As chronic kidney disease (CKD) and end-stage renal disease (ESRD) are increasing in incidence worldwide, the importance of renal replacement therapies (RRT) increases. Modes of RRT include peritoneal dialysis, renal transplant, and hemodialysis (HD),1,9,10 with hemodialysis involving autogenous arteriovenous fistula (AVF), arteriovenous grafts (AVG) and central venous catheter (CVC). Based on the 2019 KDOQI Clinical Practice Guidance Statement, selecting an appropriate RRT modality for each patient has significant implications for improving their lifetime outcomes.1

Although AVF is the preferred modality for RRT, it is far from optimal. A cautious approach is necessary to avoid indiscriminate recommendations that may result in AVF failure and subsequent interventions.2,3 Guidelines recommended various strategies to prevent AVF failure and extend its lifespan, including appropriate pre-operative patient selection.3 We intend to evaluate whether AVF is the appropriate modality for the patient and whether the patient is suitable for AVF before surgery. Due to the high incidence of primary maturation failure in AVFs, previous studies have predominantly focused on investigating the AVF maturation, with relatively limited research conducted on patency, most of which has centered around postoperative parameters.4–8 One study examined the impact of postoperative measurements on primary patency and found that AVF diameter was the only predictor of patency.8 Another study on the prediction of AVF dialysis use by ultrasound examination after fistula creation concluded that a diameter above 5mm and an arterial end-diastolic velocity above 110cm/s were the best predictors of dialysis use.4 To our knowledge, there is a scarcity of research exploring how preoperative parameters influence the outcome of AVF.

Additionally, different literatures employed various terminologies to describe outcomes of AVF1,9 and three failure outcomes of AVF were considered: early thrombosis, failure to mature, and late failure.3 This study chose to assess the functional primary patency of RCAVFs following unassisted maturation without early thrombosis, with a particular focus on late failure. And it should be noted that AVF encompasses different types such as radiocephalic fistula (RCAVF), brachiocephalic fistula (BCAVF) and brachiobasilic fistula (BBAVF), etc., and some previous studies did not specify a particular type.4,5,8 Considering that RCAVF is the preferred method with the highest case volume in our hospital, this study will primarily focus on investigating it.

In conclusion, we aim to identify sensitive indicators capable of predicting the functional primary patency of RCAVFs, with the intention of enhancing preoperative modality selection and patient screening. Patients with a lower likehood of long-term functional patency could avoid unnecessary surgery and benefit from alternative methods of RRT.

A total of 257 patients with ESRD underwent successful placement of RCAVF from July 2017 to December 2019, with pre-operative ultrasound. Follow up for 6 months, 75 (29.2%) failed to mature for the first time. The final cohort included the remaining 182 patients (70.8%) with a RCAVF that matured without any assisted invention or surgery. The interval between the access placement and RCAVF maturation ranged from 1 to 6 months (mean 2.4 months). Among the 182 patients, failure was observed in 57 cases during the follow-up period, primarily attributed to thrombosis (n=2614.3%) or stenosis (n=31.17.0%). The functional primary patency of RCAVFs last on average 30.9 months (range, 2–66). At the end of study, 81 remained on RCAVFs, 30 had died, 11 had been lost to follow-up and 3 had received a kidney transplant.

The study population comprised mostly of males (65.9%) with a mean age of 58 years (range, 24–90). All AVFs were placed in the forearm, with 84% placed on the left side. Among the patients, 116 (64%) had diabetes and 155 (85%) had hypertension for an average duration of 9.8 and 9.9 years respectively. The demographic data is presented in Tables 1 and 2.

According to the cut-points determined by ROC analyses, the parameters were divided into 2 groups: > cut-point and ≦ cut-point groups. Eleven parameters show significant differences between groups, as summarized in Table 1. Among the aforementioned eleven parameters, there were significant differences observed in the functional primary patency rates of six parameters across different years. At 2 years, the group with a PSV of radial artery near the elbow ≤59cm/s exhibited a higher primary patency rate compared to >59cm/s group (78.4% vs 57.5%, P=0.026) (Fig. 2). In the >59cm/s group, there was a significantly higher probability of thrombosis or stenosis formation (45.8% patients, 22 out of 48) compared to the ≤59cm/s group (26.1%, 35 out of 134), with a statistically significant difference observed (P=0.012). Additional relevant data can be found in Table 3.

