BASIC AND CLINICAL ASPECTS OF VERTIGO AND DIZZINESS On-Road Assessment of Driving Performance in Bilateral Vestibular-Deficient Patients Hamish G. MacDougall, a Steven T. Moore, b Ross A. Black, c Neryla Jolly, d and Ian S. Curthoys a a Vestibular Research Laboratory, School of Psychology, University of Sydney, New South Wales, Australia b Human Aerospace Laboratory, Mount Sinai School of Medicine, New York, New York 10029, USA c Department of Occupational Therapy, Prince of Wales Hospital, Kensington, New South Wales, Australia d School of Applied Vision Sciences, University of Sydney, New South Wales, Australia This study measured on-road driving behavior in subjects with bilateral vestibular loss (BVL). Data included point-of-regard (what the driver is looking at and attending to), gaze stability (the performance of the vestibulo-ocular reflex), and head movement, during complex maneuvers such as changing lanes, cornering, pulling into traffic, and parking. Subjective and objective measures showed few differences between BVL subjects and age-matched controls, and that it is possible to drive well with little or no peripheral vestibular function. This has important implications for driver licensing, road-safety policy, and for the potential successful rehabilitation of vestibular patients. Patients with unilateral vestibular dysfunction may have more difficulty driving than their bilateral counterparts. Key words: vestibular; driving; human performance; oscillopsia; vestibular loss; vestibular rehabilitation Introduction Significant concerns have been expressed with regard to the driving safety of patients with vestibular disease. 1 3 A recent survey 4 found that patients with vestibular deficits reported considerable difficulty driving in reduced visibility (such as at night or during rain), and in visually complex environments (high-traffic roads, large intersections). Vertigo and disturbed vision were triggered by driving tasks requiring rapid head motion, such as checking for traffic prior to changing lanes, entering an intersection, or merging with traffic. Despite Address for correspondence: Dr. Hamish G. MacDougall, School of Psychology, University of Sydney, Sydney, NSW 2006 Australia. hamish@psych.usyd.edu.au the potential impact on community safety, little applied research has been done to assess the impact of vestibular disease on driver ability, and there are no criteria with regard to vestibular disease in obtaining a driver s license. This study focused on functional assessment of on-road driving performance in patients over the age of 50 years with bilateral vestibular loss as shown by clinical testing. We expected that drivers with bilaterally absent vestibular function would adopt a strap-down strategy (similar to that observed in astronauts postflight), where the head is essentially locked to the trunk to limit the range of head motion, especially during tasks such as cornering, changing lanes, and entering an intersection. In short, we expected vestibular patients to exhibit smaller head movements with respect to healthy controls to minimize Basic and Clinical Aspects of Vertigo and Dizziness: Ann. N.Y. Acad. Sci. 1164: 413 418 (2009). doi: 10.1111/j.1749-6632.2008.03733.x C 2009 New York Academy of Sciences. 413
414 Annals of the New York Academy of Sciences Figure 1. The equipment used to measure head eye coordination during on-road driving performance. The video goggles (65 g) can be worn over prescription glasses and include two infrared cameras to record images of both eyes illuminated by infrared LEDs; a color scene camera for analysis of the subject s point-of-regard; a laser projector for eye-movement calibration; and a microphone for audio recording. A digital video recorder (640 480 pixels, 30 Hz, 200 g) stores images of both eyes and the scene combined by a picture-inpicture processor (200 g). Head movements are measured by an inertial measurement unit (3-D linear acceleration and angular velocity) on a plastic band (18 g), and recorded by a data logger (12-bit analog-to-digital converter [ADC], 100 Hz, 70 g) that contains a second inertial measurement unit (IMU) to sense vehicle motion. A synched GPS data logger records vehicle position, heading, speed, and so forth. retinal slip. In addition, we expected that dynamic visual acuity in vestibular-impaired drivers would be degraded (due to oscillopsia), with a consequent degradation in the ability to read signs while driving. Materials and Methods Three patients with bilateral vestibular loss (BVL) (including one with complete bilateral surgical vestibular loss) holding an unrestricted driver s license were recruited. Subject 1 (69 yr, male) history: Off balance ever since developing Gentamicin vestibulotoxicity in 2003. He notices the balance is particularly bad if the ground is uneven or he closes his eyes... On examination, there was no spontaneous, head shaking, gaze evoked or positioning nystagmus. There was no positioning vertigo, but a severe deficiency of horizontal and vertical vestibulo-ocular reflexes on impulse testing. The Romberg test was negative, but the Fromberg test was positive. Vestibular function studies showed no responses to zero degree caloric irrigation, some residual
