Timothy C. Hain, MD Page last modified: November 18, 2018
Torsion is movement of the eye about its visual axis (note that we do not define it as about the front-back axis of the head). The term "rotatory nystagmus" is used interchangeably. There are many videos of torsional nystagmus on other pages on this site.
Torsional eye movements may be jerk or pendular.
When jerk, there are several methods of designating the direction that it jerks. We favor the unambiguous method of designating it by the direction that the top of the eye jumps - -left or right. Others use "clockwise" or "counterclockwise", but the problem here is deciding whether the clock is to be considered on top of the patient's eye, or an external reference on the wall. We suggest just use the left/right nomenclature.
The most useful technique for seeing primary positional torsional nystagmus is fundoscopy. After that, video-frenzel goggles.
|1 hz||CTT lesions such as ocular palatal myoclonus, MS|
|Gaze dependent||Congenital nystagmus|
|3+ hz||Retinal lesions such as albinism and rod/cone dystrophies.|
|Pulse synchronous (1 hz)||Dehiscences such as superior canal dehiscence|
Torsional nystagmus in primary position (the eyes centered, person sitting upright) is very rare, and the few reports about it generally lump it in with other types of nystagmus (e.g. Lopez et al, 1995). Because primary positional torsional nystagmus is so rare, it is very specific, and often worth seeing a subspecialist (e.g. otoneurologist or neuro-ophthalmologist, or both). Generally, even subspecialists will have seen very few of these cases.
It is seen in disorders of the medulla such as syringomyelia, in degenerative disorders of the nervous system, in persons with palatal myoclonus, in multiple sclerosis, in superior canal dehiscence syndrome, and in persons with midbrain lesions (Helmchen et al, 2002). Most of these are probably due to lesions in the central tegmental tract.
See-saw nystagmus is a conjugate pendular torsional oscillation with a superimposed disjunctive vertical movement. The intorting eye rises and the opposite extorting eye falls. Most patients with see-saw have bitemporal hemianopia consequent to large parasellar tumors expanding within the 3rd ventricle. See-saw is exceedingly rare.
The author has seen several cases of torsional pendular nystagmus in the context of atrial fibrillation and treatment with antiarrythmic medications such as amiodarone. Sometimes this is accompanied by palatal myoclonus, and it seems likely that this is a subspecies of pendular nystagmus associated with lesions of the central tegmental tract.
The author has also seen a case of a rapid (i.e. 3 hz) pure pendular torsional nystagmus. This seems most likely to be due to retinal disease. (Perez-Carpinell et all, 1992; Pieh et al, 2008). Perhaps caused by an attempt to reduce retinal fading ?
The author has also seen a case of a slow pendular torsional nystagmus, present congenitally, without significiant visual loss. This is likely a variant of CN.
Torsional nystagmus also occurs (rarely) in superior canal dehiscence syndrome, when it may be pulse synchronous (Hain and Cherchi, 2008). This nystagmus is due to pulsations in spinal fluid pressure that directly affect the cupula of the superior semicircular canal. Most cases of SCD have no such nystagmus.
Jerk Torsional nystagmus is much more common, especially when it is positional. Jerk torsion is commonly elicited by positional maneuvers such as the Dix-Hallpike test. In that context, the nystagmus is transient, and not continuous. Torsional nystagmus in this context is generally attributed to benign paroxysmal positional vertigo (BPPV).
Jerk torsion also occurs in Migraine, in brainstem lesions that affect the pathways for the vertical semicircular canals, and occasionally in lesions of the cerebellar nodulus.
Torsional spontaneous nystagmus (Jerk type) is usually associated with lesions of either the medulla or the cerebellar peduncles (Leigh and Zee, 1995, 2015).
Medullary torsional nystagmus
The lateral medullary syndrome is reported to cause torsional nystagmus(Jeong et al., 2017, Morrow and Sharpe, 1988). Lopez et al (Lopez et al., 1992) reported 35 cases of torsional nystagmus, and reported MRI lesions often occurred in the area of the contralateral vestibular nucleus. Thus in theory, medullary lesion associated torsional jerk nystagmus should be a central vestibular nystagmus. While this is said to be the most common type, we have not seen this often in our otoneurology clinic setting.
Midbrain jerk torsion nystagmus.
In midbrain lesions, the nystagmus is related to the drivers for torsion in the ocular motor nucleus (i.e. III) and related structures such as the interstitial nucleus of Cajal. Small amounts of jerk torsional nystagmus may last for years. In midbrain lesions, vertical saccades are often slowed. (Helmchen et al, 2002).
Midbrain Vascular malformation in patient with nystagmus shown above. Thalamic portion of same vascular malformation
In the case illustrated in the movie above, there was a large thalamic/midbrain vascular malformation combined with a vertical tropia, a torsional nystagmus (fast phases up and to the right), and a shimmering nystagmus resembing that of spasmus nutans. This was associated with strong vertigo. Voluntary vertical saccades were not possible but the vertical VOR was functional. There was no palatal myoclonus.
Many of these patients with thalamic vascular lesions have a similar torsional jerk nystagmus. We have not previously seen a shimmering nystagmus like this in this context.
Torsion is not easily measured. Most practical in 2016, is just to make a video recording with an infrared camera. Scleral eye coil recordings can potentially do a far better job, but it is generally impractical to maintain a system like this for the rare occasions where it is useful.
Contemporary VENG systems do a fairly good job of recording horizontal and vertical but have no capability of recording torsion. A "trick" to get torsion on paper, is to have the person with purely torsional nystagmus to look to the side or vertically. Torsional movement of the eyes is intrinsically in "eye in head" coordinates, rather than "moving with eye" coordinates. In other words, when one looks to the side, the eyeball (the sphere) continues to twist, but the pupil (which has been carried by the eye) now picks up some horizontal or vertical movement. One doesn't need to get too technical about this, but horizontal eye displacement results in vertical pupil movement, and vertical, horizontal. This trick can be used to record torsion.
The recording above is from a patient with a rapid pure-torsional pendular nystagmus, who was looking far to one side.