Nystagmus in Infancy and Childhood

Shop

Clinical Education /

Education /

Disease Reviews

MAR 21, 2018

Nystagmus in Infancy and Childhood

By Louis Frank Dell'Osso PhD

Nystagmus

Add to My Bookmarks

View

Mark Complete

Remove

Comments

Introduction

Early nystagmus terminology was permeated with misleading, often-incorrect clinical descriptions and presumptions. Therefore, to avoid the misunderstandings and misrepresentations resulting from that older terminology, the newer terminology and descriptions established by a workshop held at the turn of this century are used.¹ Thus, INS (infantile nystagmus syndrome) replaces “CN” (congenital nystagmus) and FMNS (fusion maldevelopment nystagmus syndrome) replaces “LMLN” (latent/manifest latent nystagmus).

Ocular motor research in the past half-century has produced remarkable advances in our understanding of the underlying mechanisms of the several types of nystagmus seen in infancy and childhood. The confluence of accurate methods of recording eye movements, the application of the rules of physics, waveform analysis techniques, and control-systems analysis allowed replacement of inaccurate terminology, erroneous concepts, and the ensuing diagnostic and therapeutic quagmire. As a result, we now have a science-based approach to understanding the underlying cause and mechanisms of nystagmus and to the diagnosis and treatment of patients with nystagmus. For instance:

the INS in all patients, regardless of any visual system comorbidity, is directly caused by instability in the damping portion of the smooth pursuit system

that simple, pendular, velocity instability gives rise to all of the predominant INS waveforms (pendular and jerk) by control mechanisms inherent to the normal ocular motor system (OMS);

INS can be definitively diagnosed by eye-movement recordings;

combinations of INS and other types of nystagmus can also be definitively diagnosed by eye-movement recordings;

the use of a mathematical function (the eXpanded Nystagmus Acuity Function, NAFX) allows a priori prediction of which patients can be provided with improved visual function by specific therapies;

the NAFX and its derived function, the Longest Foveation Domain (LFD) can provide an estimate of the amount of improvement that will be produced by current accepted therapies; and

INS therapies should be aimed at damping the primary velocity instability to allow the OMS to improve the quality of the foveation periods within INS waveforms but should neither diminish nor destroy any of the OMS functions necessary for appropriate real-world visual function.

Specifically, the attempt to improve cosmesis by damping or eliminating the nystagmus itself should not override the patients’ ability to advantageously use their OMS for all normal functions. Overzealous parents of children especially should be educated to understand that the appearance of INS is not the major concern of either the patients (as they grow up) or of the child’s physician; the physican’s primary concern should be the visual well-being of the young patient and the physician should be the child’s adult advocate. Crippling the OMS to eradicate INS (even if that were possible) should never be an option imposed on a child who cannot give informed consent or even an adult until they are informed of the symptomatic deficits and potential dangers such a procedure would produce.

This chapter will concentrate on translating the above research results to better enable the clinician to make accurate diagnoses and provide safe and effective treatment (ie, multiple improvements in visual function) of patients with nystagmus. It is both an introduction to and overview of nystagmus in infancy and childhood and provides a data-based foundation for clinical diagnosis and treatment. Hopefully, it will also pique the interest of future researchers. A more thorough and fully referenced presentation of ocular motor research foundations, diagnostic findings, clinical “pearls,” and therapeutic approaches and outcomes may be found in the book on the subject by Hertle and Dell’Osso² and the many scientific citations contained therein.

Prior to the application of ocular motor research to the nystagmus of infancy, only clinical observation (albeit astute observation by pioneers like Kestenbaum and Anderson)^3-5 was available. Unfortunately, a convoluted and factually unsupported ophthalmological mythology evolved, was taught to pediatric ophthalmology residents, and subsequently used to guide their clinical practice. It and the contents of most early medical texts on infantile nystagmus were based solely on those clinical impressions, unwarranted presumptions, and contradictions to the laws of physics. Some of the putative “truths” they presented were:

Congenital Nystagmus “CN” is an oscillation of the eyes across the line of regard (target);

There are 2 types of “CN,” “sensory” and “motor;”

“Sensory CN” is caused by a visual sensory deficit and has a pendular waveform;

“Motor (aka “idiopathic”) CN” has no known cause and a jerk waveform;

There are no treatments for “CN;” and

Surgically shifting the “CN” null to correct a head turn will not improve either the “CN” or visual acuity.” That statement disregarded Anderson’s observations.⁵

Ocular motor research demonstrated that all of the above ”truths” were false. That is:

INS is an oscillation away from and back to the target, as simple physics dictates. Those with INS can have good and even normal visual acuity; therefore, their eyes must come to rest when the fovea is on target, not pass the target at maximum velocity, as the prior description necessitates;

All patients with INS have the same motor oscillation whose characteristics fall within idiosyncratic ranges (ie, types of waveforms, frequencies, amplitudes, and gaze-angle variations). Thus, INS is the same condition in all of the clinical variations encountered in patients;

INS in most patients has an underlying pendular oscillation that is transformed into the various pathognomonic waveforms by different types of normal ocular motor attempts to maintain target foveation;

All INS has the same primary cause (an undamped smooth-pursuit oscillation) and is modified to an idiosyncratic extent (eg, the position and sharpness of the INS “null”) by visuo-vestibular, constant-velocity, slow-phase movement—thus, no patient’s INS can accurately be called “idiopathic,” a misnomer derived from the mistaken designations of “sensory” and “motor” CN. Most patients with solely jerk waveforms reflect the underlying pendular pursuit oscillation; only those rare cases with solely linear slow phases reflect the visuo-vestibular tonic imbalance;

There are many treatments (optical, pharmacological, and surgical) that significantly improve visual function in those with INS, with or without associated sensory visual deficits; and

Surgically shifting the point of maximum foveation quality will usually improve both the foveation quality of the INS waveform and visual acuity across a broader range of gaze angles.

Chapters

Relevant Anatomy and Physiology

Infranuclear

This portion of the ocular motor system consists of the extraocular muscles (EOM) and their pulleys, orbital tissues, and the muscle and orbit nervous anatomy. There are 3 pairs of EOM: the horizontal rectus muscles (lateral and medial); the vertical rectus muscles (superior and inferior); and the oblique muscles (superior and inferior). Each pair operates in a push-pull manner to rotate the eye in their plane of action. A muscle pulley system maintains the stability of the muscle paths. The pulleys change their positions as a function of gaze; their exact role in ocular kinematics is unknown.

Several cranial nerves, the oculomotor (III), trochlear (IV), trigeminal (V), and abducens (VI) innervate the EOM. Cranial nerve III carries somatic motor fibers to the levator palpebrae, inferior rectus, medial rectus, superior rectus, inferior oblique, and sympathetic efferent fibers (preganglionic fibers) to the ciliary ganglion. Cranial nerve IV contains somatic motor fibers only. It supplies the superior oblique muscle of the eye. Cranial nerve V contains somatic motor and somatic sensory fibers. It plays a complex role in proprioception and proprioceptive control of the EOM responses. Cranial nerve VI contains somatic motor fibers only that supply the lateral rectus muscle of the eye.

Supranuclear

This consists of the frontal eye fields, superior colliculus, brainstem nuclei, vestibular nuclei, and cerebellum. The posterior parietal cortex, which contains neurons that are modulated by visual attention, is involved in the visual guidance of saccades by shaping the visual inputs to the superior colliculus. The frontal eye fields can selectively activate superior colliculus neurons, playing a role in the selection and production of voluntary saccades. The substantia nigra pars reticulata funnels inputs from the frontal cortex and acts as a gate for the voluntary control of saccades, keeping in check the superior colliculus activity. The parietal lobe seems mainly concerned with privileged target selection. Peristriate and parietal cortical areas may encode the information relevant to smooth-pursuit generation.

The superior colliculus provides both the motor command to the burst neurons of the paramedian pontine reticular formation (PPRF) and the trigger command to the omnipause neurons. It sends projections to both the horizontal (PPRF) and vertical gaze centers (rostral interstitial nucleus of the medial longitudinal fasciculus, riMLF), providing the motor command to move the eye to an intended new position for the foveation of a visual stimulus.

The vestibular neurons are bipolar with their cell bodies located in Scarpa’s ganglion in the internal auditory meatus. The superior and inferior vestibular nerves join to form a common bundle that enters the brainstem. From the vestibular nuclei, projections go to the cerebellum, extraocular muscle nuclei, antigravity muscles, and opposite vestibular nuclei. The semicircular canals detect head rotation and drive the rotational vestibulo-ocular reflex (VOR), whereas the otoliths detect head translation and drive the translational VOR.

The cerebellum plays an important role in eye movements. Together with several brainstem structures, including the nucleus prepositus hypoglossi and the medial vestibular nucleus, it appears to convert velocity signals to position signals for all conjugate eye movements through mathematical integration. Because of this, all of the structures involved in this process are often referred to as the “neural integrator.” Various types of image-stabilizing reflexes are also “cerebellar” functions. Pursuit, VOR cancellation, and holding the eye steady for fixation, both immediately after saccades and in eccentric positions of gaze, are controlled by the flocculus (and probably paraflocculus). The nodulus (and ventral uvula) modulates "low-frequency" aspects of vestibular responses and hence controls the duration (time constant) of the VOR. The dorsal vermis and underlying (posterior) fastigial nuclei participate in the control of the size of the saccadic pulse of innervation and hence saccadic accuracy.

Afferent

This consists of the retina/optic nerve, optic nerve, lateral geniculate, geniculostriate, association cortex, and ocular motor proprioception. The retina is organized both vertically (in columns) and horizontally (in layers). The principal "vertically oriented" elements are receptors (rods and cones), the bipolar cells and the ganglion cells. Rod and cone photoreceptors are easily distinguished by their outer segments that contain photopigment in free-floating disks (rods) or folded layers (cones).

The optic nerve (cranial nerve II) is part of the central nervous system. It is composed of retinal ganglion cell axons and support cells and leaves the orbit via the optic canal, running postero-medially toward the optic chiasm, where there is a partial decussation (crossing) of fibers from the temporal visual fields of both eyes. Most of these fibers terminate in the lateral geniculate nucleus (LGN).

The LGN serves as a relay station in the projection of the visual pathway to the striate cortex. It consists of six layers with each alternating layer receiving inputs from a different eye, 3 layers for each eye. The outer 4 layers are composed of small cells, and correspondingly, receive inputs from the small ganglion cells of the retina (parvocellular); these cells dominate the fovea, are color sensitive, and are "fine-grained," meaning their receptive fields are small enough that they can pick up a high level of detail. The 2 most ventral layers are the magnocellular layers and are composed of large cells, which receive their input from large ganglion cells (magnocellular). These cells receive information from a wide radius of bipolar cells. They are mostly found in the peripheral retina, are insensitive to color, and are "coarse-grained" (relatively insensitive to detail), but are sensitive to motion. Therefore, 2 types of information, motion vs. color and form, are kept in separate layers in the LGN.

From the LGN, fibers of the optic radiation pass to the visual cortex in the occipital lobe of the brain. The projections of the magnocellular and parvocellular layers of the LGN terminate in separate sublaminae of layer IV of striate cortex; a more superficial projection of the parvocellular layers to a narrow strip at the base of layer III. As visual information passes forward through the visual hierarchy, the complexity of the neural representations increases. Whereas a primary visual cortex (V1) neuron may respond selectively to a line segment of a particular orientation in a particular retinotopic location, neurons in the lateral occipital complex respond selectively to complete object, and neurons in visual association cortex may respond selectively to human faces, or to a particular object. Along with this increasing complexity of neural representation may come a level of specialization of processing into 2 distinct pathways: the dorsal stream and the ventral stream. The dorsal stream (the "where" or “how” stream) is involved in spatial attention and communicates with regions that control eye and hand movements, and the ventral stream (the "what" stream) is involved in the recognition, identification, and categorization of visual stimuli.

Eye-muscle proprioceptive signals provide information used in normal sensorimotor functions; these include various aspects of perception, and of the control of eye movement. The importance of proprioception in ocular motor control has been either ignored or underestimated, possibly because accurate saccades (non-linear, all-or-none motor signals) may be made in its absence. Muscle spindles are found within the proximal and distal regions of human infant and adult extraocular muscles and are located at the junction of the orbital and global layers. Palisade endings, a class of muscle receptor found exclusively within extraocular muscles, including those of humans, are located at the distal myotendonous junction of the multiply innervated non-twitch fibers of the global layer. They may be the principal source of proprioceptive feedback from extraocular muscles. Each layer of the extraocular muscles has its own type of sensory receptor to generate afferent signals, with the orbital layer utilizing muscle spindles and the global layer relying on palisade endings. Sensory signals from palisade endings appear to form part of a proprioceptive feedback network that modulates the non-twitch motor neurons that innervate the slow non-twitch extraocular muscle fibers. Thus, proprioceptive control of static muscle tension could affect the small-signal (linear, push-pull motor signals) gain of the EOM, and reducing that gain could damp an ongoing nystagmus—that is precisely the hypothetical action by which EOM surgery damps INS. A more thorough understanding of eye-muscle proprioception is necessary for understanding the physiology and pathophysiology of eye-movement control, as well as therapeutic intervention.

