- Email: [email protected]
Fundus Photographic and Fluorescein Angiographic Characteristics of Pseudoholes of the Macula in Eyes with Epiretinal Membranes
Fundus Photographic and Fluorescein Angiographic Characteristics of Pseudoholes of the Macula in Eyes with Epiretinal Membranes Betty R. Klein, MD, Cheryl]. Hiner, Bert M. Glaser, MD, Robert P. Murphy, MD, Raymond N. Sjaarda, MD, John T. Thompson, MD
Background: The fluorescein angiographic characteristics in eyes with pseudoholes of the macula associated with epiretinal membranes have not been studied extensively. Methods: Stereo photographs and fluorescein angiograms from 83 consecutive eyes of 80 patients with pseudoholes of the macula were evaluated by two independent graders for epiretinal membrane opacity, fluorescence in the base of the pseudohole, and late perifoveal pooling of dye. Results: Hyperfluorescence in synchrony with choroidal fluorescence appeared within the base of the pseudohole in 52 (63%) of the 83 eyes studied. The hyperfluorescence was smaller than the pseudohole in 37 (45%) eyes. Diffuse hyperfluorescence filled the pseudohole in 15 (18%) eyes. No fluorescence was seen in 20 (24%) eyes. Eleven (13%) eyes could not be graded due to coexistent macular disease or media opacity. Fluorescence within the area of pseudohole was less common in eyes with opaque epiretinal membranes than in eyes with visible or transparent epiretinal membranes (P = 0.002). Fluorescence within the area of the pseudohole was also less common in eyes with evidence of macular edema on fluorescein angiography (P < 0.001). The mean visual acuity was better for eyes with hyperfluorescence within the area of the pseudohole than for eyes without hyperfluorescence (P < 0.01). Conclusion: A common fluorescein angiographic characteristic associated with pseudoholes of the macula is early hyperfluorescence within the area of the pseudohole. This hyperfluorescence coincides with choroidal filling and appears to be a form of transmission defect rather than a blocking of surrounding choroidal fluorescence by the epiretinal membrane. This central hyperfluorescence may result in misdiagnosis of the macular pseudohole as a full-thickness macular hole. Ophthalmology 1995;102:768-774
Originally received: April 4, 1994. Revision accepted: November 23, 1994. From The Retina Institute of Maryland, Baltimore. Presented as a poster at the American Academy of Ophthalmology Annual Meeting, Dallas, November 1992. Reprint requests to Bert M. Glaser, MD, The Retina Institute of Maryland, 7505 Osler Dr, Suite 103, Baltimore, MD 21204.
768
Fluorescein angiography often is used to evaluate eyes with macular holes and pseudo holes. The largest fluorescein angiographic study of pseudo holes of the macula in eyes with epiretinal membranes in the literature has been limited to 11 cases.F A larger systematic study might better characterize and explain the fluorescein angiographic changes in eyes with pseudoholes of the macula and allow comparison with the angiographic character-
Klein et al . Pseudoholes of the Macula and Epiretinal Membranes photographs and fluorescein angiograms of the eyes with pseudoholes were graded for membrane opacity, hyperfluorescence within the base of the pseudohole, and late pooling of dye in the perifoveal macula. After establishing definitions for these characteristics, all photographs and angiograms were graded by two independent graders (BRK and CJH) using the criteria described below. These data were compared after all angiograms were graded. Differences were adjudicated by both graders.
Grading Criteria for Fluorescence of Pseudohole
Figure 1. C omplete hyperfluorcscen ce of base of pseudohole. The h yperfluore scence was graded as complete if it appeared to fillthe pseudohole in the early phases of the angiogram.
istics of macular holes. We report the fluorescein angiographic findings in 83 eyes from 80 patients with pseudoholes of the macula. The presence and pattern of fluorescence in the base ofthe pseudohole were determined by grading fluorescein angiograms from 83 eyes. Previous reports have suggested that apparent fluorescence in the base of pseudoholes reflects a screening effect by surrounding epiretinal membrane. ' We pro vide data suggesting that the hyperfluorescence represents a true transmission defect in the base ofsome pseudo holes. To test our hypothesis , we compared hyperfluorescence of the base of the pseudohole in eyes with varying degrees of epiretinal membrane opacity. We also compared -fluorescence of the pseudohole in eyes with and without late perifoveal pooling of dye. If the hyperfluorescence were a form of screening effect, we would expect it to be emphasized in eyes with more opaque epiretinal membranes and independent of macular thickening as one would see when there is perifoveal pooling of dye.
Early and mid phases of the fluorescein angiogram negatives were compared with color stereo fundus photographs to grade location and extent of fluorescence in the base of the pseudohole. The following definitions were used for the purpose of grading. Complete Hyperfluorescence. Diffusive hyperfluorescence matching the size, location, and shape of the base of the pseudohole (Fig I). Patchy Hyperfluorescence. Uneven hyperfluorescence within and smaller than the base of the pseudohole (Fig 2). No Hyperfluorescence. Those eyes with no evidence of fluorescence in the base of the pseudohole (Fig 3). Cannot Be Graded. Eyes with defects or attenuation of macular pigmentation from coexistent macular disease were considered ungradable.
