A case report of rigidity and recurrent lower limb myoclonus: progressive encephalomyelitis rigidity and myoclonus syndrome, a chameleon

Background Progressive encephalomyelitis with rigidity and myoclonus (PERM) syndrome is a rare neurological condition. Its clinical characteristics include axial and limb muscle rigidity, myoclonus, painful spasms and hyperekplexia. Diagnosis of this disease can be very challenging and optimal long-term treatment is unclear. Case presentation We report a case of a 62 year old patient admitted for repetitive myoclonus and rigidity in the lower limbs progressing since 10 years, associated with a fluctuating encephalopathy requiring stays in Intensive Care Unit. Multiple diagnostics and treatment were proposed, unsuccessfully, before the diagnosis of PERM syndrome was established. In association with the clinical presentation, a strong positive result for GAD (glutamic acid decarboxylase) antibodies lead to the diagnosis of PERM syndrome. Conclusions PERM syndrome is a rare disease and its diagnosis is not easy. Once the diagnosis is established, the correct treatment should follow and could be lifesaving, regardless of a delayed diagnosis. Maintenance of long-term oral corticotherapy is suggested to prevent relapses.


Case presentation
The clinical story of this 62 year old woman started in 2007 with lower back pain, lumbar muscular contractures and back rigidity. After being examined by an orthopaedic surgeon, she was diagnosed with congenital scoliosis. A lumbar interbody fusion was subsequently performed with no significant impact on the symptoms.
Between 2007 and 2014, she presented multiple episodes of lower limb myoclonus. Episodes were often associated with hyperthermia. A diagnosis of spinal myoclonus was suggested and a treatment with clonazepam began.
In 2014, she presented with lower limb myoclonus, confusion, hyperthermia and acute respiratory failure. She was admitted to the intensive care unit (ICU). Differential diagnoses at the time included NMS versus meningoencephalitis. Nevertheless, no past history of neuroleptic use was identified and multiple attempts of lumbar puncture were unsuccessful. Large spectrum antibiotics were administered with favourable clinical evolution.
She finally presented in our hospital in our emergency department (ED) in January 2017 for recurrent myoclonus and a sudden onset of bilateral leg weakness in the context of an influenza infection. Upon arrival at the ED, moderate hypotension and fever (39.4°C) were observed. Neurological examination showed bradyphrenia, however she followed one-step commands. There was a motor deficit of the lower extremities scored 4 on the Medical Research Council scale (MRC) [25], diminished and symmetrical knee and ankle jerks, myoclonus and rigidity in the lower limbs. Initial laboratory tests showed normal white cell count with moderately increased C reactive protein (CRP; 39 mg/l, normal: < 5 mg/l). Creatine kinase (CPK) was elevated at 52144 U/ L (normal: 10-170 UI/l) with an acute renal insufficiency (clearance of 14 ml/min). Clonazepam doses were increased from 1 mg to 2 mg three times daily and antibioc therapy with amoxicillin/clavulanic acid was empirically started.
The day after admission in the neurology ward alteration of the level of consciousness, persistent hypotension and respiratory distress were observed and the patient was subsequently admitted to ICU. In the following days, there was a progressive deterioration of the motor function in the lower limbs until paraplegia was reached while lower limb rigidity and myoclonus persisted. Patient contact was lost with respiratory failure, necessitating invasive mechanical ventilation during one week.
Brain and spinal magnetic resonance imaging (MRI) revealed a pituitary adenoma, cervical osteoarthritis and thoraco-lumbar discopathy without stenosis. EMG showed continuous rhythmic motor unit activity within the axial muscles. CPK continued to rise in the following days to peak of 74,410 U/L. CSF analysis was normal, polymerase chain reaction (PCR) for herpes simplex virus 1 (HSV1), HSV 2 and enterovirus were negative (Table 1).
In doubt, anti-epileptic treatment with levetiracetam (2 g twice daily) was started. Nonetheless, a consecutive twenty-four-hour electroencephalogram (EEG) demonstrated numerous muscular artefacts but no epileptic activity (Table 1).