Efferent

This consists of the smooth pursuit, saccadic, vergence, and visuo-vestibular (vestibular and optokinetic) subsystems. Ocular motor research has identified the direct causes/sources of the 2 major types of nystagmus of infancy as: 1) a failure in the development of adequate damping of the relatively unstable (underdamped) normal smooth pursuit system and 2) a tonic imbalance in the visuo-vestibular system modified by the Alexander’s law variation of the resulting slow-phase velocity. Smooth pursuit permits us to maintain a steady image of a moving object on our foveas and to thereby track moving targets with clear vision. It has evolved as an underdamped control system, meaning that the responses to changes in velocity commands are in the form of a damped oscillation about the new output velocity. The failure in calibration of this damping has been hypothesized as the primary (direct) cause of all major, complex INS waveforms (the other being an imbalance in the visuo-vestibular system that produces linear slow-phase waveforms).

Several forms of saccades, the fastest eye movements, can be observed: voluntary saccades to objects of interest, reflex saccades to unexpected new stimuli, spontaneous saccades that occur in normal inactive subjects, saccades that form the quick phases of vestibular and optokinetic nystagmus, and braking saccades that stop or reverse accelerating slow phases of nystagmus.

Vergence eye movements turn the eyes in opposite directions (convergence, divergence, and cyclovergence) so that images of objects will fall on corresponding retinal points. Three major stimuli elicit vergences: (1) retinal disparity that leads to fusional vergence; (2) retinal blur that evokes accommodative vergence; and (3) motion that induces both disparity and accommodative vergence. Conjugate and vergence signals are generated independently and are combined at the extraocular motoneurons. Both convergence and divergence cells are found intermixed in the mesencephalic reticular formation within 1–2 mm of the oculomotor nucleus. The vergence system has both a fast and a slow subsystem, each with different properties. The fast vergence system is best elicited by stimuli with large retinal disparity errors and/or velocities larger than 4°/sec. The slow vergence system is elicited by small disparity errors and/or disparity velocities of less than 3°/sec. Tectal and pretectal midbrain areas contribute to the near triad (simultaneous convergence, accommodation of the lens, and miosis), occurring during shifts in fixation between distance and near.

The vestibular apparatus drives reflex eye movements, which keep images of the world steady on the retinas during head or body motion. The eyes move in the opposite direction to the movement of the head so that they remain in a steady position in space. The semicircular canals are the end organs that provide the innervation to the vestibular nuclei, which in turn drive cranial nerves III, IV, and VI to compensate for rotations of the head. In contrast, the otoliths respond to linear accelerations of the head and to gravity when the head is tilted. The optokinetic system serves to stabilize slow movement (the optokinetic reflex, OKR) of the visual surround on the retina by generating conjugate eye movement whose velocity equals that of the surround.

Despite the complex waveforms found in INS patients, their saccadic, pursuit, vestibular, and optokinetic systems function normally^7-11 and, because they can use efference copy of the motor signals driving those complex waveforms, they do not experience oscillopsia.¹²

Infantile Nystagmus Syndrome

Characteristics

INS has been documented to occur at birth, especially in families with hereditary INS; it usually appears in infancy or childhood, and rarely, even later in life.¹³ The manifestation of INS coincides with the attempt to use or direct the eyes (“fixation attempt”—even in the dark) and it may disappear entirely with inattention or sleep.¹⁴ Stress exacerbates the nystagmus,¹⁵ as does gaze eccentric from its “null” position. The null position is that range of gaze angles where the nystagmus waveform has minimal amplitude and has the highest foveation quality. On either side of the null, the amplitude grows and, more importantly, foveation worsens. This true null (in the mathematical sense) differs from the Alexander’s-law induced damping occurring in other types of nystagmus (see Section 3). A preferred head position indicates a narrow range of gaze angles with higher acuity and may result in scanning head motion during reading.

There are 12 basic INS waveforms, 9 of which are pathognomonic and provide definitive INS diagnosis as well as ruling out various types of acquired nystagmus. Some of the INS waveforms may have a low-amplitude, high-frequency, pendular oscillation (thought to be due to instability in the circuitry of the nucleus of the optic tract, NOT) superimposed on them; these additional “dual” waveforms (4 recorded thus far) may be pendular or jerk and 3 may also be considered pathognomonic (16 waveforms, 12 pathognomonic).

INS waveform characteristics may vary with gaze angle, convergence angle, eye velocity, fixating eye, or time.^{8-10; 16} The amounts and types of variation (eg, amplitude, frequency, foveation quality, etc.) are idiosyncratic and determine the types of therapeutic approaches that would maximize improvements. Gaze-angle variations usually result in a single, static null position where the waveforms yield maximal foveation quality. In cases of INS that is damped by convergence, better foveation quality is usually achieved with convergence than from gaze-angle variation.¹⁷ Higher acuity at near than at distance indicates a convergence null. Eye velocity generated by smooth pursuit, the VOR, or OKR may also shift the position of the static null resulting in a dynamic null that differs from the static null. If the INS also varies with the fixating eye, the nystagmus has a latent component and if it varies with time, it is asymmetric (a)periodic alternating nystagmus (APAN). Because the dynamic null position is a combination of the static, gaze-angle null modified by any or all of these factors in a given patient, it sometimes appears clinically that more than one null is present, even to the extent that more than one head position may be adopted. However, in INS there is only one null but it is subject to shifting under different viewing, ocular motor, or time conditions.

Because the period of APAN is long compared to acquired PAN, it requires observation over 10-15 minutes of attempted fixation in primary position to identify it. This should be done with one eye occluded to prevent shifts in the fixating eye from confounding the observation. In some cases, the APAN is so asymmetric that the interval exhibiting jerk nystagmus in one direction may far exceed that in the other. Also, as the name, “aperiodic” implies, the times taken for each full period (jerk left/neutral/jerk right/neutral) may vary from one full cycle to the next. Large changes in visual acuity from one office visit to another may be due to APAN; also, head turns may be exhibited in cases of APAN that are highly asymmetric.

Careful recordings have revealed that although INS may appear clinically horizontal, there are usually smaller, subclinical torsional and vertical components. The conjugate torsional waveforms are in synchronization with the horizontal nystagmus but the vertical are disjugate (ie, a see-saw nystagmus).¹⁸ Some cases of INS are multiplanar with conjugate, horizontal, vertical, and torsional components (eg, circular, elliptical, or oblique nystagmus). In rare cases of human achiasma (or hypochiasma), there is a prominent see-saw component.¹⁹

Etiology

It is important to identify the direct=primary=if-and-only-if cause of INS and to determine why such a predominantly horizontal ocular motor oscillation is so prone to occur in the presence of any one of a number of afferent visual deficits or no visual deficit at all. It is significant that these associated deficits are unrelated to each other, occur in different planes (or in no plane), and are mechanistically unrelated to horizontal nystagmus. For far too long, simplistic and lazy thinking has prevailed that inappropriately and indiscriminately attributed causality to all but the actual cause(s) of INS. The resulting adjectives, so called “sensory defect” and “motor defect” INS, are erroneous and redundant, respectively—nor were they intended by David Cogan, to whom they were incorrectly attributed.²⁰

Consider the following:

If an adult falls on a concrete sidewalk and fractures his arm, it is equally irresistible and false to attribute the cause of either bone fracture in general or, in that specific case, to the fall. What if: 1) the same fall took place in a soft bed of grass and no fracture resulted; 2) a child fell on the sidewalk and no fracture resulted; or 3) an elderly person fell in the grass and a fracture resulted?—same fall, different results. Clearly, falling is neither the direct cause nor the specific cause of bone fracture; the direct cause in all of these cases and in others (eg, being struck by a hard object) is deformation of the bone beyond its limit of elasticity.

Similarly, in patients who exhibit any of the afferent visual deficits normally associated with INS, none of the former is the direct cause of INS, nor should that logical fallacy (post hoc ergo propter hoc) be passed on to the patient or parent. It is no more difficult, and far more accurate, to say, “INS is caused by a failure of the ocular motor system (or, the part of the brain that controls eye movement) to keep the eyes stable.” The description of any associated afferent visual deficit, if present, should of course be included as an additional part of the patient’s overall condition including, “Although that deficit did not cause the INS, it may have facilitated it by interfering with the development of stable eye control.”

Because the normal smooth pursuit system is quasi-stable (ie, underdamped), it is prone to continuous oscillation if the calibration that normally occurs during early development fails. Such a control system may become unstable whether or not there are associated afferent visual deficits that could interfere with calibration. Both ocular motor research and computer modeling support the conclusion that INS is caused (ie, direct=primary=if-and-only-if) by a failure in calibration of smooth-pursuit damping. This explanation is also supported by the normal function of all of the ocular motor subsystems in the presence of INS, by the wide variety of patients with INS, and by the demonstration that the INS of all patients is the same entity (ie, the same combinations of waveforms and characteristics during fixation, pursuit, or other tasks) regardless of the clinical picture in which it appears. Finally, the computer model’s emergent characteristics (ie, not designed into the model) that mimicked those measured in the eye movements of INS patients provide strong support for this being the causal mechanism.^21-25 Thus, the correct description of this nystagmus in all patients is “INS” without any of the prior adjectives attached to it and without the word, “idiopathic” (ie, cause is unknown), which cannot be used to describe INS, a condition for which the cause has been identified.

Deficits and Measurements

The accuracy, repeatability, and duration of the foveation periods are the most critical features of INS waveforms' effect on visual acuity. Studies of the dynamics of the foveation periods in INS led to the development of a nystagmus foveation function (NFF) that was the first indicator of the gaze angle of best acuity in an individual and a means of correlating waveform characteristics and acuity between subjects, something that IN intensity cannot do. A more sensitive function, the NAF, followed, which, after testing and use in a number of INS patients, was altered to accommodate the larger foveation windows required by many INS patients. The result was the NAFX, which has undergone over 15 years of testing, improvements, automation, and use in hundreds of patients with uniplanar or biplanar INS.^{26; 27} Since, like its predecessors, it contains the 3 waveform factors (foveation time, position variation, and velocity variation) that define well-developed foveation periods and affect acuity the most, its value is linearly proportional to the best-corrected visual acuity of INS patients with no afferent deficits. That is, because the NAFX is solely a function of the INS waveform characteristics, it is independent of the state of the patient’s visual sensory system. More specifically, only the foveation characteristics of the waveform are used in its calculation; all other, non-acuity–affecting, portions of the waveforms (eg, amplitude or peak slow-phase velocity) are automatically discarded. Clearly, for a given INS waveform, the resulting visual acuity will be better when the SD of the foveation-period position is small than when it is large. The same applies for the SD of the foveation-period velocity. Since they may vary independently, both are necessary factors in a function that is related to potential visual acuity and functions that do not contain both will not accurately reflect acuity in all patients.

The inclusion of the above factors makes the NAFX a powerful tool in determining, a priori, what improvements in visual function may be expected by INS therapies that affect only the nystagmus (eg, EOM surgery, prisms, soft contact lenses, topical or systemic drugs) and do not affect any afferent deficit that might or might not also be present. Using the NAFX, the physician can, for the first time, estimate not only improvements in peak visual acuity but also the more important improvements in the range of gaze angles over which the patient has their highest acuity (see section 7, Fig. 8). These estimations can then form data-driven foundations for therapeutic decisions to be made by the physician and the patient. Used properly, the NAFX can more specifically delineate expected visual function improvements, and thereby guide the patient’s expectations. This accurate measure of the factors controlling high visual acuity that are not under the subject’s control far exceeds the use of head-posture measurements, which are not as accurate and are under the patient’s control.

Treatments

There are several approaches to INS treatment: physical/optical; pharmacological; and surgical. They are complementary and may be combined for maximal therapeutic improvements in specific patients. The aims of INS treatments should be to: 1) maximize visual function while preserving all other ocular motor functions; 2) alleviate the need for head postures that may cause secondary problems; and 3) damp the nystagmus for cosmetic reasons. I have listed these in order of their important effects on the quality of life of the patient rather than the concerns of the patient’s parents, which often are overly concentrated on the cosmetic appearance of the eye motion without truly understanding the importance of the visual and ocular motor functions for which the eye and ocular motor system evolved. Thus, I intentionally rule out any therapy that effectively or actually cripples the ocular motor system by either maximally recessing or, worse yet, irreversibly removing (extirpation) or severing (myectomy) the horizontal rectus muscles.

