Relationship between nailfold capillary morphology and retinal thickness and retinal vessel density in primary open-angle and angle-closure glaucoma

Purpose To compare the morphological characteristics of the nailfold capillaries in glaucoma patients with those in healthy subjects. To evaluate the relationship of nailfold capillary morphology with retinal thickness and retinal vessel density in patients with primary open-angle (POAG) and primary angle-closure glaucoma (PACG). Methods Fifty-one patients with normal-tension glaucoma, thirty-two patients with hypertension glaucoma, thirty-two PACG patients and sixty-one healthy subjects were included in this cross-sectional study. Patients with connective tissue diseases or taking anticoagulation or antiplatelet agents were excluded. Nailfold capillaroscopy, spectral-domain optical coherence tomography and optical coherence tomography angiography were used to evaluate the nailfold capillary morphology, retinal thickness and retinal vessel density, respectively. Results Nailfold capillary morphological features, including lower capillary density, greater tortuosity, more dilated capillaries >50 mu m and more avascular zones >100 mu m, were more common in POAG and PACG patients than in control subjects (all p < 0.001). In the POAG group, nailfold vessel density was negatively correlated with ganglion cell complex (GCC) thickness (odds ratio [OR] = 0.908, p = 0.010) and radial peripapillary capillary (RPC) density (OR = 0.787, p = 0.025). Tortuosity degree was negatively associated with RPC density (OR = 0.804, p = 0.016). In the PACG group, no evident relationship between nailfold capillary morphology and retinal thickness or retinal vessel density was detected (all p > 0.05). Conclusions Nailfold capillary bed abnormalities were observed in both POAG and PACG patients. The nailfold capillary features, especially microvascular density and tortuosity degree, had a significant association with GCC thickness and RPC density in subjects with POAG but not in those with PACG.