Purpose: To evaluate the reliability of intensity of light scatter (ILS) for diagnosing mild-to-moderate uveitis, which is difficult to score by the subjective standardized uveitis nomenclature (SUN) grading. Methods: ILS (in millivolts) was measured in healthy subjects and uveitis patients by a custom-made Spot fluorometer. The reliability was assessed by noise analysis of the instrument using turbid samples and intra- and interobserver variabilities. Results: ILS increased with the severity of inflammation: healthy (0.1498 ± 0.0564) and SUN0 (0.1590 ± 0.0564) < SUN1+ (0.2050 ± 0.0887) < SUN2+ (0.4283 ± 0.1895). The increase in ILS was more significant between healthy and SUN1+ (P < 0.0001) compared to the difference between SUN0 and SUN1+ (P = 0.0006). The instrument noise was small (σ = 0.002). Intraobserver measurements revealed no variability (P > 0.98) between measurements and predicted repeatability coefficients of 0.0754, 0.0522, and 0.0532 in healthy, SUN0, and SUN1+, respectively. Interobserver measurements also showed no variability (P > 0.88) and predicted intraclass correlation coefficient > 0.63 in healthy and uveitis subjects. Conclusions: The fluorometer enables precise measurement of ILS with negligible instrument noise, and the measurements are unaffected by intra- and interobserver variabilities. However, the difference in mean ILS between SUN0 and SUN1+ is greater than the instrument noise and comparable to intra- and interobserver variabilities. Thus, ILS measurements are likely impacted by blinking, eye movements, and focusing jitter. The impact of these factors must be reduced before using the fluorometer to distinguish mild-to-moderate uveitis.
Keywords: Bland–Altman analysis; Spot fluorometer; aqueous flare; intraclass correlation coefficient; repeatability; uveitis.