Such radical therapies sacrifice both visual and ocular motor function (eg, accurate saccades to targets, binocular alignment for stereopsis, and both the optokinetic and vestibulo-ocular reflexes necessary to maintain stable vision in the presence of either environmental or body motion). In my opinion, they are so debilitating that they should not be employed on adults without a thorough explanation of their detrimental effects and never on children until they are old enough to make a competent decision for themselves, regardless of their parents’ desire for cosmetic improvement. I regard such ego-driven (man conquering nature and evolution) procedures as not only lacking any scientific foundation (or even understanding) but also as essentially harmful to the patient’s visual well being (ie, the iatrogenic, symptomatic deficits rise to the level of malpractice).

Fusion Maldevelopment Nystagmus Syndrome

Characteristics

FMNS exhibits: a jerk nystagmus with either a linear or decreasing-velocity exponential slow phase identical to that of gaze-paretic nystagmus; strabismus; alternating hyperphoria/dissociated vertical deviation; and pendular torsional nystagmus in primary position.¹ The constantly present, conjugate, horizontal, jerk nystagmus increases in intensity by monocular occlusion, blurring, or reducing image brightness. A jerk nystagmus with a linear slow phase may be present when both eyes are closed. Rarely, the nystagmus is only evoked by the "pure" or "true" latent condition (“LN”) and occurs only with uniocular viewing (ie, the other eye being occluded). That is, there is no nystagmus when both eyes are viewing, but when one eye is occluded, jerk nystagmus develops in both eyes, with the fast phases toward the uncovered eye. FMN may clinically resemble other types of nystagmus (eg, INS with a latent component) and require eye-movement recordings to differentiate it. Indeed, it can even present as spasmus nutans (see Section 4).²⁸

Patients with FMNS always have strabismus and, to eliminate diplopia, vision from the tropic eye is suppressed (“occluded”) in the cortex. The term, FMN refers to this single type of nystagmus that is present in most FMNS patients with both eyes open while one is fixating but, in some patients, may only be present when one eye is occluded. The slow phase of FMN is either linear or a slight decreasing-velocity exponential (initiating the nystagmus) or a prominent decreasing-velocity exponential (following saccadic pulses), and the fast phase is always in the direction of the eye that is fixating, the straight eye. That is, the slow-phase rotation of the fixing eye is always in adduction and the fast-phase is in abduction; fixation with the right eye generates a right-beating nystagmus of both eyes, while fixation with the left eye produces a left-beating nystagmus of both eyes.

Just as in INS, FMNS may also coexist with NOT nystagmus. The nystagmus of patients with strabismus, alternating fixation, and FMNS with both eyes open has fast phases that are always in the direction of the fixating eye. Such patients may be easily misdiagnosed as having INS, because the nystagmus is present with both eyes open. Recordings are required to document the decelerating or linear slow-phase waveforms characteristic of FMNS from the accelerating slow phases predominant in the INS. In addition to occlusion, intent or darkness also may alter FMN; the direction of FMN depends on the eye intended for fixation.

Accurate studies of FMN foveation in a patient with 20/15 acuity, revealed a dual strategy.²⁹ During the low-amplitude, linear-slow-phase FMNS waveform, the saccadic fast phases foveate the target, and the low-velocity slow phases take the eye away from the target with little effect on acuity. During the higher amplitude, decelerating slow-phase FMNS waveform, the saccadic fast phases defoveate the target, allowing foveation during the low-velocity, tail ends of the slow phases; this ensures the best acuity possible and was the first recorded demonstration of the saccadic system acting deliberately to defoveate the target.

The intensity of FMNS is maximal in abduction and minimal in adduction of the fixing eye, causing an “adduction” damping and not a true gaze (eye in orbit) “null” position. To take full advantage of Alexander’s law in order to minimize the amplitudes of their nystagmus, many patients with FMNS alternate their fixing eye such that it is always the adducting eye. Thus, they fixate with the left eye when looking right and vice versa. To view targets that are straight ahead, head turns usually result. In addition to causing the anomalous head posture, this may give the mistaken clinical impression of a nystagmus (usually INS) with two “nulls” if the clinician fails to detect the change in the fixating eye or realize that there is no true null (ie, the nystagmus does not increase as the eye is rotated further in adduction).

Despite the lack of extended foveation periods in FMN with linear or slightly decelerating slow phases, the low amplitudes of these slow phases may still allow for good visual acuity. As long as sufficient data points fall within the foveation window of the NAFX (centered at the ends of the fast phases) for each FMN cycle, the NAFX value, and visual acuity, will be high. The roles of efference copy and foveation in the suppression of oscillopsia in INS are equally applicable in FMNS where oscillopsia is not normally experienced because of the efference-copy signal of the FMN command.

Etiology

FMNS is “congenital” in the same sense as INS (ie, a congenital predisposition). FMNS occurs in patients with strabismus who, although viewing with both eyes open, are fixing monocularly. Strabismus is a necessary (but not sufficient) condition for FMNS. Various theories have been advanced to explain FMNS. These include a primitive vestibular tone imbalance, poor egocentric localization, a subcortical optokinetic system anomaly, a subcortical maldevelopment of retinal slip control, abnormal cortical motion processing, a disorder of proprioception and a phylogenic persistence of the dominance of the nasal half of the retina.^30-38 The direct cause of FMN is a tonic imbalance that drives the eyes with constant velocity (producing linear slow phases). The precipitator of the imbalance may be the naso-temporal asymmetry present in normal infants that disappears as fusion develops or, as we hypothesized, egocentric direction confusion.

Treatments

When considering therapy for FMNS, it is important to understand the different components that contribute to the nystagmus and how (and at what site) each proposed therapy works. Similar to INS, there are both sensory and motor components. The absence of fusion is the sensory component that contributes to the tonic motor imbalance that drives FMN. The effect of gaze angle (Alexander’s law) is the motor component modulating the FMN. Finally, the proprioceptively controlled small-signal gain of the extraocular muscles (a motor component) also can modulate the FMN. Therefore, different therapies and adaptations by the patient can act in distinct mechanistic ways to damp the FMN and, in some cases, restore fusion.

The primary treatment of FMNS is a combination of medical, optical and surgical therapy to create, restore or improve binocular function. Since the intensity of the FMN is related to the degree of binocular function, improving fusion will damp the FMN and improve visual function. The addition of nystagmus surgery, in the form of tenotomy and reattachment (T&R) of any horizontal rectus muscles not operated on to realign the eyes, should also further damp the FMN.

Because of Alexander’s law, it would be disadvantageous to move the fixating eye medially as it would require additional abduction innervation to fixate targets in either primary position or further in abduction. In those exotropic FMNS patients for whom fusion is impossible, recessions of all four horizontal recti with a large differential (more on the lateral than the medial recti) has proven successful in treating both the exotropia and the motor component of the nystagmus. It is also possible to surgically enhance the acuity of some patients with FMNS at the same time as correcting the strabismus and head posture. These operations are considered broadly as treatment of “strabismus and nystagmus with or without an anomalous head posture.”

In summary, to address both the FMN, head posture (if not alternating), and strabismus in the same procedure, the two horizontal rectus muscles on the eye responsible for the head posture are recessed and resected to straighten the head while the two horizontal recti on the non-fixing eye are recessed and resected to correct the resulting strabismus. The advantage of operating on all four horizontal recti is that, in addition to treating the strabismus and head posture there is the potential of fusion and further damping of the FMN produced by the T&R effect on the proprioceptive control of the small-signal gain of the extraocular muscles. Another advantage of the T&R additions to the strabismus procedure is that, in addition to damping the FMN, those muscles receiving a T&R are left intact and may be used for future strabismus adjustments that might become necessary.

Other Types of Nystagmus of Infancy and Childhood

Nystagmus Blockage Syndrome

The nystagmus blockage syndrome (NBS) was defined clinically as one in which a patient with nystagmus damps or changes that nystagmus by willfully deviating his fixating eye inward (eg, “convergence”). The lack of specificity regarding the nystagmus types or waveforms present both pre- and post-“convergence” led to problematic diagnoses and interpretations of the clinical symptoms. Thus, the NBS remains both a poorly understood and an over-diagnosed phenomenon related to INS. As the name suggests, the nystagmus of these patients diminishes or disappears clinically with the act of willed esotropia while fixating a distant target. This should not be confused with the damping of IN during true convergence on a near target. Based on our research, the NBS can now be more accurately defined as a syndrome in which a patient with INS plus a variable esotropia willfully deviates the fixating eye into adduction to accomplish either a damping of the IN or a switch from IN to a low-amplitude FMN.¹

The first characteristic revealed by eye-movement data was that there were two mechanisms by which blockage of the ongoing nystagmus can be accomplished; that resulted in two different types of patients with the NBS.^{39; 40} Patients with both types had IN when their eyes were aligned. In Type I, the IN either damped or stopped entirely upon willful esotropia, in much the same way as with true convergence. In Type II of NBS, the INS waveform converts to a low-amplitude FMNS waveform with the onset of the strabismus. Normally, the substitution of the FMNS slow phases for the INS waveforms that allow for better foveation would not be advantageous. However, in these few patients, the low FMNS amplitude results in better acuity than the larger INS amplitude. NBS is often misdiagnosed in FMNS patients with a strong Alexander's law variation of their nystagmus, which causes them to fixate with their adducting eye. Thus, the NBS encompasses two different types of infantile nystagmus and the ability to willfully change the amount of esotropia present to improve the nystagmus waveform and thereby, visual acuity. The NBS has been reported in an exotropic patient⁴¹ and also in a congenitally blind patient.⁴²

Target foveation depends on which type of waveform (INS or FMNS) is present. In some cases, the nystagmus is totally blocked by the purposive esotropia and foveation is essentially the same as in unaffected individuals. Foveation accuracy also depends on which type of waveform (INS or FMNS) is present. In the case of complete nystagmus blockage, the accuracy is essentially the same as in unaffected individuals.

A useful clinical sign in differentiating NBS from other forms of convergence excess esotropia is the absence of pupillary constriction. These patients are not using their accommodative vergence mechanism to block the nystagmus but instead, are depending upon some other mechanism to bring about damping of the nystagmus. Prior to the purposive esotropia (ie, during the binocular IN phase) there is not likely to be a head turn. However, post-esotropia (in both the NBS Types I and II), the fixating eye is the purposively esotropic eye, necessitating a head turn in the opposite direction. In Type II, because of the Alexander’s law variation of FMN, a large head turn is common.

Individuals with FMNS often fixate stationary targets with their adducting eye; this strategy is also used during smooth pursuit. Initially, as the target crosses the midline, the right eye takes up pursuit to the left, while the left eye remains esotropic, albeit moving with the same pursuit velocity. As the now rightward moving target again crosses the midline, the left eye takes up smooth pursuit while the right eye remains esotropic.

Without accurate, monocularly calibrated eye-movement data the relative positions of each eye, the determination of the fixating eye, and differentiating between INS with a latent component, FMNS, or the NBS would not be possible. In Type I, the extended foveation periods will increase the NAFX over that measured during binocular fixation. The eye-movement data are indistinguishable from convergence damping in a binocular INS patient with the exception that only one eye is adducted in the NBS. In Type II, the very low-amplitude FMN will also yield higher values of the NAFX. During the purposive esotropia, target foveation is accurate, the NAFX is high, and visual acuity is at its maximum for the NBS patient regardless of whether it is Type I or II. The built-in efference copy of motor commands in the OMS prevents the perception of oscillopsia in either IN or FMN. Thus, patients with the NBS do not normally experience oscillopsia regardless of whether it is Type I or II.

One must first determine whether the patient is Type I or II. The therapeutic choice will depend on the patient’s individual nystagmus type(s) and characteristics. The medical and surgical therapies applied to IN and FMN utilize the respective characteristics of the nystagmus. In much the same manner, INS patients turn their heads to exploit the gaze-angle null of IN and FMNS patients do the same to exploit the Alexander’s law variation of their FMN. Patients with NBS also apply these same “therapeutic” maneuvers. Head turns, for whichever type of nystagmus they are used may be cosmetically unattractive or even may lead to neck problems when severe. However, head turns are not defects associated with the INS, FMNS, or NBS, but rather, constitute purposive and therapeutic patient-administered “therapy.” Successful amelioration of a head turn can only occur if its advantages, vis a vis better visual function, are otherwise achieved (eg, surgically moving the IN null to primary position or inducing convergence that both damps and broadens the IN null).

