Significance Statement
As currently used, preoperative ultrasound mapping for vascular access planning has not improved arteriovenous fistula (AVF) maturation rates. In their retrospective analysis involving 300 patients receiving a new AVF, the authors found that preoperative vascular diameter demonstrated a linear association with AVF maturation and did not correspond to a single threshold value. They also found that the preoperative arterial diameter, not venous diameter as generally believed, was the most significant predictor of AVF maturation. In addition, two previously unidentified factors, systolic BP and left ventricular ejection fraction, predicted unassisted AVF maturation. These findings suggest that a greater emphasis on the preoperative arterial diameter as a continuous variable, as well as consideration of the patient’s baseline systolic BP and cardiac function, may improve AVF maturation rates.
Keywords: arteriovenous fistula, vascular mapping, ultrasound, maturation, arterial diameter
Abstract
Background
Preoperative ultrasound mapping is routinely used to select vessels meeting minimal threshold diameters for surgical arteriovenous fistula (AVF) creation but fails to improve AVF maturation rates. This suggests a need to reassess the preoperative ultrasound criteria used to optimize AVF maturation.
Methods
We retrospectively identified 300 catheter-dependent patients on hemodialysis with a new AVF created between 2010 and 2016. We then evaluated the associations of preoperative vascular measurements and hemodynamic factors with unassisted AVF maturation (successful use for dialysis without prior intervention) and overall maturation (successful use with or without prior intervention). Multivariable logistic regression was used to identify preoperative factors associated with unassisted and overall AVF maturation.
Results
Unassisted AVF maturation associated with preoperative arterial diameter (adjusted odds ratio [aOR], 1.50 per 1-mm increase; 95% confidence interval [95% CI], 1.23 to 1.83), preoperative systolic BP (aOR, 1.16 per 10-mm Hg increase; 95% CI, 1.05 to 1.28), and left ventricular ejection fraction (aOR, 1.07 per 5% increase; 95% CI, 1.01 to 1.13). Overall AVF maturation associated with preoperative arterial diameter (aOR, 1.36 per 1-mm increase; 95% CI, 1.10 to 1.66) and preoperative systolic BP (aOR, 1.17; 95% CI, 1.06 to 1.30). Using receiver operating curves, the combination of preoperative arterial diameter, systolic BP, and left ventricular ejection fraction was fairly predictive of unassisted maturation (area under the curve, 0.69). Patient age, sex, race, diabetes, vascular disease, obesity, and AVF location were not associated with maturation.
Conclusions
Preoperative arterial diameter may be an under-recognized predictor of AVF maturation. Further study evaluating the effect of preoperative arterial diameter and other hemodynamic factors on AVF maturation is needed.
Although arteriovenous fistulas (AVFs) remain the preferred vascular access for hemodialysis,1 a substantial proportion (30%–60%) fail to mature adequately for dialysis use.2,3 Preoperative ultrasound vascular mapping is commonly used to select blood vessels meeting minimal threshold diameter criteria, with the underlying presumption that creating an AVF with such vessels will improve its likelihood of maturation.4 However, the existing criteria of ≥2.0-mm minimal diameter for the artery and ≥2.5-mm minimal diameter for the vein1 are on the basis of limited published data.5 Furthermore, in practice, increased use of these preoperative vascular parameters has not translated into an improvement in AVF maturation rates.6–8 For example, Allon et al.8 demonstrated higher rates of AVF creation (34%–64%) following the incorporation of routine preoperative vascular mapping, but the rate of AVF maturation was not increased (46%–54%, P=0.34). Similarly, a recent systematic review of four randomized, controlled trials reported that preoperative vascular mapping did not result in improved AVF maturation rates.9 These disappointing observations suggest the need to reassess the preoperative ultrasound criteria used to optimize AVF maturation and provide the rationale for this study.
