Journal of Neuro-Ophthalmology, June 2005, Volume 25, Issue 2

Improving Vision in a Patient with Homonymous Hemianopia

POINT COUNTER POINT Editor's note: In this section, two experts have debated a controversial issue based on case material and questions directed to them via e- mail. Neither contestant was aware of the other's responses until all of the questions were answered. Then they were shown all responses and asked to write a rebuttal. The editor's summary appears at the end. Bernhard A. Sabel, PhD Institute of Medical Psychology, Otto- v.- Cuericke University of Magdeburg, Magdeburg, Germany Susanne Trauzettel- Klosinksi, MD University Eye Hospital Tuebingen, Department of Pathophysiology of Vision and Neuro- Ophthalmology, Tuebingen, Germany Improving Vision in a Patient With Homonymous Hemianopia A 54- year- old woman has a right occipital stroke that causes a complete left homonymous hemianopia. Six months after the stroke, her ophthalmologic and neurologic examinations are entirely normal except for the persistent hemianopia, which, on Humphrey perimetry, shows no macular sparing. She reports that her reading has slowed down and that because of her visual field defect, she has been forbidden by her doctor from driving a car. Is there any intervention that would help her? Bernhard A. Sabel, PhD: A closer examination is first indicated to determine if the patient has a presumed " complete" hemianopia or whether the visual field examination actually shows some visual sparing, " areas of residual vision" or " relative defects" ( 1). I would recommend re- evaluating the patient for resid­ual vision with additional repeated suprathreshold high-resolution perimetry to hunt for such areas of residual vi­sion. If some are present, vision restoration therapy ( VRT), a computer- based training program, might improve residual vision ( 2,3). Even small improvements in visual function offered by VRT can be beneficial to the patient. If VRT proves ineffective or when no further im­provements after VRT are noted, saccadic training could be tried. This might improve the patient's ability to scan visual scenes with the intact hemifield and this, in turn, might improve visual orientation ( 4) but not reading or driving. Susanne Trauzettel- Klosinski, MD: Six months after the stroke, the patient still has a chance of spontaneous recovery of at least parts of the field defect. After 1 year, spontaneous recovery can no longer be expected. After that point, any improvements could be attributed to training. However, except in the research setting, I suggest that training begin within the first few months of onset of the hemianopia. There is yet no evidence that any intervention will enlarge the visual field defect in a relevant way. Reports of visual field enlargement after training ( 3,5) have not been confirmed in controlled studies ( 6,7). Therefore, I would suggest training to improve saccadic exploration toward the blind hemifield, which has been shown to be effective in compensating for the visual field defect ( 4,8) and in im­proving the functional visual field essential in everyday life activities. However, the chance of regaining the prerequi­sites for driving is low. To overcome the reading disorder, training of pre­dictive saccades during the return sweep can reduce the difficulties in finding the beginning of the next text line. The use of a slightly eccentric fixation locus improves read­ing by increasing the perceptual span toward the blind hemifield ( 9,10). Some have recommended that the pa­tient be taught to read vertically by turning the text although there are no scientific studies to support that recommendation. J Neuro- Ophthalmol, Vol. 25, No. 2, 2005 143 J Neuro- Ophthalmol, Vol. 25, No. 2, 2005 POINT COUNTER POINT Are there any studies that compare the outcomes in patients who have undergone saccadic training to patients who have not undergone such training? Bernhard A. Sabel, PhD: No studies are available yet in which two inde­pendent groups have been compared- one with saccadic training and the other without. Instead, within- subject experimental designs have been used. Thus, the role of the patients' expectations and the general effect of working on a computer monitor cannot be separated from a specific " saccadic training" effect. Rigorous experimental stan­dards would mandate a double- blind, randomized placebo-controlled clinical trial before establishing efficacy of " saccadic training." Saccadic training can follow different strategies. One strategy is to train patients to make broader searches (" visual search field") in the blind hemifield. A second approach is to train patients to make large- scale eye movements