Clinical Significance of Saccade Analysis in Early Active Graves’ Ophthalmopathy

Abstract

purpose. To assess whether saccadic eye movements show distinct changes in patients with early active Graves’ ophthalmopathy (GO), which could serve as a diagnostic tool for early detection and treatment.

methods. Each of two prospective studies included 10 patients with early acute GO and 10 age- and sex-matched control subjects. In the explorative study (ES) 15 dynamic parameters of saccades were analyzed. In the comparative study (CS) only those parameters were evaluated, which in ES had shown significant differences between patients and controls. Horizontal and vertical saccades of 10°, 20°, and 40° including a fatigue test were recorded binocularly using the induction scleral search coil.

results. The differences of saccadic dynamics between patients and controls were small, whereas intra- and interindividual standard deviations were large. In ES, 7.1% of the parameters showed significant differences at a level of P ≤ 0.05. In CS, 2.1% of all parameters revealed repetitive significant differences. Despite statistical significance, individual data did not allow differentiation between patients and healthy individuals due to high standard deviations.

conclusions. In early active GO no clinically relevant saccadic changes were detected. These findings may be based on adaptation of the central saccadic generator. Inclusion of patients with fibrotic muscle changes due to long-standing disease could explain the contrasting results of previous studies. Consequently, analysis of saccades does not serve as a diagnostic tool during early active GO.

Graves’ ophthalmopathy (GO) is a complication of Graves’ disease and represents the most frequent inflammatory condition of the orbital tissues and extraocular muscles. Potential sequelae of GO include cosmetic alterations due to lid changes and exophthalmos, motility disturbances with or without diplopia due to inflammatory eye muscle changes, corneal ulceration due to insufficient lid closure, and decreased visual acuity up to complete loss of vision due to optic nerve compression. No specific treatment is available to date, ¹ and one can only try to avoid these sequelae by early administration of nonspecific anti–inflammatory agents such as systemic corticosteroids or orbital ionizing irradiation. 

Timing of treatment depends on the degree of inflammatory activity, which remains difficult to assess. The course of the disease includes a phase of increasing activity, a plateau phase with little change in activity, a stage of spontaneous regression, and a postinflammatory stage. ² Once the late stage has been reached, most changes have become refractory to anti–inflammatory treatment and are thus irreversible. ^{3 4 5} Although it is advisable to initiate treatment before irreversible changes have taken place, the possibility of spontaneous regression has to be taken into account. To date, there is a general lack of functional studies that would reliably help to quantify the inflammatory activity and thus assist in the timing of anti–inflammatory treatment. 

Intraorbital inflammatory changes can cause marked swelling of the extraocular muscles as demonstrated by imaging procedures such as ultrasound, computed tomography, and magnetic resonance imaging. These structural changes suggest the presence of functional changes that during the early stages are not commonly apparent on clinical examination. Because a high degree of extraocular muscle fiber activity is required for rapid eye movements, one would expect that dynamic analysis of saccades could assist in the detection of subtle functional changes. 

In the literature of the past two decades there have been only a few reports on alterations of saccadic properties due to GO. In these studies, various eye movement recording techniques such as electro-nystagmogramm (ENG), infrared and search coil technique, as well as different paradigms have been used. The results of these studies which focused primarily on saccadic velocity were contradictory (Table 1) . In patients with GO, saccadic velocity was reported to be increased, ^{6 7 8} decreased, ^{9 10 11 12 13 14} or unchanged. ^{6 15 16 17} Furthermore, the existence of pathologic fatigability of the saccadic system has remained controversial. Absence of muscle fatigue was reported in healthy subjects ^{18 19} whereas a fatigue effect was detected in patients with GO. ^{19 20} However, one study also reported significant saccadic fatigue in normal individuals. ²¹  

The aim of this study was to assess whether patients with early acute GO exhibit significant changes of saccadic properties which can reliably be distinguished from those of healthy individuals. To this purpose we used the induction scleral search coil method, as first described by Robinson. ²² We selected patients in the early active stages of GO to assess whether saccadic changes can help quantify the degree of inflammatory activity, and whether saccadic analysis can serve as a diagnostic guide for early initiation of anti–inflammatory treatment and, thus, prevention of irreversible functional disturbances. 

