Abstract
Purpose :
Age-related macular degeneration (AMD) is a leading cause of blindness in the elderly. Chronic kidney disease (CKD) is also commonly associated with age. These two diseases share risk factors such as age, smoking, and hypertension as well as pathophysiologic similarities affecting the glomerular capillaries and the retinal choroid. This study investigates potential correlations between renal function and AMD features as assessed with multimodal retinal imaging.
Methods :
Participants were part of a study that investigated dark adaptation (DA) function in those ≥ 50 years of age with varying AMD severities (NCT01352975) who also had eGFR (ml/min/1.73m2) values obtained from renal function laboratory testing of serum creatinine and cystatin C. Multimodal imaging included color fundus photography and optical coherence tomography volumes from visit dates associated with serum samples and were graded by the Wisconsin Reading Center for the presence of AMD features. Associations of eGFR with AMD features and severity grades, age, smoker status, and DA function represented by rod-intercept time (RIT) were investigated. Simple univariate analyses, age-corrected multivariate analyses, and a feature-selecting LASSO regression were performed for eGFR as a continuous dependent variable.
Results :
A total of 110 patients (mean age ± SD 75.1 ± 9.4 years [range 53–95]; mean eGFR ± SD 70.7 ± 18.2 ml/min/1.73m2 [range 19.3–109.8]) were included. 95 patients had eGFRs < 90 and 36 patients had eGFRs < 60. In univariate analyses age (estimate -1.16 units/year, [CI -1.46 – -0.87], p<0.0001), RIT (estimate -0.54 units/minute, [CI -0.81 – -0.27], p<0.001) and reticular pseudodrusen (RPD) (-11.12 units for RPD presence in either eye, [CI -20.23 – -2.01], p=0.017) were associated with decreased renal function. However, in age-corrected multivariate models, age was the only significant variable associated with renal function, confirmed by LASSO regression.
Conclusions :
Age was found to be strongly correlated with decreased renal function in both univariate and age-corrected multivariate analyses. While one of the strengths of this study is the well-phenotyped AMD cohort with serum values estimating renal function, it was a relatively small study. Further studies investigating the relationship between CKD and AMD phenotypes will need larger populations to better understand possible correlations between these diseases.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.