RESUMEN
PRCIS: Microvasculature dropout in the parapapillary choroidal layer was a more important biomarker of glaucomatous nerve fiber layer thinning when it presented with deep-layer microvasculature of the optic disc rather than when it presents by itself. PURPOSE: To characterize open angle glaucoma eyes with optic nerve head deep-layer microvasculature dropout (MvD-D) and parapapillary choroidal layer microvasculature dropout (MvD-P) and compare their retinal nerve fiber layer (RNFL) thinning rate. MATERIAL AND METHODS: This study included 122 open angle glaucoma eyes that underwent ≥5 serial spectral-domain optical coherence tomography scans during a mean follow-up of 5.4 years. Swept-source optical coherence tomography angiography was used to evaluate MvD-P and MvD-D. Subjects were classified into 3 groups according to the presence of MvD-P and MvD-D: (1) no dropout (n=37); (2) solely MvD-P (n=40), and (3) both MvD-P and MvD-D (n=45). The RNFL thinning rate was compared among the 3 groups, and the associated factors were assessed by Cox proportional hazard analysis. RESULTS: RNFL thinning rates were highest in the group with both MvD-P and MvD-D, followed by the group with solely MvD-P and finally by the no dropout group (-0.24 vs. -0.65 vs. -1.20 µm/y, P <0.001). Thinner central corneal thickness [hazard ratio (HR)0.990, P =0.003], presence of disc hemorrhage (HR=1.802, P =0.035), and coexistence of MvD-P and MvD-D (HR=2.941, P <0.001) were the factors associated with RNFL thinning. CONCLUSIONS: The coexistence of MvD-P and MvD-D was associated with faster RNFL thinning than MvD-P alone or no dropout, which suggested that observing the optic disc deep microvasculature along with parapapillary choroidal layer using Swept-source optical coherence tomography angiography may be clinically relevant in monitoring glaucoma progression.
