Forecasting Stroke-like Episodes in Mitochondrial Disease

In our recent WCMR Science Seminar Series, we heard from Dr Alasdair Blain about his research that involves forecasting stroke-like episodes and outcomes in mitochondrial disease. Read on to find out more.

Stroke remains the second leading cause of death world-wide and the principal cause of serious long-term disability, with the prevalence of self-reported stroke in the general population estimated at 2.5%. Up to 30% of patients with suspected stroke have stroke ‘mimics’, representing a significant proportion of all acute hospital admissions. A discrete group of individuals with mitochondrial disease experience a stroke ‘mimic’, termed stroke-like episode, as part of mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (MELAS) syndrome. Stroke-like episodes among people with mitochondrial disease are often a devastating clinical event and a characteristic feature of MELAS syndrome.

The aim of this research was to devise a new risk prediction model for stroke-like episodes in patients with genetically defined mitochondrial disease. This involved the retrospective collection of clinical, demographic and neuropathological data on all cases of stroke-like episodes across a 20 year period (1998-2018), with data for patients without MELAS syndrome retrieved from the UK Mitochondrial Disease Patient Cohort.

We identified 111 patients with genetically-determined mitochondrial disease who developed stroke-like episodes. The most common genetic cause of stroke-like episodes was the m.3243A>G variant in MT-TL1 (n=66), followed by recessive pathogenic POLG variants (n= 22), and 11 other rarer pathogenic mitochondrial DNA (mtDNA) variants (n=23). The mean age of first stroke-like episode was 31.8, with the median interval between first and second stroke-like episodes 1.33 years. 43% of patients developed recurrent stroke-like episodes within 12 months.

Survival analysis suggested 11 predictors of stroke-like episodes, which were assessed for suitability within a risk prediction model. After simplification and model assessment, four predictors (Body Mass Index, Blood Heteroplasmy, Serum Lactate, NMDAS hearing subscore) were found amongst m.3243A>G carriers and they formed the basis of a 6-point prediction model of stoke-like episode risk that classified patients as low, intermediate or high risk. Analysis of subsequent stoke-like episodes following the occurrence of a first event demonstrated no significant differences in the time to reoccurrence. This model can help inform more tailored genetic counselling and prognosis in routine clinical practice.

To read the full publication: