Smooth Pursuit Eye Movement Deficits After J Neurol Neurosurg Psychiatry: First Published As 10.1136/Jnnp.56.7.799 on 1 July 1993

Journal ofNeurology, Neurosurgery, and Psychiatty 1993;56:799-807 799

Smooth pursuit eye movement deficits after J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from pontine nuclei lesions in humans

B Gaymard, C Pierrot-Deseilligny, S Rivaud, S Velut

Abstract This lesion was located along the lateral edge Eye movements were recorded electro- of the basis pontis, and affected about two oculographically in four patients with thirds of the rostrocaudal extent of the pons basal pontine lesions, demonstrated by (fig 2). When eye movement recordings were MRI. The most prominent eye move- made, one week after surgery, the left hemi- ment abnormality observed was mild to plegia had not improved, though the patient severe impairment of smooth pursuit was alert and cooperative. and optokinetic nystagmus, mainly ipsilateral to the lesion. This abnormality is PATENT 2 thought to result from damage to the A 59 year old woman was hospitalised after pontine nuclei, which form a crucial the sudden onset of right hemiplegia. On ini- relay between the cerebral cortex and the tial examination, severe right hemiplegia was cerebellum controlling smooth pursuit. observed. On the left side, strength was nor- Abnormalities of saccades and the mal but there was dysmetria in the finger to vestibulo-ocular reflex in one patient are nose test. All sensory examination was nor- also discussed. mal. MRI showed a left pontine lesion, compatible with an infarct (fig 3). A transverse (9Neurol Neurosurg Psychiatry 1993;56:799-807) section showed that the lesion affected almost all the left basis pontis, and slightly crossed the midline (fig 3A). The pontine tegmentum was spared, excepc-for a limited part of the The pontine nuclei (PN) consist of several median structures, on which a very narrow groups of neurons in the basis pontis relaying posterior extent of the lesion impinged (fig 2). signals from the cerebral cortex towards the Sagittal sections clearly showed that the cerebellum. Some of them, namely the dorso- lesion had a short rostrocaudal extent, and lateral, the lateral, and the dorsomedian PN, consisted of a thin horizontal slice (about could be specifically involved in smooth pur- 5 mm) at the mid-pons level (fig 3B). MRI suit eye movements, and represent a crucial did not show any lesions in the superior or link in the corticopontine pathways mediating inferior cerebellar peduncles or in the cerebel- this eye movement.'23 However, the location lum. Eye movement recordings were per- and function of the PN involved in smooth formed five months after the stroke. At that http://jnnp.bmj.com/ pursuit still have to be precisely determined time, the motor deficit and the cerebellar in humans. The first human case with a selec- syndrome persisted. The patient was alert tive PN lesion and including eye movement and cooperative. recordings was recently reported.4 It provided data consistent with the results of experimen- PATENT 3 tal lesion studies involving the dorsolateral A 34 year old woman with no medical back-

pontine nuclei (DLPN) in the monkey. We ground was hospitalised after the sudden on September 27, 2021 by guest. Protected copyright. report four patients with diverse vascular onset of pure left hemiparesis. MRI showed a lesions affecting the basis pontis, resulting in lesion restricted to the basis pontis, compati- various impairments of smooth pursuit and ble with an infarct. It was located just below optokinetic nystagnus (OKN). The relation the mid-pons level, along the midline, with a Clinique Neurologique between smooth pursuit impairment and the limited extent on the right side (fig 2). On and Unit6 INSERM location of the lesions is discussed. sagittal sections, it occupied approximately 289, Hopital de la Salpetriere, Paris, and 30% of the whole rostrocaudal extent of the Service de pons. Eye movement recordings were per- Neurochirurgie, Case reports formed one month after the stroke. At that H6pital Bretonneau, Tours, France PATENT 1 time, there was no improvement in the motor B Gaymard A 73 year old woman had been complaining deficit, though the patient was alert and C Pierrot-Deseilligny for a year and a half of a right trigeminal neu- cooperative. S Rivaud S Velut ralgia despite carbamazepine therapy. She thus to a and PATINT 4 Correspondence to: agreed neurosurgical procedure, B Gaymard, INSERM 289, a microvascular decompression of the fifth A 66 year old woman with a long history of H6pital de la Salpetriere, 47 nerve was surgery, the experienced isolated Bd de l'H6pital, 75651 performed. After hypertension suddenly Paris, Cedex 13, France patient awoke with left hemiplegia. There left hemiplegia. MRI showed a lesion com- Received 30 March 1992 was no cerebellar symptoms, no sensory loss, patible with an infarct, located on the right and in revised from no a side of the at the level. 11 August 1992. and diplopia. MRI showed right pontine brainstem, mid-pons Accepted 25 September 1992 lesion, compatible with a haematoma (fig 1). The lesion was restricted to the basis pontis, 800 Gaymard, Pierrot-Deseilligny, Rivaud, Velut J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from http://jnnp.bmj.com/ on September 27, 2021 by guest. Protected copyright.

