- Email: [email protected]
Five-year cumulative incidence and progression of epiretinal membranes
Five-Year Cumulative Incidence and Progression of Epiretinal Membranes The Blue Mountains Eye Study Samantha Fraser-Bell, MBBS, Magdalena Guzowski, MBBS, Elena Rochtchina, MAppl Stat, Jie Jin Wang, MMed, PhD, Paul Mitchell, MD, PhD Purpose: To describe the 5-year cumulative incidence and change in epiretinal membranes in a defined older Australian population. Design: Population-based cohort study. Participants: Three thousand six hundred fifty-four persons 49 years of age or older, living in the Blue Mountains area, west of Sydney, Australia, participated in the baseline survey during 1992 to 1994. The cohort was reexamined after 5 years in 1997 to 1999. Excluding persons (543) who died since the baseline, 75% of survivors (n ⫽ 2335) attended the follow-up examination. Methods: All participants underwent a detailed eye examination, including stereo retinal photography. Epiretinal membranes were diagnosed from grading of baseline and 5-year retinal photographs. Main Outcome Measures: Epiretinal membranes were classified as either preretinal macular fibrosis (PMF), with retinal folds, or as a less severe form, termed cellophane macular reflex (CMR), without retinal folds. The incidence of epiretinal membranes was determined if either lesion was found in eyes with no preexisting epiretinal membrane at baseline. Progression was defined if the area of involvement increased by more than 25%, regression if it decreased by more than 25%, and stable if it changed by less than 25%. Results: Epiretinal membranes developed in the first eye of 108 of 2030 participants who had no sign of this condition in either eye at baseline, 5.3%, 95% confidence interval (CI) 4.4 to 6.4. Five-year cumulative incidence rates for PMF and CMR were 1.5% and 3.8%, respectively. Of those participants with epiretinal membranes in one eye at baseline, 18 of 133 (13.5%) developed this sign in their second eye after 5 years. New epiretinal membranes (mostly CMR) occurred in 15 of 165 subjects (9.1%; CI, 5.2–14.6) who had undergone cataract surgery since the Blue Mountains Eye Study I. This rate was significantly higher than in the nonsurgical group, 92 of 1861 (4.9%; CI, 4.0 – 6.0) of whom developed epiretinal membranes. Progression from CMR to PMF was observed in 17 of 183 eyes (9.3%). Existing epiretinal membranes progressed, regressed, or remained stable in 28.6%, 25.7%, and 38.8% of eyes, respectively. Conclusions: This study has documented the 5-year cumulative incidence and the natural history of epiretinal membranes in an older population. Ophthalmology 2003;110:34 – 40 © 2003 by the American Academy of Ophthalmology.
Epiretinal membranes are a relatively common finding in older persons and have been associated with various ocular pathology, including retinal detachment, retinal vascular occlusions, macular holes, and cataract surgery.1–5 However, most cases occur in the absence of ocular disease processes other than posterior vitreous detachment6 – 8 and are termed “idiopathic.” They have also been variously
Originally received: September 17, 2001. Accepted: April 30, 2002. Manuscript no. 210786. From the Department of Ophthalmology and the Save Sight and Westmead Millennium Institutes, the University of Sydney, Sydney, Australia. Supported by the Australian National Health & Medical Research Council, Canberra, Australia (grant no 974159). Reprint requests to Paul Mitchell, MD, PhD, Department of Ophthalmology, University of Sydney, Hawkesbury Rd, Westmead, NSW, Australia, 2145.
34
© 2003 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
termed “macular pucker,”1,5,9,10 “preretinal macular fibrosis,”1,4,6,11–13 “epiretinal fibrosis,”14 “gliosis,”1,12,15 “surface wrinkling retinopathy,”15 or “cellophane maculopathy.”16 Epiretinal membranes have been reported to be mostly slowly progressive, with only a small proportion causing significant effects on vision, including reduced visual acuity, metamorphopsia, and, occasionally, monocular diplopia.11,15,17 Uncommonly, the tangential traction on the macular retina causes sufficient visual disturbance to require surgical peeling.14,16 Epiretinal membranes are proliferations at the vitreoretinal junction, after posterior vitreous detachment. The predominant cellular populations found on ultrastructural examination include retinal pigment epithelium, fibrous astrocytes, fibrocytes, and myofibroblasts.9,14,16 In idiopathic cases, glial cells are thought to migrate through interruptions in the internal limiting membrane after posteISSN 0161-6420/03/$–see front matter PII S0161-6420(02)01443-4
Fraser-Bell et al 䡠 Five-Year Incidence and Progression of Epiretinal Membranes rior vitreous detachment.9,12,14,18 Retinal pigment epithelium cell migration may also occur through retinal breaks and detachments,14 or perhaps through an intact, but physiologically altered, retina.9 In patients with retinal vein occlusion or diabetic retinopathy, proliferative factors may leak through leaky vessels.14 There have been two recent population-based estimates of the prevalence of epiretinal membranes in older adults: the Beaver Dam Eye Study,19 which reported a prevalence of 11.8%, and the Blue Mountains Eye Study (BMES), which found this sign in 7% of participants.20 In both reports, epiretinal membranes were associated with diabetes (after excluding cases with retinopathy), previous cataract surgery, and retinal vein occlusion. To our knowledge, there have been no previous population-based studies reporting the incidence of epiretinal membranes. The purpose of this study was to document the 5-year cumulative incidence of epiretinal membranes in a large, representative population-based cohort of older Australians, and to describe the course of preexisting epiretinal membranes over this period.
Material and Methods Population Studied The BMES is a population-based survey of vision and common eye diseases in an area west of Sydney, Australia. The population is representative of Australia for income and education and has been previously described.21,22 All noninstitutionalized permanent residents with birth dates before January 1, 1943, were invited to participate in the study. From 4433 eligible persons, 3654 (82.4%) participated in the baseline eye examinations (BMES I) during 1992 to 1994. Five-year follow-up eye examinations (BMES II) were conducted during 1997 to 1999. Of those seen at baseline, 383 (10.5%) had moved, 394 (10.8%) refused to participate, and 543 (14.9%) had died. Thus, 2334 (75.0%) persons returned for a follow-up examination. Participants provided written, informed consent on both occasions. The Western Sydney Human Ethics Committee provided ethical approval. The age-gender profile of participants at BMES II is shown in Figure 1.
Procedures At both examinations, the participants were administered a detailed questionnaire regarding their medical and ophthalmic history, including a history of diabetes, hypertension, and systemic vascular disease. The presence of diabetes was defined as a selfreported history of diabetes, treatment with hypoglycemic medication, and/or a fasting glucose of greater than 7.0 mmol/l. All subjects underwent a comprehensive eye examination, which included subjective refraction using a logarithm of the minimum angle of resolution chart. Stereoscopic retinal photographs of the macula (30°) and other retinal fields of both eyes were taken, as described previously,20 using a Zeiss FF3 camera (Carl Zeiss, Oberkochen, Germany).
