Background: To compare retinal sensitivity measurements obtained on two microperimeters: the CenterVue MAIA and the Nidek MP-1.
Design: A prospective study conducted in a private ophthalmology clinic.
Participants: 17 participants with a range of stable macular function.
Methods: Microperimetry in one eye with identical test strategy in both devices, with randomised testing order.
Main Outcome Measures: Differences in mean sensitivity (MS), point-wise sensitivity (PWS) and duration. Limits of agreement (LoA) in MS and pooled PWS were calculated. Concordance in scotoma assessment was analysed by kappa (κ) agreement of local defect classification (LDC), LoA in number of scotomatous loci detected and congruence in scotoma localisation (CSL).
Results: Median (IQR) MS of the MP-1 and MAIA was 13.3 (5.6 – 18.1) and 21.2 (14.5 – 27.0) dB, (p < 0.05). Mean difference in PWS was 7.3 dB (MAIA > MP-1). Median (IQR) duration for the MP-1 and MAIA was 10’28” (7’17” – 11’53”) and 8’46” (8’30” – 9’06”), (p = 0.21). LoA for MS and pooled PWS was 1.4 to 13.3 dB and -3.9 to 18.5 dB. There was moderate agreement between the devices for LDC (weighted κ = 0.55, p < 0.05). LoA in number of scotomatous loci detected was -13 to 18. CSL varied from 0 – 100% and correlated strongly with increasing scotoma size.
Conclusions: The large LoA and variation in scotoma mapping concordance suggests that the same microperimeter should be used for follow-up examination. We recommend caution in comparing results derived from different types of microperimeters.