Cost-Effectiveness of Vascular Access Referral Policies Varies by Age

The gold standard of vascular access for hemodialysis has long been an arteriovenous fistula (AVF). Use of an AVF for hemodialysis access is associated with lower mortality and morbidity as well as higher quality of life compared with access via an arteriovenous graft (AVG) or central venous catheter (CVC).

Of late, there has been debate over the issue of whether or when patients with chronic kidney disease (CKD) should be referred for an AVF. Factors informing the debate include AVFs not being immediately ready, the percentage of AVFs that fail to mature, the invasive nature of AVF creation compared with insertion of a CVC, and the possibility that the patient may die prior to actual need for hemodialysis. These issues are particularly relevant for elderly patients.

AVG placement can be performed closer to the time of initiation of hemodialysis; however, rates of long-term failure and need for intervention are higher for AVGs versus AVFs. There have been suggestions that factors specific to individual patients should be considered in the vascular access referral decision-making process, including remaining expected lifetime and quality of life. Costs of various referral policies are also a factor from a system perspective, in order to manage the healthcare budgets of agencies managing governmental insurance providers.

Researchers led by Steven M. Shechter, PhD, recently developed a simulation model based on available data on the vascular access decision process. The model was designed to be suited to capture the uncertainty and complexity present in managing the progression of CKD; costs and benefits associated with varying referral policies were also considered. Further, the researchers conducted an evaluation of the effectiveness of referral policies across age groups, with particular emphasis on the uncertainties specific to vascular access referral in elderly patients. The model was described in an article in the American Journal of Kidney Diseases [2017;70(3):368-376].

The interventions of interest were CVC only, AVF or AVG referral upon initiation of hemodialysis, AVF or AVG referral when hemodialysis is estimated to begin within 12 months for AVF or 3 months for AVG, and AVF or AVG referral when estimated glomerular filtration rate is <15 mL/min/1.73 m2 for AVF or <10 mL/min/1.73 m2 for AVG. The outcomes of interest were incremental cost-effectiveness ratios (ICERs) in 2014 US dollars per quality-adjusted life-year (QALY) gained.

Compared with using only a CVC, the ICER of AVF referral within 12 months of estimated initiation of hemodialysis and AVG referral within 3 months of estimated initiation of hemodialysis was ~$105 k/QALY and ~$101k/QALY, respectively, at a population level (costs of hemodialysis included). The cost-effectiveness of AVF or AVG referral decreased with advancing age, to ~$132 k/QALY for patients 80 to 90 years of age.

In sensitivity analyses, the results were more sensitive to the costs of erythropoietin, ongoing costs of hemodialysis, and patients’ utilities for hemodialysis. In analyses that excluded costs of hemodialysis, policies favoring AVF prior to hemodialysis initiation dominated both AVG prior to hemodialysis and policies calling for CVC only.

Limitations to the study cited by the authors included taking a payer perspective rather than a societal perspective, which did not account for patient or family lost wages that may initially be greater for AVF or AVG surgery versus CVC placement; including those costs would make AVF referral less cost-effective. In addition, the estimates for survival based on vascular access were obtained from publications based on observational data, which are likely to be confounded. Third, the lack of data on quality of life related to vascular access and to hemodialysis treatment itself had a major impact on the results. Finally, the study was conducted in a universal healthcare system, potentially making the results invalid in other models of healthcare delivery.

“In conclusion, the cost-effectiveness of vascular access referral is largely driven by access-based utilities and annual hemodialysis and erythropoietin costs and varies with age. Our results suggest that further research is needed in the field of vascular access and in hemodialysis-related quality of life to inform decision making regarding referral for vascular access creation,” the researchers said.

Takeaway Points

  1. Researchers developed a Monte Carlo data-driven simulation model of the process involved in decision making regarding referral for vascular access prior to hemodialysis. Three types of access were included: arteriovenous fistula (AVF), arteriovenous graft (AVG), and central venous catheter (CVC).
  2. The outcomes of interest were incremental cost-effectiveness ratios in 2014 US dollars per quality-adjusted life-year.

The cost-effectiveness of vascular access referral was driven primarily by the annual costs of hemodialysis, costs of erythropoietin, and access-specific utilities.