Multimodal Cerebral Imaging is required to Detect Stroke-like Lesions in NDUFS8 Mutation Carriers

We read with interest Gowda et al's article on a 17-months-old female with mitochondrial disorder (MID) due to the variant c.304C>T in NDUFS8 [1] .The NDUFS8 variant manifested phenotypically with developmental delay, generalised hypotonia, quadrispasticity, lactic acidosis, leucoencephalopathy, stroke-like episodes (SLEs), and hypertrophic cardiomyopathy complicated by heart failure [1] .The patient benefited from a mitochondrial cocktail and heart failure treatment [1] .The study is impressive, but several points require discussion.The first point is that only a limited number of MRI modalities (T1, T2, diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC) have been performed [1] .To characterise a stroke-like lesion (SLL), which represents the morphological equivalent of a SLE, it is imperative to perform additional MRI modalities that typically characterise SLL [2] .These include perfusion-weighted imaging (PWI), oxygen-extraction fraction (OEF), magnetic resonance angiography (MRA), and magnetic resonance spectroscopy (MRS) [2] .SLLs are typically hyperintense on PWI but hypointense on OEF MRI [2] .MRA is usually normal but MRS typically shows a lactate peak and a reduced N-acetyl aspartate (NAA) peak.Perfusion single photon emission computed tomography (SPECT) can show hyperperfusion, but fluordeoxy glucose (FDG) positron emission tomography (PET) can show hypometabolism [3] .Another typical feature of SLLs is their dynamic expansion up to a nadir and their regression after reaching this nadir.SLLs can terminate as normal brain, or as structural lesion in the form of white matter lesions, cysts, focal atrophy, laminar cortical necrosis, or as toenail sign [2] .The second point is that cerebrospinal fluid (CSF) studies are not mentioned even though the patient had cerebral disease [1] .It is of particular interest to know whether CSF lactate was elevated or not.The third point is that it was not mentioned whether the patient suffered from myopathy or not [1] .Since MIDs often manifest with myopathy [4] , we should know whether needle electromyography (EMG) recorded on weak muscles was abnormal (myogenic, neurogenic, non-specific).Normal creatine-kinase, as in the index patient, does not rule out myopathy.A fourth point that electroencephalography (EEG) recordings have not been reported [1] .Since the patient presented with dullness, had SLE, and SLEs often manifest with seizures, we should know whether the electroencephalography (EEG) recordings showed epileptiform discharges or not.In particular, non-convulsive status epilepticus (NCSE) should be excluded.A fifth point is that cerebral vasculitis is unlikely as a differential diagnosis of MID because SLLs are not limited to one vascular territory and therefore do not correspond to a vascular lesion.A sixth point is that the statement that MELAS does not manifest with respiratory chain complex deficiencies is not supported.Usually, biochemical studies in MELAS show multiple respiratory chain defects, as previously reported [5] .However, MELAS may manifest with a single respiratory chain defect if it is due to a nuclearly encoded mitochondrial gene [6] .In summary, the excellent study has limitations that should be addressed before drawing final conclusions.Clarifying the weaknesses would strengthen the conclusions and improve the study.Documentation of SLL in non-syndromic MID due to a pathogenic variant in NDUFS8 requires multimodal MRI, MRA, MRS, and FDG-PET.