Purchase this article with an account.
S. Walia, P. S. Silva, J. D. Cavallerano, J. K. Sun, C. Dunn, S. E. Bursell, L. M. Aiello, L. P. Aiello; Comparison of MegaVision Low Light Digital Imaging With 35mm ETDRS 7-Standard Field Stereo Color Fundus Photographs and Clinical Examination. Invest. Ophthalmol. Vis. Sci. 2010;51(13):2282.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The MegaVision (Santa Barbara, CA) E1-i-RIC digital camera back is specially optimized for low light exposure nonmydriatic retinal imaging specific for the Joslin Vision Network (JVN) telemedicine program. The system provides 1000x1000 pixel resolution at low light level output of 1.7 watt-sec or 1.3 mW per exposure (~5x lower than other commercially available systems). Low light levels improve patient comfort and allow shorter imaging time. This study compares clinical severity of diabetic retinopathy (DR) and macular edema (DME) read from MegaVision E1-i-RIC images (MVI) to both ETDRS photographs and clinical examination (exam) and compares subject comfort and image acquisition time with traditional mydriatic imaging.
Nonmydriatic MVI were obtained according to previously published JVN protocol. ETDRS photos were obtained after pupil dilation and fundus examination was performed by a retina specialist. Masked readers graded the image sets independently with adjudication by an independent retina specialist.
In 126 eyes of 67 diabetic patients, DR was detected in 109 eyes (86%) by MVI, 112 eyes (89%) by ETDRS photos and 114 eyes (90%) by exam. There was substantial agreement between MVI and ETDRS photos for DR severity (weighted Κ = 0.68) with exact agreement in 75 (60%) and within 1 level of DR in 107 eyes (85%). Similar correlation existed with exam (wΚ = 0.79). MVI tended to overdiagnose severe NPDR (27% vs 12%) and underdiagnose early PDR (6% vs 11%); however, 90% of undiagnosed PDR was present outside the field covered by the MVI, resulting in 80% exact match and 100% match within 1 step for PDR in the fields covered by MVI. DME severity on MVI had moderate correlation with both ETDRS photos (wΚ = 0.51) and exam (wΚ = 0.48). MVI from 4 eyes (3%) could not be graded for DR and 15 eyes (12%) could not be graded for DME. Median time to acquire and sort 22 MVI of both eyes was 10 minutes, compared with 27 minutes to dilate and acquire ETDRS photos. In a substudy of 10 subjects, 60% considered MVI more comfortable than nonmydriatic imaging using 3.3 watt-sec and 30% felt comfort level was the same.
Images captured using a low light digital camera compared favorably with ETDRS photography and clinical examination for grading DR/DME. Furthermore, this imaging was faster and more comfortable as compared with standard mydriatic ETDRS fundus photography.
This PDF is available to Subscribers Only