-
A total of 5,440 participants were recruited as baseline from 2010 to 2011. Cognitive function was assessed by the MMSE, and retinal metrics were collected by retinal photography and spectral-domain optical coherence tomography during the first follow-up period between 2012 to 2013, with 4,004 participants attending this appointment. Of these subjects, 76 and 143 did not finish the cognitive function assessment or retinal-related metrics assessment, respectively, leaving 3,785 participants in the final analysis (CI group 108, non-CI group 3,677, Figure 1).
The mean age of the participants was 53.7 years, with 2,143 (56.6%) being male. The baseline characteristics of the two groups are shown in Table 1. In the CI group, the participants were older and had a lower educational level than that of the non-CI group. The percentage of glaucoma was also higher in the CI group than in the non-CI group. The other epidemiological characteristics and medical histories were not statistically different between the two groups (Table 1).
Parameters Missing (n) CI (n = 108) Non-CI (n = 3,677) P value Age (quartile), y n = 0 57.8 (51.8, 72.6) 51.4 (45.2, 59.1) < 0.001 Male (n), % n = 0 71 (65.7) 2,072 (56.4) 0.061 Han ethnic (n), % n = 0 107 (99.1) 3,628 (98.7) 0.900 Educational level (n), % n = 0 0.0074 Illiterate 5 (4.6) 46 (1.3) elementary school 11 (10.2) 297 (8.1) middle school or above 92 (85.2) 3,334 (90.7) Smoking (n), % n = 0 38 (35.2) 1,164 (31.7) 0.463 Alcohol Drinking (n), % n = 0 13 (12.0) 530 (14.4) 0.578 Past medical history (n), % Hypertension n = 0 56 (51.9) 1,677 (45.6) 0.204 Diabetes mellitus n = 0 14 (13.0) 398 (10.8) 0.436 Dyslipidemia n = 0 49 (45.4) 1,802 (49.0) 0.495 Glaucoma n = 754 4 (4.8) 31 (1.1) 0.003 Cataract surgery n = 756 2 (2.4) 62 (2.1) 0.698 Note. CI: cognitive impairment. Table 1. Basic characteristics of participants regarding the prevalence of cognitive impairment
In the second step, we analyzed the relationship between cognitive function and the different retinal metrics. Compared with the non-CI group, the CI group had significantly thinner RNFL thickness (odds ratio (OR): 0.973, 95% confidence interval (CI): 0.956–0.990). After adjustment for the relevant risk factors including age, sex, han ethnicity, educational level, smoking habit, alcohol drinking, past history of hypertension, diabetes mellitus, dyslipidemia, and glaucoma and cataract surgery, the difference is still valid (OR: 0.973, 95% CI: 0.953–0.994). The CRAE and AVR were not significantly different between the two groups after adjusting for the basic risk factors (Table 2).
Parameters Including OR 95% CI P value RNFL n = 3,332 0.973 0.956–0.990 0.002 Adjusted 0.973 0.953–0.994 0.013 CRAE n = 3,554 0.988 0.979–0.998 0.019 Adjusted 0.993 0.982–1.005 0.268 AVR n = 3,554 0.194 0.024–1.596 0.127 Adjusted 0.313 0.029–3.421 0.341 Note. Adjusted: adjusted by age, sex, han ethnicity, educational level, smoking, alcohol drinking, past history of hypertension, diabetes mellitus, dyslipidemia, glaucoma and cataract surgery. OR: Odds Ratio; RNFL: retinal nerve fiber layer; CI: confidence interval; CRAE: central retinal arteriolar equivalents; AVR: central retinal arteriolar equivalents /central retinal venous equivalents. Table 2. Adjusted odds ratios for cognitive function and different retinal metrics
To determine whether different arterial stenoses had different effects on retinal metrics for CI, we separated the participants according to their ICAS, ECAS, and PAD status. For RNFL, we showed that RNFL was significantly thinner in CI patients than in the non-ICAS and non-ECAS subgroups, with this association remaining for RNFL and CI after adjusting for risk factors shown in Table 1 (non-ICAS group: OR: 0.974, 95% CI: 0.956–0.993 for crude, OR: 0.973, 95% confidence interval: 0.950–0.996 for adjusted; non-ECAS group: OR: 0.973, 95% CI: 0.956–0.991 for crude, OR: 0.973, 95% CI: 0.953–0.994 for adjusted). Interestingly the association also existed in the PAD subgroup (OR: 0.947, 95% CI: 0.912–0.984 for crude, and OR: 0.943, 95% CI: 0.899–0.990 for adjusted). For CRAE and AVR, no significant difference was observed after adjusting for the risk factors (Figure 2 A for RNFL, B for CRAE, C for AVR).
Relationship of Retinal Nerve Fiber Layer Thickness and Retinal Vessel Calibers with Cognitive Impairment in the Asymptomatic Polyvascular Abnormalities Population
doi: 10.3967/bes2024.020
- Received Date: 2023-05-15
- Accepted Date: 2023-08-15
Abstract:
There is no competing interest in our study.
Citation: | WANG Dan Dan, WANG An Xin, ZHANG Xiao Li, WEI Wen Bin, WU Shou Ling, ZHAO Xing Quan. Relationship of Retinal Nerve Fiber Layer Thickness and Retinal Vessel Calibers with Cognitive Impairment in the Asymptomatic Polyvascular Abnormalities Population[J]. Biomedical and Environmental Sciences, 2024, 37(2): 196-203. doi: 10.3967/bes2024.020 |