Saudi Journal of Kidney Diseases and Transplantation (Jan 2019)

Prediction model for successful radiocephalic arteriovenous fistula creation in patients with diabetic nephropathy

  • J Suresh Kumar,
  • K S Sajeev Kumar,
  • E T Arun Thomas,
  • K G Hareesh,
  • Jacob George

DOI
https://doi.org/10.4103/1319-2442.270261
Journal volume & issue
Vol. 30, no. 5
pp. 1058 – 1064

Abstract

Read online

Arteriovenous fistula (AVF) is the most appropriate vascular access for all chronic kidney disease patients for hemodialysis. However, patients with diabetic nephropathy are at increased risk for primary failure after AVF creation, mainly due to atherosclerosis and calcification of blood vessels. We conducted this study to find out the risk factors for primary failure of radiocephalic AVF in end-stage renal disease (ESRD) patients due to diabetic nephropathy and develop a risk predicting model. This study was conducted at a tertiary care teaching hospital of South India. Patients with ESRD due to diabetic nephropathy whom underwent left radiocephalic AVF at wrist were enrolled. Risk factors for primary failure were analyzed by univariate and multivariate logistic regression models. Sixty-six patients were included in the study. Thirty-one patients had a primary failure. Independent risk factors for primary failure were palpable vessel wall of the radial artery (P = 0.003, odds ratio [OR] = 15.317), smaller radial artery diameter (P = 0.001, OR = 16.526), radial artery peak systolic velocity (PSV) <45 cm/s (P = 0.005, OR = 8.494), and linear radial artery calcification (P = 0.006, OR = 7.942). The risk predicting model obtained by adding the score given for each risk factors (vessel wall not palpable = 0, palpable = 1, no linear calcification in digital X-ray = 0, linear calcification = 1, PSV ≥45 cm/s = 0, <45 cm/s = 1 and 2.5 - radial artery diameter in mm) had an area under receiver-operating characteristic curve of 0.886. Cutoff score of 1.5 had sensitivity of 83.9% and specificity of 80.0% for primary failure. Risk predicting model for primary failure based on condition of the vessel wall on palpation, radial artery diameter, flow velocity, and calcification may be helpful for suitable patient selection.