MELAS syndrome: an acute stroke-like episode complicated by renal tubular acidosis
- Dearbhail Ni Cathain ,
- Emmet Browne ,
- Kieran Skehan and
- Karl Boyle
- Beaumont Hospital, Dublin, Ireland
- Correspondence to Dr Dearbhail Ni Cathain; dearbhailnicathain@rcsi.ie
Abstract
MELAS, a mitochondrially inherited multisystem disorder, can present with acute stroke-like episodes. The literature thus far supports the use of L-arginine therapy in acute MELAS flares to alleviate and shorten the duration of symptoms. This is the case of a patient who presented with ataxia and worsening confusion on a background of genetically confirmed MELAS syndrome. In this instance, intravenous L-arginine therapy, along with corticosteroids, was administered in keeping with best practice. However, in a metabolically vulnerable patient, L-arginine therapy resulted in a further deterioration in his clinical status and the development of a non-anion gap metabolic acidosis.
Background
MELAS is one of the most common mitochondrially inherited disorders. Research studies have supported the use of intravenous L-arginine in acute MELAS flares.1 However, the many studies discussing the proposed benefits associated with intravenous L-arginine therapy fail to acknowledge a risk of renal tubular acidosis. MELAS syndrome predisposes to a metabolically vulnerable state, given the propensity in this patient cohort to lactic acidosis, encephalopathy and often difficult to control diabetes mellitus (with associated high risk for diabetic nephropathy/diabetic ketoacidosis). In MELAS, caution is advised with the use of medications known to increase the risk of acidosis or encephalopathy, such as certain antiepileptic medications or metformin. This case adds to the evolving literature cautioning on the potential risks associated with intravenous L-arginine therapy.
Case presentation
This is the case of a 55-year-old man who presented to a tertiary centre with worsening confusion and ataxia on a background of genetically confirmed MELAS syndrome (3243A>G mutation) complicated by bilateral sensorineural hearing loss, seizure disorder, type 2 diabetes, heart failure (necessitating an ICD) and three previous presentations with stroke-like episodes.
The patient was initially triaged as a potential stroke and underwent formal stroke team assessment. Neurological examination revealed an inability to correctly answer questions, left-sided weakness, ataxia, neglect, aphasia and dysarthria (NIHSS score 7). A non-contrast CT brain scan demonstrated a new right superior parietal area of low attenuation with associated sulcal effacement. This gentleman presented to hospital outside of the time window for thrombolysis (which has yet to be proven beneficial in patients with MELAS) and was not a candidate for thrombectomy. Given the severity of his deficits and the large area of infarction on CT with surrounding oedema, the patient was commenced on intravenous L-arginine therapy—a bolus dose followed by a continuous infusion—along with intravenous corticosteroids. The use of L-arginine and corticosteroids in MELAS flares with CT evidence of oedema is supported in the literature1 2(table 1).
Summary of medications
| Medical reconciliation | |
| Levetiracetam | 1500 mg orally two times per day |
| Aspirin | 75 mg orally once daily |
| Folic acid | 5 mg orally once daily |
| Atorvastatin | 40 mg orally once daily |
| Galfer | 305 mg orally once daily |
| Coenzyme q10 | 200 mg orally three times a day |
| Riboflavin | 100 mg three times a day |
| Calcichew D3 | 500 mg/400 IU orally once daily |
| Clobazam | 10 mg nocte |
| Lansoprazole | 30 mg once daily |
| Lantus | 14units nocte |
| Novorapid | Three times a day with meals |
| Dapagliflozin | 10 mg orally once daily |
| Sitagliptin | 5 mg orally once daily |
| L-carnitine | 1 g orally two times per day |
| L-arginine | 0.5 g/kg bolus plus continued infusion |
| Dexamethasone | 4 mg orally once daily |
The severity of the patient’s new neurological deficits became apparent through continued inpatient assessments. He had severe motor and speech apraxia and also had significant left-sided visuospatial neglect, as demonstrated by a reduced score on the Sunnybrook Neglect Assessment Procedure (64/100) (figure 1).3
Left-sided visuospatial neglect.
Despite L-arginine therapy and ongoing supportive care from the multidisciplinary team, the patient deteriorated neurologically 1 week into his admission, becoming increasingly encephalopathic with a fluctuating level of consciousness ranging from a Glasgow Coma Scale of 9–14. There were no new localising neurological signs on examination and he remained haemodynamically stable.
