Scanning Protocol Choice Affects Optical Coherence Tomography (OCT-3) Measurements

Purpose:To determine the optimal OCT-3 scanning protocol. Methods:Normal subjects, ocular hypertensive, and open-angle glaucoma patients were enrolled. The “fast” RNFL scan protocol consists of 3 consecutive circular scans, each containing 256 A-scans obtained in a single 1.92 second session. The “regular” RNFL scan protocol contains 3 separate circular scans, each of which consists of 512 A-scans and is obtained in 1.28 seconds (total 3.84 seconds). A customized 256-point scan was created from the 512-point “regular” RNFL scan by extracting the RNFL measurements from every second point. Mean RNFL thickness measurements and mean coefficients of variation (CV) were calculated for each of the scan types. Results:Thirty-one eyes with visual field defects (31 glaucoma eyes) and 30 eyes with normal fields (10 normal and 20 ocular hypertensive eyes) were enrolled. There was no difference in RNFL thickness or CV between fast, regular, or customized RNFL scans in the eyes with VF defects (P > 0.05 for all comparisons). Although eyes with normal fields showed no difference in CV between fast, regular, and customized RNFL scans (P > 0.05 for all comparisons), there was a thicker RNFL measurement using the fast RNFL scan (P = 0.01). Conclusions:Optimal sampling density for OCT-3 assessment of parapapillary RNFL thickness is not more than 256 axial determinations. Although both scanning protocols have good reproducibility, the scanning protocol should remain constant during serial OCT-3 evaluation.