MAIA microperimeter for short-duration fixation stability measurements in central vision loss: Repeatability and comparison with the Nidek MP1

Purpose This study reports the repeatability of 20 s-duration fixation stability measurements recorded with the Macular Integrity Assessment (MAIA) microperimeter in patients with central vision loss, in contrast to the Nidek MP1 microperimeter. Methods Fixation stability was recorded in 39 eyes of 25 patients with macular disease using MAIA and the MP1 for 20 s intervals, twice for each eye, with each instrument. Twenty eyes were identified as the better eye (BE) and 19 eyes as the worse eye (WE). Fixation stability was quantified with the 95% bivariate contour ellipse area (BCEA), logarithmically transformed. Bland-Altman plots were used to determine the 95% limits of agreement. Results For MAIA, the 95% limits of agreement were +/- 0.84 log deg(2) for the BE and +/- 0.66 log deg(2) for the WE. Similarly, for the MP1 these limits were +/- 0.48 log deg(2) for the BE and +/- 0.72 log deg(2) for the WE. Inter-device repeatability was modest, +/- 1.09 log deg(2) for the BE and +/- 1.01 log deg(2) for the WE, and a proportional bias was detected. Occasionally, MAIA did not register all the expected number of data points, and included far outliers in the BCEA calculation; the inter-device repeatability did not improve when these outliers were removed. Conclusions Repeatability of 20 s-duration fixation stability examination in patients with central vision loss is specific to the instrument used. We recommend that only data from same type of microperimeter with the same fixation duration should be compared when using fixation stability as an outcome measure to monitor disease progression, effect of treatment or in clinical trials.