Osteoporosis was associated with increased odds of dry age-related macular degeneration and with overall age-related macular degeneration in women, according to a meta-analysis and bidirectional Mendelian randomization study published in Experimental Gerontology.

In pooled observational data from three studies including 58,057 patients, osteoporosis was associated with 1.5 times the odds of overall age-related macular degeneration (AMD) in women. Two studies also showed 1.6 times the odds of early AMD in women, whereas the association with advanced AMD in women was not statistically significant. In men, osteoporosis was not significantly associated with AMD, and no statistically significant associations were observed for early AMD. Sensitivity analyses indicated that the relationship between osteoporosis and AMD in women, as well as between osteoporosis in men and AMD, was not robust in leave-one-out analyses

The meta-analysis followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and included prospective and retrospective cohort designs. Osteoporosis was defined as a bone mineral density T-score of −2.5 or lower at the femoral neck or other skeletal sites. Study quality, assessed with the Newcastle-Ottawa Scale, was high across all three studies.

To address causality, the investigators conducted bidirectional two-sample Mendelian randomization analyses using genome-wide association study data from European-ancestry populations. For osteoporosis, data from genome-wide association studies were selected from the FinnGen project, with a total sample size of nearly 484,000 patients.

AMD subtype data were obtained from the International AMD Genomics Consortium, the IEU Open genome-wide association studies platform, and FinnGen. The dry AMD data set included nearly 338,000 patients, and the wet AMD data set included nearly 338,000 patients.

Using the inverse variance weighted method as the primary estimator, genetically predicted osteoporosis was associated with 1.3 times the odds of dry AMD. Bayesian weighted Mendelian randomization yielded a similar estimate. No statistically significant causal effects were observed for wet AMD, early AMD, or advanced AMD. In addition, reverse Mendelian randomization analysis did not indicate any causal effect of AMD on osteoporosis, and no statistically significant horizontal pleiotropy or heterogeneity was detected in any of the analyses.

Following Bonferroni correction for four outcomes, only the osteoporosis-to–dry AMD association met the adjusted significance threshold. Multivariable Mendelian randomization adjusting for smoking, type 1 diabetes, type 2 diabetes, and glaucoma remained associated with 1.2 times the odds of dry AMD.

Genetic instruments were selected using standard genome-wide significance criteria and met strength thresholds, with no evidence of horizontal pleiotropy or substantial heterogeneity. Leave-one-out analyses generally supported result stability, although findings for early AMD were less consistent.

The researchers noted several limitations, including the small number of observational studies, heterogeneity in pooled analyses, potential sample overlap in FinnGen data for exposure and outcome, differences in data sources for advanced AMD, and restriction to European-ancestry populations. They also acknowledged the possibility of limited statistical power for certain AMD subtypes.

“This body of evidence collectively points toward a unidirectional causal pathway from [osteoporosis] to dry AMD,” the researchers wrote, adding that “patients with [osteoporosis]—especially elderly women—[should be considered] as a potential high-risk group for dry AMD in clinical practice.”

Disclosures: The researchers reported no known competing financial interests or personal relationships that could have influenced the work. No funding was reported.

Source: Experimental Gerontology