Creutzfeldt-Jakob disease (CJD) is a progressive spongiform encephalopathy that causes dementia, insomnia, and focal neurological signs such as ataxia, visual loss, and motor weakness.1, 2 Generalized or partial seizures are less common, but also occur in some CJD patients.1, 3 One of the characteristic symptoms of CJD is myoclonus involving the extremities, trunk, head, and face, which occurs in association with periodic sharp wave complexes (PSWCs) or periodic synchronous discharges on the electroencephalogram (EEG).4 However, rhythmic oscillation of the pupils and eyelids in correspondence with PSWCs have never been reported before in this disease. Here we report two CJD patients with rhythmic pupillary and palpebral oscillation who both had the same mutation of prion protein (PrP) codon 200.
A 68-year-old woman developed tinnitus and deterioration of memory. Two weeks later, she presented to her physician with disorientation, unsteadiness, poor concentration, drowsiness, and a startle reaction. Three weeks after the onset, she developed generalized seizures and was admitted to a local hospital. Phenytoin (500 mg/day) was administered intravenously. Four weeks later, her generalized seizures, startle reaction, and drowsiness had progressed, so she was transferred to our hospital. She had never received a blood transfusion or growth hormone therapy, and had no history of surgical treatment. She had never been overseas. Her older sister had died after progression of dementia and ataxia over 18 months.
On admission, she was disoriented and drowsy. General examination showed nothing noteworthy. The startle reaction and recurrent partial or generalized seizure persisted despite an increase in the dose of phenytoin to 1,000 mg/day. Both pupils were round and isocoric, and the light reflex was normal. Eye movements were also normal without nystagmus. The deep tendon reflexes were normal and pathologic reflexes were absent in all four limbs. However, she had myoclonus of the head, face, and limbs. Coordination was disturbed in all four limbs, but was worse on the left side. There were no signs of autonomic dysfunction, such as abnormalities of blood pressure, temperature, or respiration. Laboratory tests revealed no abnormalities. Cerebrospinal fluid examination showed a normal cell count and a normal total protein concentration, but an increase of neuron-specific enolase (NSE) 66.97 ng/mL (normal range: 8.6 ± 2.9 ng/mL) was observed. There was also an increase of tau (11,900 pg/mL) and she was positive for 14-3-3 protein. Genetic analysis of PrP revealed a mutation of codon 200 that resulted in substitution of lysine for glutamate (Glu/Lys). In addition, she was homozygous for methionine at codon 129 and for glutamine at codon 219 of the PrP gene (polymorphism at codon 129 is considered to influence susceptibility and is correlated with the clinicopathological phenotype of the disease, while polymorphism at codon 219 is also known to alter susceptibility). The EEG demonstrated PSWCs (1 c/s) (see Video: Case 1–Segment 1), as well as spike bursts that mainly occurred over the right hemisphere. Brain magnetic resonance imaging (MRI) showed abnormal high-intensity areas in the bilateral, fronto-parietal cortex (especially on the right side), thalamus, and posterior limb of the internal capsule on diffusion-weighted images (DWI) (Fig. 1A). [(18)F]2-fluoro-2-deoxy-d-glucose-positron emission tomography revealed an increase of metabolism in the right fronto-parietal cortex corresponding with the MRI abnormality, as well as in the left cerebellar hemisphere.
Eight weeks after the onset, her symptoms showed marked progression. At that time, rhythmic oscillation of the pupils (alternating dilation and constriction), and upper eyelids (elevation and descent) were observed bilaterally (see Video: Case 1–Segments 2 and 3). Observation of the patient together with the EEG revealed that rhythmic pupillary dilation and palpebral elevation corresponded with the PSWCs; i.e., the sound of the EEG pen moving to record PSWCs coincided exactly with the involuntary movements (Fig. 2A,B). Involuntary oscillation of the pupils continued for 1 week and involuntary eyelid movements persisted until death. During this period, the light reflex was preserved and there were no abnormal eye movements or nystagmus. Tc-99m ethyl cysteinate dimer-single photon emission computed tomography (ECD-SPECT) revealed hypoperfusion of the right, fronto-parietal cortex. She died of cachexia at 20 weeks after the onset. Recurrent partial or generalized seizures persisted until the terminal stage.
