Segmental reproducibility of retinal blood flow velocity measurements using retinal function imager
To evaluate the reproducibility of blood flow velocity measurements of individual retinal blood vessel segments using retinal function imager (RFI).
Eighteen eyes of 15 healthy subjects were enrolled prospectively at three centers. All subjects underwent RFI imaging in two separate sessions 15 min apart by a single experienced photographer at each center. An average of five to seven serial RFI images were obtained. All images were transferred electronically to one center, and were analyzed by a single observer. Multiple blood vessel segments (each shorter than 100 μm) were co-localized on first and second session images taken at different times of the same fundus using built-in software. Velocities of corresponding segments were determined, and then the inter-session reproducibility of flow velocity was assessed by the concordance correlation co-efficient (CCC), coefficient of reproducibility (CR), and coefficient of variance (CV).
Inter-session CCC for flow velocity was 0.97 (95% confidence interval (CI), 0.966 to 0.9797). The CR was 1.49 mm/sec (95% CI, 1.39 to 1.59 mm/sec), and CV was 10.9%. The average arterial blood flow velocity was 3.16 mm/sec, and average venous blood flow velocity was 3.15 mm/sec. The CR for arterial and venous blood flow velocity was 1.61 mm/sec and 1.27 mm/sec respectively.
RFI provides reproducible measurements for retinal blood flow velocity for individual blood vessel segments, with 10.9% variability.
KeywordsRetinal blood flow Retinal function imager Retinal blood flow velocity
This study was supported by an Unrestricted Research Fund to Jacobs Retina Center at Shiley Eye Center, University of California, San Diego (LC), NIH-EY 007366 (WRF), NIH EY 020617 (LC), NIH- EY 018589 (WRF), RPB Inc. New York (WRF), NIH-EY 016323 (DUB) and Retina Research Foundation of Wills Eye Institute (SJG).