Outcomes of vascular access for hemodialysis: A systematic review and meta-analysis

Jehad Almasri, Mouaz Alsawas, Maria Mainou, Reem A. Mustafa, Zhen Wang, Karen Woo, David L. Cull, Mohammad H Murad

Research output: Contribution to journalArticle

45 Scopus citations

Abstract

Background The decision about the type and location of a hemodialysis vascular access is challenging and can be affected by multiple factors. We explored the effect of several a priori chosen patient characteristics on access outcomes. Methods We searched MEDLINE, Embase, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Scopus through November 13, 2014. We included studies that evaluated patency, mortality, access infection, and maturation of vascular access in adults requiring long-term dialysis. Pairs of reviewers working independently selected the studies and extracted the data. Outcomes were pooled across studies using the random-effects model. Results Two hundred studies met the eligibility criteria reporting on 875,269 vascular accesses. Overall, studies appeared to have provided incidence rates at low to moderate risk of bias. The overall primary patency at 2 years was higher for fistulas than for grafts and catheters (55%, 40%, and 50%, respectively). Patency was lower in individuals with diabetes, coronary artery disease, older individuals, and in women. Mortality at 2 years was highest with catheters, followed by grafts then fistulas (26%, 17%, and 15%, respectively). Conclusions The current evidence remains in support of autogenous access as the best approach when feasible. We provide incidence rates in various subgroups to inform shared decision making and facilitate the conversation with patients about access planning.

Original languageEnglish (US)
Pages (from-to)236-243
Number of pages8
JournalJournal of Vascular Surgery
Volume64
Issue number1
DOIs
StatePublished - Jul 1 2016

    Fingerprint

ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine
  • Surgery

Cite this