If the patient fixates with the straight eye in Type I NBS, there will be no head turn. If the newly esotropic eye is used, a head turn will be necessary and will depend on the amount of esotropia employed by the patient. The amount of Alexander’s law variation with gaze angle is idiosyncratic. Therefore, the presence or amount of head turn will also be idiosyncratic and could affect the optimal therapy. The surgery for NBS is aimed at reducing the head turn required after the patient utilizes esotropia to damp (Type I) or damp and change (Type II) the nystagmus. Because the patients are able to fuse and only manifest esotropia when they desire, standard strabismus-surgery approaches do not apply. If they have strong fusion reflexes, perhaps a bimedial recession procedure would produce the required damping for Type I NBS patients in the same manner as in binocular INS patients. For Type II NBS patients, if the INS damping produced by realigning the eyes is sufficient, perhaps the switch to FMN would not occur; if the underlying mechanism in Type II is not a switch to FMN but a manifestation of a low-amplitude FMN that coexisted with a more prominent INS, the FMN would then also appear. If fixation remains with the stationary eye as the other is deviated inward, no head turn would be necessary and surgery would be problematic. If, however, fixation is taken up by the deviated eye (the most common scenario), a head turn results, and surgery can be tailored to reduce that turn.

In the NBS, it is possible that the bimedial recession procedure could be therapeutically beneficial because the purposive esotropia is different from ordinary strabismus; the latter is not under conscious control. Although it has been suggested that bimedial recession may help those with NBS, published studies are lacking. Similarly, the Faden operation may benefit those with NBS. Simple surgical correction in NBS has not proven successful.

Spasmus Nutans Syndrome

The spasmus nutans syndrome (SNS)¹ includes ocular oscillations, head nodding, and anomalous head positions that begins in infancy (usually between 4 and 18 months of age) and disappears clinically in childhood (usually before 3 years or age). The nystagmus is generally bilateral (but can differ in each eye and may even be strictly monocular), and it oscillates in horizontal, torsional, or vertical directions. The average duration of SNS is 12 to 24 months; rarely, it lasts a number of years. SNS has been confused with other entities, but eye-movement recordings allow accurate differentiation (eg, FMN or uniocular pendular nystagmus).²⁸ Prior to subsequent ocular motility studies, diagnosis was delayed until the clinical symptoms of nystagmus and anomalous head posture resolved.

SNS appears as a high-frequency, asymmetric, disjugate ocular oscillation. It is usually horizontal in direction but may also be vertical or torsional. It is often described as an intermittent nystagmus that is asymmetrical in appearance and occasionally monocular. A key eye-movement recording observation is the variable phase difference between the 2 eyes, which is reflected clinically as an asymmetry in the oscillations between the 2 eyes. On lateral gaze, the dissociation may increase, with nystagmus of the abducting eye predominating. Some case series suggest an increased prevalence of esotropia in SNS. Gottlob et al found a high incidence of esotropia, latent nystagmus, dissociated vertical divergence, and amblyopia in children with SNS.⁴³ Conversely, rare patients with infantile esotropia display horizontal or vertical head oscillations that resolve following surgical realignment of the eyes. In contradistinction to INS, visual acuity is minimally affected in SNS.

The nystagmus waveform in SNS is a dissociated pendular nystagmus, and this dissociation may be so great that the nystagmus is uniocular.^{44; 45} Ages of onset of the 7 patients studied ranged from birth to 14 months; 5 had head nodding. The dissociated nystagmus is usually of a higher frequency than INS nystagmus, and the result can be disjugate, conjugate, or uniocular. Early reports considered SNS to be pathogenically related to diverse causes that included light deprivation, dietary factors, season, rickets, epilepsy, auto-arousal, and poor socio-economic conditions. We hypothesize that SNS reflects a yoking abnormality, perhaps due to delayed development. Recordings show that SNS nystagmus may not disappear completely but may recede to a subclinical level; neither INS nor FMNS disappears with age.⁴⁵

The nystagmus of the SNS tends to be asymmetric in the two eyes, to vary in different directions of gaze, and to be rapid and of small amplitude. The interocular phase differences can appear from minute-to-minute and during the child's development. The pendular oscillations of each eye may be phase-locked, slightly out of phase, or even totally out of phase, and this can vary during the course of the recording anywhere from pure conjugacy to pure disjugacy (0–180° phase shift). The phase variation usually varies from minute to minute. Distinguishing this variable phase relationship between the pendular oscillations of both eyes requires DC-coupled, high-bandwidth recordings of both eyes simultaneously.

Head nodding in SNS is curious. It is inconstant and irregular and can be horizontal or vertical, or both. The head nodding associated with SNS is a combination of vertical head nodding, together with a lateral shaking of the head in an unpredictable pattern.^46-48 The head nodding is of lower frequency than the nystagmus and becomes prominent when the child attempts to inspect something of interest. It disappears during sleep but may persist when the child is lying down. Since some children with INS also have head nodding, this finding alone cannot be used to confirm the diagnosis of SNS in the child with nystagmus. Studies of quantitative head-and eye-movement recordings indicate that the head movement may, using the normal VOR, actually serve to abolish the eye movements.⁴⁹ In some patients, it may be only compensatory with suppression of the VOR. Compare this to INS, where the head oscillation is an extension of the nystagmus and the VOR is normal. Gresty et al examined patients with SNS in whom head nodding abolished the nystagmus, and a normal VOR stabilized the eyes during head movements.⁵⁰ They demonstrated with eye-movement recordings that the head nodding in SNS is an adaptive behavior that serves to improve visual acuity by suppressing the nystagmus, rather than a separate pathological phenomenon. Eye-movement recordings from these patients demonstrated that the head nodding in SNS functions to abolish the nystagmus through some mechanism independent of the vestibulo-ocular response. There is now general agreement that head nodding in SNS is compensatory.

Because the VOR of these patients is normal, by willful shaking of the head, the nystagmus is switched off and the eyes become stable in space because of a good VOR. A patient may have convergence nystagmus, one eye going left, the other eye going right at the same time (180° out of phase), while the head is still. When the patient starts shaking his/her head, the nystagmus stops and, owing to the normal VOR, the eyes begin moving conjugately equally and oppositely to the head, so gaze remains constant.

“Cultured” Clinical Pearl-Based the observation that head nodding is compensatory in the SNS, if further research on the eye movements of the “SN-like” nystagmus associated with brain stem gliomas demonstrates that no head nodding is exhibited by these patients, the presence of deliberate, compensatory head nodding is an indication of SNS and is benign.

The head tilt in SNS is a variable finding that is present in less than half of cases. Although early authors stated that the head nodding was the first sign of SNS to appear and the last to resolve, it is now generally agreed that the nystagmus is the most constant feature of SNS and that it probably precedes the head nodding, although the head nodding may be the abnormality that first attracts attention. Weissman et al.⁴⁵ found persistence of the nystagmus in some of their patients, and Gottlob et al found persistence of nystagmus using eye movement recordings in all patients who had clinical resolution of the condition, suggesting that the nystagmus diminishes to a subclinical level but does not entirely resolve.⁴³

In patients with “SN-like” nystagmus, accurate diagnosis is the most important factor. Therefore, until definitive eye-movement-based criteria are identified to differentiate SNS from other, more neurologically problematic conditions, imaging is a necessity. Because SNS is benign, once it is definitively diagnosed, SNS requires no treatment.

Differential Diagnosis of Nystagmus In Infancy and Childhood

Fortunately, INS is easily identified by any of several pathognomonic waveforms. Unfortunately, they cannot be identified by clinical observation; eye-movement data are required. Until eye-movement recordings become part of the standard workup of INS patients, arriving at the correct diagnosis will remain problematic and dependent on the specific combination of a number of factors, including peak visual acuity (VApk), high-acuity, gaze-angle range (HAgar) , VA far vs. near, VApk gaze position, and strabismus.⁵¹

Benign

These types of nystagmus exhibit pathognomonic waveforms (INS) and characteristics (INS, FMNS, NBS, and SNS) that allow both differential and reliable diagnosis of each.

INS

Familiarity with the clinical features of INS is essential. It is an ocular motor disorder with the hypothesized etiology of undamped smooth pursuit. It presents at birth or early infancy and is clinically characterized by involuntary oscillations of the eyes. Estimations of the incidence of INS vary enormously from 1 in 350 to 1 in 20,000, although the generally quoted estimated incidence is 1 in 6,550 or .015%. INS is usually horizontal, with a small torsional component and may (rarely) have a vertical component. The intensity of INS increases on lateral gaze and becomes right beating in right gaze and left beating in left gaze. The fact that INS "disobeys" Alexander's law under binocular conditions (which states that, in peripheral vestibular nystagmus, the direction of the nystagmus increases in the direction of the fast phase and decreases but never reverses in the direction of the slow phase) is often useful in distinguishing it from horizontal peripheral vestibular nystagmus. Other clinical characteristics of INS, with variable association, include: INS remains horizontal in up gaze (in contrast to acquired and/or vestibular nystagmus, which changes direction in vertical gaze); INS increases intensity with fixation attempt or stress and decreases with sleep or inattention; INS had variable intensity in different positions of gaze (usually about a null position); INS changes direction in different positions of gaze (about a neutral position); INS has decreased intensity (damps) with convergence; INS exhibits anomalous head posturing; INS is associated with strabismus; and INS is associated with an increased incidence of significant refractive errors. This history is useful in further distinguishing the nystagmus from peripheral vestibular nystagmus, which becomes worse with occlusion and is damped by fixation. To confirm this observation, the examiner can observe one optic disc with the direct ophthalmoscope while periodically occluding the other eye. Increased nystagmus intensity with occlusion suggests a peripheral vestibular nystagmus, whereas no change, a direction reversal, or a decrease in nystagmus intensity suggests INS or FMNS. Anxiety or fatigue will also increase the INS intensity and thereby degrade visual acuity. When evaluating an infant or child with INS for the first time, historical points suggest afferent visual pathway dysfunction. If the child's eyes are light sensitive this suggests the presence of a congenital retinal dystrophy, degeneration, or albinism. If the child sees better in daytime or at night, this suggests congenital stationary night blindness or a rod-cone dystrophy. Figure 1 illustrates a clinical algorithm that could lead to a putative diagnosis of the type of nystagmus present in a patient.

Figure 1. Flowchart demonstrating how clinical observations and tests may be used to arrive at a nystagmus diagnosis. Note that, unlike when using waveform analysis, all paths do not lead to a definitive diagnosis and there is no reliable path to acquired nystagmus. AN, acquired nystagmus; FMNS, fusion maldevelopment nystagmus syndrome; INS, infantile nystagmus syndrome; NBS, nystagmus blockage syndrome; SNS, the spasmus nutans syndrome; VN, vestibular nystagmus.
[From Figure D.3, Hertle, R. W. & Dell'Osso, L. F. (2013) Nystagmus in Infancy and Childhood]

The presence or absence of an underlying visual sensory deficit does not affect the time of onset of INS. Infants are often initially evaluated in the first to third months of life when irregular eye movements are noted. When INS first appears, it is often arrhythmic and intermittent, consisting of a series of irregular horizontal and oblique deviations of the eyes from side to side. At this stage, the erratic eye movements may simulate opsoclonus.

INS often occurs in association with congenital or early onset (first 6 months of life) acquired defects in the visual sensory system (eg, systemic and ocular albinism, achromatopsia, aniridia, congenital retinal dystrophies and degenerations, visual cortex anomalies, and congenital cataracts, glaucoma, and corneal diseases). Children with this condition frequently present with a head turn, which is used to maintain the eyes in the position of gaze of the null point (point of minimum nystagmus). This is particularly prominent when the child is concentrating on a distant object, since this form of nystagmus tends to worsen with attempted fixation. The head turn is an attempt to improve visual function under these conditions. In most individuals with INS, the head position corresponds roughly with the minimal intensity zone of the nystagmus. A clinical algorithm to aid in the determination of the etiology of anomalous head turns in the presence of strabismus and nystagmus is shown in Figure 2.

When the angle of the null zone exceeds 15°, however, the angle of the head turn may fall short of the null zone. In some children, the anomalous head position appears to be dictated by the velocity distribution of the slow phase (ie, the percentage of time that the slow phase is less than or equal to 10° per second) and the nystagmus beat direction (which can be influenced both by the prior position of gaze and by the length of time a subject has maintained a fixed gaze position). Head oscillations are common in INS, but are not used as the strategy to improve vision, except in those rare patients with abnormal gain of their vestibulo-ocular reflex.