Methods

Studies

To explore the reliability of results, two technically identical studies were carried out. The first study was designed as an explorative study (ES). A variety of parameters characterizing the dynamics of saccades were analyzed and compared between patients and healthy individuals. In the second study, the comparative study (CS), only those parameters that were significantly different between the two groups in ES were assessed in a new group of patients and control subjects. If detection of pathologic saccadic changes has any clinical benefit, these changes should be reproducibly present because they should be clearly distinguishable from those of a normal population. 

Subjects

In ES, 10 patients were selected who attended the outpatient clinic because of recent onset of active GO. None of these patients had received any form of anti–inflammatory treatment. There were 8 women and 2 men. Their mean age was 35.5 years, ranging from 23 to 45 years. Inclusion criteria were as follows: clinically marked and unambiguous signs and symptoms of GO, an onset of symptoms not longer than 6 months ago, proven Graves’ disease as assessed by analysis on serum autoantibodies (Anti–TPO-Ab, TSH-receptor-Ab) and thyroid function tests (T3, T4, TSH), ⁴ absence of strabismus and diplopia within a range of 25° around primary position, and a stage of disease defined as L > 0, E > 2, M = 1 or 2, and O < 3 according to the “LEMO” classification ²³ (Table 2) . Exclusion criteria were any signs of myasthenia gravis, congenital strabismus and other ophthalmological diseases or operations, and neurologic disorders. 

Ten age- (±2 years) and sex-matched healthy individuals were selected as controls. They were recruited from the university staff. Mean age of these 8 female and 2 male control subjects was 34.5 years, ranging from 23 to 43 years. Inclusion criteria were as follows: the absence of any ophthalmological, endocrine, or neurologic disorder; unimpaired visual acuity and binocular function; and a TSH serum level within the normal range. 

In CS, another 10 patients and 10 healthy individuals were investigated, using the same inclusion and exclusion criteria. Mean age of the patients (9 females and 1 male) was 49.5 years, ranging from 23 to 58 years. The duration and severity of disease were comparable to that of the patients in ES. Mean age of the 10 controls was 50.2 years, ranging from 25 to 57 years. 

This research adhered to the tenets of the Declaration of Helsinki; informed consent was obtained from all patients after the nature and possible consequences of the study were explained. 

Clinical Investigations

In both ES and CS the patients underwent a number of clinical investigations to assess the stage of the disease and the degree of inflammatory activity. The ophthalmologic examination consisted of the following: inspection of the lids and measurement of the lid fissure, biomicroscopy of anterior and posterior ocular segments, measurement of intraocular pressure in relaxed gaze, and intended upgaze, ^{23 24} Hertel readings for measurement of proptosis, tear secretion analysis and vital staining of the conjunctiva and cornea, and an orthoptic investigation including visual acuity, binocular alignment and binocular function testing, measurement of monocular excursions, and assessment of the field of binocular single vision. Additionally, we performed a standardized ultrasound investigation of the extraocular muscles and the retrobulbar space ²⁵ and a computerized perimetry of 30°. 

Eye Movement Recording Technique

The saccadic eye movements were recorded binocularly with a commercially available two-field search coil system (Eye Position Meter 3020; Skalar Instruments, Delft, The Netherlands). This system generates two quadrature-modulated alternating magnetic fields at a frequency of 20 kHz, which are approximately spatially perpendicular to each other. During the experiments the head of the subject was placed in the center of the cube frame (70-cm side length), which contained the rectangular coils for generating the magnetic field. Head movements were restricted by using a chin-rest and a forehead support. The system’s deviation of the linearity was less than 1% at 20°. The cross talk between the horizontal and the vertical channels was below 1%. 

A commercially available silicone annulus with a standard induction coil according to Collewijn et al. ²⁶ was applied to the limbus of both the right and left eyes after local administration of 0.4% oxybuprocainhydrochloride. Horizontal and vertical output signals generated by the coils were digitized with a 12-bit AD converter (DAP 1200) at a sampling rate of 500 Hz and further processed in a laboratory computer (PC 486, 50 MHz). The signal noise level was less than 57 seconds of arc. 

Data Analysis

The recordings were calibrated during binocular viewing of the subject. Before starting the recording session, every individual was carefully tested to ensure they were diplopia free within the required range of the paradigm. A polynomial continuous calibration throughout each paradigm was established using late and stable fixation periods, with a velocity of less than 3°/sec over at least 130 msec within a time interval of 500 msec before the next target step. Thus, major coil slippage or changes in ocular alignment could be observed; records so affected were excluded from further analysis. Saccades were detected and marked on the basis of velocity criteria. Eye movement velocity was calculated using a symmetrical two-point differentiator after low-pass filtering with a Gaussian FIR filter with a cutoff frequency of 50 Hz. For calculation of acceleration and deceleration, the position signal was filtered using a Gaussian low-pass filter with a cutoff frequency of 28 Hz. Subsequently, it was numerically differentiated twice. Starting from the time of peak velocity, the algorithm searched forward and backward until velocity dropped below 10% of maximum; these points were defined as initiation and end of saccade, respectively. To reduce the influence of anticipatory and spontaneous eye movements, only saccades with a latency between 100 and 300 msec were included. 

Fifteen parameters of every saccade were analyzed both for the right and the left eyes; these parameters are listed in Table 3 . Correcting saccades and glissades were excluded. A commercially available computer program (SPSS, version 6.0.1) was used for further data analyses. For each saccadic parameter, aggregate means were computed for each eye of each individual. The aggregates from each eye then were averaged to obtain a single representative value for that subject because changes of GO occur bilaterally. Kolmogorov–Smirnov analyses determined that aggregate data from patients and control subjects were normally distributed. Differences between groups were assessed using the parametric Student’s t-test for independent samples. Significance was accepted at P ≤ 0.05. 

Paradigms

The visual target was a small He–Ne laser spot with a diameter of 0.2° which was projected onto a white screen at a distance of 120 cm from the subject’s corneas. Change of position of the laser spot was achieved by a horizontal and vertical rapid mirror galvanometer (General Scanning G100PD; Watertown, MA) at a velocity of 10°/msec. The subject was instructed to carefully fix and follow the laser spot. 

Three paradigms were carried out in each subject, always after the same order. In the 1st paradigm (verhor) both horizontal and vertical saccades of 10° and 20° were performed (Fig. 1) . Unlike the 2nd and the 3rd paradigms, all saccades started from (centrifugal [cfug]) and returned to (centripetal [cpet]) the primary position. For statistical reasons each saccade was evoked 4 times, and the mean of these 4 saccades was calculated for each parameter. The course of the paradigm was randomized with respect to direction and time; the mean interval between two saccades was 1.7 seconds, ranging between 1.4 and 2.0 seconds. Total duration was 100 seconds. The 2nd and 3rd paradigms were each designed as a fatigue test based on the description by Mauri et al. ¹⁹ It was our purpose to find out whether continuous performance of saccades results in dynamic changes due to fatigue of the oculomotor system. In both vertical (2nd paradigm [fatver]) and horizontal (3rd paradigm[ fathor]) directions, a sequence of 40° saccades crossing the center position had to be carried out over 3 minutes with an intersaccadic mean interval of 1.2 seconds, ranging between 0.7 and 1.7 seconds. Ten 20° saccades were performed both before and after the fatigue test. To reduce anticipation and automatic eye movements, 3 to 5 randomized fixations in primary position of approximately 200 msec duration were included in both paradigms. 

Results

Clinical Findings

In both ES and CS at least one of three typical signs of GO, ²⁸ the Dalrymple (upper lid retraction), von Graefe (upper eyelid lag on downgaze), and Möbius (deficient convergence) sign, was present in every patient. On ultrasound investigation, marked swelling of at least one of the recti muscles was found among all patients. All patients showed moderate to severe lid swellings, the mean L value of the LEMO classification ²³ being 3.0 in ES and 2.5 in CS. An exophthalmos causing permanent conjunctival irritation was present in all patients of ES and in 9 patients of CS; 1 patient of CS presented with a corneal ulcer due to proptosis and consecutive insufficient lid closure. The mean E value was 3.0 in both studies. In both ES and CS all but one patient had mild to moderate pseudoparesis due to inflammatory muscle changes, the mean M being 1.9. According to the inclusion criteria, no patient complained of diplopia within 25° around the primary position. In ES no optic nerve compression was detected (O = 0), whereas in CS 4 patients had signs of mild to moderate optic nerve affection (O = 1 or 2). This difference did not violate the study protocol because it was the purpose to search for saccadic changes at any acute stage of recent-onset GO. 