Investigations
On initial presentation, non-contrast CT brain imaging demonstrated a slightly expansile area of low attenuation change within the right superior parietal lobe with local sulcal effacement. This imaging appearance was felt to be in keeping with a MELAS flare (figures 2 and 3). MRI brain imaging was not possible due to the patient’s ICD. Following the patient’s subsequent neurological deterioration 1 week into his hospital admission, repeat CT brain imaging demonstrated no significant change. Similarly, there was no difference between the patient’s admission electroencephalogram (EEG) and repeat EEG—both of which demonstrated lateralised periodic discharges without seizure activity, consistent with a right-sided acute structural lesion.
CT brain demonstrating classical MELAS findings of bilateral basal ganglia calcifications.
CT brain demonstrating a slightly expansile area of low attenuation change within the right superior parietal lobe with local sulcal effacement.
The patient underwent full biochemical assessment including arterial/venous blood gases, serum and urine electrolyte and serum bicarbonate monitoring throughout the admission (table 2). A new-onset metabolic acidosis coincided with the patient’s in-hospital clinical deterioration that could not be explained by the patient’s baseline raised venous lactate. As would be expected in the case of renal tubular acidosis, the patient had a non-anion gap metabolic acidosis (Anion gap 14.3, Delta gap 2.3 mmol/L, Delta ratio 0.2), a positive urine anion gap (7.7 mmol/L) and an acute drop in serum phosphate.
Blood gases, electrolytes
| Laboratory analysis | |||
| Serum markers | Admission | Deterioration | Normal range |
| pH | 7.38 | 7.11 | 7.35–7.45 |
| PaCO2 | 5.6 | 5.3 | 4.6–6.4 |
| PaO2 | 4.5 | 4.1 | 2.3–5.5 |
| Sodium | 136 | 132 | 136–145 |
| Potassium | 4.6 | 4.1 | 3.4–4.5 |
| Chloride | 104 | 105 | 98–107 |
| Calcium | 1.2 | 1.4 | 1.15–1.27 |
| Glucose | 6.8 | 5 | 3.5–7.7 |
| Lactate | 2.3 | 3.7 | 0.5–1.3 |
| Bicarbonate | 24.8 | 12.7 | 22–26 |
| Phosphate | 0.98 | 0.72 | 0.81–1.45 |
-
PaCO2, partial pressure of carbon dioxide; PaO2, partial pressure of oxygen.
Differential diagnosis
The patient’s neurological deterioration during his hospital admission required a complete reassessment.
From a neurological perspective there was a need to exclude a further stroke-like episode, a worsening of the intracranial oedema or haemorrhagic transformation. Additionally, as the patient had a history of seizures, non-convulsive status epilepticus required consideration and investigation.
In terms of metabolic differentials, the patient had multiple risk factors for the development of a metabolic encephalopathy. His fluctuating levels of consciousness and non-localising deterioration were in keeping with the clinical presentation of encephalopathy. The consideration of worsening lactic acidosis, hyperosmolar hyperglycaemia syndrome, diabetic ketoacidosis, euglycaemic ketoacidosis and hypoglycaemia were warranted. The challenge of achieving glycaemic control was compounded by the patient’s recent course of corticosteroids which required uptitration of his insulin therapy and therefore increased the risk of glycaemic derangement. Given his history of SGLT-2 inhibitor use, euglycaemic ketoacidosis was another consideration.
However, the arterial blood gas that was performed when investigating the above differentials revealed an unexpected non-anion gap metabolic acidosis with evidence of hypobicarbonataemia. No recent change in the patient’s bowel habit made gastrointestinal losses of bicarbonate unlikely. As a result, renal causes for this change in his acid–base status had to be examined. Despite a history of diabetic nephropathy and chronic kidney disease, his creatinine clearance had remained stable throughout admission. His positive urine anion gap and serum electrolyte derangement was in keeping with proximal renal tubular acidosis. Iatrogenic causes of renal tubular acidosis were examined by a thorough drug reconciliation, focusing particularly on newly prescribed medications (table 1). Based on a literature review and advice from the renal service, the metabolic acidosis was deemed likely to be secondary to the commencement of intravenous L-arginine therapy.
