Hints on Eye Movement Exams

The variety of extra-ocular movement (EOM) disorders literally boggles the mind. After 40+ years as a clinical neuro-ophthalmologist at a large hospital, I still often encounter variants that I have not seen previously. Experienced examiners may find little difficulty getting a grasp on novel variants but newcomers to the field will soon find that the learning threshold is steep. The following hints aim to aid the newcomer to take his or her first steps across the morass. Experienced examiners will not find anything new in this presentation.

From the current perspective, EOM disorders fall into two broad categories: obtrusive and unobtrusive. The obtrusive ones are those that are apparent already from history and/or plain observation, for example, cranial nerve palsies, gaze palsies, gaze deviations, and nystagmus, as exemplified in the EOM+ Video Library. The unobtrusive ones are those that need goal-directed searching for identification. It is the unobtrusive disorders that form the main topic here but a few remarks on obtrusive disorders also may be useful.

Obtrusive disorders

Cranial nerve palsies belong to the bread and butter of most ophthalmic texts and need not be treated here. What may deserve mention is the utility of using a strictly neutral terminology when describing the findings in the individual case. Consider, for example, a unilateral abduction deficit. It is all too easy to label the defect as a 6th cranial nerve lesion and to fail to recognize that there is a host of other disorders that may produce much the same picture. From front to back, common alternatives to cranial neuropathy include

* Showing the effect of 20 minutes of total eye rest ("closet test") on myasthenic ptosis on the right.

An orderly front-to-back consideration of alternatives helps to safeguard against erroneous topical and etiological diagnoses. Be wary of diagnoses with catchy names, e g, ophthalmoplegic migraine and Tolosa-Hunt syndrome. These are very rare conditions and special rules apply for diagnosis.

The best way to learn about the appearances of cranial nerve palsies is to peruse a modern systematic text and to view videos from real patients. There are many good texts available on the Internet, e g, by the Canadian Neuro-ophthalmology Group. The Neuro-ophthalmology Virtual Education Library, NOVEL, is a very rich source of videos.

Obtrusive nystagmus

There are some 40 named variants of nystagmus. Whereas experienced examiners may make the correct diagnosis at a glance, newcomers to the field are recommended to make a careful description of the appearance of the abnormality, using neutral terminology. Once a neutral description is available, it is possible to search for the correct diagnosis in standard neuro-ophthalmology texts, or via the links given in the preceding paragraph. Alternatively, use the Nystagmus Diagnostic Aid.

Unobtrusive EOM disorders

When examining patients for possible defects of eye movement control it is helpful to use a well-reasoned orderly approach. Every examiner has his or her preferred mode of examination. I prefer a "double-cross" algorithm, using a flashlight as the fixation object. A dim surround illumination helps to minimize the number of competing fixation objects.

The double-cross term refers to a specific sequence of target movements. The starting point is the primary position. Step 1 is to execute a slow elevation. (In the diagram, a blue line indicates a slow movement, a red line a fast movement.) A slow elevation appears to be the simplest way to show the subject that eye and head movements can be (and must be) dissociated. The primary goal is to check that full elevation is possible and that diplopia is lacking. A lack of full elevation is common in high age (and then has no diagnostic value in isolation), in endocrine ophthalmopathy (and is then nearly universally associated with vertical diplopia), and in advanced stages of the aqueduct syndrome (see further).

Step 2 is a slow movement to extreme lateral gaze. As in the initial elevation, the nature of the movement is disregarded for now. The main objective is to obtain a full horizontal deviation, without an associated head turn. Check that the amplitude is full and that nystagmus and diplopia are lacking. A horizontal diplopia in extreme lateral gaze has little or no diagnostic interest provided that it is asymptomatic.

