Determinants of Test Variability in Scotopic Microperimetry: Effects of Dark Adaptation and Test Indices

To test the effect of different dark adaptation conditions and reliability indices on the variability of two color scotopic microperimetry.

We analyzed data from 22 consecutive visually healthy adults. Scotopic microperimetry was performed (Macular Integrity Assessment microperimeter, CenterVue, Padua, Italy) with two wavelength stimuli, cyan (505 nm) and red (627 nm), after a dark adaptation time of 10, 20, or 30 minutes. All tests were repeated twice to measure test-retest variability with Bland-Altman plots. We also provide a method to more accurately quantify the false-positive (FP) responses based on response data (button pressing) from the device, similar to FP responses used in standard static perimetry. Data on fixation stability (95% bivariate contour ellipse area) and blind spot responses were also extracted. Their relationship with measured sensitivity (in decibels) and test-retest variability was quantified through linear mixed effect models.

Dark adaptation had a significant effect on the sensitivity (dB) measured with the cyan stimulus (P < 0.001), but no effect on the red stimulus. Of the three metrics, the novel FP responses showed the best association with test-retest variability and was the only predictor consistently significant for all tests (P < 0.01).

Dark adaptation protocols should be carefully standardized for scotopic testing, especially if a cyan stimulus is used. The proposed FP responses should be used to assess reliability of microperimetry examinations instead of other metrics.

We developed a method to calculate a more accurate estimate of the FP responses using data available to all researchers, generalizable to all Macular Integrity Assessment microperimeter tests.