- Email: [email protected]
Incidence of idiopathic full-thickness macular holes in fellow eyes
Incidence of Idiopathic Full-thickness Macular Holes in Fellow Eyes A 5-year Prospective Natural History Study Eric Ezra, FRCS, FRCOphth,I John A. Wells, MD,I Roger H. Gray, FRCS, FRCOphth, Frank M. P. Kinsella, FRCS, FRCOphth, I Gavin M. Orr, FRCS, FRCOphth, s John Grego, PhD, e Geoffrey B. Arden, PhD, ~ Zdenek J. Gregor, FRCS, FRCOphth I Purpose: This study aimed to determine the incidence of idiopathic full-thickness macular hole (FTMH) in normal fellow eyes and to evaluate the role of electrodiagnostic and psychophysical tests in identifying eyes at risk. Patients and Methods: A prospective longitudinal natural history study of a cohort of patients with unilateral holes and normal, asymptomatic fellow eyes without posterior vitreous detachment was conducted. Subjects underwent baseline examination, pattern reversal electroretinography, electro-oculography (EOG), and color contrast sensitivity (CCS) testing for protan, deutan, and tritan thresholds and were recalled for clinical examination at 18 months and 5 years. Results: At baseline, 114 patients were examined. Eighty were available for review at 18 months, of whom 6 had full-thickness macular holes develop in the fellow eye. At 5 years, 67 of the remaining 74 patients who had not developed holes at 18 months were re-examined and a further 5 were found to have holes develop in the fellow eye. A posterior vitreous detachment without hole formation had developed in 20 fellow eyes at 5 years. Although mean pattern reversal electroretinography and EOG responses were within normal limits in affected and fellow eyes at baseline, mean CCS protan, deutan, and tritan thresholds were elevated significantly in affected eyes at baseline (P = 0.0001). Unaffected fellow eyes showed normal mean protan and deutan thresholds, but significantly elevated mean tritan thresholds (P = 0.01) at baseline. Mean tritan CCS loss was, however, similar in fellow eyes in which holes later developed and in fellow eyes in which holes did not. Conclusions: The Kaplan-Meier estimated risk of fellow eye involvement is 15.6% (range, 8.4%-22.3%; P = 0.05) at 5 years. Although electrodiagnostic and psychophysical testing was not predictive of fellow eye involvement, tritan CCS loss at baseline, in apparently normal fellow eyes, may indicate subclinical foveal dysfunction, the nature of which is unclear. Ophthalmology 1998; 105:353-359 Idiopathic full-thickness macular hole (FTMH) is an important cause of central visual loss, with a prevalence of 3.3 per 1000, with approximately 70% of patients being female. ~ After the proposal by Gass 2 and Johnson and Gass 3 that tangential vitreofoveal traction may lead to the formation of idiopathic FTMH, interest has focused on the surgical treatment of FTMH. Although surgery for Originally received: January 21, 1997. Revision accepted: August 22, 1997. Vitreoretinal Unit, Moorfields Eye Hospital, London, England. 2 Department of Statistics, University of South Carolina, Columbia, South Carolina. 3 Department of Electrodiagnostics, Moorfields Eye Hospital, London, England. Presented in part as a poster at the Annual Meeting of the American Academy of Ophthalmology, October-November 1995. Supported by the Association for the Guide Dogs for thg Blind (Grant 301), and by the Stringer Bequest to the Special Trustees of Moorfields Eye Hospital (Grant 31 I). Address correspondence to Zdenek J. Gregor, FRCS, FRCOphth, Vitreoretinal Unit, Moorfields Eye Hospital, City Road, London ECIV 2PD, UK.
impending FTMH has not been shown to be preventative,4 several studies have confirmed that pars plana vitrectomy, cortical vitreous peeling, and long-acting gas tamponade, with or without local adjunctive substances, for FTMH may achieve both anatomic closure, in 58% to 100% of cases, and improved vision in many patients. 5-~4 Many subjects undergoing macular hole surgery, however, have unilateral involvement with asymptomatic and clinically normal fellow eyes, and the risks of surgery and anesthesia must be considered carefully. In addition, the rigorous posturing requirements in the postoperative period may present difficulties, especially for the elderly. An estimate of the risk of fellow eye involvement is, therefore, an important factor for both the surgeon and the patient in arriving at a decision regarding the benefits of surgery. Various studies, mostly retrospective with small sample sizes, have reported widely varying risks, between 0% and 29%, ~5-26 in the fellow eye. This study was undertaken to determine prospectively the incidence of FTMH in apparently normal asymptomatic fellow eyes, without posterior vitreous detachment (PVD), of patients with unilateral FTMH. The value of
353
Ophthalmology Volume 105, Number 2, February 1998 e l e c t r o d i a g n o s t i c testing, using pattern r e v e r s a l e l e c t r o r e t i n o g r a p h y ( P E R G ) and e l e c t r o - o c u l o g r a m ( E O G ) , and c o l o r contrast sensitivity ( C C S ) thresholds, in i d e n t i f y i n g at-risk f e l l o w eyes, b e f o r e the d e v e l o p m e n t o f an F T M H , also was assessed.
Patients and Methods Patients were enrolled into the study from the outpatient clinics at Moorfields Eye Hospital between January 1989 and December 1990. The study was approved by the Ethical Committee of Moorfields Hospital, and informed consent was obtained from each patient before entry into the study. Only patients with idiopathic FTMH in one eye and a normal asymptomatic fellow eye with normal visual acuity (20/30 or better), without a PVD, were included. Patients with retinal vascular disorders, such as diabetes, known color-vision defects, high myopia ( > 6 diopters of myopia), age-related macular degeneration, or history of ocular trauma or surgery in either eye were excluded. All patients meeting the eligibility criteria were enrolled into the study consecutively. At the time of enrollment, surgery had not yet been established as a viable form of treatment for FTMH. Examination at each timepoint included determination of the best-corrected Snellen visual acuity and slit-lamp biomicroscopy,, including an examination with a Goldmann fundus contact lens. The W a t z k e - A l l e n test was used in all patients to confirm the diagnosis of FTMH in the affected eye, and any patient reporting either a distortion or a break in the slit beam in the fellow eye was excluded from the study. The configuration of the posterior vitreous face was determined in all cases at baseline and re-evaluated at follow-up on the basis of slit-lamp biomicroscopy, including Goldmann fundus contact lens examination. A complete PVD was defined as complete separation of the posterior vitreous face with the presence of a Weis ring, and a partial PVD was defined as the presence of vitreous separation with residual attachment to the optic disc. Patients with complete or partial vitreous separation at baseline were excluded from the study. All patients underwent PERG testing for each eye at baseline. The technique for recording PERGs has been described elsewhere. 27 The PERG was recorded bilaterally using a goldfoil electrode and a reference electrode on the skin at the lateral canthus. The stimulus subtended 16° x 22 °, and the patternreversing checks were of 30-minute subtense and of near 100% contrast. Six reversals per second were used in each case. Approximately 250 sweeps were averaged for a response, and the mean of 3 such responses was used as the value. Both eyes were tested in each patient. The initial corneal positive response (P50) was measured, and a value of 1.5/.tV was taken as normal for this age range in our laboratory. 27 The EOG also was recorded for each eye, using the technique described previously, 2s for both eyes in each subject. The normal value for the light peak-dark trough (Arden index), expressed as a percentage, was taken as i 80% in our laboratory. Color contrast sensitivity was recorded for each eye of each patient at baseline by the method described by Arden et al.29 Briefly, a computer graphics peripheral controls the image on a high-quality color monitor. A program calculates the relative voltages required to produce any color (within the limits of the television phosphors) in terms of the Commission Internationale d'Eclairage 1931 diagram of color space. Hues were modulated along color confusion lines for trichromatic vision. Thus, protan, deutan, and tritan color axes were used. The program produces gratings (or other images) in which there is no spatial variation
354
of luminance, and changes in color contrast are the only clues to image recognition. To improve precision, individual variations from the Commission Internationale d'Eclairage standard observer are first measured. The patient makes a heterochromic flicker brightness match between red, green, and blue television guns. This match is compared with that of the normal observer. The output then is adjusted to produce equiluminous colors for the patient tested, compensating for lens changes or macular pigmentation. When color contrast was 0, all parts of the grating had the same hue as the background. A maximum contrast of 100% was defined arbitrarily as the most extreme colors in the color axis displayed. The screen target subtends a 5 ° solid angle centered on the fovea when viewed at 1.5 m. The results of CCS testing were compared with those of previously published normative data for the same age range? ° The K a p l a n - M e i e r estimator was calculated for baseline and the two follow-up timepoints to determine the cumulative incidence of FTMH in fellow eyes. Multivariate analyses of variance were performed to determine (1) whether visual acuity, PERG, EOG, and CCS (protan, deutan, tritan), for both affected first eye and unaffected fellow eye at baseline, were predictive of macular hole formation in fellow eyes; and (2) the difference in the same parameters between affected first eye and unaffected fellow eye at baseline. A P value of 0.05 was taken as representing statistical significance. The least squares means (means adjusted for the unbalanced model design) were used and will, hence, be quoted in the Results section.
Results A total of 114 patients were enrolled. The mean age of the cohort at baseline was 66 years (range, 4 2 - 1 0 0 years), and 72% were women and 28% were men. At baseline, there was a significant difference in the median best-corrected Snellen visual acuity mean visual acuity between affected eyes (median, 20/120) and normal fellow eyes (median, 20/20) (P = 0.0001). Eighty patients were available for follow-up at 18 months, at which time 6 fellow eyes had FTMH develop. At the 5-year follow-up, 67 of the remaining 74 patients, without fellow eye holes at the 18-month interval, were available for re-examination. At this time, a further five fellow eyes had holes develop. From the K a p l a n - M e i e r calculation, the incidence of fellow eye involvement was estimated to be 7.5% (range, 2 . 2 % - 12.5%; P = 0.05) at 18 months and 15.6% (range, 8.4%-22.3%; P = 0.05) at 5 years. Of the 11 fellow eyes in which FTMH had developed at 5 years, 2 had a stage 2 hole, 8 had a stage 3 hole, and 1 had a stage 4 hole. Of the 62 fellow eyes without holes that were examined at 5 years, 2 had a stage 1 foveolar detachment develop, 2 had a lamellar hole and a complete PVD develop, 1 had a foveal cyst develop, and 1 had an epiretinal membrane with pseudohole with a complete PVD develop. All six eyes retained visual acuities of 20/30 or better. During the course of the study, a total of 41 patients were lost to follow-up (34 at 18 months and a further 7 at 5 years), of whom 5 had died and the remaining 36 had either declined to return for examination or were not contactable. At 5 years, a complete PVD without FTMH was present in 18 of 62 fellow eyes without FTMH, and a partial PVD was present in a further 2 of 62 eyes. At 5 years, 10 of 11 fellow eyes in which an FTMH had developed still had an attached posterior vitreous (2 stage 2 and 8 stage 3). The remaining fellow eye with an FTMH had a PVD develop (one stage 4). In this eye, a stage 3 F T M H had been present at 18 months
Ezra et al • Five-year Incidence of FTMH in Fellow Eyes Table 1. Mean Values of Test at Baseline for Affected Eyes and Fellow Eyes Normal Value (age adjusted)
Objective Test Electro-oculogram (EOG) (%) Pattern electroretinogram (PERG) (P50)* (/iV) Color contrast sensitivity (CCS) (%) Protan Deutan Tritan
Affected Eye
Fellow Eye
P
180 1.5
197 1.50 (0.37)
203 1.63 (0.41)
0.86 0.82
7.5 7.5 8.5
23.0 (17.0) 26.7 (21.0) 42.2 (27.0)
8.97 (7.0) 15.4 (9.9) 11.2 (9.2)
0.0001 0.0001 0.0001
Vah, es in parentheses indicate standard deviation. A value of P = 0.05 or less was taken as indicating statistical significance. * P50 indicates the initial corneal positive response.
(i.e., in the presence of an attached vitreous). Thus, in no eye did an FTMH develop after the occurrence of a complete or partial PVD. The mean values of each test at baseline for both affected eyes and normal fellow eyes are summarized in Table 1. As can be seen, the mean EOG light p e a k - d a r k trough ratio (Arden index) expressed as a percentage and the P E R G - P 5 0 values were comparable for both groups and did not differ significantly from normal values. The CCS thresholds to protan, tritan, and deutan stimuli, however, were elevated significantly in affected eyes compared with those of normal fellow eyes at baseline. The thresholds for normal fellow eyes were not significantly different for protan and deutan stimuli, compared with those of normative data, but were elevated significantly for tritan (P = 0.01). No predictive value could be identified for any individual baseline test, whether performed on the affected eye or on the
normal fellow eye, in terms of identifying fellow eyes in which FTMH would develop at 5 years (Table 2). In addition, there was no difference in total baseline CCS score (protan + deutan + tritan) or the total baseline performance score (VA + EOG + PERG/P50 + protan + deutan + tritan) for both the affected eye (P = 0.73, P = 0.22) and the normal fellow eye (P = 0.98, P = 0.52)) between subjects in w h o m F T M H developed in the fellow eye and those who did not at 5 years. No correlation was present between baseline visual acuity of the affected eye and the development of FTMH in the fellow eye (P = 0.78). Other baseline characteristics, such as patient age and sex, were not predictive of F T M H formation in the fellow eye.
Discussion The role of vitreous traction in the underlying mechanism of idiopathic FTMH formation was proposed initially
Table 2. Predictive Value of Each Test in Identifying At-risk Fellow Eyes*
Baseline Test EOG (AE) EOG (FE) PERG/P50 (AE) PERG/P50 (FE) CCS/PRO (AE) CCS/PRO (FE) CCS/DEU (AE) CCS/DEU (FE) CCS/TRI (AE) CCS/TRI (FE)
Outcome Group at 5 yrs
LS Mean
Standard Error
FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH
216.8% 195.9% 212.0% 203.1% 14.70/iV 14.44/iV 17.60 #V 19.89/iV 19.39% 23.94% 7.42% 9.07% 30.31% 27.17% 10.53% 11.18% 50.5% 41.98% 17.64% 15.48%
18.25 6.02 16.61 5.48 2.01 0.63 2.23 0.70 5.55 1.74 2.22 0.70 6.98 2.18 2.92 O.91 8.88 2.78 3.16 0.99
Test for Predictive Value for FTMH in Fellow Eyes (P value) 0.28 0.61 0.90 0.33 0.44 0.48 0.67 0.83 0.36 0.52
LS Mean = least square mean value; FrI'/vlH = full-thickness macular hole. * At baseline, AE = affected first eyes, FE = fellow eyes. For 5 year outcome groups, FEH = fellow eyes which developed holes (n = 11), FENH = fellow eyes with no holes at 5 years (n = 62). Data for electro-oculography (EOG), pattern electroretinography (PERG) P50, and color contrast sensitivity (CCS) protan, deutan, and tritan (PRO, DEU, TRI) thresholds are summarized.
355
Ophthalmology
Volume 105, Number 2, February 1998
T a b l e 3. S u m m a r y of t h e M a j o r Studies Estimating t h e I n c i d e n c e of F T M H in Fellow Eyes
Study (year)
Study Design
No. of Fellow Eyes
Follow-up (mean yrs)
Incidence (%) of F T M H in Fellow Eyes
Aaberg et al (1970)
Retrospective
64
1.6
11
Bronstein et al ( 1981)
Retrospective
25
3.4
12
McDonnell et al (1982)
Retrospective
21
2.2
0
Trempe et al (1986)
Retrospective
49
3.9
James and Feman (1980)
Retrospective/prospective follow-up
22
2.0
9
Morgan and Schatz (1986)
Retrospective
64
4.3
4
Guyer et al (1992)
Retrospective
80
4.7
1.2
Fisher et al (1994)
Prospective
46
2.8
Lewis et al (1996)
Retrospective
340
4.0
Ezra et al (curl'ent study)
Prospective
114
5.0
16
(no PVD = 29, PVD+ = O)
2
(no PVD = 3.5, PVD+ = 0) 13 (no PVD = 14%, PVD+ = 2%) 15.6 (8.4-22.3, P = 0.05)
Comments 1. Fellow eyes without holes, no further differentiation 2. PVD stares not determined 1. Normal fellow eyes only 2. PVD status not determined 1. Fellow eyes "without hole or cyst"; heterogeneous group consisted of normals, drusen, RPE changes at fovea 2. PVD status not determined 1. Fellow eyes "without hole"; no differentiation between normals, cysts, RPE changes, etc. 2. PVD status determined in all eyes. 1. "Uninvolved fellow eyes"; no differentiation between normals, cysts, etc. 2. PVD status not defined in fellow eyes at start of followup I. Normal fellow eyes only 2. PVD status in normal fellow eyes not stated 1. Normal fellow eyes only 2. Presence of PVD not determined 1. Normal fellow eyes only 2. PVD status determined I. Normal fellow eyes only. 2. PVD status not determined in 111 fellow eyes. 1. Normal fellow eyes without PVD 2. PVD status determined
PVD =. posterior vitreous detachment; RPE = retinal pigment epithelium.
