The Guillain–Mollaret triangle comprises the ipsilateral red nucleus in the midbrain, the inferior olive in the medulla and the contralateral dentate nucleus in the cerebellum: together, these form the dentato-rubro-olivary pathway. Pathology in this triangle disinhibits (and so activates) the inferior olivary nucleus. The olivary nucleus then hypertrophies and its rhythmical discharges may manifest clinically as oculopalatal tremor. We describe three cases with either oculopalatal tremor or MRI evidence of olivary hypertrophy caused by vascular insults to this triangle. It is not clear why only some patients have the oculopalatal tremor. Olivary hypertrophy can be confused with demyelination if the imaging is not put into clinical context. Oculopalatal tremor may occur without olivary hypertrophy since the nucleus atrophies with time. Oculopalatal tremor does not respond to medical treatment. A better understanding of the mechanism of the discharge at a cellular level may lead to more targeted medical treatments.
Statistics from Altmetric.com
The Guillain–Mollaret triangle
The Guillain–Mollaret triangle comprises the ipsilateral red nucleus in the midbrain, the inferior olive in the medulla and contralateral dentate nucleus in the cerebellum nucleus; together, these form the dentato-rubro-olivary pathway (figure 1). Damage to these areas or the connections between them can cause oculopalatal tremor. The cerebellum is predominantly excitatory, but the dentate nucleus within it has the primary function of inhibiting the inferior olivary nucleus via GABAergic projections.1 If the triangle is interrupted by pathology (vascular insult, demyelination or tumour), the inferior olivary nucleus is no longer inhibited and it hypertrophies. The nucleus contains many gap junctions that rhythmically discharge at 0.5–3.0 Hz, causing oculopalatal tremor at that frequency. Olivary nucleus hypertrophy can occur without the patient developing an ocular or palatal tremor.
The ocular nystagmus can be vertical, horizontal or torsional. It is described as ‘pendular’, meaning the eye movements are smooth with equal speed in both directions, unlike jerk nystagmus where the deviation phase is slow and the corrective phase is fast. The frequency of the ocular tremor is 1–3 Hz and is in tandem with the palatal tremor.
The palatal tremor is rhythmic and patients may complain of hearing a clicking sound. It does not usually cause difficulty with speech or swallowing. Unlike other movement disorders, palatal tremor continues during sleep, probably because it is brainstem-mediated, rather than from pathology in the basal ganglia.
The initial purpose of imaging is to look for pathology interrupting the Guillain–Mollaret triangle and to look at the inferior olivary nuclei. Imaging also shows the consequences of the lesion in the pathway. Whatever the cause, the inferior olive shows an increased T2 signal change about 1 month after an insult to the Guillain–Mollaret triangle and this can continue for 3–4 years or longer. The olivary hypertrophy develops 6 months after the insult and usually resolves within 3–4 years. After this, the olivary nucleus may still have high T2 signal change and then may atrophy.2
A 56-year-old man developed acute diplopia and dizziness. An MR scan of brain showed an acute infarct in the dorsal medial midbrain. Figure 2 shows DWI hyperintensity with suppression of the apparent diffusion coefficient map, in keeping with an acute infarct. He made a good recovery. However, 4 years later he presented to neuro-ophthalmology feeling unsteady, particularly when walking. He also noticed a rhythmic clicking in his ears. On examination, he had vertical and torsional pendular nystagmus in tandem with palatal myoclonus. We diagnosed oculopalatal tremor.
His MR scan of brain showed cerebral atrophy but no evidence of his old midbrain infarct. There was no olivary hypertrophy. However, the nucleus atrophies over time, and so a normal appearance of the olive in such cases is expected. His oscillopsia has not responded to treatment, and he has tried memantine, baclofen, clonazepam, levetiracetam, 3–4 diaminopyridine, gabapentin and isoniazid without success. He is considering botulinum toxin injections into extraocular muscles to immobilise his eyes. Supplementary videos of the patient's nystagmus and palatal myoclonus are available online.