Classification of Epiretinal Membranes and Subjects Included The classification and grading system for epiretinal membranes used in the follow-up study were the same as in the BMES baseline
Figure 1. Age-gender distribution of 2334 participants seen in the 5-year follow-up Blue Mountains Eye Study examinations, 1997 to 1999.
study,20 adopted from Klein et al.19 Two types of epiretinal membranes were identified: a more severe form, termed “preretinal macular fibrosis” (PMF), in which retinal folds were identified; and a less severe form termed “cellophane macular reflex” (CMR), without visible retinal folds. Eyes with both CMR and PMF present were classified as having PMF. Initially, all photographs were graded in a masked manner. When epiretinal membranes were identified at either examination, a side-by-side grading of the baseline and 5-year photographs was conducted. The site of the epiretinal membranes was classified using a transparent grid developed for grading of age-related maculopathy,23 provided by Dr. Ronald Klein (University of Wisconsin-Madison). The grid radius is 3000 m and corresponds to the anatomic macula. It includes three concentric circles surrounding the central macula. Four radial lines divide the overall grid area into a central subfield, with four inner and four outer subfields. The percentage of each subfield involved by epiretinal membrane was estimated. Incident cases in the first eye were defined when epiretinal membranes were present at 5-year follow-up in either eye of subjects who had no such lesions in either eye at BMES I (Figs 2 and 3). Cumulative incidence rates for the second eye were assessed among participants who had epiretinal membranes in one eye only at baseline. All subjects with CMR initially who later developed PMF were classified as incident cases of PMF. Progression of epiretinal membranes was defined when the area of PMF or CMR at BMES I increased by more than 25% in BMES II, as shown in Figure 4. Regression of epiretinal membranes was defined if (1) there was a decrease of more than 25% in the area involved (Fig 5.), (2) the epiretinal membrane disappeared, or (3) PMF changed to CMR. Epiretinal membranes were graded as stable if there was less than 25% change in the area involved. Of 2334 subjects examined, 2218 had photographs considered gradable for epiretinal membranes. Excluded were 88 persons with no photographs at either the BMES I or BMES II examinations, 9 had ungradable photographs, and 19 had signs of late age-related macular degeneration, considered likely to confound grading for epiretinal membranes. Of the 2218 subjects seen in BMES II, 2030 had no epiretinal membrane in either eye at baseline. Epiretinal membranes have been associated with many other disease processes. Incident cases were classified in a similar fashion to the BMES I report20 into idiopathic and secondary epiretinal membranes. Secondary cases included subjects with previous cataract surgery, retinal vascular disease (including retinal vein occlusion or diabetic retinopathy), retinal detachment, or other retinal
35
36
Fraser-Bell et al 䡠 Five-Year Incidence and Progression of Epiretinal Membranes Figure 2. A, Left eye of 61-year-old man with normal macula at baseline. B, After 5 years, early preretinal macular fibrosis with focus superior to macula (visual acuity 20/20). Figure 3. A, Left eye of 72-year-old woman with normal macula at baseline (visual acuity 20/20). B, After 5 years, marked preretinal macular fibrosis with dragged appearance of central vasculature (visual acuity 20/80). Figure 4. A, Right eye of 71-year-old woman with early preretinal macular fibrosis at the baseline examination (visual acuity 20/20). B, After 5 years, marked increase in epiretinal membrane (visual acuity 20/30). Figure 5. A, Left eye of 77-year-old woman with preretinal macular fibrosis at the baseline examination (visual acuity 20/25). B, After 5 years, marked decrease in epiretinal membrane (visual acuity 20/25). Š
lesions.4,16,19,24 Epiretinal membranes were classified as idiopathic when present in subjects without any secondary causes.
Statistical Methods Statistical Analysis System (SAS Institute Inc., Cary, NC) was used for tabulation and statistical analysis, including chi-square tests, logistic regression analysis, and analysis of covariance. Odds ratios (OR) and 95% confidence intervals (CI) are reported. Regression analyses incorporating data from both eyes of a subject were performed using the method described by Zeger et al.25
Results Over the 5-year follow-up period, incident epiretinal membranes were observed in 108 eyes of the 2030 participants with no preexisting lesions in either eye at baseline (5.3%; CI, 4.4 – 6.4). All incident cases were unilateral, except for one participant who developed CMR in one eye and PMF in the fellow eye. The age-specific cumulative incidence of epiretinal membranes is shown in Table 1 and Figure 6. The incidence was 3.7% among participants younger than 60 years of age, 6.6% for persons aged 60 to 69 years, 6.1% for persons aged 70 to 79 years, and 1.1% for persons aged 80 years or older. There was no statistically significant association between age and epiretinal membrane incidence, gender-adjusted OR 1.1 (CI, 0.9 –1.5). The incidence was 5.7% in women and 4.8% in men, but this difference was not statistically significant after adjusting for age, OR 1.3 (CI, 0.8 –2.0). PMF cumulative incidence (1.5%; CI, 1.0 –2.1) was less than half the CMR incidence (3.8%; CI, 3.0 – 4.7), as shown in Table 1. Incident PMF cases were all unilateral, whereas 15 incident CMR cases (19.5%) were bilateral. Epiretinal membrane incidence peaked between the ages of 60 and 79 years. Both types of epiretinal membranes occurred slightly more frequently in women than men: for PMF, 1.6% versus 1.5%; and for CMR, 4.1% versus 3.4%. Cumulative incidence rates for the idiopathic group and for subjects with secondary causes are shown in Table 2. The epiretinal membrane incidence among 179 persons who had potential secondary causes was only slightly higher (5.6%; CI, 2.7–10.1) than the incidence among the remainder of the study population (n
⫽ 1851), who were considered to have developed idiopathic epiretinal membranes (5.2%; CI, 4.2– 6.3). Among incident cases with a secondary cause, two had retinopathy, two had retinal vein occlusion, and four had previous cataract surgery at the BMES I examination. New epiretinal membranes were found in 15 of 165 subjects (9.1%; CI, 5.2–14.6) who had undergone cataract surgery since BMES I, including 3 cases of PMF (1.8%) and 12 cases of CMR (7.3%). This incidence rate was significantly higher than for the nonsurgical group, P ⫽ 0.02; whereas 92 of 1865 nonsurgical cases (4.9%; CI, 4.0 – 6.0) developed epiretinal membranes, including 28 cases of PMF (1.5%) and 64 cases of CMR (3.4%). Second eye involvement is shown in Table 3; 13.5% of participants with epiretinal membranes in their first eye developed membranes in their second eye over 5 years. The incidence of second eye involvement was 2.5 times higher (P ⬍ 0.001) than the incidence for first eye involvement. After excluding cases with signs of diabetic retinopathy or other known secondary causes, logistic regression analyses were performed to assess factors associated with incident idiopathic epiretinal membranes. After adjusting for age and gender, presence of diabetes was not significantly associated with incident epiretinal membranes, but there was relatively low power to examine this association. There was also no association found between PMF and either fasting plasma glucose (OR, 0.6; CI, 0.3–1.3) or hypertension (OR, 0.6; CI, 0.3–1.4). Incident CMR was also not significantly associated with diabetes (OR, 1.4; CI, 0.5– 4.1), fasting plasma glucose (OR, 0.9; CI, 0.6 –1.2), or hypertension (OR, 0.7; CI, 0.4 –1.2). Combining the two membrane types failed to demonstrate any association with diabetes, fasting glucose, or hypertension. Overall, the mean visual acuity level was only slightly affected by incident PMF in this study and was unaffected by incident CMR. The average reduction of visual acuity after 5 years in the worse eye was 5.7 letters (CI, 5.3– 6.1) in subjects without incident epiretinal membranes, 7.4 letters (CI, 4.7–10.1) in eyes with incident PMF, and 2.8 letters (CI, 1.6 – 4.0) in eyes with incident CMR. Corresponding age-adjusted and gender-adjusted mean differences in visual acuity between the baseline and 5-year examination were 7.5 letters in the worse eye of subjects without epiretinal membranes, 9.0 letters for eyes with incident PMF, and 4.4 letters in eyes with incident CMR. However, these differences
Table 1. Five-Year Cumulative Incidence (%) of Epiretinal Membranes by Age and Gender in the Blue Mountains Eye Study (n ⫽ 2030) Age Group Type of Epiretinal Membrane
⬍60 (n ⫽ 682)
60–69 (n ⫽ 817)
70–79 (n ⫽ 442)
80⫹ (n ⫽ 89)
Women (n ⫽ 1150)
Men (n ⫽ 880)
All (n ⫽ 2030)
Preretinal macular fibrosis Cellophane macular reflex Any
1.2 2.5 3.7
1.7 4.9 6.6
2.0 4.3 6.1
0.0 1.1 1.1
1.6 4.1 5.7
1.5 3.4 4.8
1.5 3.8 5.3
37
Ophthalmology Volume 110, Number 1, January 2003 Table 3. Five-Year Incidence of Epiretinal Membranes in the Second Eye of Participants with Epiretinal Membrane in One Eye at Baseline Compared with De Novo Incident Cases Type of Epiretinal Membrane Number at risk
Preretinal Macular Fibrosis
Cellophane Macular Reflex
Any
2030
1.5 (1.0–2.1)
3.8 (3.0–4.7)
5.3 (4.4–6.4)
Incidence in first eye % (95% CI) Incidence in second eye % (95% CI)
133
6.0 (2.6–11.5) 7.5 (3.7–13.4) 13.5 (8.2–20.5)
CI ⫽ confidence interval. Figure 6. Five-year incidence of epiretinal membranes by age and gender in 2030 participants with no preexisting lesion at baseline and who had gradable retinal photographs.
were not statistically significant. A reduction of visual acuity by 5 or more letters (1 line or greater) was observed in 42.9% of eyes with incident PMF compared with 28.3% of eyes that had not developed epiretinal membranes. Table 4 shows the distribution of the change in visual acuity (number of letters lost) in eyes with incident PMF. There were 7.1% of participants with incident PMF in either eye who complained of distortion in the affected eye, whereas only 2% of participants with incident CMR and 2.7% of participants without incident epiretinal membrane gave this symptom. Table 5 shows the 5-year course of epiretinal membranes in this study. Of 56 participants (62 eyes) with PMF at baseline, this progressed in 16.1% and regressed in 25.8%. Of 142 participants (183 eyes) with CMR at baseline, this progressed in 32.8% and regressed in 25.7%, including 17 eyes in which CMR progressed to PMF (9.3%). Among regressed CMR cases, the area of involvement decreased by more than 25% in 23 cases (12.6%) and disappeared completely in 24 cases (13%). There was no significant association between ocular and systemic risk factors detected to predict progression of PMF, but this was largely due to low study power, because only 10 eyes of 62 progressed during the 5-year follow-up. Table 2. Incidence of Idiopathic and Secondary Epiretinal Membranes in Participants with No Preexisting Lesion at Baseline Type of Epiretinal Membrane
Cases
Incidence % (95% Confidence Interval)
Idiopathic epiretinal membranes (n ⫽ 1851) PMF 29 1.6 (1.1–2.3) CMR 69 3.7 (2.9–4.7) Any 97* 5.2 (4.2–6.3) Secondary epiretinal membranes (n ⫽ 179) PMF 2 1.1 (0.1–4.0) CMR 8 4.5 (2.0–8.7) Any 10 5.6 (2.7–10.1) *One participant developed CMR in one eye and PMF in the other eye. CMR ⫽ cellophane macular reflex; PMF ⫽ preretinal macular fibrosis.
38
Regression analyses, adjusting for age and gender, was used to assess whether the finding of an epiretinal membrane at baseline predicted incident visual impairment after 5 years. This analysis found no significant increase in the proportion of the 62 eyes with PMF at BMES I that developed incident visual impairment ⬍20/40 compared with eyes without PMF, OR 0.4 (CI, 0.1–1.6).
Discussion Although the prevalence of epiretinal membranes has been described in two recent population-based reports,19,20 no population data have yet described the incidence and course of this relatively common lesion over time. Although most epiretinal membranes do not cause significant visual disability, some cases lead to progressive visual loss and may require surgical intervention. Cumulative incidence data on the development and progression of this common lesion, together with effects on visual acuity, are provided by this study. The report has also examined ocular and systemic associations for incident epiretinal membranes. The 5-year incidence of epiretinal membranes in this older population sample was 5.3%. The cumulative incidence of PMF was less than half the CMR cumulative incidence (1.5% vs. 3.8%), which corresponds closely with findings reported from both BMES I20 and Beaver Dam.19 These studies reported PMF:CMR prevalences of 2.2%: 4.8% from BMES I and corresponding rates of 2.8%:9.0% for Beaver Dam. The age-related increase in epiretinal membrane inciTable 4. Change in Visual Acuity in Any Eyes with Incident Preretinal Macular Fibrosis (n ⫽ 56)
5-Year Change in Visual Acuity
Eyes with Incident Preretinal Macular Fibrosis (n ⴝ 56)
No change or improved by ⬍5 letters Decreased by 1–4 letters Decreased by 5–9 letters Decreased by 10–19 letters Decreased by 20 or more letters
14 18 16 3 5
Fraser-Bell et al 䡠 Five-Year Incidence and Progression of Epiretinal Membranes Table 5. Change Observed in Eyes with Preexisting Epiretinal Membranes during the 5 Year Period Type of Epiretinal Membrane at Baseline
Change over 5-year Period Progressed Regressed Stable CMR to PMF*
Preretinal Macular Fibrosis (n ⫽ 62 eyes)
Cellophane Macular Reflex (n ⫽ 183 eyes)
10 (16.1%) 16 (25.8%) 36 (58.1%)
60 (32.8%) 47 (25.7%) 59 (32.2%) 17 (9.3%)
Any (n ⫽ 245 eyes) 70 (28.6%) 63 (25.7%) 95 (38.8%)
*CMR preexisting cases only. CMR ⫽ cellophane macular reflex. PMF ⫽ preretinal macular fibrosis.
