Quantifying the Afferent Pupillary Defect: The Afferent Pupillary Response as Measured by Automated Pupillometry in Relation to Inter-Eye Average Retinal Nerve Fiber Layer Thickness in Glaucoma

Abstract

Purpose : Glaucoma may often lead to an assymetric optic neuropathy, thereby resulting in a central disparity of afferent input and a resultant clinically detectable pupillary defect. Here we investigate correlations between quantitative differences between the inter-eye afferent pupillary response and inter-eye differences in average retinal nerve fiber layer (RFNL) thickness.

Methods : This cross sectional study included 19 patients with a known diagnosis of primary open angle glaucoma. Patients had an RFNL SD-OCT performed within 6 months of a clinic visit during which measurement with a pupillometer was performed by a single clinician along with clinical judgement of APD. Patients were excluded if they had poor visual acuity (20/100 or worse in either eye) or other major eye diseases (i.e. central nervous system disease, history of diabetic retinopathy). Pupillary constriction amplitudes were measured and an RAPD score was assigned by calculation of percent change in pupil size:
RAPD Score = [Pupil Diameter (resting) – Pupil Diameter (constricted)]/Pupil Diameter (constricted)
Inter-eye differences between RAPD score and inter-eye differences in SD-OCT RFNL thickness were measured and correlated using the Spearman correlation coefficient, with p-value measured by two-tailed t-test.

Results : A linear correlation was found between asymmetric RFNL thinning and the afferent pupillary response (see Figure 1). Notably, inter-eye differences in RFNL thickness correlated with both RAPD and clinical grading of APD. However, positive measurement of RAPD score was more sensitive in detecting asymmetry in RFNL thickness in values < 15µm (Figure 2).

Conclusions : Automated pupillometry is a facile means of quantifying asymmetry in the afferent system, the results of which correlates with SD-OCT, an established test for monitoring glaucoma. Automated pupillometry was also more reliable and consistent than clinical judgement of APD in cases of subtle asymmetry in OCT RFNL thickness.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.

 

Figure 1: Inter-eye difference in RAPD score versus inter-eye difference in average RFNL thickness. Spearman correlation coefficient Rs = 0.7769, P = 0.000091

 

Figure 2: Grading of APD (blue diamonds) as compared to pupilometer-measured Inter-eye difference in RAPD score (red squares) as correlated with inter-eye difference in average RFNL thickness.