To validate cut-off values of quantitative high-resolution temporal artery compression sonography (TCS) for the diagnosis of cranial GCA (cGCA) in patients with acute arterial ocular occlusions and in an independent control group.Consecutive patients who underwent TCS as part of the diagnostic workup of acute arterial ocular occlusions and controls not suffering from ocular ischaemia/systemic vasculitis were included. The diagnostic accuracy of the established TCS cut-off value of maximum temporal artery wall thickness (≥0.7 mm) and a novel numeric TCS score incorporating the degree of wall thickening in the four temporal artery segments assessed (0-3 points per segment) was tested by receiver operating characteristics analysis. Subgroup analyses were performed for female and male patients and patients older and younger than age of 70 years.Of 114 patients with acute ocular arterial occlusions, 30 patients received a final clinical diagnosis of cGCA. The sensitivity and specificity of the ≥0.7 mm TCS cut-off for the diagnosis of cGCA were 100 and 84.5% in the overall cohort. The TCS score did not improve the diagnostic yield (cut-off ≥5; sensitivity 100%, specificity 85.7%). In male patients >70 years of age, the specificity of TCS was limited, secondary to age- and sex-related differences in temporal artery wall thickness, which we confirmed in the independent control group.TCS yields high diagnostic accuracy in the diagnosis of cGCA in patients with acute ocular arterial occlusions. Age- and sex-related differences in temporal artery wall thickness influence the diagnostic accuracy of TCS.