Grading Criteria for Epiretinal Membrane Opacity Epiretinal membrane opacity was graded from stereo color photographic pairs . The membranes were described as
Materials and Methods We reviewed the charts of all new patients referred with macular epiretinal membranes between March 1, 1989, and February 24, 1992. A total of 415 patients were identified. Of these 415 patients, 82 (20%) had a pseudohole of the macula documented on clinical and biomicroscopic examination. The referring diagnosis was noted for all eyes found to have pseudoholes clinically . Two patients with pseudoholes were excluded from the fluorescein angiogram portion of the study because fluorescein angiograms were not available on those patients. Three patients had bilateral macular pseudoholes. Therefore, a total of 83 fluorescein angiograms were available for review. Stereo
Figure 2. Patch y hyperfluorescence of base of pseudohole. The hyperfluorescence was graded as patchy if there were both hyperfluorescence and hypoflu orescent areas within the pseudohole during early phases of the an giogram.
769
Ophthalmology
Volume 102, Number 5, May 1995 opaque, visible, or transparent according to the following definitions. Opaque. Some part of the epiretinal membrane obscured underlying retinal vessels or retinal pigment epithelium (Fig 4). Visible. Some part of the epiretinal membrane could be seen on the photographs but was not obscuring retinal vascular or retinal pigment epithelial features (Fig 5). Transparent. Membranes evidenced by retinal striae or vessel distortion but not themselves visible (Fig 6).
Grading Criteria for Perifoveal Pooling of Fluorescein
Figure 3. No hyperfluorescence of base of pseudohole. This grade was applied if the pseudohole centered on the fovea was hypofluorescent during early phases of the angiogram.
Leakage of dye from retinal capillaries into the perifoveal neurosensory retina was graded from the late phase of the fluorescein angiogram (Fig 7). Any staining of the perifoveal retina was considered positive or pooling present. Total absence oflate staining was graded as absent.
Correlation with Visual Acuity Snellen visual acuity was converted to log scale and mean visual acuities for eyes with patchy, complete, and no pseudohole fluorescence were calculated. The three means were compared using analysis of variance and Bonferroni post tests.
Top left, Figure 4. Opaque epiretinal membrane. A membrane was graded as opaque if it obscured some of the retinal vascular or pigment epithelial details on the color photographs. Top right, Figure 5. Visible epiretinal membrane. A membrane was graded as visible if it could be seen but was not obscuring retinal or retinal pigment epithelial detail on the color photographs. Bottom, Figure 6. Transparent epiretinal membrane. A membrane was graded as transparent if it was apparent on the color photographs only by the presence of retinal striae, pseudohole formation, and vessel distortion.
770
Klein et al . Pseudoholes of the Macula and Epiretinal Membranes Mac Ho le (8)
Im•• ~
V
Lamellar Hole (6) Other (6)
(10)
-~!@~~~r
None 15)
ERM (43)
Mac Cyst (4)
Figure 8. Referring diagnosis. This information was obtained from cha rt review. All patients seen with macular pseudohole on clinical examination betwee n March 1, 1989, and February 24, 1992 (n = 82), were included in this aspect of the study. ERM = ep iret inal membrane.
Figure 7. Late perifoveal pooling of fluor escein dye or angiographic macular edema. This was graded from late angiogram frames.
Epiretinal Membrane Opacity and Pseudohole Fluorescence
Results Referring
for pattern of pseudohole fluorescence between the two graders.
Dia~nosis
The referring diagnosis was available in 77 (94%) of the 82 patients with macular pseudoholes (Fig 8). Of the 82 patients, 18 (22%) were referred with a diagnosis of m~c ular hole or impending macular hole and 10 (12%) with a diagnosis of lamellar hole or macular cyst. The correct referring diagnosis of epiretinal membrane or epiretinal membrane with pseudohole was given in 43 (52%) patients . The referral was for an unrelated problem in six (7%) patients. Two patients whose fluorescein angiograms were unavailable are included in the analysis of the referring diagnosis because the diagnosis of macular pseudohole was made by clinical examination with fundus biomicroscopy.