Saccadic Properties: Explorative Study

According to the three paradigms, there were 32 different saccades to be analyzed (Table 4 , first column). Eight different kinds of saccades were carried out in the four secondary positions of gaze (upgaze [up]; downgaze [down]; abduction [ab]; adduction [ad]): 10° saccades (10) starting from primary position (centrifugal [cfug]); 10° saccades (10) returning to primary position (centripetal [cpet]); 20° saccades (20) starting from primary position (centrifugal [cfug]); 20° saccades (20) returning to primary position (centripetal [cpet]); 20° saccades (20) crossing primary position before fatigue (early); 20° saccades (20) crossing primary position after fatigue (late); 40° saccades (40) crossing primary position before fatigue (early); and 40° saccades (40) crossing primary position after fatigue (late). The early saccades were represented by the mean of the first five saccades of the 2nd and 3rd paradigms, the late saccades by the mean of the last five saccades of the 2nd and 3rd paradigms. Analysis of 15 parameters of every saccade resulted in 480 data entries for each individual as illustrated in Table 4 . 

The eye movement recordings of individual patients and their control subjects were not obviously different, as illustrated by Figure 1 . Due to the scattered data distribution within both the patient and the control groups, the standard deviations of all 15 parameters were substantial. As demonstrated in Table 4 , 34 of 480 data entries (7.1%) showed significant differences between groups. The significant differences occurred sporadically and did not adhere to any systematic pattern with regard to the various kinds of eye movements and to fatigue. There was no preference of certain directions of gaze, saccadic amplitudes, or state of fatigue. The individual data of each significantly different saccadic parameter are listed in Table 5 . 

An intraindividual saccadic fatigue effect appeared sporadically and was more pronounced in healthy individuals than in GO patients (Table 6) . Among the control subjects, 33 of 480 data entries (6.9%) showed significant changes after fatigue, whereas among the patient group 9 of 480 data entries (1.9%) were influenced significantly by fatigue. Maximum velocity (V_max), was significantly reduced in the 40° abducting, adducting, and upgaze saccades of the control subjects only. Among the patient group there was no significant change in V_max. Patient and control groups showed a significant increase of amplitude (Amp), after 40° abducting and adducting saccades and a decrease after 40° downgaze saccades. Disconjugacy (Disconj), was reduced after 20° upgaze saccades in the controls only and was never significantly changed by fatigue in the patients. 

Saccadic Properties: Comparative Study

In CS, replication of significant differences occurred in 5 of the 15 analyzed parameters. As illustrated in Table 4 , 10 of 480 data entries (2.1%) showed significant differences between the patients and the controls in both ES and CS. The individual data of these 10 differences are listed in Table 7 . As shown in Tables 4 and 7 , 7 of 10 recurrent significant differences were observed among the horizontal 40° saccades (3rd paradigm), none occurred in the 1st paradigm. Each 5 of 10 differences were found before and after fatigue. As also demonstrated in Table 7 , the standard deviations of the 10 different mean values of the patients and the controls were of a magnitude comparable to that of the differences. Subsequently, overlapping of individual data of patients and controls occurred, thus preventing the data from being used to classify the individuals according to their group. 

Discussion

The aim of this study was to determine whether or not the dynamic properties of saccades are influenced by early inflammatory eye muscle changes in patients with GO. The clinical impact of this question is significant in view of the lack of a reliable method by which inflammatory active GO can be diagnosed and treated early to avoid the disabling and potentially blinding sequelae of this disease. This issue was addressed in a systematic fashion using a highly accurate technique of eye movement recordings, the induction scleral search coil, combined with a strict selection of subjects with early but clinically marked disease. Moreover, our study used matched pairs as control individuals, examined a wide range of dynamic parameters on horizontal and vertical saccades, and assessed the reliability of results by performing an extension study under the same conditions. 

In ES, only a few differences of saccade properties were detected between healthy individuals and patients with GO. In CS, less than one third of the differences observed in ES could be confirmed. None of the 15 parameters characterizing saccades was systematically changed among all 32 conditions studied. Maximum velocity, amplitude, and binocular disconjugacy have been reported to be changed in GO, ^{6 10 13 14} but these parameters failed to reveal characteristic changes in our studies. The few changes that we observed occurred mainly in horizontal saccades of large amplitudes. These changes affected parameters that characterize the development of velocity, such as skewness and time at V_max, rather than V_max itself. The total differences occurring in both studies was 2.1% of 480 data entries, thus falling within the allowed limits of chance (5% at P ≤ 0.05). Despite statistical significance, these differences could not be used to distinguish an individual with GO from healthy individuals. We attribute this to large scatter of individual data and substantial overlap of data between groups. 

The current results are in accordance with several previous studies, which did not detect saccadic changes in patients with early, nonfibrotic, GO. ^{6 15 16 17 20} However, our findings differ from several reports of characteristic changes in GO, ^{7 8 9 10 11 12 13 14 19} which mainly involve saccadic velocity and amplitude or their relationship called main sequence. ²⁹  

The search coil technique was only applied in the most recent study. ¹⁴ All other authors used the infrared or ENG method, which can increase the probability of error. ³⁰ Some of these reports were based on single observations, ^{7 11 12} thus lacking a control population or statistical support. Except for the study of Feldon et al., ¹⁰ in which 49 patients were investigated, the sample size of the present study with 20 patients in two sets is greater than previous studies whose sample sizes ranged between 8 and 15. ^{9 13 14 19}  

Perhaps the most important methodological differences between the present study and those that reported saccadic changes are the stage and duration of the ophthalmopathy. Only early stages were included in our study. Previous studies that detected saccadic changes were not restricted to early disease. Inclusion of patients with long-standing disease increased the likelihood of fibrotic muscle changes and marked motility restriction. Saccadic velocity can drop substantially when muscular fibrosis occurs as a result of long-standing disease, which has been reported to cause tailing off at the end of the saccade. ¹⁵ Accordingly, Neumann et al. ²⁰ found no velocity changes in 10 patients with GO without restrictive myopathy but a significant decrease in 2 patients with fibrotic motility restriction. 

The question arises why the dynamic properties of rapid eye movements did not change in our patients with early GO despite evidence of eye muscle swelling and a previous report ³¹ that contractile properties are substantially altered in such individuals. We hypothesize that adaptation of the central saccadic generator compensated for the muscular changes by adjustment of the central innervational pattern to maintain rapid and precise foveation of a new target. Kommerell et al. ³² and Optican and Robinson ³³ previously reported on the remarkable adaptational capacity of the saccadic system, which was demonstrated to be mainly associated with the cerebellum. ^{33 34 35} The phenomenon of saccadic adaptation was also discussed by Acheson et al., ³⁶ who doubted the importance of this mechanism because of the disconjugacy between both eyes. However, unilateral saccadic adaptation was reported in asymmetrical changes within the ocular plant, ^{37 38 39} a feature that also applies to the commonly asymmetrical distribution of eye muscle affections in GO. ⁴⁰  

In summary, the present results were obtained in two consecutive studies that included strict selection for early stages of GO, age- and sex-matched controls, and a reliable recording technique. Our analysis of saccades did not identify clinically relevant saccadic changes in early stages of GO, when detection of inflammatory activity is critical for the initiation of anti-inflammatory treatment. Preliminary studies in our laboratory indicate that saccadic changes may become apparent in more advanced stages of GO. However, such changes are of limited clinical importance because these later stages are readily assessed by the severity of inflammation or motility restrictions. We conclude that analysis of saccades is not a useful diagnostic tool during the early inflammatory active stage of GO. 

 Supported by Johann Peter Frank–Gesellschaft, Rodalben, Germany, and Münchner Medizinische Wochenschrift, Munich, Germany.

 Submitted for publication June 21, 1999; revised October 12 and November 30, 1999; accepted December 20, 1999.

 Commercial relationships policy: N.

 

Table 1.

