Comparison between the PlusOptix PowerRef 3 and the Grand Seiko WAM-5500 for Measuring Accommodation

Abstract

Purpose : We aimed to compare repeatability, agreement and reliability of two commercial instruments mainly used in the literature for measuring the accommodation. We tested the PlusOptix PowerRef 3 (PR3) based on the eccentric photorefraction and the Grand Seiko WAM-5500 (WAM) using the Scheiner’s disc principle operating at 50 and 5 Hz respectively.

Methods : In a 1^st procedure, twenty young adults (20-25 years) viewed a Maltese cross on a screen located at 5m (accommodative demand of 0.20 D) in binocular vision. Accommodation was measured three times for sixty seconds with the PR3 and the WAM in a random order. In a 2^nd procedure, a calibration was conducted on five participants in a separated visit. The left eye viewed a Maltese cross at 5 m while the right eye was occluded with an IR filter and trial lenses ranging from -4 to +4 D in 1 D steps were placed on the same eye. Three measures of ten seconds on the right eye were recorded for each lens with both instruments in a random order. Repeatability, agreement and reliability were assessed using the mean standard deviation within three measures (MSD), the Bland-Altman plot with 95% limits of agreement (LoA) and the intraclass correlation coefficient (ICC).

Results : In the 1^st procedure, a mean bias of 0.64 D showed more hyperopic measurements with the PR3 and LoA ranging from -0.43 to 1.72 D were found. Repeatability of both instruments were significantly different (p = 0.004) with a MSD (± inter-individual SD) of 0.25 ± 0.19 D and 0.09 ± 0.10 D for the PR3 and the WAM respectively. Agreement between instruments was poor (ICC: 0.29). In the 2^nd procedure, the WAM was perfectly able to measure the power of each induced lens (y=x+0.36, R² = 1.00) despite a slight overestimation of refraction while the PR3 always underestimated the power of the induced lenses (y=0.88x-0.14, R² = 0.99).

Conclusions : The PR3 is less repeatable and gives more hyperopic measurement than the WAM as previously demonstrated in the literature. According to the calibration procedure, the PR3 is also less reliable. Difference between instruments could be explained by the different operating principles or sampling frequencies.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.

 

Bland-Altman analysis on accommodative error with the WAM and the PR3 for accommodative demand of 0.20 D.

 

Measured refractive errors as a function of the induced lenses. Error bars shows the mean standard deviation within three measures for each lens.