| OCR Text |
Show !oumal of Clinical Neuro-ophthalmology 130): 59-66, 1993. Quantitative Longitudinal Assessment of Saccades in Huntington's Disease Allen J. Rubin, M.D., W. Michael King, Ph.D., Kirk Alan Reinbold, M.S., and Ira Shoulson, M.D. © 1993 Raven Press, Ltd., New York While participating in a controlled study of baclofen as protective therapy, 39 Huntington's disease (HD) patients underwent measurements of horizontal saccade latency and velocity, repeated longitudinally over a 2-year period. Significant worsening of saccade latency and of mean velocity was detected in untreated patients. Although individual variation was great, initial velocity impairment was found to be more prominent in younger patients. Factors are identified that may affect the rate of decline in supranuclear oculomotor function, including age and the severity of illness at the time of initial assessment. We propose that serial quantitative measurement of saccade performance is a useful clinical marker of the rate of disease progression against which the efficacy of treatments may be tested. Key Words: Huntington chorea-Saccadic eye movements-- Movement disorders. From the Department of Medicine (Neurology) (A.J.R.), University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School at Camden, Camden New Jersey; Departments of Physiology (W.M.K.) and Neurology (1.5.), University of Rochester, Rochester, New York; Department of Bioengineering (K.A.R.), University of Pennsylvama, PhIladelphia, Pennsylvania, U.S.A. Essential equipment for this project was donated by the Center for Visual Science of the University of Rochester. Efforts contributing to this project were supported by grants, including USPHS NS-17978 (1984-88), NEI: R03-EY06052-01 (1985) and by fellowship support from the United Parkinson Foundation. 59 Slowing of saccades in Huntington's disease (HD) has been reported as an early clinical feature useful in detecting onset of illness (1-3), described qualitatively, and shown to progress as a feature of advancing disease (4). Quantitative studies have shown that early age of onset of HD is associated with increases in horizontal saccade latency and decreases in velocity (5), but quantitative measures of the rates of decline in saccadic latency and velocity have not been established, Using a technique of oculomotor measurement that is noninvasive and readily accessible in a clinical setting (6), we seek to establish the rates of change of saccadic performance by serial quantification. We propose that this technique provides a neurophysiologic marker of disease progression. The reliability of such a marker for longitudinal study, in contrast to clinical examination of oculomotor or other motor systems, would derive from the intrinsic regularity in the performance of brainstem saccade generators. Except for minor spontaneous variations in performance (7), the saccadic system assessed in standardized test conditions is expected to provide the regularity of performance required for quantitative longitudinal analysis. SUBJECTS We measured horizontal saccade performance in patients who participated in a double-blind placebo- controlled study of baclofen as protective therapy in HD (8). Of 39 patients tested longitudinally, 17 were on placebo, and an additional 7 pilot study patients were followed whose treatment status remains unknown. All patients were recruited while in early stages (1 and II) of disease (Shoulson- Fahn Total Functional Capacity (TFC) scale> 6) (9,10), and in most cases were tested yearly over the course of the study. The initial test occurred at 60 A. J. RUBIN ET AL. a mean 17.1 months after recruitment (range: 6-24 months). Paired quantitative comparisons of saccadic performance over a 2-year interval were achieved in 39 patients for most variables. In the data presented, missing values account for instances in which total sample size is reported to be <39. For the sample, the age of onset of HO symptoms was 34.7 ± 10.7 years (mean ± SO), and age at the time of our first assessment was 42.2 ± 10.1 years. Ouration of illness was 8.1 ± 3.4 years, mean ± SO. TFC scores, as averaged among seven independent observers, ranged from 13 to 6 (mean 10.0 ± 1.7 SO) when patients were initially tested. TFC scores progressed in a 2-year interval to range from 12.8 to 2.8 (mean 8.3 ± 2.2 SO), encompassing functional Stages I-IV of disease. Because baclofen, a GABA analog, is clinically active in ocular motor systems (II), and the related GABA neurotransmitter system is physiologically involved in the premotor system for saccade generation (12), patients receiving baclofen were treated