A 66-year-old man developed myoclonus of the left upper extremity, followed by insomnia, ataxia of gait and speech, and hallucinations along with an increase of myoclonus. Subsequently, he showed a left homonymous hemianopsia, memory loss, and disorientation. He had no family history of any neurological disorders, including CJD.
Two months after the onset, he was admitted to our hospital. General examination showed nothing noteworthy. He developed the startle reaction, and spontaneous fluctuation in the size of both pupils was observed in synchronization with PSWCs, as occurred in Case 1 (see Video: Case 2–Segment 2). His upper eyelids also showed spontaneous elevation and descent, which coincided with dilation and constriction of the pupils and with myoclonus of the head and limbs (see Video: Case 2–Segment 4). In the dark, mydriasis was noted and pupillary oscillation was absent, but the palpebral movements persisted (see Video: Case 2-Segment 3). The pupils were round and isocoric with a normal light reflex. Deep tendon reflexes were normal and pathologic reflexes were absent in all the four limbs. There was no evidence of autonomic dysfunction, such as changes of blood pressure, temperature, or respiration, and laboratory tests revealed no abnormalities. Cerebrospinal fluid examination showed a normal cell count and a normal total protein level, but there was an increase of NSE (51 ng/mL) and tau (>1,300 pg/mL), and 14-3-3 protein was positive. Genetic analysis of PrP revealed the Glu/Lys mutation at codon 200. Brain MRI with DWI showed abnormal, high-intensity areas in the right temporo-parietal cortex and the bilateral striatum (predominantly on the right) (Fig. 1B). ECD-SPECT revealed hypoperfusion of the right temporo-parietal cortex that corresponded with the MRI abnormalities. The EEG demonstrated PSWCs (0.5–1 c/s) (see Video: Case 2–Segment 1) that were predominantly detected over the right hemisphere.
Four months after the onset, the patient's symptoms showed marked progression to akinetic mutism accompanied by diminished myoclonus of the face and four limbs (predominantly on the left side). At that time, rhythmic pupillary and palpebral oscillations were still detected. He died of cachexia at 22 weeks after the onset.
In our two patients, CJD was diagnosed from their symptoms, the clinical course, and the detection of a mutation of PrP. In both patients, rhythmic pupillary and palpebral oscillation were observed in relation to PSWCs and myoclonus affecting other parts of the body. Such rhythmic oscillation of the pupils and eyelids have never been reported before in patients with CJD.
The sphincter muscle of the pupil is controlled by parasympathetic nerves that run with the oculomotor nerve, and the dilator muscle is regulated by the sympathetic system. In our patients, the light reflex was intact and pupil function was normal, even while pupillary oscillation was present. Accordingly, both the sympathetic and parasympathetic nerves to the pupils were intact, as were the Edinger-Westphal nuclei and the efferent peripheral pathway. In both of our patients, dilation of the pupils corresponded with the PSWCs. Case 2 showed no pupillary oscillation with mydriasis when examined in the dark, but eyelid movements continued in correspondence with the PSWCs. If pupillary oscillation was triggered by fluctuations of parasympathetic activity that led to rhythmic pupillary constriction, then oscillation would still be observed along with physiological dilation of the pupils in the dark. Accordingly, it seems more likely that pupillary oscillation was triggered by fluctuations of sympathetic activity that led to rhythmic pupillary dilation. Synchronization of pupillary dilation and upper eyelid elevation, which are regulated by the sympathetic nervous system, also suggests the predominant role of sympathetic activity. Both sympathetic efferent conduction and contraction of the smooth muscles of the pupils and eyelids are slower than conduction via alpha axons and skeletal muscle contraction, respectively. Therefore, the rhythmic pupillary and palpebral oscillation seen in these patients should not be referred to as myoclonus, although the timing of the oscillations was in accordance with the occurrence of myoclonus elsewhere in the body. It is possible, however, that both pupillary/palpebral oscillation and myoclonus were generated at the same site in the brain.