A 49-year-old woman presented with vomiting and vertigo; her CT scan of head showed a left cerebellar haemorrhage. There was no underlying vascular abnormality. She improved but had a persisting ataxic gait. MR brain imaging at 6 months found an abnormality within her right medulla, initially reported as a low-grade glioma or demyelination. However, putting the medullary lesion in context with her cerebellar damage it became clear that the lesion was her enlarged inferior olivary nucleus (figure 3).
Although she has inferior olivary hypertrophy, she does not have oculopalatal tremor. Olivary hypertrophy occurs weeks to months after an insult in the triangle. She may develop symptoms in the future or her brain may remodel sufficiently that the connection will not be completely lost.
A 59-year-old man developed a sudden right-sided headache and left-sided weakness, with a Glasgow coma scale score of 13/15. His CT scan of head showed a pontine haemorrhage extending into the right red nucleus (figure 4). There was no underlying vascular malformation, and he made good recovery. An MR scan of brain at 6 months found right olivary hypertrophy. He remains asymptomatic (figure 5).
The Guillain–Mollaret triangle seems an eloquent model, where disruption of the pathway leads to structural changes on imaging; occasionally, this can be translated into a specific clinical syndrome. However, it is not clear why the palate and extraocular muscles are affected specifically or why the inferior olivary hypertrophy only sometimes has a clinical expression. Perhaps the nearby nucleus ambiguus—which is involved in palatal innervation—may cause the palatal tremor. The nearby caudal dorsal cap of the inferior olive or pontomedullary tegmentum may cause the ocular tremor due to failure of neural integrators.
The pendular eye movements are smoother than would be expected in a pure inferior olivary disease and cerebellar plasticity may cause the smoothing of eye movements that worsens symptoms as it makes the nystagmus more extreme.
Most patients are not disabled by the palatal tremor, and it can be treated by botulinum toxin injections. The oscillopsia can be very disabling, with patients unable to focus on a single object due to constant eye movements, causing a feeling of unsteadiness and loss of visual acuity. It is notoriously difficult to treat: several medications, mostly antiepileptic drugs, have been tried with varying success but with only small numbers in the studies. Gabapentin has the best evidence, followed by memantine.3
Table 1 shows the variety of drugs and their underlying mechanisms. None of these directly treats the problem of the misfiring gap junctions within the olivary nucleus; perhaps if this could be targeted specifically, treatment would probably be more successful. Antimalarials such as quinine block gap junction connexin 36, and it may be that these drugs are the way forward in treating oculopalatal myoclonus.
Another way to prevent oscillopsia is to stop the extraocular muscles from moving with surgical procedures or botulinum toxin injection. However, this is a crude way to arrest eye movements and leads to the patient having to turn the whole head to look at an object. Unsteadiness may still be a problem, particularly when the patient moves due to loss of vestibulo-ocular reflex.
The dentato-rubro-olivary pathway is an eloquent pathway that normally disinhibits the inferior olivary nucleus.
Oculopalatal tremor can be a delayed manifestation of lesions to the Guillain–Mollaret triangle.
Inferior olivary hypertrophy develops after lesions to the Guillain–Mollaret triangle and can be mistaken for demyelination or tumour on MRI.
Correction notice This paper has been amended since it was published Online First. Figure 3B was missing from the original version. Also, some supplementary videos have been published to go with this paper. The first Key point has been corrected.
Twitter Follow Pushkar Shah at @pushkar_shah
Contributors SM wrote the text and carried out the background research. PS provided the first two cases and his expert opinion regarding oculopalatal tremor and the Guillain–Mollaret triangle. He also proofread the paper. RJ provided the third case, his expert opinion on the imaging and proofread the paper.
Competing interests None declared.
Patient consent Obtained.
Provenance and peer review Commissioned; externally peer reviewed. This paper was reviewed by Sui Wong, Liverpool, UK.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.