Background: Early identification of dysfunctional arteriovenous haemodialysis (HD) vascular access (VA) is important for timely referral and intervention. Method: We retrospectively calculated VA risk score using Vasc-Alert surveillance software technology from HD treatment sessions in 2 satellite HD units over 18 months. We included in the analysis HD patients dialysing with arteriovenous fistula or graft (AVF/G) with available Vasc-Alert data for≥ 2 months. For group one (eventful) that included patients who developed vascular access thrombosis or stenosis over the study period, we collected Vasc-Alert risk score 2 months prior to the event and, for group two (uneventful), over 5 consecutive months. Vasc-Alert technology utilises routinely collected data during HD to calculate VA risk score and triggers an alert if the score is ≥7 in 3 consecutive dialysis sessions. Patients with >2 alerts (vascular access score ≥7) per month were considered to have positive alerts. Results: From 140 HD patients, 81 patients dialysed via AVF/G. 77/81 had available Vasc-Alert data and were included in the final analysis. Out of 17 eventful patients, 11 (64.7%) had positive alerts 2 months prior to the vascular event. Out of the 60 patients without vascular events, 20 patients (33.3%) had positive alert. Vasc-Alert’s sensitivity and specificity for vascular events were 64.7% and 66.6%, respectively. Within the 6 patients with thrombosed access, 2 patients (33.3%) detected by Vasc-Alert were not detected with clinical monitoring. Conclusion: Vascular access risk score can be a useful non-invasive vascular access surveillance method to assist clinical decision making.

Vascular access dysfunction commonly stems from underlying stenosis or thrombosis in haemodialysis (HD) patients. The early identification of significant stenosis, coupled with timely referral for intervention, can improve VA longevity and reduce the incidence of acute thrombosis and the need for urgent intervention and/or temporary dialysis catheter placement to manage thrombotic complications. Recognising this importance, the National Kidney Foundation's Kidney Disease Outcomes Quality Initiative (KDOQI) advocates routine clinical assessments for clinically significant stenosis, administered by proficient and well-trained clinical staff [1]. However, establishing a rigorous and frequent clinical surveillance program often proves challenging due to limited medical and adequately trained nursing personnel.

Over the past years, numerous studies have explored non-invasive methods for predicting clinically significant stenosis, given that the gold standard for diagnosis, angiography, is both invasive and costly. Non-invasive tests generally fall under clinical monitoring and vascular access surveillance. Vascular access surveillance using non-invasive techniques is often challenging, requiring specialised equipment, trained staff, and additional resources to be deployed successfully.

Several surveillance techniques employing access flow measurements have been developed, primarily relying on HD needle reversal, thus necessitating additional time and dedicated staff resources. Access blood flow can be assessed using ultrasound technology, transcutaneous flow monitoring, thermodilution, glucose infusion, and conductivity. The ultrasound dilution (Transonic) method is probably the most used method for vascular access surveillance. It requires line reversal and injection of normal saline and calculates vascular access flow based on ultrasound wave velocity during this process with the aid of a bespoke software package. Low flow rates or significant change from baseline measurement indicate access dysfunction. Repeated measurements are required to improve the reliability of this method. Transonic evaluations should be performed monthly when used for VA surveillance purposes. The Duplex ultrasound method can provide flow and access to anatomy information [2‒6]. However, flow measurements are operator dependent and very sensitive to the angle of insonation [7‒9]. On the other hand, clinical monitoring involves readily available tests through physical examination or routine laboratory studies in the dialysis unit and remains the commonest standard practice [4].

Vasc-Alert, an FDA-approved software technology for dialysis access surveillance, computes a risk score for dysfunctional access using routinely collected intradialytic data, obviating the need for needle reversal or additional staff involvement. This software utilises readily accessible intradialytic data like blood pump flow to the dialyser, venous and negative arterial pressure, and mean arterial pressure to produce a risk score and has shown promising outcomes in a cohort of 6,163 HD patients in the USA [10].

