THE ROLE OF ERG/VEP AND EYE MOVEMENT RECORDINGS IN CHILDREN WITH OCULAR MOTOR APRAXIA FATIMA S. SHAWKAT, CHRISTOPHER M. HARRIS, DAVID S. I. TAYLOR and ANTHONY KRISS London SUMMARY several reports of OMA or saccade failure occurring Ocular motor apraxia (OMA) is characterised by an congenitally, with no other clinical entity?-5 How­ intermittent inability to initiate voluntary sacca des, and ever, it can also occur as part of a wider neurological a failure to produce optokinetic and vestibular quick disorder: for example with structural brain abnorm­ phases. Some patients have no other abnormalities alities, such as agenesis of the corpus callosum6 and (idiopathic OMA), whereas in others it appears vermis hypoplasia;7 with neurodegenerative condi­ associated with a variety of neurological conditions tions;8 and with acquired neurological disease such as which may affect the sensory visual pathway. Electro­ posterior fossa tumours,9 ataxia telangiectasia,lO retinograms (ERGs), flash and pattern visual evoked fronto-parietal lesions,l1.l2 occipital cortex lesions,13 potentials (VEPs) and eye movements were assessed in cerebellar and brains tern neoplasm14 and olivoponto­ 53 children with OMA (age range 17 days to 14 years) cerebellar degeneration. 15.16 The inability to gener­ to determine their efficacy in helping to distinguish ate saccades often leads to the development of between idiopathic and non-idiopathic cases. Seven patients (13.2%) had idiopathic OMA and the remain­ compensatory behaviour to shift direction of gaze; ing 46 (86.8%) had other associated clinical conditions. this includes headthrusting, blinking and tilted head All patients had episodes of absent quick phases ('lock posture, which enables the use of vertical eye up') during optokinetic nystagmus (OKN) and/or movements that are usually unaffected. Hypometric vestibular testing. Flash ERGs were abnormal in-only saccades, poor optokinetic nystagmus (OKN) and 7 patients (13.2%); 6 had syndromes involving a contraversive deviation of the eyes on vestibular pigmentary retinopathy (Joubert's, Bardet-Biedl, testing (due to the failure of OKN and vestibular infantile Refsum's, Kearns-Sayre's), and the seventh quick phases) have also been described in ,1 , had a cone dystrophy with vermis hypoplasia. VEPs OMA. 2 7 18 were normal in all 7 patients with idiopathic OMA. Earlier electrophysiological studies in patients with Thirty-three (72%) patients with OMA in association congenital OMA reported normal flash and pattern with neurological conditions had abnormal VEPs and visual evoked responses (VEPs) and electroretino­ 13 had normal VEPs (28%). There was a significant grams (ERGs).4,J3 However, a recent studiO found 2 positive correlation between VEP abnormality and poor OKN gain. VEPIERG testing and eye movement of 5 children diagnosed with 'Cogan's ocular motor studies are useful when OMA is suspected as they help apraxia' to have absent ERGs but normal flash in distinguishing isolated idiopathic cases from those VEPs. On the basis of these findings the authors with more widespread neurological abnormalities. suggest the existence of two variants of the condition, one with and the other without retinal dysfunction. Coganl in 1952, first used the term congenital ocular This prompted us to re-examine data from 53 motor apraxia (OMA) to describe the clinical signs children in whom saccade failure had been objec­ of 4 patients who had difficulties in generating tively identified and electrophysiological measure­ horizontal saccades. Since then, there have been ments had been made. Electro-oculography (EO G) and video monitoring was used to assess the eye Correspondence to: Fatima S. Shawkat, Department of movements. The extent of saccade failure was Ophthalmology, Great Ormond Street Hospital for Children, London WCI 3JH, UK. Fax: 0171 829 8647. quantified and related to the YEP and ERG findings. Eye (1996) 10,53-60 © 1996 Royal College of Ophthalmologists 54 F. S. SHAWKAT ET AL. PATIENTS AND METHODS phase) was difficult to assess accurately because of Subjects difficulties in eliciting saccades for calibration. Over a period of 2 years, 53 patients referred to the However, it was clearly evident that in certain cases ophthalmology department with OMA underwent OKN was of extremely low gain or absent when both ERGNEP assessment and formal eye move­ compared with control subjects recorded in our 19 ment recording. Age at testing ranged from 17 days laboratory. Thus, OKN gain abnormalities were 0 = 1 = to 14 years, with a mean age of 3.5 years. There were qualitatively ordinally scaled into normal, 1 25 males and 28 females. Forty-one patients (77.4%) abnormal. An abnormal OKN score of was given to 25 were scanned (24 MRI, 17 CT). those with conspicuously