Article Text
PDF
Statistics from Altmetric.com
A 23-year-old man developed general malaise and subacute neurological decline, including dysarthria and generalised weakness. He had mild intellectual disability but had been independent for activities of daily living. When aged 9 years, he had a severe episode of hypoglycaemia with associated presumed provoked seizure. Subsequent investigations for relapsing episodes of muscle weakness and speech problems found no cause. When aged 19, he underwent total thyroidectomy for minimally invasive follicular carcinoma.
On examination, he had generalised muscle weakness. His serum creatine kinase was 23 280 U/L (24–195), with normal inflammatory markers. An ECG showed Brugada pattern type I with prolonged QTc at 516 ms (figure 1). He was given hyperhydration for rhabdomyolysis.
ECG showing Brugada type I pattern characterised by coved type ST-elevation in V1 and inverted T-waves in V1–3, together with diffuse ST abnormalities and prolonged QTc (516 ms).
What is rhabdomyolysis and what are the most likely causes?
What are the main differential diagnoses?
Rhabdomyolysis is a potentially life-threatening syndrome characterised by skeletal muscle destruction, releasing intracellular enzymes into the bloodstream and causing systemic complications.1 Symptoms include muscle pain, muscle weakness and ‘tea-coloured’ urine.1 2 With a non-specific presentation, the diagnosis is based on a serum creatine kinase being five times the upper limit of normal (ie, >1000 U/L).1 2 The differential diagnosis includes acquired and inherited causes (table 1).1 3
Differential diagnosis of rhabdomyolysis2
We measured his cardiac enzymes to refine the differential diagnosis and to exclude myocardial infarction. His serum troponin-I was 55 ng/L; this remained relatively stable, peaking at 310 ng/L at 2 weeks.
Tests for infective causes (HIV, cytomegalovirus, Epstein-Barr virus, toxoplasmosis, Lyme disease) were negative, as were serum antinuclear antibodies, antineutrophil cytoplasmic antibodies and paraneoplastic antibodies. We requested tests for myositis blot including antisignal recognition particle and anti-3-hydroxy-3-methyl-glutaryl-coenzyme A reductase antibodies (anti-HMG-CoA).
MR scans of whole-body muscle showed muscle oedema in the shoulders, pelvic girdle, thighs and calves, but with no atrophy or fat infiltration. Cardiac MRI showed normal ventricular function (ejection fraction 60% left ventricle and 62% right ventricle) without delayed enhancement. The normal ventricular wall thickness and motion excluded myocarditis. Muscle biopsy of the vastus lateralis showed a destructive myopathy without lymphocytic infiltration.
We felt this was most likely an idiopathic inflammatory myopathy, specifically an immune mediated necrotising myopathy.
Having tentatively diagnosed immune-mediated necrotising myopathy, we started corticosteroids (prednisolone 1 mg/kg daily) with prophylactic alendronate, calcium, pantoprazole and cotrimoxazole. Five days later, he was found on the floor, initially unresponsive (Glasgow Coma Scale score E1, M1, V1) with urinary incontinence. Within minutes, he moved all limbs symmetrically but remained anxious, confused and emotional.
What is the differential diagnosis of this new clinical presentation?
What is the best way to differentiate?
The collapse was caused either by an epileptic seizure or by a convulsive syncope from bradycardia with hypotension or from tachyarrhythmia. His indifferent crying after the episode suggested a seizure with postevent delirium. The history of epilepsy increased the likelihood of a seizure. He had no classical syncopal prodromal features, such as light headedness, sweating, palpitation, low blood pressure, slow pulse or provocation by standing.4 Also there were no classical features to suggest epilepsy, such as auras of epigastric rising, olfaction or déjà vu, lateral tongue biting or focal neurological deficits.4
Note that urinary incontinence does not help to distinguish seizure from syncope.
EEG showed general slowing without epileptiform abnormalities. CT scan of head was normal. We started levetiracetam. Later that day, he had another witnessed brief episode of unresponsiveness, during which the cardiac monitor showed ventricular tachycardia and torsades de pointes. His QTc interval was prolonged to 554 ms.
Are there iatrogenic factors that might contribute to this clinical presentation?
Both levetiracetam and cotrimoxazole potentially may prolong the cardiac QTc interval, and should be used with caution in those with risk factors such as a congenital long QT syndrome and electrolyte disturbances.5 6 Because the witnessed event had been associated with a cardiac arrhythmia, we stopped the levetiracetam. Torsade de pointes is recognised as a possible side effect of cotrimoxazole,6 but there were no documented electrolyte disturbances.
One week after starting corticosteroids, the myositis blot and anti-HMGCoA reductase antibodies returned as negative and his serum creatine kinase dropped dramatically from a maximum of 37 400–10 750 U/L, suggesting the diagnosis was not immune-mediated necrotising myopathy.
Are there elements in the history that you would look into?
Would you reassess any of the investigations already performed?
We suspected an inherited metabolic myopathy as part of a metabolic disorder had caused the rhabdomyolysis, for two reasons: first, because of the recent complications during hospitalisation (known epilepsy in combination with cardiac arrhythmia) and second because of the remarkable medical history since childhood.
