ľֱ

Vascular Access Matters in Survival on Dialysis

— Central venous catheters may account for the poorer survival among hemodialysis patients compared with peritoneal dialysis, according to a registry study.

MedpageToday

Central venous catheters may account for the poorer survival among hemodialysis patients compared with peritoneal dialysis, according to a registry study.

Use of a central port for dialysis was associated with 80% elevated one-year mortality risk compared with peritoneal dialysis (adjusted hazard ratio 1.8, 95% confidence interval 1.6 to 1.9), Jeffrey Perl, MD, of Saint Michael's Hospital in Toronto, and colleagues found.

But hemodialysis through an arteriovenous fistula or graft yielded similar one-year survival to that of peritoneal dialysis (adjusted HR 0.9, 95% CI 0.8 to 1.1), they reported online in the Journal of the American Society of Nephrology.

Action Points

  • Point out that this study demonstrates a significant difference in mortality in association with type of access used for initiation of hemodialysis and this difference may account in part for differences in mortality in the first year of dialysis when comparing initiation of dialysis with either peritoneal or vascular access.
  • Note that the study is subject to several limitations including selection bias introduced by nonrandom allocation as well as confounding on the basis of unmeasured differences between patients that may influence both incident vascular access and dialysis modality choice.

Prior studies have found a survival advantage to peritoneal dialysis compared with hemodialysis overall, which has been chalked up to better preservation of residual kidney function with peritoneal dialysis.

But the new findings, stratified by vascular access type, challenge the conclusion that peritoneal dialysis itself is better, Perl's group suggested.

"This suggests that vascular access-related morbidity/mortality and case-mix differences that coincide with hemodialysis vascular access type are more likely to explain the higher early mortality attributed to hemodialysis," they wrote in JASN.

Central venous catheters boost the risk of sepsis and hospitalization, and may also indicate comorbidities and other factors in eligibility for surgical vascular access, Perl and colleagues noted.

Their study included 40,526 adults in the Canadian Organ Replacement Register who started dialysis from 2001 through 2008.

Starting dialysis with a central venous catheter was the most common route, with only 19% of the total population getting peritoneal dialysis, and 21.4% of the hemodialysis population initiating it with an arteriovenous fistula or graft.

During the entire course of follow-up, 31% of the 7,412 peritoneal dialysis patients died, as did 44.1% of the 24,437 who started hemodialysis with a central venous catheter and 33.9% of the 6,663 who started with an arteriovenous fistula or graft.

As seen in prior studies, hemodialysis was associated with higher adjusted one-year mortality compared with peritoneal dialysis (HR 1.5, 95% CI 1.4 to 1.7).

Five-year cumulative mortality remained higher with hemodialysis port users (adjusted HR 1.2, 95% CI 1.0 to 1.3) but lower with arteriovenous fistulas or grafts (adjusted HR 0.80, 95% CI 0.8 to 0.9) compared with peritoneal dialysis.

Sensitivity analyses showed similar results when the researchers excluded patients who died soon after starting dialysis or those who were referred late, or when they censored those who changed dialysis modalities.

The researchers noted that Canada has one of the highest rates of central venous catheter use for dialysis among developed countries, which may be contributing to early hemodialysis-related mortality.

Notably, survival among hemodialysis port users appeared to be worsening over time relative to peritoneal dialysis patients when comparing eras within the study period (2005-2008 versus 2001-2004), which Perl's group speculated "reflect a more contemporary hemodialysis patient population characterized by both a higher burden of comorbidities and higher rates of incident central venous catheter use."

The group cautioned, though, that the study was limited by selection bias from non-random allocation of patients to type of vascular access and type of dialysis, and by residual confounding and the use of an administrative database that didn't record changes in vascular access type.

Disclosures

Perl reported having received speaking honoraria from Amgen Canada and Baxter Healthcare Canada, and holding an unrestricted educational fellowship from Baxter Healthcare Canada.

Primary Source

Journal of the American Society of Nephrology

Perl J, et al "Hemodialysis vascular access modifies the association between dialysis modality and survival" J Am Soc Nephrol 2011; DOI: 10.1681/ASN.2010111155.