Figure I MRI ofpatient 1 (T, weighted, TR = 500 ms, TE = 10 ms) shows a rightpontine lesion, compatible with a haematoma, located in the lateralpart ofthe basis pontis, with a marked rostrocaudal extent. (A) pontomesencephalic junction, (B) upper pons, (C) mid-pons, (D) lower pons.

except on the midline where it extended Eye movements slightly into the tegmentum (fig 2). On sagit- For all patients, eye movements were record- tal sections of the brainstem, it occupied ed using direct current electro-oculography approximately 40% of the rostrocaudal extent (EOG), with the head immobilised. Two of the pons. Eye movements were recorded temporal electrodes were used for horizontal six weeks after the stroke. By this time, the eye movements, and two others were located motor deficit had improved moderately, and above and below the right eye for vertical eye the patient was alert and cooperative. movements. Smooth pursuit eye movement deficits after pontine nuclei lesions in humans 801

E 1'atient 1 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from

Z Patient 2 A

Patient 3

Patient 4 B

Figure 2 Drawing of the lesions of thefourpatients based on MRI scans (drawn by the authors). (A) midbrain, (B) upper pons, (C) mid-pons, (D) lower pons, an = abducens nucleus, ml = medial lemniscus, mlf = medial longitudinalfasciculus, R = right. The dorsoventral broken line passes through the peduncular PN and clearly separates the DMPN, laterally, and the DMPN, medially. http://jnnp.bmj.com/ Saccades and smooth pursuit were elicited in a dimmed room with a ramp of light emit- ting diodes (LED), located 80 cm in front of the subject. Horizontal visually guided saccades were studied by asking the subject to Figure 3 MRI ofpatient 2 (T, weighted, TR = 600 ms, as fast as possible at a suddenly appear- TE = 12 ms) shows a left pontine lesion, compatible with look an infarct. (A) horizontal section at the mid-pons level. ing 20° lateral target. Horizontal smooth pur- The lesion involved almost all of the left basis pontis, with suit was induced by asking the subject to slight extension, medially into the tegmentum and on September 27, 2021 by guest. Protected copyright. track an LED moving sinusoidally with an contralaterally. (B) sagittal section. The lesion had a amplitude of ±200, at various speeds (peak limited rostrocaudal extent. velocities of 22-5, 29, and 45°/s for patient 1, and 14-5, 22-5, 29, and 45°/s for the other patients). For each speed, smooth pursuit For patient 1, horizontal OKN was gain (defined as the ratio of peak eye velocity obtained by the manual displacement of alter- to peak stimulus velocity), was calculated for nate black and white stripes over a large part each lateral direction by averaging the gain of the visual field. The horizontal vestibulo- obtained in 10 successive cycles. The calcula- ocular reflex (VOR) was studied in a dimmed tion of smooth pursuit gain was made from room on an armchair rotating with an ampli- the chart (speed of the chart paper 10 mm/s). tude of ± 200, without stimulation being Furthermore, an index of asymmetry of recorded, and VOR suppression was tested by smooth pursuit was defined as: asking the patient to fixate a target rotating right gain - left gain with the armchair. These different eye move- x 100 ments were analysed qualitatively. right gain + left gain For the other patients, OKN was elicited Vertical saccades and smooth pursuit were by vertical black and white stripes, covering recorded under the same conditions as for 500 of the visual field and moving regularly at horizontal eye movements, but only qualita- 10, 20, and 300/s. The horizontal VOR was tively analysed. studied in complete darkness, using a chair 802 Gaymard, Pierrot-Deseilligny, Rivaud, Velut

rotating at 0-25 and 0 40 Hz with a 200 t I~~ amplitude on each side of the midline, under J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from two conditions. Under the first condition, the patient was asked to fixate an earth-fixed visual target while being rotated, which tested e both the VOR and smooth pursuit. Under the A second condition, the visual target was turned off, but the patient was asked to imagine it while being rotated, which tested only the VOR. Ability to suppress the VOR was also studied, the patient being asked to fixate a target rotating with the armchair. Mean gains were calculated for OKN, the VOR, and VOR suppression (ratio of peak eye velocity to peak head velocity). Two control groups, each of 10 normal subjects, were used, because patient 3 was markedly younger than the other three. The B first control group had a mean age of 60 years (range: 51-67 years), and the second a mean age of 35 years (range: 25-44 years). The results of and 4 were .wo + -40 . > 0000 patients 1, 2, compared a with those of the first control group, and the results of patient 3 with those of the second control group.

