Journal of Neuro-Ophthalmology, December 1997, Volume 17, Issue 4

Aspirin Reduces the Incidence of Second Eye NAION

The objective of this study was to determine if aspirin reduces the incidence of second eye involvement after nonarteritic anterior ischemic optic neuropathy (NAION) in one eye. Records were reviewed of 131 patients who sustained unilateral NAION. Of these, the 33 patients who sustained second eye NAION were compared to those followed for a minimum of 2 years without sustaining a second eye NAION (67). Thirty-one of the 131 patients were excluded because of inadequate follow-up. Except for diabetes (relative risk [RR] 1.43, p = 0.05), the incidence of second eye NAION was independent of gender, age, cup/disk, hypertension, anemia, and migraine. The degree of visual acuity or field dysfunction in the first eye correlated poorly with the acuity (r = 0.28) and field (r = 0.33) loss in the second eye. Aspirin (65-1,300 mg) taken two or more times per week decreased the incidence (17.5% vs. 53.5%) and relative risk (RR = 0.44, p = 0.0002) of second eye AION regardless of the usual risk factors. Even after eliminating those patients who had bilateral disease when first referred, ASA still reduced the incidence of second eye involvement (35% vs. 13%, RR = 0.74, p = 0.01). Aspirin may be an effective means of reducing second eye NAION.

Journal of Neuro- Ophthalmology 17( 4): 250- 253, 1997. © 1997 Lippincott- Raven Publishers, Philadelphia Aspirin Reduces the Incidence of Second Eye NAION: A Retrospective Study Mark J. Kupersmith, M. D., Larry Frohman, M. D., Matthew Sanderson, M. D., Jennifer Jacobs, M. D., James Hirschfeld, M. D., Christie Ku, M. D., and Floyd A. Warren, M. D. The objective of this study was to determine if aspirin reduces the incidence of second eye involvement after nonarteritic an­terior ischemic optic neuropathy ( NAION) in one eye. Records were reviewed of 131 patients who sustained unilateral NAION. Of these, the 33 patients who sustained second eye NAION were compared to those followed for a minimum of 2 years without sustaining a second eye NAION ( 67). Thirty- one of the 131 patients were excluded because of inadequate fol­low- up. Except for diabetes ( relative risk [ RR] 1.43, p = 0.05), the incidence of second eye NAION was independent of gen­der, age, cup/ disk, hypertension, anemia, and migraine. The degree of visual acuity or field dysfunction in the first eye correlated poorly with the acuity ( r = 0.28) and field ( r = 0.33) loss in the second eye. Aspirin ( 65- 1,300 mg) taken two or more times per week decreased the incidence ( 17.5% vs. 53.5%) and relative risk ( RR = 0.44, p = 0.0002) of second eye AION regardless of the usual risk factors. Even after elimi­nating those patients who had bilateral disease when first re­ferred, ASA still reduced the incidence of second eye involve­ment ( 35% vs. 13%, RR = 0.74, p = 0.01). Aspirin may be an effective means of reducing second eye NAION. Key Words: Aspirin- Nonarteritic anterior ischemic optic neuropathy. Nonarteritic anterior ischemic optic neuropathy ( NAION) has an annual incidence of approximately 2.30/ 100,000 in individuals over 50 years of age ( 1), causing visual loss in 1,500- 6,000 new cases per year in the United States ( 2). There is no proven therapy that will reverse the visual disturbance once it occurs ( 3). The resulting visual loss is often severe because > 41% of affected eyes have visual acuity < 20/ 100 ( 4,5). Because second involvement occurs in 24% ( 5) to 48% of af­fected subjects ( 6), this disorder frequently has a major impact on daily living functions. Manuscript received September 16,1996; accepted August 15,1997. From INN, Beth Israel Medical Center ( M. J. K.) and the Departments of Ophthalmology and Neurology, New York University School of Medicine ( M. J. K., M. S., J. H., C. K., F. A. W.), New York, New York, and New Jersey Medical School ( L. F.), Newark, New Jersey, U. S. A. Address correspondence and reprint requests to Dr. Mark J. Kupersmith, INN, Beth Israel North, 170 East End Avenue, New York, NY 10128, U. S. A. The authors have no commercial interest in the results of this study. Although there is no pathologic confirmation of a thrombosed artery in this disorder, NAION appears to result from ischemic damage and probably infarction of the anterior portion of the optic nerve ( 7). In addition, the risk of developing NAION appears to be increased by the same factors, such as diabetes mellitus and systemic hy­pertension, which also increase the risk of myocardial and cerebral infarct ( 8,9). Crowding of the optic nerve, associated with a small cup: disk ratio, may compromise arterial perfusion and seems to predispose or increase the risk of optic nerve ischemia ( 10,11). Measures that re­duce intravascular thrombosis might reduce the risk of developing NAION in patients who appear at risk be­cause of a small cup: disk ratio, systemic hypertension, or diabetes mellitus. However, because individuals without visual symptoms may not be routinely examined before an episode, it would be difficult to prescribe an interven­tion to prevent first eye NAION. If an agent could be found that lowers the incidence of second eye NAION, it would be relatively easy to identify and treat patients after first eye involvement. It would also be helpful to identify those patients, who, because of high risk factors, are most likely to experience second eye NAION, and therefore might benefit most from early prophylactic therapy. METHODS At the time of record review, we identified 131 pa­tients diagnosed as having nonarteritic anterior ischemic optic neuropathy when examined at NYUMC or NJDMJ by a neuro- ophthalmologist ( M. J. K., F. A. W., or L. F.). One hundred patients were found who met the criteria for evaluation in this study. These included age > 49 years, definite NAION, and follow- up for a minimum of 2 years by a neuro- ophthalmologist or referring ophthal­mologist, or if followed for a shorter duration with di­agnosed second eye NAION. Patients were considered to have NAION if they had symptoms of sudden visual loss, a relative afferent pupillary defect, swelling of the optic disk without vitreous cells if seen acutely, and pal­lor of the optic disk if seen after the event. At the time of diagnosis ( before chart review), patients were not con- 250 SECOND EYE NAION 251 sidered to have NAION if they had symptoms, signs, or laboratory evidence of giant cell arteritis, narrow- angle glaucoma, persistent intraocular pressure > 30 mm Hg and glaucomatous cupping of the optic disk, a temporal quadrantanopia or hemianopia, significant pain on eye movement, a normal appearance of the optic disk, posi­tive serologic test for syphilis, signs of ischemic ocul­opathy or acute retinal arterial or venous occlusion, or diagnosed multiple sclerosis. Because NAION may be an ischemic disease, it was suggested that patients take daily aspirin after the first eye event, if they did not have a condition that was a contraindication. Every patient was queried as to the dose and frequency of aspirin use, but there was no additional verification process. Those pa­tients not taking aspirin at least twice weekly or not within 2 weeks of the second eye NAION made up the nonaspirin group. No patients were given other antico­agulants. For those patients who did not return for a minimum follow- up of 2 years or did not have a current neuro-ophthalmologic evaluation at the time of record review, telephone contact to the patient and referring doctor were attempted. Follow- up examinations were arranged with one of the authors ( M. J. K., F. A. W., L. F.) or the patient's ophthalmologist. In addition, a questionnaire was sent to the referring doctors. Despite these efforts, 31 patients who were seen early enough to allow for 2 years of follow- up were lost to follow- up to the neur­ophthalmology services and their referring ophthalmolo­gists. The clinical evaluation included the best corrected Snellen acuity; color vision testing using pseudoisochro-matic plates; visual field analysis using tangent perim­etry at 1 m, Goldmann perimetry, or threshold perimetry ( the same technique was used for follow- up evaluations); pupillary response to a bright hand- held light, applana­tion tonometry, slit- lamp biomicroscopy, and ophthal­moscopy after dilation, noting the cup: disk ratio in the unaffected eye at baseline. The visual fields were cat­egorized as 0 ( normal), 1 ( arcuate nerve fiber bundle defect), 2 ( relative central [< 6 degrees], centrocecal sc­otoma, or altitudinal defect); 3 ( altitudinal field defect plus additional loss); or 4 ( no light perception). The vi­sual acuity was expressed in decimal form, and finger counting, hand motion, light perception, and no light perception vision were assigned values of 0.01, 0.005, 0.001, and 0, respectively. The data for each subject were summarized using a uniform questionnaire. Without prior grouping for treat­ment or second eye events, individual patient data were entered into a computer for analysis by a data manager. Analysis was performed using the statistical package in Epi Info, Version 6 ( 12), and the relative risks ( RR) with 95% confidence intervals ( CI) of developing second eye NAION were calculated ( 13). Regression analysis was performed, and correlation coefficients with 95% CI were calculated for visual acuity field data ( 14). A one-tailed Fisher exact test was used to calculate probabilities for data with values less than five. RESULTS Of the 100 patients, 57 ( 36 men, 21 women) were taking aspirin and 43 ( 21 men, 22 women) were not. The group taking aspirin had a mean age of 66.5 years, whereas the second group had a mean age of 62.5 years ( no statistical difference). Ten patients taking aspirin had a second eye NAION event, whereas 23 of those not taking aspirin had their second eye affected. Aspirin re­duced the relative risk of second eye NAION to 0.44 ( CI, 0.40- 0.79). If one assumes that the 31 patients lost to follow- up did not have a second eye event and did not take aspirin, the group not taking aspirin would have had a 32% incidence of second eye NAION, which is still considerably higher than the 17.5% incidence in the as­pirin- treated group. Even after eliminating those patients who had bilateral optic neuropathy when referred, aspirin reduced the relative risk of second eye NAION ( RR = 0.74, p = 0.01). The relative risk and significance of the treatment effect are minimally altered after adjusting for possible confounding factors, such as age > 60 years ( RR = 0.61, p = 0.0001), cup: disk ratio ^ 0.2 ( RR = 0.53, p = 0.002), female gender ( RR = 0.55, p = 0.004), or diabetes mellitus ( RR = 0.32, p = 0.04). The dose of aspirin used was too variable to determine whether the frequency or dose had an effect on the inci­dence or severity or acuity or field loss ( Table 1). Aspirin may have reduced the severity of visual loss in second eye NAION. The visual acuity was ^ 20/ 400 in one aspirin- treated patient compared with six patients not taking aspirin. A grade 3 visual field or worse occurred in the second eye of four patients on aspirin and nine patients not on aspirin. Regression analysis of the visual acuity of the first eye NAION at presentation and at 3 years gave a correlation coefficient of 0.79 ( CI, 0.69- 0.87). The acuity was more than two lines better, more than two lines worse, and unchanged in 18.1%, 9.7%, and 72.2% of eyes, respec­tively. In contrast, the visual field defect at presentation and at 3 years ( 69 eyes) had a lower correlation coeffi­cient of 0.60 ( CI, 0.42- 0.73). The visual field defect was at least one grade better, worse, and unchanged in 18.8%, 14.5%, and 66.7%, respectively. Regression analysis of the visual acuity of the first eye NAION at presentation and the acuity of the second eye NAION showed poor correlation ( r = 0.28; CI, - 0.07 to 0.57) for each patient. The second eye acuity was three lines better, three lines worse, and equivalent to the first TABLE 1. Dose of aspirin (> 3 times/ week) and second eye NAION Second eye NAION Aspirin ( mg) No Yes 65 1 1 130 1 0 195 0 1 325 32 5 650 12 3 1300 1 0 J Neuro- Ophthalmol, Vol. 17, No. 4, 1997 252 M. J. KUPERSMITH ET AL. eye acuity in 21.2%, 45.5%, and 33.3% of eyes, respec­tively. The correlation coefficient for the visual field de­fect for the first and second eye NAION ( 32 cases) was 0.33 ( CI, - 0.02 to 0.61). The second eye visual field defect was at least one grade better, worse, and equiva­lent in 31.3%, 15.6%, and 53.1%, respectively. The second eye episode developed within 0.02- 16 years after the first eye NAION, with 22 of 33 events occurring within the first year ( Table 2). The frequency of second eye AION appeared to be unaffected by age, hypertension, cup: disk ratio, or gender ( Table 3). The number of patients with significant anemia or migraine was too low to evaluate. Of the risk factors usually as­sociated with the development of NAION, only diabetes mellitus increased the relative risk of developing second eye NAION ( RR = 1.43; CI, 0.93- 2.22) ( Table 4). DISCUSSION Aspirin appears to significantly reduce the incidence of second eye NAION. This effect seems independent of the risk factors usually considered important for devel­oping first eye NAION. In fact, except for diabetes mel­litus, the occurrence of second eye NAION is also inde­pendent of these risk factors, including age, small cup: disk ratio, and hypertension. Our finding that incidence of bilateral NAION is not increased for patients with systemic hypertension over those patients with normo-tension has been previously noted ( 8). These data suggest that once an individual experiences an episode of NAION, the same risk factors for having the first eye affected are less important for developing a subsequent second eye attack. Alternatively, a more complex rela­tionship exists for these or additional elements as risks for second eye NAION. There is no immediate explanation why aspirin ap­pears to have such a profound benefit on the incidence of second eye NAION but has not been found to reduce the incidence of first eye NAION ( 15). Also, although aspi­rin had no effect on the development of severe visual field loss, visual acuity loss of =£ 20/ 400 occurred in only 10% of treated versus 26% of untreated patients with second eye events. The data are inadequate to determine whether the dose or frequency of aspirin could have an effect on the second eye incidence or severity of attack. The optimal dose of aspirin remains to be determined and cannot be inferred from prior trials in stroke preven­tion because the debate on aspirin dosing to prevent ce­rebral infarct remains unresolved. North American neu­rologists recommend a daily dose of 650- 1,300 mg ( 16), and European neurologists recommend as little as 30 mg TABLE 2. Years from first to second eye NAION Range No. of patients 0.02- 0.5 11 0.51- 1.0 11 1.1- 2.0 5 3.0- 5.0 3 8.0- 16 3 TABLE 3. Risk factors for NAION in 100 patients at baseline NAION in only NAION developed one eye in second eye ( no. of patients) ( no. of patients) Age ( yr) « 60 > 60 Gender F M Cup: disk unaffected eye < 0.2 > 0.2 Migraine headaches Migraine headaches No migraine headaches Blood pressure Hypertensive Normotensive Diabetes mellitus Diabetic Not diabetic Anemia ( Hct < 32) Anemic Not anemic Aspirin Aspirin use No aspirin Aspirin'' Aspirin use No aspirin 52 15 30 37 44 ( 50)" 16 ( 18)" 4 29 27 40 11 56 0 33 47 20 47 20 21 12 13 20 23 ( 26)° 5 ( 5.7)° 0 67 12 21 11 22 0 67 10 23 . 7 11 " Values in parentheses are percentage of patients; 12 additional subjects described as only having a normal second eye cup: disk ratio were not included. * Includes only those patients who did not have bilateral optic neu­ropathy when first evaluated by M. J. K. or L. F. ( 17). However, stroke that is often due to emboli from large extracranial or intracranial arteries differs from NAION, which does not appear to be caused by emboli. NAION may be similar to lacunar infarcts, which may be more adequately prevented with 600- 1,300 mg of daily aspirin compared with a lower dose ( 18). The degree of first eye visual acuity loss did not cor­relate with the acuity in the affected second eye, and only 33.3% had similar visual acuities in both eyes. This is contrasted with the results of a recent study, where 57% had similar visual acuity outcome for the two eyes ( 19). However, the dissimilarity between the acuity of both eyes has been reported before ( 4). Although patients ( 18.1% in our study) with all levels of acuity loss can TABLE 4. Factors that may affect relative risk of developing second eye NAION Aspirin Hypertension Diabetes mellitus Cup: disk < 0.2 Age > 60 yr Gender ( women) RR" 0.44 0.94 1.43 1.12 1.28 0.93 P 0.0002 0.7 0.05 0.82 0.14 0.61 95% CI 0.40- 0.79 0.72- 1.25 0.93- 2.22 0.89- 1.41 0.89- 1.85 0.71- 1.22 " RR < 1.0 suggests reduced risk; RR > 1.0 suggests increased risk. J Neuro- Ophthalmol, Vol. 17, No. 4, 1997 SECOND EYE NAION 253 demonstrate small amounts of recovery ( 4,6), the acuity in - 72% of first eye NAION at 3 years in our study and in 100% at 5 years in the study of Repka et al. ( 5) were similar to the baseline acuity loss of the first eye. Although the second eye NAION occurred as late as 16 years after the first eye attack in our patients who developed second eye NAION, 68.8% occurred within 1 year of the first eye, which is significantly higher than the 33% reported by Boghen and Glaser ( 4), slightly higher than the 51% incidence for the same duration reported by Beri et al. ( 8), but similar to the 59% reported by Sawle et al. ( 9). Our incidence of 53.5% of second eye NAION in un­treated subjects seems too high. Although the reported range of second eye attacks varies, an incidence of 35% ( 8) to 40% ( 20) is typical. Our data might be biased because patients who experience a second eye episode and worse overall visual functioning would tend to return for evaluation and treatment. In fact, many patients had optic neuropathy involving both eyes at the time of re­ferral for neuro- ophthalmologic evaluation. After ex­cluding these patients, 13% of patients using aspirin and 35% of patients not using aspirin experienced second eye NAION. Our results may suffer from the factors that can affect any retrospective study. Patients not included because of inadequate follow- up could provide a significant source of bias. However, there is no reason to assume that they would preferentially fall into either the aspirin treatment or no treatment group. Although the aspirin- treated and untreated groups appeared to have similar characteristics, some additional factor such as the motivation to take aspirin or perform other measures that might benefit their health could have influenced the results. Also, patients took their own aspirin; the frequency and doses used were not verified. The data, although retrospective and incomplete, sug­gest that it might be reasonable to give aspirin to patients with first eye NAION. A prospective clinical trial that corroborates these results and determines the safest dose of aspirin or some other platelet aggregation blocking agent could have a potentially profound effect on the treatment of this disorder and reduce long- term visual disability. Acknowledgment: This work was supported by R. L. Kohns Foundation and the Research to Prevent Blindness, Inc. REFERENCES 1. Johnson L, Arnold A. Incidence of non- arteritic anterior ischemic optic neuropathy: population based study in the state of Missouri and Los Angeles County, California. J Neuroophthalmol 1994; 14: 38- 44. 2. Ischemic Optic Neuropathy Decompression trial ( IONDT) Re­search Group. Manual of Operations. Baltimore: University of Maryland, 1992, chap 1: 12. 3. Ischemic Optic Neuropathy Decompression Trial Research Group. Optic nerve decompression surgery for nonarteritic anterior isch­emic optic neuropathy ( NAION) is not effective and may be harm­ful. JAMA 1995; 273: 625- 32. 4. Boghen DR, Glaser JS. Ischaemic optic neuropathy. Brain 1975; 98: 689- 708. 5. Repka M, Savino P, Schatz N, Sergott R. Clinical profile and long- term implications of anterior ischemic optic neuropathy. Am J Ophthalmol 1983; 96: 478- 83. 6. Ellenberger C, Keltner JL, Burde RM. Acute optic neuropathy in older patients. Arch Neurol 1973; 28: 182- 5. 7. Cogan DG. Neurology of the Visual System. Springfield, IL: Charles C Thomas, 1972: 186. 8. Beri M, Klugman MR, Kohler JA, Hayreh SS. Anterior ischemic optic neuropathy. VII. Incidence of bilaterality and various influ­encing factors. Ophthalmology 1987; 94: 1020- 1028. 9. Sawle G, James C, Ross Russell R. J Neurol Neurosurg Psychiatry 1990; 530: 830- 3. 10. Beck RW, Savino PJ, Repka MX, Schatz NJ, Sergott RC. Optic disc structure in anterior ischemic optic neuropathy. Ophthalmol­ogy 1984; 91: 1334- 7. 11. Doro S, Lessell S. Cup- disc ratio and ischemic neuropathy. Arch Ophthalmol 1985; 103: 1143- 4. 12. Dean AG, Dean JA, Coulombier D, et al. Epi Info, Version 6. Atlanta: Division of Surveillance and Epidemiology, Epidemiol­ogy Program Office, Centers for Disease Control and Prevention, 1994: 73- 546. 13. Fleiss JL. Statistical Methods for Rates and Proportions. New York: Wiley, 1981: 56- 82. 14. Hassard TH. Understanding Biostatistics. St. Louis: Mosby- Year Book, 1991: 121- 51. 15. Jay WM, Zdral JD. Anterior ischemic optic neuropathy and daily aspirin antiplatelet therapy: a retrospective study. Invest Ophthal­mol Vis Sci 1995; 35: 1542. 16. Dyken ML, Barnett HJM, Fields WS, et al. Low- dose aspirin and stroke: " It ain't necessarily so." Stroke 1992; 23: 1395- 9. 17. The Dutch TIA Trial Study Group. A comparison of tow doses of aspirin ( 30 mg vs. 283 mg a day) in patients after a transient ischemic attack or minor ischemic stroke. N Engl J Med 1991 ; 325: 1261- 6. 18. Weisberg L. Retrospective analysis of aspirin and ticlopidine pre­venting recurrent stroke following an initial lacunar infarct. J Stroke Cerebrovasc Dis 1995; 5: 44- 48. 19. Boone J, Massry G, Holds J, Chung S. Comparison of visual outcome in bilateral non- arteritic anterior ischemic optic neurop­athy. Ophthalmology 1994; 101( suppl 9A): 112. 20. Hayreh SS, Podhajsky P. Visual field defects in anterior ischemic optic neuropathy. Doc Ophthalmol Proc Ser 1979; 19: 53- 71. J Neuro- Ophthalmol, Vol. 17, No. 4, 1997