dence observed for persons aged less than 70 years corresponded with the increase in prevalence up to this age found in both BMES I20 and in Beaver Dam.19 We previously argued that the lower prevalence rate found for older age groups could represent potential underreporting as a result of increasing lens opacity.20 The corresponding decrease in incidence found in this study is consistent with this argument. We further observed no statistically significant gender difference in epiretinal membrane incidence, consistent with our prevalence report.20 This study also addresses the bilaterality of epiretinal membranes, found in BMES I to be 31% and in the Beaver Dam Eye Study to be 19.5%.19,20 Our incidence data suggest a relatively similar rate; the second eye incidence was more than double that in the first eye, 13.5% versus 5.3%, and second eyes were more likely than first eyes to develop PMF. PMF incidence in second eyes was four times that for first eyes (6% vs. 1.5%), and CMR second eye incidence was twice that for first eyes (7.5% vs. 3.8%). These data suggest a symmetric trend for this lesion, as seen with many other age-related conditions such as age-related maculopathy.26 Its major precursor, posterior vitreous detachment, is also frequently bilateral. Our study identified a statistically significant association between cataract surgery and incident epiretinal membranes, particularly the less severe form, CMR. The cumulative incidence of epiretinal membranes among participants who had cataract surgery during the 5-year period was almost double the nonsurgical group incidence. It is possible that, in some cases, these epiretinal membranes were present before surgery, but were not detected in the preoperative period because they were obscured by the lens opacity. The timing of incident epiretinal membranes in relation to cataract surgery during the follow-up period is also not known. Nevertheless, our finding of an association between cataract surgery and incident membranes supports findings from a number of previous reports,1,5,19 including our previous study.20 We are aware of only one report documenting the incidence of epiretinal membranes after cataract surgery by Jahn et al.5 In this prospective study of uneventful extracapsular cataract extraction with posterior chamber intraocular lens implantation in 332 eyes, 10% of eyes
developed new epiretinal membranes over a 6-month postoperative period. This is very close to our rate of 9.1% among subjects after cataract surgery. Although the two previous prevalence studies reported an association between diabetes and epiretinal membranes,19,20 we were not able to demonstrate this in our incidence data. In contrast with our prevalence study, epiretinal membrane incidence was not substantially different between eyes with and without secondary causes (5.6% vs. 5.2%), possibly because of the relatively small number of incident cases or because of age effects. Our study demonstrated that in most epiretinal membrane cases, no progression occurred over 5 years, a finding consistent with many previous clinic-based reports.11,17 Indeed, one in four cases actually demonstrated regression or even disappearance of the epiretinal membrane over this period, whereas only around 1 in 10 cases of CMR progressed to PMF. A limitation of our study is that only a relatively small number of subjects exhibited progression of PMF (only 10 of 62 baseline cases). Furthermore, larger studies will be needed to assess factors associated with progression. Previous studies have indicated that less than 15% of patients with epiretinal membranes have visual acuities worse than 20/70.8,11 One study reported no further decrease in visual acuity in 87% of 47 eyes with PMF followed for 2 to 4 years.11 Our study demonstrated a small reduction in mean visual acuity over the follow-up period in eyes with incident PMF compared with eyes not developing epiretinal membranes, but this was not found in eyes with incident CMR. Although we found no significant worsening of vision in eyes demonstrated to have progression of PMF, the power of our study to examine this was low, because only 16.1% of these epiretinal membranes progressed during follow-up. In conclusion, the current report provides useful information regarding the 5-year incidence and natural history of epiretinal membranes in an older population. The more severe type (PMF) was found to be relatively stable over this period, and only few cases developed incident visual impairment. This relative stability of epiretinal membranes and their limited effect on vision should be taken into account when considering surgical intervention with its potential for complications.
References 1. Appiah AP, Hirose T. Secondary causes of premacular fibrosis. Ophthalmology 1989;96:389 –92. 2. Margherio RM. Epiretinal macular membranes. In: Albert DM, Jakobiec FA, eds. Principles and Practice of Ophthalmology: Clinical Practice. Philadelphia: W.B. Saunders, 1994; v. 2, chap. 74. 3. Lobes LA Jr, Burton TC. The incidence of macular pucker after retinal detachment surgery. Am J Ophthalmol 1978;85: 72–7. 4. Kraushar MF, Morse PH. The relationship between retina surgery and preretinal macular fibrosis. Ophthalmic Surg 1988;19:843– 8. 5. Jahn CE, Minich V, Moldaschel S, et al. Epiretinal membranes
39
Ophthalmology Volume 110, Number 1, January 2003
6. 7. 8. 9.
10. 11. 12. 13. 14. 15. 16.
after extracapsular cataract surgery. J Cataract Refract Surg 2001;27:753– 60. Wise GN. Relationship of idiopathic preretinal macular fibrosis to posterior vitreous detachment. Am J Ophthalmol 1975; 79:358 – 62. Wiznia RA. Posterior vitreous detachment and idiopathic preretinal macular gliosis. Am J Ophthalmol 1986;102:196 – 8. Hirokawa H, Jalkh AE, Takahashi M, et al. Role of the vitreous in idiopathic preretinal macular fibrosis. Am J Ophthalmol 1986;101:166 –9. Smiddy WE, Maguire AM, Green WR, et al. Idiopathic epiretinal membranes. Ultrastructural characteristics and clinicopathologic correlation. Ophthalmology 1989;96:811–20; discussion 821. Tanenbaum HL, Schepens CL, Elzeneiny I, Freeman HM. Macular pucker following retinal detachment surgery. Arch Ophthalmol 1970;83:286 –93. Wiznia RA. Natural history of idiopathic preretinal macular fibrosis. Ann Ophthalmol 1982;14:876 – 8. Appiah AP, Hirose T, Kado M. A review of 324 cases of idiopathic premacular gliosis. Am J Ophthalmol 1988;106: 533–5. Sponsel WE, Kaufman PL, Blum FG Jr. Association of retinal capillary perfusion with visual status during chronic glaucoma therapy. Ophthalmology 1997;104:1026 –32. Ciulla TA, Pesavento RD. Epiretinal fibrosis. Ophthalmic Surg Lasers 1997;28:670 –9. Pearlstone AD. The incidence of idiopathic preretinal macular gliosis. Ann Ophthalmol 1985;17:378 – 80. Pournaras CJ, Donati G, Brazitikos PD, et al. Macular epiretinal membranes. Semin Ophthalmol 2000;15:100 –7.
17. Wise GN. Clinical features of idiopathic preretinal macular fibrosis. Schoenberg Lecture. Am J Ophthalmol 1975;79:349 –7. 18. Roth AM, Foos RY. Surface wrinkling retinopathy in eyes enucleated at autopsy. Trans Am Acad Ophthalmol Otolaryngol 1971;75:1047–58. 19. Klein R, Klein BEK, Wang Q, Moss SE. The epidemiology of epiretinal membranes. Trans Am Ophthalmol Soc 1994;92: 403–25; discussion 425–30. 20. Mitchell P, Smith W, Chey T, et al. Prevalence and associations of epiretinal membranes. The Blue Mountains Eye Study, Australia. Ophthalmology 1997;104:1033– 40. 21. Mitchell P, Smith W, Attebo K, Wang JJ. Prevalence of age-related maculopathy in Australia. The Blue Mountains Eye Study. Ophthalmology 1995;102:1450 – 60. 22. Attebo K, Mitchell P, Smith W. Visual acuity and the causes of visual loss in Australia. The Blue Mountains Eye Study. Ophthalmology 1996;103:357– 64. 23. Klein R, Davis MD, Magli YL, et al. The Wisconsin agerelated maculopathy grading system. Ophthalmology 1991;98: 1128 –34. 24. Smith W, Mitchell P, Wang JJ. Gender, oestrogen, hormone replacement and age-related macular degeneration: results from the Blue Mountains Eye Study. Aust N Z J Ophthalmol 1997;25(suppl 1):S13–5. 25. Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a generalized estimating equation approach [published erratum appears in Biometrics 1989;45:347]. Biometrics 1988;44: 1049 – 60. 26. Wang JJ, Mitchell P, Smith W, Cumming RG. Bilateral involvement by age related maculopathy lesions in a population. Br J Ophthalmol 1998;82:743–7.
