Comprehensive Studies are needed before High-frequency Oscillations are Recommended as a Biomarker for Epilepsia Partialis Continua in MELAS

We were interested to read the article by Maeda et al. on a 13-year-old boy with mitochondrial encephalopathy, lactic acidosis and stroke-like episode (MELAS) due to the variant m.3243A>G, in which scalp EEG recordings were analyzed for highfrequency oscillations (HFOs) before (n=1) and after (n=5) the onset of the fifth stroke-like episode (SLE), which manifested as visual disturbance, headache and epilepsia partialis continua (EPC) [1] .No scalp HFOs were recorded prior to SLE, but on day 1 of SLE, 1.20 scalp HFOs per minute were detected [1] .The scalp HFO detection rate decreased with the disappearance of the EPC, but the epileptic discharges remained on the EEG after the disappearance of the EPC [1] .No scalp HFOs appeared on the EEG, although visual disturbances continued to be observed [1] .It was concluded that HFOs on the scalp could be a biomarker for epileptic seizures in MELAS patients [1] .The study is remarkable, but some inconsistencies should be clarified.The first point is that information about the genetic background of the disease is scarce [1] .We should know in which tissue the m.3243A>G variant was detected, at what heteroplasmy rate, and whether heteroplasmy rates differed between clinically affected and unaffected tissues.There is also no mention of whether the mtDNA variant was inherited or occurred de novo.The second issue is that no family history information was provided [1] .We should know whether another family member was also a carrier of the m.3243A>G variant, and whether another family member was also clinically affected, either with MELAS or another phenotype associated with the m.3243A>G variant.The third point is that no cerebral MRI images were shown.In order to assess whether the morphologic abnormalities were indeed a stroke-like lesion (SLL), which is the correlate of SLE in imaging, it would have been necessary to see brain imaging.It would also have been important to provide an MRI follow-up.SLLs can have a very different outcome.They can regress completely without leaving a residual lesion, but they can also transform into a white matter lesion, cysts, laminar cortical necrosis, focal atrophy or the toenail sign [2] .The fourth point is that the reason why the patient received a single bolus of phenobarbital (PB) on admission was not mentioned [1] .PB is known to be potentially mitochondrial toxic [3] and could worsen the clinical picture including seizure intensity.Was PB the cause of HFOs?The fifth point is that the statement "in the present case, scalp HFOs were frequently observed in the EEG at the time of EPC onset, which disappeared after the EPC disappeared" is not supported by the reported data.According to Figure 1, EPCs occurred only once on the first hospital day but not thereafter, while HFOs also occurred on the eighth hospital day without the occurrence of an EPC [1] .This discrepancy should be clarified.The sixth point is that the conclusion that scalp HFOs have the potential to be a useful biomarker for the occurrence of EPC in patients with MELAS is not supported by the data presented [1] .Such a conclusion cannot be drawn from a single case report.Before HFOs can truly be recommended as a diagnostic biomarker for EPC in the setting of SLE, appropriately designed studies in a large MELAS cohort need to be conducted.In addition, HFOs must fulfill the criteria for a diagnostic biomarker [4] .In summary, this interesting study has limitations that affect the results and their interpretation.Addressing these limitations could strengthen the conclusions and support the message of the study.The detection of EPCs in the context of SLE does not require a biomarker.EPCs can be detected clinically and by a routine scalp EEG.Therapeutic management of EPC in the setting of SLE may require the use of not just a single anti-seizure medication (ASM), but multiple ASMs and even non-drug interventions.