from retrospective studies that showed that the incidence was low in eyes with vitreous detachment. 2"3''7-19'23m-37 Other reports have shown that an attached posterior vitreous is associated with a worse visual prognosis 3s and a greater risk of hole enlargement, 39 lending further support to the theory of tangential vitreofoveal traction. Although surgical peeling of the posterior vitreous cortex with gas tamponade appears to be beneficial in most patients, complications such as postoperative cataract, 4° visual field defects,4~-43 and retinal detachment are not insignificant. 4,a The recent interest in the surgical management of this condition has therefore prompted increased scrutiny of the natural history and bilaterality of macular FTMH, so that the risks and benefits of surgery might be put into perspective. Although bilateral FTMH can lead to registerable blindness, the majority occur unilaterally jt-ts'2°'21.23'24"37 with many patients remaining asymptomatic until the fellow eye becomes affected. Clearly, an estimate of the overall risk to the fellow eye is important for patients considering surgery for unilateral holes. The majority of
356
previous studies have attempted to estimate the relative risk to the fellow eye using retrospective analysis js-2s (Table 3), and have reported a wide variation in fellow eye involvement of between 0% and 29%. The interpretation of these data has been complicated further because some studies have not differentiated normal fellow eyes from fellow eyes with predisposing lesions, such as macular cysts and impending holes, or have not determined the presence or ab-~ence of a PVD, making direct comparison between them difficult. In a retrospective review of 69 patients, Aaberg et al~7 found that 7% were bilateral on initial examination and a further 11% of those with unilateral holes went on to have holes develop in normal fellow eyes during a mean follow-up interval of 19 months. However, they did not differentiate fellow eyes with normal foveae from those with cysts or premacular hole lesions. In another study, Bronstein et al 2° reported a bilaterality of 7% at initial examination, with 12% of patients with normal fellow eyes having contralateral holes develop in a mean follow-up of 57 months. In contrast, McDonnell et al 2t found that only 2% were bilateral on
Ezra et al • Five-year Incidence of F T M H in Fellow Eyes
initial examination, and of 36 patients with normal fellow eyes initially, none of the 21 that were available for follow-up had holes develop in the fellow eye (mean followup, 27 months). In neither study, however, was the presence or absence of a PVD in normal fellow eyes reported and correlated to the development of an FTMH. In another study of 49 fellow eyes, Trempe et a123 found a bilaterality of 3% at presentation. Subsequently, over an average of 47 months' follow-up, 16% had holes develop, all of which had no PVD at presentation. The incidence in eyes with a PVD was 0 (0%) of 21 compared with that of 8 (29%) of 28 eyes without a PVD. However, in this study, no differentiation was made between normal foveae and those with predisposing lesions. The study showed that the presence of an attached vitreous was a significant risk factor in fellow eyes, but it did not determine the incidence of holes in normal fellow eyes. Lewis et al, 24 in a retrospective review of 340 patients with an FTMH in one eye and a normal fellow eye, showed a cumulative incidence of fellow eye FTMH of 13% at 4 years using survival analysis. However, in 111 patients, the status of the vitreous had not been recorded at the initial visit. In contrast to these reports, which are retrospective, the current study has reported prospective data on a cohort of consecutive patients with normal fellow eyes and an attached posterior vitreous at presentation and represents the largest prospective series to date. The best estimate, as determined by the K a p l a n - M e i e r method, for fellow eye FTMH formation was 7.5% (range, 2.2%-12.5%; P = 0.05) at 18 months and 15.6% (range, 8.4%-22.3%; P = 0.05) at 5 years and represents a significant risk to the fellow eye. Although this estimate is comparable to that of Lewis et a124 and some other studies cited above, it is somewhat higher than the 3.5% at 2.8 years reported by Fisher and associates, -'6 who reported prospectively 46 normal fellow eyes of patients with unilateral holes. Of these, 16 had a PVD, of which none had holes or premacular hole lesions develop, and 30 eyes had an attached posterior vitreous. If the 16 eyes with a PVD are excluded, 1 (3.5%) of 30 had FTMH develop and 2 (6%) of 30 had stage 1 lesions develop. Thirty-three percent of fellow eyes with an initially attached posterior vitreous subsequently had a PVD without FTMH develop. However, a comparison between the two studies is difficult, because Fisher and associates did not specify the mean follow-up period in the subgroup of normal fellow eyes without PVD. In addition, the series was not strictly consecutive. A number of patients were excluded because of lack of follow-up, and no K a p l a n - M e i e r calculation was available. If some of these patients subsequently had impending holes or FTMH develop and were treated elsewhere, the incidence would be artificially low. Furthermore, another 2 of 30 eyes had impending holes develop. If the follow-up period, which is not specified, for these two eyes was short, one or both subsequently may have had an FTMH develop. For these reasons, because of the small sample size of their cohort and the lack of more detailed stratification of follow-up data, a direct compari-
son between the study of Fisher and associates and our study is difficult. The observation in our study that in no case did an FTMH develop after a PVD is consistent with the findings of Fisher and associates and lends further support to the theory of vitreoretinai traction as the cause of FTMH. The objective electrodiagnostic and psychophysical tests used in this study were unable to identify fellow eyes at higher risk of having an FTMH develop. The normality of the EOG findings in both affected eyes and normal fellow eyes at baseline was consistent with the focal nature of the hole and its affects on the retina rather than the pigment epithelium and choroid. The normal pattern electroretinogram (PERG) P50 in affected and normal fellow eyes, at baseline, is consistent with an earlier study, 45 probably reflects the fact that the PERG is generated achromatically by large ganglion cells, 27 and suggests that these cells are largely intact and functional in eyes with FTMH. This also would be consistent with the focal nature of the disease (the checkerboard subtends a 17° solid cone at the fovea, which is far larger than the area of even the largest holes), whereby macular function remains largely intact around the hole, and would support the recent suggestion that in most cases, FFMH arises from a foveal dehiscence without loss of retinal tissue. 46 The preservation of the PERG is in contrast to most other macular diseases that affect a greater proportion of the macula in which the PERG declines as visual acuity becomes impaired. 47 A previous study using the focal ERG (FERG), with a white flicker stimulus stimulating a 3 ° area of fovea, has shown significantly reduced FERGs in eyes with existing FTMH, as well as in fellow eyes in which FTMH later developed. 48 The authors concluded that the reduced FERGs could result from decreased optical density because of shortened or misaligned cone outer segments, again supporting the view that this could arise from tangential vitreomacular traction. Although the authors suggested that FERG may have a role in predicting which fellow eyes were at higher risk, no further data have been forthcoming. The use of CCS testing in our study showed markedly elevated protan, deutan, and tritan thresholds in affected eyes at baseline. This is perhaps not unexpected, as the stimulus subtended a 5* cone at the fovea and represents a focal stimulus of similar size to the fovea itself. The loss of CCS probably reflects reduced foveal function due to either the loss of significant numbers of cones in the operculum or to the separation of foveal cones from the retinal pigment epithelium around the hole, without necessarily any avulsion of cones. Decreased foveal function without necessarily a loss of retinal tissue would be consistent with these results. More surprising, perhaps, is the elevation of thresholds to tritan stimuli at baseline in apparently normal fellow eyes. Because the tritan threshold testing involved a correction for blue wavelength absorption by lens pigment or xanthophyll between individuals, it is likely that the increase in threshold reflects a genuine color dysfunction. In apparently normal fellow eyes, this may be a reflection
357
Ophthalmology Volume 105, Number 2, February 1998 of incipient or subclinical traction on the fovea, causing subclinical receptoral or photoreceptor dysfunction. The lack of protan and deutan CCS abnormality in fellow eyes may reflect the fact that tritan function is the most susceptible in the early stages of vitreofoveal traction. In fact, isolated tritan CCS loss has been shown in subjects with ocular hypertension and early glaucoma, 49 and it has been speculated that postreceptorai pathways subserving tritan function may be the most susceptible to early damage in this condition. 49"5° The normality of the PERG in fellow eyes suggests that tritan dysfunction probably represents dysfunction of the photoreceptors rather than postreceptoral level. The reason for susceptibility blue cones, per se, however, remains unclear. If subclinical vitreomacular traction is the underlying cause for tritan loss, the lack of correlation between tritan loss and FTMH formation in the fellow eye suggests that the severity of traction may not be the sole determining factor for hole formation. In summary, this study has shown an estimated 5-year incidence of macular hole formation in fellow eyes of 15.6% and that a significant number of patients will have a PVD without hole formation develop. The study also emphasizes the importance of determining the state of the vitreous in normal fellow eyes so that an accurate risk prediction may be given to the patient. The relatively high incidence of PVD and the finding that CCS tritan thresholds are elevated in normal fellow eyes support the theory that an abnormality at the vitreoretinal interface exists and that subclinical foveal dysfunction or cone misalignment may exist even in apparently normal fellow eyes. Acknowledgment. The authors thank Dr. C. Hogg for technical support.