MacDougall et al.: Driving Performance in Bilateral Vestibular-Deficient Patients 415 Figure 2. The distribution of all head positions adopted by bilateral vestibular-loss (BVL) subjects while driving a 20-min on-road course is similar to those from age-matched control subjects. The X-axis shows horizontal head position, and the Y-axis shows vertical head positions in 1-degree bins. The Z-axis (color intensity) shows the number of head positions in that bin. responses to rotation and some residual responses on impulse testing, during the leftward rotations. Subject 2 (63 yr, male) history: About 1973 developed severe headaches, ataxia, failing vision, vomiting but no hearing loss found to have a large right cerebellar-pontine angle tumor and operated by Dr. [ ] who had to remove some cerebellum in order to remove the tumor. Well apart from right hearing loss and not so good balance, until 10 years ago when he started to notice left hearing loss and was found to have left sided Schwannoma. No significant vestibular response to rotation, head impulses. Subject 3 (63 yr, male) history: A history of a traumatic foot drop, following a tibial fracture in 1972, and a total hip replacement 13 years ago, and a traumatic subdural haemorrhage in 1980. According to his wife, the patient was 35-yearsoldatthetime,andhestruckhishead on the canopy of the truck, and presented three months later with severe headaches and dizziness. He complains of at least 8 year history of disequilibrium consisting of a tilting sensation, but without true vertigo. No significant vestibular response to Caloric testing or VEMP. These BVL subjects and three healthy agematched control drivers (59, 64, and 67 yr) were tested while driving on city roads in an urban environment (Sydney, Australia). We measured head eye coordination using a custom-designed video eye movement system that acquired head, eye, and vehicle
416 Annals of the New York Academy of Sciences TABLE 1. The Number (Count) and Amplitude (Mean and SD) of Head Movements a Normal Control Subjects and Two Bilateral Vestibular Loss Subjects b for the Three Peak Horizontal Peak Vertical Peak Horizontal Peak Vertical Position >20 deg Position >10 deg Velocity >200 deg/s Velocity >50 deg/s Left Right Down Up Left Right Down Up NORMAL1 Count 87 107 36 47 23 25 71 62 Mean 33.3 34.6 13.0 12.8 303.1 274.6 66.5 65.7 SD 14.7 14.2 2.9 3.5 56.8 68.1 13.4 14.6 NORMAL2 Count 105 128 47 47 46 26 17 24 Mean 48.9 41.1 13.2 14.1 298.5 315.9 74.2 68.9 SD 29.0 26.4 2.1 5.3 93.0 95.1 11.9 15.6 NORMAL3 Count 88 105 22 25 24 22 11 9 Mean 38.6 39.4 14.6 14.4 306.0 291.1 70.0 67.6 SD 21.0 22.5 3.8 5.0 60.7 65.5 17.6 20.3 BVL1 Count 126 145 41 37 50 42 25 13 Mean 37.3 33.1 14.4 15.9 320.0 335.8 75.0 74.2 SD 18.5 14.8 5.0 6.3 95.2 107.0 27.6 23.4 BVL2 Count 104 117 34 21 53 57 25 11 Mean 38.7 38.7 15.6 14.1 336.6 320.2 86.2 80.7 SD 20.0 18.9 6.0 7.4 102.4 104.8 38.1 37.5 a With angular position greater than 20 deg horizontal, angular position greater than 10 deg vertical, angular velocity greater than 200 deg/s horizontal, and angular velocity greater than 50 deg/s vertical. b No data for BVL3. movement during vehicle operation (Fig. 1). We have demonstrated previously that the headmounted equipment does not affect natural head eye coordination. 5 Data included continuous time series of point-of-regard, gaze stability and head movement, during complex maneuvers, such as changing lanes, cornering, pulling into traffic, and parking. Driving performance was assessed both by objective measures of the vehicle and subjective assessments made by an occupational therapist and an orthoptist experienced in evaluating drivers with disabilities. Results The results were surprising: all 3 patients with bilateral vestibular loss drove with a high level of proficiency. Objective measures of performance, such as average vehicle speed measured with the GPS, also showed similar results (23.6, 17.7, and 19.0 km/h for normal subjects; 17.4, 18.8, and 18.4 km/h for BVL subjects). Head movements measured with the inertial measurement unit (IMU) did not show large differences between these patients and age-matched controls, and there was no evidence of a strap-down strategy (Fig. 2). An analysis of the number of active head movements and their amplitudes for normal and BVL subjects while driving over a 20-min course also showed few differences (Table 1). One-way ANOVA showed that horizontal head movements above 20 deg were significantly larger for normal subjects (M = 39.55 deg) than for bilateral subjects (M = 36.69 deg; F = 5.25, P =.022); however, the effect size
