Non-Mydriatic Ultra-Widefield Imaging Compared With Single-Field Imaging in the Evaluation of Peripheral Retinal Pathology

Size: px

Start display at page:

Download "Non-Mydriatic Ultra-Widefield Imaging Compared With Single-Field Imaging in the Evaluation of Peripheral Retinal Pathology"

Asher Jenkins
6 years ago
Views:

1 CLINICAL SCIENCE Non-Mydriatic Ultra-Widefield Imaging Compared With Single-Field Imaging in the Evaluation of Peripheral Retinal Pathology Mehreen Adhi, MD; Fabiana Q. Silva, MD; Richard Lang, MD, MPH; Raul Seballos, MD; Roxanne B. Sukol, MD; Steven Feinleib, MD; Rishi P. Singh, MD INTRODUCTION BACKGROUND AND OBJECTIVE: To report clinical feasibility of non-mydriatic ultra-widefield (NMU- WF) imaging and determine the prevalence of peripheral retinal pathology in comparison to standard single-field imaging in a primary care setting. PATIENTS AND METHODS: Six hundred and thirtytwo subjects (1,260 eyes) who underwent NMUWF imaging during annual health screening from October 2015 through March 2016 were retrospectively identified. An automated algorithm processed the raw images into: (1) NMUWF image with mask/grid outline that delineates the center 45 field simulating standard single-field photograph and (2) singlefield image comprising 45 posterior pole extracted from the corresponding NMUWF image. RESULTS: Mean age of patients was 59.6 years ± 7.5 years. Of the 1,260 eyes, 1,238 eyes (98.3%) were considered optimum for grading. NMUWF images detected peripheral retinal pathology in 228 eyes (18.4%) that were not visible on corresponding single-field images. CONCLUSIONS: NMUWF imaging is feasible in a primary care setting, allows improved visualization of peripheral retinal pathology, and may therefore be useful for telemedicine screening. [Ophthalmic Surg Lasers Imaging Retina. 2017;48: ] Retinal imaging is an important adjunct to the diagnosis, management, and monitoring of various posterior segment conditions. Traditionally, singlefield fundus photography is used to evaluate and monitor diseases primarily involving the posterior pole, such as macular degeneration, glaucoma, and optic neuropathy, among others. It effectively captures the posterior pole, including the macula and optic nerve, covering a 20 to 50 field of view. It is also routinely used for the diagnosis, treatment, and monitoring of posterior segment conditions that may extend beyond the posterior pole, such as diabetic retinopathy, choroidal nevi, retinal tears, and retinal vascular occlusions. This modality, however, is limited in that it is unable to adequately image the retinal periphery that is an important site of pathology in a multitude of diseases. To date, the evaluation of the retinal periphery relies on a dilated fundus examination using indirect ophthalmoscopy with scleral depression. However, during the past two decades, there have been significant advances in imaging of the retinal periphery using widefield fundus imaging through hardware and software innovations. In the early 1990s, the Early Treatment Diabetic Retinopathy Study (ETDRS) detailed a systematic protocol of acquiring a series of seven single-field fundus images each covering a 30 retinal field that were then montaged manually to create an approximately 75 field of view up to the midperiphery to classify diabetic retinopathy. 1,2 However, this protocol is limited in a clinical setting in that From Cole Eye Institute, Cleveland Clinic, Cleveland (MA, FQS, RPS); the Department of Ophthalmology and Visual Sciences, Kentucky Lions Eye Center, University of Louisville School of Medicine, Louisville, KY (MA); and the Department of Preventive Medicine, Cleveland Clinic, Cleveland (RL, RS, RBS, SF). Originally submitted March 14, Revision received March 14, Accepted for publication June 2, This study was presented at the Association for Research in Vision and Ophthalmology Annual Meeting, May 7-11, 2017, in Baltimore. Optos provided the camera and software for the analysis. The sponsor did not participate in the design, execution, data review, or publishing of this manuscript. Dr. Singh has received grants and personal fees from Genentech and Regeneron; grants from Alcon and Apellis; and personal fees from Shire, Optos, and Zeiss during the conduct of the study. The remaining authors report no relevant financial disclosures. Address correspondence to Rishi P. Singh, MD, 9500 Euclid Avenue, Desk i32, Cleveland, OH 44195; SINGHR@ccf.org. doi: / Ophthalmic Surgery, Lasers & Imaging Retina Healio.com/OSLIRetina

2 Figure 1. (A) Raw image extracted from the Optos ProView software before processing. (B) Image aligned for processing using an automated algorithm. (C) Image with mask/grid outline that comprises the ultra-widefield fundus image with a superimposed outline that delineates a center 45 field processed using an automated algorithm. (D) Image with mask/grid that extracts the 45 center field mimicking a standard single-field fundus photograph processed using an automated algorithm. there is need for excellent pupillary dilation, relatively clear media, and good patient cooperation for fixating appropriately to image the desired peripheral retinal location. It also relies heavily on trained photographers to obtain high-quality images and manual creation of the montage and/or independent evaluation of each of the seven single-field images, which can be both time-consuming as well prone to errors. Even so, although this protocol is able to provide visualization of the fundus up to the midperiphery, it is unable to image the far periphery of the retina. During the past decades, several innovative approaches to widefield fundus imaging have been introduced. These include the Pomerantzeff camera, the Panoret-100, the Staurenghi lens, and the RetCam3 system (Clarity Medical Systems, Pleasanton, CA). 3-8 These devices can provide 100 to 160 panoramic photographs using either traditional fundus photography or confocal scanning laser ophthalmoscopy (cslo). However, these platforms are limited by the need of a contact lens that requires a skilled photographer to hold the camera and lens in place during image acquisition. Recently, two non-contact ultra-widefield imaging systems have been introduced: the Heidelberg Spectralis (Heidelberg Engineering, Heidelberg, Germany) and the Optos (Optos, Dunfermline, UK). 9 The Optos imaging system is able to obtain a 200 internal view of the retina, representing approximately 82.5% of the total retinal surface in a single capture. This device uses cslo technology and a unique ellipsoid mirror to create a virtual focal point inside the eye that enables non-contact imaging of the central and peripheral retina in tandem through a single image without a need for pupillary dilation. It uses two wavelength lasers (red 633 nm and green 532 nm) instead of the full-spectrum white light to enhance visualization of retinal tissue. To date, it offers the widest imaging capability of any available digital retinal imaging system. 9 Using non-mydriatic ultra-widefield imaging (NMUWF), visualization of peripheral retinal pathology and its correlation with various chorioretinal conditions and their treatment has been extensively investigated recently The present study aims to analyze the feasibility of performing NMUWF in a primary care setting and to determine the prevalence of peripheral retinal pathology in comparison to standard single-field fundus photography. PATIENTS AND METHODS Study Design This retrospective study was performed at the Cleveland Clinic Cole Eye Institute (Cleveland, OH) after approval from the Cleveland Clinic Investigational Review Board. Because of the retrospective nature of the study with minimal patient risk, written informed consent was considered exempt. All studyrelated procedures were performed in accordance with best practices and adhered to the Health Insurance Portability and Accountability Act. Participants Healthy subjects who underwent NMUWF imaging during their annual primary care visit at Cleveland Clinic from October 1, 2015, through March 31, 2016, were included in this study. A total of 632 subjects (1,260 eyes) were identified. The demographic and clinical data including age, gender, race, and comorbidities (including diabetes, hypertension, and hyperlipidemia) were extracted from the electronic health record system. Non-Mydriatic Ultra-Widefield Fundus Imaging All subjects underwent NMUWF imaging using the Optos Daytona (Optos, Dunfermline, UK) device as part of their annual primary care evaluation. A medical assistant (non-ophthalmic technician) that December 2017 Vol. 48, No

3 Figure 2. (A1, B1, C1, D1) Single-field fundus images of four subjects. (A2, B2, C2, D2) Corresponding ultra-widefield images with mask/grid outlines showing peripheral retinal pathologies that were not visible on the single-field fundus images ([A2] Choroidal nevus with drusen/ possible melanoma; [B2] Chorioretinal scar; [C2] Chorioretinal scars; [D2] Horseshoe tear and laser scars). 964 Ophthalmic Surgery, Lasers & Imaging Retina Healio.com/OSLIRetina

4 TABLE 1 Demographic and Clinical Characteristics of the Study Cohort Total n (Patients; Eyes) 632; 1,260 Age (Mean ± SD) Gender (n; %) 59.6 years ± 7.5 years Female 111; 17.5% Male 521; 82.5% Race (n; %) White 516; 81.6% Black 60; 9.5% Hispanic 28; 4.5% Asian/Pacific Islander 28; 4.5% Comorbidities (n; %) Diabetes mellitus 60; 9.5% Hypertension 54; 8.5% Hyperlipidemia 157; 24.8% Renal impairment 11; 1.7% SD = standard deviation was trained to perform the photography during a 3- to 4-hour training session performed all the imaging. Image Processing and Analysis The NMUWF images of all eyes were extracted from the Optos ProView software and processed using an automated algorithm that first aligns the images and then creates two sets of images from each eye: (1) image with mask/grid outline that comprises the NMUWF image with a superimposed outline simulating the size of a standard single-field fundus photograph (45 field) delineating the posterior pole (macula and the optic nerve) from the peripheral retina, and (2) image with mask/grid that extracts the center 45 field comprising the posterior pole that resembles a standard single-field fundus photograph and masks the peripheral retina entirely (single-field image) (Figure 1). To be considered fully gradable, the NMUWF images needed to have at least a 50% or more peripheral view beyond the superimposed outline at 45 delineating the posterior pole (total: approximately 145 to 150 field of view or more, including the posterior pole). NMUWF images that had some obscuration of the retinal field due to suboptimum resolution secondary to motion and/or eyelash artifacts (up to 25% of the fundus view within the posterior pole) but at least a 25% peripheral retinal field visible beyond the outline (total: 95 to 100 field of view or more, including the posterior pole) were considered gradable. Those that had obscuration of more than 25% of the fundus view within the posterior pole and/or less than 25% peripheral retinal field visible beyond the outline were considered non-gradable. Single-field images were considered fully gradable when a complete 45 view of the posterior pole was visible, gradable when less than 25% view of the posterior pole was obscured owing to suboptimum resolution secondary to motion/eyelash artifacts, and non-gradable when more than 25% view of the posterior pole was obscured. Two independent readers (MA, FQS) analyzed the NMUWF images and the single-field images from each eye to characterize the peripheral retinal findings and determine their prevalence. RESULTS A total of 632 subjects (1,260 eyes) were included in the study. The demographic and clinical characteristics of the subjects are presented in Table 1. Of the 1,260 eyes, one eye failed alignment during image processing and, hence, could not be analyzed due to unavailability of its corresponding NMUWF and single-field fundus images. In addition, 21 eyes (21 of 1,260; 1.7%) were considered non-gradable according to the criteria for grading described above. The images of the remaining 1,238 eyes (1,238 of 1,260; 98.3%) were gradable, demonstrating the feasibility December 2017 Vol. 48, No

5 TABLE 2 Peripheral Retinal Findings Observed and Their Prevalence in the Study Cohort Peripheral Findings Number of Eyes Percent Prevalence Dot / blot hemorrhage % White without pressure % Lattice degeneration % Choroidal nevus % Paving stones degeneration % Drusen % RPE hypertrophy % Diffuse RPE atrophy 12 1% Chorioretinal atrophy 9 0.7% Retinal tear 9 0.7% Scleral buckle 5 0.4% Floater 4 0.3% Chorioretinal scars 3 0.2% Bear tracks 3 0.2% Snail track degeneration 2 0.2% Ghost vessels 1 0.1% Retinoschisis 1 0.1% Bones spicles 1 0.1% Choroidal nevus/ possible melanoma 1 0.1% Total 228 (of 1,238 eyes) 18.4% RPE = retinal pigment epithelium of performing NMUWF imaging by a medical assistant (nonophthalmic technician) in a primary care setting, and were used for the purposes of analysis. The NMUWF images were able to detect retinal pathology in 228 of 1,238 eyes (18.4%) that were not visible on the corresponding single-field images. These pathologies included dot-blot hemorrhages (44 of 1,238; 3.5%), peripheral retinal degenerations (48 of 1,238; 3.9%), choroidal nevi (23 of 1,238; 1.9%), drusen (16 of 1,238; 1.3%), retinal tear (nine of 1,238; 0.7%), chorioretinal atrophy (nine of 1,238; 0.7%), choroidal lesion suspicious of melanoma (one of 1,238; 0.1%) and retinoschisis (one of 1,238; 0.1%), among others (Figure 2). Table 2 delineates the various peripheral retinal pathologies and findings observed with their corresponding prevalence in the study cohort. DISCUSSION This study presents the utility of NMUWF fundus imaging comprising up to 200 field of view compared with standard single-field fundus imaging (45 field of view) for evaluation of peripheral retinal pathology in a large cohort of healthy subjects undergoing imaging as part of their annual health screening in a nonophthalmological setting. It demonstrates the clinical feasibility of NMUWF imaging in a wide variety of clinical conditions. The NMUWF images had excellent quality and resolution (Figure 2) with only 1.7% of non-gradable images (21/1,260 eyes). Since the device used does not require contact lens, image acquisition can be adequately performed by a nonophthalmic technician as demonstrated in the present study. The clinical feasibility of NMU- WF imaging in varied clinical settings has also been reported previously by Nagiel et al. in a review that illustrates its utility in a variety of disorders. 21 Klufas et al. has also reported the feasibility of NMUWF imaging in a study using NMUWF indocyanine angiography with the Optos system. 22 Additionally, NMUWF imaging revealed abnormalities in the peripheral retina that may otherwise 966 Ophthalmic Surgery, Lasers & Imaging Retina Healio.com/OSLIRetina

6 be missed using conventional fundus photography. The present study shows that up to 18.4% of healthy subjects have peripheral retinal findings that may not be delineated using standard singlefield fundus photography. Though some of these findings may be non-referable, others may require careful monitoring with a dilated fundus examination and further ancillary testing in an ophthalmological setting. Although other devices to image the fundus are available, to date, there is no other imaging modality that allows single-image analysis of the fundus as far out in the periphery as offered by the Optos NMUWF imaging device. Recent studies have validated the clinical utility of NMUWF imaging for diagnosis, follow-up, and documentation of peripheral retinal pathologies ,23,24 In a prospective study of 380 diabetic patients (759 eyes), Wilson et al. found a 83.6% sensitivity for NMUWF images in identifying diabetic retinopathy when compared to a sensitivity of 82.9% for single-field images and 82.9% for dualfield images (P >.2), suggesting that NMUWF imaging can be appropriately used for screening purposes. 23 NMUWF imaging is also a useful adjunct to clinical examination for characterizing retinal detachments and detecting peripheral degenerative changes. In a retrospective study that evaluated the usefulness of NMUWF imaging to identify peripheral retinal lesions in patients with myopia in a clinical setting, Lee et al. reported a 90.8% diagnostic sensitivity in patients with myopia, 91.0% in patients with high myopia, and 90.9% in totally myopic patients. 24 In conclusion, NMUWF technology can be very useful for telemedicine screening and disease management programs especially in the underserved areas. Telemedicine based diagnosis of retinal diseases has already been validated in various studies for retinopathy of prematurity, diabetic retinopathy, and cytomegalovirus retinitis using conventional standard field fundus photography The use of NMUWF imaging for this purpose would enable a nonophthalmic technician to perform non-contact photography of a wider field than traditional fundus imaging without need for mydriasis and allow improved visualization of peripheral retinal pathology. REFERENCES 1. No authors listed. Fundus photographic risk factors for progression of diabetic retinopathy. ETDRS report number 12. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology. 1991;98(Suppl 5): No authors listed. Grading diabetic retinopathy from stereoscopic color fundus photographs an extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology. 1991;98(Suppl 5): Pomerantzeff O. Equator-plus camera. Invest Ophthalmol. 1975;14(5): Shields CL, Materin M, Shields JA. Panoramic imaging of the ocular fundus. Arch Ophthalmol. 2003;121(11): Staurenghi G, Viola F, Mainster MA, Graham RD, Harrington PG. Scanning laser ophthalmoscopy and angiography with a wide-field contact lens system. Arch Ophthalmol. 2005;123(2): Reeves GM, Kumar N, Beare NA, Pearce IA. Use of Staurenghi lens angiography in the management of posterior uveitis. Acta Ophthalmol. 2013;91(1): Mantel I, Schalenbourg A, Zografos L. Peripheral exudative hemorrhagic chorioretinopathy: Polypoidal choroidal vasculopathy and hemodynamic modifications. Am J Ophthalmol. 2013;153(5): e2. 8. Dhaliwal C, Wright E, Graham C, McIntosh N, Fleck BW. Widefield digital retinal imaging versus binocular indirect ophthalmoscopy for retinopathy of prematurity screening: A two-observer prospective, randomized comparison. Br J Ophthalmol. 2009;93(3): Witmer MT, Parlitsis G, Patel S, Kiss S. Comparison of ultrawidefield fluorescein angiography with the Heidelberg Spectralis noncontact ultra-widefield module versus the Optos Optomap. Clin Ophthalmol. 2013;7: Nicholson BP, Nigam D, Miller D, et al. Comparison of wide-field fluorescein angiography and nine-field montage angiography in uveitis. Am J Ophthalmol. 2014;157(3): Wessel MM, Aaker GD, Parlitsis G, Cho M, D Amico DJ, Kiss S. Ultra-wide-field angiography improves the detection and classification of diabetic retinopathy. Retina. 2012;32(4): Silva PS, Cavallerano JD, Sun JK, Soliman AZ, Aiello LM, Aiello LP. Peripheral lesions identified by mydriatic ultrawide field imaging: Distribution and potential impact on diabetic retinopathy severity. Ophthalmology. 2013;120(12): Liegl R, Liegl K, Ceklic L, et al. Nonmydriatic ultra-wide-field scanning laser ophthalmoscopy (optomap) versus two-field fundus photography in diabetic retinopathy. Ophthalmologica. 2014;231(1): Tsui I, Franco-Cardenas V, Hubschman J, Yu F, Schwartz SD. Ultra wide field fluorescein angiography can detect macular pathology in central retinal vein occlusion. Ophthalmic Surg Lasers Imaging. 2012;43(3): Brown K, Sewell JM, Trempe C, Peto T, Travison TG. Comparison of image-assisted versus traditional fundus examination. Eye Brain. 2013;5(1): Leder HA, Campbell JP, Sepah YJ, et al. Ultra-wide-field retinal imaging in the management of non-infectious retinal vasculitis. J Ophth Inflamm Infect. 2013;3(1): Singer M, Tan CS, Bell D, et al. Area of peripheral retinal nonperfusion and treatment response in branch and central retinal vein occlusion. Retina. 2014;34(9): Campbell JP, Leder HA, Sepah YJ, et al. Wide-field retinal imaging in the management of noninfectious posterior uveitis. Am J Ophthalmol. 2012;154(5): e Kernt M, Hadi I, Pinter F, et al. Assessment of diabetic retinopathy using nonmydriatic ultra-widefield scanning laser ophthalmoscopy (optomap) compared with ETDRS 7-field stereo photography. Diabetes Care. 2012;35(12): Silva PS, Cavallerano JD, Sun JK, Noble J, Aiello LM, Aiello LP. Nonmydriatic ultrawide field retinal imaging compared with dilated standard 7-field 35-mm photography and retinal specialist examination for evaluation of diabetic retinopathy. Am J Ophthalmol. 2012;154(3): e Nagiel A, Lalane RA, Sadda SR, Schwartz SD. Ultra-widefield fundus imaging: A review of clinical applications and future trends. Retina. 2016;36(4): Klufas MA, Yannuzzi NA, Pang CE, et al. Feasibility and clinical utility of ultra-widefield indocyanine green angiography. Retina. 2015;35(3): Wilson PJ, Ellis JD, MacEwen CJ, Ellingford A, Talbot J, Leese December 2017 Vol. 48, No

