What are the observed procedural costs of vascular access surgery? Protocol for a systematic review


  • This systematic review of the procedural costings of vascular access surgery will use standardised and reproducible methodology.

  • Findings from this review will indicate whether NHS reference costs can reliably be used as a marker of procedural cost in vascular access surgery for future economic analyses.

  • The analysis will focus on direct procedural costs, excluding preprocedural or postprocedural costs and non-medical expenses that may constitute a significant proportion of the cost to be considered in any future economic analysis.

  • Conclusions will be drawn from a UK perspective and any conclusions may not be directly applicable to international practice.


A central component in the introduction of a novel surgical procedure or technique is an evaluation of its cost efficiency when compared with a benchmark standard of care.1 Accurate assessment of costs is thus essential in ensuring appropriate allocation of resources within a healthcare system.2–4 Healthcare costs are a function of resources consumed and the unit costs associated with those resources.

The economics of costing are complex, and various methods exist with varying degrees of specificity at an individual level. ‘Gross-costing’ defines the cost of a resource at a highly aggregated level by bundling items for example, the average cost per hospital day, which reduces the workload of cost estimation whilst sacrificing specificity in the resulting cost estimate. ‘Microcosting’ involves the direct enumeration and costing of every input consumed in the treatment of a particular patient, producing a more specific estimate of cost but is a labour-intensive process. In addition, there are two approaches to determining the resource allocation of non-itemised aspects. The ‘top-down’ approach to costing assigns total costs for a healthcare system to individual units, for example, dividing the annual operating theatre budget by the number of procedures performed to estimate an average cost per procedure, which lacks specificity of procedural costs. The ‘bottom-up’ approach identifies the resource use for each individual patient, providing more specific cost estimates, but requires significant investigator effort and is dependent on similar procedures consuming similar resources.5 6 There is no standardised, universal method of health-economic analysis performed, with most approaches falling somewhere on a spectrum between ‘top-down gross-costing’ and ‘bottom-up microcosting’. For example, the National Health Service (NHS) in England employs a ‘top-down gross-costing’ methodology to estimate resource use per healthcare resource group. These costs represent the average cost per inpatient episode for bundled groups of conditions or procedures, such as ‘peripheral vascular disorders’ or ‘complex abdominal procedures’. Such generalisations preclude the use of these reference costs in making meaningful comparisons between different surgical procedures within the same procedure group.7

It is now recognised that a significant volume of healthcare resources required in the treatment of kidney failure is in providing vascular access provision, and this is the main modifiable cost.8 However, the costs of providing vascular access surgery are oversimplified and may not reflect the true cost of the service. For example, the two main surgical procedures (arteriovenous fistula (AVF) and arteriovenous graft (AVG)) are defined by the same reference cost (YQ42Z—Open Arteriovenous Fistula, Graft or Shunt Procedures), allowing no distinction between their costs.

The aim of this systematic review will be to assess the reporting of procedural costs in all published economic analyses of vascular access surgery, and a comparison of the reported procedural costs involved in AVF and AVG creation. This will provide an estimate as to the accuracy of the NHS reference costs in this field.

Methods and analysis

This systematic review will be reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.9 10


To conduct a systematic review of the procedural costs of AVF and AVG creation in the published literature.

Review questions

This systematic review will address the following research question: what are the reported procedural costs of vascular access surgery and how do they compare to the NHS reference costs?

Criteria for considering studies

Inclusion criteria

Costing studies published in English, with full text available, reporting procedural costs for AVF or AVG creation for the purpose of vascular access for haemodialysis were included.

Exclusion criteria

Studies which do not report procedural costs of vascular access creation or only report non-surgical costs were excluded. Studies reporting annualised costs per access type or the costs of dialysis provision without specific reporting of procedural costs were excluded as it is not possible to compare procedural costs in this context. Studies only reporting surgical revision or the cost of vascular access maintenance without reporting procedural costs of the index surgery were excluded.

Type of outcome

The outcomes of interest will be the procedural costs of AVF and AVG creation.

Information sources and search strategy

A systematic search will be performed on the MEDLINE, Embase and Cochrane databases to identify studies published in English from 1 January 2000 to 30 August 2023. The search strategy will include keywords to describe healthcare economic evaluations, combined with Medical Subject Heading terms relating to vascular access surgery for haemodialysis (online supplemental appendix). This will be supplemented by a manual search of reference lists from the identified studies, review articles and systematic reviews, and any relevant grey literature will be reviewed if referenced.

Supplemental material

Converting health service costs over space and time relies on several assumptions, for instance, that underlying technical or relative factor prices remain constant. The search window was selected to minimise potential inaccuracies arising from converting prices in earlier studies.11

NHS reference costs will be obtained from the most recent version of the National Schedule of NHS Costs and pertain to the total unit cost for currency code ‘YQ42Z—Open Arteriovenous Fistula, Graft or Shunt Procedures’.7

Selection of studies for inclusion

Titles and abstracts identified using the search strategy will be screened independently by two reviewers (BE/CJ) to identify appropriate studies for eligibility assessment. Reasons for exclusion of studies will be collected during abstract screening. Full-text articles of potentially suitable studies will be retrieved and independently assessed for eligibility by the same reviewers. Screening conflicts will be resolved by a third independent reviewer (DK).