The goal of this study was to evaluate preoperative ultrasound and hemodynamic factors predictive of AVF maturation. We analyzed 300 central venous catheter (CVC)–dependent patients on chronic hemodialysis at a large dialysis center who received a new AVF after obtaining preoperative ultrasound vascular mapping. This study sought to determine (1) whether treating preoperative blood vessel diameters as continuous rather than categorical variables would be associated with better prediction of AVF maturation and (2) whether other hemodynamic factors would also affect AVF maturation rates.
Methods
Patient Population and Clinical Resources
The University of Alabama at Birmingham (UAB) Division of Nephrology manages the care of approximately 550 patients on hemodialysis in the Birmingham metropolitan area, with nephrologists on the university faculty serving as medical directors of 12 units in the region. Preoperative and postoperative ultrasounds were performed by credentialed sonographers in an American College of Radiology–accredited ultrasound department and interpreted by radiologists experienced in vascular access for hemodialysis imaging. Four transplant surgeons with expertise in vascular access created all AVFs and performed all necessary surgical revisions. Percutaneous interventions, including thrombectomies and angioplasties, were performed by interventional radiologists and nephrologists at UAB Hospital.
Overview of Study Methods
Patients underwent standardized preoperative sonographic vascular mapping measuring arterial diameter, venous diameter, and brachial artery blood flow.10 Surgeons used ultrasound measurements to assist in vascular access planning. Per our institutional protocol, patients were scheduled for a standardized postoperative ultrasound approximately 6 weeks after AVF creation to assess sonographic AVF maturation. Clinical AVF maturation (suitability for cannulation) was determined on the basis of clinical assessment by surgeons, nephrologists, and hemodialysis nurses. Echocardiograms completed within 3 years prior to AVF surgery were reviewed to determine left ventricular ejection fraction (LVEF). The UAB Institutional Review Board approved this study and provided a waiver for obtaining informed consent.
Preoperative Vascular Mapping
Ultrasound vascular mapping prior to access placement is standard at our institution. The nondominant arm was generally preferred for AVF creation; however, the dominant arm was used if no suitable blood vessels were identified on ultrasound examination of the nondominant arm. Thrombosis or stenosis of the proposed vessels precluded AVF placement at that site. Threshold vessel criteria for AVF placement included arterial diameter ≥2.0 mm and venous diameter ≥2.5 mm.3,4,8 Brachial arterial blood flow was assessed in all patients so as to have a standard blood flow measurement encompassing both forearm and upper arm AVFs. It was recorded as a series of three flow measurements obtained 2 cm proximal to the antecubital fossa. Instances of high brachial artery bifurcation were noted as normal anatomic variants in 29 patients (10%).11 In such cases, brachial artery blood flow was calculated as the sum of the radial and ulnar artery blood flows 2 cm cranial to the antecubital fossa, and brachial arterial diameter was calculated as the sum of the radial and ulnar artery diameters, all measured 2 cm proximal to the antecubital fossa.12 Ultrasound reports were provided to the surgeons to assist in vascular access planning, but the ultimate selection of blood vessels and the final AVF location (forearm versus upper arm) remained at the surgeon’s discretion.
Postoperative Ultrasound
Nearly 80% of the study patients underwent 6-week postoperative ultrasounds to assess sonographic AVF maturation following a standardized protocol also used in the Hemodialysis Fistula Maturation (HFM) study.12,13 Inner draining vein diameter was measured at 2, 5, 10, and 15 cm proximal to the anastomosis. Blood flow through the AVF was measured three to five times in the draining vein 10 cm proximal to the anastomosis, and the results were averaged. AVFs with diameters ≥4 mm and blood flows ≥500 ml/min were deemed sonographically mature.13,14
AVF Clinical Maturation
Clinical AVF maturation was defined as removal of the CVC within 180 days of the postoperative ultrasound, such that the AVF was the only vascular access being used for hemodialysis.15 Surgical or percutaneous procedures, such as revision of the anastomosis, angioplasty, or thrombectomy, performed prior to CVC removal were considered efforts to assist AVF maturation. Isolated transposition of the vessels (i.e., planned two-step procedure) was not counted as a procedure to assist maturation.15 An AVF that matured without prior interventions was determined to have unassisted maturation. An AVF that remained unable to be cannulated successfully for hemodialysis 180 days after its creation was counted as failed to mature.