MacDougall et al.: Driving Performance in Bilateral Vestibular-Deficient Patients 417 Figure 3. Example frames from a bilateral vestibular-loss (BVL) subject s point-of-regard video. BVL subjects showed good head eye coordination, dynamic visual acuity, and a lack of oscillopsia, demonstrated by their ability to read small and distant road signs even during vehicle perturbations such as those produced by speed bumps. was small (eta sq. = 0.005). Vertical head movements above 10 deg were significantly larger for bilateral subjects (M = 15.08 deg) than for normal subjects (M = 13.55 deg; F = 8.67, P = 0.003); however, the effect size was small (eta sq. = 0.024). Horizontal head movements above 200 deg/s were significantly faster for normal subjects (M = 298.34 deg/s) than for bilateral subjects (M = 327.68 deg/s; F = 9.46, P = 0.002); however, the effect size was small (eta sq. = 0.025). Vertical head movements above 50 deg/s were significantly faster for bilateral subjects (M = 79.49 deg/s), than for normal subjects (M = 67.49 deg/s; F = 14.34, P < 0.001); however, the effect size was small (eta sq. = 0.052). There was no effect of direction, nor were there significant interactions. Although statistically significant, these differences are not large enough or consistent enough to be of functional significance, and do not support our expectation of a large reduction in head movement amplitudes in BVL subjects. Subjective assessments by a driver-trained occupational therapist and an orthoptist noted few differences in behavior or performance. Patients ability to read out loud the content of street signs while driving was not degraded
418 Annals of the New York Academy of Sciences relative to the age-matched control drivers (Fig. 3). Discussion We expected BVL patients would experience significant difficulties behind the wheel, as well as clear differences with age-matched controls on the many objective variables that were measured. However, no systematic differences in eye head coordination, dynamic visual acuity, or driver performance as rated by an experienced driver therapist were found. This result raises an interesting question: Does driving proficiency require an intact vestibular system? Based on prior (subjective) studies, 4 we expected that drivers with bilateral vestibular deficiency would exhibit decrements in performance when the vestibular system was challenged (turns, maneuvers requiring rapid head movements, ability to read road signs while moving, etc.), consistent with subjective reports from vestibulopathic drivers. 4 Our results suggest that well-adapted bilaterally deficient drivers may rely on sensory substitution (such as visual input or neck proprioception) to compensate for the lack of vestibular input, and that this strategy is successful in maintaining driving ability at a standard comparable with healthy controls. A possible explanation for the disparity in subjective reports of driving difficulties in vestibulopathic patients 4 and the proficiency displayed by our bilateral vestibular-deficient drivers is the nature of the underlying vestibular pathology. In the survey of Cohen et al. 4 the majority of subjects (82/127) had unilateral paroxysmal vestibular dysfunction (benign paroxysmal positioning vertigo [BPPV], or Meniere s disease), which may be more likely to disrupt driving ability during sudden vestibular challenges. Acknowledgments This project is supported by the Australian Research Council Discovery Grant: Functional Assessment of Head Eye Coordination During Driving (DP0665402), and by the National Aeronautics and Space Administration Grant: Head Eye Coordination During Simulated Orbiter Landing (NNJ04HF51G). Conflicts of Interest The authors declare no conflicts of interest. References 1. McKiernan, D. & D. Jonathan. 2001. Driving and vertigo. Clin. Otolaryngol. Allied Sci. 26: 1 2. 2. Parnes, L.S. & R. Sindwani. 1997. Impact of vestibular disorders on fitness to drive: a census of the American Neurotology Society. Am.J.Otol.18: 79 85. 3. Sindwani, R. & L.S. Parnes, 1997. Reporting of vestibular patients who are unfit to drive: survey of Canadian otolaryngologists. J. Otolaryngol. 26: 104 111. 4. Cohen, H.S., J. Wells, K.T. Kimball, et al. 2003. Driving disability and dizziness. J. Safety Res. 34: 361 369. 5. MacDougall, H.G. & S.T. Moore. 2005. Functional assessment of head-eye coordination during vehicle operation. Optom. Vision Sci. 82: 706 715.