9. 10.
11. 12. 13.
14.
15. 16. 17. 18. 19. 20. 21. 22.
References I. McDonnell PJ, Fine SL, Hillis AI. Clinical features of idiopathic macular cysts and holes. Am J Ophthalmol 1982;93: 777-86. 2. Gass JD. Idiopathic senile macular hole. Its early stages arid pathogenesis. Arch Ophthalmol 1988; 106:629-39. 3. Johnson RN, Gass JD. Idiopathic macular holes. Observations, stages of formation, and implications for surgical intervention. Ophthalmology 1988;95:917-24. 4. De Bustros S. Vitrectomy for prevention of macular holes. Results of a randomized multicenter clinical trial. Ophthalmology 1994; 101:1055-60. 5. Kelly NE, Wendel RT. Vitreous surgery for idiopathic macular holes. Results of a pilot study. Arch Ophthalmol 1991; 190:654-9. 6. Wendel RT, Patel AC, Kelly NE, et al. Vitreous surgery for macular holes. Ophthalmology 1993;100:1671-6. Comment in: Ophthalmology 1993; 100:1607-8. 7. Glaser BM, Michels RG, Kuppermann BD, et al. Transforming growth factor-/32 for the treatment of full-thickness macular holes. A prospective randomized study. Ophthalmology 1992;99:1162-73. 8. Lansing MB, Glaser BM, Liss H, et al. The effect of pars plana vitrectomy and transforming growth factor-beta 2
358
23. 24. 25. 26.
27.
28. 29. 30.
without epiretinal membrane peeling on full-thickness macular holes. Ophthalmology 1993; 100:868-72. Wells JA, Gregor ZJ. Surgical treatnlent of full-thickness macular holes using autologous serum. Eye 1996; 10:593-9. Smiddy WE, Glaser BM, Thompson JT, et al. Transforming growth factor-/32 significantly enhances the ability to flatten the rim of subretinal fluid surrounding macular holes: preliminary anatomic results of a multicenter prospective randomized study. Retina 1993; 13:296-301. Ryan EH, Jr, Gilbert HD. Results of surgical treatment of recent-onset full-thickness idiopathic macular holes. Arch Ophthalmol 1994; 112:1545-53. Liggett PE, Skolik S, Horio B, et al. Human autologous serum for the treatment of full-thickness macular holes: a preliminary study. Ophthalmology 1995; 102:1071-6. Gaudric A, Massin P, Paques M, et al. Autologous platelet concentrate for the treatment of full-thickness macular holes. Graefes Arch Clin Exp Ophthalmol 1995;233:54954. Freeman WR, Azen SP, Kim JW, et al. Vitrectomy for the treatment of full-thickness stage 3 or 4 macular holes. Resuits of a multicentered randomized clinical trial. Arch Ophthalmol 1997; 115:11-21. Croll LJ, Croll M. Hole in the macula. Am J Ophthalmol 1950; 33:248-53. Yaoeda H. Clinical observation on the macular hole. Acta Soc Ophthal Jpn 1967;71:1723-36. Aaberg TM, Blair CJ, Gass JD. Macular holes. Am J Ophthalmol 1970;69:555-62. Aaberg TM. Macular holes: a review. Surv Ophthalmol 1970; 15:139-62. James M, Feman SS. Macular holes. Albrecht Von Graefes Arch Klin Exp Ophthalmol 1980;215:59-63. Bronstein MA, Trempe CL, Freeman HM. Fellow eyes of eyes with macular holes. Am J Ophthalmol 1981;92:75761. McDonnell PJ, Fine SL, Hillis AI. Clinical features of idiopathic macular cysts and holes. Am J Ophthalmol 1982; 93:777-86. Morgan CM, Schatz H. Idiopathic macular holes. Am J Ophthalmol 1985;99:437-44. Trempe CL, Weiter JJ, Furukawa H. Fellow eyes in cases of macular hole. Biomicroscopic study of the vitreous. Arch Ophthalmol 1986; 104:93-5. Lewis ML, Cohen SM, Smiddy WE, Gass JD. Bilaterality of idiopathic macular holes. Graefes Arch Clin Exp Ophthalmol 1996;234:241-5. Guyer DR, de Bustros S, Diener-West M, Fine SL. Observations on patients with idiopathic macular holes and cysts. Arch Ophthalmol 1992; 110:1264-8. Fisher YL, Slakter JS, Yannuzzi LA, Guyer DR. A prospective naturalhistory study and kinetic ultrasound evaluation of idiopathic macular holes. Ophthalmology 1994; 101:511. Arden GB, Vaegan, Hogg CR. Clinical and experimental evidence that the pattern electroretinogram (PERG) is generated in more proximal retinal layers than the focal electroretinogram (FERG). Ann N Y Acad Sci 1982;388:580601. Arden GB, Barrada A, Kelsey JH. New clinical test of retinal function based upon the standing potential of the eye. Br J Ophthalmol 1962;46:449-67. Arden GB, GUndtiz K, Perry S. Colour vision testing with a computer graphics system. Clinical Vision Sciences 1988; 2:303 -20. Arden GB, Berninger T, Hogg CR, Perry S. A survey of
Ezra et al • F i v e - y e a r I n c i d e n c e o f F T M H in F e l l o w Eyes
31. 32. 33. 34. 35. 36. 37. 38. 39. 40.
color discrimination in German ophthalmologists. Changes associated with the use of lasers and operating microscopes. Ophthalmology 1991 ; 98:567-75. Schepens CL. Fundus changes caused by alterations of the vitreous body. Am J Ophthalmol 1955;39:631-3. Maumenee AE. Further advances in the study of the macula. Arch Ophthalmol 1967; 78:151-65. Yoshioka H. Clinical studies on the macular holes. 3. On the pathogenesis of the senile macular hole. Nippon Ganka Gakka: Zasohi 1968;72:575-84. Reese AB, Jones IS, Cooper WC. Macular changes secondary to vitreous traction. Am J Ophthalmol 1967;64(Suppl): 544-9. Margherio RR, Schepens CL. Macular breaks: 1. Diagnosis, etiology and observations. Am J Ophthalmol 1972;74:21932. Tolentino FI, Schepens CL, Freeman HM. Vitreoretinal disorders: diagnosis and management. Philadelphia: Saunders, 1976;405. Avila MP, Jalkh AE, Murakami K, et al. Biomicroscopic study of the vitreous in macular breaks. Ophthalmology 1983; 90:1277- 83. Hikichi T, Akiba J, Trempe CL. Effect of the vitreous on the prognosis of full-thickness idiopathic macular hole. Am J Ophthalmol 1993; 116:273-8. Hikichi T, Trempe CL. Risk of decreased visual acuity in full-thickness: idiopathic macular-holes. Am J Ophthalmol 1993; 116:708-12. Thompson JT, Glaser BM, Sjaarda RN, Murphy RP. Progression of nuclear sclerosis and long-term visual results of vitrectomy with transforming growth factor beta-2 for macular holes. Am J Ophthalmol 1995; 199:48-54.