Initial treatment was therefore started with gammaglobulins (0.4 g/kg/day) during 5 days with poor results. Concomitant corticotherapy was started with IV hydrocortisone (50 mg, three times daily) for 10 days, which was then continued orally. The treatment was continued with five sessions of plasma exchange (2 L). Recovery of her consciousness followed within days and myoclonus ceased. She was extubated one week after the end of plasma exchange sessions. Total duration of the ICU stay accounted 33 days. The patient was then transferred to the neurology ward where she recovered progressively with respect to motor function; she was able to walk with a walker when discharged to a rehabilitation centre.
While still in rehabilitation, our patient had a short relapse of the myoclonus in March 2017, only after she was already shortened on her corticosteroids and anti-epileptics. Decision was made to keep her on a regimen of levetiracetam 2 g twice a day and hydrocortisone 20 mg twice daily. She left the rehabilitation centre in May 2017 with a favourable evolution despite persistent mild lower proximal limb weakness (scored 4 on the MRC scale) and gait ataxia.
In August 2017, seven months after admission, she had no relapse of the myoclonus nor the rigidity and was able to walk without assistance. Hydrocortisone was reduced to 10 mg twice daily.
In October, she was again admitted for intermittent episodes of myoclonus (less than 60 min). Myoclonus rapidly disappeared after treatment with IV methylprednisolone (1 g) for three consecutive days. Steroids dose were increased again to 20 mg twice daily. She is now followed monthly at the outpatient clinic and levetiracetam treatment has been progressively stopped while hydrocortisone is maintained at the same dose.

Discussion and conclusions
PERM is a challenging diagnosis due to several reasons. The condition, firstly described by Whitely et al. in 1976, is relatively rare [7]. Diagnosis is mainly clinical, as (in most cases) there is an absence of signs on imaging. Increased signal intensity of the cervical cord and lower brainstem in T2 MRI sequences was reported in only two cases [8,26]. EMG and antibodies in serum and CSF could help to confirm the diagnosis. Initial clinical presentation can be very unspecific and uncommon, and this case is no exception. Common differential diagnoses of PERM syndrome are stiff person syndrome, paraneoplastic SPS and NMS [9,17]. In this case, we can also add spinal myoclonus to the list and potentially scoliosis.
In retrospect, the initial lower back pain and rigidity could have been the premises of the PERM syndrome symptoms. Lumbar spinal fusion was performed in regards to the scoliosis indication. Unfortunately, it had no efficacy on the rigidity symptoms. Moreover, the patient later developed recurrent episodes of lower limb myoclonus. These were, in association with the past history of spinal fusion, wrongly diagnosed as spinal myoclonus. The episode of myoclonus presented by our patient in 2014, associated with hyperthermia, was considered as an episode of NMS. It was previously reported by Xu et al. that PERM could be mimicking NMS due to its significant overlap in symptoms [17]. In this case, muscle rigidity, fever, autonomic nervous system dysfunction and altered mental status, common to both entities, could have been taken in favour of NMS. Nevertheless, generalised myoclonus, especially its history of recurrent pattern, and even more the lack of evidence of neuroleptic use should have alerted the clinician to intensify the diagnostics.
Elevation of CPK in NMS is reported up to at least four times the upper limit. In the context of PERM, elevation of CPK is less frequently described. Literature is not particularly rich on the topic, but we identified three reported cases with normal CPK values [5,21,27], while five described elevated CPK values between 408 IU/l and 4779 IU/l [4,11,17,23,28].
Besides the clinical presentation, antibodies can be determinative in diagnosing this syndrome. Initially both SPS and PERM syndromes were associated with the GAD antibodies. The most recent literature estimates that there is a 60-80% correlation between GAD positive antibodies and SPS [2,10,18,19,23,29,30]. According to Schmidt et al. GAD positive antibodies would only be present in 25% of the PERM cases [19,30]. Discovered in association with PERM cases in 2008, GlyR antibodies are absent in most cases of SPS [31]. Unfortunately, GlyR antibodies testing was not available in our institution at the time.
