Author: Jan Henrik Terheyden (Germany)
Co-authors: Hannah Dunbar, Charlotte Behning, David P. Crabb, Stephen Poor, Ulrich F.O. Luhmann, Robert P. Finger
Purpose
Individuals with intermediate age-related macular degeneration (iAMD) have visual function (VF) impairments which are most prominent in low-luminance and low-contrast surroundings. The Vision Impairment in Low Luminance (VILL) questionnaire has been specifically developed to capture the impact of these functional deficits on vision-related quality of life. The relationship between measureable impairment in functional tests and patient-reported outcome (PRO) measures has been sparsely studied but represents an important prerequisite to understand how the visual deficit impacts people with iAMD in their daily lives. We investigated associations between VILL-based self-report and visual function in the MACUSTAR study.
Setting/Venue
Multi-centre cohort study in 20 ophthalmologic referral centres from seven European countries.
Methods
All participants of the iAMD and no AMD groups within the cross-sectional part of the MACUSTAR study were included. PRO measures included the reading and mobility subscales of the VILL questionnaire (Rasch person measures). Function assessments included best-corrected visual acuity (BCVA), low-luminance visual acuity (LLVA), Moorfields acuity chart test (MA), Pelli-Robson contrast sensitivity (CS), average threshold of mean sensitivity in mesopic and scotopic fundus-controlled perimetry (mAT and sAT; Macular Integrity Assessment), rod intercept time in dark adaptometry (AdaptDx) and International Reading Speed Texts reading speed (IReST). AMD was categorized according to the Beckman classification. Participants who reported difficulties on the subscales of the VILL and poor performers of VF assessments were defined based on test performance within the no AMD group. The threshold was set per individual assessment, at a test result worse than the no AMD group’s mean ± 1.64 standard deviations (SD) for normally distributed parameters or the 5th percentile for non-normally distributed parameters. Overlaps between individuals with poor VILL scores and poor performers in function tests were compared. In iAMD participants, linear regression controlled for age, sex and number of comorbidities, were used to assess the association of VF and self-report (i.e. VILL scores).
Results
A total of 168 participants with iAMD (63% female, mean age 71±8 years) and 56 people with no AMD (comparator group; 59% female, mean age 68±6 years) were included in the analysis. Thirty-four individuals with iAMD (20%) reported impairment outside the defined limits in at least one VILL subscale and 16 (10%) reported impairment in both subscales. 116 participants with iAMD (69%) were poor performers in at least one function test, 67 (40%) were poor performers in at least two function tests. 31 (86%) people with low VILL scores also performed poorly in at least 1 functional test, 18 (53%) people with low VILL scores also performed poorly in at least 2 functional tests. The functional tests most associated with low VILL person measures were LLVA (overlap with low VILL person measures in 15 individuals, 44%) sAT (overlap with low VILL person measures in 14 individuals, 41%), CS and MA (overlap in 13 individuals, 38%, respectively). In multivariable regression models across all participants with iAMD, the VF assessments BCVA, CS, sAT and IReST were significantly associated with the VILL reading subscale (p≤0.043). BCVA, mAT and sAT were significantly associated with the VILL mobility subscale (p≤0.014) in all iAMD participants.
Conlusions
Self-reported visual functioning and vision-related quality of life under challenging light conditions including low luminance and low contrast situations corresponds to VF measures in iAMD in the MACUSTAR study. The function measures LLVA, CS, MA, sAT, mAT, BCVA and IReST were associated most strongly with the reading and mobility subscales of the VILL. However, no single VF assessment alone could explain this relationship between PRO measures and overall VF. Thus, considering more than one functional measure seems necessary to capture the patient-relevant VF deficit in iAMD. The reliability of the chosen thresholds and the association between functional assessments and structural biomarkers in the context of a structurally-defined disease classification (Beckman classification) remain to be investigated.
Financial Disclosure
JHT: Heidelberg Engineering, Optos, Carl Zeiss Meditec, CenterVue (devices); HD: Boehringer Ingelheim; CB: None; SP: Novartis (Employee); DPC: Allergan, Bayer, Santen, Thea (Honoraria), Allergan, Apellis, Roche, Santen (Unrestricted funding), Centervue (Consultant); UFOL: Roche (Employee); RPF: Bayer, Novartis, Santen, Opthea, Novelion, Santhera, Inositec, Alimera, RetinaImplant, Allergan, Boehringer Ingelheim (Consultant); Bayer, Ellex, Alimera (Funding); Heidelberg Engineering, Optos, Carl Zeiss Meditec, CenterVue (Devices)
Comments
-