 

Reports on Saccadic Velocity Changes in Patients with Graves’ Ophthalmopathy

First Author, Year	Recording Technique	No. of Patients	Direction of Saccade	Saccadic Velocity
Metz, 1977 15	ENG	15	Horizontal, vertical	Unchanged
Kirsch, 1990 7	Infrared	1	Horizontal	Unchanged (mild disease)
Neumann, 1981 20	ENG	14	Horizontal	Unchanged (without motility restriction)
Mauri, 1984 19	Infrared	25	Horizontal	Unchanged (without fatigue)
Schworm, 1997 17	Infrared	20	Horizontal, vertical	Unchanged
Kirsch, 1990 7	Infrared	1	Horizontal	Increased (severe disease)
Nishino, 1996 6	Infrared	19	Horizontal	Increased
Schworm, 1997 8	Infrared	13	Horizontal, vertical	Increased
Hermann, 1982 11	ENG	2	Vertical	Reduced
Feldon, 1982 9	Infrared	13	Horizontal	Reduced
Mauri, 1984 19	Infrared	25	Horizontal	Reduced (after fatigue)
Konen, 1984 12	Infrared	1	Horizontal	Reduced
Feldon, 1990 10	Infrared	49	Horizontal	Reduced
Tian, 1997 13	ENG	8	Horizontal, vertical	Reduced
Wouters, 1998 14	Search coil	12	Horizontal, vertical	Reduced

 ENG, electro-nystagmogramm.

Table 2.

 

Clinical Classification (“LEMO” Classification) of Graves’ Ophthalmopathy According to Boergen ²³

Category	Grade	Clinical Feature
L, Lid changes	0	Absent
	1	Edema only
	2	Retraction only
	3	Retraction+ upper lid edema
	4	Retraction+ upper+ lower lid edema
E, Exophthalmus	0	Absent
	1	Without lid closure deficiency
	2	Conjunctival irritation in the morning only
	3	Permanent conjunctival irritation
	4	Corneal complications
M, Muscular changes	0	Absent
	1	Only visible on ultrasound/CT/MRI
	2	Pseudoparesis
	3	Pseudoparalysis
O, Optic nerve involvement	0	Absent
	1	On color sensitivity and VECP only
	2	Peripheral visual field defects
	3	Central visual field defects

 CT, computed tomography; MRI, magnetic resonance imaging; VECP, visually evoked cortical potential.

Table 3.

 

Fifteen Parameters Selected for Quantitative Analysis Characterizing the Dynamics of Saccadic Eye Movements

	Abbreviation	Unit	Saccadic Parameter, Definition
1	Amp	deg	Amplitude
2	Dur	msec	Duration
3	Lat	msec	Latency between stimulus movement and saccade onset
4	Vmax	deg/sec	Maximum velocity
5	Vave	deg/sec	Average velocity
6	CoV		Velocity constant (Vave/Vmax) 27
7	Amax	deg/sec2	Maximum acceleration
8	Dmax	deg/sec2	Maximum deceleration
9	TVmax	msec	Duration from onset to maximum velocity
10	TAmax	msec	Duration from onset to maximum acceleration
11	TDmax	msec	Duration from onset to maximum deceleration
12	VTAmax	deg/sec	Velocity at time of maximum acceleration
13	VTDmax	deg/sec	Velocity at time of maximum deceleration
14	Disconj	deg	Disconjugacy (abs [Amp RE− Amp LE])
15	Skew		Skewness (TVmax/Dur)

Figure 1.

 

Example of eye movement recordings of the 1st paradigm (target: Laser Ver/Hor; eye position: Right/Left Eye Ver/Hor). (A) Patient 1, ES; (B) control subject 1, ES.

Table 4.