Although INS is a lifelong condition, it may remain variable over time. This includes minute-to-minute variability and long-term changes associated with age and other ocular and systemic conditions. There are no good long-term, multi-subject studies (greater than 10 years) that characterize the aging process and its effects on INS, but we do know that changes can occur in the INS oscillation as a result of aging, medications, and degenerative and neurological illnesses. This concept of INS as a dynamic disease process may be useful when evaluating and caring for patients with this condition over their lifetime. However, eye-movement data taken over decades in some subjects with INS (one, over more than a 55-year span) showed no overall changes in the important waveform characteristics governing visual function.

Figure 2. A clinical algorithm to aid in diagnosing the etiology of an anomalous head posture in the presence of strabismus and nystagmus. The first step is to reposition the patient’s head to the neutral position and observe the consequences on the strabismus or the nystagmus (Step 1) and whether the change improves or worsens the condition (Step 2). The major clinical differentiation occurs at this step. If the nystagmus is worse then differentiating between a “gaze” null due to INS and an “adduction” null due to FMNS can be made by looking at the fixing eye’s (Step 3) and the opposite non-fixing eye’s (Step 4) effect on the oscillation in aBDuction and aDDuction. [From Figure 5.1, Hertle, R. W. & Dell'Osso, L. F. (2013) Nystagmus in Infancy and Childhood. Current Concepts in Mechanisms, Diagnoses, and Management. Oxford University Press: Oxford.]

FMNS

There is always associated strabismus and ocular motor recordings show four types of slow phases with jerk in direction of fixing eye. The oscillations appear conjugate, horizontal, uniplanar and there are usually no associated sensory system deficits (eg, albinism, achromatopsia). They may change with exaggerated convergence resulting in a head posture associated with fixing eye in adduction. Waveform is also the diagnostic criterion for FMNS; recordings show a linear or decreasing-velocity exponential slow phase. All patients with FMNS have strabismus. Included in the definition of strabismus, is latent strabismus (ie, the phoria resulting when you cover an eye). Thus, FMNS includes a pure latent form, where the eyes are straight with no nystagmus when both eyes are open and upon occlusion of one eye, an eso- or exophoria develops followed by manifest FMN in both eyes. The, more common, manifest form of FMN is present with both eyes open and mimics the latent form exactly if it is bidirectional. When it is unidirectional and the patient fixates with one eye, there will be no nystagmus and the other eye will be esotropic but when they fixate with the other eye and the formerly fixating eye is esotropic, they will have FMN. The latent form of FMN (ie, no nystagmus with both eyes open) is rare. If you occlude the left eye and the right eye is fixating, jerk-right FMN with linear or decreasing-velocity slow phases results and vice-versa.

The small group of patients with both INS and FMNS present a diagnostic nightmare. Some have mostly INS (designated, “INS/FMNS”) and their waveforms are any of the INS waveforms (ie, pendular or increasing-velocity slow phases, see Table) and other waveforms called "dual jerk" or “dual pendular”. The latter are waveforms where a low-amplitude, high-frequency pendular nystagmus is superimposed on either a decreasing-velocity slow phase jerk waveform (dual jerk) or a slow pendular waveform (dual pendular). They do not exhibit the pure FMNS waveform (ie, decreasing-velocity slow phases); therefore, INS is predominant. The other group has mostly FMNS (FMNS/INS) and their waveforms are FMNS and dual jerk FMN. There are some who have INS and FMNS equally. At various times they exhibit the INS waveform, FMNS waveform, or the dual jerk waveform. A linear slow phase is not diagnostic of either INS or FMNS. When a pendular waveform is superimposed on a jerk waveform and the slow phase is accelerating, it is a dual jerk IN; if the slow phase is decelerating, it is a dual jerk FMN. One has to carefully determine what is happening to the axis of the pendular slow phase (ie, whether it is decelerating or accelerating) to properly categorize the nystagmus. This small but difficult group of patients must be recorded for accurate diagnosis. Distinguishing the INS and FMNS waveforms from the combination waveforms requires DC-coupled, high-bandwidth recordings of both eyes simultaneously.

NBS

The NBS is the source of some misunderstanding. Individuals with NBS use a convergence-like movement to damp their INS while viewing a distant target. Thus, the waveforms in NBS are those of INS when the patient is looking in the distance and the eyes are straight.³⁹ With the imposition of the purposive esotropia (this is not a strabismus that occurs transiently but one that the patient either willfully or reflexively imposes because they have found that acuity is better under that condition) the waveform can either become a damped INS waveform (Type I NBS) or a small-amplitude FMNS (Type II NBS). Thus, the NBS implies ocular motor deficits capable of producing both IN and FMN waveforms (see Table). Even though the FMNS waveform is usually unsuitable for good acuity (because there are no long, post-saccadic foveation periods), it is of very low amplitude and acuity increases. Thus, there are two types of blockage syndrome. NBS is often misdiagnosed in patients with FMNS and alternating fixation who place their fixating eye in adduction to reduce their FMNS. Since they do not have INS, they do not have NBS. Distinguishing the INS from the FMNS waveforms and the purposive esotropia of one eye requires DC-coupled, high-bandwidth recordings of both eyes simultaneously.

Table 1. Comparative Characteristics of Infantile and Fusion Maldevelopment Nystagmus

[From Table 5.3, Hertle, R. W. & Dell'Osso, L. F. (2013) Nystagmus in Infancy and Childhood. Current Concepts in Mechanisms, Diagnoses, and Management. Oxford University Press: Oxford.]

Summarizing, within the different types of benign infantile nystagmus, there is a large category of pure INS, a significant category of pure FMNS and a small category that is a mixture of the two; there are also individuals with the nystagmus blockage syndrome (NBS) or spasmus nutans syndrome (SNS). All are easily diagnosed with the aid of ocular motility recordings and just as easily misdiagnosed without them. There are twelve INS waveforms, two mixed INS waveforms (dual jerk and dual pendular IN), two FMNS waveforms and one mixed FMNS waveform (dual jerk FMN). If a patient walks into your office with wiggling eyes and you wish to guess what they have before you record them, your best guess would be INS. A large percentage (80%) will be INS and 15% FMNS with only a small percentage, mixtures. If you could consider just INS patients, 94% will be pure INS and only 6% a mixture. If you could restrict the population to FMNS patients, 3/4 of them will have only FMNS, but many will have mixtures. Thus, more patients with predominantly FMNS will also have some INS than patients with predominantly INS having FMNS.

SNS

The independent oscillation of one or both eyes is a characteristic of SNS. When both are oscillating, there is a time-variable phase relationship between the nystagmus in each eye. SNS may appear at or after birth and usually, but not necessarily, ceases clinically (diminishes) by the age of three. The nystagmus is pendular, usually monocular or disconjugate and is commonly accompanied by a head oscillation. Reports of "spasmus nutans" accompanying neurological disease (eg, craniopharyngioma, optic nerve and chiasmal glioma^52-54 or cerebrocerebellar degeneration⁵⁵) have not all included eye-movement recordings to prove that the nystagmus was the same as that of spasmus nutans. Although the nystagmus may clinically resemble that recorded in SNS, until a proper study comparing the actual waveforms of SNS with those recorded in children with known neurological disease, they should not be presumed to be identical. The terminology, “SNS” should be reserved for the specific, benign nystagmus that has been documented (see above) and should not also be used to describe the nystagmus accompanying neurological disease, even if the nystagmus itself is subsequently found to be identical. Rather, a descriptive name for the nystagmus should be used for the latter. Other signs are needed to distinguish true SNS from similar looking nystagmus associated with central nervous system disease.⁵⁶

Strabismus

The incidence of strabismus in general population has been reported to be between 0.80% and 6.0%. Those eye care professionals who care for patients with strabismus are as likely, if not more so, as any in health care to be confronted with the disorders of nystagmus and other ocular oscillations.

Estimates of the prevalence of strabismus in INS range from 16% to 50%. Strabismus is essential for FMNS but incidental to INS. Although FMNS is intrinsically more likely to be associated with strabismus than is INS, the greater incidence of INS means that any given patient with strabismus will still be more likely to have INS (53%) than FMNS (35%). The presence and nature of an underlying sensory visual disorder seems to influence the likelihood of associated strabismus.

FMNS implies strabismus but the converse is not true; strabismus does not imply FMNS (50% have no nystagmus at all), but if an individual has FMNS, he also has strabismus. Even if it is not evident clinically (it may be a micro strabismus that can be recorded) it is there. Distinguishing the FMNS waveform and the tropia of the non-fixating eye requires DC-coupled, high-bandwidth recordings of both eyes simultaneously. In logical terminology, strabismus is a necessary (but not sufficient) condition for FMNS (ie, all individuals with FMNS have strabismus but not all with strabismus have FMNS). Summarizing, INS can occur with or without strabismus; all FMNS patients have strabismus.

A variable strabismus is a necessary but not sufficient condition for the NBS since a purposive esotropia is required to damp the IN and/or transform it into FMN. That is, all patients with NBS have strabismus but clearly all with strabismus do not have NBS.

There may be a high incidence of esotropia, latent nystagmus, dissociated vertical divergence, and amblyopia in children with SNS. Conversely, rare patients with infantile esotropia display horizontal or vertical head oscillations that resolve following surgical realignment of the eyes.

Symptomatic

The pathognomonic waveforms of INS allow us to more accurately diagnose patients with acquired nystagmus by eliminating those with INS. The same is not true for nystagmus types associated with neurological disease. They include: several types of vestibular nystagmus; gaze-holding nystagmus; visual loss nystagmus (chiasmal and pre- and post-chiasmal); pendular nystagmus associated with central myelin disease (tremor, myoclonus, and pendular vergence nystagmus); convergence and convergence-evoked nystagmus; upbeat and downbeat nystagmus; torsional and see-saw nystagmus; and lid nystagmus. A thorough discussion of these or the many types of saccadic intrusions and oscillations is beyond the scope of this chapter; the reader is referred to two excellent sources for this material.^{2; 6}

Examination Procedures

Before discussing clinical office procedures (or even treatments for INS) it is necessary to fully understand the primary deficit in visual function faced by the nystagmus patient. It is not, as is commonly believed, a lowered peak best-corrected visual acuity (BCVApk or VApk). The primary deficit in real-world visual function caused by INS is the dramatic reduction in high-acuity, gaze-angle range, HAgar (ie, the acuity present when looking at targets in the visual field that are lateral to the angle with the highest BCVA and are within 10% of that peak value).

Figure 3 (Pre) illustrates the major visual deficit in INS (top) and its alleviation following T&R surgery (Post). It should be stressed that this photograph does not depict the falling off of visual acuity as distance from the fovea increases. Rather, it illustrates the diminished foveal visual acuity at different gaze angles lateral to that with the peak acuity (here shown as primary position). Prior to T&R therapy, the identification of individuals to both sides is very difficult.

It should be noted that the therapeutic improvements to that INS-induced acuity reduction may reach 1600%.⁵⁷ Decrements from VApk are dependent solely on reductions from peak INS foveation quality. They may be measured by the NAFX at different gaze angles. The resulting decrement is expressed as the longest foveation domain, LFD, which is defined as the range of gaze angles whose NAFX is within 10% of NAFXpk; the LFD is independent of any associated visual deficit. During clinical examination, HAgar can be determined directly from BCVA measurements made at different gaze angles; it is the clinical and numerical equivalent of the INS-waveform-determined LFD (HAgar=LFD). Because the primary, real-world deficit in INS is a low HAgar, measuring BCVA only at the patient’s best gaze angle, either pre- or post-therapy, can neither fully describe the total INS deficit nor provide information necessary to evaluate its possible amelioration and improvements in visual function.

Figure 3. Illustrations of the clarity of portions of the visual field in an INS patient with a narrow, central, null (sharp, central NAFX peak) before (Pre) and after (Post) tenotomy and reattachment surgery. This procedure can also be used in cases where there are no nulls.

Another visual function deficit not measured in the office but amenable to therapy is foveation time (ie, target acquisition time).^{58; 59} This is important in every-day activities like, quickly scanning a room for familiar faces, reading street and traffic signs while driving, and all sports.

Figures 4 – 6 show the plots of the NAFX vs. gaze angle for cases with high (Fig. 4, solid), low (Fig. 5, solid) and mid-range (Fig. 6, solid) peak visual acuities. In Figures 5 and 6, the solid grey curves show possible high and low curves for NAFX (Fig. 5) and high curve for NAFX (Fig. 6) that could accompany patients with these respective visual acuity profiles; these more complex relationships in patients illustrated by Figures 5 and 6 make determination of therapeutic improvements more difficult (see Section 7). Note that the solid curve of Figure 4 corresponds to the visual picture of Figure 3 (Pre) and the dashed curve corresponds to Figure 3 (Post).