toward the blind hemifield. A third approach is to train patients to make small- scale eye movements with the goal of improving read­ing ( 11). However, such compensation strategies might actually activate attention and restoration in the border zones, making it difficult to decide how improvements should be interpreted. According to the work of Kerkhoff ( 12), approxi­mately 95% of hemianopic patients achieve significant improvements with these saccadic treatment techniques. Al­though this appears to be an overestimate, the visual field size does increase by an average of 6.7° in 30% to 50% of patients; in other words, compensation may induce restora­tion ( 13). Other studies, in contrast, have found enlarge­ment of the visual search field without enlargement of the visual field itself ( 14,15). Although some clinics in Germany use saccadic compensation strategies to treat patients with visual field defects, there is currently no firm science that supports widespread use of saccadic training. Compensation might actually inhibit restoration because the patient is trained to use the intact visual field sector only. Therefore, the most rational approach would be to first attempt restoration and then try saccadic training if it is needed. Saccadic training should only be applied to patients when restoration therapy is completed or to those who have no residual vision in the " blind" hemifield, a situation that applies to fewer than 10% of all hemianopic patients. Susanne Trauzettel- Klosinski, MD: Several studies have shown an improvement of saccadic performance after a systematic saccadic training ( 4,8,13,15). Kerkhoff et al ( 4,13) reported an increase in visual search field size ( mean 30°), as well as an improve­ment of identifying objects visually on a table and a sub­jective improvement in vision as rated by the patients. Zihl ( 8) found shorter search times, fewer fixations, and lower repetition rates in the scan path after saccadic training. Pambakian et al ( 15) described shorter reaction times, faster performance of activities of daily living, and subjective improvement after training. These studies have shown sig­nificant improvement in visual search tasks when perfor­mance was compared before and after training. However, none of these studies compared the results with an untreated control group. For assessing a specific training effect, a randomized controlled trial with a con­trol group would be necessary. The study of Zihl ( 8) is also limited by the relatively brief interval from onset of the lesion to testing ( 6- 18 weeks, mean 11 weeks), so that spontaneous recovery cannot be safely excluded ( 16,17). Would your approach to this patient differ if high- resolution perimetry disclosed small islands of intact visual filed close to fixation in the impaired hemifield? Bernhard A. Sabel, PhD: Such incomplete hemianopias are typically found in approximately two- thirds of patients. In such patients, I recommend vision restoration therapy ( VRT) ( 2,3,19), which is known to increase visual field size by shifting the absolute visual field border and improving detection ability in areas of residual vision. Although such border shifts can easily be seen with routine perimetric testing 144 © 2005 Lippincott Williams & Wilkins POINT COUOTER POINT J Neuro- Ophthalmol, Vol. 25, No. 2, 2005 ( 2,3,19), the change cannot be measured with the more difficult task of the scanning laser ophthalmoscope ( SLO), such as the one used in the study of Reinhard et al ( 7). To appreciate the issue of task difficulty of the SLO, consider the following: standard perimetry uses simple detection tasks of near- threshold or suprathreshold single dots on a dark or gray background and patients have to respond to the stimulus by pressing a button. In contrast, in the SLO task used in the Reinhard study ( 7), three black dots ( a reverse stimulus) were presented on a bright red background, which perceptually flickers because it is created by parallel laser lines that produce lateral interferences ( the " McKay effect"). Furthermore, patients had to verbally report what they were seeing and the experimenter inter­preted their verbal reports. This is a more difficult task for a damaged visual system, as evidenced by the fact that vi­sual deficits, as defined by the absolute visual field border, are significantly greater when measured with the SLO than when measured with well- established and standardized perimetry ( 19). VRT also improves reaction time ( 19) and is of practical benefit in subjective vision as assessed by patient testimonials ( 20). Typically two- thirds of patients benefit from VRT ( 20). If patients focus their attention on specific subsectors of the