Purchase this article with an account.
Tyson N. Kim, Suman S Thapa, Indira Paudyal, Leslie Niziol, Shreya Shrestha, Ian F Pitha, Pradeep Y Ramulu, Alan L Robin, Maria A Woodward, Paula Anne Newman-Casey; Portable non-mydriatic imaging: a useful tool in a tele-glaucoma assessment. Invest. Ophthalmol. Vis. Sci. 201657(12):.
Download citation file:
© 2017 Association for Research in Vision and Ophthalmology.
Portable fundus photography has the potential to improve glaucoma care in underserved populations through telemedicine. We compared images between two fundus cameras: a portable non-mydriatic Pictor [Volk, Mentor, OH] and a “gold-standard” table-top mydriatic Topcon [Topcon, Oakland, NJ] to determine the reliability and accuracy of portable fundus photography of the optic nerve in rural Nepal.
We performed a cross-sectional study of 422 eyes of 211 new patients recruited from the glaucoma clinic at the Tilganga Institute of Ophthalmology (Kathmandu, Nepal). A glaucoma specialist evaluated all subjects using a dilated fundus exam along with Sita Standard Humphrey Perimetry to establish the diagnosis of glaucoma (gold standard). Fundus photographs were taken using both a Pictor hand-held camera before dilation and a Topcon tabletop fundus camera after dilation for comparison. Two masked glaucoma specialists (PANC; IP) graded images for the cup-to-disc ratio (CDR), and for a diagnosis of glaucoma based on a CDR ≥0.7, or the presence of a notch or disc hemorrhage. Agreement between CDR measurements was assessed between cameras (for the same grader) and between graders (for the same camera). Remote readers’ diagnosis of glaucoma was compared to the true clinical diagnosis to estimate sensitivity and specificity.
Of 422 eyes, 196 (46.5%) were diagnosed with glaucoma and 226 (53.5%) were diagnosed as normal. When comparing the Topcon and Pictor images, Reader 1 had CDR measurements within +/- 0.2 units 87% and 88% of the time for the first and second read, respectively, and Reader 2 had CDR measurements within +/- 0.2 units 91% and 91% of the time. Reader 1 had CDR measurements within +/- 0.1 units 67% and 68% of time for the first and second read, respectively, and Reader 2 had CDR measurements within +/- 0.2 units 72% and 76% of the time. The Pictor had a sensitivity of 48.5-61.5% and specificity of 44.1-62.8% to remotely diagnose glaucoma when compared with the clinical examination. Similarly, the Topcon had a sensitivity of 47.5-53.9% and specificity of 55.6-64.2% when compared with clinical examination.
We demonstrate high fidelity between cup-to-disc measurements using a hand-held portable, non-mydriatic camera and an established mydriatic fundus camera, though utilizing cup-to-disc measurements alone to diagnose glaucoma without ancillary testing had modest sensitivity and specificity.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only