Predictors of early arteriovenous fistula failure in patients with end stage renal disease on hemodialysis: a systematic review and meta-analysis

Abstract

Background

Hemodialysis is the most prevalent modality of renal replacement therapy. The durability of hemodialysis and its quality depend on consistent and reliable access to the patient’s vascular system. In this systematic review we provide a comprehensive analysis of the predictors of primary arteriovenous fistula failure.

Methods

A systematic search was conducted in PubMed, Web of Science, ScienceDirect, and the WHO Virtual Health Library Regional Portal from database inception through March 2025. Search terms included combinations of ‘arteriovenous fistula’, ‘failure’, ‘hemodialysis’, ‘end-stage renal disease’, and ‘risk factors’. Four reviewers independently screened titles and abstracts, with full texts assessed according to predefined inclusion criteria. Studies were eligible if they examined risk factors for primary arteriovenous fistula failure in adult patients on hemodialysis.

Results

A total of 38 studies were included in the systematic review. Early arteriovenous fistula failure was significantly associated with a distal location of the fistula, small arterial diameter, small vein diameter, low serum albumin level, female gender, diabetes mellitus, and decreased systolic and diastolic blood pressure.

Conclusion

This systematic review identified several key risk factors for early arteriovenous fistula failure in patients with end-stage renal disease (ESRD) undergoing hemodialysis. These factors should be considered from a patient safety perspective and included in the shared decision-making process with patients who are candidates for arteriovenous fistula surgery. Optimizing these factors may enhance fistula maturation and reduce the need for repeat vascular access procedures. However, recognizing that some risk factors may not be readily modifiable in ESRD patients, clinicians should balance optimization efforts with the urgency of establishing vascular access.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13037-025-00449-9.

Background

Recent data indicate that about 2.5 million people worldwide have chronic renal failure, and around 80% depend on hemodialysis as their primary renal replacement therapy [1]. Data from Europe show that more than half a million patients with end-stage renal disease (ESRD) rely on hemodialysis, and by 2030, this number is expected to increase by more than 50% [1]. For delivering and maintaining adequate hemodialysis for ESRD patients, a well-functioning vascular access is crucial [1, 2].

A native arteriovenous fistula is the optimal vascular access, allowing connection between the patient’s blood and the hemodialysis machine, compared to central venous catheters [1–3]. arteriovenous fistula offers longer durability and fewer complications than central venous catheters or arteriovenous grafts [2]. An arteriovenous fistula is created by anastomosing a vein to an artery, facilitating dilatation and arterialization of the vein (maturation) and making it suitable for needle insertion during hemodialysis. This anastomosis leads to arterialization of the vein (thickening of its wall) and increase in his diameter making it suitable for future canulation for the purpose of hemodialysis. It can be created in the upper limb (distally or proximally) and, rarely, in the lower limbs [2].

Normally, arteriovenous fistula maturation process takes few weeks. However, arteriovenous fistula maturation does not occur in some cases. Studies attributed non-maturation to various factors, such as venous stenosis due to neointimal hyperplasia, surgical errors, and patient-related factors like hypotension and peripheral arterial disease [4, 5]. Unfortunately, arteriovenous fistula have a primary failure rate of 20–50% [6]. Studies estimate that $1–3 billion is spent annually managing failing or dysfunctional vascular access [7], highlighting the importance of identifying risk factors for primary failure to guide interventions.

In this systematic review, we aimed to identify factors contributing to arteriovenous fistula on-maturation in ESRD patients undergoing hemodialysis and provide recommendations for future research to improve patient outcomes.

Methods

The Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines were followed reporting of this systematic review [8]. The review protocol was previously registered on the Open Science Framework platform (https://osf.io/7ys32/). All available literature were systematically searched through March 2025.