C Results PATIENT 1 Horizontal saccades were accurate and had latency and velocity analogous to those of the first control group (fig 4A and table 1). Smooth pursuit gain was markedly decreased bilaterally, even for a low stimulus velocity (fig 4B and table 2). Furthermore, there was marked asymmetry for each speed (fig 7) with D a more severe deficit for rightward smooth pursuit (fig 4B and table 2). Therefore, many catch up saccades were needed to follow the target. OKN slow phase was clearly asymmetrical, with a slower rightward slow phase (fig

4G). http://jnnp.bmj.com/ The VOR was symmetrical and qualitatively normal (fig 4D). VOR suppression was impaired bilaterally, as shown by the presence of a vestibular nystagmus, even for slow rota- 200 tions, without obvious qualitative right-left asymmetry (fig 4E). is Vertical saccades appeared normal, with no Figure 4 Horizontal eye movement recordings ofpatient 1. In figures 4, 5, and 6, dysmetria. Both upward and downward on September 27, 2021 by guest. Protected copyright. rightward eye movements are represented by an upward trace, and leftward eye movements smooth pursuit appeared severely impaired, by a downward trace. (A) Saccades, with normal latency, amplitude, and velocity. (B) Smooth pursuit, saccadic bilaterally though mainly rightwards. (C) OKN, asymmetrical, being almost entirely replaced by catch up with slower rightward slow phase (horizontal arrows indicate the direction ofstimulation, saccades. and vertical arrows a change in the direction ofstimulation). (D) VOR, normal. (E) VOR suppression, impaired bilaterally, without visible asymmetry. e = eye, t = target. PATIENT 2 Horizontal visually guided saccades were accurate, with latencies analogous to those of Table 1 Horizontal saccades latencies and velocity the first control group, but with markedly Latncy (ms) Veloc4y reduced velocities, bilaterally (fig 5A and (rIs) table 1). R L R L Smooth pursuit was impaired bilaterally Patient 1 300 270 402 399 (fig 5B and table 2), but predominantly Patient 2 225 240 120* 144* Patient 3 180 200 335 340 towards the left side, with marked asymmetry Patient 4 228 230 355 410 for 145, 22-5 and 290/s target velocities (fig First control group 237 (47) 236 (59) 327 (72) 325 (72) 7). This asymmetry disappeared at higher tar- Second control get velocities. The OKN slow phase gain was group 197 (30) 192 (28) 346 (60) 331 (58) decreased bilaterally, though more markedly In this and all subsequent tables patients 1, 2, and 4 are com- leftwards (fig 5C and table 3). pared with the first control group and patient 3 is compared with the second control group. L = left, R = right; *Outside Under the first condition (the patient fixat- the normal range (mean (2 SD)). ing an earth-fixed target), the VOR gain was Smooth pursuit eye movement deficits afterpontine nuclei lesions in humans 803

Table 2 Horizontal smooth pursuit gain (SD)

Patient 1 Patient 2 Patient 3 Patient 4 First control group Second control group J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from Gain Gain Gain Gain Mean gain Mean gain Target (SD) (SD) (SD) (SD) (SD) (SD) peak velocity R L R L R L R L R L R L 14 5°/s 0-69* 0.50* 0-71* 0-94 0-96 0-98 0-98 1.01 1 00 1 00 (0-17) (0-15) (0-12) (0 10) (0 05) (0 05) (0 06) (0 07) (0 05) (0 07) 22-5°/s 0-13* 0-28* 0-62* 0-45* 0-68* 0-83* 0-78* 0-83* 1-02 0.99 1-05 1-06 (0 10) (0 10) (0-12) (0-13) (0 11) (0-15) (0-08) (0-07) (0-05) (0-06) (0 06) (0-06) 29°/s 0.10* 0-28* 0-62* 0-26* 0-63* 0.75* 0.70* 0.73* 094 0-93 1 00 1-03 (0-05) (0-09) (0-16) (0-07) (0-13) (0-13) (0 08) (0-08) (0 09) (0 09) (0 06) (0-04) 45°/s 0.05* 0-20* 0 19* 0-22* 0-56* 0-64* 0.53* 0-61 0-85 0-82 091 0-92 (003) (007) (0 10) (0-13) (0 11) (0 11) (008) (006) (006) (0-13) (005) (004) The side of the lesion is denoted in bold. (See also the legend ofTable 1.)

