Background: Retinal neurodegeneration, an easily accessible biomarker of dementia risk, is exacerbated by age and hypertension; however, the relative roles of systolic and diastolic blood pressure (SBP and DPB) remain unclear. This study aimed to determine the cross-sectional and longitudinal associations between BP and atherosclerosis levels along with the different retinal neurodegeneration parameters.
Methods: This study used cross-sectional data from the United Kingdom (UK) BioBank (UKB) and longitudinal data from the Chinese Ocular Imaging Project (COIP). The macular ganglion cell-inner plexiform layer thickness (mGCIPLT) and macular retinal nerve fiber layer thickness (mRNFLT) were measured using spectral domain optical coherence tomography imaging. Swept-source optical coherence tomography was performed at each follow-up visit to obtain the longitudinal trajectory of the mGCIPLT and peripapillary RNFLT (pRNFLT) in the COIP cohort. Multivariable linear models were used to analyze the cross-sectional and longitudinal associations between BP metrics and retinal measurements.
Results: In a cross-sectional analysis of 22,801 participants from the UKB, thinner mGCIPLT was related to older age (per 10 years, β = −0.274, 95% confidence interval (CI): −0.358 to −0.189, p < 0.001), female sex (female vs male, β = −0.897, 95% CI: −1.031 to −0.763, p = 0.000), higher SBP (per 10 mmHg increase, β = −0.085, 95% CI: −0.125 to −0.045, p = 0.000), and higher DBP (per 10 mmHg increase, β = −0.105, 95% CI: −0.174 to −0.036, p = 0.003), and was significantly associated with higher mean arterial pressure (MAP per 10 mmHg increase, β = −0.116, 95% CI: −0.176, −0.056, p = 0.000) and higher mean pulse pressure (MPP per 10 mmHg increase, β = −0.099, 95% CI: −0.155, −0.043, p = 0.001). In a longitudinal analysis of 2,012 eligible COIP participants, higher levels of baseline SBP, DBP, MAP, and MPP were associated with faster thinning in mGCIPLT and pRNFLT (all p < 0.001). The strongest association was with MAP, which produced an effect on mGCIPLT (β = −0.118, 95% CI: −0.175 to −0.061, p < 0.001) per 10 mmHg increase, comparable to a 5-year increase in age (β = −0.210, 95% CI: −0.282 to −0.138, p < 0.001). The results of the analysis of mRNFL and pRNFL were consistent with those of mGCIPLT.
Conclusion: BP levels were independently and consistently associated with various retinal neurodegenerative exacerbations, both cross-sectionally and longitudinally, regardless of the race and disease status. BP plays a key role in neurodegeneration, and long-term prevention in the population requires the control of BP levels.