Longitudinal Assessment of Low Luminance Questionnaire Scores in Early and Intermediate AMD.

PURPOSE: To characterize changes in Low Luminance Questionnaire (LLQ) scores and their correlation with visual function (VF) tests over 24 months in patients with early (eAMD) and intermediate (iAMD) AMD. METHODS: We conducted a prospective longitudinal observational study of eAMD (n = 22), iAMD (n = 31), and controls (n = 17). Subjects completed the LLQ at baseline and 24 months. VF testing included best-corrected visual acuity, Early Treatment Diabetic Retinopathy and computerized low-luminance visual acuity, microperimetry, cone contrast testing, and rod intercept time (RIT). Correlations between change in LLQ scores and VF measures were assessed using Spearman coefficients. RESULTS: Within-group differences in LLQ composite scores over 24 months were not significant in controls, eAMD, or iAMD. In an outlier analysis, subjects with a baseline LLQ composite score 2 SDs below the control baseline mean (cut off 74.2) were classified as outliers. A significant decline in composite LLQ score over 24 months was observed only in iAMD nonoutliers (median, -5.5 [minimum, maximum, -17.2, 3.1]; P = 0.008). Worsening LLQ composite and subscale scores correlated with RIT prolongation in iAMD nonoutliers. CONCLUSIONS: A significant decline in LLQ scores over 24 months was observed only in iAMD subjects with preserved baseline function. The absence of longitudinal change in the composite LLQ scores among participants with lower baseline LLQ scores suggests a floor effect. Among iAMD subjects with preserved baseline function, worsening LLQ scores correlated with RIT prolongation. This finding supports the potential utility of the LLQ as a patient-reported outcome sensitive to worsening dark adaptation. Confirmation in larger cohorts is needed.

Duke Scholars

Author Sandra Sue Stinnett Biostatistics & Bioinformatics, Division of Biostatistics