 

Illustration of the Significant Differences between Patients and Controls

Saccade	Amp	Dur	Lat	Vmax	Vave	CoV	Amax	Dmax	TVmax	TAmax	TDmax	VTAmax	VTDmax	Disconj	Skew
1st Paradigm
10 ad cfug
10 ab cfug
10 up cfug
10 down cfug
10 ad cpet
10 ab cpet
10 up cpet
10 down cpet
20 ad cfug
20 ab cfug
20 up cfug
20 down cfug
20 ad cpet
20 ab cpet
20 up cpet
20 down cpet
2nd and 3rd paradigms
20 ad early
20 ab early
20 up early	▾
20 down early
20 ad late
20 ab late									▴
20 up late	▾
20 down late
40 ad early															▴
40 ab early											▴		▾		▴
40 up early
40 down early
40 ad late									▴						▴
40 ab late									▴
40 up late
40 down late

 Explorative study, highlighted by a grey field; both explorative and comparative studies, black triangle, the tip indicating the direction of change in the patients. Differences concerning the mean values of 480 data entries (15 saccadic parameters, 32 different eye movement conditions). Up, upgaze; down, downgaze; ad, ad-duction; ab, ab-duction. See text and Table 3 for other abbreviations.

Table 5.

 

Summary of 34 Significantly Different Saccadic Parameters Obtained in the Explorative Study

Saccade Condition	Parameter	Patient (mean)	Control (mean)	Difference	P	SD Patients	SD Controls
10 ad cfug	CoV	0.63	0.65	−0.02	0.027	0.02	0.02
10 ad cpet	CoV	0.62	0.64	−0.02	0.027	0.02	0.02
10 up cpet	Disconj	0.28	0.12	+0.16	0.024	0.18	0.05
10 down cfug	Amp	10.3	9.5	+0.8	0.009	0.8	0.4
10 down cpet	CoV	0.61	0.64	−0.03	0.016	0.03	0.01
20 ab cfug	CoV	0.67	0.69	−0.02	0.015	0.02	0.02
20 ad cpet	Lat	238	210	+28	0.034	25	28
20 ab cpet	Lat	238	210	+28	0.030	25	28
20 up cfug	Lat	298	240	+58	0.006	50	24
20 up cfug	Amp	16.4	17.8	−1.4	0.012	1.0	1.2
20 up cpet	Amp	17.9	19.2	−1.3	0.009	1.0	0.8
20 down cpet	Dmax	13038	9086	+3953	0.026	3504	3782
20 ad early	CoV	0.66	0.69	−0.03	0.006	0.02	0.02
20 ad early	TAmax	10.2	9.5	+0.7	0.001	0.4	0.4
20 ad early	VTDmax	170	140	+30	0.026	33	20
20 up early	Amp	17.8	18.7	−0.9	0.046	0.9	1.0
20 ab late	TVmax	34.7	32.2	+2.5	0.047	2.9	2.5
20 up late	Amp	18.3	19.3	−1.0	0.011	0.8	0.8
40 ad early	Skew	0.423	0.330	+0.093	0.001	0.061	0.042
40 ad early	Dmax	12451	7972	+4479	0.039	5263	3438
40 ad early	VTDmax	265	333	−68	0.033	76	54
40 ab early	Skew	0.371	0.311	+0.060	0.019	0.052	0.052
40 ab early	Dmax	10005	6181	+3823	0.008	3363	2015
40 ab early	TDmax	71.4	59.1	+12.3	0.006	10.0	7.5
40 ab early	VTDmax	331	399	−68	0.009	62	34
40 ad late	Skew	0.392	0.322	+0.070	0.020	0.073	0.043
40 ad late	TVmax	43.3	36.8	+6.5	0.020	6.9	3.7
40 ab late	Skew	0.355	0.299	+0.056	0.032	0.059	0.049
40 ab late	TVmax	39.1	34.0	+5.1	0.037	5.4	4.6
40 hor late	Disconj	0.47	0.30	+0.17	0.043	0.20	0.2
40 up late	Dur	113	137	−24	0.042	11	30
40 up late	Amp	35.9	37.4	−1.5	0.050	1.8	1.0
40 up late	Vmax	524	460	+64	0.043	48	79
40 up late	Dmax	8810	6296	+2514	0.038	1462	3234

Table 6.