41. Melberg NS, Thomas MA. Visual field loss after pars plana vitrectomy with air/fluid exchange. Am J Ophthalmol 1995; 120:386-8. 42. Ezra E, Arden GB, Riordan-Eva P, et al. Visual field loss following vitrectomy for stage 2 and 3 macular holes. Br J Ophthalmol 1996; 80:519-25. 43. Hutton WL, Fuller DG, Snyder WB, et al. Visual field defects after macular hole surgery. A new finding. Ophthalmology 1996; 103:2152-8. 44. Fine SL. Vitreous surgery for macular hole in perspective. Is there an indication? [editorial]. Arch Ophthalmol 1991; 109:635-6. 45. Smith RG, Brimlow GM, Lea SJ, Galloway NR. Evoked responses in patients with macular holes. Doc Ophthalmol 1990;75:135-44. 46. Gass JD. Reappraisal of biomicroscopic classification of stages of development of a macular hole. Am J Ophthalmol 1995; 119:752-9. 47. Arden GB, Carter RM, Macfarlan A. Pattern and Ganzfeld electroretinograms in macular disease. Br J Ophthalmol 1984; 68:878-84. 48. Birch DG, Jost BF, Fish GE. The focal electroretinogram in fellow eyes of patients with idiopathic macular holes. Arch Ophthalmol 1988; 106:1558-63. 49. Gtindtiz K, Arden GB, Perry S, et al. Color vision defects in ocular hypertension and glaucoma. Quantification with a computer-driven color television system. Arch Ophthalmol 1988; 106:929-35. 50. Weinstein GW, Arden GB, Hitchings RA, et al. The pattern electroretinogram (PERG) in ocular hypertension and glaucoma. Arch Ophthalmol 1988; 106:923-8.
359
impending FTMH has not been shown to be preventative,4 several studies have confirmed that pars plana vitrectomy, cortical vitreous peeling, and long-acting gas tamponade, with or without local adjunctive substances, for FTMH may achieve both anatomic closure, in 58% to 100% of cases, and improved vision in many patients. 5-~4 Many subjects undergoing macular hole surgery, however, have unilateral involvement with asymptomatic and clinically normal fellow eyes, and the risks of surgery and anesthesia must be considered carefully. In addition, the rigorous posturing requirements in the postoperative period may present difficulties, especially for the elderly. An estimate of the risk of fellow eye involvement is, therefore, an important factor for both the surgeon and the patient in arriving at a decision regarding the benefits of surgery. Various studies, mostly retrospective with small sample sizes, have reported widely varying risks, between 0% and 29%, ~5-26 in the fellow eye. This study was undertaken to determine prospectively the incidence of FTMH in apparently normal asymptomatic fellow eyes, without posterior vitreous detachment (PVD), of patients with unilateral FTMH. The value of
353
Ophthalmology Volume 105, Number 2, February 1998 e l e c t r o d i a g n o s t i c testing, using pattern r e v e r s a l e l e c t r o r e t i n o g r a p h y ( P E R G ) and e l e c t r o - o c u l o g r a m ( E O G ) , and c o l o r contrast sensitivity ( C C S ) thresholds, in i d e n t i f y i n g at-risk f e l l o w eyes, b e f o r e the d e v e l o p m e n t o f an F T M H , also was assessed.
Patients and Methods Patients were enrolled into the study from the outpatient clinics at Moorfields Eye Hospital between January 1989 and December 1990. The study was approved by the Ethical Committee of Moorfields Hospital, and informed consent was obtained from each patient before entry into the study. Only patients with idiopathic FTMH in one eye and a normal asymptomatic fellow eye with normal visual acuity (20/30 or better), without a PVD, were included. Patients with retinal vascular disorders, such as diabetes, known color-vision defects, high myopia ( > 6 diopters of myopia), age-related macular degeneration, or history of ocular trauma or surgery in either eye were excluded. All patients meeting the eligibility criteria were enrolled into the study consecutively. At the time of enrollment, surgery had not yet been established as a viable form of treatment for FTMH. Examination at each timepoint included determination of the best-corrected Snellen visual acuity and slit-lamp biomicroscopy,, including an examination with a Goldmann fundus contact lens. The W a t z k e - A l l e n test was used in all patients to confirm the diagnosis of FTMH in the affected eye, and any patient reporting either a distortion or a break in the slit beam in the fellow eye was excluded from the study. The configuration of the posterior vitreous face was determined in all cases at baseline and re-evaluated at follow-up on the basis of slit-lamp biomicroscopy, including Goldmann fundus contact lens examination. A complete PVD was defined as complete separation of the posterior vitreous face with the presence of a Weis ring, and a partial PVD was defined as the presence of vitreous separation with residual attachment to the optic disc. Patients with complete or partial vitreous separation at baseline were excluded from the study. All patients underwent PERG testing for each eye at baseline. The technique for recording PERGs has been described elsewhere. 27 The PERG was recorded bilaterally using a goldfoil electrode and a reference electrode on the skin at the lateral canthus. The stimulus subtended 16° x 22 °, and the patternreversing checks were of 30-minute subtense and of near 100% contrast. Six reversals per second were used in each case. Approximately 250 sweeps were averaged for a response, and the mean of 3 such responses was used as the value. Both eyes were tested in each patient. The initial corneal positive response (P50) was measured, and a value of 1.5/.tV was taken as normal for this age range in our laboratory. 27 The EOG also was recorded for each eye, using the technique described previously, 2s for both eyes in each subject. The normal value for the light peak-dark trough (Arden index), expressed as a percentage, was taken as i 80% in our laboratory. Color contrast sensitivity was recorded for each eye of each patient at baseline by the method described by Arden et al.29 Briefly, a computer graphics peripheral controls the image on a high-quality color monitor. A program calculates the relative voltages required to produce any color (within the limits of the television phosphors) in terms of the Commission Internationale d'Eclairage 1931 diagram of color space. Hues were modulated along color confusion lines for trichromatic vision. Thus, protan, deutan, and tritan color axes were used. The program produces gratings (or other images) in which there is no spatial variation
354
of luminance, and changes in color contrast are the only clues to image recognition. To improve precision, individual variations from the Commission Internationale d'Eclairage standard observer are first measured. The patient makes a heterochromic flicker brightness match between red, green, and blue television guns. This match is compared with that of the normal observer. The output then is adjusted to produce equiluminous colors for the patient tested, compensating for lens changes or macular pigmentation. When color contrast was 0, all parts of the grating had the same hue as the background. A maximum contrast of 100% was defined arbitrarily as the most extreme colors in the color axis displayed. The screen target subtends a 5 ° solid angle centered on the fovea when viewed at 1.5 m. The results of CCS testing were compared with those of previously published normative data for the same age range? ° The K a p l a n - M e i e r estimator was calculated for baseline and the two follow-up timepoints to determine the cumulative incidence of FTMH in fellow eyes. Multivariate analyses of variance were performed to determine (1) whether visual acuity, PERG, EOG, and CCS (protan, deutan, tritan), for both affected first eye and unaffected fellow eye at baseline, were predictive of macular hole formation in fellow eyes; and (2) the difference in the same parameters between affected first eye and unaffected fellow eye at baseline. A P value of 0.05 was taken as representing statistical significance. The least squares means (means adjusted for the unbalanced model design) were used and will, hence, be quoted in the Results section.
