The diagnostic accuracy of visual stimuli in multifocal pupillographic objective perimetry (mfPOP) for detection of early-stage diabetic retinopathy — ASN Events

The diagnostic accuracy of visual stimuli in multifocal pupillographic objective perimetry (mfPOP) for detection of early-stage diabetic retinopathy (#321)

Lachlan J Cormick 1 , Faran Sabeti 2 , Christopher J Nolan 3 , Rohan W Essex 2 , Corinne F Carle 2 , Andrew C James 2 , Rakesh Mallikarjunan 2 , Ted Maddess 2
  1. The Australian National University Medical School, Canberra, ACT, Australia
  2. John Curtin School of Medical Research, The Australian National University, , Canberra, ACT, Australia
  3. Endocrinology, The Canberra Hospital, Canberra, ACT, Australia

Aim: Multifocal pupillographic objective perimetry (mfPOP) is a functional eye test, which has potential use in detecting early-stage diabetic retinopathy (DR). The aim of this study was to determine which of three different mfPOP stimulus protocols had the greatest diagnostic power in detecting early-stage DR in type 2 diabetes (T2D) subjects.

Methods: Thirty-five T2D subjects with varying levels of non-proliferative DR (NPDR) and twenty age-matched controls were tested using three different mfPOP stimuli, varying in eccentricity and colour. The protocols tested pupil responses to stimuli at 44 regions per eye, measuring amplitude and time to peak contraction of the pupil. The diagnostic power of mfPOP was determined by receiver operator characteristics (ROC).

Results: Analysis showed that the Macular protocol consistently had the greatest diagnostic accuracy in detecting severity of early-stage DR, achieving an ROC area under the curve (AUC) of 100% ± 0 (mean ± SE). High ROC-AUC values were achieved when comparing controls to diabetics without DR with a maximum of 89.67% ± 3.21.

Conclusion: mfPOP has good diagnostic accuracy for detecting early-stage DR and may become a useful test in providing clinical information for the diagnosis and staging of early-stage DR.