Purpose: In vitro models suggest that Heidelberg retina flowmeter (HRF) measurements are affected by changes in photodetector sensitivity. We measured blood flow in a single volume of human retinal tissue in vivo at various sensitivity (DC) levels.
Methods: The peripapillary retinal regions of 12 normal subjects were examined by HRF under five different sensitivity settings: (1) average DC range below 100; (2) average DC range below 125; (3) average DC range near 150 (normal sensitivity); (4) average DC range above 175; and (5) average DC range above 200 or extremely overexposed. The distributions of flow values were examined by pointwise analysis. All pixels from a common tissue location were analyzed, and the effect of their brightness on the flow measurement was evaluated by ANOVA with Fisher's protected least significant difference model.
Results: ANOVA analysis of image DC level showed that significantly different DC levels were achieved for each of the five sensitivity settings (P < 0.0001). Flow values decreased with increasing DC for each of the 25th percentile, 50th percentile (P: < 0.0001 for each), 75th percentile (P: = 0.0026), 90th percentile (P: = 0.0216), and mean (P: = 0.0004) flow values. The percentage of pixels with values of zero (avascular tissue) increased with increasing photodetector sensitivity (P< 0.0001).
Conclusions: Improper sensitivity settings alter the detected percentage of avascular tissue and the blood flow measurements in tissue containing capillaries. Consistent assessment of retinal blood flow requires consistent photodetector sensitivity settings between longitudinal images.