To our knowledge, this technology has not been evaluated yet in the UK, where dialysis practice is characterised by longer dialysis sessions and lower blood flow rates [11] compared to the US approach [12]. To this end, we conducted a retrospective study aimed at assessing the predictive efficacy of the Vasc-Alert technology in detecting dysfunctional vascular access within two UK dialysis facilities.

Study Design: A Single-Centre Retrospective Observational Study

Study Population

We included all adult patients over 18 years, on maintenance HD patients (>3 months on HD), receiving their dialysis via a permanent VA arteriovenous fistula or graft (AVF/G) at 2 satellite dialysis units of our centre over the past 18 months. Patients are mainly prescribed thrice weekly 4-h HD therapy and occasional patients are prescribed 4-h HD therapy twice weekly. Exclusion criteria were HD via central venous catheter (CVC) (temporary or long-term lines), and hybrid dialysis access (simultaneous use of both arteriovenous access and CVC for dialysis). We collected from the electronic patient record demographic data (age, sex, body mass index), co-morbidities (diabetes mellites, hypertension, cardio-or cerebrovascular disorders [including history of ischaemic heart disease, heart failure, peripheral vascular disease, atherosclerosis, stroke, or transient ischaemic attacks], HD vintage, prior access stenosis [identified as previous access angioplasty]). We obtained information regarding clinical signs of access dysfunction, as per KDOQI guidelines (shown in Table 1), by reviewing the latest clinical letter authored by the patients’ designated consultant during their bimonthly evaluation, as well as the records from the most recent vascular access multidisciplinary (VA-MDT) meeting. The VA-MDT involves consultant nephrologists, vascular surgeons, vascular access specialist nurses, and dialysis specialist nurses and is held biweekly to discuss cases with challenging vascular access. Clinical concerns from attending staff are escalated to the VA-MDT.

Table 1.

Clinical indicators (signs and symptoms) suggesting underlying clinically significant lesions during access monitoring

ProcedureClinical indicators
Physical examination 
  • Ipsilateral extremity oedema

  • Alterations in the pulse, with a weak or resistant pulse, difficult to compress, in the area of stenosis

  • Abnormal thrill (weak and/or discontinuous) with only a systolic component in the region of stenosis

  • Abnormal bruit (high pitched with a systolic component in the area of stenosis)

  • Failure of the fistula to collapse when the arm is elevated (outflow stenosis) and lack of pulse augmentation (inflow stenosis)

  • Excessive collapse of the venous segment upon arm elevation

  • Aneurysm

 
Dialysis parameters 
  • New difficulty with cannulation when previously not a problem

  • Aspiration of clots

  • Inability to achieve the target dialysis blood flow

  • Prolonged bleeding beyond usual for that patient from the needle puncture sites for 3 consecutive dialysis sessions

  • Unexplained (>0.2 units) decrease in the delivered dialysis dose (Kt/V) on a constant dialysis prescription without prolongation of dialysis duration

 
ProcedureClinical indicators
Physical examination 
  • Ipsilateral extremity oedema

  • Alterations in the pulse, with a weak or resistant pulse, difficult to compress, in the area of stenosis

  • Abnormal thrill (weak and/or discontinuous) with only a systolic component in the region of stenosis

  • Abnormal bruit (high pitched with a systolic component in the area of stenosis)

  • Failure of the fistula to collapse when the arm is elevated (outflow stenosis) and lack of pulse augmentation (inflow stenosis)

  • Excessive collapse of the venous segment upon arm elevation

  • Aneurysm

 
Dialysis parameters 
  • New difficulty with cannulation when previously not a problem

  • Aspiration of clots

  • Inability to achieve the target dialysis blood flow

  • Prolonged bleeding beyond usual for that patient from the needle puncture sites for 3 consecutive dialysis sessions

  • Unexplained (>0.2 units) decrease in the delivered dialysis dose (Kt/V) on a constant dialysis prescription without prolongation of dialysis duration

 