We aimed to identify significant factors associated with primary arteriovenous fistula failure (non-maturation or dysfunction). Databases searched included PubMed, Web of Science, ScienceDirect, and the WHO Virtual Health Library Regional Portal. The keywords used were (Arteriovenous Fistula*) AND (primary failure OR early failure OR dysfunction OR Immaturity OR immature OR maturation OR poor maturation) AND (hemodialysis OR ESRD OR end stage renal disease OR renal failure OR chronic kidney disease) AND (risk factor* OR predict* factor* OR related factor OR factor* OR predict* OR associate* factor OR risk*). The initial screening of titles and abstracts was done by uploading relevant articles to Endnote software, with duplicates removed.

All cross-sectional, cohort or case control studies that reported sufficient data on factors associated with primary arteriovenous fistula failure. We excluded studies reporting data on failure and complications associated with central venous catheters, abstracts, letters, reviews, case reports, non-English articles, and studies without sufficient data of interest. The selection process was firstly done by initial screening with titles and abstracts of all identified articles for identifying relevant articles. Then, a comprehensive full-text review was done based on the inclusion criteria.

Critical appraisal checklist provided by “Joanna Biggs Institute” was used for evaluation (http://jbi.global/critical-appraisal-tools). These tools assess for possibility of bias in the study design, conduct, and analysis of data. Three independent reviewers extracted relevant data from each article. Inconsistency and discrepancies between reviewers were resolved via discussion and agreement.

We summarized data on all potential risk factors that were identified across the included studies. In addition, quantitative meta-analyses were conducted for factors that were presented in studies reporting comparable quantitative effect estimates suitable for pooling. Other factors that were reported infrequently or lacking suitable quantitative data for meta-analysis were summarized qualitatively. The extracted data included: authors, year of publication, country, age groups, number of patients, patients’ characteristics, methods used for determining of primary arteriovenous fistula failure, and associated factors with primary arteriovenous fistula failure.

Statistical analyses were carried out by calculating the pooled odd ratio (OR) and standardized mean difference (SMD) estimates with 95% confidence intervals (CI) to assess associations with primary arteriovenous fistula failure. To compensate for heterogeneity between studies, random effect model was used. The I² statistic test was used to evaluate heterogeneity among studies, which quantifies the percentage of variation in effect estimates attributable to heterogeneity rather than chance ( [9, 10]). Publication bias was assessed using Begg’s and Egger’s tests ( [9, 10]). The Duval and Tweedie trim-and-fill method was applied to account for potentially missing studies when there is evidence of a publication bias [11]. The significance level for both tests was set at 0.05. The analyses were conducted using the meta package in R version 4.4.3.

Results

Initial search revealed a total of 2349 records. after removal of duplicates, 1760 studies were submitted for tittle and abstract screening. Of which, 1722 were excluded due to irrelevance. Following full text assessment, a total of 38 studies met the eligibility criteria and were submitted for synthesis of evidence [12–49]. The schematic flow of studies identification and selection were illustrated in Fig. 1. The studies were conducted across diverse geographic regions, including Europe, Asia, North America, and Africa. All of the included studies were conducted on adult populations, with mean or median participant ages ranging from the fourth to seventh decades of life (Table 1). Commonly used methods to define the outcome included absence of thrill or bruit, failure of successful needle cannulation, or inadequate blood flow for dialysis. Most studies assessed functional usability of the fistula within a period of 1 to 3 months post-creation (Table 2).

Fig. 1.

Flow chart for studies selection process

Table 1.

