Rates of Retinal Nerve Fiber Layer Loss in Contralateral Eyes of Glaucoma Patients with Unilateral Progression by Conventional Methods

Abstract

Purpose: To determine whether progressive retinal nerve fiber layer (RNFL) loss occurs in the contralateral eye of patients with glaucoma showing unilateral progression according to conventional diagnostic methods.

Design: Prospective, longitudinal, observational cohort study.

Participants: Three hundred forty-six eyes of 173 patients (118 eyes with glaucoma and 228 eyes with suspect glaucoma at baseline) followed up for an average of 3.5±0.7 years.

Methods: All subjects underwent standard automated perimetry (SAP; Humphrey Field Analyzer; Carl Zeiss Meditec, Dublin, CA) and spectral-domain (SD) optical coherence tomography (OCT; Spectralis; Heidelberg Engineering, Inc., Carlsbad, CA) in both eyes at 6-month intervals. Eyes were determined as progressing by conventional methods if there was progression on masked grading of optic disc stereophotographs or SAP Guided Progression Analysis (GPA; Carl Zeiss Meditec; "likely progression"). Rates of change in SD OCT average RNFL thickness were obtained using a linear mixed effects model. Rate of global loss was calculated using a random coefficient model and compared for nonprogressing patients, progressing eyes, and fellow eyes of unilateral progressing patients.

Main outcomes measures: Rate of change in global RNFL thickness.

Results: Thirty-nine subjects showed evidence of unilateral progression by GPA, disc photographs, or both during follow-up. Mean ± standard error rate of RNFL loss in eyes progressing by conventional methods was -0.89±0.22 μm/year (P<0.001). The contralateral eyes of these subjects also showed significant loss of RNFL over time (-1.00±0.20 μm/year; P<0.001). One hundred thirty-four subjects did not show progression by conventional methods in either eye. These eyes also showed a significant decline over time in average RNFL thickness (-0.71±0.09 μm/year; P<0.001); however, the rate of change in these eyes was slower than that of the contralateral eye of patients showing unilateral progression (P<0.001).

Conclusions: Loss of RNFL thickness was seen in a substantial number of contralateral eyes of glaucoma patients showing unilateral progression by conventional methods. These findings indicate that assessment of RNFL thickness by SD OCT may show progressive glaucomatous damage that is not detected by visual fields or optic disc stereophotography.

Figure 1

Flow chart showing the number of patients enrolled and analyzed.

Figure 2

Histogram showing the distribution of global retinal nerve fiber layer thickness in progressing eyes, fellow eyes of unilateral progressing patients, and control eyes at baseline.

Figure 3

Histogram showing the distribution of rates of change in global retinal nerve fiber layer thickness during follow-up in control eyes compared to progressing eyes of progressing patient (A) and in control eyes compared to fellow eyes of progressing patients (B).

Figure 4

Example of a patient included in the study with progression by Guided Progression Analysis in the right eye (OD). The fellow eye (left eye, OS) showed no progression by conventional parameters, however, spectral domain optical coherence tomography (SDOCT) showed a significant slope of change in global retinal nerve fiber layer thickness during follow-up in the fellow eye (OS). Abbreviations: GHT: Glaucoma Hemifield Test; GPA: Guided Progression Analysis; ONL: Outside Normal Limits; WNL: Within Normal Limits; NPD: No Progression Detected.