analogous to that of the control group (fig 5D I J tv I ;_ and table 4). Under the second condition, rightward VOR gain increased, but leftward VOR gain was normal, resulting in significant right left asymmetry (p < 0-001, Student's test) (fig 5D and table 4). VOR suppression gain was impaired bilaterally, though mainly rightwards (fig 5E and table 4). The ratio of A leftward gain to rightward gain was approximately 1 for the VOR under the first condition, it was 0-36 for the VOR in the second condition, and 0 35 for VOR suppression. Vertical saccades (200 amplitude) had normal latency and accuracy. Their velocity seemed qualitatively normal (about 2400/s B upwards and 245°/s downwards). Smooth pursuit was almost entirely saccadic both

e + upwards and downwards, and velocity of vertical OKN slow phases was markedly C reduced. PATENTS 3 AND 4 Velocity and amplitude of horizontal saccades aS of patient 3 (fig 6A) and patient 4 were analogous to those of the second and first control groups, respectively (table 1). Smooth pursuit was impaired, mainly at higher target veloci- http://jnnp.bmj.com/ ties, slightly more ipsilaterally (fig 6B and table 2). Asymmetry was slight for targets moving at 14 5°/s, 22-5°/s, and 45°/s for patient 3, and the index of asymmetry was within the normal range for patient 4 (fig 7). b Horizontal OKN was similarly impaired, with

no (patient 3) (fig 6C and table 3) or slight on September 27, 2021 by guest. Protected copyright. asymmetry (patient 4) (table 3). D In both patients, the VOR was normal under both conditions, and VOR suppression was impaired, with no asymmetry (fig 6 D and E, and table 4). Vertical smooth pursuit was qualitatively mildly impaired (that is, slightly saccadic), both upwards and down- E wards. Discussion 100L All the patients had a limited vascular lesion Is located in the basis pontis. Two structures, therefore, were damaged, at least partially: Figure S Horizontal eye movement recordings ofpatient 2. (A) Saccades, with nornal the tract and the PN. The pon- latency and amplitude, but bilateral decreased velocity. (B) Smooth pursuit, saccadic corticospinal bilaterally, mainly leftwards. (C) OKN, asymmetrical with slower leftward slow phase tine tegmentum was spared in patients 1 and (horizontal arrows indicate the direction ofstimulation, and the vertical arrow a change in 3, and slightly damaged in patients 2 and 4, the direction ofstimulation). (D) VOR (a) under thefirst condition, with normal gain; on the midline. The contralateral motor (b) under the second condition, asymmetrical with normal leftward gain but increased rightward gain. (E) VOR suppression, impaired bilaterally, though mainly rightwards. deficit existing in all four patients could easily e = eye, s = stimulation, t = target. be ascribed to damage to the corticospinal 804 Gaymard, Pierrot-Deseilligny, Rivaud, Velut

Table 3 Horizontal optokinetic nystagmus (OKN) gain

Patient 2 Patient 3 Patient 4 First control group Second control group J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from Gain Gain Gain Mean gain (SD) Mean gain (SD) Stimulus velocity R L R L R L R L R L 10°/s 0-69 0-46 0.40* 0 53 0-58 0-63 0 77 (0-17) 0-78 (0-19) 0-78 (0-11) 0 77 (0-14) 20°/s 0.40* 0-15* 0-41* 0-43* 0 50* 0-68 0-74 (0-11) 0-73 (0-14) 0-69 (0 07) 0-71 (0-11) 300/s 0.35* 0-12* 0-38* 0.39* 0-36* 0-68 0-72 (0-10) 0 75 (0-12) 0-71 (0 06) 0-69 (0 07) The stated gain is that of OKN slow phase. The side of the lesion is denoted in bold. (See also the legend of Table 1.)

Table 4 Horizontal VOR and VOR suppression gain Patient 2 Patient 3 Patient 4 First control group Second control group Stimulus Gain Gain Gain Mean gain (SD) Mean gain (SD) frequency (Hz) R L R L R L R L R L VOR, first condition 0-25 0-81 0-79 0-91 0-82 0-67 0-70 0-81 (0-08) 0-81 (0-10) 0-73 (0-10) 0-71 (0-12) 0-40 0-70 0-88 0-76 0-90 0-71 0-80 0-82 (0-18) 0-84 (0-18) 0-75 (0-19) 0-77 (0-19) VOR, second condition 0-25 2-20* 0-80 0-66 0-79 0-84 0-99 0-85 (0-12) 0-80 (0-12) 0-80 (0-18) 0-78 (0-16) 0-40 1-35* 0-85 0-71 0-72 0-90 0-85 0-76 (0-12) 0-76 (0-07) 0-79 (0-17) 0-77 (0-14) VOR suppression 0-25 0-26* 0-09* 0-14* 0-12* 0-09* 0-09* 0-04 (0-02) 0-03 (0-02) 0-03 (0-02) 0-02 (0-01) The side of the lesion is denoted in bold. (See also the legend of Table 1.)