Baseline characteristics of the included studies

Study	Country	Age group	Total No. of patients	Patients with primary AVF failure	Quality assessment
Abreu et al. 2018	Portugal	69 ± 18 years	117	23	9/11
Abreu et al. 2022	Portugal	Median of 71 years	155	15	6/9
Ahmadi et al. 2022	Iran	52.82 ± 16.94 years	400	30	6/8
Akin et al. 2016	Turkey	55.90 ± 17.5 years	119	27	9/11
Aitken et al. 2014	UK	60.5 ± 0.9	569	163	9/11
Anh et al. 2022	Taiwan	> 20 years	952	232	10/10
Bahadi et al. 2012	Morocco	49.28 ± 17.32 years	115	33	9/11
Bashar et al. 2015	Ireland	60.9 ± 16.9	86	52	10/10
Bhuwania et al. 2024	India	58.17 ± 13.42	91	41	10/10
Bojakowski et al. 2012	Poland	62 ± 15	68	11	10/11
Bosanquet et al. 2015	UK	Median of 72 years	247	52	9/11
De Winter et al. 2024	Netherlands	63.5612.4 years	55	14	9/11
Ernandez et al. 2005	Switzerland	60.28 ± 14.8 years	119	34	8/8
Eslami et al. 2016	United States	63.8 ± 14.0 years	376	66	7/9
Farber et al. 2016	United States	55.1 ± 13.4 years​	602	32	10/10
Feldman et al. 2003	United States	56 ± 17 years	348	120	7/9
Fila et al. 2014	Croatia	64.1 years ± 12.8	93	15	7/9
Genek et al. 2015	Turkey	68.5 ± 14.9 years	30	14	5/9
Gao et al. 2016	China.	58.97 ± 15.27 years	98	32	9/10
Ghosh et al. 2022	India	40.7 ± 13.08	205	55	9/11
Huijbregts et al. 2007	Netherlands	65 ± 14	395	131	11/11
Hod et al. 2014	United States	76.2 ± 6.02	20,360	4,579	11/11
Kim et al. 2001	South Korea	53 ± 14 years	53	10	9/11
Kim MH et al. 2015	South Korea	58.5 ± 13.9years	134	30	9/11
Kim JK et al. 2015	South Korea	62.1 ± 13.3 years	252	100	9/10
Martinez-mier et al. 2019	Mexico	36.3 ± 12.9 years	78	19	8/11
Mayer et al. 2016	Germany.	50 ± 87 years	41	6	10/11
Ozhan et al. 2020	Turkey	58.2 ± 13.8 years	122	34	7/10
Pandey et al. 2019	India	44.95 ± 11.7 years	224	62	9/11
Park et al. 2015	South Korea	55.86 ± 15.2 years	371	22	9/11
Pratama et al. 2023	Indonesia	53.2 ± 11.8 years	83	33	10/11
Ramanan et al. 2022	India	53.10 ± 14.54 years	460	90	7/9
Sari et al. 2018	Turkey	51.1 ± 15.4 years	36	10	7/9
Saucy et al. 2009	Switzerland	63 ± 14.3 years	58	12	8/10
Tubail et al. 2024	France	Median of 68 years	105	63	10/10
Wongmahisorn et al. 2019	Thailand	61.2 +_14.5 years	396	121	10/11
Yan et al. 2018	China	50.88 ± 15	1051	41	8/11
Zadeh et al. 2008	Iran	Median of 60 years	100	23	8/11

Table 2.