Results A total of 114 patients were enrolled. The mean age of the cohort at baseline was 66 years (range, 4 2 - 1 0 0 years), and 72% were women and 28% were men. At baseline, there was a significant difference in the median best-corrected Snellen visual acuity mean visual acuity between affected eyes (median, 20/120) and normal fellow eyes (median, 20/20) (P = 0.0001). Eighty patients were available for follow-up at 18 months, at which time 6 fellow eyes had FTMH develop. At the 5-year follow-up, 67 of the remaining 74 patients, without fellow eye holes at the 18-month interval, were available for re-examination. At this time, a further five fellow eyes had holes develop. From the K a p l a n - M e i e r calculation, the incidence of fellow eye involvement was estimated to be 7.5% (range, 2 . 2 % - 12.5%; P = 0.05) at 18 months and 15.6% (range, 8.4%-22.3%; P = 0.05) at 5 years. Of the 11 fellow eyes in which FTMH had developed at 5 years, 2 had a stage 2 hole, 8 had a stage 3 hole, and 1 had a stage 4 hole. Of the 62 fellow eyes without holes that were examined at 5 years, 2 had a stage 1 foveolar detachment develop, 2 had a lamellar hole and a complete PVD develop, 1 had a foveal cyst develop, and 1 had an epiretinal membrane with pseudohole with a complete PVD develop. All six eyes retained visual acuities of 20/30 or better. During the course of the study, a total of 41 patients were lost to follow-up (34 at 18 months and a further 7 at 5 years), of whom 5 had died and the remaining 36 had either declined to return for examination or were not contactable. At 5 years, a complete PVD without FTMH was present in 18 of 62 fellow eyes without FTMH, and a partial PVD was present in a further 2 of 62 eyes. At 5 years, 10 of 11 fellow eyes in which an FTMH had developed still had an attached posterior vitreous (2 stage 2 and 8 stage 3). The remaining fellow eye with an FTMH had a PVD develop (one stage 4). In this eye, a stage 3 F T M H had been present at 18 months
Ezra et al • Five-year Incidence of FTMH in Fellow Eyes Table 1. Mean Values of Test at Baseline for Affected Eyes and Fellow Eyes Normal Value (age adjusted)
Objective Test Electro-oculogram (EOG) (%) Pattern electroretinogram (PERG) (P50)* (/iV) Color contrast sensitivity (CCS) (%) Protan Deutan Tritan
Affected Eye
Fellow Eye
P
180 1.5
197 1.50 (0.37)
203 1.63 (0.41)
0.86 0.82
7.5 7.5 8.5
23.0 (17.0) 26.7 (21.0) 42.2 (27.0)
8.97 (7.0) 15.4 (9.9) 11.2 (9.2)
0.0001 0.0001 0.0001
Vah, es in parentheses indicate standard deviation. A value of P = 0.05 or less was taken as indicating statistical significance. * P50 indicates the initial corneal positive response.
(i.e., in the presence of an attached vitreous). Thus, in no eye did an FTMH develop after the occurrence of a complete or partial PVD. The mean values of each test at baseline for both affected eyes and normal fellow eyes are summarized in Table 1. As can be seen, the mean EOG light p e a k - d a r k trough ratio (Arden index) expressed as a percentage and the P E R G - P 5 0 values were comparable for both groups and did not differ significantly from normal values. The CCS thresholds to protan, tritan, and deutan stimuli, however, were elevated significantly in affected eyes compared with those of normal fellow eyes at baseline. The thresholds for normal fellow eyes were not significantly different for protan and deutan stimuli, compared with those of normative data, but were elevated significantly for tritan (P = 0.01). No predictive value could be identified for any individual baseline test, whether performed on the affected eye or on the
normal fellow eye, in terms of identifying fellow eyes in which FTMH would develop at 5 years (Table 2). In addition, there was no difference in total baseline CCS score (protan + deutan + tritan) or the total baseline performance score (VA + EOG + PERG/P50 + protan + deutan + tritan) for both the affected eye (P = 0.73, P = 0.22) and the normal fellow eye (P = 0.98, P = 0.52)) between subjects in w h o m F T M H developed in the fellow eye and those who did not at 5 years. No correlation was present between baseline visual acuity of the affected eye and the development of FTMH in the fellow eye (P = 0.78). Other baseline characteristics, such as patient age and sex, were not predictive of F T M H formation in the fellow eye.
Discussion The role of vitreous traction in the underlying mechanism of idiopathic FTMH formation was proposed initially
Table 2. Predictive Value of Each Test in Identifying At-risk Fellow Eyes*
Baseline Test EOG (AE) EOG (FE) PERG/P50 (AE) PERG/P50 (FE) CCS/PRO (AE) CCS/PRO (FE) CCS/DEU (AE) CCS/DEU (FE) CCS/TRI (AE) CCS/TRI (FE)
Outcome Group at 5 yrs
LS Mean
Standard Error
FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH FEH FENH
216.8% 195.9% 212.0% 203.1% 14.70/iV 14.44/iV 17.60 #V 19.89/iV 19.39% 23.94% 7.42% 9.07% 30.31% 27.17% 10.53% 11.18% 50.5% 41.98% 17.64% 15.48%
18.25 6.02 16.61 5.48 2.01 0.63 2.23 0.70 5.55 1.74 2.22 0.70 6.98 2.18 2.92 O.91 8.88 2.78 3.16 0.99
Test for Predictive Value for FTMH in Fellow Eyes (P value) 0.28 0.61 0.90 0.33 0.44 0.48 0.67 0.83 0.36 0.52
LS Mean = least square mean value; FrI'/vlH = full-thickness macular hole. * At baseline, AE = affected first eyes, FE = fellow eyes. For 5 year outcome groups, FEH = fellow eyes which developed holes (n = 11), FENH = fellow eyes with no holes at 5 years (n = 62). Data for electro-oculography (EOG), pattern electroretinography (PERG) P50, and color contrast sensitivity (CCS) protan, deutan, and tritan (PRO, DEU, TRI) thresholds are summarized.
355
Ophthalmology
Volume 105, Number 2, February 1998
T a b l e 3. S u m m a r y of t h e M a j o r Studies Estimating t h e I n c i d e n c e of F T M H in Fellow Eyes
Study (year)
Study Design
No. of Fellow Eyes
Follow-up (mean yrs)
Incidence (%) of F T M H in Fellow Eyes
Aaberg et al (1970)
Retrospective
64
1.6
11
Bronstein et al ( 1981)
Retrospective
25
3.4
12
McDonnell et al (1982)
Retrospective
21
2.2
0
Trempe et al (1986)
Retrospective
49
3.9
James and Feman (1980)
Retrospective/prospective follow-up
22
2.0
9
Morgan and Schatz (1986)
Retrospective
64
4.3
4
Guyer et al (1992)
Retrospective
80
4.7
1.2
Fisher et al (1994)
Prospective
46
2.8
Lewis et al (1996)
Retrospective
340
4.0
Ezra et al (curl'ent study)
Prospective
114
5.0
16
(no PVD = 29, PVD+ = O)
2
(no PVD = 3.5, PVD+ = 0) 13 (no PVD = 14%, PVD+ = 2%) 15.6 (8.4-22.3, P = 0.05)
Comments 1. Fellow eyes without holes, no further differentiation 2. PVD stares not determined 1. Normal fellow eyes only 2. PVD status not determined 1. Fellow eyes "without hole or cyst"; heterogeneous group consisted of normals, drusen, RPE changes at fovea 2. PVD status not determined 1. Fellow eyes "without hole"; no differentiation between normals, cysts, RPE changes, etc. 2. PVD status determined in all eyes. 1. "Uninvolved fellow eyes"; no differentiation between normals, cysts, etc. 2. PVD status not defined in fellow eyes at start of followup I. Normal fellow eyes only 2. PVD status in normal fellow eyes not stated 1. Normal fellow eyes only 2. Presence of PVD not determined 1. Normal fellow eyes only 2. PVD status determined I. Normal fellow eyes only. 2. PVD status not determined in 111 fellow eyes. 1. Normal fellow eyes without PVD 2. PVD status determined
PVD =. posterior vitreous detachment; RPE = retinal pigment epithelium.
