The causal effect of hypertension, intraocular pressure, and diabetic retinopathy: a Mendelian randomization study.

Background: Previous research has indicated a vital association between hypertension, intraocular pressure (IOP), and diabetic retinopathy (DR); however, the relationship has not been elucidated. In this study, we aim to investigate the causal association of hypertension, IOP, and DR.

Methods: The genome-wide association study (GWAS) IDs for DR, hypertension, and IOP were identified from the Integrative Epidemiology Unit (IEU) Open GWAS database. There were 33,519,037 single-nucleotide polymorphisms (SNPs) and a sample size of 1,030,836 for DR. There were 16,380,466 SNPs and 218,754 participants in the hypertension experiment. There were 9,851,867 SNPs and a sample size of 97,465 for IOP. Univariable, multivariable, and bidirectional Mendelian randomization (MR) studies were conducted to estimate the risk of hypertension and IOP in DR. Moreover, causality was examined using the inverse variance weighted method, and MR results were verified by numerous sensitivity analyses.

Results: A total of 62 SNPs at the genome-wide significance level were selected as instrumental variables (IVs) for hypertension-DR. The results of univariable MR analysis suggested a causal relationship between hypertension and DR and regarded hypertension as a risk factor for DR [ = 0.006, odds ratio (OR) = 1.080]. A total of 95 SNPs at the genome-wide significance level were selected as IVs for IOP-DR. Similarly, IOP was causally associated with DR and was a risk factor for DR ( = 0.029, OR = 1.090). The results of reverse MR analysis showed that DR was a risk factor for hypertension ( = 1.27×10, OR = 1.119), but there was no causal relationship between DR and IOP ( > 0.05). The results of multivariate MR analysis revealed that hypertension and IOP were risk factors for DR, which exhibited higher risk scores ( = 0.001, OR = 1.121 and = 0.030, OR = 1.124, respectively) than those in univariable MR analysis. Therefore, hypertension remained a risk factor for DR after excluding the interference of IOP, and IOP was still a risk factor for DR after excluding the interference of hypertension.

Conclusion: This study validated the potential causal relationship between hypertension, IOP, and DR using MR analysis, providing a reference for the targeted prevention of DR.