Figure 4. Measured VA vs. Gaze Angle plots for patients with high VApk = NAFXpk in or near primary position and low HAgar = LFD including: VAf = VAn, ± strabismus and VAf < VAn, no strabismus. In Figure legends 4 – 6, f = far, and n = near, T&R = tenotomy and reattachment, BMR = bimedial rectus recessions, and BOPr = base-out prism. Solid = pre-therapy curve, and dashed and dot-dashed = post-therapy T&R and BMR/BOPr curves respectively, Because Figures 5 and 6 include different possible combinations of INS characteristics, they are more complex than the simple plot from an individual patient, which, like Figure 4, would only contain that patient’s personal pre- and post-therapy acuities.

Figure 5. Measured VA vs. Gaze Angle plots for patients with low VApk ≤ NAFXpk in or near primary position, low HAgar = LFD, and VAf = VAn ±strabismus including the following possibilities: a) low NAFXpk at near and far; b) mid-range NAFXpk at near and far; or c) high NAFXpk at near and far. For clarity in the more complex cases of Figures 5 and 6, individual VA measurements were deemphasized in favor of the 2^nd-order trend lines for some of the possible conditions. Pre-therapy NAFX curves of equal LFD values and post-therapy VA curves for conditions b) and c) are shown in grey. Solid = pre-therapy VA and possible NAFX curves and dashed = post-therapy T&R curves. Here and in Figure 6, h = high NAFXpk (>0.6), m = mid-range NAFXpk (0.25 ≤ NAFXpk ≤ 0.6), and l = low NAFXpk (<0.25).

Figure 6. Measured VA vs. Gaze Angle plots for patients with mid-range VApk ≤ NAFXpk in or near primary position, high HAgar = LFD, including: VAf = VAn, ± strabismus or VAf < VAn, no strabismus. The following possibilities are also included: b) mid-range NAFXpk at far; or c) high NAFXpk at near and far. Pre-therapy NAFX curves of equal LFD values and post-therapy VA curves for conditions b) and c) are shown in grey. Solid = pre-therapy VA and possible NAFX curves and dashed and dot-dashed = post-therapy T&R and BMR±T&R/BOPr curves respectively.

Clinical

To adequately document their visual function deficits, all patients with nystagmus (INS or FMNS) must undergo measurement of their gaze-angle BCVA OU; that is, measurement of BCVA at different gaze angles. That will document the major symptomatic deficit in INS, the inability to maintain their BCVA when looking to the right or left of center (or of their idiosyncratic “null” angle). In cases of INS with a latent component, gaze-angle BCVA should also be measured OD and OS. Methods of measuring BCVA at different gaze angles in the clinical office may be found elsewhere.^{2; 51} Only by making gaze-angle acuity measurements the standard of care for nystagmus patients, can adequate data be collected from which prospective therapies may be chosen and their visual function improvements determined. For binocular (ie, no strabismus) INS patients whose nystagmus damps with convergence, the gaze-angle BCVA should be measured OU with 7 PD BO prisms and -1.00S added OU to their refraction (the -1.00S should not be added for presbyopic patients). In fact, once monocular refraction is determined and monocular acuity measured, the prisms and -1.00S should be added. Thereafter, one should not attempt to measure monocular acuity unless the prisms and -1.00S are removed. Doing so will yield extraneous, lower values of acuity since the prisms will effectively shift the targets to lateral gaze (right gaze for OS viewing and left gaze for OD).

Note: Although greater INS damping with higher foveation-period quality may be achieved with higher values of convergence,⁶⁰ the amount of prism prescribed should be limited to 7-9 PD BO OU for distance acuity. That will provide most of the improvement in distance BCVA available to the patient while still allowing further convergence for middle-distance and near targets (see Section 7).

In addition to gaze-angle acuity, near acuity, stereo acuity, and ocular alignment (tropias and phorias) should be determined. Contrast sensitivity, color vision, time-dependent acuity, and visual field testing may also be helpful. Finally, a thorough examination of the retina, including the foveal area and optic nerve should be performed.

Electrophysiological

The visual evoked potential (VEP) is an evoked electrophysiological potential that can be extracted, using signal averaging, from the electro-encephalographic activity recorded at the scalp. The VEP can provide important diagnostic information regarding the functional integrity of the visual system. Because chiasmal and retrochiasmal diseases may be missed using a single channel, three channels using the midline and two lateral active electrodes are suggested for INS patients. Pattern reversal is the preferred technique for most clinical purposes but is less reliable in patients with unstable fixation or nystagmus. The results of pattern-reversal stimuli are less variable in waveform and timing than the results elicited by other stimuli. The pattern-onset/offset technique can be more useful in patients with nystagmus, and the flash VEP is particularly useful when optical factors or poor cooperation make the use of pattern stimulation unreliable.

Flash VEPs are much more variable across subjects than pattern responses but show little interocular asymmetry. They may be useful in patients who are unable or unwilling to cooperate for pattern VEPs, and when optical factors such as media opacities prevent the valid use of pattern stimuli. The visual evoked potential to flash stimulation consists of a series of negative and positive waves. The earliest detectable response has a peak latency of approximately 30 ms post-stimulus and components are recordable with peak latencies of up to 300 ms. Peaks are designated as negative and positive in a numerical sequence. This nomenclature is recommended to automatically differentiate the flash VEP from the pattern reversal VEP. For the flash VEP, the most robust components are the N2 and P2 peaks. Measurements of P2 amplitude should be made from the positive P2 peak at around 120ms to the preceding N2 negative peak at around 90 ms.

VEPs should be recorded when the infant or child is in an attentive behavioral state. Direct interaction with the child can help maintain attention and fixation, and two testers are beneficial, one to work with the child and the other to control data acquisition. The order of stimulus presentation also should be flexible and selected to ensure that responses most critical to the diagnostic question are obtained within an individual child’s attention span. Binocular pattern stimulation, which facilitates attention and fixation, may be useful to evaluate overall visual function. Monocular testing to at least one stimulus is desirable to assess the function of each eye. It is particularly important to obtain replicate responses from children to assure that the response measured is a reliable signal and not an artifact. As for adults, additional channels of recording may be important for diagnosis of chiasmal and post-chiasmal dysfunction. Pattern-onset/offset stimuli can be helpful in gaining attention of children and are usually more robust in cases of nystagmus, but the waveform components change with age.

Electroretinography (ERG) evaluation of children with nystagmus has both diagnostic and prognostic value. In patients with nystagmus and a normal clinical examination alone, a true diagnosis of INS without associated sensory system deficits cannot be inferred. Other visual system diseases associated with nystagmus in infancy include: Leber amaurosis, delayed visual maturation, albinism, optic nerve hypoplasia, achromatopsia, and X-linked congenital stationary night blindness. ERG’s have been shown to help in distinguishing between these conditions.

The global or full-field ERG is a mass electrical response of the retina to photic stimulation and is used worldwide to assess the status of the retina in eye diseases in human patients and in laboratory animals used as models of retinal disease. The basic method of recording the electrical response known as the global or full-field ERG is by stimulating the eye with a bright light source such as a flash produced by a strobe lamp. The intense flash of light elicits a biphasic waveform recordable at the cornea. The two components that are most often measured are the a- and b-waves. The a-wave is the first large negative component, followed by the b-wave which is corneal positive and usually larger in amplitude. The amplitude from the baseline to the negative trough of the a-wave, the amplitude of the b-wave measured from the trough of the a-wave to the following peak of the b-wave, the time from flash onset to the trough of the a-wave, and the time from flash onset to the peak of the b-wave are the most common measures. These times, reflecting peak latency, are referred to as "implicit times." The a-wave, sometimes called the "late receptor potential," reflects the general physiological health of the photoreceptors in the outer retina. In contrast, the b-wave reflects the health of the inner layers of the retina, including the “on” bipolar cells and the Muller cells. Two other waveforms that are sometimes recorded for clinical purposes are the c-wave originating in the pigment epithelium and the d-wave indicating activity of the “off” bipolar cells. There are also wavelets that occur on the rising phase of the b-wave known as oscillatory potentials, thought to reflect activity in amacrine cells.

The ERG of a normal full-term infant looks similar to a mature ERG. The ERG attains peak amplitude in adolescence and slowly declines in amplitude throughout life. The ERG can be recorded several ways. The pupil is usually dilated. There are a number of corneal ERG electrodes that are in common use. Some are speculum structures that hold the eye open and have a contact lens with a wire ring that "floats" on the cornea supported by a small spring. Some versions use carbon, wire, or gold foil to record electrical activity. There are also cotton wick electrodes. There are yet other simpler ERG recording devices using gold Mylar tape that can be inserted between the lower lid and sclera/cornea. Most electrodes are monopolar, ie, they are referred to another electrode site, most commonly on the forehead. There are also several methods of stimulating the eye. Most laboratories use a Ganzfeld (globe) with a chin rest and fixation points. The Ganzfeld allows the best control of background illumination and stimulus flash intensity. Either strobe lamp or Ganzfeld methods of flash presentation can be used to record the ERG following a single flash or to average responses to several flashes with the aid of a computer.

Infants up to about 6 months years of age can usually be tested without sedation by the parent holding them bundled in a blanket. Most of our ERG testing is performed as part of a more extensive exam under anesthesia. Anesthesia affects the ERG varying with type and depth of anesthesia. Some anesthetics can attenuate b-wave amplitude as much as 50%. Light levels of anesthesia have little affect and most anesthetics do not usually affect a-waves or implicit times.

Most disorders of the retina are detected by an attenuation of amplitude. Implicit times, of both a- and b-waves are also affected in some conditions. The patient should be dark-adapted for about 30 minutes then the electrodes attached using dim red illumination. The ERG should be recorded using single, scotopic white flashes. Afterwards, turn on moderately high background illumination for about 10 minutes and record ERGs using 30-Hertz flicker and bright white flashes.

Imaging

Optical coherence tomography (OCT) is analogous to ultrasound except that near-infrared light waves instead of acoustic waves are used to measure distances of specific structures. OCT depends on optical ranging; in other words, shining a beam of light onto the object, then recording the echo time delay of light measure distances. Since the velocity of light is so high, it is not possible to directly measure the echo time delay of reflections; therefore, a technique known as low-coherence interferometry compares reflected light from the eye to that reflected from a reference path of known length. Different internal structures produce different time delays, and scanning the incident optical beam can generate cross-sectional images of the structures. These two-dimensional scans are then displayed in a color scale where ‘warm’ colors (red to white) represent areas of high optical reflectivity, and ‘cool’ colors (blue to black) represent areas of low reflectivity. The current technology permits excellent resolution and has been used to help diagnose and guide treatment for macular edema, macular and lamellar holes, epiretinal membranes and pseudoholes, central serous chorioretinopathy, age-related macular degeneration, glaucoma, and optic disc lesions.

Evaluation of the retina in patients with nystagmus can occasionally be difficult because of the constant motion of the eye. Eye motion has impeded the implementation of TD-OCT for diagnostic purposes in the nystagmus population. In addition to eye motion, compromised vision and the shortened attention span of children often compound the challenge of quality TD-OCT macular image acquisition.

The new SD-OCT has allowed a more reliable study of patients with nystagmus. Its use for macular imaging has been reported in patients with albinism and nystagmus. It has demonstrated several features in this population, including the: persistence of an abnormal, reflective nerve fiber layer band; persistence of multiple inner retinal layers; loss of the normal thickened photoreceptor nerve layer; and increased reflectivity of the choroid due to decreased pigmentation. Most external retinal layers, specifically the external limiting membrane, photoreceptor inner segment layer, and photoreceptor outer segment layer, appear normal, in agreement with histopathological studies and less detailed TD-OCT reported findings of either absent or rudimentary foveal pits in oculocutaneous albinism. Observations of the retinal morphology in the nystagmus population reveal that all cases of clinical foveal hypoplasia demonstrate a persistence of the outer plexiform layer, inner nuclear layer, inner plexiform layer, ganglion cell layer, and nerve fiber layer.⁶¹

Protocols are currently being designed to align images with scanning laser ophthalmoscope fundus landmarks, such as blood vessels, to remove eye motion artifact caused by micro saccades. Applying this software to the extreme case of eye motion, nystagmus, and incorporating it into a technology that is already available to clinicians would be an exciting and powerful application of the SD-OCT. Because the nystagmus population can also fall victim to common ocular diseases, such as diabetes, glaucoma, and macular degeneration, the use of SD-OCT, especially with software that may reduce eye motion and help recover 3-dimensional spatial integrity, would be an important diagnostic and management tool.