visual field during VRT, restoration is significantly greater in those regions ( 18). If this patient has reading difficulties, there is a 50% chance that reading would improve after VRT. Such an optimistic prognosis is based on the observation that approx­imately half of the patients report being better able to read again after VRT ( 20). Also, in the study of Reinhard et al Bernhard A. Sabel, PhD: There are no optical devices proven to help the vision of hemianopic patients. Occasionally prisms are used that project the images from the deficient part of the visual field onto the intact side. However, patients tend to get confused by double images on the same side of visual space. Such devices cannot be recommended for clinical use. Susanne Trauzettel- Klosinski, MD: Most patients are confused by the double images and disturbances in spatial orientation induced by optical de­vices used in treatment of hemianopia. Binocular sector prisms result in field relocation or a shift of the position of the field loss ( 21). They are not ( 7), reading improved significantly in one test but not another. I would expect reading to improve with VRT because the average border shift after VRT is approximately 5° of visual angle, a relevant change if it occurs near the fixation point. Even a 2° border shift near the fovea could improve reading. Some have speculated that eye movements explain the border shift. However, in unpublished research, when eye movements were monitored with an eye- tracker, no change in eye movement patterns or fixation position was observed after VRT despite significant border shifts in the same patients. The lack of a border shift measured with the SLO ( 7) cannot be cited as proof that vision restoration does not take place because: ( 1) the SLO is insensitive to residual vision; ( 2) in the same patients, a significant border shift in stan­dard perimetric measures occurs; and ( 3) subjective ben­efits are reliably reported by the majority of the patients ( 20), which is also described in the Reinhard et al study ( 7). Susanne Trauzettel- Klosinski, MD: My approach would not differ even if there were small islands of intact visual field close to fixation in the impaired hemifield. Stimulation in this area could provoke saccadic eye movements towards the stimulus, which can be misinterpreted as a visual field recovery, and it would be more effective to apply saccadic training. Such spared islands of vision would of course, im­prove reading if they were along the horizontal meridian. effective in hemianopic patients but have been shown beneficial in patients with hemineglect ( 22). Monocular mirrors and prisms have been used to shift the image of the blind hemifield into the normal one to expand the field ( 23). Monocular sector prisms cause diplopia and confusion. Confusion is the intended effect, because it indicates the appearance of an object, which would not be visible without the prism, inducing eye or head movements toward the blind side. However, the dip­lopia in the center was very unpleasant to the patients ( 23). Hedges ( 24) did report a benefit in 20% of his patients. Monocular sector prisms limited to the peripheral field, placed across the whole width of the lens, have been reported in a small group of patients to expand the field without central diplopia ( 25,26). Are there any optical devices that would be helpful to this patient? 145 J Neuro- Ophthalmol, Vol. 25, No. 2, 2005 POINT COUNTER POINT Do you agree that patients with isolated complete homonymous hemianopias should not be permitted to drive? Are there national regulations in Germany about this? If so, who developed them and are they consistent internationally? How much sparing of the hemianopic field would be necessary to allow safe driving? Bernhard A. Sabel, PhD: Patients with complete homonymous hemianopias should not be permitted to drive. The national regulations in Germany require 120° horizontally when the damage is binocular and a normal field when the damage is monocu­lar. These standards are not the same internationally; in fact, such regulations vary within each state in the United States. How much sparing is sufficient for driving is unclear. Patients with visual field defects have performed remarkably well in driving simulators, in which they showed little or no deficit compared with age- matched controls in one study ( 27). This suggests that their ability to perform driving- like tasks may be better than one would expect. Susanne Trauzettel- Klosinski, MD: I agree that patients with isolated complete homon­ymous hemianopia should not be permitted to drive. The regulations about driving with visual field defects in Germany are based on three levels: Bernhard A. Sabel, PhD: To be able to read an ideal scenario would be to have a " reading window" of intact visual field 5° on each side of fixation. Susanne Trauzettel- Klosinski, MD: The minimum visual field required for reading is 2° to the left and right offixation ( 31- 33). This is the area where the text is seen clearly and covers 10 to 12 letters of news­paper print at a distance of 25 cm. For fluent reading, this " visual span" has to be extended in the reading direction by parafoveal information processing ( 34) up to 5° or 15 letters. 