Intrasession repeatability of 6×6 mm swept-source OCT angiography macular vessel density measurements in high myopia: a prospective methodological study

Abstract Background Quantitative OCT angiography (OCTA) requires well-defined measurement error for longitudinal interpretation, particularly in high myopia. We evaluated intrasession repeatability of automated 6×6 mm swept-source OCTA macular vessel-density measurements in highly myopic eyes. Methods Repeated lens-off macular scans were obtained with the DRI OCT Triton 3D Plus using OCTARA. Thirteen participants contributed 23 anatomical eyes and 32 paired comparisons. Superficial and deep capillary plexus (SCP/DCP) vessel density was analyzed centrally and in four parafoveal sectors. Repeatability was assessed with two-way random-effects, single-measure, absolute-agreement ICC [ICC(A,1)], within-subject SD (Sw), coefficient of variation (CV), repeatability coefficient (RC), Bland-Altman limits of agreement, participant-cluster bootstrap confidence intervals, and a one-pair-per-eye sensitivity analysis. Results Central reliability was excellent (SCP ICC 0.926, 95% CI 0.643–0.957; DCP ICC 0.934, 0.832–0.961). Sectoral ICCs were lower (0.399–0.691), whereas sectoral CVs remained 2.24%–3.35% and RCs 3.08–4.89 percentage points. Absolute mean bias was ≤ 0.57 percentage points in all regions. Sensitivity analysis preserved excellent central reliability. Eight lens-on pairs were analyzed only exploratorily. Conclusions Central 6×6 mm Triton OCTA vessel-density measurements were highly repeatable in high myopia, but parafoveal ICCs were more heterogeneous despite modest absolute error. Region-specific repeatability and standardized acquisition conditions should be considered when interpreting small longitudinal changes.