Histopathological Investigation of Progressive Encephalomyelitis with Rigidity and Myoclonus: An Autopsy Case Characterized by Oculomotor Dysfunction and Autonomic Failure

Abstract

Progressive encephalomyelitis with rigidity and myoclonus (PERM) is a rare disease associated with the presence of anti-glycine receptor (GlyR) antibodies. We herein report an autopsy case of an 80-year-old man diagnosed with anti-GlyR antibody-positive PERM who presented with symptoms of oculomotor dysfunction and autonomic failure. Despite intensive immunotherapy, the neurological symptoms showed almost no improvement, and the patient succumbed to aspiration pneumonia and bacterial translocation. Postmortem pathology revealed mild inflammatory changes and neuronal loss that were disproportionate to a severe clinical presentation. These results suggest that the clinical symptoms of PERM may result from antibody-mediated GlyR internalization, leading to neuronal disinhibition, rather than a neuroinflammatory signature.

Introduction

Progressive encephalomyelitis with rigidity and myoclonus (PERM) is a rare disease characterized by bulbar palsy, ophthalmoplegia, hyperekplexia, rigidity, myoclonus, and autonomic failure. This syndrome was first described in 1976 by Whiteley et al (1). In 2008, the anti-glycine receptor (GlyR) antibody was discovered for the first time in a case of PERM. It was presumed to disrupt glycinergic inhibition by binding to GlyRs (2). Further research revealed that GlyRs are highly and abundantly expressed along the membranes of neuronal somata and dendrites in various regions, including the basal ganglia, substantia nigra, pontine regions, rostral medulla oblongata, and cervical spinal cord in humans (3). Carvajal-González et al. investigated the clinical spectrum of GlyR antibody-positive patients and found that approximately 70% of them had PERM, whereas the remaining 30% had diseases such as stiff person syndrome, limbic encephalitis, epileptic encephalopathy, brainstem features, and demyelinating optic neuropathy (4). Crisp et al. elucidated the pathogenic role of the anti-GlyR antibody and correlated it with clinical manifestations of the disease in 2019 (5).

In 1998, Barker et al. presented an autopsy case of PERM in which the presence of an anti-GlyR antibody was not examined. The primary pathological findings in this case included perivascular cuffing, microglial nodules, astrogliosis, and neuronal loss, predominantly in the limbic area and brainstem (6). Following the recognition of anti-GlyR antibodies as a cause of PERM, only two anti-GlyR antibody-positive autopsy cases of PERM have been reported. However, their clinical courses and pathological findings showed significant differences (4,7,8). The therapeutic response and prognosis of PERM are thus controversial (4,9,10), requiring further autopsy studies to gain more insight and clarify the pathogenesis of PERM.

We herein report an autopsy case of PERM that was positive for anti-GlyR antibodies, characterized by oculomotor dysfunction and autonomic failure. We focused on the discrepancy between clinical symptoms and pathological findings.

Case Report

The patient was a Japanese man who had been diagnosed with atrial fibrillation and chronic heart failure at 79 years old. The treatment regimen included atenolol, apixaban, and furosemide. The patient had no family history of neurological disorders. At 80 years old, he developed dysarthria abruptly, followed by dysphagia 6 days later. He was admitted to another hospital eight days after the onset of symptoms. On admission, a neurological examination revealed spastic paraplegia, ophthalmoplegia, and bulbar palsy but no sensory disturbance or ataxia. Fifteen days later, he was diagnosed with severe aspiration pneumonia necessitating ampicillin/sulbactam and mechanical ventilation. A cerebrospinal fluid (CSF) examination revealed normal cell counts and protein levels, and magnetic resonance imaging (MRI) of the cerebrum, brainstem, cerebellum, or spinal cord revealed no significant abnormalities (Fig. 1A-D). Suspecting neurological autoimmune disorders, such as Bickerstaff's brainstem encephalitis or Guillain-Barré syndrome, he received intravenous immunoglobulin (IVIg) therapy (5,000 mg/day, for 3 days) and intravenous methylprednisolone (IVMP) therapy (500 mg/day, for 3 days). With no improvement observed, the patient was transferred to our hospital 23 days after the symptom onset.

Figure 1.

Magnetic resonance imaging (MRI) performed nine days after the onset. MRI revealed no remarkable changes in the cerebrum, brainstem, cerebellum, or spinal cord.

