Device-based day-to-day and observer variability to quantify dilation capacity in the retinal microcirculation

Area of Science:

• Ophthalmology and visual sciences
• Medical technology and devices
• Cardiovascular research

Background:

• Dynamic retinal analysis (DVA) is a validated method for quantifying microvascular endothelial function.
• Assessing the reliability of DVA is crucial for its clinical implementation.
• Understanding variability is key to standardizing DVA protocols.

Purpose of the Study:

• To evaluate day-to-day, intra- and interobserver variability of DVA.
• To compare DVA measurements between two device generations (DVA 2.0 and DVA 3.0).
• To identify factors influencing DVA reliability.

Main Methods:

• DVA measurements were taken on two separate days using DVA 2.0 and DVA 3.0 devices.
• Intraobserver variability was assessed by one reader analyzing 20 arteriolar (aFID) and venular flicker-light induced dilation (vFID) signals twice.
• Interobserver variability was assessed by a second reader independently analyzing 20 aFID and vFID signals. Interclass correlation coefficient (ICC) was used.

Main Results:

• Moderate to good day-to-day variability was observed for both aFID and vFID across both DVA versions.
• Excellent intraobserver (aFID/vFID: 0.999) and interobserver (aFID: 0.997, vFID: 0.998) reliability was demonstrated.
• Moderate interdevice variability was found between DVA 2.0 and DVA 3.0 (aFID: 0.76, vFID: 0.87). ICC improved after correcting for baseline diameter.

Conclusions:

• DVA exhibits reliable intra- and interobserver performance.
• Day-to-day and interdevice variability exist, suggesting a need for standardization.
• Accounting for arteriolar baseline diameter variations can enhance DVA reliability in research and clinical settings.