AOS-KNAPP FUND FELLOWSHIP FUNDING The Knapp Fund, a supporting organization of the American Ophthalmological Society, provides funding for a second or third year of postgraduate study in ophthalmology. An annual stipend of $20,000 is available for postgraduate study beginning on or after July 1, 2003. To be eligible, applicants must be United States or Canadian citizens, the fellowship training must be conducted in the United States and applicants must have completed a residency program in ophthalmology accredited by the ACGME or the Royal College of Physicians and Surgeons of Canada. Deadline for receipt of applications is March 15, 2003. Froncie A. Gutman, M.D. Cleveland Clinic Foundation 9500 Euclid Avenue, Desk i-32 Cleveland, OH 44195 216-444-8968 (fax)
40
Epiretinal membranes are a relatively common finding in older persons and have been associated with various ocular pathology, including retinal detachment, retinal vascular occlusions, macular holes, and cataract surgery.1–5 However, most cases occur in the absence of ocular disease processes other than posterior vitreous detachment6 – 8 and are termed “idiopathic.” They have also been variously
Originally received: September 17, 2001. Accepted: April 30, 2002. Manuscript no. 210786. From the Department of Ophthalmology and the Save Sight and Westmead Millennium Institutes, the University of Sydney, Sydney, Australia. Supported by the Australian National Health & Medical Research Council, Canberra, Australia (grant no 974159). Reprint requests to Paul Mitchell, MD, PhD, Department of Ophthalmology, University of Sydney, Hawkesbury Rd, Westmead, NSW, Australia, 2145.
34
© 2003 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
termed “macular pucker,”1,5,9,10 “preretinal macular fibrosis,”1,4,6,11–13 “epiretinal fibrosis,”14 “gliosis,”1,12,15 “surface wrinkling retinopathy,”15 or “cellophane maculopathy.”16 Epiretinal membranes have been reported to be mostly slowly progressive, with only a small proportion causing significant effects on vision, including reduced visual acuity, metamorphopsia, and, occasionally, monocular diplopia.11,15,17 Uncommonly, the tangential traction on the macular retina causes sufficient visual disturbance to require surgical peeling.14,16 Epiretinal membranes are proliferations at the vitreoretinal junction, after posterior vitreous detachment. The predominant cellular populations found on ultrastructural examination include retinal pigment epithelium, fibrous astrocytes, fibrocytes, and myofibroblasts.9,14,16 In idiopathic cases, glial cells are thought to migrate through interruptions in the internal limiting membrane after posteISSN 0161-6420/03/$–see front matter PII S0161-6420(02)01443-4
Fraser-Bell et al 䡠 Five-Year Incidence and Progression of Epiretinal Membranes rior vitreous detachment.9,12,14,18 Retinal pigment epithelium cell migration may also occur through retinal breaks and detachments,14 or perhaps through an intact, but physiologically altered, retina.9 In patients with retinal vein occlusion or diabetic retinopathy, proliferative factors may leak through leaky vessels.14 There have been two recent population-based estimates of the prevalence of epiretinal membranes in older adults: the Beaver Dam Eye Study,19 which reported a prevalence of 11.8%, and the Blue Mountains Eye Study (BMES), which found this sign in 7% of participants.20 In both reports, epiretinal membranes were associated with diabetes (after excluding cases with retinopathy), previous cataract surgery, and retinal vein occlusion. To our knowledge, there have been no previous population-based studies reporting the incidence of epiretinal membranes. The purpose of this study was to document the 5-year cumulative incidence of epiretinal membranes in a large, representative population-based cohort of older Australians, and to describe the course of preexisting epiretinal membranes over this period.
Material and Methods Population Studied The BMES is a population-based survey of vision and common eye diseases in an area west of Sydney, Australia. The population is representative of Australia for income and education and has been previously described.21,22 All noninstitutionalized permanent residents with birth dates before January 1, 1943, were invited to participate in the study. From 4433 eligible persons, 3654 (82.4%) participated in the baseline eye examinations (BMES I) during 1992 to 1994. Five-year follow-up eye examinations (BMES II) were conducted during 1997 to 1999. Of those seen at baseline, 383 (10.5%) had moved, 394 (10.8%) refused to participate, and 543 (14.9%) had died. Thus, 2334 (75.0%) persons returned for a follow-up examination. Participants provided written, informed consent on both occasions. The Western Sydney Human Ethics Committee provided ethical approval. The age-gender profile of participants at BMES II is shown in Figure 1.
Procedures At both examinations, the participants were administered a detailed questionnaire regarding their medical and ophthalmic history, including a history of diabetes, hypertension, and systemic vascular disease. The presence of diabetes was defined as a selfreported history of diabetes, treatment with hypoglycemic medication, and/or a fasting glucose of greater than 7.0 mmol/l. All subjects underwent a comprehensive eye examination, which included subjective refraction using a logarithm of the minimum angle of resolution chart. Stereoscopic retinal photographs of the macula (30°) and other retinal fields of both eyes were taken, as described previously,20 using a Zeiss FF3 camera (Carl Zeiss, Oberkochen, Germany).
Classification of Epiretinal Membranes and Subjects Included The classification and grading system for epiretinal membranes used in the follow-up study were the same as in the BMES baseline
Figure 1. Age-gender distribution of 2334 participants seen in the 5-year follow-up Blue Mountains Eye Study examinations, 1997 to 1999.
