PURPOSE: To assess the agreement among three masked examiners on central retinal artery (CRA) and ophthalmic artery (OA) blood flow velocity measurements performed with colour Doppler imaging (CDI) in healthy volunteers.METHODS: The study included 30 eyes of 15 healthy volunteers. Prior to the study, all examiners underwent intensive CDI training by an expert to facilitate uniformity in performing measurements according to a specific protocol. Following the eligibility visit, three masked examiners performed CDI measurements assessing the CRA and OA in both eyes of all subjects. All CDI images were analysed by a masked grader. Intraclass correlation coefficients (ICC) were calculated for peak systolic velocity (PSV) and end diastolic velocity (EDV) in the CRA and OA among the examiners. p-values < 0.05 were considered statistically significant.RESULTS: The study included seven men and eight women (mean age 30.9 ± 4.5 years). In right eyes, ICCs for PSV and EDV in the CRA ranged from 0.917 to 0.961 and from 0.937 to 0.980, respectively. ICCs for PSV and EDV in the OA ranged from 0.802 to 0.926 and from 0.611 to 0.891, respectively. In left eyes, all ICCs for PSV and EDV were >0.9, both in the CRA and the OA. All p-values calculated for ICCs were statistically significant.CONCLUSION: Expert training and execution of a specific protocol for CDI of ocular blood flow velocity measurements provide highly reproducible results in healthy volunteers. This is important for long-term studies assessing ocular hemodynamics, where multiple examiners may be involved.