Description of the condition
Haemodialysis, the most common treatment for kidney injury globally, requires reliable access to the bloodstream by means of a vascular access. The native arteriovenous (AV) fistula and the AV graft (AV accesses) are universally recognised (Allon 2008; Besarab 2008) as the best types of access because large observational studies have shown that their use is associated with better patient outcomes than central venous catheters (Dhingra 2001; Pastan 2002). This belief was reinforced by a recent meta-analysis by our group, in which we showed that people using catheters had the highest risk of death from all causes, infections and cardiovascular disease (Ravani 2013). AV accesses are internal accesses because they are inserted in the arm of a patient and consist of a direct surgical connection between an artery and a vein (fistula), or a bridge conduit between an artery and a vein (graft), usually made of synthetic material.
Description of the intervention
Remodelling of blood vessels after surgical creation of an AV access is an important part of the maturation and enlargement of the access (Roy-Chaudhury 2006) to facilitate flow sufficient for haemodialysis. However, scarring and abnormal remodelling may also contribute to vascular access stenosis, dysfunction and thrombosis, a leading cause of hospitalisation in haemodialysis patients (Manns 2005). While early complications within the first few months after access creation are more common in fistulas than in grafts (Dember 2008) late complications of an established and functioning access are more common in grafts, and include venous outflow, graft-vein, graft-artery and intra-graft lesions in grafts, and venous outflow and arterial inflow in fistulas (Roy-Chaudhury 2006). Additional procedures are performed in 80% of newly-created fistulas to salvage access function (1.45 to 3.3 procedures/access-year, three times higher than in grafts), with an overall cost to facilitate fistula maturation (CAD 7740 in the first year) similar to that of a successful graft creation (CAD 8130 in the first year) (Lok 2007). About half of the fistulas that mature without additional interventions (20% of all fistulas) require some revisions in the long-term (0.17 to 0.57 procedures/access-year) (Lok 2007). Since these rates are six times lower than in grafts in the long-term, fistulas are considered the most cost-effective access (USD 4862/year) (Besarab 2008) available for dialysis access.
How the intervention might work
In most haemodialysis programs, different methods are routinely used for early identification of a haemodynamically significant stenosis of a fistula or graft (NKF 2006). These include physical examination (monitoring) and different surveillance techniques, such as sequential measurements with trend analysis of intra-access flow measurements, measured or calculated static arterial and venous dialysis pressure, and/or duplex ultrasound. In the presence of a trend toward critical values of access flow or pressure, guidelines recommend patient referral for an imaging study and pre-emptive correction of stenoses greater than 50% to prevent further progression of vessel narrowing and thrombosis of the access (Jindal 2006; NKF 2006; Polkinghorne 2008; Tordoir 2007). These pre-emptive radiological intervention procedures (angioplasty with or without stenting) or surgical corrections of a stenosis are expected to result in increased longevity of the AV access, improved suitability for dialysis, increased dialysis dose delivery, and reduced subsequent access-related procedures and cost. However, the evidence in support of such practice is controversial (Tonelli 2008). Implementation of access screening strategies has been found to reduce the risk of thrombosis but not patency rates in fistulas, and to have no benefits in grafts (Tonelli 2008). In addition, information regarding access loss, costs, and hospitalisation rates following implementation of these screening strategies (a measure of potential harm) is insufficient.
Why it is important to do this review
Current guidelines recommend routine screening of permanent AV access (NKF 2006) but several trials and a systematic review (Tonelli 2008) have been published since these guidelines were last updated. Given the considerable morbidity associated with access dysfunction but also the important resource implications and uncertain benefits and harms of these screening strategies (Tonelli 2008) knowledge synthesis about optimal ways to improve outcomes of a functioning access remains a priority for patients, clinicians and health policy makers. In this systematic review we will summarise data from clinical trials on the benefits and harms of pre-emptive correction of AV access stenosis on the risk of major patient-centred endpoints.