Predicting Outcomes in Native AV Forearm Radio-cephalic Fistulae; the CAVeA 2 T 2 Scoring System

Risk prediction models for autogenous arteriovenous fistula failure in maintenance hemodialysis patients: A systematic review and meta-analysis

BACKGROUND

The aim of this study was to systematically retrieve and evaluate published risk prediction models for autogenous arteriovenous fistula (AVF) failure post-formation in maintenance hemodialysis (MHD) patients, with the goal of assisting healthcare providers in selecting or developing appropriate risk assessment tools and providing a reference for future research.

METHODS

A systematic search of relevant studies was conducted in PubMed, Web of Science, Cochrane Library, CINAHL, Embase, CNKI, Wanfang Database, VIP Database, and CBM Database up to February 1, 2024. Two researchers independently performed literature screening, data extraction, and methodological quality assessment using the Prediction Model Risk of bias (ROB) Assessment Tool.

RESULTS

A total of 4869 studies were identified, from which 25 studies with 28 prediction models were ultimately included. The incidence of autogenous AVF failure in MHD patients ranged from 3.9% to 39%. The most commonly used predictors were age, vein diameter, history of diabetes, AVF blood flow, and sex. The reported area under the curve (AUC) ranged from 0.61 to 0.911. All studies were found to have a high ROB, primarily due to inappropriate data sources and a lack of rigorous reporting in the analysis domain. The pooled AUC of five validation models was 0.80 (95% confidence interval: 0.79-0.81), indicating good predictive accuracy.

CONCLUSION

The included studies indicated that the predictive models for AVF failure post-formation in MHD patients are biased to some extent. Future research should focus on developing new models with larger sample sizes, strict adherence to reporting procedures, and external validation across multiple centers.

A systematic review of prediction models on arteriovenous fistula: Risk scores and machine learning approaches

OBJECTIVE

Failure-to-mature and early stenosis remains the Achille's heel of hemodialysis arteriovenous fistula (AVF) creation. The maturation and patency of an AVF can be influenced by a variety of demographic, comorbidity, and anatomical factors. This study aims to review the prediction models of AVF maturation and patency with various risk scores and machine learning models.

DATA SOURCES AND REVIEW METHODS

Literature search was performed on PubMed, Scopus, and Embase to identify eligible articles. The quality of the studies was assessed using the Prediction model Risk Of Bias ASsessment (PROBAST) Tool. The performance (discrimination and calibration) of the included studies were extracted.

RESULTS

Fourteen studies (seven studies used risk score approaches; seven studies used machine learning approaches) were included in the review. Among them, 12 studies were rated as high or unclear "risk of bias." Six studies were rated as high concern or unclear for "applicability." C-statistics (Model discrimination metric) was reported in five studies using risk score approach (0.70-0.886) and three utilized machine learning methods (0.80-0.85). Model calibration was reported in three studies. Failure-to-mature risk score developed by one of the studies has been externally validated in three different patient populations, however the model discrimination degraded significantly (C-statistics: 0.519-0.53).

CONCLUSION

The performance of existing predictive models for AVF maturation/patency is underreported. They showed satisfactory performance in their own study population. However, there was high risk of bias in methodology used to build some of the models. The reviewed models also lack external validation or had reduced performance in external cohort.

Pre-operative Patient Specific Flow Predictions to Improve Haemodialysis Arteriovenous Fistula Maturation (Shunt Simulation Study): A Randomised Controlled Trial

OBJECTIVE

An arteriovenous fistula (AVF) needs to mature before it becomes suitable to cannulate for haemodialysis treatment. Maturation importantly depends on the post-operative flow increase. Unfortunately, 20-40% of AVFs fail to mature (FTM). A patient specific computational model that predicts immediate post-operative flow was developed, and it was hypothesised that providing information from this model for planning of fistula creation might reduce FTM rates.

METHODS

A multicentre, randomised controlled trial in nine Dutch hospitals was conducted in which patients with renal failure who were referred for AVF creation, were recruited. Patients were randomly assigned (1:1) to the control or computer simulation group. Both groups underwent a work up, with physical and duplex ultrasonography (DUS) examination. In the simulation group the data from the DUS examination were used for model simulations, and based on the immediate post-operative flow prediction, the ideal AVF configuration was recommended. The primary endpoint was AVF maturation defined as an AVF flow ≥500 mL/min and a vein inner diameter of ≥4 mm six weeks post-operatively. The secondary endpoint was model performance (i.e. comparisons between measured and predicted flows, and (multivariable) regression analysis for maturation probability with accompanying area under the receiver operator characteristic curve [AUC]).

RESULTS

A total of 236 patients were randomly assigned (116 in the control and 120 in the simulation group), of whom 205 (100 and 105 respectively) were analysed for the primary endpoint. There was no difference in FTM rates between the groups (29% and 32% respectively). Immediate post-operative flow prediction had an OR of 1.15 (1.06-1.26; p < .001) per 100 mL/min for maturation, and the accompanying AUC was 0.67 (0.59-0.75).