We re-evaluated his metabolic investigations from childhood. In his family history, there had been consanguinity (his mother and father were cousins), but without other affected relatives. MR scan of brain showed no white or grey matter abnormalities apart from a subtle white matter hyperintensity at the left cerebral peduncle. His acylcarnitine profile did not suggest a fatty acid oxidation disorder. Enzyme activity of carnitine palmitoyl transferase-II and long-chain 3-hydroxyacyl-coenzyme A dehydrogenase were normal. His urinary acylcarnitine profile suggested fasting; his serum organic acids were normal. Serum amino acid, creatine, mucopolysaccharide, purine and pyrimidine metabolite, and sialic acid concentrations were all normal. Chromosome analysis was normal and genetic testing excluded fragile X syndrome and showed no deletion or duplication in the Duchenne muscular dystrophy (dystrophin) gene.
Reassessment of his muscle biopsy confirmed active necrosis without lymphocytic infiltration, and so was consistent with the clinical presentation of rhabdomyolysis, but without additional clues to its cause. Generally, muscle biopsy at the time of acute rhabdomyolysis is not recommended, since overwhelming necrosis associated with rhabdomyolysis can obscure any underlying pathology that might have indicated certain inherited disorders.3 However, when suspecting myositis, rapid biopsy may be able to confirm the diagnosis and allow urgent initiation of immunosuppressive agents.3
His hospitalisation was complicated by further neurological decline. Basic communication became impossible. His serum markers of infection and inflammation were increasing, such that we suspected a septic encephalopathy. Our differential diagnosis was of a metabolic crisis or encephalopathy. We cultured Raoultella species from his blood. His blood biochemistry showed a mild hyperammonaemia of 41 µmol/L. We presumed that a central venous line infection had caused his temporary deterioration.
He remained encephalopathic despite antibiotics. There were recurrent non-sustained ventricular tachycardias for which we started propranolol. Unfortunately, he still had an episode of polymorphic ventricular tachycardia, after which he turned his eyes, made orofacial automatisms, with myoclonic jerks and impaired awareness, and then had a postictal episode complicated by sustained ventricular tachycardia with loss of output. We started advanced life support, with spontaneous return circulation after 12 min, and he was transferred to the intensive care unit.
Is further genetic screening indicated?
Are there genes of interest that can guide diagnosis?
His family history of consanguinity raised suspicion of an autosomal recessive congenital disorder. The combination of metabolic crises, rhabdomyolysis with a suspicion of an underlying metabolic myopathy, neurodegeneration, epilepsy and cardiac arrhythmias in a patient with underlying long QTc-interval and Brugada pattern suggested a mutation affecting transport and Golgi organisation 2 (TANGO2). Single gene testing confirmed biallelic deletion in 22q11.21 of exons 3–9 of the TANGO2 gene (figure 2). His clinical characteristics and reports of similar case series encouraged single-gene testing before whole exome sequencing.
Schematic overview of biallelic alterations found by single gene next generation sequencing (NGS) missing coverage of exons 3–7 of 22q11.21.
This patient did not recover completely and remained dependent on his relatives for activities of daily, living only a further 3 months despite having no postanoxic changes on MR scan of brain. Given this poor neurological recovery, we withheld implanting a cardioverter defibrillator.
Discussion
TANGO2 mutations are associated with a multisystem disorder that includes metabolic crises, intellectual disability, rhabdomyolysis, seizures and arrhythmias. Optic nerve atrophy is also described, as is hypoglycaemia and markedly high serum thyroid stimulating hormone.7–10 The metabolic crises can be precipitated by infections, fasting, ketogenic diet or dehydration and typically present with elevated serum lactate, mild hyperammonaemia and hypoglycaemia.7 An expert committee suggested B-complex vitamins supplementation, including vitamins B5 and B9, for patients with genetically confirmed TANGO2-deficiency disorder.11 This patient’s presentation seemed to suggest that some metabolic crises can increase disability. Cardiac arrhythmias can be fatal. Although we do not know the pathophysiology of cardiac arrhythmia in the context of TANGO2, long QT and Brugada ECG patterns are well described and normally occur in hereditary disorders that affect cardiac ion channels, especially those involving potassium and sodium currents.7–10
This patient had thyroid function abnormalities attributed to an underlying minimally invasive follicular carcinoma.7–10 As these patients may develop hypothyroidism, this might be part of the clinical phenotype, though may be coincidental.
This case highlights the pitfalls of clinical and technical investigations. A false positive diagnosis of an immune-mediated necrotising myopathy led to immunosuppressive therapy that might have worsened the central venous line infection, worsened the QTc-prolongation and led to cardiac arrhythmia. Through clinical reasoning, we diagnosed an extremely rare genetic condition, highlighting the relevance of low-grade evidence in the context of rare diseases. The importance of a definite diagnosis cannot be underestimated, given the consequences on prognosis and limited therapeutic options.
Data availability statement
No data are available.
Ethics statements
Patient consent for publication
Footnotes
Twitter @jonastoeback
Contributors All authors contributed.
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 by Jon Walters, Swansea, UK.