A functional vascular access is crucial to the successful delivery of maintenance hemodialysis therapy. National organizations agree that a well-functioning arteriovenous (AV) fistula (AVF) is the preferred access for patients on hemodialysis, while also agreeing that vascular access should be tailored to an individual patient’s health status and other considerations.
DOPPS (Dialysis Outcomes and Practice Patterns Study), an international prospective cohort study of hemodialysis practices and patient outcomes, has been ongoing since 1996. In each participating country, patients ≥18 years of age and hemodialysis facilities are randomly selected to be representative of all hemodialysis patients and facilities. There were variations in outcomes from DOPPS phase 5 (2012-2014) across 21 countries. Use of an AVF varied from 49% to 92%, use of an AV graft (AVG) varied from 2% to 18%, and use of a central venous catheter varied from 2% to 49%.
There were also differences in survival of AVFs following successful first use (cumulative functional survival) across the three continental DOPPS regions; Japan reported longer AVF survival than Europe, Australia, New Zealand (Europe/ANZ), and North America. Both dialysis and practitioner practices may affect AVF survival.
In light of these international variations in vascular access type as well as outcomes, Ronald L. Pisoni, PhD, MS, and colleagues conducted an analysis designed to examine international differences in (1) AVF creation location, i.e., upper versus lower arm; (2) trends in AVF location from DOPPS phases 1 to 5; and (3) successful use of newly created AVFs and AVGs (together referred to as “AV accesses”). Results of the analysis were reported in the American Journal of Kidney Diseases [2018;71(4):469-478].
Individuals participating in DOPPS and included in the current analysis were from the United States, Japan, and Europe/ANZ [Belgium, France, Germany, Italy, Spain, Sweden, United Kingdom, Australia, and New Zealand]. Of the participants, 3850 received 4247 new AVFs and 842 received 1129 new AVGs in the period 2009 to 2015. Location trends for AVFs were calculated based on 38,868 AVFs recorded in cross-sections from DOPPS phases 1 to 5 (1996-2015).
In DOPPS phases 1 to 5, >96% of all AVFs were located in an upper-extremity location. There were substantial international differences in the location of arm AVFs: in Japan, ≥93% were in the lower arm (DOPPS phases 1 to 5); in Europe/ANZ, 77% were in the lower arm (DOPPS phases 1 to 2, declining to ~65% in DOPPS phases 4 to 5); in the United States, 70% were in the lower arm in DOPPS phase 1 while only 32% were in the lower arm in DOPPS phase 5. When the analysis was restricted to patients of dialysis vintage <90 days, similar international trends were seen.
Facility variation in percentage of lower-arm AVFs in DOPPS phases 4 to 5 indicated a tight distribution within Japan, with 95% of AVFs in the lower arm in the median facility; broad variation across facilities in the 11 Europe/ANZ countries, with 67% of arm AVFs located in the lower arm in the median facility; and large variation across facilities in the United States, with 38% of arm AVFs located in the lower arm in the median facility.
Additional aspects of location placement of AV access were examined in DOPPS phases 4 and 5; most patients (86%) contributed only one AV access to those analyses. Lower-arm location among AVFs was highest in Japan (89%), compared with 60% in Europe/ANZ, and 34% in the United States. The lower figure in the United States was seen despite US patients having the lowest mean age. US patients had the highest prevalence of diabetes, and there was little difference among patients in the United States having lower- versus upper-arm AVF placement.
In DOPPS phases 4 to 5, 12% to 13% of all created AV access in Europe/ANZ and Japan was AVGs, compared with 25% in the United States. Compared with patients with a newly created AVF, those with a newly created AVG had longer median dialysis vintage and lower percentage of males in all three regions; mean age was slightly older in the United States and Europe/ANZ, and higher in Japan.
DOPPS phases 4 to 5 defined successful AV access use as a newly created AV access being used for ≥30 continuous days (for typical thrice-weekly hemodialysis) during follow-up. In analyses without follow-up time requirement beyond 30 days, successful AVF use was 87% in Japan, 67% in Europe/ANZ, and 64% in the United States. Results of a secondary calculation allowing AVFs to have up to 7 months after creation to demonstrate successful AVF use, the proportions of successful use were 89% in Japan, 76% in Europe/ANZ, and 77% in the United States.
Successful use of AVGs created in DOPPS phases 4 and 5 were assessed in each international region. In analyses without time requirements, successful AVG use was 86% in Japan, 75% in Europe/ANZ, and 78% in the United States. In alternative calculations that allowed AVGs up to 3 months following creation to demonstrate successful use, the proportions of successful use were 89% in Japan, 78% in Europe/ANZ, and 83% in the United States. In Europe/ANZ and the United States, 3% to 6% of newly created AVGs failed within 30 days of first use.
There was wide variation by international region in time until the first day of successful AVF use among newly created AVFs that were successfully used for ≥30 days in DOPPS phases 4 to 5: Japan, median of 10 days; Europe/ANZ, median of 46 days; and United States, median 82 days. Among newly created AVGs, median time until first day of successful AVG use was 6 days in Japan, 24 days in Europe/ANZ, and 29 days in the United States.
A limitation to the analyses cited by the authors was dependence on accurate reporting of vascular access data from each study site.
In conclusion, the authors said, “Large international differences exist in the percentage of arm AVFs created in the lower arm, predictors of lower- versus upper-arm AVF location, successful use of newly created AVFs, and time to first use of newly created AVFs and AVGs. These findings, coupled with large differences in AVF survival across these international regions, raise important questions regarding what is best practice and how it is best achieved to optimize vascular access outcomes. Importantly, the large shift in predominantly lower- to upper-arm AVF creation in the United States, even in ‘ideal’ young patients for AVF creation, raises the question of whether this practice shift may place patients at greater risk for exhausting available sites for future AVF creation when needed. This could have substantial health implications for patients and economic implications for healthcare systems.”
- There is wide international variation in the use of arteriovenous fistulas (AVFs) and arteriovenous grafts (AVGs), and in maturation time and location for AVFs.
- Analyses of data from DOPPS (Dialysis Outcomes and Practice Patterns Study) included dialysis patients from 21 countries; 3850 patients who received new AVFs and 842 patients who received new AVGs in 2009 to 2015.
- The international variations seen in the analyses “challenge what constitutes best practice,” the researchers said.