Fig. 2.

Kaplan–Meier curves of the PSV of the radial arteries near the elbow and the primary patency.

(0.08MB).

The univariate Cox proportional hazard models showed the parameters with a P-value of less than 0.1 were gender, the diameter of radial artery near the elbow, the PSV of radial artery near the elbow, and the diameter of brachial artery near the elbow (gender: HR=0.604, 95%CI 0.358–1.020, P=0.059, the diameter of radial artery near the elbow: HR=0.582, 95%CI 0.356–0.952, P=0.031, the PSV of radial artery near the elbow: HR=1.015, 95%CI 1.001–1.030, P=0.037, the diameter of brachial artery near the elbow: HR=0.678, 95%CI 0.442–1.039, P=0.074). All four parameters were included in the multivariable Cox regression analysis, and only the effect of the PSV of radial artery near the elbow had a statistically significant effect on RCAVF functional primary patency (HR=1.017, 95%CI 1.002–1.031, P=0.021). The other three parameters did not show statistical significance (P>0.05). The details can be found in Table 4.

Based on the findings of this study, the PSV of the radial artery near the elbow can be considered an independent risk factor of functional primary patency for RCAVFs. The hazard ratio is greater than 1, indicating a 1.7% increased risk of primary late-failure in RCAVF per unit increase in PSV. Using a cut-off value of 59cm/s for PSV in the radial artery near the elbow, evidence suggests that patients with ≤59cm/s have better overall primary patency rates compared to those with >59cm/s, particularly at 2 years where a statistically significant difference was observed. PSV of radial arteries near the elbow above 59cm/s could predict relatively shorter functional primary patency of RCAVF.

The Spanish Clinical Guidelines for Vascular Access in Hemodialysis10 propose that peak systolic velocity, as a flow characteristic at the arterial level, exhibits a certain degree of correlation in predicting future arteriovenous fistula (AVF) development. However, there is limited existing research on the assessment of peak systolic velocity of arteries. According to the literature available, only one earlier study briefly mentioned a contrary finding proposing that a PSV of radial artery at least 50cm/s could serve as a criterion for AVF placement.11 This criterion was also briefly mentioned in two reviews published in 201612 and 2022.13 The likely reason is that the minimum standard for PSV aims to optimize blood flow and reduce the likelihood of blood clots, which exerts a more pronounced influence on AVF creation and early thrombosis prevention.14 However, our study specifically focused on the post-maturation patency, which may account for the divergent findings.

A previous study proposed that systolic spikes and high pulsatility index (PI) were identified as risk factors for AVF failure, with a higher probability of failure observed in AVFs with PI greater than 1.4.15 It's probably because that an elevated systolic spikes may contribute to a more rapid increase and fluctuation in wall shear stress (WSS), which refers to the traction stress induced by velocity gradient on the luminal vessel surface.16 Andrea et al.16 suggested that unsteady and disturbed WSS patterns could stimulate endothelial cells (ECs), leading to intimal hyperplasia development and induction of a proinflammatory state, thereby promoting inward remodeling instead of outward remodeling of vessels, ultimately impeding AVF maturation and resulting in primary failure. Perhaps this is the underlying reason why the higher PSV of the radial artery hinders the suitability of utilizing RCAVF.

Notably, the measurement location in the previous studies was typically set near the wrist. A previous research had identified that the diameter of radial artery and cephalic vein (measured at about 3–6cm above the wrist) were risk factors for surgical failure.15 Another research concluded that the diameter of the artery was the sole critical factor associated with the maturation of AVF, in which the measurement point was established at a distance of 5cm above the wrist.17 The diameter near the wrist may have a more significant impact on surgical procedures and maturation, but it may not be associated with subsequent patency. As observed in our study, there was no correlation between the diameter near the wrist and functional primary patency in either multivariable or univariate analysis. Similarly, when compared to assessing PSV near the wrist, evaluating PSV near the elbow has potential for providing a more comprehensive evaluation of subsequent arterial supply and expansion. Regarding the radial artery diameter near the elbow, it only showed a significant correlation (P-value<0.05) in univariate Cox regression analysis, but no correlation was found in multivariable analysis. This discrepancy could be attributed to an indirect association formed by PSV with patency or possibly due to limited sample size. Unfortunately, there have been limited studies that have separately investigated the two positions, near the wrist and near the elbow. This dearth of research may partially explain why previous studies did not identify a significant correlation with PSV.

