| OCR Text |
Show Journal of Neuro- Ophthalmology 15( 2): 61- 62, 1995. © 1995 Raven Press, Ltd., New York Editorial Optic Nerve Sheath Fenestration for Anterior Ischemic Optic Neuropathy? The Answer Is In Roy W. Beck, M . D . , Ph. D. The role of optic nerve sheath fenestration ( ONSF) in nonarteritic anterior ischemic optic neuropathy ( NAION) has been controversial since the initial report of Sergott et al. ( 1) purporting its effectiveness in stabilizing or improving visual function. Those authors reported that visual function improved in 12 of 14 patients with progressive visual loss from NAION who underwent ONSF compared with no visual improvement in 12 untreated controls ( nonrandomized, from same time period). Additional anecdotal reports of the proce-dures's success in NAION followed ( 2,3). Not only did the procedure become popularized but also its indication was broadened by many to include not just the progressive form of NAION as reported by Sergott et al. ( 1) but also the more common nonprogressive form. Recent reports questioning the benefit of surgery have not seemed to dampen the enthusiasm for the procedure ( 4,5). Despite the lack of a reasonable hypothesis as to why ONSF should be effective in NAION and the fact that its benefit had not yet been scientifically validated, the procedure gained widespread acceptance. This led to the development of the Ischemic Optic Neuropathy Decompression Trial ( IONDT) spearheaded by Shalom Kelman, Kay Dickersin, and others ( 6). Recruitment toward a goal of 300 patients began in October 1992. In addition to a clinical syndrome consistent with NAION, eligibility criteria included age 50 years or older, visual symptoms for less than 14 days from onset, and visual acuity of 20/ 64 or worse. Patients seen within the first 14 days with visual acuity better than 20/ 64 could be entered if acuity worsened to this level before 30 days from the onset of symptoms. Patients were Manuscript accepted March 21, 1995. From the Jaeb Center for Health Research, Tampa, Florida, U. S. A. Address correspondence to Dr. Roy W. Beck, Jaeb Center for Health Research, Inc., 3010 E. 138th Ave., Suite 9, Tampa, FL 33613, U. S. A. randomly assigned to be in either a surgery group or a control group. The primary outcome measure was a three or more line improvement in visual acuity after 6 months. Visual field mean deviation assessed with the Humphrey Field Analyzer program 24- 2 was a secondary outcome measure. Vision testers were masked to treatment group. When recruitment was halted in September 1994 at the recommendation of the Data and Safety Monitoring Committee, 119 patients had been randomized to the surgery group and 125 to the control group. Six- month follow- up vision testing had been completed by 92 and 89 patients in the two groups, respectively, and 3- month follow- up by 103 and 109, respectively. At 6 months, 32.6% of the surgery group compared with 42.7% of the control group had improved three or more lines of visual acuity; more patients in the surgery group than in the control group had worsened three or more lines of visual acuity: 23.9% vs. 12.4%. There was no indication of benefit when visual acuity data were analyzed as a continuous measure or with different definitions of improvement, nor was there any indication of benefit in the visual field data. The 3- month and 12- month data mirrored the 6- month data quite closely. The Data and Safety Monitoring Committee, of which I was a member, approved of the study design prior to the start of patient accrual and evaluated the data throughout the course of the trial. IONDT was a well- conceived and well- conducted trial. Losses to follow- up were minimal and the methodology followed to assure that the data collected would be valid was rigorous. There were no apparent sources of bias or confounding that might have masked a treatment effect. There was not even a glimmer of benefit in any subgrouping by patient or clinical factors, including the experience of the surgeon. At no point throughout the course of the study, did interim analyses even suggest that treatment might be effective. The decision to recommend early termination of the trial 61 62 R. W. BECK was reached only after the committee was convinced that the trial would not find surgery to be beneficial even in the remote circumstance of a dramatic shift in favor of surgery with additional recruitment and follow- up. One issue surely to be raised by critics is that only a portion of the IONDT's patients had progressive visual loss, the type of cases that formed the basis of Sergott et al.' s original report ( 1). However, assessment of treatment effect among those patients with progressive