Abstract
Purpose :
To evaluate the agreement and repeatability of the new MAIA (MAIA3) microperimeter compared with MAIA 2013 EDITION (MAIA2) on healthy subjects
Methods :
Thirty healthy subjects (mean Sensitivity Range [25.50, 29.18] dB for MAIA3, [25.37, 29.38] dB for MAIA2) underwent a total of four microperimetric (MP) exams on one randomly selected eye. MP exams were performed in two sessions within one week (one exam per device in each session). MP tests were collected using a 4-2 strategy and a 10-2 grid (68 locations) plus the fovea. MP exams with Fixation Losses >30% and average pupil size during tests >2.8 mm (based on MAIA3) were considered unreliable and excluded. A total of 120 MP exams were analyzed. The agreement between the two devices and the intra-device test-retest (TRT) repeatability was evaluated both globally with Mean Sensitivity (MS) and pointwise (PWS) with Bland-Altman technique (95% Confidence Interval (CI) for Mean Difference (MD), Limits of Agreement (LoA), Limits of Repeatability (LoR)). The exam durations for both devices (average difference and distribution) were investigated, as shown in Figure2
Results :
Regarding the agreement, the MD between MAIA3 and MAIA2 was -0.36 dB. Thus, on average, MAIA3 slightly underestimated MAIA2 by a non-clinically significant amount (< 1 dB). MAIA3 95%CI LoA (MS [-2.27, +1.55] dB, PWS [-4.59, +3.87] dB) were respectively narrower than MAIA2 LoR (MS [-2.04, +1.93] dB, PWS [-4.00, +3.89] dB) by -0.16 dB for MS and wider by +0.57 dB for PWS.
Intra-device MD was close to 0dB (-0.04dB MAIA3, -0.05dB MAIA2). MAIA3 95%CI LoR (MS [-1.56, +1.48] dB, PWS [-3.84, +3.76] dB) were narrower than MAIA2 LoR (MS [-2.04, +1.93] dB, PWS [-4.00, +3.89] dB) by -0.94dB for MS and by -0.30 dB for PWS, as reported in Figure1.
The average difference in exam duration between MAIA2 and MAIA3 (494.08 and 472.97 seconds) was 21.11 seconds.
Conclusions :
MAIA3 agreed well with MAIA2 exams globally and at the point-wise level with a minimal bias (<0.5 dB). MAIA3 achieved better TRT repeatability than MAIA2, narrowing the LoR by almost 1 dB. The average exam duration was reduced by more than 20 seconds for MAIA3. These data on healthy subjects suggest that MAIA3 and MAIA2 can be used interchangeably. The external validation study in patients is ongoing (NCT06071546)
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.