It was reported5 that pupillary dilation occurred after transcranial magnetic stimulation (TMS) of the frontal, central, and parieto-occipital regions of each hemisphere, but stimulation of the right central region led to greater dilation than stimulation of the left side. In both of our patients, the right cerebral hemisphere (including the central region) was predominantly involved on MRI. Therefore, pupillary dilation corresponding with PSWCs, which could act like TMS, may have originated from right hemispheric abnormalities in our patients. The authors of the TMS study5 concluded that pupillary dilation after TMS appeared to mainly reflect nonspecific activation of the sympathetic nervous system. In our patients, pupillary dilation may also have occurred in response to sympathetic activation. On the other hand, the myoclonus observed in CJD patients is periodic myoclonus, which is considered to have a subcortical origin.6 Moreover, PSWCs observed in a CJD patient, who showed photo-stimulated giant spikes that suppressed PSWCs on the EEG and loss of the pupillary light reflex during the course of the illness, were reported to originate from the thalamus, lateral geniculate body, or brain stem reticular formation.7 Whether rhythmic pupillary oscillation is a direct response to a subcortical generator or an indirect response mediated via the cortex and accompanied by PSWCs is unclear, but such pupillary oscillation may at least occur by the latter mechanism.
Pupillary hippus is the spontaneous bilateral synchronous rhythmic constriction and dilation of the pupils, which is observed in normal individuals and in various pathological conditions.8 Analysis of the pupillogram shows that low-frequency changes of the pupils reflect the hippus observed in most alert subjects.9 It was reported that pupillary hippus and the basic EEG rhythm showed the same frequency in an unconscious patient with epileptic seizures, chronic alcoholism with hepatopathy, and primidon intoxication.10 The rhythmic pupillary oscillation observed in our cases seems to be similar to pupillary hippus.
Pupillary dysfunction is observed in various generalized disorders. During the course of tonic–clonic seizures, pupillary, constriction in the tonic phase, and dilation in the clonic phase is known to occur.11 Although pupillary abnormalities have also been reported in association with other types of seizure, none of the patients showed rhythmic oscillation.11 In Case 1, rhythmic pupillary oscillation was observed in the interictal period and the frequency was about 1 c/s, coinciding with the PSWCs. Thus, pupillary oscillations have no correlation with seizures in this patient. Cheyne-Stokes respiration is characterized by alternating hyperpnea and apnea. In patients with this condition, the pupils show dilation during the hyperpneic phase and constriction during the apneic phase.12 In our patients, snoring was frequent and there were also episodes of choking, but Cheyne-Stokes respiration was not present. Moreover, the episodes of choking did not occur in correspondence with pupillary changes, so there was no relation between pupillary oscillation and respiration.
The clinical course of familial CJD with the Glu/Lys mutation of codon 200 resembles that of sporadic CJD.13 In both of our patients, the disease followed a rapidly progressive course with seizures and the startle reaction, unlike other types of familial CJD. Therefore, rhythmic pupillary oscillation may be observed in rapidly progressive CJD, whether familial or non-familial, and might be related to severe right cerebral hemispheric impairment. Moreover, the abnormalities in the posterior limb of the internal capsule on DWI in Case 1 were atypical for CJD. These changes may have reflected the effects of pyramidal tract over-excitation secondary to cortical lesions that caused frequent startle responses, seizures, and myoclonus, all of which persisted until death. That is, these involuntary movements were more severe and persisted over a longer period than is usual for CJD, although a startle response and myoclonus are common in this disease. Accordingly, the MRI findings may also indicate the severity of our cases.
Legends to the Video
Segment 1. The EEG demonstrates PSWCs (1 c/s).
Segment 2. Case 1 displays rhythmic dilation and constriction of the pupil. The onset of pupillary dilation seems to correspond with the PSWCs, i.e., the sound of the EEG pen moving to record PSWCs coincides with the pupillary changes.
Segment 3. Rhythmic dilation and constriction of the bilateral pupils coincides with rhythmic elevation and descent of the bilateral upper eyelids.
Segment 1. The EEG demonstrates PSWCs (1–0.5 c/s).
Segment 2. Case 2 also had bilateral rhythmic pupillary and palpebral oscillation in synchronization with PSWCs when wearing Frenzel glasses under bright light, as seen in Case 1.
Segment 3. In the dark, mydriasis is noted and pupillary oscillation is absent, but the movement of the eyelids persists. (The video was recorded with an infrared camera, so some objects are bright despite the darkened room.)
Segment 4. Palpebral elevation seems to occur in correspondence with myoclonus of the head and extremities.