On admission, the patient was sedated with propofol. A neurological examination revealed intact pupillary reflexes, bilateral ptosis, right-gaze nystagmus, and bilateral ophthalmoplegia with preserved left eye adduction. The right eye was abducted to the primary position. Oculocephalic reflexes were absent, but eyelash reflexes were exaggerated, with hyperekplexia of the chin and neck. He exhibited spastic quadriplegia, exaggerated biceps and patellar tendon reflexes, bilateral positive Babinski signs, and right positive Chaddock's sign. He showed pronounced rigidity in the neck and upper extremities and milder rigidity in the lower extremities. In addition, myoclonus in the right upper extremity, abdominal distension, constipation, and facial hyperhidrosis were observed. The sensory functions, including touch and cold temperature perception, were intact. Ataxia and the urinary function (because of catheterization) could not be assessed. Blood tests revealed anemia (Hb 9.6 g/dL), elevated white blood cell count (10,000/mm³), procoagulant state (PT-INR 1.24, D-dimer 4.5 μg/mL), mild renal dysfunction (Cre 1.05 mg/dL), and inflammation (CRP 3.19 mg/dL). Tests for various autoantibodies, including anti-nuclear antibody, anti-double-stranded DNA antibody, anti-Sjögren's syndrome (SS)-A antibody, anti-SS-B antibody, proteinase 3 anti-neutrophil cytoplasmic antibody (ANCA), myeloperoxidase ANCA, anti-glutamic acid decarboxylase antibody, anti-acetylcholine receptor antibody, and anti-GQ1b antibody, were all negative. A CSF examination indicated a normal cell count (5 /μL, all lymphocytes), slightly elevated protein levels (42 mg/dL), and normal glucose levels (94 mg/dL, blood glucose level: 159 mg/dL). However, the IgG index was high (0.82) and oligoclonal bands were detected. Myelin basic protein levels were within the normal range (33.7 pg/mL). Based on these findings and the symptoms of rigidity, myoclonus, brainstem dysfunction, autonomic failure, and hyperekplexia, a clinical diagnosis of PERM was made.

The patient was treated with IVIg (25,000 mg/day, for 5 days) and IVMP (1,000 mg/day, for 3 days), followed by oral prednisolone (50 mg/day). Gradual improvements were observed in the abnormal primary eye position of the right eye, horizontal ophthalmoplegia to the left eye, and bilateral ptosis. His critical condition and concurrent infections prevented plasma exchange. The hyperekplexia, spastic quadriplegia, marked rigidity, myoclonus, bulbar palsy, and severe autonomic failure persisted. He experienced repeated septic shock due to aspiration pneumonia and bacterial translocation from the intestinal tract, leading to death two months after symptom onset. Autoantibodies against neuronal surface antigens, including GlyR, were examined at the laboratory of Josep Dalmau (IDIBAPS Hospital Clínic, Barcelona, Spain) with established assays, such as rat brain immunohistochemistry or a cell-based assay when appropriate, and were reportedly negative for NMDAR, AMPRA, GABAaR, GABAbR, or dipeptidyl peptidase-like protein 6, but the patient's serum and CSF were both strongly positive for anti-GlyR antibodies, confirming the diagnosis of anti-GlyR antibody-positive PERM. In addition, autoantibodies against classical paraneoplastic intracellular antigens (CV2/CRMP5, Ma2, Ri, Yo, Hu, amphiphysin) and glutamic acid decarboxylase 65 were measured in serum with immunoblotting (Euroimmun AG) and were negative for all examined antigens.

Pathological findings

Written informed consent to perform an autopsy was obtained from the patient's family. A systemic organ evaluation revealed intestinal pneumatosis concurrent with intestinal lymphangiectasia and evidence of aspiration pneumonia. The brain weighed 1,340 g and showed mild bilateral swelling in the medial temporal lobes. A gross inspection revealed no significant changes in the cerebrum, cerebellum, brainstem, or spinal cord and no notable meningeal opacification or cerebrovascular arteriosclerosis. The brain and spinal cord were cut and trimmed using established methods (11). Coronal sections of the cerebrum, axial sections of the brainstem and right cerebellum, and sagittal sections of the left cerebellum showed no discoloration or softening of lesions. Axial sections of the spinal cord revealed no obvious abnormalities.

