Profiles of paediatric patients experiencing stroke-like episodes associated with mitochondrial disease

Abstract

Introduction:

Stroke-like episodes (SLE) are defined as events characterized by the sudden onset of neurological symptoms with clinical manifestations similar to those of a stroke. However, they are distinguished by the presence of radiological lesions that do not conform to single vascular territory. MELAS syndrome, which is characterized by metabolic encephalopathy, lactic acidosis, and SLE, has been identified as the first genetically defined and most widely known mitochondrial cause of SLE. It has been demonstrated that SLE may occur in the course of a variety of mitochondrial diseases, including those that are the result of nuclear DNA mutations.

Objective:

In this retrospective, multicenter, observational cohort study, we sought to determine the clinical, radiological, EEG, and genetic characteristics of patients with mitochondrial gene mutations presenting with SLE and the frequency and treatment of SLE.

Methods:

Thirty-four patients with a genetically diagnosed mitochondrial disease from 9 paediatric metabolic disease centres in the Marmara Region of Turkey were included in the study, of whom 13 pateints had SLEs. Demographic characteristics, symptoms, clinical features, cranial MRI, EEG findings, and genetic characteristics were evaluated.

Conclusion:

In this study, stroke-like episodes in genetically defined mitochondrial disorders were most frequently observed in MELAS and POLG mutations, and rarely in CoQ10 deficiency, Leigh syndrome cases. Cranial MRI findings are often frontotemporal in location and inconsistent with vascular distribution, and focal epileptiform activity on EEG are diagnostically significant. In MELAS, clinical improvement was observed in patients when L-arginine was initiated in the acute period. The findings emphasise that SLE should be evaluated in the differential diagnosis of sudden onset neurological symptoms in mitochondrial diseases.

Introduction

Stroke is an important neurologic emergency that may lead to serious morbidity and mortality, although it is observed more rarely in childhood than in adults. The World Health Organization defines stroke as “a rapidly developing set of clinical findings due to focal or global impairment of cerebral functions, lasting longer than 24 h or resulting in death” (1). While the etiologies of stroke in the pediatric population spread over a wide spectrum, hereditary metabolic diseases are among the rarer causes. Nevertheless, they have special importance in terms of the need for early diagnosis and specific treatment (2, 3). SLE are characterized by acute or subacute onset neurologic findings do not conform to single vascular territory (4). The best-defined example in the mitochondrial disease group with SLE is MELAS (mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes) syndrome (5). In addition, Kearns-Sayre syndrome, Leigh syndrome, MERRF, LHON (Leber’s hereditary optic neuropathy), and non-syndromic mitochondrial diseases, especially POLG mutations, have also been associated with similar clinical pictures (6, 7). The clinical spectrum is quite broad. While migraine-like headaches and epileptic seizures are frequently seen before the episode, various symptoms, including altered consciousness, hemiparesis, aphasia, cortical visual loss, and neuropsychiatric findings, may occur in the acute phase of the episode (7, 8). On imaging, lesions are frequently localized in occipital, parietal, or posterior temporal lobes, limited to the cortex, asymmetric, and incompatible with vascular distribution (4, 9). Electroencephalography (EEG) findings contribute to the diagnosis of encephalopathic changes and focal epileptiform activity (10). Treatment aims to reduce metabolic stress, ensure seizure control, and avoid agents with the potential for mitochondrial toxicity. L-arginine treatment administered in the acute period has been reported to provide symptomatic improvement in some patients (11, 12).

In this study, we retrospectively evaluated the clinical, laboratory, imaging, and EEG characteristics of SLE in children with molecularly confirmed mitochondrial disease who were followed up in nine different pediatric metabolic centers across Turkey.