Summary of the main resuls

Study	Outcome definition	Factors associated with the outcome
Abreu et al. 2018	Thrombosis within one month of vascular access creation	Older age, vein diameter, and peak systolic velocity of radial and ulnar arteries.
Abreu et al. 2022	Lack of vein diameter > 4 mm, depth < 6 mm, and blood flow > 500 mL/min	Pulsatility index, peak systolic velocity, blood flow, and resistance index.
Ahmadi et al. 2022	Lack of thrill on the palpation or vascular murmur in the AVF examination within one month	Low BMI and weight
Akin et al. 2016	Lack of successful cannulation with two fistula needles with a blood flow of 250 mL/min for at least one complete dialysis session, after 4 weeks	Presence of thrill, amount of daily proteinuria, insulin levels, Homa-IR (Insulin Resistance evaluated by homeostasis model assessment), and serum albumin level.
Anh et al. 2022	Failure of AVF to be provided with prescribed HD consistently with two-needle insertion for at least two-thirds of HD sessions in four consecutive weeks	High doses and long-term use of contact laxatives, advanced age, diabetes, and use of contact laxative in combined with certain medications (e.g., erythropoiesis-stimulating agents, beta-blockers, calcium channel blockers) were associated with increased failure risks.
Aitken et al. 2014	Lack of thrill, bruit, or adequate maturation to permit needle cannulation	Uremia was associated with Early AVF Failure
Bahadi et al. 2012	primary non-function due to thrombosis (before puncturing) or non-maturation	Diabetes and distal site of creation of AVF.
Bashar et al. 2015	Use of AVF at least six successive occasions for hemodialysis.	Female gender
Bhuwania et al. 2024	Inability to use AVF for dialysis or non-maturity at 6 weeks	Male sex and less diabetes associated with a matured RC-AVF outcome. Preoperative vascular evaluation and artery and vein diameter is associated with poor AVF maturation, calcification of feeding artery and poor vein increment have prominent role in deciding maturation
Bojakowski et al. 2012	Failure of AVF to mature or thrombosis or stenosis within 2 months	High inflammatory markers
Bosanquet et al. 2015	Loss of AVF patency within 6 weeks	Ipsilateral CVC access, Age > 73 years, anastomosing Vein diameter < 2.2 mm, previous lower limb Angioplasty for peripheral vascular disease (PVD), and absent intra-operative Thrill .
De Winter et al. 2024	Lack of venous diameter of 4 mm with a blood flow of 500 mL/min, an easily compressible vein, an audible bruit, a palpable thrill, and an adequate venous length and depth for receiving two separate needles for hemodialysis within 12 weeks.	Interaoperative blood flow equal or lower than the determined cutoff point of 160 ml\min
Ernandez et al. 2005	Non-functioning AVF (thrombosis or absence of fistula maturation) within 6 months of creation	Distal location, female gender, diabetes mellitus, and limited surgical expertise.
Eslami et al. 2016	Loss of patency or lack of maturation within 3 months post operation.	COPD, AVF site on radial artery and cephalic vein, and smaller mean diameters of both inflow artery
Farber et al. 2016	Early thrombosis diagnosed by physical exam or ultrasound within 18 days of AVF creation.	Female gender, forearm AVF, smaller arterial size, draining vein diameter of 2–3 mm, and protamine use, Surgeon experience
Feldman et al. 2003	Use of AVF at least six successive occasions for hemodialysis.	History of hypertension, myocardial infarction, coronary artery bypass surgery; cerebrovascular events, thrombophlebitis or pulmonary embolism, dialysis dependence at the time of surgery (reduced odds of maturation), mean arterial pressure < 85 mmHg (associated with 48% reduction in odds of success), high intraoperative heparin doses, vein diameter, and side-to-side anastomosis.
Fila et al. 2014	AVF that thrombosed before the first cannulation or did not reach functional status. Immediate failure was defined as failed AVF function within the first 24 h. Early patency or failure was defined by thrill and/or bruit 4 weeks post-surgery.	BMI, vein diameter, and artery diameter
Genek et al. 2015	Absence of thrill and bruit by physical examination at 30 days postoperatively.​	Higher radial artery and brachial artery resistive index (RI) and lower peak systolic flow rate in failed fistulas. Older age and female gender, the narrower brachial artery diameter were associated with failure.​ The brachial artery peak systolic flow rate was significantly reduced in patients in the radio cephalic fistula group that developed early and late-term failure.
Gao et al. 2016	Absence of thrill and bruit by examination at 4–6 week.​	Cephalic vein and radial artery small diameter, preoperative blood pressure and PTH .
