| OCR Text |
Show Journal of Neiiro- Ophthalmology 18( 1): 36- 39, 1998. © 1998 Lippincotl- Raven Publishers, Philadelphia A New Clinical Technique for Demonstrating Changes in Eye Acceleration During Horizontal Saccades in Patients with Partial Internuclear Ophthalmoplegias Peter Brown, M. D. The eyelid- mounted accelerometer can pick up the acceleration waveform of the eye during horizontal eye movements. The acceleration profile comprises high- amplitude pulsatile activity in the saccade and changes in the level of background ocular microtremor related to eye position. Adducting saccades of 20° were recorded in eight patients with partial internuclear ophthalmoplegias caused by multiple sclerosis and in eight age-matched healthy subjects. The initial pulse of acceleration activity was reduced by 85% in the patients. In the worst- affected cases, adducting saccades were associated only with an increase in the level of background ocular microtremor in the acceleration trace. The results confirm the hypothesis that an internuclear ophthalmoplegia is due to the loss of the pulse signal to ocular motor neurons, with preservation of the step signal in an adducting saccade. Key Words: Eye acceleration- Internuclear ophthalmoplegia- Ocular microtremor. Changes in the acceleration recorded during a saccade give an idea of the variation in forces acting on the eye. The fusion frequency of extraocular muscle is - 400 Hz ( 1), so synchronous fluctuations in the firing frequency of motor units at well below this should be discernible as fluctuations of force and therefore acceleration. Brown and Day ( 2) recently described two techniques for the recording of eye acceleration. In one, a lightweight accelerometer was mounted to a suction contact lens and, in the other, an identical accelerometer was fixed over the corneal convexity through the closed eyelid. Both techniques gave very similar acceleration profiles during horizontal saccades. In addition, surface- recorded extraocular muscle electromyography mirrored the acceleration activity in both the time and frequency domains, suggesting that the various features in the acceleration waveform were due to the rhythmic and synchronous modulation of eye muscle discharge. Saccades took two Manuscript accepted 4/ 25/ 97. From the MRC Human Movement and Balance Unit, Institute of Neurology, London, England. Address correspondence and reprint requests to Dr. P. Brown, MRC Human Movement and Balance Unit, Institute of Neurology, London WC1 3BG, U. K. general forms: some consisted of 2- 3 large pulses of activity, whereas others consisted of an initial big pulse followed by polyphasic activity of higher frequency that continued for the rest of the saccade. An accelerometer can also be used to record the background microtremor of the eye in the different positions of horizontal gaze ( 3,4). As with other methods, the predominant frequencies represented in the microtremor are around 40 and 80 Hz ( 5,6). These frequency peaks may reflect the rhythmic discharge of extraocular eye muscle, because the eye is heavily overdamped and unlikely to exhibit much in the way of resonance phenomena ( 7,8). The eyelid accelerometer technique may provide a simple means of recording acceleration profiles during patients' horizontal eye movements. The present investigation was aimed at further validating the technique while demonstrating that this method of recording can be used in patients. With these objectives in mind, patients with partial internuclear ophthalmoplegias ( INOs) were studied. The results confirm the hypothesis that an INO is caused by the loss of the pulse in the normal saccadic pulse step signal to ocular motoneurons. METHODS Studies were performed with the understanding and written consent of each subject and with approval of the local ethics committee. Eight patients with clinically definite multiple sclerosis were studied ( mean age, 43; age range, 24- 60 years; four women). Three patients had unilateral and five had bilateral partial INOs. The worst-affected eye was recorded in each case ( the mean peak velocity of the adducting eye in horizontal saccades was 1677s; range, 63- 2207s). Eye movements were also recorded in eight healthy volunteers ( mean age, 43; age range, 28- 65 years; three women). Acceleration was recorded by using a linear accelerometer because this was lightweight and yet recorded the pattern of acceleration of the eye ( although it did not provide direct recordings of rotational acceleration). The technique has been previously reported in detail ( 2). The eyelid of one eye was taped gently shut and the acceler- 36 EYE ACCELERATION IN INTERNUCLEAR OPHTHALMOPLEGIA 37 ometer mounted over the eyelid at the point of maximal convexity ( with the cornea and pupil underlying this). Double- sided tape was used to secure the accelerometer to the lid, and the accelerometer lead was taped to the forehead. The accelerometer was orientated so as to be most sensitive to horizontal movements. The accelerometer consisted of a semiconductor strain gauge bonded to a simple cantilever