Abstract
BACKGROUND: Arteriovenous (AV) access, an essential component of dialysis treatment, has been implicated in the development of cardiovascular abnormalities. However, whether AV access blood flow and the duration of exposure to this flow, as represented by the dialysis vintage, have independent or interactive effects on cardiac outcomes remains unclear. METHODS: In this multicenter cross-sectional study, we enrolled hemodialysis patients who had undergone both transthoracic echocardiography and measurement of AV access flow, between April 2014 and January 2024. Left ventricular mass index (LVMI) was used as a surrogate marker of cardiovascular morbidity. AV access flow was quantified as brachial artery flow volume (FV) using pulsed Doppler ultrasonography. Multivariate regression models were applied to investigate the associations of FV and dialysis vintage with LVMI, and to assess potential interactions between these variables. RESULTS: A total of 241 patients were included. The mean FV was 0.71 ± 0.29 L/min, and the median dialysis vintage was 5.0 years (interquartile range, 3.0-11.0 years). Higher FV was significantly associated with increased LVMI (mean difference = 22.83; 95% confidence interval [CI]: 0.37 to 45.29; p = 0.046), whereas dialysis vintage was not (mean difference = 0.51; 95% CI: -0.41 to 1.45; p = 0.27). Moreover, no statistically significant interaction was observed (mean difference = 0.13; 95% CI: -3.14 to 3.40; p = 0.94). CONCLUSION: AV access flow may serve as a candidate variable in future left ventricular hypertrophy prediction algorithms and cardio-renal machine learning models because of its linear and independently significant association with LVMI.