Abstract
Abstract: :
Purpose: To determine the validity and test–retest reliability of the Lighthouse Letter Contrast Sensitivity Test (LHT) in healthy subjects, patients with chronic open–angle glaucoma (COAG), and patients with age–related macular degeneration (ARMD). Methods: The sample contained 47 healthy subjects (mean age 48, range 22 to 77 years; mean visual acuity [VA] –0.01, range –0.16 to +0.24 logMAR), 27 COAG patients (mean VA +0.04, range –0.16 to +0.28 logMAR; mean MD –6.4, range –31.1 to +1.4 dB), and 12 ARMD patients (mean VA +0.87, range +0.22 to +1.56 logMAR). For the healthy group, one eye was randomly selected. For the COAG and ARMD groups, the eye with worse MD or VA, respectively, was selected as the study eye. The study eye was tested twice with the Pelli–Robson Contrast Sensitivity Test (PRT) and LHT, in random order. The median time between tests was 7 days. Results: Mean contrast sensitivity (CS) with the LHT was 1.62 log units (5th and 95th %tile, 1.56 and 1.72 log units) and 1.59 log units (5th and 95th %tile, 1.44 and 1.72 log units), for healthy subjects aged <55 years and >55 years, respectively. LHT 95% test–retest limits of agreement (LOA) were ±0.13 log units (3 letters), ±0.19 log units (5 letters) and ±0.26 log units (6–7 letters) for the healthy, COAG and ARMD groups, respectively. In comparison, the PRT 95% test–retest LOA were ±0.18, ±0.19 and ±0.26 log units for the healthy, COAG and ARMD groups, respectively. Mean test–retest differences indicated a small learning effect with both the LHT and PRT. With each of the tests, group mean learning effects were <0.02 log units, except for the ARMD group (LHT mean test–retest difference 0.11 log units [3 letters], 95% CI 0.02 to 0.19 log units, P=0.02). A strong correlation was found between CS with the LHT and PRT (r=0.95, P<0.001). However, CS with the LHT was significantly less than with the PRT (mean difference 0.12 log units, 95% CI 0.09 to 0.14 log units, P<0.001), and differences were greater at the upper compared with the lower end of the CS scale (P<0.001). Conclusions: The findings suggest that the LHT is a valid test of CS. Also, the 95% test–retest LOAs indicate that the reliability of the LHT is comparable to the PRT for healthy subjects, patients with COAG and patients with ARMD. However, there are systematic differences between the LHT and PRT, and normative values are different for each. Clinical advantages of the LHT CS include its compact size, portability and simplicity of administration.
Keywords: contrast sensitivity • low vision • clinical research methodology