With interest we read the article by Birtel et al. about the retinal abnormalities in 21 patients with a syndromic (n=9) or non-syndromic (n=12) mitochondrial disorder (MID) [1]. Among the syndromic MIDs, 7 were diagnosed with chronic progressive external ophthalmoplegia (CPEO), 1 with mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (MELAS) syndrome, and 1 patient with neuropathy, ataxia, and retinitis pigmentosa (NARP) syndrome [1]. Three retinal phenotypes were differentiated in the 21 patients: type 1 was characterised by mild, focal pigmentary abnormalities, type 2 was characterized by multifocal white-yellowish sub-retinal deposits and pigment changes limited to the posterior pole, and type 3 was characterised by widespread granular pigment alterations [1]. Advanced type 2 and 3 retinopathy deteriorated to chorio-retinal atrophy that typically started in the peri-papillary and para-central areas with foveal sparing [1]. It was concluded that distinct subtypes of mitochondrial retinopathy can be delineated even in the absence of clinical manifestations and that recognition of these subtypes may facilitate the diagnosis of a MID [1]. The study is appealing but raises comments and concerns.

Seven of the 21 patients carried the m.3243A>G variant in MT-TL1 with heteroplasmy rates ranging from 10 to 64% [1]. The most frequent phenotype of this genotype is MELAS. However, none of these 7 patients was diagnosed with MELAS. We should be told if the absence of the typical phenotype was attributable to the low heteroplasmy rates in 6 of these patients or to the fact that heteroplasmy rates were determined in blood lymphocytes, which are usually not affected. We should know if these seven patients manifested with features other than diabetic retinopathy, retinal dystrophy, myopathy, acute, posterior, multifocal, placoid pigment epitheliopathy, or age-related macular degeneration. Common phenotypic features of m.3243A>G carriers include stroke-like episodes (SLEs), seizures, cognitive decline, confusion, headache, hypoacusis, fatigue, myopathy, nausea/vomiting, and lactic acidosis [2].

Nine patients carried single mtDNA deletions, which were classified as sporadic [1]. Though single mtDNA deletions are sporadic in the majority of the cases, they are maternally transmitted in 4% of the cases [3]. Thus, we should know in how many of the nine patients carrying single mtDNA deletions the family history was positive for a MID. Single mtDNA deletions manifest phenotypically as CPEO, Pearson syndrome, or Kearns-Sayre syndrome (KSS). However, only 7 of these patients manifested with CPEO. We should know if the two remaining patients manifested with features other than ptosis, retinopathy, or “muscular dystrophy” [1]. Particularly we should be told if these two patients with a non-syndromic phenotype presented with features of Pearson syndrome or KSS.

The most frequent phenotype of the variant m.8344A>G is myoclonic epilepsy with ragged-red fiber (MERRF) syndrome [4]. However, the patient carrying the m.8344A>G variant, manifested with MELAS [1]. This is not unusual but surprising given the high heteroplasmy rate of 85% [1]. We should be told if patient 3 presented with any of the four canonical features of MERRF (myoclonic epilepsy, ataxia, myopathy, myocloni).or other features of MERRF plus.

Missing is an in-depth discussion of macular abnormalities in m.3243A>G carriers, which include peri-macular and peri-papillary retinal pigment epithelium abnormalities with mottled dye auto-fluorescence and retinal pigment epithelium atrophy and deposits on optic coherence tomography [5].

Overall, the interesting study has several limitations which challenge the results and their interpretation. An explanation should be provided why m.3243A>G carriers and mtDNA deletion carriers manifested minimally and non-syndromically.