The Prevalence of Relative Afferent Pupillary Defects in Patients with Normal Visual Fields

Abstract

Purpose: To determine the frequency with which a relative afferent pupillary defect (RAPD) occurs in patients with glaucoma but without VF loss when tested with SAP.

Methods: A retrospective study: The electronic medical records of 672 patients who presented to the Wills Eye Glaucoma Service from January to May 2012 were reviewed. All patients were included except those with: 1) Any condition preventing adequate visualization and examination of the pupil or optic nerve 2) Active infection of the anterior or posterior segments of the eye 3) Any intraocular surgical or laser procedure within the previous 4 weeks 4) Any cause other than glaucoma known to cause an RAPD 5) Only one eye, or fixed pupils Clinical parameters such as VA, IOP, C/D ratio, DDLS, or VF mean defect within 6 months of the documented RAPD testing were tabulated. A documented RAPD was considered a “true” RAPD if it correlated with inter-eye structural asymmetry (such as visual field mean defect, DDLS score, or cup/disc ratio). The visual fields were measured either with Humphrey or Octopus perimetry. Normal VF (Humphrey) was defined as an MD score of -6dB or more positive, WITHOUT any of the following: 3 adjacent data points of <5% with at least one point <1%; PSD significant at <.05 (or 5%); Glaucoma hemifield test "Outside normal limits". For Octopus, normal VF was defined as an MD score of -0.8 or more negative, with a Bebie curve line not intersecting with the lower confidence limit line.

Results: The analyzed sample size was 152, 28 of whom had visual field via Humphrey and 124 via Octopus. Among 152 analyzed subjects, 89 were RAPD negative, 63 were RAPD positive. Among the 63 positive subjects, 60 subjects had a “true” RAPD. There was high agreement between RAPD and "true" RAPD (kappa=0.96). 85% of the patients (Humphrey testing) had abnormal VF and 93% by Octopus. 55% of patients with abnormal VF (Humphrey) had a positive RAPD; 50% of those with normal VF had a positive RAPD. 39% of patients (Octopus) with abnormal VF had a positive RAPD, while 33% of those with normal VF had a positive RAPD. There was no association between presence of an RAPD and whether or not the field was normal (p=0.40 - Humphrey) (p=0.27 - Octopus).

Conclusions: Around 1/3 to 1/2 of patients in a tertiary glaucoma practice have a positive RAPD in the absence of a visual field defect being detected by Octopus or Humphrey perimetry.

Keywords: 463 clinical (human) or epidemiologic studies: prevalence/incidence • 668 pupillary reflex • 758 visual fields