Statistics from Altmetric.com
A 70 year old male presented in February 2000 with persistent cough. Chest radiograph revealed a mediastinal abnormality. A computed tomography (CT) scan of the thorax confirmed an anterior mediastinal mass. CT guided mediastinal biopsy was performed but proved inconclusive. The patient declined further invasive investigation for 2 years. After follow up CT showed significant enlargement he underwent surgery in September 2002. An excellent post operative recovery was made. Histology confirmed a minimally invasive thymoma.
However, 6 weeks later he developed progressive proximal leg weakness, lethargy, and night sweats. He described a sensation like “someone touching his legs” and became agitated and restless at night. He was readmitted in November 2002 and on the first night became acutely confused and markedly disorientated with visual hallucinations.
On examination he was agitated and confused. His eyes opened spontaneously, there was no coherent verbalisation, and he withdrew to pain. He was apyrexial although there was marked hyperhidrosis and hypersalivation. Blood pressure was 120/70 mm Hg, and pulse was 98 bpm and regular. He had profuse diarrhoea although his abdomen was soft and non-tender. There were irregular myoclonic jerks of all four limbs predominating in his upper limbs as elbow flexion jerks. There was no startle reaction. Cranial nerve examination was unremarkable. Power and deep tendon reflexes were normal and plantar responses flexor. Intravenous phenytoin was administered with resolution of the myoclonic movements.
Laboratory investigations including electrolytes, auto-antibody screen, and anti-neuronal antibodies were unremarkable. Cerebral spinal fluid (CSF) was clear, colourless, and acellular, CSF glucose was 4.0 mmol/l (2.2–4.4) (serum 6.7), and CSF protein 0.25 g/l (0–0.40). Oligoclonal bands were negative in CSF and serum and no abnormality of the immunoglobulin G pattern was detected. CSF analysis for the 14-3-3 (prion) protein was negative.
Chest radiograph showed postoperative changes. Brain magnetic resonance imaging (MRI) disclosed multiple areas of periventicular and subcortical white matter signal change especially within the right parietal region, felt likely to represent diffuse cerebrovascular changes.
Serial electrocardiograms (ECGs) demonstrated sinus tachycardia. Serial electroencephalograms (EEGs) showed diffuse slowing only. Nerve conduction studies revealed mild motor conduction slowing (right median 49 m/s, right common peroneal 43 m/s). Sensory nerve amplitudes were small with normal velocities. Repetitive discharges were noted following evoked compound muscle action potential in upper and lower limb motor nerves. Doublet, triplet, and multiplet were recorded in abductor pollicis brevis, tibialis anterior, and extensor digitorum brevis indicating motor axon membrane instability compatible with neuromyotonia (fig 1).
Confusion, visual hallucinations, insomnia, anxiety, sweating, diarrhoea, a slurred dysarthria, and abnormal muscle activity fluctuated dramatically from day to day, but there was a steady overall deterioration. In a lucid interval he scored 68/100 using Addenbrooke’s bedside testing of cognitive function (equivalent to a Mini Mental State Examination (MMSE) score of 18/30), performing poorly in orientation, attention/concentration, verbal fluency, and visuospatial abilities. He became chair bound despite little clinical weakness either proximally or distally, and later bed bound. Although never as prominent as on first admission, there was frequent muscle twitching, at best resembling myokymia, at worst multifocal myoclonus. He was transferred to a high dependency unit.
The diagnosis of Morvan’s syndrome was based on the clinical phenotype and supported by the nerve conduction studies and the presence of antibodies to voltage gated potassium channel antibodies (VGKC) (165 pmol; normal <100 pmol), 10 weeks post-thymectomy.
At 10 weeks post-thymectomy he had 5 days of plasma exchange (PE). There was rapid improvement in orientation and memory and a cessation of abnormal muscle movements within 5 days. Prednisolone was commenced 40 mg daily and phenytoin continued. He made a dramatic clinical recovery, standing unassisted 10 days post-exchange and mobilising independently by 14 days, with no further muscle twitching. Addenbrooke’s test 2 weeks post-PE revealed a score of 78/100 (MMSE 24/30).
Post-PE nerve conduction studies and EMG showed an increase in the motor conduction velocities (right median 56 m/s, right common peroneal 45 m/s) and resolution of the previous features of neuromyotonia. Repeat serum VGKC antibody levels at 1 and 4 months post-PE were within normal limit. EEGs both 2 and 6 months after treatment showed little improvement. MRI 6 months after treatment also showed no change.
At 8 months post-PE, on 10 mg prednisolone daily he remained asymptomatic aside from fatigue. His wife reported that apart from irritability his personality and memory were essentially back to normal, although on repeat Addenbrooke’s he scored only 82/100. He has even returned to playing golf.
Since Morvan’s original publication only a few cases have been described. The majority appear to be a paraneoplastic manifestation.1,2 The discovery of VGKC antibodies in several of these cases, as seen in many cases with neuromyotonia,3 has suggested that Morvan’s syndrome may be an autoimmune disorder. Whether VGKC antibodies play a pathogenic role in the encephalopathy as they do in the peripheral nervous symptoms is as yet unclear. Others1,2 have suggested that the VGKC antibodies may cross the blood–brain barrier and act centrally, binding predominantly to thalamic and striatal neurons2 causing encephalopathic and autonomic features. The reversibility of the encephalopathy with plasmapheresis does suggest that the encephalopathy is also mediated by serum factors. Liguori et al reported the presence of weak CSF oligoclonal bands, absent in the serum, supporting a central immunological role.2 There are also reports of non-paraneoplastic limbic encephalitis associated with raised serum VGKC4 suggesting that these antibodies may give rise to a spectrum of neurological disease presenting with symptoms arising peripherally, centrally, or both.5 However in our case and the case reported by Lee et al1 oligoclonal bands were absent in CSF and serum, and CSF immunoglobulin profiles were unremarkable.
The natural history of Morvan’s is highly variable. Two cases have been reported to remit spontaneously. In one of these cases remission was associated with a fall in the serum level of VGKC antibodies. Others have required a combination of plasmapheresis and long term immunosuppression,1,2 although in one of these cases the patient died shortly after receiving PE.2 Other fatalities without remission have been described by, amongst others, Morvan himself. Cardiac involvement in some cases may increase vulnerability to sudden death.
Thymectomy has previously been a proposed treatment for Morvan’s syndrome. This is the first reported case of Morvan’s syndrome presenting post-thymectomy. Morvan’s syndrome normally presents with a slow insidious onset over months to years.1,2 Our case is unique in that presentation was over days, just weeks post-thymectomy, and responded to a single PE course with low dose immunosuppression. We hypothesise that surgery may have precipitated a rise in the serum VGKC antibodies levels, which were cleared by one course of PE resulting in remission as supported by the drop in serum VGKC levels. Although potentially a low risk of thymectomy, it is an important complication to recognise because of the dramatic reversibility to treatment.