10-Diopter Fixation Test for Amblyopia

Transcription

1 10-Diopter Fixation Test for Amblyopia Kenneth W. Wright, MD; Frances Walonker, DBO, CO, COT; Paula Edelman, CO \s=b\fixation-preference testing has been useful in detecting amblyopia in children unable to give reliable visual acuities, except for children with small-angle tropias and those without manifest deviations. This study introduces a method of fixation testing that includes all children, regardless of the presence or size of the deviation. By placing a 10-diopter vertical prism over one eye, we induce a vertical deviation. Once the eyes are dissociated, fixation preference is evaluated and used to predict the presence of amblyopia. We prospectively studied fixation patterns in 98 children with either small-angle tropia or essentially straight eyes. All children with amblyopia of two lines or more showed abnormal fixation patterns, and those without amblyopia showed normal patterns. We were able to obtain fixation patterns on all children studied, including 13 infants less than 24 months of age. This test proved to be a reliable, objective method for detecting amblyopia in children with straight eyes or in children with microstrabismus of almost any age. (Arch Ophthalmol 1981;99: ) "[functional amblyopia is a disease that affects approximately 1% to 3% of the general population.'4 Of the preschool children with amblyopia, 46% will have no overt sign of strabis mus and the condition in the majori ty will go undetected until the chil dren enter school.'-' Recent literature Accepted for publication Oct 7, From the Department of Ophthalmology, Los Angeles County-University of Southern California Medial Center, and the Children's Hospital of Los Angeles. Reprint requests to Los Angeles County-University of Southern California Medical Center, 1200 N State St, Box 568, Los Angeles, CA (Dr Wright). has emphasized the importance of early detection and treatment of amblyo pia and many authors now believe that for optimum results, therapy should be initiated before 3 years of age.''- "'" Most of the methods pres ently used to detect amblyopia depend heavily on subjective responses and produce unreliable results, especially when the preschool age groups being are tested.--6"8 In young patients with strabismus for whom reliable subjective visual acuities cannot be obtained, binocular fixation patterns have been useful in diagnosing blyopia.8"11 am This test measures fixa tion preference under binocular condi tions in patients with a manifest deviation. The ability to freely alter nate fixation shows equal visual pref erence and strongly militates against the presence of amblyopia. Strong grades of fixation preference, con versely, suggest amblyopia in the nonpreferred eye. Zipf8" showed that binocular fixa tion patterns could be used to reliably predict strabismic amblyopia in young children, with certain important limi tations. Only those patients with a manifest deviation could be studied. This limitation excludes nonstrabismic amblyopia, such as amblyopia associated with anisometropia, ocular trauma, congenital cataract, and con genital glaucoma. Essentially straight postoperative strabismus also has been excluded from binocular fixation testing, leaving amblyopia undetected until these patients become old enough for a reliable test of visual acuity. Another difficulty with bino cular fixation testing has been the diagnosis of amblyopia in cases of small-angle tropias.81" Zipf8 has pointed out that small-angle esotropes show strong fixation preference even in the absence of an amblyopia. His conclusion was that binocular fixation patterns are unreliable in diagnosing amblyopia in the patient with smallangle esotropia. In the present study, we present a new method for evaluating binocular fixation patterns that will enable us to reliably diagnose amblyopia in young children and infants regardless of the size or presence of the deviation. SUBJECTS AND METHODS We prospectively studied the sensory status and binocular fixation patterns in 98 children with manifest deviations of 10 PD or less: 31 nonstrabismic normal control subjects; 41 with small-angle esotropia; three with small-angle exotropia; ten with essentially straight, postoperative strabis mus; nine with straight anisometropic amblyopia; one with Brown's syndrome; and three nonstrabismic patients with cor neal trauma where amblyopia was a con sideration. The ages ranged from 6 months to 16 years and 42% of the children were aged 4 years or less. The patients were seen by three of us independently at two sepa rate institutions-the Department of Oph thalmology, Los Angeles County-Universi ty of Southern California Medical Center, and the Children's Hospital of Los Angeles. Our new method for evaluating binocu lar fixation patterns (the 10-diopter fixa tion test) was used on all subjects. A 10-D vertical prism was placed over one eye to induce a vertical deviation. Once the eyes were dissociated, the degree of fixation preference was studied. Fixation was attracted using only accommodative tar gets, such as wiggle toys, finger puppets, and cartoon movies. The ability to sponta neously alternate fixation shows equal visual preference and we recorded this as "alternates." If ocular dominance was present, the degree of preference was quantitated by momentarily covering the fixing, dominant eye to force refixation of

2 the nonpreferred eye. After the cover was removed, if the nonpreferred eye could hold steady fixation for at least 5 s, hold through smooth-pursuit movement, or hold through a blink before refixation to the dominant eye, we would record "prefers one eye but 'holds well' with the nonpre ferred eye." This characteristic indicates the ability of the nonpreferred eye to maintain fixation under binocular condi tions. For simplicity, the patterns of "alter nates" and "holds well" will be referred to as "normal" throughout the text. Refixation to the dominant eye that occurs promptly after cover removal shows a strong disregard for the nondominant eye and indicates an abnormally high grade of ocular preference. The strongest grade of preference is when the nondominant eye will not hold fixation at all and immediately refixes to the dominant eye. In this case, we recorded "will not hold." If the nondominant eye could hold fixation for 1 to 2 s, we recorded "holds momentari ly." When refixation was delayed 3 s, yet the nondominant eye would not hold fixa tion through smooth pursuit or a blink, we recorded "holds briefly." These strong grades of fixation preference of "will not hold," "holds momentarily," and "holds briefly" will be referred to as "abnormal" patterns. Ocular alignment was determined by cover-uncover test and alternate-cover testing at distance and near. Sensory test ing consisted of visual acuity (ie, Allen cards, -game, or letters), 4-D base-out prism test, stereo acuity (titmus), and Worth four-dot testing. Testing was per formed at distance and near; however, only near responses were used for comparison because of the poor attention paid to dis tant targets by the younger children. In those patients who were unable to cooper ate with a given test or when the 4-D base-out responses were atypical, we recorded "inconclusive." Visual acuity could be established in 79 of the 98 children, allowing for direct correlation of fixation patterns with the presence of amblyopia. In those for whom visual acuity could not be obtained and with fixation patterns indicative of amblyopia, patching was ini tiated on the basis of Follow-up information was obtained in 12 treated patients to document changes in RESULTS Fixation Patterns With The 10-D Fixation Test Of the 98 patients studied, 79 were able to have their visual acuity deter mined (Fig 1) and therefore fixation could be correlated with the patterns presence of amblyopia. Fixation pat terns were taken with the 10-D fixa tion test and then were compared with the degree of amblyopia, ie, amblyo pia (> two lines of difference), bor derline (one line of difference), and no amblyopia (< one line of difference) in visual acuity. Amblyopia. Of 79 patients, 13 had amblyopia and all 13 showed abnormal Borderline Amblyopia. Six patients had borderline amblyopia; four showed normal and two showed abnormal No Amblyopia. Sixty patients were nonamblyopic and they all had normal Patterns With and Without The 10-D Fixation Test Of the 79 children with obtainable visual acuities, 16 had deviations large enough to permit evaluation of binoc ular fixation patterns without the aid of the 10-D fixation test (Fig 2). Fixa tion patterns were taken with and without the vertical prism and then were compared with the presence of amblyopia as determined by visual acuity testing. Amblyopia.-Of the 16 patients, four had two or more lines of amblyopia. Fixation patterns taken with and without the prism test were similar, both methods demonstrating abnor mal Borderline Amblyopia. None of the 16 patients had one line of amblyo pia. No Amblyopia. Twelve children were not amblyopic, having less than one line of difference in visual acuity. When studied without the 10-D fixa tion test, ten of the nonamblyopic patients gave abnormal fixation pat terns. Fixation patterns were re peated with the 10-D prism, and all previously abnormal patterns changed to normal. Follow-up Data on 12 Treated Children with Amblyopia We obtained follow-up data on 12 patients with amblyopia who were started on occlusion therapy during our study (Fig 3). The treatment peri od ranged from one to five months and consisted of patching the nonpre ferred eye. Amblyopia in these chil dren was associated with small-angle esotropia, anisometropia, and refrac tive esotropia. Amblyopia Resolved. Of the 12 pa tients, seven showed resolution of the amblyopia after occlusion therapy. These patients also demonstrated a corresponding shift in fixation pat terns, going from abnormal to normal after patching. Residual Amblyopia. Three showed residual amblyopia of three lines or more and all continued with abnormal No Improvement. The two remain ing patients did not respond to occlu sion therapy and demonstrated no change in either fixation patterns or levels of amblyopia. Detection of Amblyopia in Preschool Children We compared the usefulness of the Worth four-dot, 4-D base-out prism test, stereo acuity, and 10-D fixation test in predicting amblyopia in pre school children (Fig 4). Of the 98 patients studied, 41 fell into this cate gory, ranging from 6 months to 4 years of age. Of the 41 preschoolers, 21 were able to give visual acuities: 17 nonamblyopic, one borderline, and three amblyopic. Worth Four-Dot.-Of the 41 children, 66% could not perform the Worth four-dot test even when simplified by allowing identification of the dots by touch at near. Nine children without amblyopia and two children with amblyopia gave reliable responses and all demonstrated fusion. 4-D Base-Out Test. Seventy-three percent gave inconclusive responses to 4-D base-out testing. Six children without amblyopia gave conclusive responses: four fusing and two show ing supression. Stereo Acuity. Fifty-six percent would not regard the stereo acuity test and gave inconclusive responses. Of 17 children without amblyopia, the 13 who gave reliable responses all showed some degree of stereopsis, 3,000 arc seconds or better. The three with amblyopia had varied responses; one flat, one 3,000 arc seconds, and one inconclusive. COMMENT We found binocular fixation pat terns taken with the 10-D fixation test to be extremely reliable in diag nosing amblyopia, both in children with small-angle tropias and in essen tially straight-eyed children (Fig 1). All patients with documented amblyo pia of two lines or more showed abnor mally high grades of fixation prefer ence. In the children without amblyo pia with equal or one-half line differ ence, all demonstrated normal binocu lar Those patients with borderline amblyopia of one line difference showed varied responses. During our study, three patients were found to have latent refractive errors when fixation patterns were normal, yet the visual acuities indi cated amblyopia. These cases demon strate some of the difficulties in using visual acuity testing to diagnose amblyopia, especially when dealing with the preschool age group."- " Refractive errors, fatigue factors, and behavioral problems can all lead to an

3 ==3 tines 2 tines 25 E < 1 tine t5.e Vi Line Equal Alternates Iff * «? H ititmmi Holds Well Il M mu lllllllllll llllllllll Holds Holds Will Not Briefly Momentarily Hold I min nm Fig 1. Correlation of fixation patterns in 79 children with either small-angle tropia or essentially straight eyes examined with 10-diopter prism test. There is linear relationship of lines of amblyopia to degree of fixation preference. Fig 3. Twelve amblyopic children with follow-up examinations after occlusion therapy. Note changes in fixation patterns corre late well with changes in lines of amblyopia. Amblyopia Resolved Residual Amblyopia Before Patching Lines of Fixation Age, yr Amblyopia Pattern 5"/2 10% NO NO No Improvement 11 HW Holds Well H B Holds Briefly HM Holds Momentarily HM After Patching Lines of Fixation Amblyopia Pattern NO HW HW ALT ALT HW Equal Vision Will Not Hold NO Not Obtainable ALT Alternates Freely Fig 2. Fixation patterns with and without 10-diopter prism in 16 children with small-angle tropia. Note inappropriately high grades of ocular preference in children with equal vision before 10-D prism was used. Fixation patterns taken with prism now correlate well with presence of amblyopia. Alternates Without Prism Holds Well Holds Momentarily Holds Briefly Will Not Hold Alternates Holds Well With Prism Holds Momentarily Holds Briefly Will Not Hold 2=3 Lines S- 2 Lines II II < 1 Line '/2 Line Equal II tm» nm mm Fig 4. Comparison of sensory test results in predicting amblyopia in 41 children between 6 months and 4 years of age. High incidence of inconclusive responses to Worth four-dot (66%), 4-diopter base-out (73%), and stereo acuity (56%) testing. Reliable responses were obtained on all patients examined with 10-D vertical prism test. Flat 3,000 Nonamblyopic Borderline No Acuity Obtainable Stereo Acuity Suppression Worth Four-Dot Test Fusion immii II II 4-D Base-Out Test nconclusive Response Suppression Fusion numi mum m««» Inconclusive Response tut«nm muni imm 10-D Base-Down Fixation Test Holds 200 «100 Inconclusive Alternates Holds Well Holds Briefly Momentarily Will Not Hold Nonamblyopic II» ««««««I Borderline No Acuity Obtainable ««II««««««««

4 5. This diagram shows fixation patterns in three subgroups of monofixational syndrome when tested with 10-diopter fixation test; ie, I Equal Visual Preference, II Ocular Dominance, Amblyopia. (A) Monofixational syndrome with right facultative scotoma Fig and peripheral fusion which is dependent on central fixation of the left eye. (B) Introduction of 10-D vertical prism over right eye displaces image inferiorly out of central facultative scotoma. Patient attempts to fuse 10-D disparity but cannot and is left with vertical diplopia. I-C (Equal Visual Preference), Spontaneous shift of central fixation to right diplopie image, moving fixation off left fovea and breaking monofixational pattern. I-D, This patient without amblyopia demonstrates the ability to freely alternate central fixation. Il-C (Ocular Dominance), Fixation is forced to right nondominant eye by temporarily occluding left eye. ll-d, When cover is removed fixation is maintained with right eye and induced deviation keeps left image in pericentral location, thus preventing reestablishment of monofixational scotoma. This patient with ocular dominance and no amblyopia will be able to hold fixation with right nondominant eye. Ill-C (Amblyopia), Fixation is forced to right amblyopic eye by temporarily occluding left eye. IIl-D, Fixation is not maintained when cover is removed, and central fixation is rapidly reestablished by left, nonamblyopic eye. This amblyopic patient will prefer left eye and will not hold fixation with right amblyopic eye. erroneous diagnosis of amblyopia. Acuities, when taken with single char acter, ie, Allen picture cards, picturematching games, and single E-games, can discount the crowding phenome non and underestimate or even miss amblyopia. Binocular fixation pat terns, conversely, measure visual preference, which is less dependent on variables such as refractive errors, fatigue, and the crowding phenome non. Since visual preference is ob served, there is no need for a subjec tive response, eliminating errors in communication and allowing amblyo pia testing in the very young. We by no means wish to discourage visual acuity testing of young children, but rather, we stress the importance of binocular fixation testing. However, what of the high falsepositive rate previously seen when binocular fixation patterns are used on small-angle tropias?*1" It is inter esting that children with small-angle no longer show exaggerated forms of fixation preference when examined with the 10-D fixation test (Fig 2). Of 12 nonamblyopic smallangle esotropías studied without the tropias prism, ten (77%) gave fictitiously abnormal amblyopic patterns. When fixation testing was repeated using the 10-D fixation test, all previously abnormal patterns changed to normal, demonstrating their true nonam blyopic state (Fig 2). Six of the ten children were able to give reliable responses to Worth four-dot, stereo acuity, and 4-D base-out testing. All showed peripheral fusion with a facul tative central suppression scotoma in the nondominant eye, indicating the presence of the monofixational syn drome.12 Our contention is that many patients with small-angle tropias have the monofixational syndrome of Parks and that a facultative central scotoma was responsible for the inap propriately high grades of fixation We believe that the 10-D fixation test rectifies these inappro preference. priate patterns by vertically displac ing the nondominant image out of the central scotoma, thereby producing diplopia and disrupting the monofixa tional suppression pattern (Fig 5). Thus, in patients with small-angle tro pias, the 10-D fixation test uncovers the true visual potential of the nondominant eye previously concealed by facultative central scotoma. We also found the 10-D fixation test to be helpful in monitoring the treatment of amblyopia, especially as an end point in deciding when the amblyopia has resolved (Fig 3). Because of the reliability demonstrated by the 10-D fixation test, in correctly diagnosing and following amblyopia, we strongly suggest that all children with abnor mal binocular patterns, ie, "will not hold," "holds momentarily," or "holds briefly," should be considered to have amblyopia, and treatment should be a

5 instituted until those patterns become normal. Likewise, those with normal responses, ie, "alternates freely," or "holds well," can be safely diagnosed as nonamblyopic. To determine the usefulness of the 10-D fixation test in screening pre school children for amblyopia, we com pared the results of the Worth four dot, 4-D base-out, stereo acuity, and the 10-D fixation test in children between the ages of 6 months and 4 years. As might be expected, those tests requiring subjective showed answers a high percentage of inconclu sive responses, ie, 66% in the Worth four-dot and 56% in the stereo acuity test. Since the 4-D base-out test does not require subjective responses, one may have anticipated increased reli ability. However, 73% gave inconclu sive responses to this test and two of six children without amblyopia showed suppression. Romano and Von Noorden7 showed similar results, with normal subjects with only six of 25 equal vision consistently giving nor mal responses to 4-D base-out testing. By using the 10-D fixation test, we were able to obtain reliable responses on all children studied, including 13 infants who were 24 months or younger. Our method of evaluating binocular fixation differs from tests previously described"'1""11 in that we specifically use a 10-D vertical prism and that the degree of fixation preference is quantitated relative to "holds through smooth pursuit" in addition to "holds through a blink." The use of a vertical prism has eliminated some of the dif ficulties encountered when examining patients with horizontal phorias or horizontal suppression scotomas. It has also reduced the size of the prism necessary to dissociate the eyes. This factor keeps the diplopie images closer to the fovea, and avoids a prisminduced blurred second image. We believe that these modifications have improved the accuracy of binocular fixation testing, especially in smallangle tropia. Ingrani- stressed the need for a new method of screening preschool children that would identify amblyopia in both the "squinter" and the "straighteyed amblyope." We believe that the 10-D fixation test fills that need. It is an inexpensive, quick, objective meth od of detecting amblyopia that is reli able in straight-eyed children and in children with small-angle tropia of almost any age. Because this test is reliable in young children, regardless of the presence of a deviation, early surgery to correct strabismus may be considered without the fear of unde tected postoperative amblyopia. Fur thermore, the 10-D fixation test should prove extremely valuable in the difficult problem of nonstrabismic amblyopia, as in congenital cataracts, latent nystagmus, congenital glauco ma, and anisometropia. We share the view of Burian1"' that functional amblyopia should not exist since true functional amblyopia is a preventable disease. It is our sincere hope that through earlier detection and treat ment, amblyopia will soon become a disease of the past. 1. Simons K, Reinecke RD: A reconsideration of amblyopia screening and stereopsis. Am J Ophthalmol 1974;78: Ingram RM: The problem of screening children for visual defects. Br J Ophthalmol 1977;61: Flom MC, Neumaier RW: Prevelance of amblyopia. Public Health Rep 1966;8: Helveston EM: The incidence of amblyopia ex anopsia in young adult males in Minnesota in Am J Ophthalmol 1965;60: Simons K, Reinecke RD: Amblyopia screening and stereopsis, in Symposium on Strabismus: References Transactions of the New Orleans Academy of Ophthalmology. St Louis, CV Mosby Co, 1977, pp Oliver M, Nawrataki I: Screening of pre\x=req-\ school children for ocular anomalies. Br J Ophthalmol 1971;55: Romano PE, Von Noorden GK: Atypical responses to the 4-diopter prism test. Am J Ophthalmol 1969;67: Zipf RF: Binocular fixation pattern. Arch Ophthalmol 1976;94: Zipf RF: Binocular fixation pattern. Arch Ophthalmol 1976;94: Knapp P, Moore S: Diagnostic procedures in an orthoptic evaluation. Am Orthopt J 1962;12: Parks MM: Ocular motility diagnosis. Int Ophthalmol Clin 1963;3: Parks MM: The monofixational syndrome. Trans Am Ophthalmol Soc 1969;67: Miller SD, Judisch GF: Persistent pupillary membrane. Arch Ophthalmol 1979;97: Burian HM: Pathophysiologic basis of amblyopia and of its treatment. Am J Ophthalmol 1969;67:1-12.