Cause-Specific Mortality and the Relationship Between Visual Acuity and Mortality in University of Colorado AMD Registry Patients

Abstract

Purpose : To: 1) compare the distribution of cause-specific mortality in a Colorado age-related macular degeneration (AMD) cohort to statewide mortality data and 2) evaluate the relationship between visual acuity and mortality in the AMD cohort.

Methods : Patients 55 and older with AMD were recruited into the University of Colorado AMD Registry (July 2014 - December 2021). AMD phenotype was determined using multi-modal imaging. Mortality status and cause of death ICD-10 codes were obtained through a collaboration with the Colorado Department of Public Health Vital Statistics Program and a review of Electronic Health Records with follow-up through October 19, 2022. Snellen habitual visual acuity (VA) was measured at the time of registry enrollment. Death rates for categories are presented as basic frequencies and proportions. Cox proportional hazards modeling was used to assess time to event for mortality. Hazards ratios, 95% confidence intervals, and p-values are presented for multivariable models.

Results : 180 (20.6%) of 876 patients died during the follow-up period with an average follow-up time of 52.5 (26.6) months. The two most common causes of death in our AMD cohort were cardiovascular diseases (excluding cerebrovascular disease) (23.9%) and malignant neoplasms (20.0%) which aligned with the distribution for persons 55 and older in the state of Colorado (Table 1). Notably the proportion of deaths from COVID-19 was similar between our study cohort (4.5%) and statewide data (3.6%). VA at time of study enrollment was associated with mortality for both the better- and worse-seeing eyes (both p<0.0001) in univariate analysis. However, in multivariable analysis after adjustment for confounding variables (Table 2), VA was not significantly associated with mortality. The covariates significantly related to mortality were age, smoking history, kidney disease, and cardiovascular disease.

Conclusions : The causes of death in our cohort of AMD patients were remarkably similar to statewide mortality data. Although visual acuity has been shown previously to be a significant risk factor for mortality, we found that VA was not significantly associated with mortality utilizing multivariable modeling in our AMD cohort.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.