 

Data of Parameters Showing an Intraindividual Saccadic Fatigue Effect in Patients and Control Subjects (Explorative Study)

Saccade Condition	Parameter	Before Fatigue	After Fatigue	Difference	P
Patients
20 deg adduction	TAmax	10.2	9.5	−0.7	0.006
20 deg upgaze	Skew	0.459	0.425	−0.034	0.022
20 deg upgaze	Dmax	16622	14226	−2396	0.049
40 deg abduction	Amp	37.1	38.0	+0.9	0.014
40 deg adduction	Amp	36.9	37.8	+0.9	0.013
40 deg adduction	Dmax	12450	9750	−2700	0.033
40 deg adduction	VTDmax	265	317	+52	0.039
40 deg downgaze	Amp	39.0	37.5	−1.5	0.022
40 deg downgaze	Vave	323	306	−17	0.012
Controls
20 deg abduction	Skew	0.461	0.446	−0.015	0.045
20 deg abduction	Vave	301	290	−11	0.025
20 deg abduction	CoV	0.682	0.669	−0.013	0.005
20 deg abduction	Dmax	18550	15989	−2561	0.032
20 deg adduction	Vave	299	286	−12	0.039
20 deg adduction	CoV	0.686	0.677	−0.009	0.007
20 deg adduction	Dmax	20344	17790	−2554	0.023
20 deg upgaze	Dur	73	79	+6	0.004
20 deg upgaze	Skew	0.422	0.382	−0.040	0.013
20 deg upgaze	Vave	265	252	−13	0.021
20 deg upgaze	CoV	0.636	0.612	−0.024	0.002
20 deg upgaze	Disconj	0.23	0.18	−0.05	0.038
20 deg upgaze	TAmax	9.9	9.4	−0.5	0.016
20 deg upgaze	VTDmax	212	232	+20	0.005
40 deg abduction	Lat	215	232	+17	0.014
40 deg abduction	Amp	37.8	38.4	+0.6	0.032
40 deg abduction	CoV	0.667	0.687	+0.020	0.000
40 deg abduction	Vmax	509	493	−16	0.010
40 deg abduction	TDmax	59.1	64.6	+5.5	0.009
40 deg adduction	Lat	216	233	+17	0.015
40 deg adduction	Amp	37.6	38.3	+0.7	0.036
40 deg adduction	CoV	0.683	0.704	+0.021	0.000
40 deg adduction	Vmax	496	477	−19	0.007
40 deg upgaze	Dur	126	137	+11	0.000
40 deg upgaze	Skew	0.298	0.268	−0.030	0.032
40 deg upgaze	Vave	313	287	−26	0.003
40 deg upgaze	Vmax	494	460	−34	0.003
40 deg upgaze	Amax	23470	21966	−1504	0.043
40 deg upgaze	Dmax	8400	6295	−2105	0.004
40 deg upgaze	VTAmax	268	245	−23	0.024
40 deg downgaze	Amp	38.4	37.4	−1.0	0.012
40 deg downgaze	Vave	297	287	−10	0.047
40 deg downgaze	Amax	22034	20745	−1289	0.001

Table 7.

 

Summary of 10 Saccadic Parameters Being Significantly Different in Both the Explorative and Comparative Studies

Saccade Condition	Parameter	Patient (mean)	Control (mean)	Difference	P	SD Patients	SD Controls
20 up early	Amp	18.2	18.9	−0.7	0.035	0.89	0.59
20 ab late	TVmax	33.3	27.6	+5.7	0.008	5.3	0.9
20 up late	Amp	18.7	19.7	−1.0	0.022	0.7	1.0
40 ad early	Skew	0.413	0.349	+0.064	0.002	0.044	0.034
40 ab early	Skew	0.357	0.309	+0.048	0.043	0.052	0.046
40 ab early	TDmax	70.5	54.1	+16.4	0.011	16.0	6.5
40 ab early	VTDmax	347	427	−80	0.010	71	50
40 ad late	Skew	0.398	0.346	+0.052	0.029	0.058	0.035
40 ad late	TVmax	46.5	35.6	+10.9	0.002	8.0	4.1
40 ad late	TVmax	41.1	30.9	+10.2	0.021	11.3	4.4

The eye movement recordings were performed in the laboratories of Thomas Brandt, MD, and Ulrich Büttner, MD, Department of Neurology of the Ludwig Maximilians University, Munich, Germany, with the invaluable support of Thomas Eggert, Klaus Bartl, and Sigrid Langer. The authors thank Roberto Bolzani, PhD, University of Bologna, Italy, for statistical advice, and Michael B. Reid, PhD, Baylor College of Medicine, Houston, Texas, for editorial advice.