Additional information can be found in Chapter 6 of Nystagmus in Infancy and Childhood. Current Concepts in Mechanisms, Diagnoses, and Management, ² which contains multiple excellent references regarding the VEP, ERG, and OCT.

Fundus photography in the patients with nystagmus can be a very important adjunct in their clinical evaluation and follow up. In those patients in whom retinal or optic nerve diseases are suspected, photography may be indispensable in assisting with diagnosis and prognosis of associated afferent system diseases. There are multiple methods of obtaining photographs. The use of stereo photos, narrow-angle, high-magnification, and wide-angle, low-magnification techniques can be combined to obtain the best views of the macula/fovea, optic nerve head, and peripheral retina. In patients too young to cooperate for photography while awake, there are multiple types of hand-held and portable intraocular cameras that can be used as part of an examination under anesthesia.

Treatment

When considering possible treatments for INS, it is important to keep several things in mind:

1) INS in isolation is not a very debilitating condition (it does not prevent stereopsis nor greatly diminish peak acuity); 2) all of the major ocular motor functions that are necessary for good dynamic visual function are intact and functioning normally (gaze-holding range, saccades to locate targets, pursuit to track moving targets, vergence to fuse near targets, and the VOR and OKR to stabilize gaze in the presence of body or environmental motion); 3) the cosmetic effects of INS are usually minimal, often go unnoticed, and should not be the major factor in determining therapy (despite possibly being the primary concern of a parent; the child with INS is the patient, not the parent). Thus, in the tradition of, “do no harm,” no treatment should be considered that diminishes or prevents good visual function in real-world situations (ie, reading an eye chart with a still head in a specific position and without time constraints does not begin to simulate real-world situations). More specifically, treatment should strive to damp only the INS to allow improved foveation quality while leaving unaffected the already normal gaze-angle range, alignment (if non-strabismic), convergence range, saccades, smooth pursuit, VOR and OKR. The sometimes-intense desire to “stop the wiggling” (especially on the part of concerned parents) must not be allowed to overrule this constraint on safe and effective INS therapy. Indeed, as will be discussed below for specific therapies, even the amount of therapy provided may have to be limited to achieve maximal effectiveness without detrimental effects on other, necessary ocular motor functions.

If eye-movement data are available, determination of the efficacy of INS therapy is accurate, repeatable, and straightforward using the NAFXpk and LFD values, as has been well documented in the literature.² Therapeutically exploitable clinical characteristics of INS are shown in Figure 7. The therapies employed may be surgical or non-surgical. Also, based on research data, it is possible to use the curves shown in Figure 8 to estimate preoperatively the postoperative improvements in both NAFXpk (and VApk) and the LFD (and HAgar).

Figure 7. Flowchart demonstrating how clinical observations and tests may be used to determine therapeutically exploitable characteristics of infantile nystagmus syndrome (INS). The relevant therapies may be surgical or nonsurgical. Note that when the peak is high and the range of high-visual acuity (Hi VA) gaze angles is broad, their values cannot be significantly increased and, therefore, no waveform foveation improvements are possible; only under these simultaneous conditions is nystagmus therapy precluded. BMR, bimedial recession; m, muscle; Rec, recession; R&R, recess and resect; Strab, strabismus; T&R, tenotomy and reattachment. [From Figure D.4, Hertle, R. W. & Dell'Osso, L. F. (2013) Nystagmus in Infancy and Childhood. Current Concepts in Mechanisms, Diagnoses, and Management. Oxford University Press: Oxford.]

If, as is mostly the case, those data are not available, it is still possible (albeit more difficult) to predict post-therapy improvements in either VApk or HAgar for some patients.⁵¹ VApk is dependent on both the sensory and motor conditions present but, without NAFX data, the relative contributions of sensory deficits and foveation quality of the INS waveform cannot be accurately determined in most cases. Fortunately, however, HAgar is solely dependent on the characteristics of INS variation with gaze angle; it is purely motor and not affected by either the presence or magnitude of associated sensory deficits. Thus, although the percent improvement in VApk is directly related to the pre-therapy NAFXpk, the actual magnitude of post-therapy VApk will be limited by the magnitude of the sensory deficit. However, both the percent improvement and magnitude of post-therapy HAgar can be predicted from its pre-therapy value—determination of gaze-angle BCVA is the single most important clinical test to perform on all INS patients. It provides clinicians with an office measure that would allow them to better assess (pre-therapy) the post-therapy probability and amount of visual function improvement possible. One can substitute Hagar for LFD in the bottom part of Figure 8.

Despite the many possible combinations of sensory deficits and foveation-quality deficits, some combinations of VApk and HAgar (along with other factors discussed above) lend themselves to educated presumptions/conclusions that are helpful in making decisions regarding therapy and even establishing reasonable expectations of the magnitudes of possible improvements. For other combinations, the probability of correctly diagnosing and assessing therapeutic success in INS is low. The complexity of making those determinations precludes discussing them in detail here but a thorough discussion and methodology are available.⁵¹ A rule of thumb that can guide therapeutic decisions is that if HAgar is low, therapy is likely to provide significant improvements in visual function regardless of the relative contributions of sensory deficits and foveation quality to peak BCVA.

Figure 3 (Post) shows that after therapy, although the peak acuity is improved, the off-peak acuities are improved to a much greater extent. That broadens the range of gaze angles within which high visual acuity is possible and makes identification of those individuals on both sides much easier.

Figure 8. Plots of estimated post-T&R percentage and numerical improvements in NAFX and LFD as functions of their pre-therapy values. Green areas indicate that therapy should produce high-percentage, red areas indicate low-percentage, and blue areas indicate intermediate-percentage improvements. The NAFX and LFD data from these Figures plus data from the effects of BMR and BOPr therapies were used to construct the illustrative examples shown in Figs. 4 – 6.

Medical

Medical – Many systemic drugs to treat INS have been tried in the past but they either did not work, had undesirable side effects, or both. Dosage is very important to maximize INS foveation improvements while minimizing other ocular motor or systemic side effects. More recently, acetazolamide showed positive ocular motor improvements⁵⁷ as has topical brinzolamide.^{62; 63} Currently, a new topical sodium pump inhibitor is undergoing safety and efficacy testing.⁶⁴

Optical

Clearly, the first means to effectively treat INS is the application of refractive correction to produce the clearest images on the fovea. For binocular patients whose INS damps with convergence, the addition of 7 PD BO and -1.00S OU is a very effective therapy, almost always exceeding gaze-angle therapy. Research has shown that INS foveation improvements increase with the amount of convergence.¹⁷ However that should not lead to the conclusion that the best effective therapy results from the largest prisms the patient will tolerate. Good visual dynamic function requires that it be maintained over a broad range of gaze and convergence angles. Too much BO prism will prevent further adduction of either eye to foveate targets in lateral gaze and, more importantly, will prevent foveation of near targets. Attempts to do either could result in broken fusion, diplopia, and loss of stereopsis. Thus, we usually prescribe enough BO prisms to sufficiently improve distance foveation/acuity while preserving the ability to further adduct either eye. Seven PD BO has worked well for most patients although 8-9 PD BO may provide higher acuity for some while preserving their gaze and convergence ranges.

Surgical

There is a variety of surgical therapies that will damp INS in specific patients without harming (and in many cases, improving) ocular motor and visual function. They include: 1) the Kestenbaum (Anderson-Kestenbaum) procedure; 2) the Anderson plus T&R procedure; 3) the BMR procedure; 4) the T&R procedure; 5) procedures on the vertical rectus and oblique muscles in combination with these INS procedures; and 6) strabismus procedures in combination with these INS procedures. The details of specific surgical therapies and of their application to patients with INS in different clinical and ocular motor settings are complex and have been presented elsewhere.^{2; 51} It is the author’s opinion that maximal recession surgery offers little benefit above and beyond the T&R procedure, that surgical research on INS should begin with animal models, and that IRB approval is needed for all research on human subjects following the Helsinki protocol and following guidelines from the Office of Human Research Protections.

Other

In all cases where INS is associated with afferent visual deficits, new therapies for the latter should always be considered to remove that precipitating factor and possibly allow the ocular motor systems of those patients to recalibrate and damp/remove the INS oscillation (eg, see the research on congenital cataracts and Leber congenital amaurosis).^65-68 The details of each specific therapy and of its application to patients with INS in different clinical and ocular motor settings are complex and have been presented elsewhere.^{2; 51}

Discussion

Despite prior, mistaken dogma based on clinical observations alone, ocular motor data analysis has shown that saccades, smooth pursuit, VOR, and OKR are normal in patients with INS and should be preserved by any therapy.

Visual acuity is diminished by both sensory and motor deficits. That is, image degradation (due to optical, retinal, or neurological deficits) and poor foveation quality (due to nystagmus waveform characteristics that reduce foveation time or increase either foveation inaccuracy or eye velocities) will cause less-than-optimal visual acuity. From a therapeutic perspective, it is the ratio of these sensory to motor deficits that will determine the efficacy of therapies aimed at either. Thus, improving foveation quality in INS can only result in improved acuity if that quality (ie, the NAFX) was lower than optimal to start with—that is not rocket science. In fact, the percent improvement in the NAFX (and acuity) will be the same with or without an associated sensory deficit; only the absolute values of both pre- and post-therapy will be less than for those with INS and no sensory deficits.

Because of the dual nature of the factors affecting visual acuity, one must carefully evaluate patients with both INS and sensory deficits (eg, those with albinism, hypoplasia of the optic nerves, etc.) and not dismiss such patients as “not likely to benefit from INS therapy” simply because of the latter. Ocular motor research had shown that, given the right ratio of sensory to motor deficit, these patients may benefit most from INS therapy.^69-72

Finally, the therapeutic improvements produced by INS therapies acting at different ocular motor sites have been shown to be multiplicative.⁷³ That is, if a drug can reduce the originating ocular motor signal driving INS by 50% and if an extraocular muscle surgery can reduce an ongoing nystagmus (driven by that ocular motor signal) by 50%, their use in combination will result in a nystagmus that is 25% of its pre-therapy value. Therefore, the physician should continue to explore other post-surgical therapies (systemic or topical drugs, optical or exteroceptive methods, etc.) to maximize the individual patient’s overall visual function. This is not an area where clinical trials and their usually inadequate/incorrect exclusion criteria and analysis statistics will provide much guidance but rather, they may falsely exclude the specific therapy needed for your patient (eg, a clinical trial that concluded incorrectly that contact lenses do not provide improved visual function when, in fact, such improvements have been decisively shown for specific patients).

References

CEMAS_Working_Group. A National Eye Institute Sponsored Workshop and Publication on The Classification of Eye Movement Abnormalities and Strabismus (CEMAS). In The National Eye Institute Publications (www.nei.nih.gov/news/statements/cemas.pdf). 2001

Hertle RW, Dell'Osso LF. Nystagmus in Infancy and Childhood. Current Concepts in Mechanisms, Diagnoses, and Management. Oxford: Oxford University Press. 2013.

Kestenbaum A. Clinical Methods of Neuro-Ophthalmologic Examination, 1st Ed. New York: Grune and Stratton. 1947.

Kestenbaum A. Clinical Methods of Neuro-Ophthalmologic Examination, 2nd Ed. New York: Grune and Stratton. 1961.

Anderson JR. The treatment of congenital nystagmus. editor. Ocular vertical deviations and the treatment of nystagmus. Philadelphia: J.B. Lippincott Co, 1959; 165-172.

Leigh RJ, Zee DS. The Neurology of Eye Movements, Edition 4 (Contemporary Neurology Series). New York: Oxford University Press. 2006.

Dell'Osso LF. Evaluation of smooth pursuit in the presence of congenital nystagmus. Neuro ophthalmol 1986;6:383-406.

Dell'Osso LF, Van der Steen J, Steinman RM, Collewijn H. Foveation dynamics in congenital nystagmus I: Fixation. Doc Ophthalmol 1992;79:1-23.

Dell'Osso LF, Van der Steen J, Steinman RM, Collewijn H. Foveation dynamics in congenital nystagmus II: Smooth pursuit. Doc Ophthalmol 1992;79:25-49.

Dell'Osso LF, Van der Steen J, Steinman RM, Collewijn H. Foveation dynamics in congenital nystagmus III: Vestibulo-ocular reflex. Doc Ophthalmol 1992;79:51-70.

Jacobs JB, Dell'Osso LF, Leigh RJ. Characteristics of braking saccades in congenital nystagmus. Doc Ophthalmol 2003;107:137-154.