Treatment
A number of treatments were instigated including pharmacological treatments and multidiscliplinary team interventions. First, the intravenous L-arginine was discontinued. In order to alleviate the metabolic stress caused by the acidosis, the patient received an intravenous isotonic sodium bicarbonate infusion. He was switched to an oral regime of sodium bicarbonate (in daily divided doses) when appropriate, with daily bicarbonate monitoring. His diabetic control was also optimised with up-titrations of his total daily insulin.
The multidisciplinary team was essential in the patient’s recovery and in the implementation of supportive measures. He required nasogastric feeding due to his fluctuating levels of consciousness; this was guided by the dietician and speech and language therapist. Due to the severe cognitive impairment (with marked language dysfunction, apraxia and visuospatial neglect) caused by the MELAS flare, and the deconditioning related to his subsequent acute deterioration, speech and language therapy, occupational therapy and physiotherapy input was invaluable to the patient’s recovery and in-hospital optimisation. Once stabilised medically, the patient was transferred to an off-site facility for further neurorehabilitation.
Outcome and follow-up
Having had a turbulent and unpredictable hospital course—in part owing to the unpredictable clinical course of a MELAS flare, and then exacerbated by the iatrogenic insult described above—the patient required extensive in-hospital rehabiliation as well as specialised neurorehabilition following discharge. In total, he remained an inpatient in a tertiary hospital for 44 days followed by 14 days of offsite neurorehabilitation. The patient completed his time in the neurorehabilitation centre and returned home having made significant progress.
Discussion
The use of intravenous L-arginine therapy in acute MELAS flares has been endorsed as current best practice with the publication of a consensus statement by the mitochondrial medicine society.1 It has been proposed that stroke-like episodes in MELAS are a complication of an inherent nitric oxide-deficient state that leads to impaired smooth muscle relaxation within the cerebral vasculature, thus resulting in reduced cerebral perfusion.4 The term ‘stroke-like’ episodes was generated as the cerebral perfusion deficits seen in MELAS flares are not in keeping with established vascular territories.5 L-arginine, via endothelial nitric oxide synthase, is a nitric oxide precursor that can increase nitric oxide synthesis and lead to improved endothelial vascular relaxation and therefore limit the associated ischaemic insult.6
Given the proposed benefits, the patient in our case was given intravenous L-arginine therapy at the recommended dosage but consequently developed a non-anion gap metabolic acidosis. Although an apparently rare phenomenon following L-arginine therapy, this is not a unique case, with previously reported events of renal tubular acidosis secondary to over-the-counter amino acid supplementation7 and another documented case in a MELAS flare.8 Previous studies into the use of L-arginine as an antihypertensive agent also demonstrated systemic acidosis,9 and several animal models have demonstrated impaired bicarbonate resorption in the proximal convoluted tubule associated with L-arginine use.10 11
Despite evidence to support renal tubular acidosis as a consequence of L-arginine therapy, clinical guidelines to date1 fail to acknowledge this risk in a patient cohort that is already vulnerable to acid–base disturbances. This case report adds to the scarce literature describing L-arginine-induced metabolic acidosis. Although the proposed benefits of L-arginine therapy may ultimately far outweigh the risk of renal tubular acidosis, it is still important to be cognisant of the associated risks of this therapy and to exercise caution in its use.
Serum bicarbonate monitoring during L-arginine therapy in patients with MELAS could be utilised to identify increased urinary bicarbonate excretion, indicating the development of renal tubular acidosis. The earlier identification of hypobicarbonataemia may facilitate the discontinuation of L-arginine therapy in appropriate cases. Further studies into the efficacy, safety and optimum dosage of L-arginine therapy are ongoing1 and will hopefully shed light on some of these concerns.
Learning points
-
Patients with MELAS syndrome are a metabolically vulnerable cohort and therefore especially susceptible to major acid–base disturbances.
-
L-arginine therapy is associated with renal tubular acidosis.
-
When acute deteriorations occur within the hospital setting, iatrogenic insults must be considered in the differential diagnosis.
Ethics statements
Patient consent for publication
Footnotes
-
Contributors DNC identified the case for reporting and wrote the body of the report with contribution from EB and KS to the clinical presentation and investigations performed. KB was the consultant supervisor for the report and reviewed the piece with amendments/edits, where appropriate.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Competing interests None declared.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2021. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer