Outcomes of Optic Nerve Sheath Decompression for Nonarteritic Ischemic Optic Neuropathy

Efficacy of optic nerve sheath decompression (ONSD) in treating non-arteritic ischemic optic neuropathy (NAION) is not clear. We retrospectively analyzed the records of 91 patients with NAION, who were examined during a two-year period, and compared the final Snellen visual acuities of eyes treated with ONSD with those of eyes that did not have surgery. Seven of 18 eyes with ONSD (39%) demonstrated increased visual acuity of two or more lines; whereas 23 of 71 eyes without surgery (32%) had increased acuity. The ONSD group and no surgery group were further subdivided into eyes with progressive visual loss and nonprogressive visual loss. No statistically significant differences in visual outcome between groups were found. We did not find the high frequency of visual improvement that has been reported in some studies of ONSD for NAION.

Journal of Neuro- Ophthalmology 14( 2): 70- 76, 1994. © 1994 Raven Press, Ltd., New York Outcomes of Optic Nerve Sheath Decompression for Nonarteritic Ischemic Optic Neuropathy Robert D. Yee, M. D., F. A. C. S., Aki K. Selky, M. D., and Valerie A. Purvin, M. D. Efficacy of optic nerve sheath decompression ( ONSD) in treating non- arteritic ischemic optic neuropathy ( NAION) is not clear. We retrospectively analyzed the records of 91 patients with NAION, who were examined during a two- year period, and compared the final Snellen visual acuities of eyes treated with ONSD with those of eyes that did not have surgery. Seven of 18 eyes with ONSD ( 39%) demonstrated increased visual acuity of two or more lines; whereas 23 of 71 eyes without surgery ( 32%) had increased acuity. The ONSD group and no surgery group were further subdivided into eyes with progressive visual loss and nonprogressive visual loss. No statistically significant differences in visual out­come between groups were found. We did not find the high frequency of visual improvement that has been re­ported in some studies of ONSD for NAION. Key Words: Anterior ischemic optic neuropathy- Optic nerve sheath decompression- Natural history- Sur­gery. From the Indiana University School of Medicine, Department of Ophthalmology, Indianapolis, Indiana, U. S. A. This work was supported by a Development Grant from Re­search To Prevent Blindness, Inc., New York, NY. A version of this paper was presented at 129th Meeting of the American Ophthalmological Society on May 24, 1993. Address correspondence and reprint requests to Dr. Robert D. Yee, 702 Rotary Circle, Indianapolis, Indiana 46202, U. S. A. Nonarteritic ischemic optic neuropathy ( NAION) is a frequently encountered disorder. However, the natural course of NAION remains largely un­defined. Several studies that retrospectively exam­ined records of large numbers of patients with NAION reported that spontaneous recovery of vi­sion was uncommon ( 1- 3). More recent studies re­ported higher frequencies of improvement ( Sawle and colleagues ( 4): 7/ 53 eyes or 13%; Movsas and colleagues ( 5): 53/ 242 eyes or 22%). Treatment of NAION has been even more controversial. Vari­ous drugs in anecdotal reports failed to improve vision and thus popularized the belief that NAION was probably not treatable. Renewed interest in treatment arose from a study by Sergott and col­leagues ( 6), who reported marked improvement of visual function after optic nerve sheath decom­pression ( ONSD), especially among NAION pa­tients who demonstrated progressive visual loss over 1- 4 weeks. However, results from subse­quent ONSD studies have been conflicting ( 7- 12). We retrospectively analyzed the results of surgical vs nonsurgical management of 91 consecutive cases of NAION. Patients and Methods Patient Groups The records of all patients with NAION exam­ined by two neuro- ophthalmologists ( R. D. Y. and V. A. P.) between December 1989 and December 1991 were retrospectively reviewed. The diagnosis of NAION was based on a clinical profile of acute, monocular, painless loss of vision and the findings of pallid disc edema, loss of visual function ( visual acuity, visual field, and color vision), and abnor­mal pupillary reaction. No patient had systemic 70 OPTIC NERVE SHEATH DECOMPRESSION FOR NAION 71 symptoms or an elevated Westergren sedimenta­tion rate suggestive of giant cell arteritis. None had a history of prior optic neuritis or demyelinating disease. We identified 91 patients and 91 eyes. However, two patients were excluded because of recent intraocular surgery ( less than 4 weeks) be­fore presentation. The remaining 89 patients comprised the subject population. There were 39 females and 50 males. Their ages ranged from 41 to 81 years with a mean of 60.8 years and standard deviation of 10.9. The following data were recorded: best corrected Snellen visual acuity at presentation and at follow-up examinations, presence of concurrent medical conditions ( hypertension, diabetes mellitus, and cardiovascular and cerebrovascular disease) and if corticosteroid treatment had been initiated. Pro­gressive NAION ( P- NAION) was defined as a fur­ther loss of visual acuity of two lines or more at 1 to 3 weeks after initial presentation. Conversely, nonprogressive NAION ( NP- NAION) was defined as no change of two or more lines. The surgically treated ( ONSD) group consisted of 18 affected eyes, 9 with P- NAION and 9 with NP- NAION. The technique of ONSD included me­dial orbitotomy and fenestration of the optic nerve sheath as described by Sergott and coworkers ( 6). The timing of ONSD was 2 to 6 weeks after initial consultation ( mean: 3 weeks). Preoperative visual acuity immediately before ONSD was compared to postoperative visual acuity at the most recent ex­amination ( range: 4 to 52 months; mean: 11.2). The patients' ages ranged from 41 to 81 years ( mean: 62.8; SD: 11.5). The nonsurgical ( NS) group con­sisted of the remaining 71 NAION eyes, 24 with P- NAION, and 47 with NP- NAION. For NP eyes, visual acuity at presentation was compared to the visual acuity at the most recent examination ( range: 3 to 47 months; mean: 9.3). For P- NAION eyes, initial visual acuity 3 weeks after presenta­tion was compared to the visual acuity at the most recent examination ( range: 4 to 46 months; mean: 10.5). Ages ranged from 42 to 76 years ( mean: 61.9; SD: 9.9). Because of the small number of ONSD eyes and the large difference in sample sizes between ONSD and NS eyes, a subgroup of NS eyes was selected to match the ONSD groups. From the NS group of 71 eyes, a subgroup of 18 eyes was se­lected. They were matched individually to each of the 18 ONSD eyes, according to type of NAION ( P and NP), visual acuity (± 1 line), age (± 2 years), and sex. For ONSD eyes, visual acuity immedi­ately before surgery was used. For matched NS eyes, visual acuity 3 weeks after presentation was selected, since most ONSD eyes had surgery then. Although color vision ( pseudoisochromatic and Ishihara plates), visual fields ( Goldmann and com­puterized, threshold fields), contrast sensitivity, and swinging flashlight test were measured in most patients, the results of these tests were not analyzed since the same methods had not been used for all patients. Statistical Analysis The primary outcome measurement analyzed was the change in Snellen visual acuity scores be­tween presentation or before surgery and the most recent examination. Improvement in visual func­tion was defined as a gain of visual acuity of two lines or more at any time after presentation or sur­gery. This criterion of change in visual acuity was chosen, since it had been used in most previously reported studies of management of NAION. Sev­eral statistical tests were conducted to test the null hypothesis that there was no difference in visual acuity change between patients who underwent ONSD and patients who did not have surgery. The alternate hypothesis was that patients who under­went ONSD had a greater improvement in visual acuity than did the control groups. One- tailed tests were used because we did not expect ONSD to decrease visual acuity. The effect of ONSD vs no surgery on final visual outcome was analyzed us­ing the one- tailed Fisher's exact test for the un­matched samples and the sign test for the matched samples. To possibly increase the statistical power, we also compared the actual changes in visual acu­ity between the ONSD and NS groups using the Wilcoxon rank sum test for the unmatched sam­ples and the signed ranked test for the matched samples. Power calculations were performed using results from previous studies. We assumed the percent­age of patients who underwent ONSD and gained improvement in visual acuity of two lines or more was 87% ( 6- 8) and that 22% of the NAION patients who did not have surgery would also exhibit an improvement of visual acuity ( 5). Based on this difference in these proportions ( 65%), our sample size provided greater than 90% power to detect the difference. In other words, if a difference of 65% in visual acuity improvement between ONSD eyes and NS eyes truly existed, there was greater than a 90% chance of detecting this difference with the sample sizes and groups in our study. RESULTS The clinical profiles of the 89 NAION eyes are summarized in Table 1. The overall frequency of / Neuro- Ophthalmol, Vol. 14, No. 2, 1994 72 R. D. YEE ET AL. diabetes mellitus was 11% ( 10/ 89); systemic hyper­tension, 38% ( 34/ 89), and generalized arterioscle­rosis, 28% ( 25/ 89). The two groups, ONSD vs NS, did not differ significantly for these three associ­ated conditions. Of all 89 NAION eyes, 32 had been treated with corticosteroids on initial presen­tation. Corticosteroids had been administered to 56% of the surgically treated eyes and 31% of the nonsurgically treated eyes. The assignment to groups and the visual out­come in each group of NAION eyes is shown in Table 2. Improvement was defined as an increase of two or more Snellen lines. Decreased visual acu­ity was defined as a loss of two or more lines, and no change was a difference of less than two lines. ONSD was performed for 18 of the 89 NAION eyes. Although the indications for ONSD were dis­cussed with the patients, 71 eyes did not have sur­gery ( NS eyes). Nine of the 18 ONSD eyes had P- NAION, and 9 ONSD eyes were classified as having NP- NAION. Among the 71 NS eyes, 24 had P- NAION, and 47 were classified as having NP- NAION. Progressive worsening of visual func­tion in the first 1- 4 weeks has been reported in up to 25% of patients with NAION ( 1,13). We found a higher prevalence of P- NAION ( 33 of 89 eyes or 37%). A sampling bias could explain this differ­ence, since some physicians might have preferen­tially referred patients with P- NAION to us based on the finding of Sergott and colleagues that ONSD might be more effective in treating P- NAION than NP- NAION ( 6). Surgically Treated Eyes Of the 18 ONSD eyes, 7 ( 39%) showed postop­erative improvement of two lines or more in visual acuity. Figure 1 displays the initial and final visual acuities. The extent of improvement was modest ( only 2- 3 lines) in most patients. One eye with NP- NAION and one eye with P- NAION regained 8 and 12 lines, respectively ( CF to 20/ 40, HM to 20/ 20). Of the 18 ONSD eyes, visual acuity de­creased in 3 eyes ( 17%) and was unchanged in 8 eyes ( 44%). Among the 9 P- NAION eyes, 5 had increased visual acuity ( 56%) and 4 ( 44%) showed no change. No eye had decreased visual acuity. Of the 9 NP- NAION eyes, 2 ( 22%) had improvement; whereas, 3 eyes ( 33%) lost visual acuity, and 4 eyes ( 44%) did not change. Eyes Without Surgery The overall improvement rate among the NS eyes was 23/ 71 or 32%. Visual acuity decreased in 4 eyes ( 6%) and did not change in 44 eyes ( 62%). Among the 18 matched NS eyes, 7 eyes ( 39%) had increased visual acuity, none had decreased acu­ity, and 11 ( 61%) had no change in acuity. In the subgroup of 9 matched P- NAION eyes, 3 ( 33%) had increased acuity, none lost acuity, and 6 ( 67%) had no acuity change. Of the 9 matched NP-NAION eyes, 4 eyes ( 44%) improved, none lost acuity, and 5 ( 56%) did not change. The initial and final visual acuities of the matched NS eyes are shown in Figure 2. Figure 3 shows the same data in the remaining unmatched NS eyes. As in the ONSD group, when improvement occurred it was usually modest, only 2- 3 lines. Three NS eyes had more substantial increases in acuity ( matched P- NAION eye, 5 lines, 20/ 60 to 2/ 20; matched NP-NAION eye, 6 lines, 20/ 60 to 20/ 15; unmatched P- NAION eye, 7 lines, 20/ 200 to 20/ 30). Statistical Analysis Table 3 shows 2 by 2 contingency tables for each group vs frequency of visual improvement. For the purposes of statistical analysis, the decreased and no change categories were combined ( no improve­ment) since the number of cases was small. Fish­er's exact probability distributions were calculated for the unmatched ONSD and NS groups. For all ONSD eyes ( N = 18) vs all NS eyes ( N = 71), no significant difference was found ( P = 0.398). Sim­ilarly, comparison of surgically treated P- NAION TABLE 1. Patient characteristics Optic nerve sheath decompression No surgery No. of patients Mean age: Diabetes Hypertension Arteriosclerosis Corticosteroid P 9 62.7 1 3 3 5 NP 9 62.9 2 1 2 5 P + NP 18 62.8 3 4 5 10 P 24 63.4 3 7 6 10 NP 47 60.1 4 23 14 12 P + NP 71 61.9 7 30 20 22 P, progressive nonarteritic ischemic optic neuropathy; NP, nonprogressive nonarteritic isch­emic optic neuropathy. / Neuro- Ophthalmol, Vol. 14, No. 2, 1994 OPTIC NERVE SHEATH DECOMPRESSION FOR NAION 73 TABLE 2. Visual outcomes ( Change Groups Surgery- P Surgery- NP Matched, no surgery- P Matched, no surgery- NP Unmatched, no surgery- P Unmatched, no surgery- NP Im 5/ 9 2/ 9 7/ 18 3/ 9 4/ 9 7/ 18 7/ 24 16/ 47 23/ 71 3 roved 56% 22% 39% 33% 44% 39% 29% 34% 32% = + 2 lines acuity) Decreased 0/ 9 3/ 9 3/ 18 0/ 9 0/ 9 0/ 18 3/ 24 1/ 47 4/ 71 0% 33% 17% 0% 0% 0% 13% 2% 6% Unchanged 4/ 9 4/ 9 8/ 18 6/ 9 5/ 9 11/ 18 14/ 24 30/ 47 44/ 71 44% 44% 44% 67% 56% 61% 58% 64% 62% P, progressive nonarteritic ischemic optic neuropathy; NP, nonprogressive nonarteritic isch­emic optic neuropathy. eyes ( N = 9) vs NS P- NAION eyes ( N = 24) showed no significant difference ( P = .159). The visual outcomes in ONSD NP- NAION eyes ( N = 9) vs NS NP- NAION eyes ( N = 47) were not sig­nificantly different ( P = .362). The sign test was used to compare visual out­comes in matched ONSD and NS groups. For all ONSD eyes ( N = 18) vs matched NS eyes ( N = 18), the outcomes were not significantly different ( P = .351). The difference between ONSD P- NAION eyes ( N = 9) and matched NS P- NAION eyes ( N = 9) was not significant ( P = .156). Finally, the difference between ONSD NP-NAION eyes ( N = 9) vs matched NS NP- NAION eyes ( N = 9) was also not significant ( P = .266). The chi- square test ( P = .603), Wilcoxon rank sum test ( P = .895) and sign test ( P = .701) failed to show significant differences in visual outcome between the ONSD ( N = 18) and NS ( N = 71) groups. In summary, none of the differences be­tween ONSD and control groups were significant when a variety of statistical tests ( chi- square, PRE- OP VISUAL ACUITY FIG. 1. Preoperative vs postoperative Snellen visual acuity in eyes with progressive NAION ( squares) and nonprogressive NAION ( circles). Wilcoxon rank sum, Fisher's exact f- test, sign and signed- rank) were used to evaluate the changes in visual acuity. Bilateral Nonarteritic Ischemic Optic Neuropathy Sergott and his colleagues ( 6) reported that vi­sual function improved in some nonsurgically treated, previously affected, fellow eyes. In our surgically treated group, 10 patients had fellow eyes that previously had NAION ( 5 P- NAION in the acutely affected, second eye; 5 NP- NAION). Visual acuity did not improve in any of the fellow eyes after surgery in the other eye. In our nonsur­gically treated group, the fellow eye had NAION in 11 patients ( 6 NP- NAION, 5 P- NAION). Among the NS fellow eyes, visual acuity increased by 2 lines in 1 eye ( other eye had P- NAION), 3 lines in 3 < O < > < z 20/ 20 - 20/ 30 • 20/ 50- - 20/ 70- - 20/ 100 - " 20/ 400- CF-HM - LP-NLP - • • • / • • • • • • • NLP LP HM CF 20/ 400 FIG. INITIAL VISUAL ACUITY 2. Initial vs final Snellen visual acuity in eyes without surgery with progressive NAION ( squares) and nonprogressive NAION, which were matched with sur­gically treated eyes. / Neuro- Ophthalmol, Vol. 14, No. 2, 1994 74 R. D. YEE ET Ah. NIP LP HM CF 20/ 400 20/ 100 20/ 70 20/ 50 20/ 30 20/ 20 INITIAL VISUAL ACUITY FIG. 3. Initial vs final Snellen visual acuity in eyes without surgery with progressive NAION ( squares) and nonprogressive NAION ( circles), which were not matched with surgically treated eyes. 1 eye ( other eye P- NAION), and 6 lines in 1 eye ( other eye NP- NAION) during the period of fol­low- up examinations. No correlations were found between the visual outcomes in the acutely af­fected eyes and the visual outcomes in their pre­viously affected, fellow eyes. Clearly, ONSD did not seem to improve visual outcome in the fellow eye. DISCUSSION Sergott and colleagues ( 6) described remarkable results of ONSD in patients with P- NAION in 1989. They reported improved visual function in 12 of 14 eyes ( 86%) with P- NAION and in 1 of 3 eyes ( 33%) with NP- NAION. Rather unexpectedly, they also noted that 2 of 7 eyes with chronically poor visual function secondary to prior NAION recovered some visual acuity after the contralateral eye had undergone ONSD. Spoor and colleagues ( 7) reported that 4 out of 5 eyes ( 80%) with P- NAION showed improved visual acuity and vi­sual field after ONSD. They and others speculated that a cycle of ischemia producing swelling and axoplasmic stasis that resulted in more ischemia might account for the progressive decline of visual function in some patients with NAION. Thus, they concluded that ONSD was an effective treat­ment for P- NAION because it relieved perineural pressure and reduced axoplasmic stasis. Kelman and Elman performed ONSD in 7 pa­tients with NAION ( 8). Five patients had P- NAION and 2 patients had NP- NAION. All pa­tients recovered some visual acuity postopera­tively. Four of five P- NAION eyes and 2 of 3 NP-NAION eyes also expanded their visual fields by at least 20 degrees. Color vision and relative afferent pupillary defect were also improved in 2 of 5 pro­gressive and 1 of 2 nonprogressive NAION eyes. They demonstrated the beneficial effects of ONSD for both P and NP- NAION among multiple param­eters of visual function. More recently, other studies described mixed re­sults of ONSD for NAION. Spoor et al reported that ONSD improved visual acuity and visual fields in P- NAION if standardized A- scan echog­raphy demonstrated increased perineural, sub­arachnoid fluid by the 30- degree test ( 12,14). How­ever, vision did not improve in P- NAION eyes without increased fluid or in NP- NAION eyes. In some patients, ischemia at the level of the lamina cribosa caused transudation and increased peri­neural fluid in the retrobulbar optic nerve. The in­creased fluid created optic nerve compression and additional damage to axons. The authors believed that ONSD can regain the level of vision that ex­isted immediately after the onset of ischemia and prior to the progressive visual loss. They did not find increased perineural fluid in patients with P- NAION who did not regain vision. Patients with NP- NAION did not have increased fluid and did not have improved vision. In contrast to the studies discussed above, Jab- Ions and colleagues ( 11) reported that ONSD did not improve vision more than what might be ex­pected without surgery in P- NAION. They pro­spectively evaluated 26 patients with P- NAION who had ONSD. Improvement was defined as an increase of two lines or more in Snellen acuity, a Goldmann isopter expansion of 20 degree, loss of a scotoma or increase in mean defect by 5 dB on Humphrey- automated perimetry. Only 7 of 26 ( 27%) eyes demonstrated postoperative improve­ment of visual function. G. S. Kosmorsky ( unpub­lished data, presented at the Upper Midwest Neuro- Ophthalmology Group Meeting, July 1992) found only 1 of 9 ( 11%) of NAION eyes regained 3 or more lines of visual acuity after ONSD. Al­though A. Epstein and R. Baker ( unpublished data, presented at the International Neuro- Ophthalmological Society Meeting, May 1992) re­ported that 13 of 24 NAION eyes " improved" after ONSD, this included patients who recovered even just 1 line of visual acuity. The disparity in re­ported postoperative improvement rates has heightened the controversy about the efficacy of / Neuw- Ophthalmol, Vol. 14, No. 2, 1994 OPTIC NERVE SHEATH DECOMPRESSION FOR NA10N 75 TABLE 3. Statistical analysis of visual improvement frequencies Groups P + NP Unmatched P Unmatched NP Matched P + NP Matched P Matched NP Surgery 7/ 18 5/ 9 2/ 9 7/ 18 5/ 9 2/ 9 No surgery 23/ 71 7/ 24 16/ 47 7/ 18 3/ 9 4/ 9 P values .398* .159* .362* .623** .344" .313" P, progressive nonarteritic ischemic optic neuropathy; NP, nonprogressive nonarteritic isch­emic optic neuropathy. * P value for right tail of Fisher's exact probability distribution for 2 by 2 tables. ** P value for sign test. ONSD for NAION. Sadun ( 15) recently summa­rized the controversy. In our retrospective analyses of 89 patients with NAION, 18 eyes underwent ONSD, and 71 eyes did not have surgery. Statistical comparisons of the improvement rates in visual acuity between the ONSD group ( 39%) and the NS group ( 32%) showed that the differences were not significant. Likewise, when the ONSD group was compared to a matched NS group, there were no significant dif­ferences in improvement rates. Even separating progressive eyes from nonprogressive eyes did not reveal a beneficial effect of ONSD over that of the nonsurgical course of NAION. The largest differ­ence between ONSD and NS groups was within eyes with P- NAION. For example, the frequency of increased visual acuity was 56% with ONSD and was 33% without surgery when the matched con­trol group was used. Nevertheless, the difference was not statistically significant. We recognize that this study is limited by the retrospective nature of its data, the small sample sizes, and the use of visual acuity as the only mea­sure of visual improvement. In addition, preoper­ative echography to show increased perineural fluid was not done. However, calculations of the statistical power of our analysis showed that our sample sizes were large enough to detect signifi­cant differences between treatment and control groups, if the true difference in rates of improve­ment in visual acuity was as large as reported in some published studies. Our calculations of statis­tical power were based on a mean improvement rate of 87% with surgery, as reported in three pub­lished studies of P- NAION ( 6- 8), and an improve­ment rate of 22% without surgery, as reported by Movsas and colleagues ( 5). If the true difference in improvement rate in P or NP- NAION eyes is less, the small sample sizes of our groups might have prevented us from detecting significant differences between ONSD and NS groups. We presented a version of this paper at the 129th Meeting of the American Ophthalmological Soci­ety ( 16). In a discussion of the paper, Dr. Robert Sergott ( 17) presented data about the timing of ONSD. Univariate logistic regression analysis ap­plied to data from previously published studies showed that ONSD performed 18 days or less from the onset of visual loss had a significantly greater chance of improving visual acuity than surgery performed later. Patients in our study had surgery 2 to 6 weeks ( mean 3 weeks) after the onset of NAION. Our findings did clearly show that the visual outcome of NAION without surgical intervention is also variable. The frequencies of changes in vi­sual acuity were remarkably similar in our study and that of Movsas and colleagues ( 5). The fre­quencies of improvement were 32% and 22%, and those of no change were 62% and 69% in our study and in the latter study, respectively. The frequency of further visual loss after initial presentation ( NP-NAION) or 3 weeks after presentation ( P- NAION) was 6% in our study, compared to 9% in the other study. Although the frequency of improvement was large, its amplitude was usually small or mod­est ( 2 to 3 lines). Several visual defects persisted, and most patients did not report having a func­tionally significant improvement in vision. More accurate delineation of the natural history of NAION and the effect of ONSD will depend on future randomized, controlled studies using large sample sizes, prospectively collected data and measurement of multiple variables. The National Eye Institute is sponsoring a multicenter, con­trolled, randomized clinical trial ( Ischemic Optic Neuropathy Decompression Trial) that should pro­vide some of this information. Acknowledgments: We thank William Nunery, M. D., and Ronald Martin, M. D., who performed surgery for some of our patients. We appreciate the assistance of Siu Hui, Ph. D., and Eleanor Boatwright, M. S., Division of Biostatistics, Department of Medicine, Indiana Univer- / Neuro- Ophthalmol, Vol. 14, No. 2, 1994 76 R. D. YEE ET AL. sity School of Medicine, in preparing the statistical anal­yses. REFERENCES 1. Boghen DR, Glaser JS. Ischemic optic neuropathy. The clin­ical profile and natural history. Brain 1975; 98: 689- 708. 2. Repka MX, Savino PJ, Schatz NJ, Sergott RC. Clinical pro­file and long term implications of anterior ischemic optic neuropathy. Am ] Ophthalmol 1983; 96: 478- 83. 3. Ellenberger C, Keltner JL, Burde RM. Acute optic neurop­athy in older patients. Arch Neurol 1973; 28: 182- 5. 4. Sawle GV, James CB, Ross Russell RW. The natural history of non- arteritic anterior ischemic optic neuropathy. / Neurol Neurosurg Psychol 1990; 53: 830- 3. 5. Movsas T, Kelman SE, Elman MJ, Miller NR, Dickersin K, Min Y. The natural course of non- arteritic ischemic optic neuropathy. Invest Ophthalmol Vis Sci ( Suppl) 1991; 32: 951. 6. Sergott RC, Cohen MS, Bosley TM, Savino PJ. Optic nerve decompression may improve the progressive form of non-arteritic ischemic optic neuropathy. Arch Ophthalmol 1989; 107: 1743- 54. 7. Spoor TC, Wilkinson MJ, Ramocki JM. Optic nerve sheath decompression for the treatment of progressive nonarteritic ischemic optic neuropathy. Am ] Ophthalmol 1991; 111: 724- 8. 8. Kelman SE, Elman MJ. Optic nerve sheath decompression for nonarteritic ischemic optic neuropathy improves multi­ple visual function measurements. Arch Ophthalmol 1991; 109: 667- 71. 9. Hayreh SS. The role of optic nerve sheath fenestration in the management of anterior ischemic optic neuropathy. Arch Ophthalmol 1990; 108: 1063- 4. 10. Wilson WB. Does optic nerve sheath decompression help progressive ischemic optic neuropathy? Arch Ophthalmol 1990; 108: 1065- 6. 11. Jablons MM, Glaser JS, Schatz NJ, Siatkowski RM, Tse DT, Kronish JW. Optic nerve sheath fenestration for treatment of progressive ischemic optic neuropathy. Arch Ophthalmol 1993; 111: 84- 87. 12. Spoor TC, McHenry JG, Lau- Sickon L. Progressive and static nonarteritic ischemic optic neuropathy treated by op­tic nerve sheath decompression. Ophthalmology 1993; 100: 306- 11. 13. Borchert M, Lessell S. Progressive and recurrent nonar­teritic anterior ischemic optic neuropathy. Am ) Ophthalmol 1988; 106: 443- 9. 14. McHenry JG, Spoor TC. The efficacy of optic nerve sheath decompression for anterior ischemic optic neuropathy and other optic neuropathies. Am ] Ophthalmol 1993; 116: 254- 5. 15. Sadun AA. The efficacy of optic nerve sheath decompres­sion for anterior ischemic optic neuropathy and other optic neuropathies. Am ] Ophthalmol 1993; 115: 384- 89. 16. Yee RD, Selky AK, Purvin VA. Outcomes of surgical and nonsurgical management of nonarteritic ischemic optic neuropathy. Trans Am Ophthalmol Soc 1993; 91: 227- 40. 17. Sergott RC. Outcomes of surgical and nonsurgical manage­ment of nonarteritic ischemic optic neuropathy. Trans Am Ophthalmol Soc 1993; 91: 240- 1. / Neuro- Ophthalmol, Vol. 14, No. 2, 1994