A comparison of arteriovenous fistulas and central venous lines for long-term chronic haemodialysis

• Published in: Pediatric nephrology (Berlin, West) West), 2013-02, Vol.28 (2), p.321-326
• Main Authors: Ma, Alison, Shroff, Rukshana, Hothi, Daljit, Lopez, Marina Munoz, Veligratli, Faidra, Calder, Francis, Rees, Lesley
• Format: Article
• Language: English
• Subjects: Adolescent; Adults; Antibiotics; Arteriovenous Shunt, Surgical - adverse effects; Care and treatment; Central venous catheters; Central Venous Catheters - adverse effects; Child; Child, Preschool; Confidence Intervals; Data collection; Female; Fistula; Fistula, Arteriovenous; Health aspects; Hemodialysis; Hospitalization; Hospitalization - statistics & numerical data; Humans; Infection - etiology; Infections; Kaplan-Meier Estimate; Laboratories; Male; Medicine; Medicine & Public Health; Nephrology; Original Article; Patient outcomes; Patients; Pediatrics; Renal Dialysis - methods; Renal Insufficiency, Chronic - therapy; Retrospective Studies; Survival analysis; Time Factors; Treatment Outcome; Urology; Venous access
• ISSN: 0931-041X; 1432-198X
• DOI: 10.1007/s00467-012-2318-2

Background Despite the Fistula First initiative there is still reluctance to use arteriovenous fistulas (AVF) for chronic haemodialysis (HD) in children. Our aim was to compare outcomes of AVFs and central venous lines (CVL) in children on chronic HD in a centre where AVF is the primary choice for vascular access. Patients and methods This was a retrospective case notes analysis of access complications, dialysis adequacy and laboratory outcomes in children who underwent dialysis for at least a year by AVF ( n  = 20, median age 14.2 years, range (2.9–16.5) and CVL ( n  = 5, median age 2.4 years, range 2.0–12.2) between January 2007 and December 2010. Results Primary access failure rate (patient-months) was 1 per 78.8 for AVF ( n  = 5) and 1 per 15.5 for CVLs ( n  = 7, p  = 0.3). Failure thereafter was 1 per 131.3 and 1 per 18.5 for AVF and CVLs respectively ( n  = 3 and 6 respectively; p  = 0.2). The annualised hospitalisation rate for access malfunction was 0.44% and 3.1% for AVFs and CVLs respectively ( p  = 0.004). Patients with AVFs had a lower infection rate of 0.25 per 100 patient-months compared with CVL at 3.2 per 100 ( p  = 0.002). There was no difference in dialysis adequacy or laboratory values between AVF and CVL groups. Access survival rates (including both primary and secondary access failure) were significantly higher for AVF compared with CVL ( p  = 0.0002, hazard ratio = 0.15, 95% confidence interval 0.04–0.37). Conclusions Patients with AVF spend less time in hospital than those dialysed by CVLs and have a much lower access infection rate. These findings emphasise the need to use AVF as first-line access for paediatric patients on chronic HD.