Study Finds Wide Geographic Variation in ESRD Care

Across the United States, there is wide regional variation in functional incident arteriovenous fistula (AVF) frequency and adjusted end-stage renal disease (ESRD) mortality. That was among the findings of a study conducted recently by Devin S. Zarkowsky, MD, and colleagues. The researchers reported results of the retrospective analysis in JAMA Surgery [2015; 150(8):764-770].

When used at first hemodialysis, survival in patients with ESRD has been shown to be improved with use of AVF access compared with other modalities. In 2005, the Fistula First Breakthrough Initiative (now known as Fistula First Catheter Last) established national guidelines for use of AVF access and set a goal of 50% AVF access frequency at the incident hemodialysis episode, a goal that has yet to be reached. As of September 2014, only 20.3% of patients received their initial hemodialysis treatment with an AVF.

In 2014 the United States Renal Data System (USRDS) released a heat map depicting significant regional variation in the prevalence of ESRD with the United States. There are also differences in therapy for ESRD: data gathered over the past three decades show population density (urban, suburban, or rural) as one contributing factor to the geographic disparities in treatment modality.

According to the researchers in the current study, there are few data on geographic trends in the population of patients with ESRD undergoing hemodialysis. The study was designed to quantify the frequency with which patients undergo hemodialysis via an AVF at the initial hemodialysis session (categorized by the ESRD Network Program). The researchers also sought to examine regional mortality hazards to test the hypothesis that incident AVF frequency and patients mortality hazard would vary inversely when stratified by ESRD Network Program within the United States.

The researchers utilized univariate analyses and multivariable logistic regressions to compare patient characteristics, incident AVF frequencies, and corrected mortality hazards between ESRD Network Programs. The programs comprise 18 states, commonwealths, and protectorates in which patients receive hemodialysis. Data on 464,547 patients who were beginning renal replacement therapy between January 1, 2006, and December 31, 2010, were analyzed; analysis began on April 1, 2013, and ended August 3, 2014.

Google’s GeoMaps were used to create a heat map depicting functional AVFs at incident hemodialysis for all 18 ESRD Network Programs in the United States. Rates of incident AVF use were highest in Network 1 (New England; 21.3%) and Network 16 (Pacific Northwest; 22.2%). The highest rates were double those in Network 7 (Florida), Network 14 (Texas), and Network 18 (Southern California; 11.1%) (P<.001).

After controlling for age and comorbidities, a Cox proportional hazards regression model showed a significant difference in mortality hazard between ESRD Network Programs. Mortality hazards for patients with ESRD were significantly lower in Network 1 (New England), Network 11 (Northern Midwest), Network 17 (Northern California), and Network 18 (Southern California) compared with Network 13 (Arkansas, Louisiana, and Oklahoma), with a 28% variation (hazard ratio [HR], 0.99; 95% confidence interval [CI], 0.96-1.03 vs HR, 1.27; 95% CI, 1.22-1.31; P<.001).

In a logistic regression model adjusted for age and comorbidities, the factors influencing hemodialysis therapy beginning with an AVF rather than a hemodialysis catheter included nephrology care and insurance coverage (Medicaid, Medicare, or private). There was a significant association between nephrology care and increased odds of incident AVF use.

The highest frequency of nephrology care in the study population was in Network 1 (New England; 74.3%); Network 1 also has a commensurately high frequency of AVF-based incident hemodialysis access (21.3%). Network 3 (New Jersey, Puerto Rico, and the US Virgin Islands) had the lowest frequency of nephrology care (54.3%) and had among the lowest frequencies of fistula-based incident hemodialysis access (13.3%).

Negative correlates for AVF-based hemodialysis included congestive heart failure (CHF) and immobility.

Limitations to the study cited by the researchers included an inability to ascertain through the USRDS database whether the reported deaths in the study population were directly caused by ESRD; the possibility that patients moved between ESRD Networks during the study period; and access selection may reflect patient-, physician- or system-level patterns the researchers could not account for.

“There is marked regional variation in functional incident AVF frequency and risk-adjusted ESRD mortality across the United States. No region achieves the 50% target for incident AVF access. This study suggests two targets for improving ESRD care delivery. First, nephrologist involvement increases functional incident AVF frequencies logarithmically, highlighting the necessity of a multidisciplinary team approach. Second, AVF construction in patients with ESRD and CHF can be achieved with superior regional mortality outcomes,” the researchers said.

Takeaway Points

  1. When used at initiation of hemodialysis, access with an arteriovenous fistula (AVF) is associated with improved survival in patients with end-stage renal disease compared with other modalities.
  2. This study was designed to quantify regional variation in AVF access at first hemodialysis and the effect of the variation on mortality in the US Renal Data System.
  3. The study found significant differences in frequency of use of incident AVF access as well as associated differences in risk-adjusted ESRD mortality. Some of the variations can be explained by differences in access to nephrology care and patient comorbidities.