-sections; one part was fixed in 3% cacodylate buffered glutaraldehyde for election microscopy; and one part was fixed in 10% buffered formalin for routine paraffin embedding sections. Cryo-sections were stained with haematoxyclin and eosin, modified Gomori trichrome stain (MGT), Masson's trichrome stain, oil red O, periodic acid Schiff with and without diastase. Enzyme histochemistry with adenosine triphosphatase (ATP) at pH 9.4, 4.6 and 4.3, succinic dehydrogenase (SDH) and NADH-Tetrazolium reductase (NADH-TR) were also done on cryo-sections. Electron microscopic studies could not be done due to its unavailability at our center.
The mean age at the time of diagnosis was 29.3 years (range 1 - 71 years) and there were 35 males and 25 females. The mean age at the onset of the disease was 19.7 years (range: birth to 59.7 yrs) and the age of the onset in each diagnostic criteria is given in [Table - 1]. The disease had a progressive course in all the patients. The mean delay between the onset of the disease and the diagnosis was 6.7 years (range 2 months to 30 yrs). Based on the clinical features, patients could be grouped into one of the clinical syndromes proposed.
[Table - 2]. The most common clinical syndrome associated with RRFs on muscle biopsy was progressive external ophthalmoplegia (PEO) with or without other signs in 38 (63%) patients. Twenty-six patients (43%) had only external ophthalmoplegia and 6 had oculopharyngeal weakness. Six patients with PEO had associated myoclonus. Five (8%) patients presented with cerebellar ataxia, 3 had associated myopathy and 2 had diffuse muscle wasting. Specific syndromes were the presenting feature in 8 (13%), Kearns-Sayre syndrome (KSS) in 4 and myoclonus epilepsy with ragged-red fibers (MERRF) in 4. Myopathy was the presenting feature in 5 (8%) and 4 presented with infantile myopathy. Of the 60 patients, 18 (30%) had proximal muscle weakness and 4 had clinical and/or electrophysiological evidence of neuropathy. Retinitis pigmentosa was noted in 8 patients including 4 patients with KSS.
Associated systemic involvement was noted in 5 patients, diabetes mellitus in 2 and cardiac involvement in 3. Electrocardiogram was done in 35 patients, complete heart block was the specific abnormality in 3 patients. Of the 4 patients with KSS, 2 had complete heart block. One of the patients with myopathy had cardiomyopathy and complete heart block. He presented with recurrent syncope. All three required permanent pacing. 2D Echo was done in 5 patients with myopathy, 4 patients with KSS, and 4 children with infantile myopathy.
Positive family history was noted in 3 probands. In one family, father and one of the sons were affected and in the other 2 families, one of the brothers and sisters were affected.
Of the 35 patients in whom serum CK values were available, moderately elevated levels were noted in 5 (300 - 700 IU/l) and all of them had clinical evidence of significant proximal muscle weakness. EMNG data was available in 31 patients, myopathic EMG features were seen in 12 patients, one had features of denervation, and 4 had evidence of neuropathy.
CT was the neuroimaging modality in all the patients with encephalopathic features. Cerebellar folial atrophy was noted in 4 (all of them had features of cerebellar ataxia), bilateral basal ganglionic calcification in 1, and cortical atrophy was noted in 2.
The diagnosis of mitochondrial disease was based on the presence of characteristic ragged-red fibers (RRF) in > 2% of muscle fibers in the muscle biopsy.,,, All muscle biopsies showed subsarcoplasmic red deposits and ragged sarcoplasm on MGT stain though the number of fibers showing RRF varied significantly [Figure - 1] and also an increase in the intensity of staining for oxidative enzymes on SDH and NADH [Figure - 2], [Figure - 3]. Type grouping suggestive of reinnervation was seen in 5 (8.3%) biopsies. ATP staining showed type-1 dominance and type-2 atrophy in 5 each. One patient showed type-1 dominance and type-2 atrophy. Significant neutral lipid accumulation was seen in 3 and none showed glycogen excess on PAS staining. Central nuclei were seen in 2 biopsies [Table - 3].
Ragged-Red Fibers - Diagnosis of Mitochondrial Disease
Of the 60 patients with RRFs on muscle biopsy, 25 (42%) patients fulfilled the definite criteria for the diagnosis of mitochondrial disease [Table - 3]. This included 4 patients with Kearns-Sayre syndrome (KSS), 4 with myoclonus epilepsy with ragged-red fibers (MERRF), 5 patients with encephalomyopathies, and 12 patients with multisystemic symptoms that was essentially pathognomonic for respiratory chain (RC) disorder and a progressive clinical course and at least three of the organ system involvement (dysmorphic features 5, diabetes mellitus 2, and cardiac involvement 1, and retinopathy 4). One patient had clinical features suggestive RC disorder on histological and enzymological criteria. While all the remaining 34 patients had symptoms that were essentially pathognomonic for RC disorder, with one or two organ system presentation, and a progressive clinical course [Table - 4].