Data Collection
Two dedicated access coordinators employed by the Division of Nephrology maintained a prospective electronic database of all dialysis access procedures performed at UAB.16 We limited our analysis to patients undergoing AVF placement after initiation of dialysis with a CVC and excluded those with pre-ESKD AVF creation for three reasons. First, approximately 80% of United States patients initiate hemodialysis with a CVC,17 and we wanted to focus on outcomes in this important subpopulation. Second, analysis of patients with pre-ESKD AVF placement is confounded by the 30% who do not initiate dialysis in the ensuing 2 years due to death or lack of progression of kidney disease.18,19 Exclusion of this population from the analysis may introduce bias in the interpretation of outcomes of patients with pre-ESKD AVFs (for example, patients who die prior to needing dialysis may be less likely to experience AVF maturation). Third, the approach to nonmaturing AVFs often differs in patients predialysis compared with those who have already initiated dialysis. Specifically, there is frequently a “wait and see” approach to AVF maturation with long delays in interventions in patients pre-ESKD. In contrast, after the patient has started dialysis with a CVC, there is a greater sense of urgency to intervene in the case of nonmaturing AVFs in order to minimize the duration of CVC exposure.
Using our access database, we identified 688 patients who underwent new AVF placement from January 1, 2010 to December 31, 2016, including 358 who underwent pre-ESKD AVF creation and 330 (48%) with AVF placement after initiation of dialysis. We derived a patient cohort meeting the following inclusion criteria: (1) declared to have ESKD prior to AVF creation, (2) solely dependent on a CVC for hemodialysis, and (3) having had preoperative ultrasound vascular mapping prior to AVF creation. From this cohort, we excluded 4 patients who recovered kidney function subsequent to AVF placement, 9 with unknown timing of first AVF use, 2 due to patient death <7 days following AVF creation, and 15 due to loss to follow-up prior to assessment of AVF maturation (Figure 1). The remaining 300 patients (91% of those meeting inclusion criteria) served as the study cohort. The final cohort size of 300 patients was a coincidence and not predetermined by the authors. Forty-one patients (14%) included in this cohort participated in the HFM study.12 Clinical and demographic patient information was obtained from the electronic medical record. The first preoperative BP obtained on the day of AVF surgery was recorded.
Figure 1.
After applying exclusion criteria, a total of 300 patients were included in the study analysis.
Statistical Analyses
The major exposures of interest were the preoperative arterial and venous diameters closest to the AVF anastomosis and the brachial artery blood flow. Secondary exposures included preoperative systolic BP, LVEF, patient demographics, and comorbidities (Table 1). The major clinical outcomes of interest were unassisted and overall clinical AVF maturation. A secondary outcome was the AVF measurements obtained in the 6-week postoperative ultrasound assessing AVF diameter and blood flow.
Table 1.
Baseline demographic, clinical, and preoperative ultrasound characteristics of all 300 patients in the study cohort
| Variable | All Patients |
|---|---|
| N of patients | 300 |
| Age, yr, mean±SD | 53±15 |
| Sex, men (%) | 185 (62%) |
| Race, black (%) | 246 (82%) |
| Diabetes mellitus, N (%) | 157 (52%) |
| Coronary artery disease, N (%) | 70 (23%) |
| Peripheral vascular disease, N (%) | 41 (14%) |
| Cerebrovascular disease, N (%) | 58 (19%) |
| LVEF<55% on echocardiogram, N (%)a | 60 (20%) |
| Obesity (BMI≥30 kg/m2), N (%) | 139 (46%) |
| AVF location, upper arm, N (%) | 217 (72%) |
| Systolic BP, mm Hg, mean±SD | 146±25 |
| Diastolic BP, mm Hg, mean±SDb | 82±17 |
| Preoperative arterial diameter, mm, mean±SDc | 3.9±1.3 |
| Preoperative venous diameter, mm, mean±SDd | 3.9±1.1 |
| Brachial artery blood flow, ml/min, median [IQR]e | 53 [35–76] |
BMI, body mass index.