from retrospective studies that showed that the incidence was low in eyes with vitreous detachment. 2"3''7-19'23m-37 Other reports have shown that an attached posterior vitreous is associated with a worse visual prognosis 3s and a greater risk of hole enlargement, 39 lending further support to the theory of tangential vitreofoveal traction. Although surgical peeling of the posterior vitreous cortex with gas tamponade appears to be beneficial in most patients, complications such as postoperative cataract, 4° visual field defects,4~-43 and retinal detachment are not insignificant. 4,a The recent interest in the surgical management of this condition has therefore prompted increased scrutiny of the natural history and bilaterality of macular FTMH, so that the risks and benefits of surgery might be put into perspective. Although bilateral FTMH can lead to registerable blindness, the majority occur unilaterally jt-ts'2°'21.23'24"37 with many patients remaining asymptomatic until the fellow eye becomes affected. Clearly, an estimate of the overall risk to the fellow eye is important for patients considering surgery for unilateral holes. The majority of
356
previous studies have attempted to estimate the relative risk to the fellow eye using retrospective analysis js-2s (Table 3), and have reported a wide variation in fellow eye involvement of between 0% and 29%. The interpretation of these data has been complicated further because some studies have not differentiated normal fellow eyes from fellow eyes with predisposing lesions, such as macular cysts and impending holes, or have not determined the presence or ab-~ence of a PVD, making direct comparison between them difficult. In a retrospective review of 69 patients, Aaberg et al~7 found that 7% were bilateral on initial examination and a further 11% of those with unilateral holes went on to have holes develop in normal fellow eyes during a mean follow-up interval of 19 months. However, they did not differentiate fellow eyes with normal foveae from those with cysts or premacular hole lesions. In another study, Bronstein et al 2° reported a bilaterality of 7% at initial examination, with 12% of patients with normal fellow eyes having contralateral holes develop in a mean follow-up of 57 months. In contrast, McDonnell et al 2t found that only 2% were bilateral on
Ezra et al • Five-year Incidence of F T M H in Fellow Eyes
initial examination, and of 36 patients with normal fellow eyes initially, none of the 21 that were available for follow-up had holes develop in the fellow eye (mean followup, 27 months). In neither study, however, was the presence or absence of a PVD in normal fellow eyes reported and correlated to the development of an FTMH. In another study of 49 fellow eyes, Trempe et a123 found a bilaterality of 3% at presentation. Subsequently, over an average of 47 months' follow-up, 16% had holes develop, all of which had no PVD at presentation. The incidence in eyes with a PVD was 0 (0%) of 21 compared with that of 8 (29%) of 28 eyes without a PVD. However, in this study, no differentiation was made between normal foveae and those with predisposing lesions. The study showed that the presence of an attached vitreous was a significant risk factor in fellow eyes, but it did not determine the incidence of holes in normal fellow eyes. Lewis et al, 24 in a retrospective review of 340 patients with an FTMH in one eye and a normal fellow eye, showed a cumulative incidence of fellow eye FTMH of 13% at 4 years using survival analysis. However, in 111 patients, the status of the vitreous had not been recorded at the initial visit. In contrast to these reports, which are retrospective, the current study has reported prospective data on a cohort of consecutive patients with normal fellow eyes and an attached posterior vitreous at presentation and represents the largest prospective series to date. The best estimate, as determined by the K a p l a n - M e i e r method, for fellow eye FTMH formation was 7.5% (range, 2.2%-12.5%; P = 0.05) at 18 months and 15.6% (range, 8.4%-22.3%; P = 0.05) at 5 years and represents a significant risk to the fellow eye. Although this estimate is comparable to that of Lewis et a124 and some other studies cited above, it is somewhat higher than the 3.5% at 2.8 years reported by Fisher and associates, -'6 who reported prospectively 46 normal fellow eyes of patients with unilateral holes. Of these, 16 had a PVD, of which none had holes or premacular hole lesions develop, and 30 eyes had an attached posterior vitreous. If the 16 eyes with a PVD are excluded, 1 (3.5%) of 30 had FTMH develop and 2 (6%) of 30 had stage 1 lesions develop. Thirty-three percent of fellow eyes with an initially attached posterior vitreous subsequently had a PVD without FTMH develop. However, a comparison between the two studies is difficult, because Fisher and associates did not specify the mean follow-up period in the subgroup of normal fellow eyes without PVD. In addition, the series was not strictly consecutive. A number of patients were excluded because of lack of follow-up, and no K a p l a n - M e i e r calculation was available. If some of these patients subsequently had impending holes or FTMH develop and were treated elsewhere, the incidence would be artificially low. Furthermore, another 2 of 30 eyes had impending holes develop. If the follow-up period, which is not specified, for these two eyes was short, one or both subsequently may have had an FTMH develop. For these reasons, because of the small sample size of their cohort and the lack of more detailed stratification of follow-up data, a direct compari-
son between the study of Fisher and associates and our study is difficult. The observation in our study that in no case did an FTMH develop after a PVD is consistent with the findings of Fisher and associates and lends further support to the theory of vitreoretinai traction as the cause of FTMH. The objective electrodiagnostic and psychophysical tests used in this study were unable to identify fellow eyes at higher risk of having an FTMH develop. The normality of the EOG findings in both affected eyes and normal fellow eyes at baseline was consistent with the focal nature of the hole and its affects on the retina rather than the pigment epithelium and choroid. The normal pattern electroretinogram (PERG) P50 in affected and normal fellow eyes, at baseline, is consistent with an earlier study, 45 probably reflects the fact that the PERG is generated achromatically by large ganglion cells, 27 and suggests that these cells are largely intact and functional in eyes with FTMH. This also would be consistent with the focal nature of the disease (the checkerboard subtends a 17° solid cone at the fovea, which is far larger than the area of even the largest holes), whereby macular function remains largely intact around the hole, and would support the recent suggestion that in most cases, FFMH arises from a foveal dehiscence without loss of retinal tissue. 46 The preservation of the PERG is in contrast to most other macular diseases that affect a greater proportion of the macula in which the PERG declines as visual acuity becomes impaired. 47 A previous study using the focal ERG (FERG), with a white flicker stimulus stimulating a 3 ° area of fovea, has shown significantly reduced FERGs in eyes with existing FTMH, as well as in fellow eyes in which FTMH later developed. 48 The authors concluded that the reduced FERGs could result from decreased optical density because of shortened or misaligned cone outer segments, again supporting the view that this could arise from tangential vitreomacular traction. Although the authors suggested that FERG may have a role in predicting which fellow eyes were at higher risk, no further data have been forthcoming. The use of CCS testing in our study showed markedly elevated protan, deutan, and tritan thresholds in affected eyes at baseline. This is perhaps not unexpected, as the stimulus subtended a 5* cone at the fovea and represents a focal stimulus of similar size to the fovea itself. The loss of CCS probably reflects reduced foveal function due to either the loss of significant numbers of cones in the operculum or to the separation of foveal cones from the retinal pigment epithelium around the hole, without necessarily any avulsion of cones. Decreased foveal function without necessarily a loss of retinal tissue would be consistent with these results. More surprising, perhaps, is the elevation of thresholds to tritan stimuli at baseline in apparently normal fellow eyes. Because the tritan threshold testing involved a correction for blue wavelength absorption by lens pigment or xanthophyll between individuals, it is likely that the increase in threshold reflects a genuine color dysfunction. In apparently normal fellow eyes, this may be a reflection
357
Ophthalmology Volume 105, Number 2, February 1998 of incipient or subclinical traction on the fovea, causing subclinical receptoral or photoreceptor dysfunction. The lack of protan and deutan CCS abnormality in fellow eyes may reflect the fact that tritan function is the most susceptible in the early stages of vitreofoveal traction. In fact, isolated tritan CCS loss has been shown in subjects with ocular hypertension and early glaucoma, 49 and it has been speculated that postreceptorai pathways subserving tritan function may be the most susceptible to early damage in this condition. 49"5° The normality of the PERG in fellow eyes suggests that tritan dysfunction probably represents dysfunction of the photoreceptors rather than postreceptoral level. The reason for susceptibility blue cones, per se, however, remains unclear. If subclinical vitreomacular traction is the underlying cause for tritan loss, the lack of correlation between tritan loss and FTMH formation in the fellow eye suggests that the severity of traction may not be the sole determining factor for hole formation. In summary, this study has shown an estimated 5-year incidence of macular hole formation in fellow eyes of 15.6% and that a significant number of patients will have a PVD without hole formation develop. The study also emphasizes the importance of determining the state of the vitreous in normal fellow eyes so that an accurate risk prediction may be given to the patient. The relatively high incidence of PVD and the finding that CCS tritan thresholds are elevated in normal fellow eyes support the theory that an abnormality at the vitreoretinal interface exists and that subclinical foveal dysfunction or cone misalignment may exist even in apparently normal fellow eyes. Acknowledgment. The authors thank Dr. C. Hogg for technical support.
9. 10.
11. 12. 13.
14.
15. 16. 17. 18. 19. 20. 21. 22.
