Association between heart failure and arteriovenous access patency in patients with end-stage renal disease on hemodialysis

Abstract

Background: Heart disease and chronic kidney disease are often comorbid conditions owing to shared risk factors, including diabetes and hypertension. However, the effect of congestive heart failure (CHF) on arteriovenous fistula (AVF) and AV graft (AVG) patency rates is poorly understood. We hypothesize preexisting HF may diminish blood flow to the developing AVF and worsen patency.

Methods: We conducted a single-institution retrospective review of 412 patients with end-stage renal disease who underwent hemodialysis access creation from 2015 to 2021. Patients were stratified based on presence of preexisting CHF, defined as clinical symptoms plus evidence of reduced left ventricular ejection fraction (EF) (<50%) or diastolic dysfunction on preoperative echocardiography. Baseline demographics, preoperative measures of cardiac function, and dialysis access-related surgical history were collected. Kaplan-Meier time-to-event analyses were performed for primary patency, primary-assisted patency, and secondary patency using standard definitions for patency from the literature. We assessed differences in patency for patients with CHF vs patients without CHF, patients with a reduced vs a normal EF, and AVG vs AVF in patients with CHF.

Results: We included 204 patients (50%) with preexisting CHF with confirmatory echocardiography. Patients with CHF were more likely to be male and have comorbidities including, diabetes, chronic obstructive pulmonary disease, hypertension, and a history of cerebrovascular accident. The groups were not significantly different in terms of prior fistula history (P = .99), body mass index (P = .74), or type of hemodialysis access created (P = .54). There was no statistically significant difference in primary patency, primary-assisted patency, or secondary patency over time in the CHF vs non-CHF group (log-rank P > .05 for all three patency measures). When stratified by preoperative left ventricular EF, patients with an EF of <50% had lower primary (38% vs 51% at 1 year), primary-assisted (76% vs 82% at 1 year), and secondary patency (86% vs 93% at 1 year) rates than those with a normal EF. Difference reached significance for secondary patency only (log-rank P = .029). AVG patency was compared against AVF patency within the CHF subgroup, with significantly lower primary-assisted (39% vs 87% at 1 year) and secondary (62% vs 95%) patency rates for AVG (P < .0001 for both).

Conclusions: In this 7-year experience of hemodialysis access creation, reduced EF is associated with lower secondary patency. Preoperative CHF (including HF with reduced EF and HF with preserved EF together) is not associated with significant differences in overall hemodialysis access patency rates over time, but patients with CHF who receive AVG have markedly worse patency than those who receive AVF. For patients with end-stage renal disease and CHF, the risks and benefits must be carefully weighed, particularly for those with low EF or lack of a suitable vein for fistula creation.

Keywords: Arteriovenous access; Arteriovenous fistula; Arteriovenous graft; Congestive heart failure; Primary patency; Primary-assisted patency; Reduced ejection fraction; Secondary patency.