A review of the predictive methods for arteriovenous fistula (AVF) failure identification

The arteriovenous fistula (AVF) is the optimal form of vascular access for end-stage renal failure patients requiring haemodialysis. However, AVFs can fail to mature into a useable conduit or develop recurrent stenoses that result in high surgical re-intervention rates to maintain patency. Here, we performed a narrative review of literature on methods of AVF failure identification such as clinical prediction models, shear stress metrics characterisation with 0D, 1D and 3D computational models and blood flow sound frequency. Clinical risk factors identified by a clinical scoring system fail to predict ongoing AVF status and patency rate. The pulse wave propagation model, although showed positive prospect to support surgical planning, has limited ability in blood flow volume prediction in the anastomotic region. Shear stress metrics were one of the most common indicators to understand flow behaviours in AVF. While transverse wall shear stress seemed promising in characterising multidirectional flow in an AVF, no definitive correlations between shear stress metrics and disease initialisation have been reported. We conclude that further assessment is required to develop a clinical diagnostic marker for AVF failure prediction as no definitive prediction indicator was proposed even though there is an apparent impact of haemodynamic factors in AVF failure.