Ghosh et al. 2022	Use of AVF at least six successive occasions for hemodialysis.	Patient’s BMI and presence of diabetes had significant impact on outcome thrill was significantly associated with successful maturation,
Huijbregts et al. 2007	insufficient access flow to maintaining dialysis at 6 weeks after surgery	Female gender, renal replacement therapy prior AVF creation, the center of access creation, AVF site at wrist, diabetes .
Hod et al. 2014	Inability to use AVF at the dialysis initiation	Older age, female gender, black race, history of diabetes, cardiac failure, and shorter duration of ESRD show similar finding of entire cohort pre hemodialysis initiation, but in elderly hemodialysis population there is association between all this variable and pre dialysis AVF failure
Kim et al. 2001	having a complete obstruction of blood flow or a severe stenosis requiring radiologic intervention or surgical correction within 4 months.	Low access blood flow, small vein diameter, and diabetes mellitus .
Kim MH et al. 2015	Failure of two-point needling hemodialysis using the AVF between 6 weeks and 3 months post creation	Female gender, previous history of AVF, and the deference between the pre and postdilation diameter equal or less than 2.2 mm.
Kim JK et al. 2015	Failure to use the AVF successfully by 3 months after its creation.	Older age, female gender, obesity and low intraoperative blood flow, history of cardiovascular disease, low serum albumin level and high CRP level.
Martinez-mier et al. 2019	AVF that cannot be used with two-needle cannulation for two-thirds or more of all dialysis runs for 1 month delivering dialysis	Radiocephalic AVF, short prothrombin time, low serum albumin and AVF site on the same side of HD catheter .
Mayer et al. 2016	Lack of a progressive remodeling of the vein, readineas for two-needle cannulation with minimum risk of damage, and flow > 400mL/min with the US assessment within 6 weeks	The vessel diameter and the distensibility of the vein also appear to play an important role in successful maturation of an AVF.
Ozhan et al. 2020	Lack of an easily compressible vein, continuous audible bruit, and palpable thrill, at least 6 mm internal diameter, maximal 6 mm depth from the skin to the outflow vein, and flow > 600mL/min with the US assessment within 6 weeks	Older age and small artery diameter
Pandey et al. 2019	Failure to achieve vascular access from the fistula within first 4 months of its creation.	Age and brachial artery diameter
Park et al. 2015	Fistula that was not ready to use based on physical examination and US assessment within 3 months.	Small vessels diameter, diabetes, BMI, CHF.
Pratama et al. 2023	Lack of blood flow > 600 ml/min, vein depth < 0.6 cm, and a diameter > 0.6 cm within 6 weeks.	Low intraoperative brachial artery blood flow rate and peak systolic velocity in brachiocephalic AVF .
Ramanan et al. 2022	Primary outcomes were immediate failure defined as an AVF that has either no appearance of or loss of bruit or thrill within 72 h of creation and primary failure defined as an AVF that was inadequate for hemodialysis at 3 months’ post creation, including immediate failure, failure to mature or early thrombosis	younger age and feeding artery diameter < 2.0 mm associated with higher specificity for failure.
Sari et al. 2018	Thrombus formation, absence of thrill or maturation failure within 6 weeks	Female gender, absence of thrill post-operation, poor vein quality, smaller cephalic vein diameter at the Brescia-Cimino region, the RI of the radial artery in the snuff box region first decreased and thereafter increased, while the peak systolic velocity and end-diastolic velocity of the cephalic vein in the snuff box region first increased and thereafter decreased. CRP, ESR, and prothrombin values was found to be higher in the group with dysfunctional fistula.
Saucy et al. 2009	Inability to use the fistula for dialysis with two needles within the first 3months.	Intraoperative blood flow < 120mL
Tubail et al. 2024	Having thrombosis, stenosis, < 500 ml/min blood flow, < 5 mm vein diameter, or ≥ 6 mm deep vein within 4 months post-surgery	Low vein-distensibility
Wongmahisorn et al. 2019	AVF was not successfully cannulated or could not be used for hemodialysis within 3 months post-surgery	High preoperative and postoperative neutrophil-to-lymphocyte ratios, low BMI, low blood pressure and low rate of AVF creation
Yan et al. 2018	Lack of thrill on palpation and a low-pitched noise on auscultation of the anastomosis within 7 days of surgery	Cephalic vein diameter and preoperative MAP
Zadeh et al. 2008	n/a	Hb < 8 g/dL, diabetes, and age > 60 years were identified as predictive factors for AVF failure and ACE inhibitor intake was found to be protective.