tract. Impairment of smooth pursuit was pursuit gain, mainly ipsilaterally.4 Other PN probably a result of the PN lesions. We will are also probably involved in smooth pursuit. mainly deal with this point, but abnormalities The lateral PN (LPN) have largely similar concerning the VOR and horizontal saccades connections to those of the DLPN, although in patient 2 will also be discussed briefly. they appear to project more heavily to the floccular region than the DLPN. 81' A lesion SMOOTH PURSUIT AND OPTOKYNETIC restricted to the LPN in the monkey results in NYSTAGMUS severe impairment of ipsilateral smooth pur- The precise location of the different PN suit.25 Recent experimental data suggest that groups is unknown in humans. In the mon- the dorsomedian PN (DMPN) are also prob- key, the PN consist of several contiguous ably involved in smooth pursuit.226 The subdivisions, classified using topographical DMPN receive unilateral afferences from criteria. The classification proposed by Nyby areas MT and MST, and bilateral afferences and Jansen,5 although obtained using old from the FEF.21' They send efferences main- anatomical techniques, is still used as a refer- ly to the vermis, bilaterally, and also to the ence by modern authors.678 They distinguish contralateral flocculus.21315 Cells in these the median PN, dorsal PN, lateral PN, ven- nuclei discharge during smooth pursuit and, tral PN, and peduncular PN, all of which are in the monkey, a lesion in this area impairs http://jnnp.bmj.com/ intermingled with the corticospinal tract. smooth pursuit.2 Lastly, the nucleus reticu- These nuclei extend from the pontomedullary laris tegmenti pontis (NRTP), located at the junction to the pontopeduncular junction.6 anterior part of the upper pontine tegmentum The subdivisions involved in smooth pursuit and difficult to differentiate anatomically have recently been investigated. The DLPN from the nearby DMPN, also appears to be have been extensively studied and appear to involved in smooth pursuit.2 Cells in the ros-

control smooth pursuit. The DLPN receive tral NRTP discharge during smooth pursuit,27 on September 27, 2021 by guest. Protected copyright. afferences from ipsilateral cortical areas and stimulation of the NRTP elicits slow eye involved in smooth pursuit: the middle tem- movements (even if the eyes are not already poral (MT) and medial superior temporal performing smooth pursuit),28 as does stimu- (MST) areas,6-9 and the frontal eye field lation of the FEF.29 The NRTP receives affer- (FEF).3"01 The DLPN project to the cere- ences from the FEF30 and projects heavily to bellar structures involved in smooth pursuit- the flocculus and also to the vermis.13 14 that is, bilaterally to the oculomotor vermis Lastly, a lesion affecting the NRTP in the and the fastigial nucleus, and also bilaterally monkey results in smooth pursuit impair- but mainly contralaterally to the floccu- ment." Therefore, two parallel pathways lUS.' 12-18 Neurons in the DLPN discharge could control smooth pursuit between the during ipsilateral or contralateral smooth pur- cerebral cortex and the pons: the first could suit.19-22 Microstimulations applied within the originate in the temporoparietal cortex (areas DLPN increase ipsilateral smooth pursuit MT and MST), and project mainly to the velocity, but do not elicit slow eye move- DLPN and LPN, whereas the second could ments if eyes are not already performing originate in the FEF and project mainly to smooth pursuit.2' It should be noted that the the NRTP and/or the DMPN. The specific same fact was observed for stimulations of roles of these two parallel pathways in smooth area MST.24 Furthermore, it has been report- pursuit remain, however, to be determined. ed, in the monkey and in one patient, that a Smooth pursuit impairment differed DLPN lesion results in a decrease in smooth between our four patients. Ipsilateral smooth Smooth pursuit eye movement deficits afterpontine nuclei lesions in humans 805

t group and the DMPN group. According to _ Schmahmann and Pandya,6 these two groups J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from are separated by another PN group, the peduncular PN, intermingled with the corticospinal tract. Therefore, the DLPN-LPN group and the DMPN group lie respectively A laterally and medially to a virtual anteroposte- rior line crossing the pons through the central part of the peduncular PN (see fig 2). Thus, t it is reasonable to assume that lesions located laterally or medially to this line involve either e the DLPN-LPN group or the DMPN group, B respectively. According to these anatomical criteria, the DLPN and the LPN were proba- 5,~~~~~ol- t -.00 bly extensively damaged in patient 1 (given the large rostrocaudal extent of this lesion), slightly damaged in patient 2, and intact in patients 3 and 4. C Conversely, the DMPN were partly damaged unilaterally in patients 3 and 4, and bilaterally in patient 2, but spared in patient 1. From a comparison of patient 1 and patient 2, it may be suggested that the exten- a;/ sive impairment of the DLPN and the LPN was responsible for the severe impairment of \ { / ipsilateral smooth pursuit in patient 1. Moreover, a comparison of patients 3 and 4 with patient 2 suggests that the more pronounced impairment of ipsilateral smooth pursuit in patient 2 could be the consequence of the involvement of the DLPN in this patient, the DMPN being partly damaged in b all these three patients. Lastly, impairment of horizontal OKN and vertical smooth pursuit in particular in patients 1 and 2, were consis- D tent with results of experimental DLPN lesions.25 Smooth pursuit impairment was less asymmetrical in patient 2 than in patient 1, partly because of a relatively more marked con- ro-- - -% tralateral smooth pursuit deficit in the former, E for target velocities (29 and

