Evaluation of agreement and repeatability between the Colvard, OPD-Scan Ⅲ and binocular dynamic pupillometer in pupil diameter measurement

Objective  To compare the agreement in scotopic pupil diameter measurements among the Colvard, OPD-Scan Ⅲ and a binocular dynamic pupilometer, and to evaluate the repeatability of the binocular dynamic pupilometer for pupil diameter measurements under different illumination intensities.

Methods  A cross-sectional study was conducted, enrolling 134 patients (134 right eyes) scheduled for refractive surgery at the Refractive Surgery Center of Tianjin Eye Hospital from October to November 2024. The same physician measured the pupil diameters of both eyes in the fixed order of Colvard, OPD-Scan Ⅲ, and the binocular dynamic pupillometer, and only the right eye data were included for analysis. Measurements with the Colvard and OPD-Scan Ⅲ were performed under an ambient illumination of 0.15-0.20 lx, while the binocular dynamic pupillometer was used under dark (0.04 lx), dusk (4 lx), and bright (40 lx) conditions. Differences in scotopic pupil diameter measurements among the three devices were compared; the correlations of the measurements were analyzed; the Bland-Altman method was used to assess the inter-device agreement; and the repeatability of the binocular dynamic pupillometer for pupil diameter measurement under different illumination intensities was evaluated. This study adhered to the Declaration of Helsinki and was approved by the Medical Ethics Committee of Tianjin Eye Hospital (KY-2025050). Informed consent was obtained from all participants.

Results  The scotopic pupil diameters measured by the Colvard, OPD-Scan Ⅲ, and binocular dynamic pupillometer were (6.97±0.68), (6.69±0.75), and (7.13±0.65)mm, respectively, with a statistically significant overall difference ( F=111.46, P<0.001).The scotopic pupil diameters measured by the binocular dynamic pupillometer were higher than those by Colvard and OPD-Scan Ⅲ, and the scotopic pupil diameters measured by Colvard were higher than those by OPD-Scan Ⅲ, slowing statistically significant differences (all P<0.001). Significant positive correlations were observed in the scotopic pupil diameters measured between the Colvard and OPD-Scan Ⅲ, Colvard and the binocular dynamic pupillometer, and the binocular dynamic pupillometer and OPD-Scan Ⅲ ( r=0.880, 0.910, 0.855; all P<0.001). Bland-Altman analysis demonstrated wide 95% limits of agreement (LoA) for scotopic pupil diameter measurements between the Colvard and OPD-Scan Ⅲ, the binocular dynamic pupillometer and Colvard, and the binocular dynamic pupillometer and OPD-Scan Ⅲ, which were -0.42 to 0.97 mm, -0.39 to 0.72 mm, and -0.32 to 1.20 mm, respectively. For the binocular dynamic pupillometer, the within-subject standard deviations at 0.04, 4, and 40 lx were 0.11, 0.17, and 0.16 mm, with the corresponding intraclass correlation coefficients of 0.97, 0.89, and 0.89, respectively.

Conclusions  While the Colvard, OPD-Scan Ⅲ, and the binocular dynamic pupillometer show high correlation in measuring scotopic pupil diameter, their agreement is suboptimal. The 95%LoA for all device comparisons exceeded the clinically acceptable margin of ±0.5 mm, indicating that their measurements are not clinically interchangeable. The binocular dynamic pupillometer demonstrated high intra-device repeatability.