INTRARETINAL FIBROSIS IN EXUDATIVE DIABETIC MACULAR EDEMA AFTER RANIBIZUMAB TREATMENTS Voraporn Chaikitmongkol, MD,*† Neil M. Bressler, MD*
Purpose: To describe the clinical and optical coherence tomography findings of an eye with diabetic macular edema that developed intraretinal fibrosis in an area previously occupied by lipid accumulating after the intravitreous ranibizumab treatment. Methods: Interventional case report. Results: An 85-year-old man presented with diabetic macular edema involving the center of the macula with a half disk area of lipid inferotemporal to the macula. He received ranibizumab treatments after the principles of the Diabetic Retinopathy Clinical Research Network retreatment guidelines. After 12 doses of intravitreous ranibizumab injections over 20 months, macular edema resolved, visual acuity improved from 20/63 to 20/40, and the central subfield thickness decreased from 404 mm to 234 mm. As the edema resolved, the area of lipid did not expand toward the fovea but was replaced by fibrosis occupying the area of lipid, only smaller in extent. Optical coherence tomography scans showed an intraretinal, dome-shaped hyperreflective area corresponding to the fibrosis. Conclusion: This case report, to our knowledge, provides the first documentation of intraretinal fibrosis replacing an area of lipid associated with diabetic macular edema after anti-vascular endothelial growth factor therapy, as had been described previously following laser photocoagulation for diabetic macular edema. Unlike some previous reports of lipid accumulating within the fovea with subsequent fibrosis corresponding to the metaplastic retinal pigment epithelium on histopathology, with or without laser treatment, the lipid in this case did not expand into the fovea before the development of fibrosis, and optical coherence tomography confirmed that the fibrosis was located in the intraretinal rather than the subretinal pigment epithelium space. RETINAL CASES & BRIEF REPORTS 8:336–339, 2014
replaced by fibrosis, associated with permanent central visual acuity impairment.2 Clinicopathologic study of exudative plaques in DME suggests that a metaplastic response of the retinal pigment epithelium (RPE) to the lipid can result in a fibrotic scar with no choroidal pathology.3,4 Despite this well-characterized sequence of events, we are unaware of any reports describing whether a similar process occurs after anti-vascular endothelial growth factor treatment of DME, which typically is the standard care for DME involving the center of the macula with vision impairment.5,6 Therefore, we report a case of lipid from DME that was replaced by fibrosis after the treatment with anti-vascular endothelial growth factor therapy to describe its course and subsequently compare
From the *Retina Division, Wilmer Eye Institute, Johns Hopkins University School of Medicine, Baltimore, Maryland; and †Retina Division, Department of Ophthalmology, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand.
L
ipid, or hard exudate, is one of the clinical features of diabetic macular edema (DME). Its presence presumably represents lipoproteins that accompany fluid accumulation in association with DME within the retinal tissue. In the Early Treatment Diabetic Retinopathy Study, individuals with the same extent of macular edema were at a higher risk of losing visual acuity depending on the extent of lipid.1 Large areas of lipid that accumulate in the foveal center may be 336
FIBROSIS AFTER ANTI-VEGF FOR DME
337
Fig. 1. A and B. April 2006, color fundus photograph and late phase fluorescein angiogram frame showing faint scars from previous focal/grid laser with no evidence of lipid, extensive atrophy from laser treatment, or choroidal neovascularization. C. October 2011, DME with lipid inferotemporal and superotemporal to the fovea. D. January 2012, edema persists with multilobular areas of lipid. E and F. April 2012, as edema resolves, fibrotic tissue develops in an area previously occupied by lipid inferotemporally (arrow). G. November 2012, the formation of fibrosis within an area previously occupied by lipid obscured somewhat by cataract. H. June 2013, after cataract surgery and 12 doses of ranibizumab, edema and lipid appear resolved; fibrosis is noted in an area previously occupied by lipid inferotemporal to the fovea.
that course with what was described after the treatment of such cases with laser photocoagulation, to expand on our understanding of the process by providing optical coherence tomography (OCT) findings.
Case Report An 85-year-old African American man with Type 2 diabetes presented in 2011 with mild blurry vision in his left eye. He had been diagnosed with a central retinal artery occlusion in the right eye, and proliferative diabetic retinopathy with DME in the left eye. In 2006, he underwent panretinal photocoagulation for proliferative diabetic retinopathy, focal/grid laser photocoagulation for DME, and several intravitreous bevacizumab injections for DME in his left eye. When DME had resolved, his left macula showed faint scars from the focal/grid laser photocoagulation with no angiographic evidence that any of the laser-treated areas were very destructive and likely to promote ingrowth of choroidal neoSupported by unrestricted research gifts to the Retina Division, Wilmer Eye Institute, Johns Hopkins University, and the James P. Gills Professorship (N.M.B.), and Department of Ophthalmology, Faculty of Medicine, Chiang Mai University, Thailand (V.C.). N. M. Bressler is PI of grants at the Johns Hopkins University School of Medicine sponsored by the following entities: Bayer Healthcare Pharmaceuticals, Inc; Genentech, Inc; Novartis Pharma AG; and Regeneron Pharmaceuticals, Inc. V. Chaikitmongkol has no conflicting interests to disclose. Reprint requests: Neil M. Bressler, MD, 600 N Wolfe Street, Maumenee 752, Baltimore, MD 21287; e-mail: nmboffi[email protected]
vascularization (Figure 1, A and B). These conditions stabilized through 2010. In October 2010, during his routine follow-up visit, his visual acuity remained 1/200 in his right eye and 20/32 in his left eye. Dilated ophthalmoscopic examination of the left eye showed macular edema with fluid and the lipid exudate located inferotemporal to the fovea. Because of relatively good visual acuity and the lack of symptoms, no treatment was recommended although close observation was planned. In October 2011, his visual acuity in the left eye decreased to 20/63 consistent with an increased area of macular edema extending to the foveal center. The central subfield (CSF) thickness was 404 mm on a spectral domain OCT. The lipid in the inferotemporal macula had increased in size. There was a new area of retinal thickening with linear exudate in the extramacular region of the same eye (Figure 1C). He received his first dose of intravitreous ranibizumab, and then was treated according to the principles of the Diabetic Retinopathy Clinical Research Network treatment guidelines.7 In January 2012, after 3 doses of ranibizumab, his visual acuity improved to 20/40 but macular edema persisted with the CSF thickness increasing to 413 mm. The area of lipid appeared dispersed into a multilobular shape that remained localized within the inferotemporal region. An additional 2 doses of ranibizumab were administered, the visual acuity deteriorated slightly to 20/63 while the CSF thickness decreased to 329 mm. The lipid started to disappear as a faint area of fibrosis was noted within the center of the lipid (Figure 1, D and E). In May 2012, after 6 doses of ranibizumab, the visual acuity remained 20/63 as a cataract, which began to diminish the view of the macula and contribute to decreased vision, began to develop in the setting of persistent macular edema, and a CSF thickness increasing to 364 mm. Ophthalmoscopic examination showed
338
RETINAL CASES & BRIEF REPORTS´ 2014 VOLUME 8 NUMBER 4
a substantial decrease in the extent of lipid as fibrosis became more apparent (Figure 1F). An additional 5 doses of ranibizumab treatments were given because of persistent but improving macular edema. In November 2012, his visual acuity improved to 20/40. The OCT showed no fluid in the macula, and ranibizumab treatment was withheld (Figure 1G). In January 2013, his left visual acuity dropped to 20/50, judged to be, predominantly, due to his cataract, and he was advised to undergo cataract surgery. In May 2013 after cataract surgery, his visual acuity improved to 20/40 with a minimal degree of macular edema judged to be DME, not postsurgical (Irvine–Gass) cystoid macular edema, and ranibizumab injections were resumed. In June 2013, after a total of 12 ranibizumab injections over 20 months, and 3 months after cataract surgery, his visual acuity was 20/40 with resolution of all thickening and lipid in the macula, while his CSF thickness decreased to 234 mm. Faint focal/grid photocoagulation lesions from the treatment as recent as 2006 could be seen in the macular region. In addition, a distinct white fibrotic scar was noted in the area previously occupied by lipid (Figure 1H). Optical coherence tomography of this area showed a dome-shaped hyperreflective substance in the outer retina corresponding to the fibrotic area seen ophthalmoscopically. The extramacular linear area of lipid noted previously in the superotemporal region also was no longer apparent except for a faint fibrotic appearance.
Discussion This report provides new evidence to our knowledge, specifically spectral domain OCT image findings, to assist our understanding regarding the origin of fibrosis within an area previously occupied by lipid from DME. The OCT findings support the hypothesis, based on histopathology of similar cases,4 that accumulation of hard exudate in the outer retina in juxtaposition to the RPE is associated with focal metaplasia of the RPE, leading to fibrotic scar formation. Specifically, the OCT in Figure 2 showed that only areas with more extensive lipid in juxtaposition to the RPE developed a fibrotic scar, while other less extensive areas of lipid within the outer retina but not in juxtaposition to the RPE showed resolution of lipid without scarring. The localization of the fibrotic lesion within the outer retinal boundary with an intact hyperreflective line between the outer retina and the choriocapillaris, presumably the attenuated RPE/Bruch membrane complex, suggested a retinal or RPE origin of the fibrotic lesion rather than the choroidal origin
Fig. 2. Red-free fundus photographs show the location of B-scan OCT across lipid (A) and subsequent fibrosis (B). Scans overtime across the areas of lipid in the inferotemporal macula (C–E) and outside the macula (F–H); when edema resolves (E and H), smaller foci of lipid no longer are apparent while the larger area of lipid, which extended to the RPE, remains visible as hyperreflective area in the outer retina and corresponds to fibrosis seen ophthalmoscopically. Noted attenuated RPE/Bruch membrane complex underneath the fibrotic area (arrow).
339
FIBROSIS AFTER ANTI-VEGF FOR DME
(such as choroidal neovascularization, or fibrovascular tissue, growing from the choriocapillaris). Also, over 4 years had elapsed from the time of his focal/grid laser treatment and the development of lipid, with no evidence of subretinal fluid or subretinal hemorrhage to suggest that the lipid was associated with choroidal neovascularization. The thin dark ring separating out a distinctly different zone within the lipid exudate, (Figure 1C) which eventually evolved into the fibrotic lesion, might suggest previous disruption of the RPE– photoreceptor boundary, possibly from antecedent focal/grid laser, or might occur concomitantly with the development of lipid with later fibrosis. Nevertheless, the absence of heavy laser treatment noted on both fundus photography and fluorescein angiography 4 years before the development of the fibrotic nodule makes it more likely that the fibrosis was a consequence of the lipid rather than choroidal neovascularization (Figure 1, A and B). Finally, the OCT appearance does not suggest that some or all of the fibrosis is under the neurosensory retina. Rather, the retina overlying the fibrosis, while elevated or thickened, does not appear to be fullthickness retina. The fibrosis appears to have replaced at least a portion of the outer retina, suggesting that the fibrosis may be “intraretinal.” At the least, the term “subretinal” fibrosis, that often has been used in the peer-reviewed literature2 previously to describe these lesions, perhaps should be used cautiously with the recognition that some or all of the fibrosis may be intraretinal. In summary, this case demonstrates that intraretinal or subretinal fibrosis may occur after anti-vascular endothelial growth factor treatment for DME in an area previously occupied by lipid, similar to what has been seen with the natural course of this pathology or
after the laser treatment of DME. However, the case shows that the lipid, and subsequently the fibrosis within areas previously occupied by the lipid, may not migrate into the center of the macula. Optical coherence tomography confirms that the fibrosis, which does form likely, is within the outer portion of the retina or subretinal or both. Key words: fibrosis, lipid, exudate, diabetic macular edema, anti-vascular endothelial growth factor, ranibizumab. References 1. Chew EY, Klein ML, Ferris FL, et al. Association of elevated serum lipid levels with retinal hard exudate in diabetic retinopathy. Early Treatment Diabetic Retinopathy Study (ETDRS) Report 22. Arch Ophthalmol 1996;114:1079–1084. 2. Fong DS, Segal PP, Myers F, et al. Subretinal fibrosis in diabetic macular edema. ETDRS report 23. Early Treatment Diabetic Retinopathy Study Research Group. Arch Ophthalmol 1997;115:873–877. 3. Begg IS, Rootman J. Clinico-pathological study of an organized plaque in exudative diabetic maculopathy. Can J Ophthalmol 1976;11:197–202. 4. Sigurdsson R, Begg IS. Organised macular plaques in exudative diabetic maculopathy. Br J Ophthalmol 1980;64:392–397. 5. Diabetic Retinopathy Clinical Research Network, Elman MJ, Aiello LP, Beck RW, et al. Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology 2010;117:1064–1077.e35. 6. Elman MJ, Bressler NM, Qin H, et al. Expanded 2-year follow-up of ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology 2011;118:609–614. 7. Diabetic Retinopathy Clinical Research Network; Writing Committee, Aiello LP, Beck RW, Bressler NM, et al. Rationale for the diabetic retinopathy clinical research network treatment protocol for center-involved diabetic macular edema. Ophthalmology 2011;118:e5–e14.
Purpose: To describe the clinical and optical coherence tomography findings of an eye with diabetic macular edema that developed intraretinal fibrosis in an area previously occupied by lipid accumulating after the intravitreous ranibizumab treatment. Methods: Interventional case report. Results: An 85-year-old man presented with diabetic macular edema involving the center of the macula with a half disk area of lipid inferotemporal to the macula. He received ranibizumab treatments after the principles of the Diabetic Retinopathy Clinical Research Network retreatment guidelines. After 12 doses of intravitreous ranibizumab injections over 20 months, macular edema resolved, visual acuity improved from 20/63 to 20/40, and the central subfield thickness decreased from 404 mm to 234 mm. As the edema resolved, the area of lipid did not expand toward the fovea but was replaced by fibrosis occupying the area of lipid, only smaller in extent. Optical coherence tomography scans showed an intraretinal, dome-shaped hyperreflective area corresponding to the fibrosis. Conclusion: This case report, to our knowledge, provides the first documentation of intraretinal fibrosis replacing an area of lipid associated with diabetic macular edema after anti-vascular endothelial growth factor therapy, as had been described previously following laser photocoagulation for diabetic macular edema. Unlike some previous reports of lipid accumulating within the fovea with subsequent fibrosis corresponding to the metaplastic retinal pigment epithelium on histopathology, with or without laser treatment, the lipid in this case did not expand into the fovea before the development of fibrosis, and optical coherence tomography confirmed that the fibrosis was located in the intraretinal rather than the subretinal pigment epithelium space. RETINAL CASES & BRIEF REPORTS 8:336–339, 2014
replaced by fibrosis, associated with permanent central visual acuity impairment.2 Clinicopathologic study of exudative plaques in DME suggests that a metaplastic response of the retinal pigment epithelium (RPE) to the lipid can result in a fibrotic scar with no choroidal pathology.3,4 Despite this well-characterized sequence of events, we are unaware of any reports describing whether a similar process occurs after anti-vascular endothelial growth factor treatment of DME, which typically is the standard care for DME involving the center of the macula with vision impairment.5,6 Therefore, we report a case of lipid from DME that was replaced by fibrosis after the treatment with anti-vascular endothelial growth factor therapy to describe its course and subsequently compare
From the *Retina Division, Wilmer Eye Institute, Johns Hopkins University School of Medicine, Baltimore, Maryland; and †Retina Division, Department of Ophthalmology, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand.
L
ipid, or hard exudate, is one of the clinical features of diabetic macular edema (DME). Its presence presumably represents lipoproteins that accompany fluid accumulation in association with DME within the retinal tissue. In the Early Treatment Diabetic Retinopathy Study, individuals with the same extent of macular edema were at a higher risk of losing visual acuity depending on the extent of lipid.1 Large areas of lipid that accumulate in the foveal center may be 336
FIBROSIS AFTER ANTI-VEGF FOR DME
337
Fig. 1. A and B. April 2006, color fundus photograph and late phase fluorescein angiogram frame showing faint scars from previous focal/grid laser with no evidence of lipid, extensive atrophy from laser treatment, or choroidal neovascularization. C. October 2011, DME with lipid inferotemporal and superotemporal to the fovea. D. January 2012, edema persists with multilobular areas of lipid. E and F. April 2012, as edema resolves, fibrotic tissue develops in an area previously occupied by lipid inferotemporally (arrow). G. November 2012, the formation of fibrosis within an area previously occupied by lipid obscured somewhat by cataract. H. June 2013, after cataract surgery and 12 doses of ranibizumab, edema and lipid appear resolved; fibrosis is noted in an area previously occupied by lipid inferotemporal to the fovea.
that course with what was described after the treatment of such cases with laser photocoagulation, to expand on our understanding of the process by providing optical coherence tomography (OCT) findings.
Case Report An 85-year-old African American man with Type 2 diabetes presented in 2011 with mild blurry vision in his left eye. He had been diagnosed with a central retinal artery occlusion in the right eye, and proliferative diabetic retinopathy with DME in the left eye. In 2006, he underwent panretinal photocoagulation for proliferative diabetic retinopathy, focal/grid laser photocoagulation for DME, and several intravitreous bevacizumab injections for DME in his left eye. When DME had resolved, his left macula showed faint scars from the focal/grid laser photocoagulation with no angiographic evidence that any of the laser-treated areas were very destructive and likely to promote ingrowth of choroidal neoSupported by unrestricted research gifts to the Retina Division, Wilmer Eye Institute, Johns Hopkins University, and the James P. Gills Professorship (N.M.B.), and Department of Ophthalmology, Faculty of Medicine, Chiang Mai University, Thailand (V.C.). N. M. Bressler is PI of grants at the Johns Hopkins University School of Medicine sponsored by the following entities: Bayer Healthcare Pharmaceuticals, Inc; Genentech, Inc; Novartis Pharma AG; and Regeneron Pharmaceuticals, Inc. V. Chaikitmongkol has no conflicting interests to disclose. Reprint requests: Neil M. Bressler, MD, 600 N Wolfe Street, Maumenee 752, Baltimore, MD 21287; e-mail: nmboffi[email protected]
vascularization (Figure 1, A and B). These conditions stabilized through 2010. In October 2010, during his routine follow-up visit, his visual acuity remained 1/200 in his right eye and 20/32 in his left eye. Dilated ophthalmoscopic examination of the left eye showed macular edema with fluid and the lipid exudate located inferotemporal to the fovea. Because of relatively good visual acuity and the lack of symptoms, no treatment was recommended although close observation was planned. In October 2011, his visual acuity in the left eye decreased to 20/63 consistent with an increased area of macular edema extending to the foveal center. The central subfield (CSF) thickness was 404 mm on a spectral domain OCT. The lipid in the inferotemporal macula had increased in size. There was a new area of retinal thickening with linear exudate in the extramacular region of the same eye (Figure 1C). He received his first dose of intravitreous ranibizumab, and then was treated according to the principles of the Diabetic Retinopathy Clinical Research Network treatment guidelines.7 In January 2012, after 3 doses of ranibizumab, his visual acuity improved to 20/40 but macular edema persisted with the CSF thickness increasing to 413 mm. The area of lipid appeared dispersed into a multilobular shape that remained localized within the inferotemporal region. An additional 2 doses of ranibizumab were administered, the visual acuity deteriorated slightly to 20/63 while the CSF thickness decreased to 329 mm. The lipid started to disappear as a faint area of fibrosis was noted within the center of the lipid (Figure 1, D and E). In May 2012, after 6 doses of ranibizumab, the visual acuity remained 20/63 as a cataract, which began to diminish the view of the macula and contribute to decreased vision, began to develop in the setting of persistent macular edema, and a CSF thickness increasing to 364 mm. Ophthalmoscopic examination showed
338
RETINAL CASES & BRIEF REPORTS´ 2014 VOLUME 8 NUMBER 4
a substantial decrease in the extent of lipid as fibrosis became more apparent (Figure 1F). An additional 5 doses of ranibizumab treatments were given because of persistent but improving macular edema. In November 2012, his visual acuity improved to 20/40. The OCT showed no fluid in the macula, and ranibizumab treatment was withheld (Figure 1G). In January 2013, his left visual acuity dropped to 20/50, judged to be, predominantly, due to his cataract, and he was advised to undergo cataract surgery. In May 2013 after cataract surgery, his visual acuity improved to 20/40 with a minimal degree of macular edema judged to be DME, not postsurgical (Irvine–Gass) cystoid macular edema, and ranibizumab injections were resumed. In June 2013, after a total of 12 ranibizumab injections over 20 months, and 3 months after cataract surgery, his visual acuity was 20/40 with resolution of all thickening and lipid in the macula, while his CSF thickness decreased to 234 mm. Faint focal/grid photocoagulation lesions from the treatment as recent as 2006 could be seen in the macular region. In addition, a distinct white fibrotic scar was noted in the area previously occupied by lipid (Figure 1H). Optical coherence tomography of this area showed a dome-shaped hyperreflective substance in the outer retina corresponding to the fibrotic area seen ophthalmoscopically. The extramacular linear area of lipid noted previously in the superotemporal region also was no longer apparent except for a faint fibrotic appearance.
Discussion This report provides new evidence to our knowledge, specifically spectral domain OCT image findings, to assist our understanding regarding the origin of fibrosis within an area previously occupied by lipid from DME. The OCT findings support the hypothesis, based on histopathology of similar cases,4 that accumulation of hard exudate in the outer retina in juxtaposition to the RPE is associated with focal metaplasia of the RPE, leading to fibrotic scar formation. Specifically, the OCT in Figure 2 showed that only areas with more extensive lipid in juxtaposition to the RPE developed a fibrotic scar, while other less extensive areas of lipid within the outer retina but not in juxtaposition to the RPE showed resolution of lipid without scarring. The localization of the fibrotic lesion within the outer retinal boundary with an intact hyperreflective line between the outer retina and the choriocapillaris, presumably the attenuated RPE/Bruch membrane complex, suggested a retinal or RPE origin of the fibrotic lesion rather than the choroidal origin
Fig. 2. Red-free fundus photographs show the location of B-scan OCT across lipid (A) and subsequent fibrosis (B). Scans overtime across the areas of lipid in the inferotemporal macula (C–E) and outside the macula (F–H); when edema resolves (E and H), smaller foci of lipid no longer are apparent while the larger area of lipid, which extended to the RPE, remains visible as hyperreflective area in the outer retina and corresponds to fibrosis seen ophthalmoscopically. Noted attenuated RPE/Bruch membrane complex underneath the fibrotic area (arrow).
339
FIBROSIS AFTER ANTI-VEGF FOR DME
(such as choroidal neovascularization, or fibrovascular tissue, growing from the choriocapillaris). Also, over 4 years had elapsed from the time of his focal/grid laser treatment and the development of lipid, with no evidence of subretinal fluid or subretinal hemorrhage to suggest that the lipid was associated with choroidal neovascularization. The thin dark ring separating out a distinctly different zone within the lipid exudate, (Figure 1C) which eventually evolved into the fibrotic lesion, might suggest previous disruption of the RPE– photoreceptor boundary, possibly from antecedent focal/grid laser, or might occur concomitantly with the development of lipid with later fibrosis. Nevertheless, the absence of heavy laser treatment noted on both fundus photography and fluorescein angiography 4 years before the development of the fibrotic nodule makes it more likely that the fibrosis was a consequence of the lipid rather than choroidal neovascularization (Figure 1, A and B). Finally, the OCT appearance does not suggest that some or all of the fibrosis is under the neurosensory retina. Rather, the retina overlying the fibrosis, while elevated or thickened, does not appear to be fullthickness retina. The fibrosis appears to have replaced at least a portion of the outer retina, suggesting that the fibrosis may be “intraretinal.” At the least, the term “subretinal” fibrosis, that often has been used in the peer-reviewed literature2 previously to describe these lesions, perhaps should be used cautiously with the recognition that some or all of the fibrosis may be intraretinal. In summary, this case demonstrates that intraretinal or subretinal fibrosis may occur after anti-vascular endothelial growth factor treatment for DME in an area previously occupied by lipid, similar to what has been seen with the natural course of this pathology or
after the laser treatment of DME. However, the case shows that the lipid, and subsequently the fibrosis within areas previously occupied by the lipid, may not migrate into the center of the macula. Optical coherence tomography confirms that the fibrosis, which does form likely, is within the outer portion of the retina or subretinal or both. Key words: fibrosis, lipid, exudate, diabetic macular edema, anti-vascular endothelial growth factor, ranibizumab. References 1. Chew EY, Klein ML, Ferris FL, et al. Association of elevated serum lipid levels with retinal hard exudate in diabetic retinopathy. Early Treatment Diabetic Retinopathy Study (ETDRS) Report 22. Arch Ophthalmol 1996;114:1079–1084. 2. Fong DS, Segal PP, Myers F, et al. Subretinal fibrosis in diabetic macular edema. ETDRS report 23. Early Treatment Diabetic Retinopathy Study Research Group. Arch Ophthalmol 1997;115:873–877. 3. Begg IS, Rootman J. Clinico-pathological study of an organized plaque in exudative diabetic maculopathy. Can J Ophthalmol 1976;11:197–202. 4. Sigurdsson R, Begg IS. Organised macular plaques in exudative diabetic maculopathy. Br J Ophthalmol 1980;64:392–397. 5. Diabetic Retinopathy Clinical Research Network, Elman MJ, Aiello LP, Beck RW, et al. Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology 2010;117:1064–1077.e35. 6. Elman MJ, Bressler NM, Qin H, et al. Expanded 2-year follow-up of ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology 2011;118:609–614. 7. Diabetic Retinopathy Clinical Research Network; Writing Committee, Aiello LP, Beck RW, Bressler NM, et al. Rationale for the diabetic retinopathy clinical research network treatment protocol for center-involved diabetic macular edema. Ophthalmology 2011;118:e5–e14.