Presumably, GAD antibodies cause a reduced brain GABA, which is consistent with the occurrence of seizures. A four channel EMG study showed defective inhibition and exaggerated excitation of the brainstem reflexes mediated by dysfunction of the inhibitory GABA and glycine neurons [32]. Interestingly, post-mortem analysis of PERM cases shows inflammation of the brainstem [23].
Brainstem involvement can be used to differentiate SPS from PERM. There is a controversy about PERM being part of the spectrum of SPS and considered as SPS-plus version of the main syndrome pain [33]. Initial clinical presentation does indeed show significant overlap in symptoms. Nevertheless, the fluctuating course of our patient with severe relapses definitely supports the diagnosis of PERM.
The presence of GAD antibodies is not specific to PERM syndrome. It is associated with a number of other conditions such as diabetes mellitus, schizophrenia or bipolar disease, Parkinson disease, cerebellar disorders and neuropathic conditions. Patients with type 1 diabetes mellitus usually show GAD65 antibody titers at only less than 20, while patients with neurological autoimmune disorders have serum antibody values greater than 20, usually even greater than 100 [22]. Level of GAD65 antibodies in our patient was increased to 95. However, its association with the clinical features of our case and its evolution leave little doubt to our final diagnosis.
Treatment recommended for PERM patients includes immunomodulation using corticosteroids, IV immunoglobulins, plasma exchange or cyclophosphamide [1,24]. Intrathecal Bacflofen has also been used successfully [1,2]. The prognosis after treatment can be fluctuating.
Literature is more limited regarding long-term clinical course of PERM. It has been suggested that the syndrome might stay latent during several years and suddenly decompensate by an infection. Up to 18% of PERM cases are reported with chronic course and acute exacerbations [14,17]. Recurrence has been reported from a few weeks after disease onset up to nine years [1,4,8,11,12,31,34,35]. In most cases, relapses are related to corticosteroid discontinuation, as in our case. Often, the dosage of corticosteroids is increased at the time of relapse [8,11,22]. However, when there is no satisfactory response, more aggressive immunosuppressive therapy may be required, such as cyclosporine, azathioprine and rituximab [5,12,14,31,35]. Also other immunotherapies have been described at the time of relapse: plasma exchange and cyclophosphamide [36]. Long-term treatment is described with oral corticosteroids variating from 8 mg/day to 20 mg/day combined with Azathioprine 50 mg/day to 150 mg/day [1,6,24,35]. The combination of oral corticosteroids with Diazepam or Clonazepam and intrathecal Baclofen has also been reported [1,24].
In the current case nearly 10 years went by, after onset of initial symptoms, before the right diagnosis was found. Cases reporting delayed diagnosis PERM are very rare in the literature. One case was identified with a six month delayed diagnosis of PERM, also with good response to immunoglobulin treatment [17]. After literature review we found three case reports with a delayed diagnosis of SPS, with delays ranging from four months up to three years. The first patient responded well to a treatment combination of diazepam, baclofen, dexamethasone, azathioprine and plasma exchange, the patient maintained moderate rigidity without spasms at a follow-up of five months [37]. The second case was treated by immunoglobulins with good response at a follow-up of two months [38]. The third case received an IV treatment of diazepam, no relapse at six months follow-up [39].
In conclusion, PERM syndrome is a rare disease and its diagnosis is not easy, which is underlined by this case report. Once the diagnosis is established, the correct treatment should follow, which could be lifesaving.
Maintenance of long term oral corticotherapy is suggested to prevent relapses. Moreover, it appears that regardless of a delayed diagnosis, a good response may be obtained with adequate treatment in autoimmune central nervous system (CNS) disorders such as PERM. However this report only describes a single case, more data need to be examined in order to generalize these conclusions.