Methods

The retrospective medical records of 34 patients who were followed up in the pediatric metabolism and nutrition divisions of nine centers in the Marmara Region in Turkey were analyzed. Inclusion criteria were children and young adults with a genetically confirmed mitochondrial disease and at least one follow-up visit in the participating centers. Exclusion criteria were incomplete molecular diagnosis or lack of accessible clinical data. Current age, age at first episode and age at diagnosis, clinical features, growth parameters, laboratory, radiological, EEG and molecular analysis findings were recorded. For cases with incomplete EEG or MRI data, the available findings were included and explicitly noted as missing when absent. To ensure comparability, standardized data collection templates were used across all nine centers, and case information was harmonized by a coordinating investigator. The data obtained in the study were analyzed using SPSS 23 (Statistical Package for Social Sciences) software. In statistical analyses, Chi-square analysis and Fisher Exact test were used to examine the relationships between categorical variables. Kruskal-Wallis test and Mann- Whitney U test were used to evaluate the differences between continuous variables since parametric test assumptions were not met. In all analyses, the significance level was determined as p < 0.05 and the confidence level was evaluated as 95%. Ethical approval was obtained from the Tekirdağ Dr. İsmail Fehmi Cumalıoğlu City Hospital Clinical Research Ethics Committee (Decision no. 2024/119). Research was conducted accordance with the Declaration of Helsinki.

Findings

Demographic characteristics

Among the 34 patients with a genetic diagnosis of mitochondrial diseases, 50% (n = 17) were female and 50% (n = 17) were male. Regarding age distribution, the highest proportion was observed in the 5.01–10-year age group at 23.5% (n = 8), whereas the lowest proportion was in the 10.01–15-year age group at 14.7% (n = 5). The most common period for the onset of initial symptoms was under 5 years of age, accounting for 58.8% (n = 20), while the lowest frequency was recorded in the 15.01–20-year age group at 3% (n = 1). In terms of age at diagnosis, 41.2% (n = 14) of the patients were diagnosed before the age of 5 years, whereas 14.7% (n = 5) were diagnosed after the age of 20 years (Tables 1, 2).

Genetic diagnoses and distribution by diagnosis

Of the 34 patients with a genetic diagnosis of mitochondrial diseases, 44.1% (n = 15) were diagnosed with MELAS, 26.5% (n = 9) with Coenzyme Q10 deficiency, 11.8% (n = 4) with Leigh syndrome and 8.8% (n = 3) with POLG mutation. There were also cases with FBXL4 mutation 5.9% (n = 2), LHON 2.9% (n = 1) (Table 1).

Stroke-like episodes

Of the 13 patients with SLE, 46.1% (n = 6) were diagnosed with MELAS and 23.1% (n = 3) with POLG. These episodes were also observed in individuals diagnosed with Coenzyme Q10 deficiency 15.4% (n = 2), Leigh syndrome 15.4% (n = 2) (Figure 1). A comparative analysis of the data reveals that the most prevalent age group experiencing SLE was 5.01–10 years (38.4%), followed by 15.01–20 years (30.8%). The highest proportion of individuals without an episode was in the age group above 20 years, with 28.5%. However, no statistically significant relationship was found between age groups and SLE (χ² = 6.703; p = 0.152) (Table 3).

Figure 1

Clinical findings associated with stroke episodes

The prevalence of seizures was found to be significantly higher in patients with a SLE compared to those without such an episode (64.3% vs. 22.7%, p < 0.05) (p = 0.013). Psychiatric disorders were also associated with SLE; being present in 44.4% of those who had an episode and 9.5% of those who did not (p = 0.049). Ophthalmoplegia was reported only in individuals who had an episode, and both findings were statistically significant (p = 0.048) (Figure 2).

Figure 2

When genetic subtypes were analyzed, migraine headache was reported 100% in FBXL4 mutations, 88.9% in Coenzyme Q10 deficiency, 75% in Leigh syndrome, and 40% in MELAS. Cognitive decline was present in all patients (100%) with Leigh, FBXL4 mutations and LHON. Decreased muscle mass was significantly observed in patients with MELAS, FBXL4. Cataracts were detected in 50% of individuals with POLG mutation only, which was distinctive in this respect. Hearing loss was observed in 50% of patients with FBXL4 mutation. An evaluation of body mass index (BMI) revealed mean BMI values of 16.49 ± 3.85 in patients with MELAS, 23.83 ± 11.47 in patients with CoQ10 deficiency, 11.98 ± 2.97 in Leigh syndrome, and 27.34 ± 15.37 in the FBXL4, LHON group. Although no significant difference was found between the diagnosis groups (χ² = 7.371; p = 0.061), the p-value is close to the significance limit and suggests that there may be BMI trends according to diagnosis.