Mitochondrial diseases can present at any age and include a combination of multisystemic symptoms. The adult criteria require muscle or neurologic involvement. The presence of PEO is very suggestive, as is the presence of CNS dysfunction such as ataxia, seizures, or sensorineural deafness. In most cases myopathy is present along with a cluster of CNS symptoms suggesting the well-defined syndromes [Table - 5]. The retina, in particular the retinal pigment epithelium, is highly vulnerable to be involved by mtDNA defect, and the retinopathy is phenotypically variable and frequently subclinical, depending to some extent on the type or site of mt DNA defect. Peripheral neuropathy is also a frequent association. The clinical features of cardiac involvement in mitochondrial diseases vary in the different subgroups of these disorders. Mitochondrial myopathy with cardiomyopathy has been re ported with a variety of point mutations in DNA. Cardiac conduction blocks are more frequent in patients with KSS. Intraventricular conduction defects show an unusually rapid progression to potentially fatal complete atrioventricular block and are an indication for prophylactic cardiac pacing.,
Clinical investigations only help, to some extent in categorizing the mitochondrial disorders into one of the specific encephalopathic syndromes. Serum CK level may be mildly elevated in mitochondrial myopathies but are often normal. Electroneuromyographic features are not pathognomonic and mitochondrial diseases should be in the electrophysiolgic differential diagnosis of a patient with myopathic EMG findings and reduced sural sensory response amplitudes., Neuroradiologic features suggestive of mitochondrial diseases include symmetric basal ganglia calcifications, brain infarcts, or cerebral and cerebellar atrophy.
RRFs on muscle biopsy are observed in a wide variety of mitochondrial diseases including mtDNA depletion, mtDNA deletions, and mitochondrial transfer RNA mutations. RRFs are not seen in disorders due to defects of substrate transport, substrate utilization, and the Kreb's cycle., RRFs are not pathognomonic of mitochondrial diseases, but can also be seen in other conditions like dystrophies, dermatomyositis and in older individuals and also in zidovudine-associated myopathy in HIV-infected patients., However, in these diseases, the intensity and the proportion of muscle biopsies showing RRFs will be sparse. Presence of more than 2% RRFs in skeletal muscle biopsy is taken as one of the criteria for the diagnosis of mitochondrial disease.,,,, Another useful histochemical probe is the stain for Cytochrome c oxidase (COX) activity. Many studies use Cytochrome c oxidase staining to identify negatively staining fibers. Cytochrome c oxidase identifies complex IV in the respiratory chain. Ragged-red fibers are often COX-negative. Although COX staining may be diffusely abnormal in primary COX deficiencies, most mitochondrial myopathies show segmental loss of COX staining in individual muscle fibers. In our study we have not used this stain. Unfortunately, most patients with mitochondrial diseases do not show any of the above characteristic muscle changes. Enzymology and molecular genetic studies are often required to establish a definite diagnosis of mitochondrial diseases [Table - 1]a & b.
Mammalian mtDNA is thought to be strictly maternally inherited. In familial cases of disease caused by gene defects in mtDNA, inheritance is nonmendelian and passes from mother to offspring. Families with a mendelian pattern of inheritance of mitochondrial gene defects have been described. However, these cases appear to be attributable to defects in the nuclear genes that encode proteins responsible for the fidelity of mtDNA replication or the maintenance of mitochondrial genomes rather than attributable to direct inheritance of defective forms of mtDNA. Nuclear DNA mechanisms of genetic transmission include autosomal dominant, autosomal recessive genetic imprinting and X-linked transmission. Recently Schwartz and Vissing (2002) described a patient with exercise intolerance, lactic acidosis after minimal exertion, and RRFs. The mitochondrial haplotype in the patient's muscle matched that of his father. The 2 bp deletion causing disease was unique to the patient, apparently representing a neq mutation arising in the paternal germ line or during embryonic development. The family with mendelian pattern of inheritance, autosomal dominant, in this study may be due to the defects in the nuclear genes.
Several diagnostic schemes were proposed to improve the diagnostic sensitivity, as the clinical features are suggestive rather than pathognomonic of mitochondrial diseases. First consensus diagnostic criteria were proposed in 1996 (Adult Criteria, AC). However non-neurologic manifestations occur in over 30% of pediatric patients. A modified version to allow for diagnosis of pediatric cases has recently been suggested (Modified Adult Criteria, MAC). Both systems include clinical, genetic, histopathological, and metabolic criteria and also a scoring criterion for the results of biochemical studies. The latter criteria appear to improve the sensitivity of the adults. Recently Wolf and Smeitink developed a scoring system, Mitochondrial Disease Criteria (MDC). MDC allows more precise definitions of clinical and metabolic items and the independent scoring of muscle biochemical investigations
Definite diagnosis of mitochondrial disease is defined as the identification of either two major criteria or one major plus two minor criteria. A probable diagnosis is defined as either one major plus one minor criterion or at least three minor criteria. A possible diagnosis is defined as either a single major criterion or two minor criteria, one of which may be clinical. In our study, only 43% of patients with RRFs fulfilled the definite criteria for the diagnosis of mitochondrial disease. The remaining 34 (57%) patients had histological criteria and two of the three clinical criteria for the diagnosis of mitochondrial diseases and thus they were probable or possible cases of mitochondrial disease. However, it is important to keep in mind that the application of these criteria is based on a detailed understanding of complex clinical and laboratory information. Most patients do not present with easily recognizable disorders. Symptoms can be difficult to assimilate into a unifying diagnosis as in the case of pediatric patients. One of the essential clinical criteria for the diagnosis of mitochondrial disease is "symptoms must include at least three of the organ system presentations". The relevance of such criteria can only be validated by genetic and enzymatic studies in patients with symptoms of one or two organ system presentation suggestive of RC disorder. To determine the sensitivity, both MAC and MDC criteria need to be tested prospectively in large cohorts of patients.
Our study also suggests that except for some specific mitochondrial encephalomyopathic syndromes, the clinical features are rarely pathognomonic for the diagnosis of mitochondrial disorders. It will be prudent to evaluate such patients with unexplained combination of multisystem symptoms and progressive clinical course for possible mitochondrial disease. In the absence of genetic and enzymatic studies, muscle biopsy for RRFs is likely to establish the diagnosis of mitochondrial disease in a proportion of such patients.