1. The binding regulations of the European Community ( 28) require that group 1 drivers ( private cars) have a horizontal field of at least 120°. In persons with only monocular vision, the visual field has to be normal. In Group 2 drivers ( public cars, busses, trucks, transport of other persons), the binocular field has to be normal. 2. The German Law for Driving License ( 29) further specifies that group 1 drivers have a normal central 30° of field. 3. The German Ophthalmological Society ( 30) has converted these regulations into a more concrete form. For homonymous hemianopia in group 1 drivers, the visual field must extend at least 120° along the hor­izontal meridian. It must be normal within 20° of fixation in all directions. It must be normal within 30° horizontally and completely normal within 10° above and below fixation. The other hemifield must be completely normal. Ophthalmologists normally follow these recommen­dations but are free to deviate in individual cases. Using conventional and SLO perimetry, we demon­strated that patients with hemianopia need a minimum of 5° to both sides of fixation to read normally. Less than that amount impairs proper reading of a given line of text by right hemianopes and ability to locate the beginning of the next line of text by left hemianopes ( 33,35,36). Some patients with macular splitting have a valu­able adaptive strategy: they use eccentric fixation, sac-rficing a bit of visual acuity to gain a perceptional area of 1° to 2° to one side of the vertical meridian. This shift of the field defect is crucial for regaining reading ability ( 9). How many degrees of visual field sparing would allow a patient to read without difficulty? 146 © 2005 Lippincott Williams & Wilkins POINT COUOTER POINT J Neuro- Ophthalmol, Vol. 25, No. 2, 2005 Rebuttals Bernhard A. Sabel, PhD: Although both of us clearly agree that compensation training currently lacks firm evidence of efficacy, we do not share the same opinion concerning the value of vision resto­ration therapy ( VRT). Building an argument on the only two studies in which vision training was only partially effective or not effective in some measures ( 6,7) is a misrepresentation. Failing to consider all the experiments showing efficacy of vision training will inevitably produce a biased view that does not do justice to the therapeutic potential of VRT. Dr. Trauzettel- Klosinski fails to acknowledge that although her own VRT- treated patients did not improve in the SLO task, they did improve significantly in near- threshold ( TAP) and supra- threshold perimetry ( HRP). This is in full agreement with prospective, double- blind randomized placebo- controlled clinical trials ( 3) and was most recently also observed in a study with 300 patients ( unpublished). As in previous studies ( 3,20), our joint study ( 7) found that approximately two- thirds of the patients clearly reported subjective improvements. In one of two reading tests, performance also improved significantly after VRT. At the center of the argument over VRT efficacy, however, is the issue of whether an increased visual field size is caused by an artefact of eye movements towards the hemianopic side. There is no empirical evidence for this claim; any claims to the contrary are speculative. Also, it is true that a small number of hemianopes may saccade toward the hemianopic side, but if they do so at all, they do so irrespective of VRT. No one has observed any additional saccades after VRT. In fact, in the Reinhard study ( 7), most patients showed rather stable fixation ability without any preferential large saccades after VRT; none showed stable eccentric fixation on SLO or TAP. Second, fixation per­formance in TAP and HRP was unchanged after VRT. Both experiments used standard, clinically verified fixation control procedures. Additionally, in 12 out of the 16 patients tested, the blind spot position remained identical after VRT, disprov­ing eccentric fixation ( 19). Finally, if eye movements were the cause of visual field expansion, one would expect the entire visual field border to shift- a phenomenon that was not seen. To put the " eye movement