The spinal cord, brainstem, cerebellum, and left hemisphere of the cerebrum were stained with Hematoxylin and Eosin (H&E) and Klüver-Barrera (KB) staining. In the entorhinal cortex, slight gliosis and a few hypertrophic astrocytes were identified (Fig. 2A). The hippocampus exhibited pyramidal cell loss in the CA1 region and subiculum. Notably, no neuronal loss or gliosis was detected in other regions of the cerebral cortex, basal ganglia, or hypothalamus (Fig. 2B). Purkinje cells and neurons of the dentate nucleus in the cerebellum were atrophic with eosinophilic cytoplasm. Mild perivascular cuffing was evident in the frontal lobe, temporal lobe, and amygdala (Fig. 2C). Subtle microglial nodules were evident in the frontal lobe, temporal lobe (Fig. 2D), parietal lobe, caudate nucleus, putamen, external capsule, thalamus, entorhinal cortex, and cerebellar cortex.

Figure 2.

Neuropathological findings associated with encephalopathy. (A) A few hypertrophic astrocytes (arrows) and slight gliosis were observed in the entorhinal cortex. (B) Neuronal loss or gliosis was not observed in the hypothalamus. (C) Mild perivascular cuffing was observed in the amygdala. (D) Mild microglial nodules were observed in the occipitotemporal gyrus. (E, F) Microglial nodules were partially immunopositive for anti-CD163 antibody in amygdala (E) and anti-Iba1 antibody in occipitotemporal gyrus (F). (G, H) Neuronal loss or gliosis was not observed in the left side of the oculomotor nucleus (G) or trochlear nucleus (H). Hematoxylin and Eosin staining: (A-D, G, H), Immunohistochemical staining: (E, F). Scale bars: 40 μm (A, C-F), 100 μm (B, G, H).

A more detailed assessment was performed using immunohistochemical staining with anti-CD163 (mouse monoclonal, clone 10D6; Leica, Newcastle Upon Tyne, UK; 1:400, Fig. 2E), anti-Iba1 (rabbit polyclonal, Cat. No. 016-20001; Wako, Osaka, Japan; 1:4,000, Fig. 2F), and anti-leukocyte common antigen (unconjugated monoclonal, clone X1699; Leica, Newcastle Upon Tyne, UK; 1:100) antibodies revealed an increased presence of microglial nodules compared with HE staining. The Edinger-Westphal nucleus, medial longitudinal fasciculus, substantia nigra, locus coeruleus, oculomotor nucleus (Fig. 2G), trochlear nucleus (Fig. 2H), trigeminal motor nucleus, dorsal nucleus of vagus, and hypoglossal nucleus did not display neuronal loss or gliosis. Myelin sheaths in the oculomotor, abducens, and facial nerves were well-preserved by KB staining. The right oculomotor and bilateral abducens nuclei were not examined. Finally, the spinal cord showed no discernible abnormalities, including neuronal loss, gliosis, perivascular cuffing, or microglial nodules.

Lewy body pathology was also observed in our case. In the substantia nigra and locus coeruleus (Fig. 3A), neurons were well preserved, as demonstrated by KB staining. However, the dorsal nucleus of the vagus exhibited the presence of Lewy bodies upon HE staining (Fig. 3B). An immunohistochemical analysis performed using an anti-α-synuclein (rabbit polyclonal, Cat. No. S 3062; Sigma Aldrich, St. Louis, USA; dilution 1:20,000) revealed the presence of Lewy bodies and Lewy neurites in the locus coeruleus (Fig. 3C), dorsal nucleus of the vagus (Fig. 3D), substantia nigra, and nucleus basalis of Meynert. In contrast, Auerbach's plexus in the gut, autonomic nerves surrounding the abdominal aorta, and sympathetic nerves adjacent to the anterior wall of the left ventricle (Fig. 3E, F) showed an absence of Lewy neurites. An immunohistochemical analysis using anti-neurofilament (mouse monoclonal, clone SMI-31; Covance, Berkeley, USA; dilution 1:5,000, Fig. 3G) and anti-tyrosine hydroxylase (mouse monoclonal, clone TH-16; Sigma Aldrich; dilution 1:3,000, Fig. 3H) antibodies confirmed the preservation of the sympathetic nerves around the anterior wall of the left ventricle. The observed distribution of Lewy pathology was consistent with Braak stage III for alpha-synuclein (12) and was classified as the brainstem-predominant type (13).

