MELAS (Myopathy, Encephalopathy, Lactic Acidosis, and Stroke-like episodes) is a progressive neurodegenerative disorder characterized by acute neurological episodes resembling strokes associated with hyperlactatemia and mitochondrial myopathy. The exact prevalence of the disease is unknown. Patients usually present during childhood or early adulthood with acute crises, which may be triggered by infection or physical exercise. These crises associate cephalalgia, vomiting and sometimes pseudo-stroke signs, such as confusion, hemiparesia and hemianopsia. They often occur in patients with chronic symptoms such as muscle weakness, deafness, diabetes, short stature, cardiomyopathy, developmental delay, learning difficulties, memory loss or attention disorders. The disease is caused by mitochondrial DNA mutations. At least 10 different mutations have been identified but 80% of the cases are due to the 3243A>G mutation in the leucine transfer RNA gene (tRNA Leu). This mutation is therefore often referred to as the MELAS mutation despite its association with diverse clinical presentations: its prevalence in the general population of Europe has been estimated at 1/6250. Mutation 3271T>C in the tRNA Leu gene is associated with the syndrome in a further 7.5% of patients. Diagnosis of MELAS syndrome relies on the clinical presentation and brain imaging. Magnetic resonance imaging may reveal numerous hyperintense T2 lesions in cerebral white and grey matter, while computerized tomography shows cerebral atrophy and basal ganglia calcifications. They show that the lesions are not confined to vascular territories and thus that the acute episodes are not typical strokes. Abnormal accumulation of lactate is frequent in blood and almost constant in cerebrospinal fluid. Muscle biopsy is abnormal in approximately 85% of patients. It shows abnormal mitochondrial proliferation (ragged red fibers) and muscle fibers with a cytochrome c oxidase defect. Analysis of muscle respiratory chain activities may reveal complex I deficiency or a combined deficiency of complexes I and IV. Identification of the causal mutation has to take into account the constant heteroplasmy i.e. its coexistence with a residual population of wild type mitochondrial DNA. Mutation proportions may differ considerably between tissues, but it is most often very high (above 90%) and may therefore be investigated in blood. Genetic counseling is very arduous in MELAS syndrome due to the heteroplasmy. Mitochondrial DNA mutations are transmitted according to maternal inheritance. An affected man cannot transmit the disease. The mutation will be transmitted along the maternal lineage but its proportion is essentially unpredictable. Although higher proportions of the mutation in the blood of the mother result in a higher risk of having a child with severe phenotype, there are many examples of extreme segregation of the mutation from mother to child, which prevent efficient genetic counseling at an individual level. The possibility heterogeneity in the proportion of the mutation between tissues theoretically hampers prenatal diagnosis. Very few proper clinical trials have been conducted with MELAS patients. A recent one has found dichloracetate to have negative effects in the medium-term. Spontaneous evolution of the disease with acute crises, remission and recurrence makes it difficult to evaluate the clinical improvement reported in some MELAS patients treated with supportive treatments (including coenzyme Q10 and its analogue idebenone, creatine monohydrate and arginine) or the deleterious impact of treatment such as valproic acid (an antiepileptic drug reported to provoke stroke-like episodes). Prognosis is poor. Patients may die during one stroke-like episode and, along with recurrent episodes, they often develop mental deterioration, loss of vision and hearing, as well as severe myopathy, potentially leading to loss of autonomy.