Seizures and EEG findings

A history of seizures was identified in 9/13 patients with stroke like episodes, while 4/13 had no history of seizures. In the evaluation of seizure types, 5/13 of patients experienced generalized seizures and 4/13 experienced focal seizures. Analysis of EEG findings revealed that EEG evaluation was not performed in 3/14 of the patients, whereas focal epileptiform activity was detected in 5/13. Additionally, generalized epileptiform activity was observed in 3/13 of patients, while 6/13 showed no evidence of generalized epileptiform activity (Table 4).

Magnetic resonance imaging (MRI) findings

Cranial MRI revealed lesions in 15/30 of the patients. Of the lesions, 11/15 affected both hemispheres, 3/14 affected the left hemisphere, and 1/15 affected the right hemisphere. Lesions were most frequently observed in the frontal (4/11), temporal (4/11) and parietal lobes (3/11), with lower rates in the occipital lobe (1/11). Regarding lesion resolution, disappearance occurred within 2–6 months in 3/8 of patients, between 6 months and 1 year in 12.5% 1/8, and after more than 1 year in 4/8 of the patients with available data (Table 5).

MR spectroscopy findings

MR spectroscopy was performed in 13 / 34 of the patients; no lactate peak was found in 8/13, a lactate peak was found in 5/13 of the patients who underwent MR spectroscopy (Table 5).

Laboratory findings

The mean plasma lactic acid level was 22.61 ± 23.24 mg/dL in individuals who had a SLE, while this value was 10.21 ± 12.39 mg/dL in those who did not have an episode. The difference between the two groups was not statistically significant (z = −1.785; p = 0.078).

L-arginin treatment

A total of 12 patients (n = 10 MELAS, n = 1 Coenzyme Q10 deficiency, n = 1 Leigh syndrome) received L-arginine treatment. Among the 13 patients who experienced SLE, 6 received L-arginine therapy during the acute phase, and 4 of them continued long-term oral L-arginine treatment (Tables 2, 6).

