Perfusion and Electrophysiological Changes in MELAS

A 35-year-old man with mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes (MELAS) due to the m.3243A>G variant in the MT-TL1 gene (OMIM # 590050) had acute onset of flashing, colorful, well-formed lines and shapes in his right visual field. A few hours later, he developed a right retro-orbital headache and nausea. Symptoms waxed and waned throughout the night, prompting his presentation to the emergency department the next morning. Although he did not specifically complain of it, initial neurological evaluation revealed right homonymous hemianopsia. Alexia without agraphia was noted later in his admission. A brain magnetic resonance imaging (MRI) was emergently obtained for a presumed MELAS stroke-like episode (SLE). The MRI of his brain showed no new signal changes on T2/fluid-attenuated inversion recovery (FLAIR), diffusion-weighted imaging (DWI), or apparent diffusion coefficient (ADC) sequences (stable, chronic T2 signal abnormalities from prior SLEs were again noted), but there was markedly increased perfusion signal within the left parieto-occipital gray matter on arterial spin labeling (ASL) sequences (Fig 1A). Asymmetry of the posterior cerebral arteries was noted with left being larger in caliber compared to right and compared to the same vessel on angiography 3 years prior, consistent with dilation and evidence of hyperperfusion in the left posterior quadrant on ASL sequences (Fig 1B). Arginine infusion was initiated,1 he was continued on home levetiracetam, lacosamide, and clonazepam, and monitored with continuous video electroencephalogram (EEG). Positive visual symptoms resolved, but right homonymous hemianopsia persisted. The EEG showed frequent to abundant 10 to 30 seconds runs of ~2 hertz (Hz) lateralized periodic discharges (LPDs) often with superimposed fast activity (Fig 1E), consistent with ictal-interictal continuum (IIC). There was also one brief electrographic seizure over the left posterior quadrant (Fig 1F). LPDs ceased after intravenous lorazepam challenge. Standing lorazepam was initiated under EEG guidance with no evidence of recurrent seizures. Repeat MRI brain 4 days later showed new gyriform restricted diffusion and increased parieto-occipital T2/FLAIR signal corresponding to the previously observed regions of ASL signal (Fig 1C). He was discharged on a lorazepam taper in addition to his standing anti-seizure medication regimen. Three months after discharge, neurological examination showed resolution of prior deficits. Repeat MRI brain showed stable chronic lesions and interval resolution of the gyriform diffusion restriction and associated increased T2/FLAIR signal in the same region (Fig 1D). Routine EEG showed rare left occipital epileptiform discharges and no LPDs. SLEs constitute the most debilitating feature of MELAS; however, their underlying pathophysiology remains unclear. There are prior reports of increased perfusion appearing concurrently with DWI or T2 bright lesions in m.3243A>G-associated SLEs.2, 3 This is, to our knowledge, the first report of the complete evolution of MRI findings beginning prior to DWI or T2 changes. Increased perfusion occurred at symptom onset, followed by diffusion restriction that gradually evolved to chronic FLAIR changes. This is also the first report of IIC in an MELAS SLE. Increased metabolic demand has been previously demonstrated by fluorodeoxyglucose-positron emission tomography (FDG-PET) in other causes of IIC.4 This case sheds light on the pathophysiology of SLEs, nominating interconnected processes of neuronal hyperexcitability, increased metabolic demand, and hyperperfusion in the setting of mitochondrial dysfunction ultimately leading to cortical injury. The authors would like to thank the patient and Dr Alex C. Bender, Dr Alexander M. Cerjanic, Dr Galina Ghehman, and Dr Leigh Rettenmaier for the excellent care and thoughts they provided for this case. Dr Walker received salary support from K08NS117889 (NINDS). B.A., C.J.C., N.S.R., and M.A.W. contributed to the conception and design of the report. B.A. prepared the figures. B.A. and M.A.W. contributed to drafting the text. No relevant conflict of interest to report. This report did not generate unique research data.