separately for analysis. In all cases, we have compared the rates of progression in eye movement measures over a 2-year epoch following randomization, avoiding the comparison of eye movement parameters between the premedication baseline and later measurements while on medication. This confines our inference for medicated patients to rates-of-change over time, controlling for direct drug effects on latency and velocity. Patients randomized to placebo were not permitted any psychoactive medication unless seriously disabled by movement or depression. None of the patients tested were treated with neuroleptics. METHODS Horizontal eye position was recorded by an infrared reflectance technique with 0.25° resolution (Gulf & Western Eyetrac Model 200). While abnormalities in saccades in HO may occur with more severe vertical than horizontal impairment (4), measurement of horizontal saccades with the infrared technique avoids the relative inaccuracy caused by the interference of the lid in vertical measurement, a potentially significant problem in HO patients who typically show an increased rate of blinking and blink-saccade synkinesis (13,14). Blink artifact was monitored independently by vertical OC electro-oculogram; saccades associated with blinks were excluded from analysis. System bandwidth was OC to 50 Hz. Testing was performed in the dark in a perimetric field, the equidistant targets viewed monocularly. The head was secured against both chorea and gaze-driven head movements in a stabilizing chin-forehead mount. I Clin Neuro-ophthalmol. Vol. 13. No. 1. 1993 Targets for saccades were red LEOs subtending OS visual angle, placed perimetrically at 5° intervals eliciting saccades 0;;40° in amplitude, and presented in pseudorandom sequence after unpredictable delays. All fixation targets were within 20° of primary gaze. Saccades of greater amplitude than 20° were therefore stimulated by peripheral targets. Eye position signals were displayed by chart recorder (Grass Model 78B polygraph) running at 60 mm/s for maximal resolution. Polygraph recordings of eye position were quantified by use of a digitizing pad (GTCG Digipad-5) operating with a LSI 11/23 computer. Eye velocity signals developed via analog differentiation were also monitored by polygraph display to improve visual resolution of onset and endpoints of the saccade epoch. For each saccade, the latency from target change, the total saccade duration, and the amplitude were derived. Such direct reading of eye movement records by visual inspection has provided consistent results by others (15). Typically, 80-120 saccades were analyzed for each patient. Summary measures were developed to characterize the saccade latency and velocity performance by an individual at each point in time. Latency is the interval (in milliseconds) from the unpredictable target change to the onset of saccadic eye movement; a mean latency for all saccades served as the summary measure. As saccade velocity varies with amplitude, we required a single value to characterize the velocity performance and allow comparison between patients who may make a paucity of saccades of a given amplitude, and allow comparisons over time. We calculated mean velocity (MY) for each saccade before any hypometric corrections, representing the duration of the total saccade divided by the total saccade amplitude, MV = amplitude/duration, and from the entire sample of saccades derived the amplitude- duration relation. This relation is adequately described by a linear equation (15). We extrapolated from this regression the velocity intercept for a 30° amplitude, designated as "projected mean velocity" (MVp)' This projected measure (MYp) is highly correlated (r = 0.88, P < .001), with the mean for each patient of actual MY of all saccades of amplitude 30° ± 5° taken together, supporting that this measure adequately describes the actual velocity performance. Linear regressions were not accepted for a patient if variability within a test performance yielded an amplitude-velocity correlation of less than 0.5. Published normal ranges for this measure would lead to expected MYp = 270-375°/s (15). Compa- SACCADES IN HUNTINGTON'S DISEASE 61 rable results were achieved using our methods in 29 normal controls showed a sample mean MVp = 249 ± 8 (mean ± SEM, range: 175--323). When we divide normal controls according to the median age (42) for HD patients, we note that older controls showed a mean MVp = 258 ± 10, and younger controls showed mean MYp = 238 ± 13. The precision of visual inspection of a polygraph record for determining saccade duration was validated at the last visit of most