CONCLUSION

Providing pre-operative patient specific flow simulations during surgical planning does not result in improved maturation rates. Further study is needed to improve the predictive power of these simulations in order to render the computational model an adjunct to surgical planning.

Development and validation of a clinical score to predict 1-year survival of arteriovenous fistula access: a diagnostic study

Purpose

To develop and validate a clinical scoring model to predict 1-year access survival among end-stage renal disease (ESRD) patients who received a new arteriovenous fistula (AVF).

Methods

The data of 195 ESRD patients in the development cohort who underwent first-time AVF creation between January 2009 and June 2013 and who had successful cannulation for dialysis use were reviewed. The clinical features that were significantly associated with 1-year AVF survival were incorporated into a clinical scoring model. The validity of this clinical score was then tested in a validation cohort of 204 ESRD patients who received a new AVF between July 2013 and December 2017.

Results

Of the 195 patients in the development cohort, 168 patients (86.2%) had a well-functioning AVF at 1 year. Absence of diabetes mellitus, no previous history of central venous catheter insertion, and absence of intervention performed to achieve access maturation were positively associated with 1-year AVF survival. These 3 factors were incorporated into a clinical scoring model, which ranged from 0 to 4 points. For a cutoff score of ≥3, the sensitivity, specificity and area under the receiver operating characteristic curve to predict 1-year AVF survival were 81.5%, 70.4%, and 0.760, respectively. The predictive performance of the clinical score was confirmed in the validation cohort, with a sensitivity of 76.1%, a specificity of 64.4% and an area under the curve of 0.703.

Conclusion

The scoring model using clinical data yielded acceptable performance in predicting 1-year access survival among patients receiving a new AVF.

Survival and Prognostic Predictors of Primary Arteriovenous Fistula for Hemodialysis

Objectives: This study aims to evaluate the survival and prognostic predictors of arteriovenous fistulas (AVFs) among patients undergoing AVF creation. The significant predictors were incorporated into a prognostic model to determine its prognostic performance for five-year AVF survival.

Materials and Methods: Data on 290 patients who underwent first-time AVF creation and who had been followed up for at least 5 years or until AVF failure were reviewed. The Kaplan–Meier survival curves and Cox proportional hazards model were generated to determine the AVF survival and associated prognostic predictors. Significant predictors were used to derive a prognostic model.

Results: The mean age of the patients was 59.7±14.6 years, and the 5-year AVF survival rate was 34.5%. Three features were found to be independent prognostic factors for the five-year AVF survival: old age, diabetes mellitus, and prior central venous catheter placement. These three significant factors were integrated into a prognostic scoring model that ranged from zero to five points. According to this model, the patients whose scores were 0, 1, 2, 3 and 4 or more had five-year AVF survival rates of 60.0%, 45.3%, 36.6%, 15.0%, and 2.9%, respectively.

Conclusion: The five-year survival rate of AVFs was modest, and a prognostic model could excellently estimate the five-year AVF survival.

Predicting Technical Success after Fistuloplasty: An Analysis of 176 Procedures

BACKGROUND

Significant stenoses in arteriovenous fistulae (AVFs) or arteriovenous grafts (AVGs) with limitation of flow and dialysis inadequacy should prompt consideration for fistuloplasty. We sought to identify fistulae, lesions, and patient-specific variables, which predict for outcomes after fistuloplasty.

METHODS

Data were extracted retrospectively from a renal access database from 2011 to 2016 of patients undergoing fistuloplasty. Demographics, comorbidities, outcomes of intervention, and flow rates documented on preintervention and postintervention duplex were collected. Secondary analysis of factors associated with postfistuloplasty flow rates of >600 mL/min, previously shown to be predictive of not requiring future intervention, was performed.

RESULTS

Of 204 attempted fistuloplasties, 176 were completed. One hundred forty (79.5%) were native AVFs and 34 (19.3%), AVGs (no data for 2). Median stenosis treated was 75%, with a majority (43.8%) in the proximal outflow vein. Flow rate on duplex after fistuloplasty was significantly better in AVFs (mean improvement 189.2 mL/min) than that in AVGs (mean improvement 51.8 mL/min; P = 0.034). Greatest flow improvement occurred for needling site stenotic lesions compared with other locations (from anastomosis to central vein) but was not significant. Brachio-brachial or brachio-axillary AVGs did significantly (P < 0.05) worse than all other fistulae types. The presence of hypertension was predicted for postfistuloplasty flow rate of >600 mL/min.

CONCLUSIONS

Flow rates after fistuloplasty vary depending on the type of fistula treated and the presence of hypertension. Knowledge of this can lead to better patient selection and counseling for fistuloplasty.

