Challenges of Predicting Arteriovenous Access Survival Prior to Conversion from Catheter.

OBJECTIVE: The decision to convert from catheter to arteriovenous access is difficult yet very important. The ability to accurately predict fistula survival prior to surgery would significantly improve the decision making process. Many previously investigated demographic and clinical features have been associated with fistula failure. However, it is not conclusively understood how reliable predictions based on these parameters are on an individual level. The aim of this study was to investigate the probability of arteriovenous fistula maturation and survival after conversion using machine learning workflows. METHODS: A retrospective cohort study on multicentre data from a large North American dialysis organisation was conducted. The study population comprised 73 031 chronic in centre haemodialysis patients. The dataset included 49 variables including demographic and clinical features. Two distinct feature selection/prediction pipelines were used: LASSO regression and Boruta followed by a random forest classifier. Predictions were facilitated for re-conversion to catheter within one year. Additionally, all cause mortality predictions were conducted to serve as a comparator. RESULTS: In total, 38 151 (52.2%) had complete data and made up the main cohort. Sensitivity analyses were conducted in 67 421 patients (92.3%) after eliminating variables with a high proportion of missing data points. Selected features diverged between datasets and workflows. A previously failed arteriovenous access appears to be the most stable predictor for subsequent failure. Prediction of re-conversion based on the demographic and clinical information resulted in an area under the receiver operating characteristic curve (ROCAUC) between 0.541 and 0.571, whereas models predicting all cause mortality performed considerably better (ROCAUC 0.662 - 0.683). CONCLUSION: While group level depiction of major adverse outcomes after catheter to arteriovenous fistula/graft conversion is possible using the included variables, patient level predictions are associated with limited performance. Factors during and after fistula creation as well as biomolecular and genetic biomarkers might be more relevant predictors of fistula survival than baseline clinical conditions.

Cephalic arch restenosis rates in hemodialysis patients with brachiocephalic fistulae: a retrospective multicenter analysis of 3301 patients.

BACKGROUND: We evaluated restenosis rates at the cephalic arch after percutaneous angioplasty and stenting procedures in patients with brachial artery to cephalic vein arteriovenous fistula (BCAVF) hemodialysis access. METHODS: We used data from adult hemodialysis patients treated at a national network of 44 outpatient interventional facilities during Oct 2011-2015. We included data from patients with BCAVF who received an exclusive angioplasty, or stent with angioplasty, for treatment of cephalic arch stenosis and had ≥1 subsequent evaluation of the cephalic arch. Median percent restenosis per month at cephalic arch and days between encounters was calculated from the 1st index to 2nd procedure, and for up to 4 subsequent encounters. Analyses were stratified by intervention and device types. RESULTS: We identified a cohort of 3301 patients (mean age 62.2 ± 13.9 years, 58.5% male, 33.2% white race) with a BCAVF who had an angioplasty, or stent, at the cephalic arch for an index and ≥ 1 follow-up procedure. Between the 1st index to 2nd procedure, patients who received an angioplasty (n = 2663) or stent (n = 933) showed a median decrease of 18.9 and 16.5% in luminal diameter per month and a median time of 93 and 91 days between encounters, respectively. Restenosis and day rates were similar for standard versus high-pressure angioplasties. Bare metal stents showed 10.1 percentage point higher restenosis rate compared to stent grafts. Restenosis rates and time to restenosis were relatively consistent across subsequent encounters. CONCLUSIONS: Findings suggest hemodialysis patients with a BCAVF who require an angioplasty or stent to treat a stenosis at the cephalic arch will have stenosis reformed at a rate of 18.9 and 16.5% per month after the first intervention, respectively. Findings suggest patients are at risk of having significant lesions at the cephalic arch within 3 months after the previous intervention.

Artificial Intelligence Methods for Rapid Vascular Access Aneurysm Classification in Remote or In-Person Settings.

BACKGROUND: Innovations in artificial intelligence (AI) have proven to be effective contributors to high-quality health care. We examined the beneficial role AI can play in noninvasively grading vascular access aneurysms to reduce high-morbidity events, such as rupture, in ESRD patients on hemodialysis. METHODS: Our AI instrument noninvasively examines and grades aneurysms in both arteriovenous fistulas and arteriovenous grafts. Aneurysm stages were adjudicated by 3 vascular specialists, based on a grading system that focuses on actions that need to be taken. Our automatic classification of aneurysms builds on 2 components: (a) the use of smartphone technology to capture aneurysm appearance and (b) the analysis of these images using a cloud-based convolutional neural network (CNN). RESULTS: There was a high degree of correlation between our noninvasive AI instrument and the results of the adjudication by the vascular experts. Our results indicate that CNN can automatically classify aneurysms. We achieved a >90% classification accuracy in the validation images. CONCLUSION: This is the first quality improvement project to show that an AI instrument can reliably grade vascular access aneurysms in a noninvasive way, allowing rapid assessments to be made on patients who would otherwise be at risk for highly morbid events. Moreover, these AI-assisted assessments can be made without having to schedule separate appointments and potentially even via telehealth.

Tract dilation to salvage failing buttonholes in arteriovenous dialysis fistulae.

INTRODUCTION: Hemodialysis patients with an arteriovenous fistula can use buttonhole techniques for cannulation. Although buttonholes generally work well, patients may report difficult and painful cannulation, and buttonholes may fail over time. We aimed to assess the effectiveness of tract dilation in treatment of failing buttonholes. METHODS: We retrospectively analyzed data from patients treated with buttonhole tract dilation at an outpatient vascular access center between January 2013 and August 2015. RESULTS: Data from 23 patients were analyzed. There were 51 tract dilation procedures during 36 encounters for failing arteriovenous fistula buttonhole tract(s). The technical success rate for established tract dilation with "blunt-recanalization" was 90% (n = 46). The five remaining buttonholes had "sharp-recanalization" to create and dilate new tract through the buttonhole. For 46 buttonholes treated with "blunt-recanalization," there was an 85% clinical success rate at one week (39 buttonholes), and one was lost to follow-up; there was a 70% clinical success rate after one month (32 buttonholes). In the five buttonholes with "sharp-recanalization," there was only one clinical success with p < 0.05 for difference in success rate compared to "blunt-recanalization" at both one week and one month. There was one complication from "sharp-recanalization" requiring abandonment of the buttonhole tract. DISCUSSION: Buttonhole tract dilation is a useful method to treat difficult cannulation and painful cannulation and has the potential to extend the life of failing buttonholes.