7 GP. Screening for diabetic retinopathy: A comparative trial of photography and scanning laser ophthalmoscopy. Ophthalmologica. 2010;224: Lee DH, Kim SS, Kim M, Koh HJ. Identifiable peripheral retinal lesions using ultra-widefield scanning laser ophthalmoscope and its usefulness in myopic patients. J Korean Ophthalmol Soc. 2014;55(12): Athikarisamy SE, Patole S, Lam GC, et al. Screening for retinopathy of prematurity (ROP) using wide-angle digital retinal photography by non-ophthalmologists: A systematic review. Br J Ophthalmol. 2015;99(3): Vinekar A, Gilbert C, Dogra M, et al. The KIDROP model of combining strategies for providing retinopathy of prematurity screening in underserved areas in India using wide-field imaging, tele-medicine, non-physician graders and smart phone reporting. Indian J Ophthalmol. 2014;62(1): Fijalkowski N, Zheng LL, Henderson MT, et al. Stanford University Network for Diagnosis of Retinopathy of Prematurity (SUNDROP): Five years of screening with telemedicine. Ophthalmic Surg Lasers Imaging Retina. 2014;45(2): Weaver DT. Telemedicine for retinopathy of prematurity. Curr Opin Ophthalmol. 2013;24(5): Quinn GE, Ying GS, Daniel E, et al. Validity of a telemedicine system for the evaluation of acute-phase retinopathy of prematurity. JAMA Ophthalmol. 2014;132(10): Owsley C, McGwin G, Lee DJ, et al. Diabetes eye screening in urban settings serving minority populations: detection of diabetic retinopathy and other ocular findings using telemedicine. JAMA Ophthalmol. 2015;133(2): Jirawison C, Yen M, Leenasirimakul P, et al. Telemedicine screening for cytomegalovirus retinitis at the point of care for human immunodeficiency virus infection. JAMA Ophthalmol. 2015;133: Yen M, Ausayakhun S, Chen J, et al. Telemedicine diagnosis of cytomegalovirus retinitis by nonophthalmologists. JAMA Ophthalmol. 2014;132(9): Ophthalmic Surgery, Lasers & Imaging Retina Healio.com/OSLIRetina

The development and implementation of ocular

The development and implementation of ocular Ocular Telehealth: A Wider View of the Need and the Opportunity in Ophthalmology Ultra-widefield retinal imaging will help ophthalmology address growing needs. BY SHACHAR TAUBER, MD The development and