Blood pressure and intraocular pressure in the Canadian Longitudinal Study on Aging: a bidirectional Mendelian randomization analysis

Purpose To investigate the association of blood pressure (BP) with intraocular pressure (IOP), determine whether these associations are modified by genetic susceptibility to elevated IOP, and assess potential causality using Mendelian randomization (MR). Methods Cross-sectional analyses were conducted using baseline data from the Canadian Longitudinal Study on Aging (CLSA) Comprehensive Cohort. Non-closely related participants of European descent with complete data for IOP and BP were eligible for individual-level data analyses ( N = 21,633). Associations of systolic BP (SBP), diastolic BP (DBP) and pulse pressure (PP) with corneal-compensated IOP were assessed using multivariable linear regression. An IOP polygenic risk score (PRS) was used to measure genetic susceptibility to elevated IOP. One-sample and two-sample MR analyses were conducted to evaluate potential causality. Results Multivariable linear regression found SBP, DBP, and PP to be positively associated with IOP after adjustment for relevant covariates ( P ≤ 3.82 × 10 −17 ). Interaction analyses did not find the IOP PRS to influence these associations ( P ≥ 0.122). One-sample MR found genetically predicted SBP to be positively associated with IOP ( β = 0.227 mm Hg per 10 mmHg increase in SBP, 95% confidence interval (CI) [0.074, 0.381], P = 0.004). Reverse one-sample MR did not provide evidence of IOP having a causal effect on BP ( P ≥ 0.103). Similar results for the observational and bidirectional one-sample MR analyses were obtained after excluding participants taking IOP-lowering or antihypertensive medications. Two-sample MR provided some evidence of a positive causal effect of SBP on IOP. Significant positive associations were observed using the inverse-variance weighted multiplicative random-effects (IVW-MRE; β = 0.216 mmHg per 10 mmHg increase in SBP, 95% CI [0.043, 0.390], P = 0.015), MR contamination mixture (MR-ConMix; β = 0.314 mmHg per 10 mmHg increase in SBP, 95% CI [0.134, 0.565], P = 0.001), and MR Pleiotropy RESidual Sum and Outlier (MR-PRESSO; β = 0.239 mmHg per 10 mmHg increase in SBP, 95% CI [0.075, 0.402], P = 0.004) methods. However, there was evidence of directional pleiotropy based on the MR-Egger intercept test ( P = 0.049). Similar two-sample MR results were obtained using IOP summary statistics from that CLSA after excluding participants taking IOP-lowering medications, although the MR-Egger intercept test was non-significant ( P = 0.062). Analyses using Steiger-filtered genetic instruments provided some support for the primary two-sample MR results, but not the primary one-sample MR results. Conclusion SBP, DBP, and PP are positively associated with IOP and genetic susceptibility to elevated IOP does not appear to modify these associations. One-sample and two-sample MR analyses provided some evidence of a potential causal effect of SBP on IOP. If confirmed, a causal effect of BP on IOP could lead ophthalmologists treating elevated IOP to encourage their patients to also have their BP under control.