Data collection and management

Included studies will be extracted for information relating to study design, country, price year, costing method and/or source of cost data, vascular access procedures studied, and cost per procedure. Data will be extracted into a standardised proforma using Microsoft Excel (V.16.76 2023, Microsoft).

When more than one variation of a procedure is reported, for example, brachiocephalic fistula and radiocephalic fistula, or when the procedure is compared between certain groups, for example, incident or prevalent haemodialysis patients, a procedural average will be calculated for the purposes of reporting costs.

When the price year is not reported, it will be assumed to be the same as year of publication. If procedural costing data have been adopted from government-derived costing schedules, the year of schedule publication will be used.

To allow meaningful comparison over time and geography, costs will be adjusted from original price year to target price year using a gross domestic product (GDP) deflator index and converted to US dollars using conversion rates based on purchasing power parities for GDP.12 The target year (2021) is selected to match the most recently published NHS reference costs.

Data analysis and quality assessment

Normality of data will be assessed using the Shapiro-Wilk’s test. Continuous variables will be analysed using the independent Students’ t-test or Mann-Whitney U test as appropriate, and multiset comparisons performed using ANOVA (Analysis of Variance) or Kruskal-Wallis tests. Continuous variables will be reported as mean with SD if normally distributed, or median with IQR in the case of non-normality. Data analysis will be performed using RStudio (V.2023.03.0+386 2022 Posit Software, PBC).

Due to variations in study designs (all study types are eligible) and the primary outcome of interest (direct procedural costs), no appropriate quality assurance checklist has been identified. Quality assurance will, therefore, be performed by the reviewers based on the following criteria:

  1. Is there a clear description of the procedure performed and does it fit the procedures eligible for inclusion?

  2. Is there a clear description of the cost definitions (procedural costs vs total hospital costs)?

  3. Is the study sample representative for the patient population studied?

Outcomes and prioritisation

The primary outcome of interest is the procedural costs of AVF and AVG creation in 2021 US dollars. Secondary outcomes include comparison with the NHS reference costs for these procedures, and whether reported costs are significantly different based on study design and country.

Patient and public involvement


Ethics and dissemination

Ethical approval is not required for this systematic review of peer-reviewed published literature. The study will be reported in accordance with PRISMA guidelines.9 10 Findings will be disseminated through peer-reviewed publications and conference presentations.


Although bottom-up microcosting is recognised as the ideal way to produce accurate costs at a patient-specific level, it is rarely done perhaps due to the time required to perform accurately. Consequently, top-down gross-costing methodology is often employed but this is less transparent and less reliable.13

There has been long-standing debate over the two main surgical methods of providing vascular access for haemodialysis: AVF and AVG. Since the inception of AVG as a novel technique, there has been greater emphasis on the assessment of outcomes rather than cost. Economic analysis has played a role in only one randomised trial in this field, which based procedural costs on NHS reference costs in the absence of a suitable alternative.14 Such an approach may allow a meaningful comparison of periprocedural care, but accurate costs of an intervention at the microlevel are required in order perform accurate economic analysis and provide a comprehensive view of outcomes alongside the cost at which they are achievable.8 15

This study will assess the methodology employed in performing health-economic evaluations of AVF and AVG in the literature and assess the prevalence of microcosting methodology within these studies. Comparison against the NHS reference costs for these procedures—the contemporary benchmark used to cost procedures in UK practice—will assess their reliability for use in future cost-efficiency analyses of vascular access modalities.

Several limitations to this review are recognised. The analysis will focus on direct procedural costs, rather than preprocedural or postprocedural costs and non-medical expense, which constitute a significant proportion of the cost to be considered in any future economic analysis. The objective, however, is to differentiate between procedural costs, as these appear more highly aggregated than those for periprocedural care. Second, by comparing against NHS reference costs, we approach this study from a UK perspective and any conclusions drawn may not be directly applicable to international practice. Third, there is potential for national tariffs used to cost procedures to be used as incentives to practitioners to target specific interventions, such as providing cost savings to centres achieving high rates of dialysis via native AVF. This may obscure the true procedural costs if cost data are sourced from national tariffs, as it may reflect a higher commissioner-to-institution reimbursement rather than a higher true procedural cost. Finally, we anticipate that in certain studies, costs (the true cost to the hospital) will be reported as charges, that is, the price paid by patients, government or third-party payers, and that the two are not directly interchangeable. However, without transparent data on the variability between costs and charges it is not possible to determine its impact on the results of this review.

Ethics statements

Patient consent for publication

This post was originally published on https://bmjopen.bmj.com