Discussion

In this multicenter observational cohort study, the clinical, radiological, EEG, and genetic features of genetically diagnosed mitochondrial diseases, the frequency of stroke-like episodes, associated findings, and L-arginine treatment were evaluated. In a study by Ng YS and colleagues on the prediction of stroke-like events and outcomes in mitochondrial diseases, 111 patients aged 1–72 years diagnosed with mitochondrial disease were retrospectively evaluated (13). This study indicated that stroke-like episodes in mitochondrial diseases are also seen in the paediatric age group. In our study, we evaluated paediatric patients who experienced stroke-like episodes associated with mitochondrial disease. Although the data obtained were generally consistent with the existing literature, they also showed some remarkable differences. SLE were most frequently observed in individuals diagnosed with MELAS (44.1%), as reported in the literature (11). In the majority of these patients (66.6%), m.3243A > G mutation was found in the MT-TL1 gene, and this rate was consistent with previously reported data (14). POLG mutation stood out as the second most common cause of SLE, and this was consistent with the frequency ranking in the literature (13). Coenzyme Q10 (CoQ10) deficiency is one of the rare mitochondrial disorders that develop due to mutations in genes involved in ubiquinone synthesis. In the literature, there are a limited number of cases and small patient series in which SLE have been reported to be associated with mutations in biosynthesis genes such as COQ8A (ADCK3) and COQ4 (15, 16). In our study, SLE were observed in two of nine patients diagnosed with CoQ10 deficiency. This finding indicates that CoQ10 deficiency may rarely lead to SLE and emphasizes that neurologic symptoms should be carefully evaluated in this patient group. SLE associated with Leigh syndrome have rarely been described in the literature. In a case report, a three-month-old female infant with post-infectious encephalopathic features and neurologic symptoms, including hypotonia, apnea, and optic atrophy was diagnosed with Leigh syndrome. This case suggests that metabolic stress may trigger neurologic findings in Leigh syndrome, and the disease may present with stroke-like episodes (17). In our series, it was observed that two of the four patients with Leigh syndrome developed stroke-like episodes in relation to metabolic stress. This observation reveals that Leigh syndrome may present not only as a progressive neurodegenerative disorder but also with acute neurologic decompensations triggered by metabolic stress. It emphasizes the necessity for early diagnosis, close follow-up, and prevention of metabolic crises in this patient group. In a study conducted by Durrleman et al. (18) on 60 pediatric patients, it was reported that the first symptoms of the patients started between 0 and 28 months, and the first SLE occurred between 17 and 124 months. In a study conducted by Ng et al. (13) on Forecasting stroke-like episodes and outcomes in mitochondrial diseases, it was noted that 32% of patients experienced their first stroke episod after the age of 40. In our study, it was observed that the first symptoms usually started before the age of 5 years, and the first SLE occurred mostly in the age range of 5.01–10 years (38.4%). The youngest patient was 1.9 years old, and the oldest patient was 40 years old. This shows that the development of episodes is not limited by age and can occur at any age. Gender distribution was equal (F/M: 7/7), and gender was not a determining factor in episodes. In this respect, it differs from some studies in the literature (18). An analysis of the clinical symptoms revealed that headache was prevalent before a SLE. This was followed by decreased muscle mass, myopathy, and seizures. In a study by Xu et al. (14) involving children with MELAS syndrome, muscle weakness was defined as an indicator of severe mitochondrial dysfunction. In studies, seizures are frequently reported as the first symptom of SLE. In mitochondria-damaged cells, the increased metabolic demand associated with excessive neuronal activity during seizures cannot be met. This leads to local energy deficits, ion imbalances, and intracellular lactic acid accumulation. This situation can cause cell damage, vascular irregularities, and associated SLE. The high seizure frequency found in our study and its significant association with stroke- like episodes support the hypothesis that these episodes may be triggered by epileptic seizures (14, 18). In a study conducted by Ng et al. (13) on MELAS patients, it was reported that 91% of the patients had symptoms indicating mitochondrial dysfunction before the SLE, and the most common symptom was sensorineural hearing loss. In our series, findings such as hearing loss, cortical blindness, recurrent vomiting, and myopathy were frequently observed before SLE in individuals diagnosed with MELAS. In addition, sensorineural hearing loss was one of the most common symptoms in these patients. Although no statistical significance was found between SLE and some demographic variables such as age, gender, and BMI, low BMI in MELAS and Leigh patients is remarkable. This indicates that short stature and nutritional deficiency also accompany these patients and is in parallel with the literature (13). Studies have shown that brain lactate levels increase with disease progression in MELAS patients, and this increase reflects a progressive shift in oxidative redox potential. These findings suggest that lactate levels may be an important biomarker for monitoring the course of the disease and establishing a diagnosis (19). While studies suggested that lactate levels may be an important biomarker for monitoring the course of the disease and establishing a diagnosis, the variation does not clearly dichotamise those with or without stroke like episodes in our study. In radiologic evaluations, Durrleman et al. (18) reported that bilateral cortical diffusion hyperintensity and hyperperfusion were prominent in cranial MR images obtained in the first 48 h, and this condition developed due to energy metabolism disorders different from classical ischemic strokes (13). In our study, bilateral lesions were observed in all seven patients imaged in the acute period, and the most common site of involvement was frontotemporal lobes. When EEG findings were analyzed, focal epileptiform activity was found to be more frequent in accordance with the literature (13). Regarding the use of L-arginine treatment, a retrospective analysis by Ganetzky et al. (20) reported that clinical improvement was achieved in 47% of patients who received intravenous (iv) arginine treatment. A better response to treatment was observed, especially in cases of hemiplegic attacks. In our series, six patients with SLE received iv arginine treatment in the acute phase of the episode, and clinical improvement was observed. Seven of our patients who did not receive L-arginine treatment (3 POLG, 2 MELAS, 1 Coenzyme Q10 deficiency, 1 Leigh) experienced prolonged headaches, nausea and visual impairment. In Durrleman et al.’s (18) study, recurrent stroke-like episodes were reported in three patients over an eight-year follow-up period, whereas in our series, six patients experienced recurrent stroke-like episodes, with five episodes recorded in one MELAS patient and four in another. In systematic reviews on the efficacy of oral L-arginine treatment, it was emphasized that the clinical benefit of acute or prophylactic use was limited, but methodological deficiencies were found in these studies (21). In a study including six MELAS patients, it was reported that oral L-arginine treatment at a dose of 0.15–0.3 g/kg/day administered for 18 months resulted in significant improvement (22). In our study, 4 of our 13 patients with SLE continued long-term oral L-arginine treatment. Six of our MELAS patients without SLE also received long-term oral L-arginine treatment. While no episodes were observed in 6 of these patients, a median of 2 attacks were observed in 4 patients.

