Per-region visual field sensitivities and delays in AMD

Bhim Rai, Rohan Essex, Faran Sabeti, Ted Maddess, Emilie Rohan, Josh van Kleef, Corinne Carle

Research output: Contribution to journalMeeting Abstractpeer-review


Purpose : To compare central vs. peripheral visual field sensitivities and delays in neovascular AMD (nAMD) and earlier-stage fellow eyes using FDA-cleared multifocal pupillographic objective perimetry (mfPOP).

Methods : We recruited 18 nAMD patients testing both eyes concurrently. Practitioner directed intravitreal anti-VEGF injection was administered monthly for 14 to 28 visits. Diagnostic procedures included mfPOP, Matrix 10-2 perimetry, and OCT. The macular mfPOP variant had 44 interleaved stimuli/eye presented from fixation to 15o eccentricity. We examined correlations between central and peripheral mfPOP per-region sensitivities and delays within eyes, and between nAMD eyes and untreated fellow eyes.

Results : Twenty-three eyes of 18 patients were tested (14 females, 77.8%). In treated eyes central sensitivity decreased over time by -2.23 ± 0.051 dB/month (p
Conclusions : We observed decreased central sensitivity, and coincident peripheral hypersensitivity. In the untreated fellow eyes, the central sensitivity decreased and delay increased progressively until treatment was needed. Based on peripheral responses the eyes could be divided into positive and negative groups. The predictive value of mfPOP improved after the positive and negative eyes were analysed separately. Diagnostic sensitivity and delay changes could precede active disease by 1 to 3 months. We conclude that mfPOP may be a potential biomarker to predict the need for anti-VEGF injections in nAMD.

This is a 2021 ARVO Annual Meeting abstract.
Original languageEnglish
Pages (from-to)317-317
Number of pages1
JournalInvestigative ophthalmology & visual science
Issue number8
Publication statusPublished - Jun 2021


Dive into the research topics of 'Per-region visual field sensitivities and delays in AMD'. Together they form a unique fingerprint.

Cite this