With interest we read the article by Sen et al. about a 27yo female with genetically confirmed mitochondrial encephalopathy, lactic acidosis, and stroke-like episode (MELAS) syndrome in whom infection with SARS-CoV-2 was assumed to have triggered a stroke-like episode (SLE) clinically manifesting as non-bilious, non-bloody emesis, inability to tolerate feeds and medications, severe abdominal pain, mental deterioration, and a seizure [1]. The patient’s SLE was managed with L-arginine and unfractionated heparin in a therapeutic dosage [1]. The report is appealing but raises the following comments and concerns.

We do not agree with the notion that the SLE was triggered by SARS-CoV-2 [1]. Due to the gastro-intestinal compromise the patient was not taking her regular antiepileptic drugs (AEDs) since one week prior to admission [1]. Since it is well-known that seizures can trigger SLEs [2], seizures more likely than the infection with SARS-CoV-2 triggered the SLE. Arguments for seizures as the trigger of the SLE are that she was not talking levetirazetam (LEV), oxcarbazepine (OXC), and lamotrigine (LTG) since one week, that she had a seizure one day prior to admission, that electroencephalography recorded sharp waves over the area of the SLE, and that SARS-CoV-2 was not documented in the cerebro-spinal fluid (CSF) [1]. Missing in this respect are serum levels of OXC, LEV, and LTG. Sub-therapeutic levels of the AEDs would further strengthen the seizure theory. 

A shortcoming of the study is that only T2/FLAIR and apparent diffusion coefficient (ADC) MRI modalities were provided. Stroke-like lesions (SLLs), the morphological equivalent of a SLE on MRI, are characterised by hyperintensity not only on T2/FLAIR but also on diffusion weighted imaging (DWI) and perfusion weighted imaging (PWI). Furthermore, a SLL may show up as hypo-intensity on oxygen-extraction fraction (OEF) MRI and as hypo-metabolism on FDG-PET [3]. Additionally, MR-spectroscopy (MRS) may show a lactate peak within the SLL [4]. Since SLLs in the chronic stage can have a similar appearance as the lesion shown in figure 1, it cannot be excluded that the presented lesion does not represent an acute SLL but rather an old regressing lesion. A further argument against an acute SLL in the index case is that she did not present with typical features of a SLE. Isolated vomiting is a rare manifestation of a SLE and mental decline could be simply interpreted as prolonged post-ictal confusion. 

Since COVID-19 frequently manifests with gastro-intestinal compromise, it cannot be excluded that the gastro-intestinal abnormalities were rather due to the infection with SARS-CoV-2 than due to the SLE. Missing in this respect are stool cultures, gastroscopy and colonoscopy, particularly with regard to the dark-coloured stool [1]. 

We should be told if quadruparesis, ataxic gait, intention tremor, generalised wasting, were present already prior to the SARS-CoV-2 infection. Description of her pre-morbid condition is crucial to assess which of the abnormalities could be attributed either to the viral infection and which to MELAS. Delineation of COVID-19 manifestations from MELAS manifestations is crucial as the treatment of both is different. We should be told why AEDs were not increased and the rationale for therapeutic anticoagulation for the SLE. 

Missing in the report is the genetic cause of MELAS. We should be told if MELAS was due to an mtDNA or nDNA variant. In case it was due to an mtDNA variant we should know the heteroplasmy rates in affected tissues. 

Overall, the interesting case has several limitations which challenge the conclusions. Acuteness of the presented SLL must be documented by multimodal MRI and MRS. To proof that SARS-CoV-2 triggered the SLE confirmation of the virus within the CSF should be provided. Exact delineation of MELAS and COVID-19 manifestations is crucial in the index patient. Whether an infection with SARS-CoV-2 truly deteriorates MELAS remains to be proven.