By Neal S. Parikh, MD, MS
Assistant Professor of Neurology and Neuroscience, Weill Cornell Medical College
Stroke symptoms in mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS) are difficult to diagnose correctly, which leads to missed opportunities to provide MELAS-specific treatment. Delay in diagnosis also complicates efforts to investigate acute treatments for MELAS. Khasminsky et al proposed clinicoradiologic criteria based on a single-center validation study. Although there are methodological limitations, the concepts highlighted by the authors are valuable.
Khasminsky V, Auriel E, Luckman J, et al. Clinicoradiologic criteria for the diagnosis of stroke-like episodes in MELAS. Neurol Genet 2023;9:e200082.
Mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS) is a mitochondrial disorder that presents with nervous system and muscular dysfunction in children and young adults. Stroke-like episodes are, as the name suggests, episodes of neurological dysfunction that resemble stroke.
These episodes are thought to be caused by impaired mitochondrial respiratory chain function, resulting in energy supply-demand mismatch in highly metabolically active tissues, such as the cerebral cortex. MELAS is a rare disorder; therefore, the diagnosis of MELAS frequently is delayed. Developing criteria to quickly distinguish patients with MELAS from patients with bland ischemic stroke is important for two reasons. First, rapid identification of MELAS allows clinicians to begin immediate treatment with intravenous (IV) L-arginine, which may improve outcomes. Second, reliable identification of MELAS would facilitate research into acute treatments beyond L-arginine.
To address this need, the authors derived and validated clinicoradiologic criteria for stroke-like episodes. In a single-center study, the authors identified 11 patients with MELAS who had a total of 17 stroke-like episodes in addition to 21 patients with ischemic stroke involving cortical structures typically seen in MELAS. No clinical matching (for example, by age and sex) was done. The authors compared clinical and radiologic characteristics of these patients and developed diagnostic criteria, which they then tested in the same patient population in the hands of blinded adjudicators.
Clinically, there were several striking features that differentiated patients with MELAS from patients with bland ischemic stroke. Patients with MELAS were substantially younger, more often women, had lower body mass index, and had a high rate of hearing loss (91% vs. 5%). Additionally, despite their younger age and lower body mass index, patients with MELAS had a high prevalence of diabetes (82%). Presenting symptoms in MELAS were more subacute, frequently included seizure and headache, and were accompanied by higher serum lactate measures.
Radiologically, MELAS had two distinct spatiotemporal patterns. In the first, imaging abnormalities evolved in an anterior pattern, from the anterior hippocampus and temporal operculum to the frontal cortex. In the second, imaging abnormalities evolved in a posterior pattern, from the hippocampal tail or cuneus or precuneus to the parieto-occipital cortex. A striking horseshoe pattern of cortical involvement was seen. Patients with MELAS had imaging evidence of prior episodes and cerebellar volume loss out of proportion to age.
None of the MELAS patients had a large vessel occlusion on computed tomography angiogram (CTA), but rather they had hyperemia and venous engorgement. Based on these findings, the authors proposed that a probable diagnosis of MELAS can be made for patients younger than 60 years of age with acute or subacute neurological symptoms and/or nausea/vomiting who meet at least one clinical criterion (body mass index < 20, history of sensorineural hearing loss, or headache and/or seizure at presentation), at least one acute radiologic criterion (cortical lesion with temporal periopercular involvement, occipital lesion with medial occipital involvement, or otherwise not defined by a vascular territory), at least one chronic imaging criterion (cerebellar atrophy, chronic cortical lesions in MELAS pattern, or hyperemia/venous engorgement on CTA), and lack of a large vessel occlusion. Blinded reviewers found these criteria to have a sensitivity of 88% and specificity of 95%.
COMMENTARY
Derivation and validation of diagnostic criteria for a rare disorder in a small, single-center cohort is fraught with methodological limitations. Sensitivity and specificity in this study overstate the true test characteristics because, although MELAS is rare at the population level, the prevalence of MELAS in the validation cohort was dramatically inflated and the blinded assessors intentionally were considering MELAS as a leading possibility. With these caveats stated, this paper is highly instructive because it underscores the unique clinical and radiologic features of MELAS. Knowledge of these concepts will help neurologists and radiologists more frequently consider the diagnosis of MELAS.