patients. Redundant recording in both polygraph and computerdigitized modes was available for a subset of 34 HD patients. (This instrumentation was not available, however, for the earlier longitudinal testing.) The visual inspection method was compared to velocity analysis in which both average and peak velocity were detected by a quantitative criterion. The choice of the end of the saccade on which the denominator of MY (saccade duration) is highly dependent, could be defined uniformly in computerized analysis as the point at which saccade velocity returned to a value 5% of the peak velocity. The correlation for MVp between the methods was .84 (p < .001). A physiologic relation between mean velocity and peak velocity is well recognized (15,16), providing the basis of further validation of our method. Saccade peak velocity (digitally derived), which is independent of the endpoint of the saccade, is also highly correlated with MYp (derived by visual inspection) in 34 HD patients (r = .78, P < .0001), and with MYp (digitally derived) (r = .81, P < .0001), where measures were developed redundantly in the same test performance. MVpis therefore correlated with a value extrapolated from the saccade "main sequence" (17). Assessment of reliability of MYp is approximated by comparison of test-retest values obtained in a subset of patients free of medication separated by the short interval of only 6 months. Mean velocity measures correlated significantly (r = .94, P< .01, n = 4) between test and retest. Latency measures also correlated significantly (r = .86, P< .005, n = 6) in this short interval. RESULTS The ability of some patients to generate saccades of large amplitude, which we define uniformly as those of ~26°, was affected selectively over a 2-year interval. In our study, 11 patients had asymmetric dropout of large saccades in one direction compared to their own performance on the same test 2 years earlier, and 3 more patients had complete dropout of large saccades, meaning that they could not generate any saccades of amplitude ~26°. As our measure of latency was affected by this selective dropout, we report longitudinal latency data for 25 patients whose capacity for generating large amplitude saccades was comparable over the 2 years. In a 2-year interval, significant worsening was detected in these HD patients of mean saccade latency, both in placebo and in baclofen- treated groups (Table 1) . Approximate rates of change are estimated from the differences in group means over 2 years. In the placebo group, an 8% per year increase was measured in mean latency of all saccades (paired t, p < .01). When we select a subset of large saccades, we find a 13% increase per year in large saccade mean latency (p = .05), and greater departure from the normal value of 208 ± 5 (mean ± SEM) obtained in 29 normal controls. These rates appeared higher in the patients randomized to the treatment group, increasing by 19% (p = .001) and 22% (p = .005) per year, respectively. Although at the first measurement there was little difference between the mean latency for all saccades and large saccades, latency was increased to 279 ± 71 (all saccades) TABLE 1. Saccade latency Latency in all saccades increases significantly in 2 years At first test 2 years later Latency in large saccades (amplitude ;;'26°) shows relatively greater rate of increase At fi rst test 2 years later Mean:!: SO (ms) Placebo Baclofen Both 223 ± 44 211 ± 44 247 ± 46 259 ± 36·· 290 :!: 46·· 279 ± 71··· N = 11 N=8 N = 25 253 ± 86 228 ± 46 250 ± 68 317 ± 73· 329:!: 63·· 314 ± 67" N = 11 N=8 N = 25 Paired t-test comparing first test to that 2 years later (*p < .05, ••p < .01, •••p < .001). Differences between placebo- and baclofen-treated groups at a given time are not significant. Combined group includes HD patients from pilot studies whose treatment status with respect to baclofen is unknown. Normal mean latency does not differ for small and large saccades: 208 ± 5 ms (mean ± SEM) in 29 controls. ] Clin Neuro-ophthalmol. Vol. 13. No. 1, 1993 62 A. J. RUBIN ET AL. and to 314 ± 67 ms (subset of large saccades only), greater than 100 ms above the normal value after 2 years. No association with the progressive impairment of latency was detected related to age, age of onset, or to duration of illness. Unexpected observations are made, however, that in some HD patients saccade latency often increases with saccade amplitude, a relationship rarely present in controls. In addition, prolongation of latency may occur asymmetrically, varying between saccades directed leftward or rightward. Similarly, in a 2-year interval significant worsening