ASSOCIATION OF RETINAL VESSEL CALIBERS AND LONGITUDINAL CHANGES IN ARTERIAL STIFFNESS: THE NAGAHAMA STUDY. (June 2018)

• Record Type: Journal Article
• Authors: Tabara, Y.; Kawashima-Kumagai, K.; Yamashiro, K.; Setoh, K.; Yoshikawa, M.; Miyake, M.; Nakayama, T.; Tsujikawa, A.; Matsuda, F.
• Is Part Of: Journal of hypertension. Volume 36(2018)Supplement 1
• Publication Date: 2018-06
• Subjects: Hypertension -- Periodicals; Hypertension -- Periodicals; 616.132005
• DOI: 10.1097/01.hjh.0000538964.73890.fc
• Languages: English
• ISSNs: 1473-5598

Abstract

Abstract : Objective: Morphological change in retinal vessel calibers has been reported as a marker of cardiovascular risk, but its association with arterial stiffening, a possible factor relating retinal vessel sings and cardiovascular outcomes, is not clear. The study aim was to clarify the relationship between retinal small vessel calibers and longitudinal change in large arterial stiffness in a sample of the general population. Design and method: The study included 6, 720 Japanese participants (52.1 ± 12.8 years). Central retinal arteriolar equivalent (CRAE) and venular equivalent (CRVE) were measured by fundus photography. Arterial stiffness was evaluated by brachial-to-ankle pulse wave velocity (baPWV) at baseline and at 5 years. Results: The overall change in baPWV (δ-baPWV) during mean follow-up 1, 814 ± 136 days was 41 ± 131 cm/sec (3.4 ± 9.9 % of baseline), and was significantly larger in men (52 ± 147 cm/sec) than in women (36 ± 122 cm/sec, p < 0.001). δ-baPWV was significantly increased in individuals with narrower CRAE (quartiles: Q1, 4.3 ± 10.6%; Q2, 3.3 ± 10.0%; Q3, 3.1 ± 9.3%; Q4, 3.1 ± 9.7%, p = 0.001), while no significant association was observed with CRVE. Although several factors also significantly differ among the CRAE quartiles, multivariate analysis identified CRAE as a significant inverse determinant of δ-baPWV (β = -0.033, p = 0.006) independent of the possible covariates including age (β = 0.433, p = < 0.001), male sex (β = 0.050, p < 0.001), BMI (β = 0.028, p = 0.024), mean BP (β = 0.059, p < 0.001), and baseline baPWV (β = -0.438, p < 0.001). The association of CRAE with δ-baPWV remained significant in a model further adjusted for CRVE (β = -0.040, p = 0.002). Further, the association between CRAE and δ-baPWV was prominent in a middle-aged (age Q2, β = -0.078, p = 0.002), but not younger (Q1, p = 0.232) or older (Q3, p = 0.427; Q4, p = 0.542) participants. Conclusions: Narrower CRAE in middle-age was associated with the long-term risk of arteriosclerosis in a general population sample.