Trust VA Monitoring and Surveillance Protocol

Patients receive a medical review every 2–3 months. Monitoring of vascular access is based on clinical examination and dialysis access parameters followed by discussion in multidisciplinary meetings as needed. Our unit does not routinely use direct or indirect vascular access surveillance tools. A monthly vascular access multidisciplinary team meeting (VA-MDT) occurs in each satellite dialysis unit to discuss concerns from routine clinical monitoring and arrange fistulogram if indicated based on KDOQI criteria [1]. Consultants can refer directly for fistulogram if they deem appropriate without prior discussion at the VA-MDT. Complex cases are discussed at separate MDT with the vascular access surgeons.

Study Groups

For the purposes of the analysis, the total population was divided in 2 groups based on the presence of significant vascular access events (thrombosis, angiographic stenosis requiring angioplasty or doppler with >50% stenosis) during the observational period (Group 1 – eventful) and (Group 2 – uneventful).

Vasc-Alert Technology

The parameters calculated by Vasc-Alert technology and scoring algorithm have been described in detail previously [10]. In brief, the software uses pressure and flow readings at 30 min intervals automatically retrieved during each session adjusted for haematocrit and systemic blood pressure to calculate the venous access pressure ratio (VAPR) and the arterial access pressure ratio (AAPR), metrics of outflow and inflow status (shown in Fig. 1). The Vasc-Alert scoring system is a risk stratification algorithm based on Vasc-Alert vascular access surveillance parameters, that ranges from 1 to 10 with values 8–10 associated with almost threefold increase in vascular access stenosis for AVFs and a cut-off point of ≥7 has reported sensitivity and specificity of 38% and 78% for AVF and 31% and 79% for AVG based only on a single treatment. The automatically collected and stored data in the 5008 Fresenius dialysis platform were transferred in pseudonymized fashion to a secure server for the application of the algorithm and the results were returned to the research team to be linked with the clinical database. An alert was triggered if vascular access score was ≥7 in 3 consecutive dialysis sessions.

Fig. 1.

Metrics for VA risk score. VAPR, venous access pressure ratio; AAPR, arterial access pressure ratio; BFR, blood flow rate.

Fig. 1.

Metrics for VA risk score. VAPR, venous access pressure ratio; AAPR, arterial access pressure ratio; BFR, blood flow rate.

Close modal

Definition of Positive and Negative Alerts

For group one (eventful), we collected the monthly access alerts 2 months prior to the event and, for group two (uneventful) over 5 consecutive months, allowing for 2 additional months of observation period for vascular events after the last Vasc-Alert data collection. Patients with >2 alerts (vascular access score ≥7) per month were considered to have positive alerts.

Statistical Analysis

The Statistical Package for Social Sciences (IBM SPSS Statistics), version 26 for Windows (IBM Corp., Armonk, NY, USA) was used for data analysis licenced to the University of Manchester. Categorical variables are presented as counts and percentages and continuous variables are presented as mean ± standard deviation (SD) or median and interquartile range depending for normally and not normally distributed data, respectively. Comparisons were performed using χ2 test. The distribution of continuous numerical variables was assessed with Shapiro-Wilk test. Numerical variables following the normal distribution were summarised as mean and standard deviation (SD). Comparisons of parametric and non-parametric data were performed with independent samples t test and Mann-Witney U test, respectively. Sensitivity of Vasc-Alert access risk score was calculated as the number of true positive (for event) over the true positive plus false negative (for alerts), specificity was calculated as the number of true negative (for event) over the true negative plus false positive (for alerts), positive predictive value was calculated as the number of true positive (for events) over the total number of positive alerts, and negative predictive value was calculated as the number of true negative (for events) over the total number of negative alerts. Values were expressed in numbers and percentages. This study was registered with the Northern Care Alliance Research and Innovation department (Ref: ID 21HIP02).