Figure 3.

Neuropathological findings associated with Lewy body pathology. (A-C) The absence of neuronal loss (A), the presence of Lewy bodies (B), and anti-α-synuclein antibody-positive structures (C) were observed in the locus coeruleus. (D) Anti-α-synuclein antibody-positive structures were observed in the dorsal nucleus of the vagus. (E-H) The sympathetic nerve around the anterior wall of the left ventricle showed no Lewy neurites using Hematoxylin and Eosin (H&E) staining (E) or immunohistochemical staining with anti-α-synuclein antibody (F) Immunoreactivity for neurofilament (G) and tyrosine hydroxylase (H) was well preserved. Klüver-Barrera staining: (A) H&E staining: (B, E) Immunohistochemical staining: (C, D, F-H). Scale bars: 200 μm (A), 20 μm (B-D), 40 μm (E-H).

Furthermore, mild tau and amyloid-β aggregation was evident in the cerebrum. These pathological findings correspond to Braak stage II for neurofibrillary tangles (14) and Thal phase 2 for amyloid-β (15).

Discussion

The present case had typical clinical symptoms of PERM, but the pathological findings were mild compared with the patient's severe clinical symptoms. Neuronal loss and gliosis, which are considered irreversible changes, were either absent or mild. Perivascular cuffing and microglial nodules, which might be considered reversible inflammatory changes, were observed but were mild. The distribution of these pathological changes was more limited than expected based on the neurological symptoms remaining at death. It is suggested that the reversible inflammatory changes were partially ameliorated by repeated immunotherapies, and the symptoms remaining at death were mainly caused by functional impairment that could not be detected by a morphological examination.

Two autopsy cases of PERM that were positive for anti-GlyR antibodies have been reported. Turner et al. (2011) reported the first case. The patient's neurological symptoms rapidly progressed and caused death within two months, despite receiving no immunotherapy. A pathological examination revealed distinct encephalomyelitis, including neuronal loss (4,7). Ozaki et al. (2018) reported the second case. The patient was diagnosed with thymoma complication. The neurological symptoms improved markedly after immunotherapy and thymectomy, and the patient died of causes other than encephalitis. Mild CD3-positive mononuclear cell infiltration around the vessels in the brainstem is the only pathological finding suggestive of encephalitis (8). In our case, the pathological findings were similar to those in the second case, except for the absence of thymoma. However, despite immunotherapy, neurological symptoms showed almost no improvement (Table). Therefore, we hypothesized that, by focusing on the discrepancy between the clinical symptoms and pathological findings in our case, we could elucidate the pathogenesis of PERM.

Table.

Summary of the Anti-GlyR Antibody-positive Autopsy Cases of PERM.

	(7)	(8)	Our case
Age at disease onset	28 years old	75 years old	80 years old
Sex	Male	Female	Male
Disease duration	2 months	Not described	2 months
Rigidity	+	+	+
Myoclonus	+	+	+
Gaze nystagmus	-	+	+
Opthalmoplegia	-	-	+
Constipation	+	+	+
Therapy	Anti-epileptic drug	PE, thymectomy	Steroid, immunoglobulin
Response to therapy	No improvement	Marked improvement	Almost no improvement
Brain MRI	No abnormality	No abnormality	No abnormality
Anti-GlyR antibody	Serum (1:1,280)	Serum (1:350)	Serum (strongly positive)
		CSF (1:40)	CSF (strongly positive)
Anti-NMDAR antibody	+	Not described	-
Pathological findings of tumor	-	Thymoma	-
Brain swelling	Mesial temporal	-	Medial temporal
	hypothalamus
Perivascular cuffing	+	-	Mild
Microglial nodules	+	-	Mild
Neuronal loss	+	-	Quite Mild
Gliosis	Not described	-	Slight

GlyR: glycine receptor, PERM: progressive encephalomyelitis with rigidity and myoclonus, PE: plasma exchange, MRI: magnetic resonance imaging, CSF: cerebrospinal fluid