The study is robust in its examination of SLE in paediatric mitochondrial cases with molecular diagnosis, utilising multicentre and detailed clinical-imaging data. However, the small sample size, retrospective design, resulting lack of imaging data, and limited treatment outcomes represent significant limitations; conclusions regarding L-arginine efficacy are purely hypothesis-generating.

Conclusion

This study showed that SLE in genetically diagnosed mitochondrial diseases are most commonly observed in MELAS and POLG mutations. However, they may also develop in rare subtypes such as CoQ10 deficiency and Leigh syndrome. During episodes, the presence of bilateral cortical lesions on cranial MRI—often located in the frontotemporal regions and inconsistent with vascular distribution—and the detection of focal epileptiform activity on EEG were diagnostically significant findings. Clinical improvement was observed with iv L-arginine treatment initiated in the acute period. The findings emphasize that Stroke-like episodes should be considered in the differential diagnosis of sudden onset neurological findings in mitochondrial diseases.

References

• 1.

  Aho K Harmsen P Hatano S Marquardsen J Smirnov VE Strasser T . Cerebrovascular disease in the community: results of a WHO collaborative study. Bull World Health Organ. (1980) 58:113–30.

  • Google Scholar

• 2.

  Pavlakis SG Kingsley PB Bialer M . Stroke in children: genetic and metabolic issues. J Child Neurol. (2000) 15:308–15. doi: 10.1177/088307380001500507

  • CrossRef
  • Google Scholar

• 3.

  Balcı MC . Neurometabolic strokes in childhood In: AydınlıNAksu UzunhanT, editors. Pediatric stroke. Ankara: Türkiye Klinikleri (2021). 14–20.

  • Google Scholar

• 4.

  Bianchi MC Sgandurra G Tosetti M Battini R Cioni G . Brain magnetic resonance in the diagnostic evaluation of mitochondrial encephalopathies. Biosci Rep. (2007) 27:69–85. doi: 10.1007/s10540-007-9046-z

  • CrossRef
  • Google Scholar

• 5.

  Goto Y Nonaka I Horai S . A mutation in the tRNA(Leu)(UUR) gene associated with the MELAS subgroup of mitochondrial encephalomyopathies. Nature. (1990) 348:651–3. doi: 10.1038/348651a0,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6.

  Brinjikji W Swanson JW Zabel C Dyck PJ Tracy JA Gavrilova RH . Stroke and stroke-like symptoms in patients with mutations in the POLG1 gene. JIMD Rep. (2011) 1:89–96. doi: 10.1007/8904_2011_22

  • CrossRef
  • Google Scholar

• 7.

  Finsterer J Wakil SM . Stroke-like episodes, peri-episodic seizures, and MELAS mutations. Eur J Paediatr Neurol. (2016) 20:824–9. doi: 10.1016/j.ejpn.2016.08.002

  • CrossRef
  • Google Scholar

• 8.