loss, regardless of whether progression was defined by the patient's report ( N = 99) or by documented worsening on examination ( N = 33), provided no indication whatsoever of a benefit to surgery. A true treatment benefit in patients with progressive loss that was not detected in the IONDT seems improbable. The IONDT provides clear and convincing evidence that ONSF is not effective and may be harmful in NAION. One difficulty with a surgery trial is that there will be surgeons who say " In my hands the procedure works." If such proponents of the procedure remain, the burden is now on them to provide data of comparable high quality to those collected in the IONDT. In retrospect, the suspicion that the reports of visual improvement after ONSF may not reflect a true treatment effect could have come from the published and unpublished observations that in some patients fellow eyes that had suffered NAION years earlier suddenly showed improvement after ONSF on the opposite eye ( 1,3). Beyond learning of the lack of benefit for ONSF in NAION, we have learned a great deal about the natural history of visual loss in NAION from the IONDT results. Once again one of our discipline's tenets- that NAION is a static condition- has been shown in a controlled study to be no more than neuro- ophthalmic mythology. An underestimation of the proportion of patients with NAION who spontaneously improve undoubtedly helped to fuel the enthusiasm for ONSF as an effective procedure. Of note, the report by Ellenberger et al. ( 7) more than 20 years ago of a 33% spontaneous improvement rate in NAION, which was never given much credence, provided data similar to that found in the IONDT. Recently, others have reported a similar spontaneous improvement rate as well ( 5,8). ONSF was first popularized for treatment of visual loss in patients with pseudotumor cerebri. Based principally on the results of uncontrolled patient series and clinical observations, in recent years the possible indications for the procedure have been extended to include not only NAION but also arteritic AION, central retinal vein occlusion, optic disc drusen, and low tension glaucoma. The value of the procedure in all of these conditions, even in pseudotumor cerebri, where there is at least a rationale for why the procedure might work, must now undergo increased scrutiny. Clinical trials have served as the foundation of much of the knowledge we have gained in the last 40 years in regard to the efficacy of treatments of various medical conditions. In very few conditions is the natural history of the disease so well defined and the outcome of intervention so evident that the benefits as well as the adverse effects of a treatment can be determined without a control group and randomization of treatment assignments. The National Eye Institute, which supported the IONDT, has been a leader among the National Institutes of Health in promoting the value of the multicenter controlled clinical trial. The IONDT's finding that ONSF was not only ineffective but was associated with an increased rate of poor outcomes underscores the enormous value and the indispensable role of the randomized clinical trial in assessing medical interventions. As Thomas Chalmers ( 9) has so aptly stated, " One has only to review the graveyard of discarded therapies to discover how many patients might have benefitted from being assigned to a control group." Such is the case for ONSF for NAION. Based on the IONDT results, it is time to stop treating NAION with ONSF. REFERENCES 1. 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. 2. Kelman SE, Elman MJ. Optic nerve sheath decompression for nonarteritic ischemic optic neuropathy improves multiple visual function measurements. Arch Ophthalmol 1991; 109: 667- 71. 3. Spoor TC, Wilkinson MJ, Ramocki JM. Optic nerve sheath decompression for the treatment of progressive nonarteritic ischemic optic neuropathy. Am J Ophthalmol 1991; 111: 724- 8. 4. Glaser JS, Teimory M, Schatz NJ. Optic nerve sheath fenestration for progressive ischemic optic neuropathy. Arch Ophthalmol 1994; 112: 1047- 50. 5. Yee RD, Selky AK, Purvin VA. Outcomes of optic nerve sheath decompression for nonarteritic ischemic optic neuropathy. / Ncuroophthalmol 1994; 14: 70- 6. 6. The 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: 525- 32. 7. Ellenberger C, Kelnter JL, Burde RM. Acute optic neuropathy in older patients. Arch Neurol 1973; 28: 182- 5. 8. Arnold AC, Hepler RS. Natural history of nonarteritic anterior ischemic optic neuropathy. /. Ncuroophthalmol 1994; 14: 66- 9. 9. Chalmers TC. Prophylactic treatment of Wilson's disease. N Engl f Med 1968; 278: 911- 2. / Ncuro- Ophlhalmol, Vol. 25, No. 2, 1995 [CLiondt] |