Clinical Utility of a New Predicting Score for Radiocephalic Arteriovenous Fistula Survival

BACKGROUND

Radiocephalic arteriovenous fistula (RC-AVF) is the recommended first choice for vascular access (VA). The CAVeA₂T₂ scoring system was recently published (ipsilateral central venous catheter access, age >73 years, vein <2.2 mm, lower limb angioplasty, and absent intraoperative thrill). The aim of the present study was to assess the clinical utility of the CAVeA₂T₂ scoring system for predicting RC-AVFs survival in our center and its subsequent application in VA management.

MATERIAL AND METHODS

In this single-center retrospective study, all RC-AVFs performed from January 2010 to July 2014 were included. The CAVeA₂T₂ was applied. Primary, assisted primary, and secondary patency rates were measured.

RESULTS

Sixty RC-AVFs were analyzed. Mean age was 64.3 ± 14.7 years. Mean CAVeA₂T₂ score was 1.23 ± 1.2. The median fistula secondary patency was 13.7 ± 1.6 months. Secondary patency was at 6 weeks and at 6, 12, and 24 months: 88.3%, 66.7%, 55%, and 31.7%, respectively. Increasing score (≥2) was associated with a decrease in primary (log-rank, χ² = 16.7, dif = 1, P = 0.0001) and secondary patency rate survival (log-rank, χ² = 5.4, dif = 1, P = 0.0001). In addition, stratification of the CAVeA₂T₂ score into 3 groups (scores 0-1, 2, and 3+) retained its significance for primary (log-rank, χ² = 19.4, dif = 2, P = 0.0001) and secondary patency rate survival (log-rank, χ² = 5.5, dif = 2, P = 0.046) at the end of the study.

CONCLUSIONS

In the present study, the CAVeA₂T₂ scoring system has proved to be a useful, easy to apply tool that is highly predictive of RC-AVF survival. Based on our results, we should avoid perform RC-AVFs, in those patients with CAVeA₂T₂ score ≥2 and late nephrology referral. Prospective studies should be designed to establish the management of patients with a higher CAVeA₂T₂ score.

Vascular access should be tailored to the patient

A cornerstone of hemodialysis treatment is the creation of a functional and durable dialysis vascular access. Every patient with chronic kidney disease should have a plan of renal replacement therapy and access site protection. Factors having a crucial impact on vascular access selection include age, comorbidity, vessel quality, prognosis, dialysis urgency, and surgeon's preferences. Our medical group have reviewed these factors in our patients and, based on recently published data, developed a clinical decision tree for dialysis access in the chronic kidney disease patient. Vascular access care should be patient-centered with the aim to maximize patient survival without loss of vascular access options; and not focused only the primary patency rates of dialysis access procedures.

The importance of success prediction in angioaccess surgery

Access to the circulation is an "Achilles' heel" of chronic hemodialysis. According to the current guidelines, autologous arteriovenous fistula is the best choice available. However, the impossibility of immediate use and the high rate of non-matured fistulas place fistula far from an ideal hemodialysis vascular access. The first attempt at constructing an angioaccess should result in functional access as much as possible. After failed attempts, patients and nephrologists lose their patience and confidence, which results in high percentage of central venous catheter use. Predictive models could help, but clinical judgment still remains crucial. Early referral to the nephrologist and vascular access surgeon, careful preoperative examinations, preparation of patients and duplex sonography mapping of the vessels are very important in the preoperative stage. In the operative stage, it is crucial to understand that angioaccess procedures should not be considered as minor procedures and these operations must be performed by surgeons with demonstrable interest and experience. In the postoperative stage, appropriate surveillance of the maturation process is also important, as well as good cannulation skills of the dialysis staff. The purpose of this review article is to stress the importance of success prediction in order to avoid unsuccessful attempts in angioaccess surgery.

Predictive models for arteriovenous fistula maturation

PURPOSE

Haemodialysis (HD) is a lifeline therapy for patients with end-stage renal disease (ESRD). A critical factor in the survival of renal dialysis patients is the surgical creation of vascular access, and international guidelines recommend arteriovenous fistulas (AVF) as the gold standard of vascular access for haemodialysis. Despite this, AVFs have been associated with high failure rates. Although risk factors for AVF failure have been identified, their utility for predicting AVF failure through predictive models remains unclear. The objectives of this review are to systematically and critically assess the methodology and reporting of studies developing prognostic predictive models for AVF outcomes and assess them for suitability in clinical practice.

METHODS

Electronic databases were searched for studies reporting prognostic predictive models for AVF outcomes. Dual review was conducted to identify studies that reported on the development or validation of a model constructed to predict AVF outcome following creation. Data were extracted on study characteristics, risk predictors, statistical methodology, model type, as well as validation process.

RESULTS

We included four different studies reporting five different predictive models. Parameters identified that were common to all scoring system were age and cardiovascular disease.

CONCLUSIONS

This review has found a small number of predictive models in vascular access. The disparity between each study limits the development of a unified predictive model.