A broad range of predictors were assessed for association with primary arteriovenous fistula failure. The factors included patient-related characteristics (female gender, diabetes mellitus, low body mass index (BMI), and hypoalbuminemia), vascular factors (smaller arterial and venous diameters, low intraoperative blood flow, poor vein distensibility, the absence of an intraoperative thrill or bruit, use of distal arteriovenous fistula sites, and low peak systolic velocities), and several inflammatory markers such as elevated C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), high neutrophil-to-lymphocyte ratio, and low hemoglobin levels (Table 2).

Meta-analysis of 12 studies revealed that both smaller vein diameter (SMD = − 0.429, 95% CI: -0.534 to -0.323; p < .001) and smaller artery diameter (SMD = -0.427, 95% CI: -0.573 to -0.282; p < .001) were significantly associated with increased risk of primary arteriovenous fistula failure. Additionally, intraoperative AFV blood flow was significantly lower in arteriovenous fistula failure group (SMD = -0.729, 95% CI: -1.181 to -0.277; p = .002) (Figs. 2, 3 and 4).

Fig. 2.

Forest plot for the association between vein diameter and primary arteriovenous fistula failure

Fig. 3.

Forest plot for the association between artery diameter and primary arteriovenous fistula failure

Fig. 4.

Forest plot for the association between arteriovenous fistula blood flow and primary arteriovenous fistula failure

Lower serum albumin levels were significantly associated with arteriovenous fistula failure (SMD = -0.423, 95% CI: -0.733 to -0.113; p = .007; I² = 69.34%). However, body mass index (BMI) was not significantly related to arteriovenous fistula failure (SMD = 0.044, 95% CI: -0.136 to 0.226; p = .628) (Figs. 5 and 6).

Fig. 5.

Forest plot for the association between serum albumin levels and primary arteriovenous fistula failure

Fig. 6.

Forest plot for the association between body mass index (BMI) and primary arteriovenous fistula failure

Regarding blood pressure, both systolic and diastolic blood pressure were inversely associated with arteriovenous fistula success. Mete-analysis of seven studies revealed that low levels of systolic blood pressures (SBP) were associated with significant higher risk for primary arteriovenous fistula failure (SMD = -0.346, 95% CI: -0.651 to -0.040; p = .026). Similarly, meta-analysis of seven studies revealed that low levels of diastolic blood pressures (DBP) were associated with significantly higher risk for primary arteriovenous fistula failure (SMD = -0.30, 95% CI: -0.565 to -0.003; P = .031) (Figs. 7 and 8).

Fig. 7.

Forest plot for the association between systolic blood pressure and primary arteriovenous fistula failure

Fig. 8.

Forest plot for the association between diastolic blood pressure and primary arteriovenous fistula failure

Female gender (OR = 1.423, 95% CI: 1.199 to 1.687; p < .001) and diabetes mellitus (OR = 1.314, 95% CI: 1.079 to 1.600; p = .006) were significantly associated with increased odds of primary arteriovenous fistula failure. Likewise, distal arteriovenous fistula location was predictive of primary arteriovenous fistula failure (OR = 1.795, 95% CI: 1.305 to 2.468; p < .001). In contrast, smoking was not significantly associated with primary arteriovenous fistula failure (OR = 1.038, 95% CI: 0.914 to 1.179; p = .568) (Additional file 1).

Results of studies heterogeneity assessment were variable across analyses, and most of the analyses showed moderate levels of heterogeneity (Table 3). Begg’s and Egger’s tests were used to assess potential for publication bias (Table 3). These tests were not significant for most of the analyses except for results of DBP. However, the Duval and Tweedie trim-and-fill indicated no absent study was found and the adjusted estimate remained consistent with the original finding.