Out of 140 HD patients, 79 patients dialysed via AVF, 2 via AVG, 58 patients via CVC, and 1 patient was dialysing via AVF and CVC simultaneously. Among the 81 patients with AVF/G, Vasc-Alert metrics were unretrievable in 4 uneventful patients. The final analysis included 77 patients. Within this cohort, 17 patients were classified in the eventful group (11 with vascular access stenosis and 6 with vascular access thrombosis). Within this group 11 (64.7%) patients had positive alerts. The uneventful group included 60 patients, among them 20 (33.3%) patients had positive alerts (shown in Fig. 2). The Vasc-Alert sensitivity to predict VA events was 64.7%, specificity was 66.6%, positive predictive value 35.4%, and negative predictive value 86.9% (shown in Table 2).

Fig. 2.

Flowchart of the study population. HD, haemodialysis; AVF, arteriovenous fistula; AVG, arteriovenous graft; CVC, central venous catheter.

Fig. 2.

Flowchart of the study population. HD, haemodialysis; AVF, arteriovenous fistula; AVG, arteriovenous graft; CVC, central venous catheter.

Close modal
Table 2.

Vascular events and Vasc-Alert alerts

VASC-Alert resultsVascular eventsTotal
positivenegative
Positive 11 20 31 
Negative 40 46 
Total 17 60 77 
VASC-Alert resultsVascular eventsTotal
positivenegative
Positive 11 20 31 
Negative 40 46 
Total 17 60 77 

Sensitivity 11/17 (64.7%).

Specificity 40/60 (66.6%).

Positive predictive value 11/31 (35.4%).

Negative predictive value 40/46 (86.9%).

There was no statistically significant difference between the eventful and uneventful groups in age, sex, body mass index, co-morbid conditions (diabetes mellites, hypertension, CVD), clinical signs of access dysfunction, HD vintage, site of VA anastomosis, or the presence of antiplatelet therapy. History of prior access stenosis was significantly higher in the eventful group compared to the uneventful group (76.5% vs. 27.1%, p < 0.001). Vasc-Alert high access risk scores were more frequently observed in the eventful group compared to the uneventful group (64.7% vs. 33.3%, p = 0.020). Within the 6 patients with thrombosed access, 2 patients (33.3%) detected by the Vasc-Alert technology were not detected with clinical monitoring (shown in Table 3). Table 4 shows a comparative analysis for patients with and without alerts. In the group with alerts, there were more patients with brachiocephalic fistulas. Examples of software Vasc-Alert output for patients uneventful, eventful with successful angioplasty and eventful with recurrent stenosis following angioplasty are presented in Figure 3a–c, respectively.

Table 3.

