Associations of rod-mediated dark adaptation measures in older adults without AMD or with early or intermediate AMD, baseline results from the ALSTAR2 study

Abstract

Purpose : The slope of the second component of rod-mediated dark adaptation (RMDA), rod-intercept time (RIT), and, more recently, the area under the dark adaptation curve (AUC) are metrics used in characterizing RMDA. Associations among all 3 measures have yet to be examined by AMD status and severity in the same persons.

Methods : RMDA testing at 5° eccentricity was completed in the ALSTAR2 baseline visit using the AdaptDx (83% equivalent bleach). Testing terminated once RIT was calculated by the AdaptDx or stopped at 45 minutes in cases where sensitivity did not recover. The slope of the second component was obtained from a linear model fit for all data points > 2.5 log sensitivity. AUC was calculated using the trapezoidal rule for all sensitivities < 3 log units for the full test time, and with test data truncated at 20, 12, and 6 minutes. The AREDS 9-step classification was used to determine AMD status and severity. Spearman correlations were used to associate the measures.

Results : Among 481 participants (mean age 71.8 years, 60% female, 91% white), there were 239 normal, 139 early AMD, and 103 intermediate AMD eyes. Slope of the second component, RIT, and all AUC measures were significantly associated among participants. Slope of the second component was more strongly associated with RIT (r_s=0.65) than the AUC measures (r_s=0.44 or less) (Table). As AUC measures were based on increasingly shorter test times, correlations with RIT and slope decreased. Compared to early and intermediate AMD, RIT and AUC associations were lower among normal eyes. The slope and AUC measures were not associated among normal eyes. The strongest associations of RIT with slope and the full AUC measure were among those with intermediate AMD (r_s=0.75 and 0.94 respectively).

Conclusions : All RMDA measures are associated with each other when examined among all participants. As AUC decreases at shorter test times, it becomes less correlated with RIT. This suggests that reductions in test time, specifically < 12 minutes, for faster screening do not produce measures that are good substitutes for RIT. The lack of association between the slope and AUC among normal eyes suggests that AUC is insufficient in conveying RMDA results for older adults without AMD. Slope and AUC based on 12 minutes or more are acceptable surrogates of RIT among those with early and intermediate AMD.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.