reduced gain at deg/s Patients were divided into groups on the basis of stimulus speeds or when OKN was totally absent. clinical history and signs, and diagnostic findings VOR was tested by rotating the patient on the (neuro-imaging and biochemical studies). This Barany chair while in complete darkness. The chair 2 categorisation was carried out independent of, and was accelerated at 18 deg/s to a speed of 80 degls prior to, the VEP/ERG assessment. Seven children and this speed was maintained for 40 seconds before 2 40 had no other clinical abnormalities apart from the decelerating at 18 deg/s to rest. After another OMA and were labelled as having 'idiopathic' OMA. seconds at rest, the chair was rotated in the opposite Two children had normal MRI results with other direction. Per- and post-rotatory vestibular nystag­ symptoms: one had early onset vertical nystagmus mus were recorded in both directions. and microphthalmos, and the other was markedly hypotonic, developmentally delayed with gastro­ Electrophysiology 13 oesophageal reflux. There were patients with VEPs were recorded using silver/silver chloride EEG non-progressive neurological deficits secondary to electrodes placed in a line across the occipital scalp. perinatal or postnatal problems (anoxia, ischaemia or A midline electrode was sited at Oz (10-20 system, meningitis). Thirteen children had progressive, approximately 3 cm above the inion) and lateral neurodegenerative conditions (including Gaucher's electrodes were placed midway between the Oz disease, Krabbe's leucodystrophy, GM gangliosido­ 1 electrode and the ear. All occipital electrodes were sis, infantile Refsum's disease, Pelizaeus Merzbacher referred to a common frontal electrode at Fz (10-20 disease, propionic acidaemia) and 18 patients had system). The electroretinogram (ERG) was recorded structural brain anomalies (including agenesis of the from a lower eyelid skin electrode also referred to Fz. corpus callosum, vermis cyst, vermis hypoplasia in Full-field pattern reversal VEPs were elicited using a isolation, and in association with Joubert's syndrome, checkerboard pattern presented on a large TV Dandy Walker malformation, hydrocephalus, hemi­ display (sub tending 28 degrees horizontal by 21 megalencephaly, porencephalic cyst). degrees vertical). Stimulus check-sizes ranged from Neuro-radiological examination revealed a wide 25 minutes to 7 degrees. A Grass PS22 photic range of abnormalities the most common of which stimulator, held 15 cm in front of the eye, was used were brainstem abnormalities, cerebellar abnormal­ to elicit flash VEPs and mixed cone/rod ERGs under ities (particularly those involving the vermis), fully darkened laboratory conditions. delayed or absent myelination, and agenesis of the corpus callosum. Patients with idiopathic OMA had RESULTS normal scans. Eye Movements Clinical examination identified head thrusting beha­ Eye Movements viour in 30 of the patients (57%). The remaining 23 Horizontal eye movements were recorded using dc (43 %) did not head thrust and OMA was identified EOG. Silver/silver chloride electrodes were attached during the formal eye movement recording. The with tape at the outer canthi of either eye with a majority of patients with idiopathic OMA (86%) reference electrode at the mid-forehead. Older showed head thrusting behaviour; in contrast, it was patients sat alone in a Barany chair; younger evident in only 10% of patients with associated patients sat on a parent's lap. The head was neurodegenerative conditions. There was no obvious restrained by a head rest, or was held by the trend in children with structural malformations of the parent. Video monitoring of the patient was carried central nervous system (CNS), of whom 56% out through the entire recording session. Infra-red exhibited head thrusting, whereas fewer (30%) monitoring was used when testing the vestibulo­ patients with peri- or postnatal problems showed ocular reflex (VOR) in absolute darkness. this compensatory behaviour. OKN was elicited by rotating a brightly coloured Saccade failure during OKN and VOR testing was and patterned, full-field curtain around the patient. identified as episodes of absent quick phases (Fig. 1). Leftward and rightward OKN at speeds of 25 and 50 When quick phases at the end of an OKN or deg/s were recorded. Gain of OKN (slope of the slow vestibular nystagmus slow phase are absent the eyes OCULAR MOTOR APRAXIA 55 A. OKN to Leftward Curtain Rotation R L B. VOR to Rightward Chair Rotation R L Fig. 1. EOG traces of optokinetic nystagmus (A) elicited by leftward curtain rotation at a speed of 25 degls, and vestibular nystagmus (B) elicited by chair rotation (open arrow) at 80 degls. Episodes of 'lock up' (filled arrows), where there is failure in generating