beam with a mass at its end. The device weighed 0.5 g and had a linear response range (± 0.5 dB) up to 500 Hz ( EGAXT- 50, En-tran, Fairfield, New Jersey). The response to acceleration in a plane other than the direction of sensitivity of the accelerometer was < 3% of the output to the same acceleration in the device's direction of sensitivity. The accelerometer signal was amplified with a very low- noise, low- distortion instrumentation amplifier ( INA 103, Burr- Brown, Tucson, Arizona). Acceleration signals were digitally low- pass filtered 40 dB down at 600 Hz. Eye movements were recorded by using the monocular electro- oculogram ( EOG). Electrodes ( 9- mm silver discs) were taped just adjacent to the inner and outer canthus of each eye, and a ground electrode was taped to the middle of the subject's forehead. The EOG was amplified ( D150, Digitimer Ltd., Welwyn Garden City, Hertfordshire, United Kingdom) with high- frequency and low- frequency responses 3 dB down at 300 Hz and 0.53 Hz, and monitored on an oscilloscope by the experimenter. The level of illumination was kept steady. Seated subjects were instructed to make self- paced saccades between two light- emitting diodes arranged so as to be straight ahead and 20° in the direction of adduction of the tested eye. They were asked to hold each position of gaze for ~ 4 s before the next saccade. Head movements were restrained. Each recording session took - 30 min to complete. All signals were digitized with 12- bit resolution ( 1401- plus analogue- to- digital converter, Cambridge Electronic Design, Cambridge, United Kingdom), and collected and analyzed on a personal computer by a software package ( CED Spike2). The sampling rate was 400° s" 1 second B EOG Primary Acceleration c EOG Primary 20° adduction Acceleration 200ms FIG. 1. Adducting and abducting saccades of the right eye of a patient with a right partial internuclear ophthalmoplegia ( INO). A: An extended record showing the monocular electro- oculogram ( EOG) and acceleration recorded with an eyelid accelerometer ( traces 2 and 3) in the saccades. The traces 4 and 5 are the high- pass- filtered acceleration ( 40 dB down at 28 Hz) and the rectified and integrated high- pass- filtered acceleration. The slope of the latter trace is proportional to the level of ocular microtremor in each eye position. Record A begins with the eye in the primary position. A pathologically slow adducting saccade is then made ( see the differentiated and smoothed EOG in the top trace). The first boxed area has been expanded in B, which shows that pulsatile activity is absent from the acceleration trace during the adducting saccade. The ocular microtremor becomes more pronounced, however, with the eye adducted, as demonstrated by the steeper slope in the bottom trace in A. After a few seconds, the eye returns to the primary position. The second boxed area, which includes the abducting saccade, is shown in greater detail in C. A large acceleration- deceleration pulse is followed by a brief burst of polyphasic activity. The ocular microtremor in A returns to its former level after the saccade. J Neuro- Ophlhalmol, Vol. 18, No. I, 1998 38 P. BROWN 2,080 Hz. Amplitude and latency were measured by visual inspection using movable cursors on screen. The peak- to- peak acceleration of the initial acceleration-deceleration pulse ( defined as the biggest wave peaking within 10 ms of the onset of the saccade as determined by the EOG) was measured for each of 10 successive sac-cades made in the same direction. The initial acceleration- deceleration pulse was the largest pulse in the sac-cades of all the controls and patients, with the exception of the adducting saccades in two patients. In these, the maximum amplitude of the later polyphasic acceleration activity was 156% and 205% greater than that of the pulse at the onset of the saccade. The peak- to- peak amplitudes of the initial pulses were averaged to give the peak- to- peak acceleration in adducting and abducting saccades for each subject. Comparisons were made using the Mann- Whitney U test. Ocular microtremor was also analyzed in the four patients with the lowest peak- to- peak accelerations in adducting saccades. Preliminary studies in four healthy subjects confirmed that the power spectrum of the ocular microtremor recorded with an eyelid- mounted acceler-ometer is very similar to that recorded with a contact-lens accelerometer ( except that the total power recorded with the latter technique is ~ 2 times greater). Spectra have two main peaks at around 40 and 80 Hz, and the total power in such spectra increases by 31- 94% when the eye is adducted or abducted 20° ( unpublished observations). Ocular microtremor was analyzed in the patients by first digitally high- pass filtering the acceleration signal so that it was 40 dB down at 28 Hz. This was necessary to remove direct- current offset and activity due to slow head and eye tremor. The high- pass- filtered acceleration signal was then digitally rectified and integrated as in Fig. 1. 1.6 t Peak I- 2 Acceleration ( ms"!) 