was detected in HD patients of saccade MVp' both in placebo and in baclofen-treated groups. For HD patients the sample mean MVp was 178 ± 52°/s declining over 2 years to 153 ± 45°/s (range: 40-2500 /s). Missing data were encountered less frequently because estimates derived from linear regression were still obtainable despite dropout of saccades of large amplitude. The rate of MVp decrease is approximately 6% per year in nonmedicated patients (paired t test, p = .02), and 10% per year in patients in the treatment group (paired t test, p = .009). Overall, in a sample of 38 HD patients an average 7% per year decline in saccade velocity is demonstrated (paired t test, p = .0002). These values are shown in Table 2. AGE EFFECTS ON SACCADE VELOCITY We and others (5,18) have reported previously that saccadic velocity defects in early stages of HD are less prominent with later age of onset. In this study, impairment in saccadic velocity is directly related to functional measures of disease progression (TFC). As shown in Fig. 1, mean saccade velocity (MVp) declines with progression of disease. Each of 38 HD patient's MVp in degrees per second is plotted against the mean TFC score obtained from 7 observers. Figure 1 also illustrates the effect of age by coding for four quartiles of age. The old- TABLE 2. Saccade mean velocity (MVp ) decreases significantly in 2 years Mean ± SD (o/s) est quartile are least impaired in saccade velocity, and do not show a decline in velocity with progressive lose of functional capacity (TFC). A regression line highlights the direct relation between functional capacity and saccade velocity that hold for the other three quartiles, and yields a correlation of r = .48, P < .01). Illustrated as well is the observation that for patients in Stages I and II of illness (i.e., TFC >6), saccade velocity often falls within the range of normal, and that this is more likely in the older patient. In this sample of early patients, age of onset was strongly associated (r = .95) with age and did not provide any additional explanatory potential. In Fig. 2 we plot the MVp for each HD patient against age, at the time of his first test and again 2 years later. For the whole sample there is an inverse relation of saccade velocity to advancing age (regression line). This demonstrated a comparison between groups of patients whose ages differ by decades. Individual to individual variation is great within any age group. Superimposed on this trend, distinct short-term losses in velocity over a 2-year epoch is apparent for many patients, sufficient in degree to account for the finding of significant longitudinal decline in velocity over time for the HD sample as a whole. An alternative illustration of rates of change in velocity is provided in Fig. 3, where MVp at the time of the first assessment is plotted against MVp as assessed 2 years later. Here the diagonal represents no change, individuals falling above the diagonal have shown small increments in velocity, and those falling below the diagonal demonstrate decrements in velocity over time. Individuals are coded in the figure for age, showing a possible trend that younger patients manifest greater velocity impairment, but that the rate of velocity decline (fall-away from the diagonal) may be greater in the older patients. Rate of decline is compared by age quartiles in Table 3. Differences attain or approach statistical significance within a 2-year epoch for the older two quartiles (mean 40 ± 48°/s decline) com- TABLE 3. Severity of saccade mean velocity (MVp) slowing and rate of decline is compared among age quartiles Paired t-test comparing first test to that 2 years later (*p < .05, **p < .01, ***p < .001). Differences between placebo- and baclofen-treated groups at a given time are not significant. Combined group includes HD patients from pilot studies whose treatment status with respect to baclofen is unknown. At first test 2 years later Placebo 181 ± 51 158 ± 39* N = 17 Baclofen 183 ± 44 147 ± 49*· N = 15 Both 182 ± 51 154 ± 43*" N = 38 Age Quartile Mean ± SD (%/s) Years N At fi rst test Two years later Decline 24-35 8 139 ± 43 136 ± 49 15 ± 28 36--41 11 152 ± 39 139 ± 35 18 ± 37 42-51 9 172 ± 43 155 ± 33 48 ± 47** 52-66 10 202 ± 47 187 ± 38 33 ± 49* Paired t-test comparing first test to that two years later (*p = .06, "p = .01). I Clill Neuro-ophthalmol, Vol. 13, No. 1, 1993 SACCADES IN HUNTINGTON'S DISEASE 63 350 ~Cl. 300 l~ > 0 TIl 1 fT 1 FIG. 1. Total functional capacity declines ::::E z '--'0 with loss of saccade velocity. Effects of ~ uw 250 age are illustrated by dividing the HD pa- u (/) To! J; fl. ~!I tients into four age quartiles: 0, ages 24- 0....J .