Table 3.

Summary of meta-analyses results:

Variables	No. of studies	Estimate	Pooled effect size	P value	I2 test	Begg’s test (P value)	Egger’s test (P value)
Vein diameter	12	SMD	-0.429 (-0.534 to -0.323)	< 0.001	0%	0.545	0.172
Artery diameter	12	SMD	-0.427 (-0.573 to -0.282)	< 0.001	41.9%	0.841	0.831
Arteriovenous fistula blood flow	3	SMD	-0.729 (-1.181 to -0.277)	0.002	48.0%	0.333	0.066
Serum Albumin	6	SMD	-0.423 (-0.733 to − 0.113)	0.007	69.2%	0.136	0.064
Systolic BP	7	SMD	-0.346 (-0.651 to -0.040)	0.026	81.2%	1.00	0.973
Diastolic BP	7	SMD	-0.30 (-0.565 to -0.025)	0.031	75.5%	0.562	0.034
BMI	12	SMD	0.044 (-0.136 to 0.226)	0.628	82.2%	0.737	0.620
Location (distal arteriovenous fistula)	11	OR	1.795 (1.305 to 2.468)	< 0.001	43.7%	0.761	0.851
Diabetes mellitus	29	OR	1.314 (1.079 to 1.600)	0.006	59.4%	0.042	0.931
Smoking	15	OR	1.038 (0.914 to 1.179)	0.568	0%	0.181	0.759
Gender (female)	26	OR	1.423 (1.199 to 1.687)	< 0.001	50.3%	0.321	0.443

Discussion

In this systematic review, we assessed factors that predispose ESRD patients on hemodialysis to non- maturation and early failure of native arteriovenous fistula created by surgically anastomosing an inflow artery to an outflow vein. The meta-analysis results identified several key risk factors of early arteriovenous fistula failure and non-maturation in ESRD patients on hemodialysis.

Hemodialysis vascular access can be achieved through three primary methods: native arteriovenous fistula, arteriovenous graft, and central venous catheter. While arteriovenous fistula is considered the gold standard due to its superior durability and lower complication rates compared to other methods, arteriovenous graft can also serve as a reliable long-term access option for some patients. In contrast, central venous catheters are associated with significant risks, including bloodstream infections, sepsis, endocarditis, and spinal abscesses, making them the least preferred option [50]. This systematic review focuses on arteriovenous fistula due to its widespread use and clinical importance, aiming to identify predictors of early failure to optimize its outcomes.

The results revealed that a distally located arteriovenous fistula is associated with high risk of early failure compared to a more proximally located one. Proximally located veins are anatomically larger than distal ones, which affect the achievement of the blood flow needed for arteriovenous fistula function. In addition, most patients undergoing arteriovenous fistula creation have multiple comorbidities such as diabetes mellitus, hypertension, and atherosclerosis, which primarily affect the arterial system. Furthermore, secondary hyperparathyroidism in chronic kidney disease contributes to dystrophic vascular calcification, particularly in distal arteries. This may lead to reduced blood flow in distal arteries, thereby diminishing the shear stress necessary for proper arterialization and distension of the vein [46, 51].

The meta-analysis confirms the role of low serum albumin as a predictive factor for early arteriovenous fistula failure in patients undergoing hemodialysis. Albumin plays a vital role in maintaining vascular integrity and capillary permeability, and its deficiency is often a marker of ongoing inflammation [52]. This pro-inflammatory state may promote blood coagulation and contribute to early occlusion of a newly created arteriovenous fistula before full maturation is achieved ( [53, 54]). Therefore, preoperative assessment and optimization of nutritional and inflammatory markers, including serum albumin, are essential in patients ESRD disease to minimize the risk of early arteriovenous fistula failure.