Patient’s characteristics and vascular events

EventsStatistical tests
total (N = 77)positive (N = 17)negative (N = 60)p value
Age1 Mean±SD 58.7±15.4 63.1±14.1 57.4±15.6 0.185 
Gender2, n (%) Male 54 (70.1) 12 (70.6) 42 (70.0) 0.963 
Female 23 (29.9) 5 (29.4) 18 (30.0) 
BMI1 Mean±SD 26.5±5.4 27.1±5.1 26.3±5.5 0.586 
HTN3, n (%) Yes 74 (96.1) 17 (100.0) 57 (95.0) 1.000 
DM2, n (%) Yes 26 (33.8) 7 (41.2) 19 (31.7) 0.464 
CVS2, n (%) Yes 37 (48.1) 9 (52.9) 28 (46.7) 0.648 
Prior stenosis2, n (%) Yes 29 (38.2) 13 (76.5) 16 (27.1) <0.001* 
Clinical2, n (%) Yes 26 (33.8) 9 (52.9) 17 (28.3) 0.058 
Alerts2, n (%) Yes 31 (40.3) 11 (64.7) 20 (33.3) 0.020* 
HD vintage4 Median [IQR] 3.0 [2.0–6.0] 4.0 [3.0–7.0] 3.0 [2.0–5.0] 0.390 
Antiplatelet 2, n (%) Yes 32 (41.5) 6 (7.7) 26 (33.7) 0.352 
Anticoagulation Yes NA 
Site of anastomosis3 N (%) 77 (100) 17 (22) 60 (78) 0.088 
 Brachiocephalic 39 (50.6) 9 (11.6) 30 (39) 
 Brachiobasilic 6 (7.8) 3 (3.8) 3 (3.9) 
 Brachioaxillary 1 (1.2) 1 (1.2) 0 (0) 
 Radiocephalic 30 (38.9) 4 (5.2) 26 (33.7) 
 Radioaxillary 1 (1.2) 0 (0) 1 (1.2) 
EventsStatistical tests
total (N = 77)positive (N = 17)negative (N = 60)p value
Age1 Mean±SD 58.7±15.4 63.1±14.1 57.4±15.6 0.185 
Gender2, n (%) Male 54 (70.1) 12 (70.6) 42 (70.0) 0.963 
Female 23 (29.9) 5 (29.4) 18 (30.0) 
BMI1 Mean±SD 26.5±5.4 27.1±5.1 26.3±5.5 0.586 
HTN3, n (%) Yes 74 (96.1) 17 (100.0) 57 (95.0) 1.000 
DM2, n (%) Yes 26 (33.8) 7 (41.2) 19 (31.7) 0.464 
CVS2, n (%) Yes 37 (48.1) 9 (52.9) 28 (46.7) 0.648 
Prior stenosis2, n (%) Yes 29 (38.2) 13 (76.5) 16 (27.1) <0.001* 
Clinical2, n (%) Yes 26 (33.8) 9 (52.9) 17 (28.3) 0.058 
Alerts2, n (%) Yes 31 (40.3) 11 (64.7) 20 (33.3) 0.020* 
HD vintage4 Median [IQR] 3.0 [2.0–6.0] 4.0 [3.0–7.0] 3.0 [2.0–5.0] 0.390 
Antiplatelet 2, n (%) Yes 32 (41.5) 6 (7.7) 26 (33.7) 0.352 
Anticoagulation Yes NA 
Site of anastomosis3 N (%) 77 (100) 17 (22) 60 (78) 0.088 
 Brachiocephalic 39 (50.6) 9 (11.6) 30 (39) 
 Brachiobasilic 6 (7.8) 3 (3.8) 3 (3.9) 
 Brachioaxillary 1 (1.2) 1 (1.2) 0 (0) 
 Radiocephalic 30 (38.9) 4 (5.2) 26 (33.7) 
 Radioaxillary 1 (1.2) 0 (0) 1 (1.2) 

1Independent samples t test.

2Pearson’s χ2 test for independence of observations.

3Fisher’s exact test.

4Mann-Whitney test.

*Significant p value <0.05.

SD, standard deviation; BMI, body mass index; HTN, hypertension; DM, diabetes mellitus; CVS, cardio-cerebrovascular disease; IQR, interquartile range (25th–75th percentile); NA, non-applicable.

Table 4.