References I. McDonnell PJ, Fine SL, Hillis AI. Clinical features of idiopathic macular cysts and holes. Am J Ophthalmol 1982;93: 777-86. 2. Gass JD. Idiopathic senile macular hole. Its early stages arid pathogenesis. Arch Ophthalmol 1988; 106:629-39. 3. Johnson RN, Gass JD. Idiopathic macular holes. Observations, stages of formation, and implications for surgical intervention. Ophthalmology 1988;95:917-24. 4. De Bustros S. Vitrectomy for prevention of macular holes. Results of a randomized multicenter clinical trial. Ophthalmology 1994; 101:1055-60. 5. Kelly NE, Wendel RT. Vitreous surgery for idiopathic macular holes. Results of a pilot study. Arch Ophthalmol 1991; 190:654-9. 6. Wendel RT, Patel AC, Kelly NE, et al. Vitreous surgery for macular holes. Ophthalmology 1993;100:1671-6. Comment in: Ophthalmology 1993; 100:1607-8. 7. Glaser BM, Michels RG, Kuppermann BD, et al. Transforming growth factor-/32 for the treatment of full-thickness macular holes. A prospective randomized study. Ophthalmology 1992;99:1162-73. 8. Lansing MB, Glaser BM, Liss H, et al. The effect of pars plana vitrectomy and transforming growth factor-beta 2
358
23. 24. 25. 26.
27.
28. 29. 30.
without epiretinal membrane peeling on full-thickness macular holes. Ophthalmology 1993; 100:868-72. Wells JA, Gregor ZJ. Surgical treatnlent of full-thickness macular holes using autologous serum. Eye 1996; 10:593-9. Smiddy WE, Glaser BM, Thompson JT, et al. Transforming growth factor-/32 significantly enhances the ability to flatten the rim of subretinal fluid surrounding macular holes: preliminary anatomic results of a multicenter prospective randomized study. Retina 1993; 13:296-301. Ryan EH, Jr, Gilbert HD. Results of surgical treatment of recent-onset full-thickness idiopathic macular holes. Arch Ophthalmol 1994; 112:1545-53. Liggett PE, Skolik S, Horio B, et al. Human autologous serum for the treatment of full-thickness macular holes: a preliminary study. Ophthalmology 1995; 102:1071-6. Gaudric A, Massin P, Paques M, et al. Autologous platelet concentrate for the treatment of full-thickness macular holes. Graefes Arch Clin Exp Ophthalmol 1995;233:54954. Freeman WR, Azen SP, Kim JW, et al. Vitrectomy for the treatment of full-thickness stage 3 or 4 macular holes. Resuits of a multicentered randomized clinical trial. Arch Ophthalmol 1997; 115:11-21. Croll LJ, Croll M. Hole in the macula. Am J Ophthalmol 1950; 33:248-53. Yaoeda H. Clinical observation on the macular hole. Acta Soc Ophthal Jpn 1967;71:1723-36. Aaberg TM, Blair CJ, Gass JD. Macular holes. Am J Ophthalmol 1970;69:555-62. Aaberg TM. Macular holes: a review. Surv Ophthalmol 1970; 15:139-62. James M, Feman SS. Macular holes. Albrecht Von Graefes Arch Klin Exp Ophthalmol 1980;215:59-63. Bronstein MA, Trempe CL, Freeman HM. Fellow eyes of eyes with macular holes. Am J Ophthalmol 1981;92:75761. McDonnell PJ, Fine SL, Hillis AI. Clinical features of idiopathic macular cysts and holes. Am J Ophthalmol 1982; 93:777-86. Morgan CM, Schatz H. Idiopathic macular holes. Am J Ophthalmol 1985;99:437-44. Trempe CL, Weiter JJ, Furukawa H. Fellow eyes in cases of macular hole. Biomicroscopic study of the vitreous. Arch Ophthalmol 1986; 104:93-5. Lewis ML, Cohen SM, Smiddy WE, Gass JD. Bilaterality of idiopathic macular holes. Graefes Arch Clin Exp Ophthalmol 1996;234:241-5. Guyer DR, de Bustros S, Diener-West M, Fine SL. Observations on patients with idiopathic macular holes and cysts. Arch Ophthalmol 1992; 110:1264-8. Fisher YL, Slakter JS, Yannuzzi LA, Guyer DR. A prospective naturalhistory study and kinetic ultrasound evaluation of idiopathic macular holes. Ophthalmology 1994; 101:511. Arden GB, Vaegan, Hogg CR. Clinical and experimental evidence that the pattern electroretinogram (PERG) is generated in more proximal retinal layers than the focal electroretinogram (FERG). Ann N Y Acad Sci 1982;388:580601. Arden GB, Barrada A, Kelsey JH. New clinical test of retinal function based upon the standing potential of the eye. Br J Ophthalmol 1962;46:449-67. Arden GB, GUndtiz K, Perry S. Colour vision testing with a computer graphics system. Clinical Vision Sciences 1988; 2:303 -20. Arden GB, Berninger T, Hogg CR, Perry S. A survey of
Ezra et al • F i v e - y e a r I n c i d e n c e o f F T M H in F e l l o w Eyes
31. 32. 33. 34. 35. 36. 37. 38. 39. 40.
color discrimination in German ophthalmologists. Changes associated with the use of lasers and operating microscopes. Ophthalmology 1991 ; 98:567-75. Schepens CL. Fundus changes caused by alterations of the vitreous body. Am J Ophthalmol 1955;39:631-3. Maumenee AE. Further advances in the study of the macula. Arch Ophthalmol 1967; 78:151-65. Yoshioka H. Clinical studies on the macular holes. 3. On the pathogenesis of the senile macular hole. Nippon Ganka Gakka: Zasohi 1968;72:575-84. Reese AB, Jones IS, Cooper WC. Macular changes secondary to vitreous traction. Am J Ophthalmol 1967;64(Suppl): 544-9. Margherio RR, Schepens CL. Macular breaks: 1. Diagnosis, etiology and observations. Am J Ophthalmol 1972;74:21932. Tolentino FI, Schepens CL, Freeman HM. Vitreoretinal disorders: diagnosis and management. Philadelphia: Saunders, 1976;405. Avila MP, Jalkh AE, Murakami K, et al. Biomicroscopic study of the vitreous in macular breaks. Ophthalmology 1983; 90:1277- 83. Hikichi T, Akiba J, Trempe CL. Effect of the vitreous on the prognosis of full-thickness idiopathic macular hole. Am J Ophthalmol 1993; 116:273-8. Hikichi T, Trempe CL. Risk of decreased visual acuity in full-thickness: idiopathic macular-holes. Am J Ophthalmol 1993; 116:708-12. Thompson JT, Glaser BM, Sjaarda RN, Murphy RP. Progression of nuclear sclerosis and long-term visual results of vitrectomy with transforming growth factor beta-2 for macular holes. Am J Ophthalmol 1995; 199:48-54.
41. Melberg NS, Thomas MA. Visual field loss after pars plana vitrectomy with air/fluid exchange. Am J Ophthalmol 1995; 120:386-8. 42. Ezra E, Arden GB, Riordan-Eva P, et al. Visual field loss following vitrectomy for stage 2 and 3 macular holes. Br J Ophthalmol 1996; 80:519-25. 43. Hutton WL, Fuller DG, Snyder WB, et al. Visual field defects after macular hole surgery. A new finding. Ophthalmology 1996; 103:2152-8. 44. Fine SL. Vitreous surgery for macular hole in perspective. Is there an indication? [editorial]. Arch Ophthalmol 1991; 109:635-6. 45. Smith RG, Brimlow GM, Lea SJ, Galloway NR. Evoked responses in patients with macular holes. Doc Ophthalmol 1990;75:135-44. 46. Gass JD. Reappraisal of biomicroscopic classification of stages of development of a macular hole. Am J Ophthalmol 1995; 119:752-9. 47. Arden GB, Carter RM, Macfarlan A. Pattern and Ganzfeld electroretinograms in macular disease. Br J Ophthalmol 1984; 68:878-84. 48. Birch DG, Jost BF, Fish GE. The focal electroretinogram in fellow eyes of patients with idiopathic macular holes. Arch Ophthalmol 1988; 106:1558-63. 49. Gtindtiz K, Arden GB, Perry S, et al. Color vision defects in ocular hypertension and glaucoma. Quantification with a computer-driven color television system. Arch Ophthalmol 1988; 106:929-35. 50. Weinstein GW, Arden GB, Hitchings RA, et al. The pattern electroretinogram (PERG) in ocular hypertension and glaucoma. Arch Ophthalmol 1988; 106:923-8.
359