Notably, our meta-analysis found no significant association between smoking and primary AVF failure, which contrasts with the conventional understanding of smoking as a risk factor for vascular dysfunction. This unexpected finding may be due to heterogeneity in how smoking status was defined or reported across studies, variations in smoking exposure (e.g., current vs. former smokers), or limited statistical power in some analyses. Future studies should standardize smoking definitions and explore dose-response relationships to clarify its role in AVF outcomes.

Another significant predictor was blood vessels diameters. Both vein and artery diameters are key predictors of arteriovenous fistula maturation. Studies have shown that larger vessel diameters are associated with higher rates of successful maturation and patency ( [46, 55, 56]). Concerning blood pressure, studies showed that lower systolic and diastolic blood pressure was significantly associated with an increased risk of primary arteriovenous fistula failure. The mechanism may involve insufficient vascular pressure, leading to reduced shear stress on the vessel wall. This can result in vascular recoil and repetitive trauma, ultimately promoting thrombosis and arteriovenous fistula failure ( [57, 58]).

The findings of this review have important clinical implications. First, the consistent association between anatomical variables (particularly vessel diameter) and arteriovenous fistula maturation supports the routine use of preoperative vascular mapping via duplex ultrasound to guide fistula site selection and improve surgical planning [59]. Second, the predictive value of hypoalbuminemia and blood pressure parameters highlights the need for multidisciplinary care, including nutritional support and cardiovascular optimization, prior to arteriovenous fistula creation. Furthermore, understanding these risk factors can aid in the development of risk stratification tools, enabling nephrologists and vascular surgeons to individualize access planning based on a patient’s profile. Finally, identifying modifiable predictors presents an opportunity for targeted interventions (e.g., blood pressure control, correction of malnutrition) to improve arteriovenous fistula success rates.

Despite the strengths, this review has several limitations. The included studies were observational in nature, which increases the risk of residual confounding. The studies included were conducted in adult populations, limiting applicability to pediatric patients. Several studies reported on additional factors potentially associated with primary AVF failure but lacked sufficient data for meta-analysis. These included inflammatory markers such as elevated CRP, ESR, and high neutrophil-to-lymphocyte ratio, as well as low hemoglobin levels. Other vital signs, such as pulse rate or temperature, were infrequently reported and inconsistently associated with AVF outcomes. These parameters warrant further investigation in future studies to determine their relevance to AVF maturation.

Conclusion

This systematic review and meta-analysis identified several significant predictors of primary arteriovenous fistula failure and non-maturation in patients with ESRD undergoing hemodialysis. Key risk factors included distal arteriovenous fistula location, low serum albumin, smaller arterial and venous diameters, and abnormal blood pressure levels. These findings indicate the multifactorial nature of primary arteriovenous fistula failure and highlight the importance of comprehensive preoperative evaluation, including vascular mapping, nutritional status, and blood pressure control, which may enhance AVF maturation while avoiding unnecessary delays in access placement. Optimizing these factors may enhance fistula maturation and reduce the need for repeat vascular access procedures. However, recognizing that some risk factors may not be readily modifiable in ESRD patients, clinicians should balance optimization efforts with the urgency of establishing vascular access. Future studies should aim to standardize outcome definitions and explore interventions targeting modifiable risk factors to improve arteriovenous fistula outcomes.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

BMI

Body Mass Index

CI

Confidence Interval

CRP

C-reactive protein

DBP

Diastolic Blood Pressure

ESR

Erythrocyte sedimentation rate

ESRD

End-Stage Renal Disease

OR

Odds Ratio

SBP

Systolic Blood Pressure

SMD

Standardized Mean Difference

Author contributions

SS conceptualized the research idea and plan. KS, MS, AAM, and YE undertook database searches and articles screening. AEA, ASA and NS undertook risk of bias assessment. KFA, AAOM, DI, SM, and AIM extracted and summarized data. SM analyzed data. SS, KM, AOM, and SM drafted the manuscript. All authors read and approved the final manuscript.

Funding

No fund.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.