Patient’s characteristics and Vasc-Alert

Vasc-AlertStatistical tests
total (N = 77)positive (N = 31)negative (N = 46)p value
Age1 Mean±SD 58.7±15.4 59.4±15.9 58.2±15.2 0.730 
Gender2 Male 54 (70.1) 22 (71.0) 32 (69.6) 0.895 
Female 23 (29.9) 9 (29.0) 14 (30.4) 
BMI1 Mean±SD 26.5±5.4 26.2±5.2 26.7±5.5 0.712 
HTN3 Yes 74 (96.1) 31 (100.0) 43 (93.5) 0.269 
DM2 Yes 26 (33.8) 11 (35.5) 15 (32.6) 0.794 
CVS2 Yes 37 (48.1) 17 (54.8) 20 (43.5) 0.328 
Prior stenosis2 Yes 29 (38.2) 15 (48.4) 14 (31.1) 0.128 
Clinical2 Yes 26 (33.8) 10 (32.3) 16 (34.8) 0.818 
Events2 Yes 17 (22.1) 11 (35.5) 6 (13.0) 0.020* 
HD vintage4 Median [IQR] 3.0 [2.0–6.0] 4.0 [3.0–7.0] 3.0 [2.0–5.0] 0.036* 
Antiplatelet2 Yes 32 (41.5) 12 (15.5) 20 (25.9) 0.173 
Anticoagulation Yes NA 
Site of anastomosis3 N (%) 77 (100) 31 (40.2) 46 (59.7) <0.0001* 
 Brachiocephalic 39 (50.6) 21 (27.2) 18 (23.4) 
 Brachiobasilic 6 (7.8) 5 (6.5) 1 (1.2) 
 Brachioaxillary 1 (1.2) 1 (1.2) 0 (0) 
 Radiocephalic 30 (38.9) 4 (5.2) 26 (33.7) 
 Radioaxillary 1 (1.2) 0 (0) 1 (1.2) 
Vasc-AlertStatistical tests
total (N = 77)positive (N = 31)negative (N = 46)p value
Age1 Mean±SD 58.7±15.4 59.4±15.9 58.2±15.2 0.730 
Gender2 Male 54 (70.1) 22 (71.0) 32 (69.6) 0.895 
Female 23 (29.9) 9 (29.0) 14 (30.4) 
BMI1 Mean±SD 26.5±5.4 26.2±5.2 26.7±5.5 0.712 
HTN3 Yes 74 (96.1) 31 (100.0) 43 (93.5) 0.269 
DM2 Yes 26 (33.8) 11 (35.5) 15 (32.6) 0.794 
CVS2 Yes 37 (48.1) 17 (54.8) 20 (43.5) 0.328 
Prior stenosis2 Yes 29 (38.2) 15 (48.4) 14 (31.1) 0.128 
Clinical2 Yes 26 (33.8) 10 (32.3) 16 (34.8) 0.818 
Events2 Yes 17 (22.1) 11 (35.5) 6 (13.0) 0.020* 
HD vintage4 Median [IQR] 3.0 [2.0–6.0] 4.0 [3.0–7.0] 3.0 [2.0–5.0] 0.036* 
Antiplatelet2 Yes 32 (41.5) 12 (15.5) 20 (25.9) 0.173 
Anticoagulation Yes NA 
Site of anastomosis3 N (%) 77 (100) 31 (40.2) 46 (59.7) <0.0001* 
 Brachiocephalic 39 (50.6) 21 (27.2) 18 (23.4) 
 Brachiobasilic 6 (7.8) 5 (6.5) 1 (1.2) 
 Brachioaxillary 1 (1.2) 1 (1.2) 0 (0) 
 Radiocephalic 30 (38.9) 4 (5.2) 26 (33.7) 
 Radioaxillary 1 (1.2) 0 (0) 1 (1.2) 

1Independent samples t test.

2Pearson’s χ2 test for independence of observations.

3Fisher’s exact test.

4Mann-Whitney test.

*Significant p value <0.05.

SD, standard deviation; BMI, body mass index; HTN, hypertension; DM, diabetes mellitus; CVS, cardio-cerebrovascular disease; IQR, interquartile range (25th–75th percentile); NA, non-applicable.

Fig. 3.

Examples of the software Vasc-Alert output. a Patient with uneventful AVF had graph showing the average VAPR below threshold (0.6) (blue line). No alerts were generated as the access risk score was below 7. b Patient with stenotic AVF had successful angioplasty in March 2023. Graph shows the average VAPR was above the threshold and after treatment of stenosis it returns to below the threshold (blue line). Continuous red lines indicate the alert generated when the vascular access score was ≥7 in 3 consecutive dialysis sessions. c Patient with stenotic AVF had angioplasty with residual stenosis in April 2023, with progressive restenosis afterwards. Graph shows the average VAPR was above the threshold, did not get below the threshold level after angioplasty as there was residual stenosis and increased with time with progressive restenosis (blue line). Continuous red lines indicate the alert generated when the vascular access score was ≥7 in 3 consecutive dialysis sessions.

Fig. 3.

