Abstract
PRÉCIS: Compared with trend-based analysis, event-based analysis detects OCT structural progression in more patients and at an earlier time point. Using event-based analysis, MDB rim thickness detects progression more often than RNFL thickness. PURPOSE: To determine whether event-based or trend-based analysis best detects glaucoma progression using spectral domain optical coherence tomography (OCT) retinal nerve fiber layer (RNFL) thickness and minimum distance band (MDB) neuroretinal rim measurements. METHODS: Over 5 years, 124 open angle glaucoma patients had yearly dilated eye examinations, disc photography, Humphrey visual field (HVF 24-2) testing, 2-dimensional (2D) OCT RNFL thickness measurements, and 3-dimensional (3D) OCT MDB rim measurements, all on the same day. One eye of each patient was analyzed, which was randomly selected if both eyes from a patient were eligible. Using global RNFL thickness and global MDB rim thickness, event-based progression was defined as change greater than normal aging change and expected inter-test variability. Trend-based analysis used linear regression with progression defined as rate of decline greater than age-related decline. RESULTS: The average follow-up for the 124 open angle glaucoma patients was 66.9±16.4 months. Event-based analysis was better than trend-based analysis because it detected progression in more patients (15.3% by RNFL event-based analysis versus 8.1% by RNFL trend-based analysis, P =0.025; 52.4% by MDB event-based analysis versus 9.7% by MDB trend-based analysis, P <0.001) and earlier (RNFL: mean 28.8 vs. 63.2 mo; P <0.001; MDB: mean 30.7 vs. 56.2 mo; P <0.003) whether using MDB rim thickness or RNFL thickness. Using event-based analysis, MDB rim thickness detected progression more often than RNFL thickness (52.4% vs. 15.3%; P <0.001). CONCLUSIONS: Compared with trend-based analysis, event-based analysis detected OCT structural progression in more glaucoma patients and at an earlier time point.