204 Ophthalmic Surgery, Lasers & Imaging Retina | Healio.com/OSLIRetina

■ C A S E R E P O R T ■

Increasing Volume of a Retinal Pigmented Epithelial Detachment as a Predictor of Submacular Hemorrhage During Anti-VEGF TherapyCarlos Alexandre de Amorim Garcia Filho, MD; Fernando Marcondes Penha, MD, PhD; Giovanni Gregori, PhD; Philip J. Rosenfeld, MD, PhD

ABSTRACT: Vascularized retinal pigment epithelium detachments (PEDs) are part of the spectrum of neovascular age-related macular degeneration (AMD). These patients with vascularized PEDs are at a higher risk of experiencing severe vision loss. This case report demonstrates the use of a new spectral-domain optical coherence tomography (SD-OCT) algorithm to measure the area and volume of PEDs. When this algorithm was applied to the scans from a patient with a vascularized PED who developed a large submacular hemorrhage while undergoing ranibizumab therapy, the authors found that the algorithm measured an increase in the area and volume of the PED that preceded the macular hemorrhage. Although further studies are needed, the increase in the volume of a PED may serve as a useful predictor of disease progression and the need for more aggressive anti-VEGF therapy.

[Ophthalmic Surg Lasers Imaging Retina. 2013;44:204-207]

INTRODUCTION

A retinal pigment epithelium detachment (PED) is formed by the accumulation of sub-RPE fluid, blood, fibrovascular tissue, or drusenoid material between the basal lamina of the retinal pigment epithelium (RPE) and the inner collagenous layer of Bruch’s membrane.1-4 PEDs can be part of the spectrum of age-related macular degeneration (AMD) and are fre-quently associated with the presence of choroidal neovascularization (CNV) and a higher risk of severe vision loss.5

The introduction of spectral-domain optical co-herence tomography (SD-OCT) imaging allows for the acquisition of high-speed, high-resolution, high-density, three-dimensional images that can capture the true geometry of the retina and RPE. Recently, a segmentation algorithm was developed to extract quantitative information from SD-OCT datasets ob-tained from scans containing elevations of the RPE. This new algorithm creates a virtual RPE floor free of any deformations and subtracts this virtual RPE floor from the true RPE segmentation. The resulting dif-ference map is referred to as the RPE elevation map, which includes both area and volume measurements of a PED. This algorithm has been shown to repro-ducibly measure the area and volume of drusen and larger PEDs.6-8

The purpose of this report is to demonstrate the potential usefulness of this algorithm in following eyes undergoing SD-OCT–guided re-treatment for neovascular AMD. We used this quantitative algo-rithm retrospectively in this case to reveal a change in the volume of the PED that was not obvious from the B-scans. This case demonstrates how the changes in the PED measurements could serve as useful pre-dictors of disease progression in similar cases.

CASE REPORT

A 74-year-old woman who was treated for dry AMD at the retina clinic of the Bascom Palmer Eye Institute presented with decreased visual acuity (VA) in her right eye for 15 days. VA was 20/40 in the right eye and 20/25 in the left eye. Anterior segment bio-microscopy and intraocular pressures were normal. Fundus examination revealed juxtafoveal choroidal neovascularization (CNV) with subretinal fluid in-volving the fovea. SD-OCT confirmed the presence of subretinal fluid with foveal involvement, but a serous PED was not present at baseline. Intravitreal injec-

From the Department of Ophthalmology, Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida.

Originally submitted May 1, 2012. Accepted for publication February 20, 2013.Drs. Garcia Filho and Rosenfeld received research support from Carl Zeiss Meditec. Dr. Gregori and the University of Miami co-own a patent that is licensed to Carl Zeiss Meditec. Dr. Rosenfeld has received honoraria for lectures from Carl Zeiss Meditec.Address correspondence to Philip J. Rosenfeld, MD, PhD, Bascom Palmer Eye Institute, 900 NW 17th street, Miami, FL, 33136; Email: prosenfeld@med.miami.edu.doi: 10.3928/23258160-20130313-12

March/April 2013 • Vol. 44, No. 2 205

tions of Lucentis (ranibizumab 0.5 mg; Genentech, South San Francisco, CA) were initiated. After 11 months of follow-up in which a treat-and-extend re-treatment regimen was used, the patient had received seven intravitreal injections of ranibizumab and the CNV was controlled. The VA was 20/30 and there was no evidence of intraretinal or subretinal fluid (Figure 1). The patient received an injection of ranibizumab at this visit and was asked to return 45 days later. At the next visit, she complained of vision loss, and her VA was now 20/40. Fundus examination revealed the presence of a PED with associated subretinal fluid. Figure 2 shows the color image, horizontal and verti-cal foveal SD-OCT B-scans, RPE segmentation map, and retinal thickness map at this visit. A PED is evi-dent on the B-scans (B, C) and the RPE segmentation map (D), while the subretinal fluid is visualized on the B-scans (B, C) and the retinal thickness map (E, yellow areas). An injection of ranibizumab was given.

The patient continued to receive monthly in-

jections of ranibizumab (0.5 mg), and the next five monthly visits are depicted in Figure 3, which shows the color fundus images, the horizontal foveal B-scans, the RPE segmentation maps, retinal thickness maps, and RPE elevation maps. At each of these follow-up visits, the amount of subretinal fluid increased and the volume of the PED increased compared with the previous visit. Despite monthly ranibizumab injec-tions for 5 months, the volume measurements of the PED increased from 0.17 mm3 to 0.24 mm3 during this interval. The VA remained stable during this period.

Within 28 days of the last injection, a submacular hemorrhage occurred (Figure 4), with VA decreasing to 20/200. Management options were discussed, and the patient elected to receive additional injections of ranibizumab rather than undergo an attempt to displace or remove the hemorrhage. After 4 monthly injections of ranibizumab, the submacular hemorrhage had almost completely resolved, a fibrotic scar formed, and VA was stable at 20/200 (Figure 5).

Figure 1. The right eye of a 74 year-old woman after 11 months and seven intravitreal injections of ranibizumab for the treatment of cho-roidal neovascularization. No evidence of intraretinal or subretinal fluid is observed. Color image (A), horizontal foveal B-scan (B), retinal thickness map (C), and retinal pigment epithelium segmentation map (D). The eighth injection of ranibizumab was given at this visit.

Figure 2. Right eye 45 days after the eighth intravitreal injection of ranibizumab. Color fundus image (A), horizontal foveal B-scan (B), vertical foveal B-scan (C), retinal thickness map (D), and retinal pigment epithelium segmentation map (E) of the pigmented epithelial detachment observed in the B-scans (B, C) and the RPE segmentation map (D), along with the presence of subretinal fluid, better visual-ized on the B-scans (B, C) and retinal thickness map (E, yellow areas). The ninth injection of ranibizumab was given at this visit.

206 Ophthalmic Surgery, Lasers & Imaging Retina | Healio.com/OSLIRetina

DISCUSSION

This report describes the retrospective use of a novel SD-OCT algorithm in following up a patient with neovascular AMD and an associated PED under-going ranibizumab therapy. While receiving monthly ranibizumab therapy, the amount of subretinal fluid in the macula increased at every visit. By analyzing the scans retrospectively, we found that the volume mea-surements of the PED increased as well, as document-ed by the SD-OCT algorithm for the measurement of RPE elevations. Despite the adequate monthly treat-ment, the increasing amount of subretinal fluid alone served as a poor prognostic sign. However, one of the most dreaded outcomes associated with vascularized PEDs is the formation of a RPE tear and hemorrhage, and tears of the RPE often result from pre-existing PEDs. Approximately 12.5% of vascularized PEDs re-sult in a tear of the RPE.9 By monitoring the change in the volume of a PED, we may have found a useful new tool for following eyes with vascularized PEDs

undergoing treatment and assessing disease progres-sion. If the volume of a PED is found to increase, then this finding might serve as a useful predictor of eyes at risk for RPE tears and hemorrhage. While the in-creasing amount of subretinal fluid alone was an omi-nous sign, we might have been more insistent that the patient receive more aggressive ranibizumab therapy, such as biweekly injections,10 had we used this algo-rithm at the time the injections were being given.

The increasing height of a PED detected using OCT imaging has previously been described as a risk factor for a RPE tear.11 This change was difficult to ap-preciate on sequential visits using the conventional B-scans and maps obtained from the SD-OCT instru-ment. This new SD-OCT algorithm, which is now commercially available in version 6.0 of the Cirrus operating software (Carl Zeiss Meditec, Dublin, CA), demonstrated a volumetric increase even before the height had unambiguously changed. Moreover, the height may vary depending on whether the B-scan

Figure 3. Five monthly visits after the ninth injection in which a ranibizumab injection was given at each visit. Color images, horizontal foveal B-scans, RPE segmentation maps, retinal thickness maps, and RPE elevation maps are shown. The interval between each visit was approximately 30 days. The amount of subretinal fluid is shown to increase at each visit compared with the previous visit as depicted in the B-scans (A2, B2, C2, D2, E2) and in the retinal thickness maps (A4, B4, C4, D4, E4). An increase in the PED volume can be observed in the B-scans (A2, B2, C2, D2, E2), the RPE segmentation maps (A3, B3, C3, D3, E3), and especially in the RPE elevation maps (A5, B5, C5, D5, E5). The PED volume measurements were 0.17 mm3, 0.19 mm3, 0.20 mm3, 0.22 mm3, and 0.24 mm3 at each visit, respectively. Visual acuity was stable during this period, and an injection was given at each visit.

March/April 2013 • Vol. 44, No. 2 207

Figure 4. Extensive submacular hem-orrhage developing 30 days after the 14th ranibizumab injection. Color im-age of the right eye (A) and horizontal and vertical B-scans showing the large submacular hemorrhage (B,C).

was reproducibly positioned from visit to visit. In addition, the volume measurements may also detect lateral spread of a PED, which may not be evident on a single B-scan image. The automatic and reproduc-ible algorithm described in the manuscript allowed us to retrospectively identify the growth of the PED at each visit prior to the hemorrhage. This algorithm is a simpler and less ambiguous approach than the man-ual measurement of PED height using a single B-scan when following these patients undergoing treatment. While PED volume is just one of several components such as fluid, contractile elements, fibrovascular tis-sue, and proteinaceous material associated with the evolution of vascularized PEDs, the volume measure-ment may serve as a useful surrogate marker for all these associated components as they increase within the sub-RPE compartment. However, all these com-ponents should be studied to see if any one provides a better indicator of hemorrhage risk than volume alone.

In summary, SD-OCT imaging can be used to follow elevations of the RPE, and this new quantitative algo-rithm may be particularly useful when following and managing patients with vascularized PEDs. Although further studies are needed, the increase in the volume of a PED may serve as a useful predictor for disease progression and the need for aggressive therapy.

REFERENCES

1. Green WR, McDonnell PJ, Yeo JH. Pathologic features of senile macular degeneration. Ophthalmology. 1985;92(5):615-627.

2. Pauleikhoff D, Harper CA, Marshall J, Bird AC. Aging changes in Bruch’s membrane. A histochemical and morphologic study. Ophthalmology. 1990;97(2):171-178.

3. Pepple K, Mruthyunjaya P. Retinal pigment epithelial detachments in age-related macular degeneration: classification and therapeutic options. Semin Ophthalmol. 2011;26(3):198-208.

4. Casswell AG, Kohen D, Bird AC. Retinal pigment epithelial detachments in the elderly: classification and outcome. Br J Ophthalmol. 1985;69(6):397-403.

5. Age-Related Eye Disease Study Research G. The Age-Related Eye Disease Study system for classifying age-related macular degeneration from ste-reoscopic color fundus photographs: the Age-Related Eye Disease Study Report Number 6. Am J Ophthalmol. 2001;132(5):668-681.

6. Yehoshua Z, Wang F, Rosenfeld PJ, Penha FM, Feuer WJ, Gregori G. Natural History of Drusen Morphology in Age-Related Macular Degen-eration Using Spectral Domain Optical Coherence Tomography. Ophthal-mology. 2011; 118(12):2434-2441.

7. Gregori G, Wang F, Rosenfeld PJ, et al. Spectral domain optical coher-ence tomography imaging of drusen in nonexudative age-related macular degeneration. Ophthalmology. 2011;118(7):1373-1379.

8. Penha FM, Rosenfeld PJ, Gregori G, et al. Quantitative imaging of retinal pigment epithelial detachments using spectral-domain optical coherence tomography. Am J Ophthalmol. 2012;153(3):515-523.

9. Pauleikhoff D, Loffert D, Spital G, et al. Pigment epithelial detachment in the elderly. Clinical differentiation, natural course and pathogenetic impli-cations. Graefes Arch Clin Exp Ophthalmol. 2002;240(7):533-538.

10. Stewart MW, Rosenfeld PJ, Penha FM, et al. Pharmacokinetic rationale for dosing every 2 weeks versus 4 weeks with intravitreal ranibizumab, bevacizumab, and aflibercept (vascular endothelial growth factor Trap-eye). Retina. 2012;32:434-457.

11. Chiang A, Chang LK, Yu F, Sarraf D. Predictors of anti-VEGF-associated retinal pigmented epithelial tear using FA and OCT analysis. Retina. 2008; 28(9):1265-1269.

Figure 5. Right eye 4 months after the bleeding episode, showing near complete resolution of the submacular hemorrhage with resulting fibrovascu-lar scar. Color fundus image (A), hori-zontal and vertical B-scans through the fovea (B, C).