Background
Evidence regarding the relationship between estimated glomerular filtration rate (eGFR) and arterial stiffness is limited, and the data analysis is not sufficient to clarify the true relationship between the two. We aimed to investigate the relationship between eGFR and brachial-ankle pulse wave velocity (baPWV) in Japanese.
Methods
The present study was a cross-sectional study. Nine hundred twelve Japanese men and women, aging 24—84 years old, received a health medical check-up program including the results from baPWV inspection and various standardized questionnaires in a health examination center in Japan. The main outcome measures included eGFR, baPWV, fatty liver and postmenopausal status. Abdominal ultrasonography was used to diagnose fatty liver. Postmenopausal state was defined as beginning 1 year after the cessation of menses.
Results
The average age of the 912 selected participants was 51.5 ± 9.6 years old, and about 57.6% of them were male. The participants’ eGFR distribution was median 69.29 (min 39, max 122.28). The results of multivariate linear regression showed eGFR was not independently associated with baPWV after adjusting potential confounders (β = − 1.11, 95%CI -2.25 to 0.03), this is inconsistent with the result of eGFR (quartile) as a categorical variable (p for trend was 0.038). A non-linear relationship was detected between eGFR and baPWV, whose point was 77.05. The effect sizes and the confidence intervals of the left and right sides of inflection point were − 2.80 (− 4.41 to − 1.19) and 1.84 (− 0.50, 4.17), respectively. Subgroup analysis showed, the change in the elderly population is more pronounced (P for interaction = 0.018; − 2.83 with ≤60 year vs − 6.12 with > 60 year). The same trend was also seen in hypertensive people (P for interaction = 0.018; − 4.55 with hypertension vs − 0.82 with non-hypertension).
Conclusion
The relationship between eGFR and baPWV is non-linear. eGFR was negatively related to baPWV when eGFR is less than 77.05.