0.8 H 0.4 Controls ( n= 8) Partial INOs ( n= 8) FIG. 2. Peak- to- peak acceleration recorded with an eyelid-mounted accelerometer in healthy subjects and in patients with partial internuclear ophthalmoplegias ( INOs) during 20° adducting saccades made from the primary position. Each value represents the mean of the peak- to- peak amplitude of the initial acceleration- deceleration pulse measured in 10 successive saccades for each subject. The interrupted horizontal lines give the means across subjects. The peak- to- peak acceleration differed in controls ( mean, 0.94 ± 0.12 ms~ 2; p < 0.01) and patients ( mean, 0.14 + 0.04 ms- 2). nounced, however, with the eye adducted, as demonstrated by the steeper slope of the bottom trace ( the rectified and integrated high- pass- filtered acceleration) in Fig. 1A. After a few seconds, the eye returns to the primary position. This abducting saccade is shown in greater detail in Fig. 1C. There is a large acceleration-deceleration pulse followed by a brief burst of polyphasic activity. The ocular microtremor in Fig. 1A returns to its former level after this saccade. RESULTS DISCUSSION The peak- to- peak acceleration recorded in the adducting saccades of the patients was reduced by 85% compared with age- matched healthy subjects ( Fig. 2). Similarly, the ratio of the peak- to- peak acceleration in abducting versus adducting saccades in the same eye was very different in patients ( mean, 4.8 + SEM 0.7) and controls ( mean, 1.5 ± 0.2; p < 0.001). Visual inspection of the acceleration traces of adducting saccades in the patient group confirmed that the initial and later pulses of acceleration activity were equally impaired. In the four most severely affected patients, pulsatile activity was absent and adducting saccades were associated only with a step increase in the background ocular tremor in the acceleration trace. This is shown in Fig. 1A for a patient with a partial right INO. The record begins with the eye in the primary position. A pathologically slow adducting saccade is then made ( see the differentiated EOG in the top trace). This saccade has been expanded in Fig. IB, which shows that pulsatile activity is absent from the acceleration trace during the movement. The ocular microtremor becomes more pro- We have previously postulated that the vibrations picked up by an accelerometer fixed over the closed eyelid reflect the synchronous discharge of many extraocular motor units ( 2). Normally, horizontal saccades are thought to be due to a pulse and step change in ocular motor neuron- firing rate in response to signals originating from neurons in the parapontine reticular formation. These signals reach the ocular motor neurons via axons running in the medial longitudinal fasciculus ( MLF). The axons of the MLF are thought to be damaged in patients with INOs. Partially damaged axons transmit low-frequency signals better than high- frequency ones ( We-densky phenomenon) and, as a result, the synchronous burst ( or bursts) of high- frequency activity comprising the pulse is particularly affected in lesions of the MLF. It is believed that the loss of the pulse leads to impaired acceleration and slow adducting saccades ( 9,10). Here we confirm that the initial acceleration pulse is indeed diminished in adducting saccades in patients with partial INOs. In the most severely affected cases, the slowed saccade seems to be accomplished solely by the step J Neuro- Ophtlmlmol, Vol. IS, No. I, 1998 EYE ACCELERATION IN INTERNUCLEAR OPHTHALMOPLEGIA 39 increase in extraocular muscle activity, manifest as an increase in the microtremor of the eye. The eyelid- mounted accelerometer technique is easy to use, taking just a few minutes to set up. It is well tolerated by patients and may provide a simple means of recording the detailed acceleration profile of the eye during horizontal eye movements. Acknowledgment: I thank Dr. P. Rudge and Dr. P. Riordan- Eva for referring some of the patients. REFERENCES 1. Fuchs AF, Luschei ES. Development of isometric tension in simian extraocular muscle. J Physiol 1971; 219: 155- 66. 2. Brown P, Day BL. Eye acceleration during large horizontal sac-cades in man. Exp Brain Res 1997; 113: 153- 7. 3. Bengi H, Thomas JG. Fixation tremor in relation to eyeball- muscle mechanics. Nature 1968; 217: 773- 4. 4. Bengi H, Thomas JG. Studies on human ocular tremor. In: Kenedi RM, ed. Perspectives in biomedical engineering. London: Mac-millan, 1972: 281- 92. 5. Coakley D. Minute eye movement and brain stem function. Boca Raton, FL: CRC, 1983. 6. Eizenman M, Hallett PE, Frecker RC. Power spectra of ocular drift and tremor. Vis Res 1985; 25: 1635- 40. 7. Carpenter RHS. Movements of the eyes. London: Pion, 1977. 8. Ashton JA, Boddy A, Donaldson 1ML, Lakie M, Walsh EG, Wright GW. Dynamics of movements of the cat's eye. J Physiol 1985; 358: 40P. 9. Pola J, Robinson DA. An explanation of eye movements seen in internuclear ophthalmoplegia. Arch Neurol 1976; 33: 447- 52. 10. Baloh RW, Yee RD, Honrubia V. Internuclear ophthalmoplegia. Arch Neurol 1978; 35: 484- 9. J Neiiro- Oplillialmol. Vol. 18, No. I, 1998 |
