Acta Ophthalmologica

Genetic study does not support vitamin D lowering eye disease risk

In Copenhagen, each increase in a vitamin D-raising genetic score had a hazard ratio, comparing diagnosis rates over time, of 1.01; 1 means no difference.

Figure 1 from Acta Ophthalmologica
Open the figure at full size
Figure 1. Beheshti et al.
Mendelian randomisation in 406,861 peoplePopulations
Follow

In 25,445 people from the Copenhagen General Population Study and 372,477 from UK Biobank, researchers examined whether initial blood vitamin D levels were linked to later age-related macular degeneration. This eye disease damages central vision. These observational analyses linked higher vitamin D levels to higher disease risk in Copenhagen but lower risk in UK Biobank.

The researchers also analysed genetic variants linked to lifelong higher vitamin D levels. These analyses included 107,793 people in Copenhagen, 406,861 in UK Biobank and 31,858 in a genetic case-control consortium. Each increase in the Copenhagen vitamin D-raising genetic score had a hazard ratio of 1.01, with similar results elsewhere. Its range of likely values included no difference. The researchers concluded that the findings did not support a causal role for vitamin D in the disease.

Why it matters

For age-related macular degeneration, the study addressed whether vitamin D itself might affect disease risk, rather than merely being associated with it.

Caveats

The study used observational and genetic analyses, not a randomised trial testing vitamin D supplements. The genetic analyses examined lifelong differences in vitamin D levels.

From the paper

Participants

Observationally, we studied 25 445 individuals with measured baseline plasma 25-hydroxyvitamin D (536 incident AMD events) in the Copenhagen General Population Study (CGPS) and 372 477 individuals (7842 incident AMD events) in the UK Biobank using Cox regression analysis for hazard ratios.

Methods

Result

Genetically, each increase in the 25-hydroxyvitamin D-increasing weighted CGPS allele score was associated with a hazard ratio of 1.01 (95% confidence interval: 1.00–1.02) in CGPS, with similar results in the other studies.

Results

Quoted word for word from the full text on onlinelibrary.wiley.com.