especially high http://jnnp.bmj.com/ 45°/s). Patients 3 and 4, with DMPN lesions, had, respectively, slight and no asymmetry in smooth pursuit (fig 7). Slight asymmetry after ioOL a DMPN lesion may be due to the medial location of these nuclei. Since the two groups Is are located on either side of the midline, and Figure 6 Horizontal eye movement recordings ofpatient 3. (A) Saccades, normal. (B) their efferences decussate through the basis Smooth pursuit, mildly impaired. (C) OKN, mildly impaired (horizontal arrows indicate pontis before projecting to the cerebellum, a on September 27, 2021 by guest. Protected copyright. the direction ofstimulation, and the vertical arrow a change in the direction of stimulation). (D) VOR (a) under thefirst condition, normal; (b) under the second single medial lesion may affect both nuclei (as condition, normal. (E) VOR suppression, slighdy impaired bilaterally. e = eye, s = in patient 2) and their efferences. It should stimulation, t = target. also be noted that these nuclei receive bilateral afferences from the FEF. The severe impairment of contralateral smooth pursuit in pursuit was severely impaired in patient 1, patients 1 and 2 appears to contradict experi- moderately impaired in patient 2, and only mental studies in the monkey, in which slightly impaired in patients 3 and 4. Smooth DLPN lesions result in only slight or no pursuit asymmetry was also pronounced in impairment of contralateral smooth pursuit.25 patient 1, less marked in patient 2, and very However, these chemically induced lesions, slight in patients 3 and 4. At first, a difference performed by excitotoxic agents, such as in the age of the lesions may account for dif- ibotenic acid, selectively damage cell soma ferent rates of recovery between the patients, but spare fibres of passage-that is, efferences especially for patient 2, recorded 5 months from the contralateral pontine nuclei. after the stroke. However, the differences Therefore, this could explain the difference could also result from the diverse locations of between the results of experimental and clini- the lesions. Although the exact topography of cal studies. the PN is unknown in humans, it is possible Our findings in patient 1 are consistent to distinguish the two groups of PN in which with the results of Thier et al.4 These authors we are interested-that is, the LPN-DLPN reported the first case of selective smooth 806 Gaymard, Pierrot-Deseilligny, Rivaud, Velut

Target 45 I velocity I -0 * J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from (°/s)

291 0 - * * 00- 225}, * 8 14.5i v

0 4.0 5.0 Index of asymmetry

14.51 0 0 C 122.5++ 0 * Patient I * Patient 2 Target O Patient 3 velocity V Patient 4 (°/s) 1--4 0 Figure 7 Index ofasymmetry ofsmooth pursuit. Valuesfor each control group are plotted as horizontal bars displaying mean index ofasymmetry ±SD. Note thatpatient 1, with a DLPN and LPN lesion, had the most marked asymmetry, whereas patients 3 and 4, with DMPN lesions, had slight and no asymmetry, respectively.

pursuit impairment after a PN lesion. observed in cerebellar degeneration,"-'4 and According to MRI, the lesion probably in the monkey, hyperactive VOR has been involved the LPN-DLPN group. Saccades produced by a reversible lesion affecting the were normal but smooth pursuit was marked- olivofloccular tracts or the inferior olivary ly impaired, bilaterally although mostly ipsi- nucleus.35 The lesion of patient 2 did not laterally, resulting in a high level of appear to involve the flocculus, the olivary asymmetry. To our knowledge, no previous nucleus, or the flocculovestibular and olivo- smooth pursuit study has been reported in a cerebellar tracts, both passing through the patient with a medially located pontine inferior cerebellar peduncle. Therefore, the lesion, considered to involve the DMPN. explanation for such a unilateral hyperactive Therefore, our results appear to confirm VOR is unclear. It may be that damage to the http://jnnp.bmj.com/ the involvement of the DLPN and LPN in afferences of the inferior olivary nucleus was smooth pursuit control, and suggest that the responsible for this abnormality. VOR sup- DMPN could be involved in this eye move- pression was likewise impaired bilaterally in ment as well, though with a minor role. As patient 2, but predominantly for the right- for the NRTP, its role in smooth pursuit in ward hyperactive VOR. However, as the humans remains to be determined, since this index of asymmetry of the VOR under the