Dell'Osso LF. The mechanism of oscillopsia and its suppression. In: Rucker J, Zee DS, editors. Basic and Clinical Ocular Motor and Vestibular Research—Ann NY Acad Sci 1233. Boston, MA: Blackwell Publishing, 2011; 298-306.

Gresty MA, Bronstein AM, Page NG, Rudge P. Congenital-type nystagmus emerging in later life. Neurology 1991;41:653-656.

Dell'Osso LF. Fixation characteristics in hereditary congenital nystagmus. Am J Optom Arch Am Acad Optom 1973;50:85-90.

Tkalcevic LA, Abel LA. The effects of increased visual task demand on foveation in congenital nystagmus. Vision Res 2005;45:1139-1146.

Daroff RB, Dell'Osso LF. Periodic alternating nystagmus and the shifting null. Can J Otolaryngol 1974;3:367-371.

Serra A, Dell'Osso LF, Jacobs JB, Burnstine RA. Combined gaze-angle and vergence variation in infantile nystagmus: two therapies that improve the high-visual acuity field and methods to measure it. Invest Ophthalmol Vis Sci 2006;47:2451-2460.

Dell'Osso LF, Jacobs JB, Serra A. The sub-clinical see-saw nystagmus embedded in infantile nystagmus. Vision Res 2007;47:393-401.

Dell'Osso LF. Original ocular motor analysis of the first human with achiasma: documentation of work done in 1994. OMLAB Report 2006;090506:1-21.

Dell'Osso LF, Hertle RW, Daroff RB. "Sensory" and "motor" nystagmus: erroneous and misleading terminology based on misinterpretation of David Cogan's observations. Arch Ophthalmol. 2007;125:1559-1561.

Dell'Osso LF, Jacobs JB. A normal ocular motor system model that simulates the dual-mode fast phases of latent/manifest latent nystagmus. Biol Cybern. 2001;85:459-471.

Dell'Osso LF. Nystagmus basics. Normal models that simulate dysfunction. In: Hung GK, Ciuffreda KJ, editors. Models of the Visual System. New York: Kluwer Academic / Plenum Publishers, 2002; 711-739.

Jacobs JB, Dell'Osso LF. Congenital nystagmus: hypothesis for its genesis and complex waveforms within a behavioral ocular motor system model. J Vis. 2004;4:604-625.

Dell'Osso LF. Biologically relevant models of infantile nystagmus syndrome: the requirement for behavioral ocular motor system models. Semin Ophthalmol. 2006;21:71-77.

Wang ZI, Dell'Osso LF. A unifying model-based hypothesis for the diverse waveforms of infantile nystagmus syndrome. J Eye Movement Res. 2011;4:1-18.

Dell'Osso LF, Jacobs JB. An expanded nystagmus acuity function: intra- and intersubject prediction of best-corrected visual acuity. Doc Ophthalmol. 2002;104:249-276.

Jacobs JB, Dell'Osso LF. Extending the expanded nystagmus acuity function for vertical and multiplanar data. Vision Res. 2010;50:271-278.

Kim JI, Dell'Osso LF, Traboulsi E. Latent/manifest latent and ‘uniocular’ acquired pendular nystagmus masquerading as spasmus nutans. J Neuro-Ophthalmol .2003;22:198-203.

Dell'Osso LF, Leigh RJ, Sheth NV, Daroff RB. Two types of foveation strategy in 'latent' nystagmus. Fixation, visual acuity and stability. Neuro Ophthalmol. 1995;15:167-186.

Dell'Osso LF, Schmidt D, Daroff RB. Latent, manifest latent and congenital nystagmus. Arch Ophthalmol.1979;97:1877-1885.

Ishikawa S. Latent nystagmus and its etiology. In: Reinecke RD, editor. Strabismus, Proceedings of the Third Meeting of the International Strabismological Association. New York: Grune and Stratton, 1979; 203-214.

Abadi RV. Pattern contrast thresholds in latent nystagmus. Acta Ophthalmol. 1980;58:210-220.

Dell'Osso LF, Daroff RB. Clinical disorders of ocular movement. In: Zuber BL, editor. Models of Oculomotor Behavior and Control. West Palm Beach: CRC Press Inc, 1981; 233-256.

van Dorp DB, Eriksson AW, Delleman JW, van Vliet AG, Collewijn H, van Balen AT, Forsius HR. Aland eye disease: no albino misrouting. Clin Genet 1985;28:526-531.

Von Noorden GK, Avilla C, Sidikaro Y, La Roche R. Latent nystagmus and strabismic amblyopia. Am J Ophthalmol. 1987;103:87-89.

Gresty MA, Metcalfe T, Timms C, Elston J, Lee J, Liu C. Neurology of latent nystagmus. Brain. 1992;115:1303-1321.

Gradstein L, Goldstein HP, Wizov SS, Hayashi T, Reinecke RD. Relationships between visual acuity demands, convergence and nystagmus in patients with manifest/latent nystagmus. J Am Assoc Pediatr Ophthalmol Strab. 1998;00:000-000.

Jacobs JB, Dell’Osso LF. A robust, normal ocular motor system model with latent/manifest latent nystagmus (LMLN) and dual-mode fast phases. In: Kaminski HJ, Leigh RJ, editors. Neurobiology of Eye Movements. From Molecules to Behavior—Ann NY Acad Sci 956. New York: NYAS, 2002; 604-607.

Dell'Osso LF, Ellenberger JC, Abel LA, Flynn JT. The nystagmus blockage syndrome: Congenital nystagmus, manifest latent nystagmus or both? Invest Ophthalmol Vis Sci. 1983;24:1580-1587.

Dell'Osso LF. Congenital, latent and manifest latent nystagmus - similarities, differences and relation to strabismus. Jpn J Ophthalmol. 1985;29:351-368.

Kommerell G, Gusek G, Gilles U. Kongenitaler Nystagmus und intermittierende Exotropie. Nystagmushemmung durch fusionale Konvergenz. Klin Mbl Augenheilk. 1992;200:210-212.

Reinecke RD. Nystagmus blockage syndrome in the unilaterally blind patient. Doc Ophthalmol .1984;58:125-130.

Gottlob I, Wjizov SS, Reinecke RD. Spasmus nutans. A long-term follow-up. Invest Ophthalmol Vis Sci. 1995;36:2768-2771.

Weissman BM, Dell'Osso LF, Abel LA, Leigh RJ. Spasmus nutans: A quantitative, prospective study. In: Keller EL, Zee DS, editors. Adaptive Processes In Visual and Oculomotor Systems. Oxford: Pergamon Press, 1986; 479-483.

Weissman BM, Dell'Osso LF, Abel LA, Leigh RJ. Spasmus nutans: A quantitative prospective study. Arch Ophthalmol. 1987;105:525-528.

Gresty MA, Halmagyi GM. Abnormal head movements. J Neurol Neurosurg Psychiatry. 1979;42:705-714.

Gottlob I, Zubcov AA, Wizov SS, Reinecke RD. Head nodding is compensatory in spasmus nutans. Ophthalmology. 1992;99:1024-1031.

Gottlob I, Wizov SS, Reinecke RD. Quantitative eye and head movement recordings of retinal disease mimicking spasmus nutans. Am J Ophthalmol. 1995;119:374-376.

Gresty MA, Leech J, Sanders MD, Eggars H. A study of head and eye movement in spasmus nutans. Br J Ophthalmol. 1976;160:652-654.

Gresty MA, Ell JJ. Spasmus nutans or congenital nystagmus? Classification according to objective criteria. Br J Ophthalmol. 1981;65:510-511.

Dell'Osso LF. Clinical INS Assessment to Determine Maximally Effective Therapy: What can the Physician Apply from the Bench to the Bedside? OMLAB Report. 2014;061214:1-44.

Kelly TW. Optic glioma presenting as spasmus nutans. Pediatrics. 1970;45:295-296.

Koenig SB, Naidid TP, Zaparackas Z. Optic glioma masquarading as spasmus nutans. J Pediatr Ophthalmol Strab. 1982;19:20-24.

Albright AL, Sclabassi RJ, Slamovits TL, Bergman I. Spasmus nutans associated with optic gliomas in infants. J Pediatr. 1984;105:778-780.

Kalyanaraman K, Jagannathan K, Ramanujam RA, Ramamurthi B. Congenital head nodding and nystagmus with cerebrocerebellar degeneration. J Pediatr. 1973;83:1023-1026.

Gottlob I, Zubkov A, Catalano RA, Reinecke RD, Koller HP, Calhoun JH, Manley DR. Signs distinguishing spasmus nutans (with and without central nervous system lesions) from infantile nystagmus. Ophthalmology. 1990;97:1166-1175.

Thurtell MJ, Dell'Osso LF, Leigh RJ, Matta M, Jacobs JB, Tomsak RL. Effects of acetazolamide on infantile nystagmus syndrome waveforms: comparisons to contact lenses and convergence in a well-studied subject. Open Ophthalmol J. 2010;4:37-46.

Wang ZI, Dell'Osso LF. Being "slow to see" is a dynamic visual function consequence of infantile nystagmus syndrome: Model predictions and patient data identify stimulus timing as its cause. Vision Res. 2007;47:1550-1560.

Wang ZI, Dell'Osso LF. Tenotomy procedure alleviates the "slow to see" phenomenon in infantile nystagmus syndrome: model prediction and patient data. Vision Res. 2008;48:1409-1419.

Dell'Osso LF. Effects on congenital nystagmus (CN) of combined gaze-angle and vergence variation: therapeutic implications. J Neuro-Ophthalmol. 2002;22:185.

Cronin TH, Hertle RW, Ishikawa H, Schuman JS. Spectral domain optical coherence tomography for detection of foveal morphology in patients with nystagmus. J AAPOS. 2009;13:563-566.

Dell'Osso LF, Hertle RW, Leigh RJ, Jacobs JB, King S, Yaniglos S. Effects of topical brinzolamide on infantile nystagmus syndrome waveforms: eye drops for nystagmus. J Neuro-Ophthalmol. 2011;31:228-233.

Hertle RW, Yang D, Adkinson T, Reed M. Topical brinzolamide (Azopt) versus placebo in the treatment of infantile nystagmus syndrome (INS). Br J Ophthalmol. 2015;99:471-476.

Hertle RW, Dunmire J, Dell'Osso LF, Jacobs JB, Dalvin LA, Yang D, Evano-Chapman M. Efficacy and Safety of New Topical Sodium Pump Inhibitor (NSPI) in Reducing Eye Oscillations in a Canine model of Infantile Nystagmus Syndrome (INS). Invest Ophthalmol Vis Sci. 2014;55:E-abstract 4546.

Cross SA, Smith JL, Norton EWD. Periodic alternating nystagmus clearing after vitrectomy. J Clin Neuro Ophthalmol 1982;1:5-11.

Yagasaki T, Sato M, Awaya S, Nakamura N. Changes in nystagmus after simultaneous surgery for bilateral congenital cataracts. Jpn J Ophthalmol. 1993;37:330-338.

Jacobs JB, Dell'Osso LF, Hertle RW, Acland GM, Bennett J. Eye movement recordings as an effectiveness indicator of gene therapy in RPE65-deficient canines: Implications for the ocular motor system. Invest Ophthalmol Vis Sci. 2006;47:2865-2875.

Jacobs JB, Dell'Osso LF, Wang ZI, Acland GM, Bennett J. Using the NAFX to measure the effectiveness over time of gene therapy in canine LCA. Invest Ophthalmol Vis Sci. 2009;50:4685-4692.

Hertle RW, Dell'Osso LF, FitzGibbon EJ, Thompson D, Yang D, Mellow SD. Horizontal rectus tenotomy in patients with congenital nystagmus. Results in 10 adults. Ophthalmology. 2003;110:2097-2105.

Hertle RW, Anninger W, Yang D, Shatnawi R, Hill VM. Effects of extraocular muscle surgery on 15 patients with oculo-cutaneous albinism (OCA) and infantile nystagmus syndrome (INS). Am J Ophthalmol. 2004;138:978-987.

Hertle RW, Dell'Osso LF, FitzGibbon EJ, Yang D, Mellow SD. Horizontal rectus muscle tenotomy in patients with infantile nystagmus syndrome: a pilot study. J AAPOS. 2004;8:539-548.

Wang Z, Dell'Osso LF, Jacobs JB, Burnstine RA, Tomsak RL. Effects of tenotomy on patients with infantile nystagmus syndrome: foveation improvement over a broadened visual field. JAAPOS. 2006;10:552-560.

Tomsak RL, Dell'Osso LF, Rucker JC, Leigh RJ, Bienfang DC, Jacobs JB. Treatment of acquired pendular nystagmus from multiple sclerosis with eye muscle surgery followed by oral memantine. DJO. 2005;11:1-11.