  Finsterer J Aliyev R . Metabolic stroke or stroke-like lesion: peculiarities of a phenomenon. J Neurol Sci. (2020) 412:116726. doi: 10.1016/j.jns.2020.116726,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9.

  Divakar M Gertler T . Stroke snapshot: metabolic stroke. Pract Neurol. (2020) 20:e1–5. doi: 10.1136/practneurol-2019-002365

  • CrossRef
  • Google Scholar

• 10.

  Finsterer J . Mitochondrial metabolic stroke: phenotype and genetics of stroke-like episodes. J Neurol Sci. (2019) 400:135–41. doi: 10.1016/j.jns.2019.03.021

  • CrossRef
  • Google Scholar

• 11.

  Sproule DM Kaufmann P . Mitochondrial encephalopathy, lactic acidosis, and strokelike episodes: basic concepts, clinical phenotype, and therapeutic management of MELAS syndrome. Ann N Y Acad Sci. (2008) 1142:133–58. doi: 10.1196/annals.1444.011

  • CrossRef
  • Google Scholar

• 12.

  Koga Y Akita Y Junko N Yatsuga S Povalko N Fukiyama R et al . Endothelial dysfunction in MELAS improved by l-arginine supplementation. Neurology. (2006) 66:1766–9. doi: 10.1212/01.wnl.0000220197.36849.1e

  • CrossRef
  • Google Scholar

• 13.

  Ng YS Lax NZ Blain AP Erskine D Baker MR Polvikoski T et al . Forecasting stroke-like episodes and outcomes in mitochondrial disease. Brain. (2022) 145:542–54. doi: 10.1093/brain/awab353,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14.

  Xu C Dai S Jiang H Liu Z Zhang Y Song T et al . Clinical characteristics and long-term prognosis of 150 children with MELAS syndrome in China. J Neurol. (2025) 272:280. doi: 10.1007/s00415-025-13009-z

  • CrossRef
  • Google Scholar

• 15.

  Ramesh R Hariharan S Sundar L . Stroke-like episodes and epilepsy in a patient with COQ8A-related coenzyme Q10 deficiency. Ann Indian Acad Neurol. (2023) 26:980–2. doi: 10.4103/aian.aian_511_23,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16.

  Bosch AM Kamsteeg EJ Rodenburg RJ van Deutekom AW Buis DR Engelen M et al . Coenzyme Q10 deficiency due to a COQ4 gene defect causes childhood-onset spinocerebellar ataxia and stroke-like episodes. Mol Genet Metab Rep. (2018) 17:19–21. doi: 10.1016/j.ymgmr.2018.09.002,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17.

  Haliloğlu B Fedakar A . Ensefalopati tablosuyla gelen Leigh sendromu. Çocuk Dergisi. (2010) 10:94–7. doi: 10.5222/j.child.2010.094

  • CrossRef
  • Google Scholar

• 18.

  Durrleman C Grevent D Aubart M Kossorotoff M Roux CJ Kaminska A et al . Clinical and radiological description of 120 pediatric stroke-like episodes. Eur J Neurol. (2023) 30:2051–61. doi: 10.1111/ene.15821,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 19.

  Gao R Gu L Zuo W Wang P . Long-term prognostic factors and outcomes in mitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes: a clinical and biochemical marker analysis. Front Neurol. (2024) 15:1491283. doi: 10.3389/fneur.2024.1491283,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20.

  Ganetzky RD Falk MJ . 8-year retrospective analysis of intravenous arginine therapy for acute metabolic strokes in pediatric mitochondrial disease. Mol Genet Metab. (2018) 123:301–8. doi: 10.1016/j.ymgme.2018.01.010,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21.

  Stefanetti RJ Ng YS Errington L Blain AP McFarland R Gorman GS . L-arginine in mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes: a systematic review. Neurology. (2022) 98:e2318–28. doi: 10.1212/WNL.0000000000200299,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22.

  Almannai M El-Hattab AW . Nitric oxide deficiency in mitochondrial disorders: the utility of arginine and citrulline. Front Mol Neurosci. (2021) 14:682780. doi: 10.3389/fnmol.2021.682780,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar