Purpose The purpose of the study was to investigate nailfold microvascular morphology in exfoliation syndrome with or without glaucoma (XFS/XFG) compared with primary open-angle glaucoma (POAG) and control subjects using nailfold capillary videomicroscopy. Patients and methods We used a JH-1004 capillaroscope to perform nailfold capillary videomicroscopy on the fourth and fifth digit of the non-dominant hand. We enrolled 56 XFS/XFG patients, 87 POAG patients, and 75 control subjects. Masked observers graded the videos for hemorrhages, avascular zones ≥ 200 microns (μm), and degree of microvascular tortuosity on a four-point subjective scale. Multivariable odds ratios, 95% confidence intervals and P-for trends for assessing the relation between morphological changes and POAG or XFS/XFG were obtained from logistic regression analyses. We also assessed this relation with XFS/XFG compared with POAG in multivariable models. Results After adjusting for multiple covariates, nailfold hemorrhages, avascular zones ≥ 200 μm, and higher degree of vascular tortuosity were more common in XFS/XFG vs controls (P-for trend ≤ 0.0001) and in POAG vs controls (P-for trend ≤ 0.01). For each 100 capillaries, the number of hemorrhages was similar (P-for trend = 0.91) between XFS/XFG and POAG patients; however, there were more avascular zones per 100 capillaries with borderline significance (P-for trend = 0.04) in the XFS/XFG group. XFS/XFG patients had more tortuosity than POAG patients; specifically, having a tortuosity score ≥ 1.5 was associated with a 4.4-fold increased odds of XFS/XFG (95% confidence interval: 1.5-13.3) relative to a tortuosity score o1.0 (P-for trend = 0.005). Conclusion A high degree of nailfold capillary tortuosity is a distinct non-ocular feature associated with XFS/XFG compared with either POAG or controls.