study,20 adopted from Klein et al.19 Two types of epiretinal membranes were identified: a more severe form, termed “preretinal macular fibrosis” (PMF), in which retinal folds were identified; and a less severe form termed “cellophane macular reflex” (CMR), without visible retinal folds. Eyes with both CMR and PMF present were classified as having PMF. Initially, all photographs were graded in a masked manner. When epiretinal membranes were identified at either examination, a side-by-side grading of the baseline and 5-year photographs was conducted. The site of the epiretinal membranes was classified using a transparent grid developed for grading of age-related maculopathy,23 provided by Dr. Ronald Klein (University of Wisconsin-Madison). The grid radius is 3000 m and corresponds to the anatomic macula. It includes three concentric circles surrounding the central macula. Four radial lines divide the overall grid area into a central subfield, with four inner and four outer subfields. The percentage of each subfield involved by epiretinal membrane was estimated. Incident cases in the first eye were defined when epiretinal membranes were present at 5-year follow-up in either eye of subjects who had no such lesions in either eye at BMES I (Figs 2 and 3). Cumulative incidence rates for the second eye were assessed among participants who had epiretinal membranes in one eye only at baseline. All subjects with CMR initially who later developed PMF were classified as incident cases of PMF. Progression of epiretinal membranes was defined when the area of PMF or CMR at BMES I increased by more than 25% in BMES II, as shown in Figure 4. Regression of epiretinal membranes was defined if (1) there was a decrease of more than 25% in the area involved (Fig 5.), (2) the epiretinal membrane disappeared, or (3) PMF changed to CMR. Epiretinal membranes were graded as stable if there was less than 25% change in the area involved. Of 2334 subjects examined, 2218 had photographs considered gradable for epiretinal membranes. Excluded were 88 persons with no photographs at either the BMES I or BMES II examinations, 9 had ungradable photographs, and 19 had signs of late age-related macular degeneration, considered likely to confound grading for epiretinal membranes. Of the 2218 subjects seen in BMES II, 2030 had no epiretinal membrane in either eye at baseline. Epiretinal membranes have been associated with many other disease processes. Incident cases were classified in a similar fashion to the BMES I report20 into idiopathic and secondary epiretinal membranes. Secondary cases included subjects with previous cataract surgery, retinal vascular disease (including retinal vein occlusion or diabetic retinopathy), retinal detachment, or other retinal
35
36
Fraser-Bell et al 䡠 Five-Year Incidence and Progression of Epiretinal Membranes Figure 2. A, Left eye of 61-year-old man with normal macula at baseline. B, After 5 years, early preretinal macular fibrosis with focus superior to macula (visual acuity 20/20). Figure 3. A, Left eye of 72-year-old woman with normal macula at baseline (visual acuity 20/20). B, After 5 years, marked preretinal macular fibrosis with dragged appearance of central vasculature (visual acuity 20/80). Figure 4. A, Right eye of 71-year-old woman with early preretinal macular fibrosis at the baseline examination (visual acuity 20/20). B, After 5 years, marked increase in epiretinal membrane (visual acuity 20/30). Figure 5. A, Left eye of 77-year-old woman with preretinal macular fibrosis at the baseline examination (visual acuity 20/25). B, After 5 years, marked decrease in epiretinal membrane (visual acuity 20/25). Š
lesions.4,16,19,24 Epiretinal membranes were classified as idiopathic when present in subjects without any secondary causes.
Statistical Methods Statistical Analysis System (SAS Institute Inc., Cary, NC) was used for tabulation and statistical analysis, including chi-square tests, logistic regression analysis, and analysis of covariance. Odds ratios (OR) and 95% confidence intervals (CI) are reported. Regression analyses incorporating data from both eyes of a subject were performed using the method described by Zeger et al.25
Results Over the 5-year follow-up period, incident epiretinal membranes were observed in 108 eyes of the 2030 participants with no preexisting lesions in either eye at baseline (5.3%; CI, 4.4 – 6.4). All incident cases were unilateral, except for one participant who developed CMR in one eye and PMF in the fellow eye. The age-specific cumulative incidence of epiretinal membranes is shown in Table 1 and Figure 6. The incidence was 3.7% among participants younger than 60 years of age, 6.6% for persons aged 60 to 69 years, 6.1% for persons aged 70 to 79 years, and 1.1% for persons aged 80 years or older. There was no statistically significant association between age and epiretinal membrane incidence, gender-adjusted OR 1.1 (CI, 0.9 –1.5). The incidence was 5.7% in women and 4.8% in men, but this difference was not statistically significant after adjusting for age, OR 1.3 (CI, 0.8 –2.0). PMF cumulative incidence (1.5%; CI, 1.0 –2.1) was less than half the CMR incidence (3.8%; CI, 3.0 – 4.7), as shown in Table 1. Incident PMF cases were all unilateral, whereas 15 incident CMR cases (19.5%) were bilateral. Epiretinal membrane incidence peaked between the ages of 60 and 79 years. Both types of epiretinal membranes occurred slightly more frequently in women than men: for PMF, 1.6% versus 1.5%; and for CMR, 4.1% versus 3.4%. Cumulative incidence rates for the idiopathic group and for subjects with secondary causes are shown in Table 2. The epiretinal membrane incidence among 179 persons who had potential secondary causes was only slightly higher (5.6%; CI, 2.7–10.1) than the incidence among the remainder of the study population (n
⫽ 1851), who were considered to have developed idiopathic epiretinal membranes (5.2%; CI, 4.2– 6.3). Among incident cases with a secondary cause, two had retinopathy, two had retinal vein occlusion, and four had previous cataract surgery at the BMES I examination. New epiretinal membranes were found in 15 of 165 subjects (9.1%; CI, 5.2–14.6) who had undergone cataract surgery since BMES I, including 3 cases of PMF (1.8%) and 12 cases of CMR (7.3%). This incidence rate was significantly higher than for the nonsurgical group, P ⫽ 0.02; whereas 92 of 1865 nonsurgical cases (4.9%; CI, 4.0 – 6.0) developed epiretinal membranes, including 28 cases of PMF (1.5%) and 64 cases of CMR (3.4%). Second eye involvement is shown in Table 3; 13.5% of participants with epiretinal membranes in their first eye developed membranes in their second eye over 5 years. The incidence of second eye involvement was 2.5 times higher (P ⬍ 0.001) than the incidence for first eye involvement. After excluding cases with signs of diabetic retinopathy or other known secondary causes, logistic regression analyses were performed to assess factors associated with incident idiopathic epiretinal membranes. After adjusting for age and gender, presence of diabetes was not significantly associated with incident epiretinal membranes, but there was relatively low power to examine this association. There was also no association found between PMF and either fasting plasma glucose (OR, 0.6; CI, 0.3–1.3) or hypertension (OR, 0.6; CI, 0.3–1.4). Incident CMR was also not significantly associated with diabetes (OR, 1.4; CI, 0.5– 4.1), fasting plasma glucose (OR, 0.9; CI, 0.6 –1.2), or hypertension (OR, 0.7; CI, 0.4 –1.2). Combining the two membrane types failed to demonstrate any association with diabetes, fasting glucose, or hypertension. Overall, the mean visual acuity level was only slightly affected by incident PMF in this study and was unaffected by incident CMR. The average reduction of visual acuity after 5 years in the worse eye was 5.7 letters (CI, 5.3– 6.1) in subjects without incident epiretinal membranes, 7.4 letters (CI, 4.7–10.1) in eyes with incident PMF, and 2.8 letters (CI, 1.6 – 4.0) in eyes with incident CMR. Corresponding age-adjusted and gender-adjusted mean differences in visual acuity between the baseline and 5-year examination were 7.5 letters in the worse eye of subjects without epiretinal membranes, 9.0 letters for eyes with incident PMF, and 4.4 letters in eyes with incident CMR. However, these differences
Table 1. Five-Year Cumulative Incidence (%) of Epiretinal Membranes by Age and Gender in the Blue Mountains Eye Study (n ⫽ 2030) Age Group Type of Epiretinal Membrane
⬍60 (n ⫽ 682)
60–69 (n ⫽ 817)
70–79 (n ⫽ 442)
80⫹ (n ⫽ 89)
Women (n ⫽ 1150)
Men (n ⫽ 880)
All (n ⫽ 2030)
Preretinal macular fibrosis Cellophane macular reflex Any
1.2 2.5 3.7
1.7 4.9 6.6
2.0 4.3 6.1
0.0 1.1 1.1
1.6 4.1 5.7
1.5 3.4 4.8
1.5 3.8 5.3
37
Ophthalmology Volume 110, Number 1, January 2003 Table 3. Five-Year Incidence of Epiretinal Membranes in the Second Eye of Participants with Epiretinal Membrane in One Eye at Baseline Compared with De Novo Incident Cases Type of Epiretinal Membrane Number at risk
Preretinal Macular Fibrosis
Cellophane Macular Reflex
Any
2030
1.5 (1.0–2.1)
3.8 (3.0–4.7)
5.3 (4.4–6.4)
Incidence in first eye % (95% CI) Incidence in second eye % (95% CI)
133
6.0 (2.6–11.5) 7.5 (3.7–13.4) 13.5 (8.2–20.5)
CI ⫽ confidence interval. Figure 6. Five-year incidence of epiretinal membranes by age and gender in 2030 participants with no preexisting lesion at baseline and who had gradable retinal photographs.