Echocardiogram data available for 235 patients (78%).
Diastolic BP data available for 298 patients (99%).
Preoperative arterial diameter data available for 292 patients (97%).
Preoperative venous diameter data available for 295 patients (98%).
Brachial artery blood flow data available for 298 patients (99%).
Complete data for patient demographics and comorbidities, preoperative systolic BP readings, and clinical AVF outcomes were available for analysis in all 300 study patients. Preoperative arterial diameter, venous diameter, and brachial artery blood flow measurements were missing in eight, five, and two patients, respectively. Preoperative diastolic BP readings were missing in two patients. Postoperative AVF ultrasounds were obtained in 238 patients (79%). Echocardiograms were done within 3 years prior to AVF creation in 235 patients (78%); of these, 81% were performed in the preceding 1 year. Analysis of LVEF was restricted to patients with available echocardiogram data. Patients were stratified by preoperative feeding arterial diameter (<3.0, 3.0–3.9, 4.0–4.9, and ≥5.0 mm), preoperative venous diameter (<3.0, 3.0–3.9, 4.0–4.9, and ≥5.0 mm), and brachial artery blood flow (<40, 40–59, 60–79, and ≥80 ml/min). Rates of unassisted AVF maturation were calculated for each subgroup and compared by the chi-squared test.
Data were analyzed using a chi-squared test for categorical variables and ANOVA or nonparametric tests for continuous variables to determine the association of baseline clinical characteristics and preoperative ultrasound measurements with postoperative ultrasound measurements and unassisted and overall AVF maturation. Multivariable logistic regression was used to evaluate the association of clinical patient characteristics and preoperative ultrasound measurements with unassisted AVF maturation and overall AVF maturation. Receiver operating characteristic curves were generated to determine the predictive value of preoperative ultrasound measurements and hemodynamic parameters on unassisted AVF maturation. P values of 0.05 were considered statistically significant.
Results
Baseline Patient Characteristics
The clinical and demographic characteristics of the entire study cohort are summarized in Table 1. Just over 60% of the patients were men, and 82% were black, consistent with the demographics of our hemodialysis population. Over 50% had diabetes mellitus, nearly 50% were obese, and 20% had an LVEF<55% on echocardiogram. A significant proportion had vascular disease. Mean systolic BP prior to surgery was 146±25 mm Hg. The majority of AVFs (72%) were created in the upper arm.
Association of Preoperative Vascular Measurements with Patient Characteristics
There was a moderate correlation between the preoperative arterial and venous diameters (R=0.38, P<0.001) and somewhat weaker correlation between preoperative arterial diameter and brachial arterial blood flow (R=0.28, P<0.001). We evaluated the relationship of clinical and demographic characteristics with preoperative arterial diameter, venous diameter, and brachial artery blood flow (Tables 2–4). Upper arm AVF location was associated with greater preoperative arterial and venous diameters. Preoperative arterial diameter was significantly associated with patient age, men, and coronary artery disease (Table 2). Median arterial diameter was 0.35 in women (interquartile range [IQR], 0.30–0.40) and 0.42 in men (IQR, 0.29–0.51; P<0.001). Preoperative venous diameter was not associated with any demographic characteristics or patient comorbidities (Table 3). Median vein diameter was 0.37 in women (IQR, 0.31–0.37) compared with 0.38 in men (IQR, 0.32–0.38; P=0.62). Preoperative brachial arterial blood flow was associated with men, obesity, and AVF location (Table 4). Women were nearly twice as likely to have an AVF placed in the upper arm: 92 of 115 women (80%) compared with 125 of 185 men (68%; odds ratio, 1.92; 95% confidence interval [95% CI], 1.10 to 3.40; P=0.02).
Table 2.
