BACKGROUND: Vascular access (VA) is essential for hemodialysis (HD) patients, and its dysfunction is a major complication. However, little is known about outcomes in patients with recurrent VA dysfunction. We explored the influence of recurrent VA dysfunction on cardiovascular (CV) events, death and VA abandonment. METHODS: This is a single-center, retrospective study conducted in patients who underwent VA surgery between 2009 and 2014. VA dysfunction was defined as VA stenosis or thrombosis requiring intervention after the first successful cannulation. Patients with ≥2 interventions within 180 days were categorized as having recurrent VA dysfunction. Outcomes were analyzed using Cox proportional hazards model before and after propensity score matching. RESULTS: Of 766 patients (ages 59.6 ± 14.3 years, 59.7% male), 10.1% were in the recurrent VA dysfunction group. Most baseline parameters after matching were similar between the recurrent and non-recurrent groups. A total of 213 propensity score-matched patients were followed for 28.7 ± 15.8 months, during which 46 (21.6%), 30 (14.1%) and 14 (6.6%) patients had de novo CV outcomes, died and abandoned VA, respectively. Recurrent VA dysfunction after adjustment remained an independent risk factor for CV events (adjusted hazards ratio (aHR), 2.71; 95% CI 1.48-4.98; p = 0.001). Moreover, recurrent VA dysfunction predicted composite all-cause mortality (ACM)/CV events (aHR 1.99; 95% CI 1.21-3.28; p = 0.007). CONCLUSIONS: Recurrent VA dysfunction was a novel independent risk factor for CV and composite ACM/CV events in HD patients, but not for VA abandonment. Patients with recurrent vascular dysfunction should be carefully monitored not only for VA patency but also for CV events.
BACKGROUND: Vascular access (VA) is essential for hemodialysis (HD) patients, and its dysfunction is a major complication. However, little is known about outcomes in patients with recurrent VA dysfunction. We explored the influence of recurrent VA dysfunction on cardiovascular (CV) events, death and VA abandonment. METHODS: This is a single-center, retrospective study conducted in patients who underwent VA surgery between 2009 and 2014. VA dysfunction was defined as VA stenosis or thrombosis requiring intervention after the first successful cannulation. Patients with ≥2 interventions within 180 days were categorized as having recurrent VA dysfunction. Outcomes were analyzed using Cox proportional hazards model before and after propensity score matching. RESULTS: Of 766 patients (ages 59.6 ± 14.3 years, 59.7% male), 10.1% were in the recurrent VA dysfunction group. Most baseline parameters after matching were similar between the recurrent and non-recurrent groups. A total of 213 propensity score-matched patients were followed for 28.7 ± 15.8 months, during which 46 (21.6%), 30 (14.1%) and 14 (6.6%) patients had de novo CV outcomes, died and abandoned VA, respectively. Recurrent VA dysfunction after adjustment remained an independent risk factor for CV events (adjusted hazards ratio (aHR), 2.71; 95% CI 1.48-4.98; p = 0.001). Moreover, recurrent VA dysfunction predicted composite all-cause mortality (ACM)/CV events (aHR 1.99; 95% CI 1.21-3.28; p = 0.007). CONCLUSIONS: Recurrent VA dysfunction was a novel independent risk factor for CV and composite ACM/CV events in HDpatients, but not for VA abandonment. Patients with recurrent vascular dysfunction should be carefully monitored not only for VA patency but also for CV events.