artefact hypothesis" finally to rest, we have now positively and conclusively shown that visual border shifts occur irrespective of eye movements ( data presented at the 2005 NANOS annual meeting). Visual fields were measured in a separate sample of 16 patients with simultaneous eye- tracker recordings to measure eye movement positions before and after VRT. This observation clearly provided evidence that visual field enlargements are not caused by eye movements. Restora­tion of vision is real. If we focus our therapeutic effort only on training patients to more vigorously move the eyes around and enlarge the search field with " compensation training," we force them to use only their intact fields. Continuing on this path misses the potential that residual vision in or near the blind regions has for plasticity and repair. Modern approaches to neurologic rehabilitation in other domains are now focusing on the " forced use" of impaired functional systems with great success. An example is constraint- induced therapy in locomotor rehabilitation. In the visual domain, we should do the same thing- offer a therapy that improves residual visual potential and not send patients home with the message that there is nothing we can do to help them. Susanne Trauzettel- Klosinski, MD: The problem with VRT is that the stimulation is performed in the " transition zone" between the seeing and the blind hemifields. An apparent increase in visual field size by a shift of the field border could easily be caused by eye movements towards the hemianopic side. SLO studies on patients with hemianopia have shown that they spon­taneously perform frequent saccades towards the hemi­anopic side ( 7,37,36). This phenomenon is even more marked if the patient expects a stimulus from the blind side. Accordingly, approximately 50% of patients in the SLO study ( 7) showed less stable fixation after VRT. The advantage of SLO perimetry is reliable fixation control. In our study, we used only those responses recorded, whereas fixation was central and stable during stimulus presentation. Another advantage of SLO perimetry is the high spatial resolution ( 0.5°) that we used in examination of the hemianopic fields. Such spatial resolution cannot be achieved with other perimetric methods. Additionally, there is no light scatter into the seeing hemifield if the inverted stimulus mode is used ( black stimulus on bright back­ground). Because the SLO study ( 7) focused on absolute field defects, only these were examined. We found no change in the absolute field defects before and after VRT. In principle, relative defects could also be measured by SLO. However, even if had there been improvement from absolute to relative defects, as was reported for high- resolution perimetry ( HRP), it is questionable if this finding would have been clinically relevant. Other effects of VRT reported by Dr. Sabel are not specific. Improved reaction times are an expected effect and should be even more pronounced if the same stimulus is used during training ( as with VRT) and outcome measure­ment ( HRP). Subjective improvements of vision cannot be judged without a control group. 147 J Neuro- Ophthalmol, Vol. 25, No. 2, 2005 POINT COUNTER POINT Because of reliable fixation control, high spatial resolution, and absence of light scatter, the SLO is the most reliable method to examine the central visual field and especially field borders. The studies on VRT have not proven a restoration of the visual field. This does not mean that restitution mechanisms should be excluded in principle. However, the effects reported by Dr. Sabel and coworkers are likely to be caused by other factors, such as explorative saccades, a raised level of attention, or other nonspecific effects ( 38,39). Editor's Summary Homonymous hemianopia, especially when it is com­plete or nearly complete, can be disabling. In most parts of the world, it means no more driving. Reading becomes extremely difficult. In most other aspects of daily living, patients adapt remarkably well if they do not have neglect and otherwise have their wits. They quickly learn that they do not see on one side and they compensate by turning their head and making frequent saccades into the blind hemifield. But in an attempt to provide them with a broader field of view, optical devices have been affixed to glasses to bring the blind hemifield into view. These efforts have been a dismal failure. They only make the patient visually con­fused. Perhaps the multiplexing approach described in this issue of the Journal ofNeuro- Ophthalmology will be better. With the apparent success of rehabilitation for neurologic deficits in other systems, it was natural that the same effort