structure was probably spared, either totally second condition and VOR suppression were on September 27, 2021 by guest. Protected copyright. or largely, in our four patients. analogous, it may be suggested that asymmetry observed in VOR suppression was VESTIBULO-OCULAR REFLEX mainly the consequence of VOR asymmetry. Since the VOR of patient 1 was not tested in In patients 3 and 4, the VOR tested under complete darkness and was analysed only both conditions was normal. VOR suppres- qualitatively, a slight abnormality of this eye sion was mildly impaired bilaterally, without movement cannot be ruled out. VOR sup- asymmetry. pression was impaired bilaterally in this patient but, unlike smooth pursuit, was sym- SACCADES metrical. A similar discrepancy between Patient 2 had reduced velocity of all horizon- smooth pursuit and VOR suppression results tal saccades. As no lesions were visible on was also observed in the patient previously MRI in the cerebral hemispheres or in the reported.4 It has been suggested that VOR midbrain, it may be suggested that this suppression does not result only from addi- abnormality was due to the pontine lesion. In tional mechanisms between the VOR and the monkey,25 as well as in the previously smooth pursuit. reported case4 or in our patient 1, DLPN or In patient 2, rightward gain was about LPN lesions did not result in a decrease in 2-20, whereas leftward VOR gain was normal. saccade velocity. Moreover, a lesion in the Increased VOR gain is uncommon. In DMPN area does not seem to reduce saccade humans, bilateral hyperactive VOR has been velocity, as shown by patients 3 and 4. Smooth pursuit eye movement deficits afterpontine nuclei lesions in humans 807