The epiretinal membrane was graded as opaque in 20 (24%) eyes, visible but not obscuring retina features in 38 (46%) eyes, and transparent in 16 (19%)eyes (Figs 10 and II). The opacity of the epiretinal membrane could not be graded from the stereo color photographs in nine (I 1%) eyes. In eyes with opaque epiretinal membranes, II (55%) had no fluorescence in the base of the pseudohole and 8 (40%) had either patch y or complete fluorescence. In eyes with visible and transparent epiretinal membranes, six (16%) and two (13%), respectively, had no fluorescence in the base of the pseudohol e, whereas 27 (71%) and 12 (75%), respectively, had either patch y or complete fluorescence patterns. Hyperfluorescence within the area of the pseudohole was less common in eyes with opaque epiretinal membranes than in eyes with visible or transPatchy (37)
Fluorescein Angiographic Findings Fluorescein angiograms were available for review in 83 eyes with macular pseudoholes (Fig 9). Fluorescence which appeared and faded in synchrony with choroidal fluorescence was noted in the base of 52 (63%) of the pseudoholes. Among these 52 eyes, the pattern of fluorescence was patchy and smaller than the total area of the pseudo hole in 37 (45%) eyes. A "complete" pattern with hyperfluorescence filling the base of the pseudohole was seen in 15 (18%) eyes. The base of the pseudo hole transmitted no fluorescence in 20 (24%) eyes. In one (1%) eye, the pseudohole fluorescence was graded as questionable. In five (6%) eyes, there was atrophic macular degeneration or healed retinochoroiditis. which confounded evaluation of hyperfluorescence in the area of the pseudohole. Five (6%) eyes could not be graded because of inadequate photographs or media opacity. There was an 83% agreement
45%
Can't Determine (11 )
None (20)
Complete (15) Figure 9. Pseudohole fluorescence pattern. This was graded from early ph ases of th e fluorescein angiogram. The fluorescence appeared to be a form of tr ansmission defect as it coincided in timing and intensity to choroidal filling and decreased with ch or oidal emptying.
771
Ophthalmology
Volume 102, Number 5, May 1995
Visible
Present
1381
1261 31%
en't Determine
141
Cen't Grede
64%
191 Opeque
Absent
1201
(531
Trensperent
Figure 12. Late macular pooling of fluorescein in eyes with pseudoholes. This was graded from late angiogram frames.
1161 N
= 83 eyes
Figure 10. Epiretinal membrane opacity in eyes with pseudoholes. This was graded from stereo color photographs.
parent epiretinal membranes (P = 0.002). There was an 80% interobserver agreement for epiretinal membrane opacity before adjudication of any differences.
Late Macular Pooling of Dye and Pseudohole Fluorescence Macular pooling of fluorescein dye was present in 26 (31%) eyes, absent in 53 (64%) eyes, graded as questionable in 1 (1%) eye, and could not be graded in 3 (4%) eyes (Figs 12 and 13). In eyes with pooling of dye, 13 (50%) had no fluorescence in the base of the pseudohole and 10 (39%) showed either patchy or complete patterns of pseudohole fluorescence. In eyes with no pooling of dye, 7 (13%) had no fluorescence in the base of the pseudohole and 41 (78%) showed either patchy or complete patterns of hyperfluorescence. Hyperfluorescence in the base of the pseudohole was less common in eyes with fluorescein pooling in the macula (P < 0.001). Interobserver agreement for macular pooling of fluorescein dye was 98%.
Pseudohole Fluorescence and Visual Acuity Visual acuities were averaged using conversion to log scale and compared using analysis of variance and Bonferroni adjusted P values (Fig 14). The mean visual acuity for
eyes with no hyperfluorescence in the base of the pseudohole was 20/70. The mean visual acuity for eyes with patchy hyperfluorescence was 20/40 and for eyes with complete hyperfluorescence was 20/30. The mean visual acuity was significantly lower for eyes with no hyperfluorescence in the base of the pseudo hole than for eyes with patchy (P < 0.01) or complete (P < 0.01) hyperfluorescence.
Discussion Pseudoholes of the macula are well-demarcated round or oval depressions usually found in the fovea in eyes with macular epiretinal membranes. Allen and Gass' and Gass' described this as a central defect in the epiretinal membrane which creates a steep drop-off around the fovea, simulating a full-thickness macular hole. Pseudoholes of the macula sometimes arouse suspicion of full-thickness or impending macular holes, as was seen in 22% of the eyes in this study and 29% in a report by Fish et a1. 2 In 1990, Gass and Joondeph" reported that 4 (22%) of 18 patients referred to Bascom Palmer Institute for suspected impending macular holes had instead epiretinal membranes with pseudoholes of the macula. Because both prognosis and management differ among these conditions, specific reliable diagnostic criteria are needed. In 1976, Allen and Gass' reported faint central hyperfluorescence in three of four eyes with pseudoholes and
Patchy
Patchy
C.O. Complete None
ERM Opaque
772
None
ERM Visible
Complete
None
ERM Transparent
Figure 11. Pseudohole fluorescence versus epiretinal membrane opacity. Hyperfluorescence within the base of the pseudohole was less common in eyes with opaque epiretinal membranes (ERM). CD = cannot determine.
Klein et al . Pseudoholes of the Macula and Epiretinal Membranes
Patchy
Complete (2)
Can't Determine (3)
(28)
atchy
Figure 13. Pseudohole fluorescence versus late macular pooling of dye. Hyperfluorescence in the base of the pseudohole was less common in eyes with angiographic macular edema.
(8)
None (7)
Can't Determine
epiretinal membranes. They attributed the central fluorescence to a screening effect by the semitransparent perifoveal membrane. I.3 Variable fluorescence was seen in 52 (63%) of the 83 macular pseudo holes in this study. The fluorescence was complete and, thus, most likely to be confused with a full-thickness macular holev? or impending macular hole 7 in 15 (18%) eyes. Fluorescence in the base of the pseudohole was more common in eyes with transparent or barely visible epiretinal membranes than in eyes with opaque membranes. This may suggest true transmission hyperfluorescence rather than a screening effect from overlying membrane. One would expect a screening effect to be emphasized in eyes with pseudoholes within more opaque epiretinal membranes. The waxing and waning of hyperfluorescence during the early and middle phases of the angiogram as observed in this study is a pattern frequently seen in cases of transmission hyperfluorescence. The basis for this transmission fluorescence may be local attenuation or flattening of the foveal xanthophyll or retinal pigment epithelium. We think the former is more likely because eyes with visible changes in the retinal pigment epithelium were eliminated from grading as having coexistent macular disease. Mean VA(201y)
80
70 60 50 40 30 20 10
---L.
None
...L-
(13)
(5)
Pooling Absent
OIL.-
None
----.-/
Patchy Pseudohole Fluorescence
Figure 14. Pseudohole fluorescence versus visual acuity (VA). Mean visual acuity was lower for eyes with no hyperfluorescence in the base of the pseudohole.
Pooling Present Pseudohole hyperfluorescence during early angiogram frames was less prevalent in eyes with late perifoveal pooling of fluorescein dye and with lower visual acuities. This may represent reduced transmission fluorescence in the presence of macular thickening in those eyes with pooling of dye in the perifoveal neurosensory retina. A report by Fish et a1 2 described window defects in only 3 of the 11 (27%)eyes with pseudoholes as compared with 63% in the current series. If standard photographs and definitions were used for evaluating fluorescein angiograms in their study, we did not find them in the report and so are unable to explain their lower incidence of window defects. We noted pseudoholes in a higher percentage (20%) of eyes with epiretinal membrane than has been reported in previous studies. Sidd et a1 8 reported an incidence of9.4% and Margherio et a1 9 8% after review of 74 and 275 eyes with epiretinal membranes, respectively. Our institution has been investigating the use of vitrectomy and transforming growth factor-In for full-thickness macular holes and thus receives many referrals of known or suspected macular holes. 10 This unusual situation may account for this difference. The results of this study do not establish the anatomic basis for hyperfluorescence of the base of pseudoholes in eyes with macular epiretinal membranes. When present, the pseudohole fluorescence behaved as a transmission defect appearing early and fading late. The fluorescence, therefore, most likely reflects an attenuation of one of the pigmented layers in the fovea, either xanthophyll or retinal pigment epithelium. Because these pseudoholes were associated with fibro-glial proliferation on the inner surface of the macula, attenuation of xanthophyll by flattening or stretching of the neurosensory retina seems a reasonable explanation for the hyperfluorescence. This hyperfluorescence may result in confusion of a macular hole with a pseudohole if the patient is not examined carefully by slit-lamp biomicroscopy.
Conclusions Pseudoholes of the macula appear concave on biomicroscopy and are sometimes confused with full-thickness
773
Ophthalmology
Volume 102, Number 5, May 1995
macular holes. Hyperfluorescence of the base of the pseudohole presented as a variable and often incomplete transmission defect in 63% of eyes studied; however, the entire base of the pseudohole was hyperfluorescent in only 18% of eyes studied. This alteration may represent local attenuation of foveal xanthophyll and/or thinning of the foveal neurosensory retina, resulting from centripetal traction forces within the epiretinal membrane.
References 1. Allen AW Jr, Gass JOM. Contraction of a perifoveal epiretinal membrane simulating a macular hole. Am J Ophthalmol 1986;82:684-91. 2. Fish RH, Anand R, Izbrand OJ. Macular pseudoholes. Clinical features and accuracy ofdiagnosis. Ophthalmology 1992;99: 1665-70. 3. Gass JDM. Stereoscopic Atlas of Macular Diseases: Diagnosis and Treatment. 3rd ed. Vol. 1. St. Louis: CV Mosby, 1987;700-3. 4. Gass JDM, Joondeph BC Observations concerning patients with suspected impending macular holes. Am J Ophthalmol 1990; 109:638-46.
774
5. Gass JDM. Lamellar macular holes: a complication of cystoid macular edema after cataract extraction: a clinicopathologic case report. Trans Am Ophthalmol Soc 1975;73:23150. 6. Gass JDM. Idiopathic senile macular hole. Its early stages and pathogenesis. Arch Ophthalmol 1988;106:629-39. 7. Akiba J, Yoshida A, Trempe CL. Risk of developing a macular hole. Arch Ophthalmol 1990;108:1088-90. 8. Sidd RJ, Fine SL, Owens 8L, Patz A. Idiopathic preretinal gliosis. Am J Ophthalmol 1982;94:44-8. 9. Margherio RR, Cox MS Jr, Trese MT, et al. Removal of epimacular membranes. Ophthalmology 1985;92: 1075-83. 10. Glaser BM, Michels RG, Kuppermann BD, et al. Transforming growth factor-S, for the treatment of full-thickness macular holes. A prospective randomized study. Ophthalmology 1992;99: 1162-73. II. Margherio RR. Discussion, 1387-8 Of: Michels RG. Vitrectomy for macular pucker. Ophthalmology 1984;91: 1384-8. 12. Appiah AP, Hirose T, Kado M. A review of 324 cases of idiopathic premacular gliosis. Am J Ophthalmol 1988;106: 533-5. 13. Guyer DR, Green WR, de Bustros S, Fine SL. Histopathologic features of idiopathic macular holes and cysts. Ophthalmology 1990;97: 1045-51.
Background: The fluorescein angiographic characteristics in eyes with pseudoholes of the macula associated with epiretinal membranes have not been studied extensively. Methods: Stereo photographs and fluorescein angiograms from 83 consecutive eyes of 80 patients with pseudoholes of the macula were evaluated by two independent graders for epiretinal membrane opacity, fluorescence in the base of the pseudohole, and late perifoveal pooling of dye. Results: Hyperfluorescence in synchrony with choroidal fluorescence appeared within the base of the pseudohole in 52 (63%) of the 83 eyes studied. The hyperfluorescence was smaller than the pseudohole in 37 (45%) eyes. Diffuse hyperfluorescence filled the pseudohole in 15 (18%) eyes. No fluorescence was seen in 20 (24%) eyes. Eleven (13%) eyes could not be graded due to coexistent macular disease or media opacity. Fluorescence within the area of pseudohole was less common in eyes with opaque epiretinal membranes than in eyes with visible or transparent epiretinal membranes (P = 0.002). Fluorescence within the area of the pseudohole was also less common in eyes with evidence of macular edema on fluorescein angiography (P < 0.001). The mean visual acuity was better for eyes with hyperfluorescence within the area of the pseudohole than for eyes without hyperfluorescence (P < 0.01). Conclusion: A common fluorescein angiographic characteristic associated with pseudoholes of the macula is early hyperfluorescence within the area of the pseudohole. This hyperfluorescence coincides with choroidal filling and appears to be a form of transmission defect rather than a blocking of surrounding choroidal fluorescence by the epiretinal membrane. This central hyperfluorescence may result in misdiagnosis of the macular pseudohole as a full-thickness macular hole. Ophthalmology 1995;102:768-774
Originally received: April 4, 1994. Revision accepted: November 23, 1994. From The Retina Institute of Maryland, Baltimore. Presented as a poster at the American Academy of Ophthalmology Annual Meeting, Dallas, November 1992. Reprint requests to Bert M. Glaser, MD, The Retina Institute of Maryland, 7505 Osler Dr, Suite 103, Baltimore, MD 21204.
768
Fluorescein angiography often is used to evaluate eyes with macular holes and pseudo holes. The largest fluorescein angiographic study of pseudo holes of the macula in eyes with epiretinal membranes in the literature has been limited to 11 cases.F A larger systematic study might better characterize and explain the fluorescein angiographic changes in eyes with pseudoholes of the macula and allow comparison with the angiographic character-
Klein et al . Pseudoholes of the Macula and Epiretinal Membranes photographs and fluorescein angiograms of the eyes with pseudoholes were graded for membrane opacity, hyperfluorescence within the base of the pseudohole, and late pooling of dye in the perifoveal macula. After establishing definitions for these characteristics, all photographs and angiograms were graded by two independent graders (BRK and CJH) using the criteria described below. These data were compared after all angiograms were graded. Differences were adjudicated by both graders.
Grading Criteria for Fluorescence of Pseudohole
Figure 1. C omplete hyperfluorcscen ce of base of pseudohole. The h yperfluore scence was graded as complete if it appeared to fillthe pseudohole in the early phases of the angiogram.
istics of macular holes. We report the fluorescein angiographic findings in 83 eyes from 80 patients with pseudoholes of the macula. The presence and pattern of fluorescence in the base ofthe pseudohole were determined by grading fluorescein angiograms from 83 eyes. Previous reports have suggested that apparent fluorescence in the base of pseudoholes reflects a screening effect by surrounding epiretinal membrane. ' We pro vide data suggesting that the hyperfluorescence represents a true transmission defect in the base ofsome pseudo holes. To test our hypothesis , we compared hyperfluorescence of the base of the pseudohole in eyes with varying degrees of epiretinal membrane opacity. We also compared -fluorescence of the pseudohole in eyes with and without late perifoveal pooling of dye. If the hyperfluorescence were a form of screening effect, we would expect it to be emphasized in eyes with more opaque epiretinal membranes and independent of macular thickening as one would see when there is perifoveal pooling of dye.
Early and mid phases of the fluorescein angiogram negatives were compared with color stereo fundus photographs to grade location and extent of fluorescence in the base of the pseudohole. The following definitions were used for the purpose of grading. Complete Hyperfluorescence. Diffusive hyperfluorescence matching the size, location, and shape of the base of the pseudohole (Fig I). Patchy Hyperfluorescence. Uneven hyperfluorescence within and smaller than the base of the pseudohole (Fig 2). No Hyperfluorescence. Those eyes with no evidence of fluorescence in the base of the pseudohole (Fig 3). Cannot Be Graded. Eyes with defects or attenuation of macular pigmentation from coexistent macular disease were considered ungradable.
Grading Criteria for Epiretinal Membrane Opacity Epiretinal membrane opacity was graded from stereo color photographic pairs . The membranes were described as
Materials and Methods We reviewed the charts of all new patients referred with macular epiretinal membranes between March 1, 1989, and February 24, 1992. A total of 415 patients were identified. Of these 415 patients, 82 (20%) had a pseudohole of the macula documented on clinical and biomicroscopic examination. The referring diagnosis was noted for all eyes found to have pseudoholes clinically . Two patients with pseudoholes were excluded from the fluorescein angiogram portion of the study because fluorescein angiograms were not available on those patients. Three patients had bilateral macular pseudoholes. Therefore, a total of 83 fluorescein angiograms were available for review. Stereo
Figure 2. Patch y hyperfluorescence of base of pseudohole. The hyperfluorescence was graded as patchy if there were both hyperfluorescence and hypoflu orescent areas within the pseudohole during early phases of the an giogram.
769
Ophthalmology
Volume 102, Number 5, May 1995 opaque, visible, or transparent according to the following definitions. Opaque. Some part of the epiretinal membrane obscured underlying retinal vessels or retinal pigment epithelium (Fig 4). Visible. Some part of the epiretinal membrane could be seen on the photographs but was not obscuring retinal vascular or retinal pigment epithelial features (Fig 5). Transparent. Membranes evidenced by retinal striae or vessel distortion but not themselves visible (Fig 6).
Grading Criteria for Perifoveal Pooling of Fluorescein
Figure 3. No hyperfluorescence of base of pseudohole. This grade was applied if the pseudohole centered on the fovea was hypofluorescent during early phases of the angiogram.
Leakage of dye from retinal capillaries into the perifoveal neurosensory retina was graded from the late phase of the fluorescein angiogram (Fig 7). Any staining of the perifoveal retina was considered positive or pooling present. Total absence oflate staining was graded as absent.
Correlation with Visual Acuity Snellen visual acuity was converted to log scale and mean visual acuities for eyes with patchy, complete, and no pseudohole fluorescence were calculated. The three means were compared using analysis of variance and Bonferroni post tests.
Top left, Figure 4. Opaque epiretinal membrane. A membrane was graded as opaque if it obscured some of the retinal vascular or pigment epithelial details on the color photographs. Top right, Figure 5. Visible epiretinal membrane. A membrane was graded as visible if it could be seen but was not obscuring retinal or retinal pigment epithelial detail on the color photographs. Bottom, Figure 6. Transparent epiretinal membrane. A membrane was graded as transparent if it was apparent on the color photographs only by the presence of retinal striae, pseudohole formation, and vessel distortion.
770
Klein et al . Pseudoholes of the Macula and Epiretinal Membranes Mac Ho le (8)
Im•• ~
V
Lamellar Hole (6) Other (6)
(10)
-~!@~~~r
None 15)
ERM (43)
Mac Cyst (4)
Figure 8. Referring diagnosis. This information was obtained from cha rt review. All patients seen with macular pseudohole on clinical examination betwee n March 1, 1989, and February 24, 1992 (n = 82), were included in this aspect of the study. ERM = ep iret inal membrane.
Figure 7. Late perifoveal pooling of fluor escein dye or angiographic macular edema. This was graded from late angiogram frames.
Epiretinal Membrane Opacity and Pseudohole Fluorescence
Results Referring
for pattern of pseudohole fluorescence between the two graders.
Dia~nosis
The referring diagnosis was available in 77 (94%) of the 82 patients with macular pseudoholes (Fig 8). Of the 82 patients, 18 (22%) were referred with a diagnosis of m~c ular hole or impending macular hole and 10 (12%) with a diagnosis of lamellar hole or macular cyst. The correct referring diagnosis of epiretinal membrane or epiretinal membrane with pseudohole was given in 43 (52%) patients . The referral was for an unrelated problem in six (7%) patients. Two patients whose fluorescein angiograms were unavailable are included in the analysis of the referring diagnosis because the diagnosis of macular pseudohole was made by clinical examination with fundus biomicroscopy.
The epiretinal membrane was graded as opaque in 20 (24%) eyes, visible but not obscuring retina features in 38 (46%) eyes, and transparent in 16 (19%)eyes (Figs 10 and II). The opacity of the epiretinal membrane could not be graded from the stereo color photographs in nine (I 1%) eyes. In eyes with opaque epiretinal membranes, II (55%) had no fluorescence in the base of the pseudohole and 8 (40%) had either patch y or complete fluorescence. In eyes with visible and transparent epiretinal membranes, six (16%) and two (13%), respectively, had no fluorescence in the base of the pseudohol e, whereas 27 (71%) and 12 (75%), respectively, had either patch y or complete fluorescence patterns. Hyperfluorescence within the area of the pseudohole was less common in eyes with opaque epiretinal membranes than in eyes with visible or transPatchy (37)
Fluorescein Angiographic Findings Fluorescein angiograms were available for review in 83 eyes with macular pseudoholes (Fig 9). Fluorescence which appeared and faded in synchrony with choroidal fluorescence was noted in the base of 52 (63%) of the pseudoholes. Among these 52 eyes, the pattern of fluorescence was patchy and smaller than the total area of the pseudo hole in 37 (45%) eyes. A "complete" pattern with hyperfluorescence filling the base of the pseudohole was seen in 15 (18%) eyes. The base of the pseudo hole transmitted no fluorescence in 20 (24%) eyes. In one (1%) eye, the pseudohole fluorescence was graded as questionable. In five (6%) eyes, there was atrophic macular degeneration or healed retinochoroiditis. which confounded evaluation of hyperfluorescence in the area of the pseudohole. Five (6%) eyes could not be graded because of inadequate photographs or media opacity. There was an 83% agreement
45%
Can't Determine (11 )
None (20)
Complete (15) Figure 9. Pseudohole fluorescence pattern. This was graded from early ph ases of th e fluorescein angiogram. The fluorescence appeared to be a form of tr ansmission defect as it coincided in timing and intensity to choroidal filling and decreased with ch or oidal emptying.
771
Ophthalmology
Volume 102, Number 5, May 1995
Visible
Present
1381
1261 31%
en't Determine
141
Cen't Grede
64%
191 Opeque
Absent
1201
(531
Trensperent
Figure 12. Late macular pooling of fluorescein in eyes with pseudoholes. This was graded from late angiogram frames.
1161 N
= 83 eyes
Figure 10. Epiretinal membrane opacity in eyes with pseudoholes. This was graded from stereo color photographs.
parent epiretinal membranes (P = 0.002). There was an 80% interobserver agreement for epiretinal membrane opacity before adjudication of any differences.
Late Macular Pooling of Dye and Pseudohole Fluorescence Macular pooling of fluorescein dye was present in 26 (31%) eyes, absent in 53 (64%) eyes, graded as questionable in 1 (1%) eye, and could not be graded in 3 (4%) eyes (Figs 12 and 13). In eyes with pooling of dye, 13 (50%) had no fluorescence in the base of the pseudohole and 10 (39%) showed either patchy or complete patterns of pseudohole fluorescence. In eyes with no pooling of dye, 7 (13%) had no fluorescence in the base of the pseudohole and 41 (78%) showed either patchy or complete patterns of hyperfluorescence. Hyperfluorescence in the base of the pseudohole was less common in eyes with fluorescein pooling in the macula (P < 0.001). Interobserver agreement for macular pooling of fluorescein dye was 98%.
Pseudohole Fluorescence and Visual Acuity Visual acuities were averaged using conversion to log scale and compared using analysis of variance and Bonferroni adjusted P values (Fig 14). The mean visual acuity for
eyes with no hyperfluorescence in the base of the pseudohole was 20/70. The mean visual acuity for eyes with patchy hyperfluorescence was 20/40 and for eyes with complete hyperfluorescence was 20/30. The mean visual acuity was significantly lower for eyes with no hyperfluorescence in the base of the pseudo hole than for eyes with patchy (P < 0.01) or complete (P < 0.01) hyperfluorescence.
Discussion Pseudoholes of the macula are well-demarcated round or oval depressions usually found in the fovea in eyes with macular epiretinal membranes. Allen and Gass' and Gass' described this as a central defect in the epiretinal membrane which creates a steep drop-off around the fovea, simulating a full-thickness macular hole. Pseudoholes of the macula sometimes arouse suspicion of full-thickness or impending macular holes, as was seen in 22% of the eyes in this study and 29% in a report by Fish et a1. 2 In 1990, Gass and Joondeph" reported that 4 (22%) of 18 patients referred to Bascom Palmer Institute for suspected impending macular holes had instead epiretinal membranes with pseudoholes of the macula. Because both prognosis and management differ among these conditions, specific reliable diagnostic criteria are needed. In 1976, Allen and Gass' reported faint central hyperfluorescence in three of four eyes with pseudoholes and
Patchy
Patchy
C.O. Complete None
ERM Opaque
772
None
ERM Visible
Complete
None
ERM Transparent
Figure 11. Pseudohole fluorescence versus epiretinal membrane opacity. Hyperfluorescence within the base of the pseudohole was less common in eyes with opaque epiretinal membranes (ERM). CD = cannot determine.
Klein et al . Pseudoholes of the Macula and Epiretinal Membranes
Patchy
Complete (2)
Can't Determine (3)
(28)
atchy
Figure 13. Pseudohole fluorescence versus late macular pooling of dye. Hyperfluorescence in the base of the pseudohole was less common in eyes with angiographic macular edema.
(8)
None (7)
Can't Determine
epiretinal membranes. They attributed the central fluorescence to a screening effect by the semitransparent perifoveal membrane. I.3 Variable fluorescence was seen in 52 (63%) of the 83 macular pseudo holes in this study. The fluorescence was complete and, thus, most likely to be confused with a full-thickness macular holev? or impending macular hole 7 in 15 (18%) eyes. Fluorescence in the base of the pseudohole was more common in eyes with transparent or barely visible epiretinal membranes than in eyes with opaque membranes. This may suggest true transmission hyperfluorescence rather than a screening effect from overlying membrane. One would expect a screening effect to be emphasized in eyes with pseudoholes within more opaque epiretinal membranes. The waxing and waning of hyperfluorescence during the early and middle phases of the angiogram as observed in this study is a pattern frequently seen in cases of transmission hyperfluorescence. The basis for this transmission fluorescence may be local attenuation or flattening of the foveal xanthophyll or retinal pigment epithelium. We think the former is more likely because eyes with visible changes in the retinal pigment epithelium were eliminated from grading as having coexistent macular disease. Mean VA(201y)
80
70 60 50 40 30 20 10
---L.
None
...L-
(13)
(5)
Pooling Absent
OIL.-
None
----.-/
Patchy Pseudohole Fluorescence
Figure 14. Pseudohole fluorescence versus visual acuity (VA). Mean visual acuity was lower for eyes with no hyperfluorescence in the base of the pseudohole.
Pooling Present Pseudohole hyperfluorescence during early angiogram frames was less prevalent in eyes with late perifoveal pooling of fluorescein dye and with lower visual acuities. This may represent reduced transmission fluorescence in the presence of macular thickening in those eyes with pooling of dye in the perifoveal neurosensory retina. A report by Fish et a1 2 described window defects in only 3 of the 11 (27%)eyes with pseudoholes as compared with 63% in the current series. If standard photographs and definitions were used for evaluating fluorescein angiograms in their study, we did not find them in the report and so are unable to explain their lower incidence of window defects. We noted pseudoholes in a higher percentage (20%) of eyes with epiretinal membrane than has been reported in previous studies. Sidd et a1 8 reported an incidence of9.4% and Margherio et a1 9 8% after review of 74 and 275 eyes with epiretinal membranes, respectively. Our institution has been investigating the use of vitrectomy and transforming growth factor-In for full-thickness macular holes and thus receives many referrals of known or suspected macular holes. 10 This unusual situation may account for this difference. The results of this study do not establish the anatomic basis for hyperfluorescence of the base of pseudoholes in eyes with macular epiretinal membranes. When present, the pseudohole fluorescence behaved as a transmission defect appearing early and fading late. The fluorescence, therefore, most likely reflects an attenuation of one of the pigmented layers in the fovea, either xanthophyll or retinal pigment epithelium. Because these pseudoholes were associated with fibro-glial proliferation on the inner surface of the macula, attenuation of xanthophyll by flattening or stretching of the neurosensory retina seems a reasonable explanation for the hyperfluorescence. This hyperfluorescence may result in confusion of a macular hole with a pseudohole if the patient is not examined carefully by slit-lamp biomicroscopy.
Conclusions Pseudoholes of the macula appear concave on biomicroscopy and are sometimes confused with full-thickness
773
Ophthalmology
Volume 102, Number 5, May 1995
macular holes. Hyperfluorescence of the base of the pseudohole presented as a variable and often incomplete transmission defect in 63% of eyes studied; however, the entire base of the pseudohole was hyperfluorescent in only 18% of eyes studied. This alteration may represent local attenuation of foveal xanthophyll and/or thinning of the foveal neurosensory retina, resulting from centripetal traction forces within the epiretinal membrane.
References 1. Allen AW Jr, Gass JOM. Contraction of a perifoveal epiretinal membrane simulating a macular hole. Am J Ophthalmol 1986;82:684-91. 2. Fish RH, Anand R, Izbrand OJ. Macular pseudoholes. Clinical features and accuracy ofdiagnosis. Ophthalmology 1992;99: 1665-70. 3. Gass JDM. Stereoscopic Atlas of Macular Diseases: Diagnosis and Treatment. 3rd ed. Vol. 1. St. Louis: CV Mosby, 1987;700-3. 4. Gass JDM, Joondeph BC Observations concerning patients with suspected impending macular holes. Am J Ophthalmol 1990; 109:638-46.
774
5. Gass JDM. Lamellar macular holes: a complication of cystoid macular edema after cataract extraction: a clinicopathologic case report. Trans Am Ophthalmol Soc 1975;73:23150. 6. Gass JDM. Idiopathic senile macular hole. Its early stages and pathogenesis. Arch Ophthalmol 1988;106:629-39. 7. Akiba J, Yoshida A, Trempe CL. Risk of developing a macular hole. Arch Ophthalmol 1990;108:1088-90. 8. Sidd RJ, Fine SL, Owens 8L, Patz A. Idiopathic preretinal gliosis. Am J Ophthalmol 1982;94:44-8. 9. Margherio RR, Cox MS Jr, Trese MT, et al. Removal of epimacular membranes. Ophthalmology 1985;92: 1075-83. 10. Glaser BM, Michels RG, Kuppermann BD, et al. Transforming growth factor-S, for the treatment of full-thickness macular holes. A prospective randomized study. Ophthalmology 1992;99: 1162-73. II. Margherio RR. Discussion, 1387-8 Of: Michels RG. Vitrectomy for macular pucker. Ophthalmology 1984;91: 1384-8. 12. Appiah AP, Hirose T, Kado M. A review of 324 cases of idiopathic premacular gliosis. Am J Ophthalmol 1988;106: 533-5. 13. Guyer DR, Green WR, de Bustros S, Fine SL. Histopathologic features of idiopathic macular holes and cysts. Ophthalmology 1990;97: 1045-51.