Comparison of pupil size measurement: Repeatability and agreement across four devices with different measurement principles under varied lighting conditions

Abstract Aim To evaluate the repeatability of pupil size measurements obtained with four different devices under varying lighting conditions and to assess the agreement between them. Methods This prospective study included 80 healthy participants with a mean age of 27.5 ± 8.8 years (range 19–58). Pupil size was measured under mesopic (4 lux) and low photopic (50 lux) conditions using two open‐field devices (tabletop and handheld PowerRefractors) and two closed‐field devices (MS‐39 and NIDEK Tonoref III). Three consecutive measurements were taken per eye for each device. Repeatability was analysed using the repeatability limit (R Limit), while the agreement was assessed with Bland–Altman analysis. Results All devices demonstrated R Limits lower than 1 mm under both low photopic and mesopic conditions. The NIDEK Tonoref III had the best repeatability under low photopic conditions (R Limit = 0.52 mm), while the tabletop PowerRefractor had the best R Limit under mesopic conditions (0.68 mm). The agreement analysis under mesopic conditions revealed smaller mean differences and narrower limits of agreement (LoA) (approximately 2 mm) among devices with similar designs (tabletop with handheld PowerRefractor and MS‐39 with NIDEK Tonoref III), while combinations of designs exhibited wider variability. Mesopic conditions yielded more consistent LoA across all device pairs. Conclusions All devices showed excellent repeatability. Instruments with similar measurement designs demonstrated better agreement, particularly under low photopic conditions. Significant variability existed when comparing open‐ and closed‐field devices. These findings underscore the importance of selecting appropriate instruments for clinical and research applications, and highlight the need for further validation of pupillometry devices.