Examples of the software Vasc-Alert output. a Patient with uneventful AVF had graph showing the average VAPR below threshold (0.6) (blue line). No alerts were generated as the access risk score was below 7. b Patient with stenotic AVF had successful angioplasty in March 2023. Graph shows the average VAPR was above the threshold and after treatment of stenosis it returns to below the threshold (blue line). Continuous red lines indicate the alert generated when the vascular access score was ≥7 in 3 consecutive dialysis sessions. c Patient with stenotic AVF had angioplasty with residual stenosis in April 2023, with progressive restenosis afterwards. Graph shows the average VAPR was above the threshold, did not get below the threshold level after angioplasty as there was residual stenosis and increased with time with progressive restenosis (blue line). Continuous red lines indicate the alert generated when the vascular access score was ≥7 in 3 consecutive dialysis sessions.

Close modal

This study shows that Vasc-Alert with positive alerts (i.e., access risk score ≥7 in 3 consecutive treatments) occurring more than twice monthly is associated with vascular access events such as thrombosis or stenosis in a cohort of UK HD patients. Our results are in line with the recently published study by Astor et al. [10] in US population. The authors described different sensitivity and specificity of the risk score depending on the type of access (AVF vs. AVG) [10]. In our cohort, there were only 2 patients dialysing via AVG and we did not perform subgroup analysis. We report a sensitivity and specificity of 64.7% and 66.6%, respectively, using a Vasc-Alert risk score ≥7 in a population predominantly using AVFs. Astor et al. [10] employed the same risk score cut-off using a single Vasc-Alert measurement and reported sensitivity and specificity of 38% and 78% for AVF. In addition to differences in the dialysis practices and the different population size of the 2 studies, it could be speculated that the requirement of >2 positive alerts in our study may have resulted in increased sensitivity in our cohort.

Within the 6 cases with subsequent thrombosis, the Vasc-Alert access risk score detected 2 cases (33.3%) that were not detected by routine clinical monitoring. Patients with thrombosed access are likely to suffer from missed dialysis therapy, need urgent intervention to restore VA patency and may require temporary line insertion for dialysis therapy if the intervention is unsuccessful or not delivered in a timely manner [1]. Identifying patients at risk for thrombosis can expedite intervention and reduce avoidable morbidity and hospitalisation in these patients. Despite the small number of cases, our finding suggests that this technology may have a role in assisting clinical surveillance pathways to avoid acute thrombotic complications.

Our retrospective study showed that Vasc-Alert technology software is a useful non-invasive tool that identified 64.7% of eventful vascular access cases whereas clinical examination detected 52.9%. Kumbar et al. [13] in a prospective study of 38 HD patients reported that high vascular access risk scores correctly identified 81% with vascular access stenosis (defined as >50% on ultrasonography or angiography) and clinical evaluation accurately detected 69% of cases.

The study has several limitations. It is a single-centre retrospective study with a relatively small number of patients and short follow up. There were only 2 patients dialysing via AVG hence we could not analyse this group separately.

In conclusion, Vasc-Alert access risk score can be a useful non-invasive vascular access surveillance method to assist clinical decision making. It is expected that incorporation of this risk stratification algorithm in the clinical vascular access surveillance pathway is likely to increase the frequency of angiograms and angioplasties. Large prospective studies are required to clarify if pre-emptive interventions will decrease the likelihood of VA access thrombosis or failure.

This is a retrospective observational study using routinely collected data in an anonymised fashion and as indicated by the NHS Health Research Authority online tool http://www.hra-decisiontools.org.uk/research does not require research Ethics Committee review. The study protocol was approved and registered (Ref: ID 21HIP02) by the Research and Innovation committee of the Northern Care Alliance NHS Group, and the need for individual patient consent was waived.

The authors have no conflict of interest.

No external funding was sought for this study.

