Eight patients with newly diagnosed, untreated NPG and eight healthy volunteers, matched for age, sex, mean arterial blood pressure, weight, and forearm length, were recruited. NPG was diagnosed on the basis of the accepted criteria: mean intraocular pressure <22 mm Hg on diurnal phasing, gonioscopically open anterior chamber angle, characteristic optic disc cupping (cup-disc ratio >0.6) with either thinning or notching of the neuroretinal rim, and glaucomatous visual field loss detected by automated perimetry (Humphrey Instruments; Carl Zeiss Meditec, Dublin, CA) with the 24-2 threshold program (average mean deviation = −5.15 dB and corrected pattern standard deviation = 9.21 dB). Where indicated, neuroimaging was used to exclude an intracranial cause of disc or field changes. None of the patients had other ocular disease, or had previously received steroid therapy. All control subjects had normal findings in a normal ocular examination, normal intraocular pressure, and normal visual fields. None of the subjects was taking any vasoactive medication. Approval was granted by the Lothian Research Ethics Committee and the tenets of the Declaration of Helsinki were upheld. Written, informed consent was obtained from all subjects before enrollment in the study.