The patient had marked oculomotor dysfunction. In particular, the right eye was abducted in the primary position and bilateral ophthalmoplegia was observed, except for adduction of the left eye. This presentation suggests impairment of the bilateral oculomotor and trochlear nuclei as well as the left abducens nucleus or its associated peripheral nerves. Intact adduction of the left eye suggests that the medial longitudinal fasciculus communicating with the left oculomotor nucleus, located caudally to the nucleus, remained intact. Iizuka discussed such multifocal regions in PERM in detail and summarized the anti-GlyR antibody-positive conditions. The PERM case reported by Iizuka showed impairments such as horizontal gaze palsy, gaze-evoked nystagmus, slow saccades, and ocular lateropulsion symptoms, comparable to those observed in our case (16). Another PERM case documented by Peeters et al., which presented with supranuclear gaze palsy, further underscores the potential disturbances in supranuclear inhibition associated with oculomotor dysfunction in PERM (17). However, the absence of an oculocephalic reflex in both our case and that of Iizuka suggests the involvement of nuclear and perinuclear lesions in the oculomotor dysfunction inherent to PERM. The anti-GlyR antibody recognizes the surface of homodimeric GlyRs, presumably leading to receptor internalization, which induces neuronal excitation (4). The absence of significant inflammation on brain MRI also contradicts the active role of inflammation in the pathogenesis of PERM (16). The absence of pathological changes in the supranuclear, nuclear, and infranuclear regions of the oculomotor system in our patient confirms this and suggests that oculomotor dysfunction results from the disinhibition of neuronal excitation due to the internalization of GlyR.

It should be noted that our patient also presented with prominent autonomic failure, including abdominal distension, constipation, and hyperhidrosis, which persisted until death. Immunohistochemical investigation revealed α-synuclein-positive structures in the locus coeruleus, dorsal nucleus of the vagus, substantia nigra, and nucleus basalis of Meynert. Although the role of Lewy body pathology may underlie autonomic failure in various contexts, our investigation argues against the involvement of Lewy body pathology in the patient's autonomic failure because there was no neuronal loss in the dorsal nucleus of the vagus or α-synuclein-positive structures in the peripheral nervous system, such as Auerbach's plexus of the gut, celiac ganglion, and sympathetic nerves around the anterior wall of the left ventricle. Barker et al. reported an autopsy case of PERM in which the anti-GlyR antibody was not examined (6). A pathological examination was performed in various regions associated with autonomic failure, such as the hypothalamus, locus coeruleus, dorsal nucleus of the vagus, intermediolateral nucleus of the spinal cord, and autonomic ganglia. They reported that encephalitic processes, such as neuronal loss and various inflammatory changes, were observed in these regions. Their report suggested that both the central and peripheral autonomic nervous systems may be affected by PERM. In contrast, we did not observe neuronal loss or inflammatory changes in such regions, suggesting that the autonomic failure in our case also resulted from neuronal disinhibition due to insufficient glycinergic neurotransmission.

PERM is considered a fatal disease, with patients typically surviving less than three years due to the lack of effective therapies (9). However, after the correlation between PERM and anti-GlyR antibodies was reported, several studies suggested the efficacy of immunotherapy for PERM (4,10). Unfortunately, despite intensive immunotherapy, the neurological symptoms in our patient showed almost no improvement. This resistance to immunotherapy may be attributed to high levels of anti-GlyR antibodies in the serum and CSF. In contrast, the pathological findings in our case, such as the absence of neuronal loss and mild gliosis, are consistent with the treatable nature of PERM. This suggests potential improvements in PERM outcomes with sustained multidisciplinary immunotherapy in the absence of concurrent severe systemic conditions.

Several limitations associated with the present study warrant mention. First, this research is based on a single autopsied case, which limits its generalizability. Second, comorbidities such as septic shock and bacterial translocation may affect both clinical and pathological interpretations. Third, owing to technical constraints, we were unable to perform a pathological study of glycine receptor distribution. Nevertheless, this case report adds to the understanding of clinicopathological correlations in PERM. Further research is imperative to elucidate the pathomechanism and enhance therapeutic interventions for this severe but treatable rare disease.

The authors state that they have no Conflict of Interest (COI).

Financial Support

This research received support from AMED under grant numbers JP18dm0107105 and JP16kk0205009, which were both awarded to M. Yoshida. Further support came from Grants-in-Aid provided by the Research Committee of CNS Degenerative Diseases and Research on Policy Planning and Evaluation for Rare and Intractable Diseases. These were made possible by Health, Labour and Welfare Sciences Research Grants from the Ministry of Health, Labour and Welfare, Japan, designated as Y. Iwasaki.