Alshymaa Eltahan performed the data curation and analysis, writing the original draft, review, and editing. Dimitrios Poulikakos performed conceptualisation, study design, data analysis, review, and editing. Rosemary Donne, David Lewis, Maharajan Raman, Zulfikar Pondor, Paul Hinchliffe, and Jan Cowperthwaite contributed to data reviewing and editing.

Data will be available upon request.

1.
Lok
CE
,
Huber
TS
,
Lee
T
,
Shenoy
S
,
Yevzlin
AS
,
Abreo
K
, et al
.
KDOQI Clinical Practice Guideline for Vascular Access: 2019 Update
.
Am J Kidney Dis
.
2020
;
75
(
4 Suppl 2
):
S1
64
.
2.
Lopot
F
,
Nejedlý
B
,
Sulková
S
,
Bláha
J
.
Comparison of different techniques of hemodialysis vascular access flow evaluation
.
Int J Artif Organs
.
2003
;
26
(
12
):
1056
63
.
3.
Moist
LM
,
Churchill
DN
,
House
AA
,
Millward
SF
,
Elliott
JE
,
Kribs
SW
, et al
.
Regular monitoring of access flow compared with monitoring of venous pressure fails to improve graft survival
.
J Am Soc Nephrol
.
2003
;
14
(
10
):
2645
53
.
4.
Allon
M
,
Robbin
ML
.
Hemodialysis vascular access monitoring: current concepts
.
Hemodial Int
.
2009
;
13
(
2
):
153
62
.
5.
Ram
SJ
,
Work
J
,
Caldito
GC
,
Eason
JM
,
Pervez
A
,
Paulson
WD
.
A randomized controlled trial of blood flow and stenosis surveillance of hemodialysis grafts
.
Kidney Int
.
2003
;
64
(
1
):
272
80
.
6.
Koirala
N
,
Anvari
E
,
McLennan
G
.
Monitoring and surveillance of hemodialysis access
.
Semin Intervent Radiol
.
2016
;
33
(
1
):
25
30
.
7.
Bay
WH
,
Henry
ML
,
Lazarus
JM
,
Lew
NL
,
Ling
J
,
Lowrie
EG
.
Predicting hemodialysis access failure with color flow Doppler ultrasound
.
Am J Nephrol
.
1998
;
18
(
4
):
296
304
.
8.
Schwarz
C
,
Mitterbauer
C
,
Boczula
M
,
Maca
T
,
Funovics
M
,
Heinze
G
, et al
.
Flow monitoring: performance characteristics of ultrasound dilution versus color Doppler ultrasound compared with fistulography
.
Am J Kidney Dis
.
2003
;
42
(
3
):
539
45
.
9.
Wiese
P
,
Nonnast-Daniel
B
.
Colour Doppler ultrasound in dialysis access
.
Nephrol Dial Transplant
.
2004
;
19
(
8
):
1956
63
.
10.
Astor
BC
,
Hirschman
K
,
Kennedy
J
,
Frinak
S
,
Besarab
A
.
Development and validation of a risk score to prioritize patients for evaluation of access stenosis
.
Semin Dial
.
2022
;
35
(
3
):
236
44
.
11.
Ashby
D
,
Borman
N
,
Burton
J
,
Corbett
R
,
Davenport
A
,
Farrington
K
, et al
.
Renal association clinical practice guideline on haemodialysis
.
BMC Nephrol
.
2019
;
20
(
1
):
379
.
12.
Tentori
F
,
Zhang
J
,
Li
Y
,
Karaboyas
A
,
Kerr
P
,
Saran
R
, et al
.
Longer dialysis session length is associated with better intermediate outcomes and survival among patients on in-center three times per week hemodialysis: results from the Dialysis Outcomes and Practice Patterns Study (DOPPS)
.
Nephrol Dial Transplant
.
2012
;
27
(
11
):
4180
8
.
13.
Kumbar
L
,
Astor
BC
,
Besarab
A
,
Provenzano
R
,
Yee
J
.
Association of risk stratification score with dialysis vascular access stenosis
.
J Vasc Access
.
2024
;
25
(
3
):
826
33
.