In addition to ultrasound parameters, this study comprehensively incorporates preoperative factors that may be relevant to AVF, including routine indicators such as patient demographics and comorbidities, as well as less commonly included factors like serum markers, aiming to provide a comprehensive assessment. Previous studies have suggested that serum markers like eGFR hold potential in reflecting the optimal timing for AVF creation.1,18 Additionally, there is literature indicating that Uremia significantly exacerbates inflammation and oxidative stress, both of which impact the formation of NIH and endothelial function, ultimately leading to a prothrombotic state.3 Furthermore, another study has also supported the influence of the uremic state in patients with ESRD on AVF outcomes, with higher levels of serum urea being associated with poorer patency at 6 weeks and inferior long-term outcomes in RCAVF.19 However, our study did not find a significant correlation between serum indicators and the functional primary patency of RCAVFs. Consistent with our findings, previous studies have also failed to establish associations between preoperative blood urea nitrogen, eGFR, and serum creatinine levels with maturation and functionality.17,20 A single or multiple serum indicators may not offer a comprehensive reflection of the overall patient condition and could be influenced by temporary dialysis before surgery.17

Conversely, arterial PSV might provide insights into the long-term preoperative state of patients. The presence of systemic abnormalities in ESRD patients has been reported to contribute to the development of accelerated atherosclerosis, vessel thickening, vascular calcification, and stiffness,3,21 which might lead to PSV of the artery elevated. A higher PSV likely indicates more severe pre-existing vascular pathology, which is closely associated with AVF maturation and patency.22 This perspective also could explain why a higher PSV negatively impacts the patency of RCAVF.

Naturally, as indicated in the Spanish Clinical Guidelines,10 it is not advisable to base vascular access decisions solely on one clinical or ultrasound parameter. It is recommended that a comprehensive evaluation of the patient's medical history, vascular physical examination, and preoperative ultrasound be conducted, taking into account individual preferences. Therefore, the decision to undergo arteriovenous fistula surgery should not rely solely on PSV of radial artery, but rather consider other factors in conjunction. A thorough assessment of various aspects of each patient's condition should be undertaken to tailor an optimal dialysis plan for them. The significance of this study's findings lies in alerting nephrologists to the potential detrimental effects excessive PSV may have on functional primary patency, as previous researches have yielded limited relevant results and primarily focused on studying diameter thresholds.

There are several limitations to our study as well. Firstly, 38 (20.9%) patients opted for transferring to another dialysis unit after maturation, their outcomes were assessed via telephone and WeChat. It should be noted that the inclusion of these patients may have potentially compromised the accuracy of the results. Secondly, the objective of this study is to evaluate the suitability of patients for creating RCAVF preoperatively, so we have exclusively included preoperative parameters while not addressing the impact of postoperative parameters on functional primary patency. However, it should be noted that postoperative parameters such as ultrasound results after surgery, puncture techniques, dialysis guidelines, prescribed flows of the pump, among others can all potentially influence the functional primary patency of RCAVFs. Therefore, we intend to incorporate these postoperative parameters in future studies. Additionally, since this study only focuses on RCAVF, it is possible that patients with more severe conditions and complications, who were unsuitable for creating RCAVF, have already been excluded. This may potentially affect the generalizability of our findings. In the future studies, we would include other different modalities of AVF to validate our results. Lastly, an increase in PSV can be expected due to decreased arterial compliance and pre-existing vascular disease associated with common risk factors such as advanced age, gender, hypertension, diabetes and presence of uremia. Although this study incorporates these variables in the analysis, it remains challenging to completely eliminate their impact on PSV of the arteries. However, PSV may serve as a more comprehensive indicator of patients’ overall physical condition, as previously discussed.

The aim of this study was to assess patients’ suitability for RCAVF by comprehensively investigating the impact of pre-operative parameters on the functional primary patency of unassisted matured RCAVF. This study concluded that the PSV of the radial artery near the elbow serve as a significant risk factor for functional primary patency of RCAVF, with PSV above 59cm/s potentially associated with reduced RCAVF longevity. The unexpected finding reminds us to not only assess artery velocity adequacy but also pay attention to excessive peak systolic velocity, although further multicenter studies are needed for evaluation.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that they have no conflicts of interest.