Patient characteristics by preoperative arterial diameter
| Arterial Diameter | <3 mm | 3–3.9 mm | 4–4.9 mm | ≥5 mm | P Value |
|---|---|---|---|---|---|
| Total, N | 73 (25%) | 80 (27%) | 81(28%) | 58 (20%) | |
| Age, yr, mean±SD | 48±14 | 52±16 | 53±15 | 58±12 | 0.002 |
| Sex, men (%) | 45 (62%) | 28 (35%) | 57 (70%) | 49 (84%) | <0.001 |
| Race, black (%) | 59 (81%) | 65 (81%) | 67 (83%) | 51 (88%) | 0.70 |
| Diabetes mellitus, N (%) | 35 (48%) | 38 (48%) | 45 (56%) | 33 (57%) | 0.60 |
| Coronary artery disease, N (%) | 8 (11%) | 17 (21%) | 25 (31%) | 18 (31%) | 0.01 |
| Peripheral vascular disease, N (%) | 6 (8%) | 12 (15%) | 11 (14%) | 10 (17%) | 0.46 |
| Cerebrovascular disease, N (%) | 10 (14%) | 15 (19%) | 18 (22%) | 14 (24%) | 0.43 |
| Obesity (BMI≥30 kg/m2), N (%) | 32 (44%) | 38 (48%) | 40 (49%) | 28 (48%) | 0.92 |
| LVEF<55%, N (%) | 13 (18%) | 16 (20%) | 17 (21%) | 12 (21%) | 0.96 |
| AVF location, upper arm, N (%) | 11 (15%) | 68 (85%) | 77 (95%) | 55 (95%) | <0.001 |
| Preoperative systolic BP, mm Hg, mean±SD | 148±27 | 142±22 | 148±27 | 147±24 | 0.42 |
| Preoperative diastolic BP, mm Hg, mean±SD | 85±15 | 76±20 | 85±17 | 83±18 | 0.004 |
n=292 (eight missing values). BMI, body mass index.
Table 4.
Patient characteristics by preoperative brachial artery blood flow
| Brachial Artery Blood Flow | <40 ml/min | 40–59 ml/min | 60–79 ml/min | ≥80 ml/min | P Value |
|---|---|---|---|---|---|
| Total, N | 94 (32%) | 72 (24%) | 65 (22%) | 67 (22%) | |
| Age, yr, mean±SD | 51±16 | 53±15 | 54±13 | 53±13 | 0.67 |
| Sex, men (%) | 46 (49%) | 41 (57%) | 43 (66%) | 54 (81%) | <0.001 |
| Race, black (%) | 80 (85%) | 56 (78%) | 54 (83%) | 55 (49%) | 0.67 |
| Diabetes mellitus, N (%) | 42 (45%) | 39 (54%) | 42 (65%) | 33 (49%) | 0.09 |
| Coronary artery disease, N (%) | 23 (24%) | 14 (19%) | 17 (26%) | 15 (22%) | 0.80 |
| Peripheral vascular disease, N (%) | 17 (18%) | 9 (13%) | 6 (9%) | 9 (13%) | 0.44 |
| Cerebrovascular disease, N (%) | 17 (18%) | 16 (22%) | 16 (25%) | 9 (13%) | 0.37 |
| Obesity (BMI≥30 kg/m2), N (%) | 31 (33%) | 34 (47%) | 35 (54%) | 38 (57%) | 0.01 |
| LVEF<55%, N (%) | 16 (17%) | 14 (19%) | 15 (23%) | 15 (22%) | 0.77 |
| AVF location, upper arm, N (%) | 74 (79%) | 52 (72%) | 51 (78%) | 55 (58%) | 0.02 |
| Preoperative systolic BP, mm Hg, mean±SD | 145±24 | 146±26 | 145±24 | 149±26 | 0.82 |
| Preoperative diastolic BP, mm Hg, mean±SD | 81±16 | 84±18 | 81±17 | 83±17 | 0.58 |
n=298 (two missing values). BMI, body mass index.
Table 3.