Therefore, it may be assumed that this the oculomotor cerebellar vermis in the macaque monkey. J Comp Neurol 1987;265:224-41. saccade abnormality was due to the small 16 Suzuki DA, Keller EL. The role of the posterior vermis of J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.56.7.799 on 1 July 1993. Downloaded from median tegmental extent of the lesion, monkey cerebellum in smooth-pursuit eye movement control. I. Eye and head movement-related activity. probably involving the pontine paramedian J Neurophysiol 1988;59:1-18. reticular formation (PPRF). The lesion of 17 Zee DS, Yamazali A, Butler PH, Gucer G. Effects of ablation of flocculus and paraflocculus on eye move- patient 4 impinged also slightly in the pontine ment in primate. JNeurophysiol 1981;46:878-99. tegmentum, but to a lesser extent than that of 18 Buttner U, Fuchs AF, Markert-Schwab G, Buckmaster P. Fastigial nucleus activity in the alert monkey during patient 2. Therefore, the PPRF was probably slow eye and head movements. J Neurophysiol 1991; spared in this patient, with saccades of nor- 65:1360-71. 19 Mustari MJ, Fuchs AF, Wallman J. Response properties mal velocity. of dorsolateral pontine units during smooth pursuit in the rhesus macaque. JNeurophysiol 1988;60:664-86. 20 Suzuki DA, Keller EL. Visual signals in the dorsolateral This work was supported by the Direction de la Recherche pontine nucleus of the alert monkey: their relationship Clinique (N 91.2215). to smooth-pursuit eye movements. Exp Brain Res 1984;53:473-8. 21 Suzuki DA, May JG, Keller EL, Yee RD. Visual motion 1 Brodal P. The pontocerebellar projection in the rhesus response properties of neurons in dorsolateral pontine monkey: an experimental study with retrograde axonal nucleus of alert monkey. JNeurophysiol 1990;63:37-59. transport of horseradish peroxidase. Neuroscience 1979; 22 Thier P, Koehler W, Buettner UW. Neuronal activity in 4:193-208. the dorsolateral pontine nucleus of the alert monkey 2 Keller EL, Heinen SJ. Generation of smooth-pursuit eye modulated by visual stimuli and eye movements. Exp movements: neuronal mechanisms and pathways. Brain Res 1988;70:496-512. Neurosci Res 1991;11:79-107. 23 May JG, Keller EL, Crandall WF. Changes in eye velocity 3 Leichnetz GR. Inferior frontal eye field projections to pur- during smooth pursuit tracking induced by microstimu- suit related dorsolateral pontine nucleus and middle lation in the dorsolateral pontine nucleus of the temporal area (MT) in the monkey. Vis Neurosci macaque. Soc NeurosciAbstr 1985;11:79. 1989;3: 171-80. 24 Komatsu H, Wurtz RH. Modulation of pursuit eye move- 4 Thier P, Bachor A, Faiss J, Dichgans J, Koenig E. ments by stimulation of cortical areas MT and MST. J Selective impairment of smooth pursuit eye movements Neurophysiol 1989;62:31-47. due to an ischemic lesion of the basal pons. Ann Neurol 25 May JG, Keller EL, Suzuki DA. Smooth pursuit eye 1991;29:443-8. movement deficits with chemical lesion in the dorsolat- 5 Nyby 0, Jansen J. An experimental investigation of the eral pontine nucleus of the monkey. J Neurophysiol corticopontine projection in Macaca mulatta. Norske 1988;59:952-77. Videnskaps Akad Oslo Nat 1951;3:1-47. 26 Keller EL and Crandall WF. Neuronal responses to opto- 6 Schmahmann JD, Pandya DN. Projections to the basis kinetic stimuli in pontine nuclei of behaving monkey. J pontis from the superior temporal sulcus and superior Neurophysiol 1983;49: 169-87. temporal region in the rhesus monkey. J Comp Neurol 27 Suzuki DA, Yamada T, Betelak KF, Yee RD. Smooth- 1991;289:224-48. pursuit eye movement related activity in monkey 7 Glickstein M, May J, Mercier BE. Corticopontine projec- nucleus reticularis tegmenti pontis. Soc Neurosci Abstr tion in the macaque: the distribution of labeled cortical 1991;17:459. cells after large injections of horseradish peroxidase in 28 Yamada T, Suzuki DA, Betelak KF, Yee RD. Initial eye the pontine nuclei. J Comp Neurol 1985;253:343-59. position dependence of eye movements evoked by 8 May JG, Andersen RA. Different patterns of corticopon- microstimulation of the monkey nucleus reticularis tine projections from separate cortical fields within the tegmenti pontis. Soc NeurosciAbstr 1991;17:459. inferior parietal lobule and dorsal prelunate gyrus of the 29 Bruce CJ, Goldberg ME. Primate frontal eye fields: I. macaque. Exp Brain Res 1986;63:265-78. Physiological and anatomical correlates of electrically 9 Ungerleider LG, Desimone R, Galkin TW, Mishkin M. evoked eye movements. J Neurophysiol 1985;54:714-34. Subcortical projections of area MT in the macaque. J 30 Stanton GB, Goldberg ME, Bruce CJ. Frontal eye field Comp Neurol 1984;223:368-86. efferents in the macaque monkey: II. Topography of ter- 10 Huerta MF, Kaas JH. Supplementary eye field as defined minal fields in midbrain and pons. Jf Neurophysiol by intracortical microstimulation in macaques. J Comp 1988;271:493-506. Neurol 1990;293:299-330. 31 Suzuki DA, Yee RD, Betelak K. Deficits in smooth- 11 Stanton GB, Goldberg ME, Bruce CJ. Frontal eye field pursuit eye movements following lidocaine injection in efferents in the macaque monkey. II. Topology of termi- monkey nucleus reticularis tegmenti pontis. Soc Neurosci nal fields in midbrain and pons. J Comp Neurol Abstr 1990;16:903. 1988;271:493-506. 32 Baloh RW, Konrad HR, Honrubia V. Vestibulo-ocular 12 Hoddevik GH, Brodal A, Kawamura K, Hashikawa T. function in patients with cerebellar atrophy. Neurology

The pontine projection to the cerebellar vermal visual 1975;25: 160-8. http://jnnp.bmj.com/ area studied by means of the retrograde axonal transport 33 Thurston SE, Leigh RJ, Abel LA, Dell'Osso LF. ofhorseradish peroxidase. Brain Res 1977;123:209-27. Hyperactive vestibulo-ocular reflex in cerebellar degen- 13 Langer T, Fuchs AF, Scudder CA, Chubb MC. Afferents eration: pathogenesis and treatment. Neurology 1987; to the flocculus of the cerebellum in the rhesus macaque 37:53-7. as revealed by retrograde transport of horseradish perox- 34 Zee DS, Yee RD, Cogan DG, Robinson DA, Engel WK. idase. J Comp Neurol 1985;235: 1-25. Ocular motor abnormalities in hereditary cerebellar 14 Noda H, Sugita S, Ikeda Y. Afferent and efferent connec- ataxia. Brain 1977;99:207-34. tions of the oculomotor region of the fastigial nucleus in 35 Demer JL, Robinson DA. Effects of reversible lesions and the macaque monkey. J Comp Neurol 1990;302:330-48. stimulation of olivocerebellar system on vestibuloocular 15 Yamada J, Noda H. Afferent and efferent connections of reflex plasticity. J Neurophysiol 1982;47:1084-107. on September 27, 2021 by guest. Protected copyright.