be devoted to hemianopia. This debate is centered on two non- optical efforts: saccadic compensation training- teaching hemianopes to make more efficient saccades into the blind hemifield- and VRT- stimulating relative field defect regions at the edge of absolute defects along the vertical meridian to improve the function of these partially preserved visual regions. Both debaters support saccadic compensation training, but they are skeptical of its benefits until a properly controlled study is performed. Even so, Dr. Trauzettel- Klosinki rec­ommends it as soon as a major hemianopia is identified. Dr. Sabel recommends it only if: ( 1) high- resolution perimetry fails to disclose areas of relatively preserved vision within the " blind hemifield;" or ( 2) patients have completed VRT. It is no surprise that Dr. Sabel, who has expertise in neuropsychologic aspects of vision, and who founded VRT, vigorously supports this restorative approach. He does so on the basis of research that shows improvement in the damaged hemifield after VRT and patient testimonials of improved vision. His approach is based on the idea that a partially damaged central visual pathway should behave like central motor, language, and attentional systems, which apparently improve somewhat with training. But motor, language, and attentional systems have redundancy in the form of supplementary regions. The success of rehabili­tation in motor and attention systems may lie in reinforcing these supplementary systems rather than in fortifying the damaged areas. The retinocortical pathway, which is designed to provide exquisite topographical representation of visual space in the brain, has no such redundancy, no back- up systems. Dr. Sabel is counting on the ability to train visual attentional systems. Whether they participate is open to question. Activation studies ( fMRI and positron emission tomography) may tell us more about that. Dr. Trauzettel- Klosinski, a physician with extensive experience in perimetry and in efforts to improve reading in low- vision patients, disputes Dr. Sabel's results. She argues that the SLO study ( 7), which allows monitoring of eye movements, did not show any enlargement of the blind hemifield after VRT that could not be accounted for by saccades made into that blind field. Moreover, she says, even if there was a tiny enlargement of the field, would it make any real world difference? And that, of course, is where this whole business must come to rest. Do any of these trainings really make a difference in the two most important activities affected by hemianopia- driving and reading? I doubt that an extra 5° in the blind hemifield will make a hemianopic person safe on the road. Reading is a different matter. As Dr. Trauzettel- Klosinki's research has shown, gaining 5° in the hemifield may make the difference between a stumbling and a smooth reader. If controlled, masked studies of reading before and after saccadic compensation training or VRT show a major difference, I could become a believer. REFERENCES 1. Kasten E, Wiist S, Sabel BA. Residual vision in transition zones in patients with cerebral blindness. J Clin Exp Neuropsychol 1998a; 20: 581- 98. 2. Kasten E, Sabel BA. Visual field enlargement after computer- training in brain damaged patients with homonymous deficits - an open pilot trial. Restorative Neurol Neurosci 1995: 8: 113- 127. 148 © 2005 Lippincott Williams & Wilkins POINT COUNTER POINT J Neuro- Ophthalmol, Vol. 25, No. 2, 2005 3. Kasten E, Wiist S, Behrens- Baumann W, Sabel BA. Computer- based training for the treatment of partial blindness. Nat Med 1998; 4: 1083- 7. 4. Kerkhoff G, Miinssinger U, Haaf E, Eberle- Strauss G, Stogerer E. Rehabilitation of homonymous scotomas in patients with post-geniculate damage of the visual system: saccadic compensation training. Restorative Neurol Neurosci 1992; 4: 245- 54. 5. Zihl J, von Cramon D. Restitution of visual function in patients with cerebral blindness. J Neurol Neurosurg Psychiatry 1979; 42: 312- 22. 6. Balliet R, Blood KM, Bach- Y- Rita P. Visual field rehabilitation in the cortically blind? J Neurol Neurosurg Psychiatry 1985; 48: 1113- 24. 7. Reinhard J, Schreiber A, Schiefer U, Kasten E, Sabel BA, Kenkel S, Vonthein R, Trauzettel- Klosinski S. Does visual restitution training change absolute homonymous visual field defect? A fundus con­trolled study. Br J Ophthalmol 2005; 89: 30- 5. 8. Zihl J. Visual scanning behavior in patients with homonymous hemianopia. Neuropsychologia 1995; 33: 287- 303. 9. Trauzettel- Klosinski S. Eccentric fixation with hemianopic field defects. A valuable strategy to improve reading ability and an indication of cortical plasticity. Neuro- Ophthalmol 1997; 18: 117- 31. 