were not statistically significant. A reduction of visual acuity by 5 or more letters (1 line or greater) was observed in 42.9% of eyes with incident PMF compared with 28.3% of eyes that had not developed epiretinal membranes. Table 4 shows the distribution of the change in visual acuity (number of letters lost) in eyes with incident PMF. There were 7.1% of participants with incident PMF in either eye who complained of distortion in the affected eye, whereas only 2% of participants with incident CMR and 2.7% of participants without incident epiretinal membrane gave this symptom. Table 5 shows the 5-year course of epiretinal membranes in this study. Of 56 participants (62 eyes) with PMF at baseline, this progressed in 16.1% and regressed in 25.8%. Of 142 participants (183 eyes) with CMR at baseline, this progressed in 32.8% and regressed in 25.7%, including 17 eyes in which CMR progressed to PMF (9.3%). Among regressed CMR cases, the area of involvement decreased by more than 25% in 23 cases (12.6%) and disappeared completely in 24 cases (13%). There was no significant association between ocular and systemic risk factors detected to predict progression of PMF, but this was largely due to low study power, because only 10 eyes of 62 progressed during the 5-year follow-up. Table 2. Incidence of Idiopathic and Secondary Epiretinal Membranes in Participants with No Preexisting Lesion at Baseline Type of Epiretinal Membrane
Cases
Incidence % (95% Confidence Interval)
Idiopathic epiretinal membranes (n ⫽ 1851) PMF 29 1.6 (1.1–2.3) CMR 69 3.7 (2.9–4.7) Any 97* 5.2 (4.2–6.3) Secondary epiretinal membranes (n ⫽ 179) PMF 2 1.1 (0.1–4.0) CMR 8 4.5 (2.0–8.7) Any 10 5.6 (2.7–10.1) *One participant developed CMR in one eye and PMF in the other eye. CMR ⫽ cellophane macular reflex; PMF ⫽ preretinal macular fibrosis.
38
Regression analyses, adjusting for age and gender, was used to assess whether the finding of an epiretinal membrane at baseline predicted incident visual impairment after 5 years. This analysis found no significant increase in the proportion of the 62 eyes with PMF at BMES I that developed incident visual impairment ⬍20/40 compared with eyes without PMF, OR 0.4 (CI, 0.1–1.6).
Discussion Although the prevalence of epiretinal membranes has been described in two recent population-based reports,19,20 no population data have yet described the incidence and course of this relatively common lesion over time. Although most epiretinal membranes do not cause significant visual disability, some cases lead to progressive visual loss and may require surgical intervention. Cumulative incidence data on the development and progression of this common lesion, together with effects on visual acuity, are provided by this study. The report has also examined ocular and systemic associations for incident epiretinal membranes. The 5-year incidence of epiretinal membranes in this older population sample was 5.3%. The cumulative incidence of PMF was less than half the CMR cumulative incidence (1.5% vs. 3.8%), which corresponds closely with findings reported from both BMES I20 and Beaver Dam.19 These studies reported PMF:CMR prevalences of 2.2%: 4.8% from BMES I and corresponding rates of 2.8%:9.0% for Beaver Dam. The age-related increase in epiretinal membrane inciTable 4. Change in Visual Acuity in Any Eyes with Incident Preretinal Macular Fibrosis (n ⫽ 56)
5-Year Change in Visual Acuity
Eyes with Incident Preretinal Macular Fibrosis (n ⴝ 56)
No change or improved by ⬍5 letters Decreased by 1–4 letters Decreased by 5–9 letters Decreased by 10–19 letters Decreased by 20 or more letters
14 18 16 3 5
Fraser-Bell et al 䡠 Five-Year Incidence and Progression of Epiretinal Membranes Table 5. Change Observed in Eyes with Preexisting Epiretinal Membranes during the 5 Year Period Type of Epiretinal Membrane at Baseline
Change over 5-year Period Progressed Regressed Stable CMR to PMF*
Preretinal Macular Fibrosis (n ⫽ 62 eyes)
Cellophane Macular Reflex (n ⫽ 183 eyes)
10 (16.1%) 16 (25.8%) 36 (58.1%)
60 (32.8%) 47 (25.7%) 59 (32.2%) 17 (9.3%)
Any (n ⫽ 245 eyes) 70 (28.6%) 63 (25.7%) 95 (38.8%)
*CMR preexisting cases only. CMR ⫽ cellophane macular reflex. PMF ⫽ preretinal macular fibrosis.