(.../..)... 35; e, ages 36-41; 0, ages 42-51; and ., w w 200 o > w l1'D~'T6 + ages 52-66. In the oldest quartile, no rela- W 0:: tion of TFC to MVp appears to hold. A 0 <.:> T uu«« 0wz 150 01 61 .--..~....."'" .'1 ?~1l' ~ T1 drealsahtieodn orefgTrFeCssitoonMlVinpe fo(Ar)thileluostthraetresthrtehee T 1 11 (/) 100 A~.~. 0 quartiles. Comparable normal MVp values T 1 are 238 j: 13 (mean j: SEM) for ages below y ¢ the median age of 42, and 258 j: 10 above 50 the median age. 4 7 10 13 TOTAL FUNCTIONAL CAPACITY [TFC) pared to the younger two quartiles (17 ± 32°/s decline). DISCUSSION While it is recognized that supranuclear oculomotor defects occur in HD and progress with advancement of disease, we have demonstrated that significant decline can be quantified in a relatively early HD sample in a 2-year epoch. Incremental change in latency was measured in 88% of patients, and decremental change in velocity measured in 72%, this degree of abnormal change achieving statistical significance for both measures. While progressive impairment was not detected in every patient, such a measurement may represent a marker of neurodegeneration against which to test the success of protective therapeutic interventions in clinical trials, for example, those built on treatment strategies that modify putative excitotoxic mechanisms of neuronal loss (19-21). The ability to detect significant change in a 2-year epoch, makes this technique practical for application in clinical drug trials. The addition of welldefined quantitative biological markers to clinical rating scales for disease progression, presents an opportunity to increase statistical power in clinical trials in HD. Noting that many patients in Stage I and II of disease will have saccade velocities within the normal range, an electrophysiological decline may be nevertheless detectable, which presumably may be undetectable by clinical observation alone. The electrophysiological measurement of saccadic eye movement therefore adds an important complement to clinical observation in the tracking of HD progression. While the impairment of supranuclear saccadic dysfunction in HD is not localized to a specific region of neuropathological involvement (22), measurement of saccade generation may represent a strategy for detecting and tracking involvement (either degenerative or neurochemical) in subsystems in basal ganglia circuits for eye 350,......--------------------, 300 250 200 150 100 50 4----+--+---+-~r___+_-_+____1r___+_-_+____t 20 25 30 35 40 45 50 55 60 65 70 AGE (YEARS) FIG. 2. Change in mean saccade velocity (MVp ) is illustrated over a 2-year epoch for each HD patient, plotted against the age of the patient. 0, placebo-treated patients; ., baclofen-treated patients. A dashed regression line illustrated the inverse relation of saccade velocity to advancing age for the entire sample (r = .46, P < .01). I C/i" Neuro-ophthalmol. Vol. 13. No.1. 1993 A. ]. RUBIN ET AL. 100 150 200 250 SACCADE VELOCITY [MVp] AT FIRST TEST IN DEGREES/SECOND • FIG. 3. For each of 38 HD patients. saccade velocity (MVp) at the time of the initial test (To) is plotted against saccade velocity obtained 2 years later (T2 ). Patients who fall below the diagonal have declining velocity over time (MVp at To > MVp at T2 ). Patients are coded for age below the median (0. ages 24-41) and above the median (eo ages 42-66). 300 • •• 0 • • • • o • o 64 300 0:: w 3 250 (/) 0:: L5 200 >- 0~ ,......, 150 11. >~ ...... 100 50 50 movement control, or projections to brainstem premotor systems (23,24). In addition to measurements of the rates of decline, a number of specific observations have emerged regarding the natural history of saccade latency and saccade velocity in HD. Changes in latency were more detectable by the selective observation of large saccades, suggesting a greater sensitivity of this measure. Similarly, both in testing and clinically, some patients exhibited a selective dropout of large saccades with passage of time, particularly patients who had greatly slowed saccade velocity. This observation implies that the supranuclear saccadic disturbance of HD may apply earlier to saccades of large amplitude. Indeed, three patients with slowed velocity had complete dropout of large saccades after a 2-year interval as to obviate our capacity to make a quantitative interval comparison. We also noted that early patients were often asymmetrically impaired, showing differences by direction of saccadic refixation, and that this asymmetry persisted over time. In contrast, this asymmetry occurred less commonly in patients with advanced impairment