10. Trauzettel- Klosinski, S, Reinhard J. The vertical field border in hemianopia and its significance for fixation and reading. Invest Ophthalmol Vis Sci 1998; 39: 2177- 86. 11. Kerkhoff G. Neurovisual rehabilitation: recent developments and future directions. J Neurol Neurosurg Psychiatry 2000; 68: 691- 706. 12. Kerkhoff G. Restorative and compensatory therapy approaches in cerebral blindness- a review. Restor Neurol Neurosci 1999; 15( 2- 3): 255- 71. 13. Kerkhoff G, Munflinger U, Meier EK. Neurovisual rehabilitation in cerebral blindness. Arch Neurol 1994; 51: 474- 81. 14. Nelles G, Esser J, Eckstein A, Tiede A, Gerhard H, Diener C. Compensatory visual field training for patients with hemianopia after stroke. Neurosci Lett 2001; 306: 189- 92. 15. Pambakian AL, Mannan SK, Hodgson TL, Kennard C. Saccadic visual search training: a treatment of patients with homonymous hemianopia. J Neurol Neurosurg Psychiatry 2004; 75: 1443- 8. 16. Kolmel HW. Die homonymen Hemianopsien. Klinik und Pathophy­siologic zentraler Sehstorungen. Berlin: Springer; 1988. 17. Trobe JD, Lorber ML, Schlezinger NS. Isolated homonymous hemi­anopia. A review of 104 cases. Arch Ophthalmol 1973; 89: 377- 81. 18. Poggel DA, Kasten E, Sabel BA. Attentional cueing improves vision restoration therapy in patients with visual field loss. Neurology 2004; 63: 2069- 76. 19. Sabel BA, Kenkel S, Kasten E. Vision restoration therapy ( VRT) efficacy as assessed by comparative perimetric analysis and subjec­tive questionnaires. Restorative Neurol Neurosci 2004; 22: 399^ 20. 20. Mueller I, Poggel DA, Kenkel S, Kasten E, Sabel BA. Vision restoration therapy ( VRT) after brain damage: subjective improve­ments of activities of daily life and their relationship to visual field enlargements. Vision Impairment Research 2003; 5: 157- 78. 21. Rossi PW, Kheyfets S, Reding Ml Fresnel prisms improve visual perception in stroke patients with homonymous hemianopia or unilateral visual neglect. Neurology 1990; 40: 1597- 9. 22. Rossetti Y, Rode G, Pisella L, Fame A, Li L, Boisson D, Perenin MX Prism adaptation to a rightward optical deviation rehabilitates left hemispatial neglect. Nature 1998; 395( 6698): 166- 9. 23. Gottlieb DD, Allen CH, Eikenberry J, Ingall- Woodruff S, Johnson M. Living with vision loss: independence, driving, and low vision solutions. Atlanta: SX Barthelemy Press; 1996. 24. Hedges XR Jr, Stunkard, J, Xwer A. Fresnel- Prismen- ihr Stellenwert in der Rehabilitation homonymer Hemianopsien. Klin Monatsbl Augenheilk 1988; 192: 568- 71. 25. Peli E. Treating hemianopia using prisms to create peripheral diplopia. In: Stuen C, Arditi A, Horowitz A, et al, eds. Vision Reha bilitation. Assessment, intervention and outcomes. Lisse: Swets and Zeitlinger; 2000: 104- 9. 26. Peli E. Field expansion for homonymous hemianopia by optically induced peripheral exotropia. Optom Vis Sci 2000; 77: 453- 64. 27. Schulte T, Strasburger H, Muller- Oehring EM, Kasten E, Sabel BA. Automobile driving performance of brain- injured patients with visual field defects. Am JPhys Med Rehabil 1999; 78: 136^ 2. 28. Amtsblatt der Europaischen Gememschaften Nr. L 2371/ 20 vom 24.08.1991, Anhang III. 29. Fahrerlaubnis- Verordnung ( Fey 01.01.1999; FeVAnd V, 23.08.2002) Driving Licence Law from 01.01.1999 and changes from 23.08.2002. 30. Empfehlung der Deutschen Ophthalmologischen Gesellschaft und des Berufsverbandes der Augenarzte Deutschlands zur Fah-reignungsbegutachtung fur den Strajienverkehr 2003, 3rd ed. 31. Aulhorn E. Uber Fixationsbreite und Fixationsfrequenz beim Lesen gerichteter Konturen. Pflilgers Arch Physiol 1953; 257: 318- 28. 32. Legge GE, Ahn SJ, Klitz XS, Luebker A. Psychophysics of reading- XVI. Xhe visual span in normal and low vision. Vision Res 1997; 37: 1999- 2010. 33. Xrauzettel- Klosinski S, Brendler K. Eye movements in reading with hemianopic field defects: the significance of clinical parameters. Graefes Arch Clin Exp Ophthalmol 1998; 236: 91- 102. 34. McConkie GW, Rayner K. Xhe span of the effective stimulus during a fixation in reading. Percept Psychophys 1975; 17: 578- 586. 35. Xrauzettel- Klosinski S. Reading disorders due to visual field defects- a neuro- ophthalmological view. Neuro- Ophthalmology 2002; 27: 79- 90. 36. Xrauzettel- Klosinski S, Reinhard J. Xhe vertical field border in hemianopia and its significance for fixation and reading. Invest Ophthalmol Vis Sci 1998; 39: 2177- 86. 37. Bischoff P, Lang J, Huber A. Macular sparing as a perimetric artifact. Am J Ophthalmol 1995; 199: 72- 80. 38. Horton JC. Disappointing results from Nova Vision's visual restoration therapy. Br J Ophthalmol 2005; 89: 1- 2. 39. Plant GX. A work out for hemianopia. Br J Ophthalmol 2005; 89: 2. 149