dence observed for persons aged less than 70 years corresponded with the increase in prevalence up to this age found in both BMES I20 and in Beaver Dam.19 We previously argued that the lower prevalence rate found for older age groups could represent potential underreporting as a result of increasing lens opacity.20 The corresponding decrease in incidence found in this study is consistent with this argument. We further observed no statistically significant gender difference in epiretinal membrane incidence, consistent with our prevalence report.20 This study also addresses the bilaterality of epiretinal membranes, found in BMES I to be 31% and in the Beaver Dam Eye Study to be 19.5%.19,20 Our incidence data suggest a relatively similar rate; the second eye incidence was more than double that in the first eye, 13.5% versus 5.3%, and second eyes were more likely than first eyes to develop PMF. PMF incidence in second eyes was four times that for first eyes (6% vs. 1.5%), and CMR second eye incidence was twice that for first eyes (7.5% vs. 3.8%). These data suggest a symmetric trend for this lesion, as seen with many other age-related conditions such as age-related maculopathy.26 Its major precursor, posterior vitreous detachment, is also frequently bilateral. Our study identified a statistically significant association between cataract surgery and incident epiretinal membranes, particularly the less severe form, CMR. The cumulative incidence of epiretinal membranes among participants who had cataract surgery during the 5-year period was almost double the nonsurgical group incidence. It is possible that, in some cases, these epiretinal membranes were present before surgery, but were not detected in the preoperative period because they were obscured by the lens opacity. The timing of incident epiretinal membranes in relation to cataract surgery during the follow-up period is also not known. Nevertheless, our finding of an association between cataract surgery and incident membranes supports findings from a number of previous reports,1,5,19 including our previous study.20 We are aware of only one report documenting the incidence of epiretinal membranes after cataract surgery by Jahn et al.5 In this prospective study of uneventful extracapsular cataract extraction with posterior chamber intraocular lens implantation in 332 eyes, 10% of eyes
developed new epiretinal membranes over a 6-month postoperative period. This is very close to our rate of 9.1% among subjects after cataract surgery. Although the two previous prevalence studies reported an association between diabetes and epiretinal membranes,19,20 we were not able to demonstrate this in our incidence data. In contrast with our prevalence study, epiretinal membrane incidence was not substantially different between eyes with and without secondary causes (5.6% vs. 5.2%), possibly because of the relatively small number of incident cases or because of age effects. Our study demonstrated that in most epiretinal membrane cases, no progression occurred over 5 years, a finding consistent with many previous clinic-based reports.11,17 Indeed, one in four cases actually demonstrated regression or even disappearance of the epiretinal membrane over this period, whereas only around 1 in 10 cases of CMR progressed to PMF. A limitation of our study is that only a relatively small number of subjects exhibited progression of PMF (only 10 of 62 baseline cases). Furthermore, larger studies will be needed to assess factors associated with progression. Previous studies have indicated that less than 15% of patients with epiretinal membranes have visual acuities worse than 20/70.8,11 One study reported no further decrease in visual acuity in 87% of 47 eyes with PMF followed for 2 to 4 years.11 Our study demonstrated a small reduction in mean visual acuity over the follow-up period in eyes with incident PMF compared with eyes not developing epiretinal membranes, but this was not found in eyes with incident CMR. Although we found no significant worsening of vision in eyes demonstrated to have progression of PMF, the power of our study to examine this was low, because only 16.1% of these epiretinal membranes progressed during follow-up. In conclusion, the current report provides useful information regarding the 5-year incidence and natural history of epiretinal membranes in an older population. The more severe type (PMF) was found to be relatively stable over this period, and only few cases developed incident visual impairment. This relative stability of epiretinal membranes and their limited effect on vision should be taken into account when considering surgical intervention with its potential for complications.
References 1. Appiah AP, Hirose T. Secondary causes of premacular fibrosis. Ophthalmology 1989;96:389 –92. 2. Margherio RM. Epiretinal macular membranes. In: Albert DM, Jakobiec FA, eds. Principles and Practice of Ophthalmology: Clinical Practice. Philadelphia: W.B. Saunders, 1994; v. 2, chap. 74. 3. Lobes LA Jr, Burton TC. The incidence of macular pucker after retinal detachment surgery. Am J Ophthalmol 1978;85: 72–7. 4. Kraushar MF, Morse PH. The relationship between retina surgery and preretinal macular fibrosis. Ophthalmic Surg 1988;19:843– 8. 5. Jahn CE, Minich V, Moldaschel S, et al. Epiretinal membranes
39
Ophthalmology Volume 110, Number 1, January 2003
6. 7. 8. 9.
10. 11. 12. 13. 14. 15. 16.
after extracapsular cataract surgery. J Cataract Refract Surg 2001;27:753– 60. Wise GN. Relationship of idiopathic preretinal macular fibrosis to posterior vitreous detachment. Am J Ophthalmol 1975; 79:358 – 62. Wiznia RA. Posterior vitreous detachment and idiopathic preretinal macular gliosis. Am J Ophthalmol 1986;102:196 – 8. Hirokawa H, Jalkh AE, Takahashi M, et al. Role of the vitreous in idiopathic preretinal macular fibrosis. Am J Ophthalmol 1986;101:166 –9. Smiddy WE, Maguire AM, Green WR, et al. Idiopathic epiretinal membranes. Ultrastructural characteristics and clinicopathologic correlation. Ophthalmology 1989;96:811–20; discussion 821. Tanenbaum HL, Schepens CL, Elzeneiny I, Freeman HM. Macular pucker following retinal detachment surgery. Arch Ophthalmol 1970;83:286 –93. Wiznia RA. Natural history of idiopathic preretinal macular fibrosis. Ann Ophthalmol 1982;14:876 – 8. Appiah AP, Hirose T, Kado M. A review of 324 cases of idiopathic premacular gliosis. Am J Ophthalmol 1988;106: 533–5. Sponsel WE, Kaufman PL, Blum FG Jr. Association of retinal capillary perfusion with visual status during chronic glaucoma therapy. Ophthalmology 1997;104:1026 –32. Ciulla TA, Pesavento RD. Epiretinal fibrosis. Ophthalmic Surg Lasers 1997;28:670 –9. Pearlstone AD. The incidence of idiopathic preretinal macular gliosis. Ann Ophthalmol 1985;17:378 – 80. Pournaras CJ, Donati G, Brazitikos PD, et al. Macular epiretinal membranes. Semin Ophthalmol 2000;15:100 –7.
17. Wise GN. Clinical features of idiopathic preretinal macular fibrosis. Schoenberg Lecture. Am J Ophthalmol 1975;79:349 –7. 18. Roth AM, Foos RY. Surface wrinkling retinopathy in eyes enucleated at autopsy. Trans Am Acad Ophthalmol Otolaryngol 1971;75:1047–58. 19. Klein R, Klein BEK, Wang Q, Moss SE. The epidemiology of epiretinal membranes. Trans Am Ophthalmol Soc 1994;92: 403–25; discussion 425–30. 20. Mitchell P, Smith W, Chey T, et al. Prevalence and associations of epiretinal membranes. The Blue Mountains Eye Study, Australia. Ophthalmology 1997;104:1033– 40. 21. Mitchell P, Smith W, Attebo K, Wang JJ. Prevalence of age-related maculopathy in Australia. The Blue Mountains Eye Study. Ophthalmology 1995;102:1450 – 60. 22. Attebo K, Mitchell P, Smith W. Visual acuity and the causes of visual loss in Australia. The Blue Mountains Eye Study. Ophthalmology 1996;103:357– 64. 23. Klein R, Davis MD, Magli YL, et al. The Wisconsin agerelated maculopathy grading system. Ophthalmology 1991;98: 1128 –34. 24. Smith W, Mitchell P, Wang JJ. Gender, oestrogen, hormone replacement and age-related macular degeneration: results from the Blue Mountains Eye Study. Aust N Z J Ophthalmol 1997;25(suppl 1):S13–5. 25. Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a generalized estimating equation approach [published erratum appears in Biometrics 1989;45:347]. Biometrics 1988;44: 1049 – 60. 26. Wang JJ, Mitchell P, Smith W, Cumming RG. Bilateral involvement by age related maculopathy lesions in a population. Br J Ophthalmol 1998;82:743–7.
AOS-KNAPP FUND FELLOWSHIP FUNDING The Knapp Fund, a supporting organization of the American Ophthalmological Society, provides funding for a second or third year of postgraduate study in ophthalmology. An annual stipend of $20,000 is available for postgraduate study beginning on or after July 1, 2003. To be eligible, applicants must be United States or Canadian citizens, the fellowship training must be conducted in the United States and applicants must have completed a residency program in ophthalmology accredited by the ACGME or the Royal College of Physicians and Surgeons of Canada. Deadline for receipt of applications is March 15, 2003. Froncie A. Gutman, M.D. Cleveland Clinic Foundation 9500 Euclid Avenue, Desk i-32 Cleveland, OH 44195 216-444-8968 (fax)
40