Aspect en IRM cérébrale des maladies mitochondriales. Algorithme décisionnel des maladies mitochondriales les plus fréquentes

Mitochondrial diseases are due to deficiency of the respiratory chain and are characterized by a broad clinical and genetic heterogeneity that makes diagnosis difficult. Some clinical presentations are highly suggestive of given gene mutations, allowing rapid genetic diagnosis. However, owing to the wide pattern of symptoms in mitochondrial disorders and the constantly growing number of disease genes, their genetic diagnosis is frequently difficult and genotype/phenotype correlations remain elusive. For this reason, brain MRI appears as a useful tool for genotype/phenotype correlations. Here, we report the most frequent neuroradiological signs in mitochondrial respiratory chain deficiency and we propose a diagnostic algorithm based on neuroimaging features, so as to direct molecular genetic tests in patients at risk of mitochondrial respiratory chain deficiency. This algorithm is based on the careful analysis of five areas on brain MRI: (1) basal ganglia (hyperintensities on T2 or calcifications); (2) cerebellum (hyperintensities on T2 or atrophy); (3) brainstem (hyperintensities on T2 or atrophy); (4) white matter (leukoencephalopathy); (5) cortex (sub-tentorial atrophy); (6) stroke-like episodes. We believe that the combination of brain MRI features is of value to support respiratory chain deficiency and direct molecular genetic tests.