Step 3 is to elicit a large horizontal saccade to the contralateral side. For the first time so far, check the nature of the movement carefully. The important parameters are velocity, magnitude, and conjugation. Normal subjects will execute a large and perfectly conjugated saccade in the direction of the new position of the fixation target. This first saccade will rarely be perfectly metric, however: it may be slightly hypermetric (over-shooting) or slightly hypometric (under-shooting). Hence, a small correcting saccade is expected and allowed. Large hypermetria or hypometria, and/or more than one correcting saccade, points toward abnormality. Similarly, a subnormal velocity or a lack of perfect conjugation is abnormal. What is a normal rate of movement has to be learned from experience. As in step 2, check that a full horizontal deviation can be obtained and that end-point nystagmus and diplopia are lacking. Slow horizontal saccades are most commonly due to spino- or olivoponto-cerebellar degeneration, progressive supranuclear palsy, Huntington chorea, or progressive external ophthalmoplegia. The most common cause of unobtrusive, horizontally disconjugated saccades is partial internuclear ophthalmoplegia (INO), with slowing of the adducting eye.

There is much at stake in this first saccade test. On repeated testing, many subjects learn to compensate for their deficits. For maximum diagnostic efficiency, the various steps in the EOM exam must be properly planned and executed, and unpredictable to the subject. Hence, step 4 is not what might be expected, a large saccade to the other side. Instead, step 4 is a slow movement to maximum down gaze. Disregard the nature of the movement but check that the amplitude is full and that nystagmus is lacking. Incidentally, down-beat nystagmus is usually most conspicuous on gazing down and to the side, like looking into a side pocket.

Step 5 is to execute a full saccade upwards. Again, carefully observe the rate, metrics, and conjugacy of the eye movements. Conjugacy may be the most important aspect as it fails in the very first stages of the aqueduct syndrome, long before the amplitude of upgaze is reduced. Conjugacy is then typically replaced by one or more convergence/retraction jerks. In late stages, the aqueduct syndrome may superpose convergence/retraction jerks also on horizontal saccades.

Step 6 is a simple transportation manouvre, to the starting point (7) of a large horizontal saccade in the hitherto untested direction. The horizontal saccade is carefully monitored as described under step 3.

Step 8 is a another simple transportation manouvre, this time to the starting point (9) of the last large saccade, downwards. Here, speed of movement is of paramount importance. A subnormal velocity of downgaze saccades is an early hallmark of progressive supranuclear palsy, PSP.

It may appear surprising that the double-cross procedure completely disregards the characteristics of pursuit (following) eye movements. This is because the pursuit system is easily disturbed by extraneous factors like a lack of sleep and side effects of various drugs. In most instances, it is only asymmetric (right-left, up-down) pursuit deficits that may have some clinical interest. Such asymmetries may be best elicited by a test of optokinetic nystagmus.

The double-cross procedure cannot reveal all types of unobtrusive EOM abnormalities. Useful complements include

One way to assess the vestibulo-ocular reflex (VOR) is to use a head-fixed fixation target and to ask the subject to smoothly turn his or her head alternatingly about 30°, about once a second, several times, while fixating on the target all the time. Try rotations around the cranio-caudal axis first, then around the transversal one. An abnormal reflex is signalled by the appearance of nystagmus during head rotation. Watch carefully for sources of error, i e, loss of fixation and "cog-wheeling" head turns. A drinking straw may serve as an improvised fixation object. It can be held in place by biting on one end or by fastening the straw to a trial lens holder.

A convenient way to documentation is to ask the patient to hold a video camera and look into the camera lens while the chair is rotated to and fro:

Note subtle upbeat nystagmus and retrocollis in the resting position. Nystagmus appears on rotation both to the right and the left. Case of ependymoma localized to the cranio-cervical junction.

Further, important information can be gained already from observing the patient's gaze pattern prior to the formal examination. Many neuro-degenerative disorders are associated with a paucity of spontaneous fixation changes. Patients suffering from neglect may favor gaze to the non-neglected side. Patients with lateropulsion may be seen to show an evanescent conjugate deviation of the eyes to one side upon eye opening.

Finally, a note of caution: patients with extensive visual field defects may have difficulties seeing rapidly moving fixation targets and may use compensatory, searching saccades. Searching eye movements must not be misinterpreted as primary defects of eye movement control.