of saccade velocity. Separating saccades by direction of refixation may augment sensitivity for detecting progression in early disease. Review of technical aspects of this study provides direction for improving future longitudinal studies. In an apparent paradox, some untreated patients showed modest improvement in velocity measures. This occurrence has questionable facevalidity in a progressive degenerative disease, and may be due to measurement error, to spontaneous variations in motor activity, or to circumstantial differences in each testing session, such as degree of alertness. An alternative technical explanation of this apparent paradox is offered as well. An estimate, such as MVp' which relies on linear regression is biased by the selective dropout of large saccades to weight regression heavily on small saccades, and may overestimate a projected intercept for a large amplitude. The bias deriving from dropout of large saccades may be averted by referring comparisons to a less eccentric amplitude as the linear intercept. While hand-digitized analysis of polygraph records of eye position was adequate for this present analysis, there would be an advantage in computerized techniques that derive saccade measures by uniform criteria, or provide detection of the absolute peak saccadic velocity. These techniques may offer advantages in reliability for longitudinal assessment. In the report of the clinical trial of baclofen, from which this sample of HD patients was drawn, a rate of TFC decline per year was reported of 0.53 ± 0.46 units per year on a 13-point scale, representing 4.1 ± 3.5% decline per year. This resembles closely the magnitude of progression of impairment we have described in saccade latency and in velocity. It was recognized that the baclofentreated patients in this study were not randomized with respect to mode of inheritance, and showed a 2:1 ratio of paternal:maternal inheritance. As we have detected more rapid rates of progressive impairment in both saccade latency and velocity in the baclofen-treated sample, we may conjecture that this effect may relate to cumulative treatment effects of medications active in the oculomotor system, to effects of mode of inheritance, or to the interaction of medication with the rate of progression of the underlying disease. Although age and age of onset were strongly correlated in this sample, an interaction may be f Clin Neuro-ophthalmol, Vol. 13, No.1, 1993 SACCADES IN HUNTINGTON'S DISEASE 65 present between age of onset and duration of illness (r = - .33, P < .06), sufficient to introduce a bias that younger and earlier onset patients have manifested illness for a longer period at the time of testing. This interaction, or the finding of generally lower TFC ratings in the younger patients, may indicate a floor effect ("burnout") for saccadic function in the patients with earlier onset and longer duration of illness. The floor effect may be present despite the controlling strategy that the study as a whole recruited patients restricted to early functional stages of disease. This interaction may account for the apparent conflict with recent studies establishing a more rapid rate of disease progression in patients with early age of onset, as measured neuropathologically (25,26). While we have confirmed an overall inverse association between age and saccadic velocity, the appearance of a relative damping of the rate of decline of saccade velocity in younger patients required control for duration of illness. Within the oldest quartile of patients, a subset of patients has relative preservation of saccade velocity or has a slower rate of decline of velocity over a 2-year interval. A reduced degree of motor dysfunction in the elderly-onset HD patient, perhaps excepting chorea, is asserted by some authors (25,27). The recognition in this sample of older patients whose total functional capacity declined despite preservation of saccade velocity, may imply that functional impairment was arising in these patients from other aspects of the disease, such as cognitive or emotional disturbances. Alternatively, eye movement disturbance may fail to reflect the degree of overall motor impairments in these patients. In summary, our study establishes the utility of quantitative longitudinal assessment of saccade latency and velocity for assessing the progression of Huntington's disease over a period of 2 years. Patients with less prominent eye movement impairment initially may appear to have a more rapid rate of measurable progression of their oculomotor dysfunction. This observation may derive from the age-effect of less impairment in