Clinical Utility of Multifocal Pupillographic Objective Perimetry in Type 1 Diabetes

Abstract

 

Purpose

 

To evaluate the sensitivity and specificity of five variants of multifocal pupillographic objective perimetry (mfPOP) for discriminating controls from patients with type 1 diabetes (T1D) presenting with nil to non-proliferative diabetic retinopathy (NPDR).

 

Methods

 

25 T1D patients (37.9 ± 2.44 years, 12 females, mean duration 22.5 ± 10.1 years) and 23 normal (36.0 ± 2.58 years, 12 females) subjects were recruited. Pupillary contraction amplitudes and times-to-peak responses to red/green, yellow and blue dichoptic stimuli with 44 test regions/eye extending, to either ±15° or ±30°. Disease severity was based on either ETDRS scores from fundus photos; or scores based or factors from correlations between 18 patient variables, which were mainly based on blood analyses, but included peripheral neuropathy, and advanced glycation end-products.

 

Results

 

Factor scores based mainly on renal function, HBA1c and BMI were the best indicator of diabetic damage as measured by mfPOP. In eyes with no or mild NPDR, red/green mfPOP stimuli extending to ±30° achieved the highest ROC area under the curve (AUC) of 0.95 ± 0.05 (mean ± SE) for delay deviations and 0.91 ± 0.05 when amplitude deviations were considered. In moderate NPDR eyes diagnostic accuracy improved to 1.00 ± 0.05 for blue and yellow protocols.

 

Conclusions

 

The results suggest that red/green stimuli presented within the central 30° of fixation achieved good diagnostic power to determine diabetes status in eyes with minimal to no retinopathy. MfPOP may be a useful clinical tool in evaluating and monitoring eyes with early NPDR.

 

Keywords: 499 diabetic retinopathy • 642 perimetry • 667 pupil