patients with later age-of-onset, or may derive from the floor effect, where change is less detectable in patients with relatively severe impairment initially. We have found for saccade latency that the measurement of changes in saccades of large amplitude may be more sensitive, and for saccade velocity that the loss of the capacity to make large saccades in patients with slowed velocity may be a measure itself of disease progression. Longitudinal research design in HD should anticipate marked individual variation, effects of age of onset, and variation of rates of change attributable to the severity of illness at the time of initial assessment. Acknowledgments: Appreciation is expressed to Kathleen Gala, Allen D. Pettee, and Christopher R. Thorp for technical assistance. REFERENCES 1. Klawans HL, Goetz CG, Perlik S. Presymptomatic and early detection in Huntington's disease. Ann Neuro/1980;8: 343-7. 2. Leigh RJ, Newman SA, King WM. Vertical gaze disorders. In: Lennerstrand G, Zee D, Keller EL, eds. Functional basis of oculomotility disorders. Oxford: Pergamon Press, 1982:25766. 3. Oepen G, Clarenbach P, Thoden U. Disturbance of eye movements in Huntington's chorea. Arch Psychiatr Nervenkr 1981;229:205-13. 4. Leigh RJ, Newman SA, Folstein SE, Lasker AG, Jensen BA. Abnormal ocular motor control in Huntington's disease. Neurology (Cleve) 1983;33:1268-75. 5. Lasker AG, Zee DS, Hain TC, Folstein SE, Singer HS. Saccades in Huntington's disease: slowing and dysmetria. Neurology 1988;38:427-31. 6. Leigh RJ, Zee DS. The neurology of eye movements. Philadelphia: FA Davis, 1983:265. 7. Becker W. Metrics. In: Wurtz RH, Goldberg ME, eds. The neurobiology of saccadic eye movements. Amsterdam: Elsevier, 1989:13-67. 8. Shoulson I, Odoroff C, Oakes D, et al. A controlled trial of baclofen as protective therapy in early Huntington's disease. Ann Neural 1989;25:252-9. 9. Shoulson I, Fahn S. Huntington's disease: clinical car and evaluation. Neurology 1979;29:1-3. 10. Shoulson I, Kurlan R, Rubin AJ, et al. Assessment of functional capacity in neurodegenerative movement disorders: Huntington's disease as a prototype. In: Munsat TL, ed. Quantification of neurologic deficit. Boston: Butterworths, 1989:271--83. 11. Halmagyi GM, Rudge P, Gresty MA, Leigh RJ, Zee DS. Treatment of periodic alternating nystagmus. Ann Neural 1980;8:609-11. 12. Hikosaka 0, Wurtz RH. Modification of saccadic eye movements by GABA-related substances: parts I and II. JNeurophysioI1985; 53:266-308. 13. Karson CN, Burns RS, LeWitt PA, Foster NL, Newman RP. Blink rates and disorders of movement. Neurology (Cleve) 1984;34:677--8. 14. Zee DS, Chu FC, Leigh RJ, et al. Blink-saccade synkinesis. Neurology 1983;33:1233-6. 15. Becker W. Metrics. In: Wurtz RH, Goldberg ME, eds. The neurobiology of eye movements. Amsterdam: Elsevier, 1989: 13-68. 16. King WM, Lisberger 5, Fuchs A. Oblique saccadic eye movement of primates. JNeurophysiol 1986;56:769-84. 17. Bahill AT, Clark MR, Stark L. The main sequence, a tool for studying human eye movements. Math Biosci 1975;24:191204. 18. Rubin AJ, King WM, Snyder L, Shoulson I. Oculomotor distractibility in early Huntington's disease. Neurology 1985;35(SuppI.1):176. 19. Shoulson I. Huntington's disease: anti-neurotoxic therapeutic strategies. In: Fuxe K, Roberts P, Schwarcz R, eds. Excitotoxins. Wenner-Gren center international symposium series. Vol. 30. London: Macmillan Press, 1983:343-53. 20. Schwarcz R, Meldrum B. Excitatory amino acid antagonists provide a novel therapeutic approach to neurological disorders. Lancet 1985;2:140-3. I Clill Neuro-ophthalmol, Vol. 13, No.1. 1993 66 A. f. RUBIN ET AL. 21. Freese A, Swartz KJ, During MJ, Martin JB. Kynurenine metabolites of tryptophan: implications for neurological diseases. Neurology 1990;40:691-5. 22. Leigh RJ, Parhad 1M, Clark AW, Buettner-Ennever JA, Folstein SE. Brainstem findings in Huntington's disease: possible mechanisms for slow vertical saccades. J Neurol Sci 1985;71:247-56. 23. Wurtz RH, Hikosaka 0. Role of the basal ganglia in the initiation of saccadic eye movements. Prog Brain Res 1986; 64:175-90. 24. Hikosaka 0, Wurtz RH. The basal ganglia. In: Wurtz RH, I Clin Neuro-ophthalmol. Vol. 13, No. 1, 1993 Goldberg ME. The neurobiology of saccadic eye movements. Amsterdam: Elsevier, 1989:257-81. 25. Myers RH, Vonsattel JP, Stevens BA, et al. Clinical and neuropathological assessment of severity of Huntington's disease. Neurology 1988;38:341-7. 26. Myers RH, Sax OS, Schoenfeld M, Bird ED, et al. Late onset of Huntington's disease. JNeurol Neurosurg PSYChUltry 1985; 48:530-4. 27. Harper PS. Huntington's disease. London: WB Saunders, 1991:51. |