Patient characteristics by preoperative venous diameter
| Venous Diameter | <3 mm | 3–3.9 mm | 4–4.9 mm | ≥5 mm | P Value |
|---|---|---|---|---|---|
| Total, N | 55 (19%) | 111 (38%) | 80 (27%) | 49 (17%) | |
| Age, yr, mean±SD | 51±14 | 54±14 | 52±16 | 53±13 | 0.82 |
| Sex, men (%) | 35 (64%) | 64 (58%) | 48 (60%) | 34 (69%) | 0.54 |
| Race, black (%) | 48 (87%) | 89 (80%) | 64 (80%) | 41 (84%) | 0.66 |
| Diabetes mellitus, N (%) | 26 (47%) | 55 (50%) | 48 (60%) | 24 (49%) | 0.39 |
| Coronary artery disease, N (%) | 9 (16%) | 21 (19%) | 24 (30%) | 16 (33%) | 0.07 |
| Peripheral vascular disease, N (%) | 9 (16%) | 13 (12%) | 9 (11%) | 8 (16%) | 0.71 |
| Cerebrovascular disease, N (%) | 11 (20%) | 23 (21%) | 15 (19%) | 7 (14%) | 0.81 |
| Obesity (BMI≥30 kg/m2), N (%) | 20 (36%) | 50 (45%) | 46 (58%) | 22 (45%) | 0.10 |
| LVEF<55%, N (%) | 9 (16%) | 23 (21%) | 18 (23%) | 9 (18%) | 0.83 |
| AVF location, upper arm, N (%) | 26 (47%) | 70 (63%) | 68 (85%) | 48 (98%) | <0.001 |
| Preoperative systolic BP, mm Hg, mean±SD | 146±26 | 148±26 | 145±23 | 145±26 | 0.87 |
| Preoperative diastolic BP, mm Hg, mean±SD | 85±18a | 83±16 | 80±17 | 80±18a | 0.27 |
n=295 (five missing values). BMI, body mass index.
Two missing values: one from each category.
Association of Preoperative Vascular and Clinical Measurements with Unassisted AVF Maturation and Overall AVF Maturation
For the entire cohort of 300 patients, the unassisted AVF maturation rate was 43% (n=129), and the overall AVF maturation rate was 65% (n=194). Upper arm and forearm AVFs had similar rates of unassisted maturation (44% versus 40%) and overall maturation (65% versus 63%). Using univariable analysis, we identified several preoperative factors associated with unassisted AVF maturation, including preoperative systolic BP, LVEF, feeding arterial diameter, and venous diameter (Table 5). Overall (unassisted or assisted) AVF maturation was associated with higher preoperative systolic BP, arterial diameter, and venous diameter but not with LVEF (Table 5). Unassisted AVF maturation and overall AVF maturation were achieved in 59% and 83% of patients with a preoperative arterial diameter ≥5 mm, respectively, compared with 33% and 59% of those with a diameter <3 mm, respectively. Unassisted AVF maturation and overall AVF maturation occurred in 55% and 75% of patients with a preoperative venous diameter ≥5 mm, respectively, compared with 31% and 56% of those with a diameter <3 mm, respectively. Likewise, unassisted AVF maturation and overall AVF maturation occurred in 57% and 81% of those with a preoperative brachial arterial blood flow ≥80 ml/min, respectively, compared with 33% and 54% of those with brachial arterial blood flow <40 ml/min, respectively (Tables 6 and 7).
Table 5.
Univariable analysis of factors associated with unassisted AVF maturation (without prior intervention) and overall AVF maturation (assisted maturation + unassisted maturation)
| Variable | Unassisted Maturation | Overall Maturation | ||
|---|---|---|---|---|
| OR (95% CI) | P Value | OR (95% CI) | P Value | |
| Age (per 10 yr) | 0.91 (0.83 to 1.11) | 0.55 | 0.97 (0.98 to 1.01) | 0.27 |
| Sex, men versus women | 1.37 (0.85 to 2.20) | 0.19 | 1.09 (0.67 to 1.77) | 0.73 |
| Race, nonblack versus black | 1.23 (0.67 to 2.25) | 0.50 | 1.20 (0.66 to 2.21) | 0.55 |
| Diabetes mellitus | 1.03 (0.65 to 1.62) | 0.91 | 1.30 (0.81 to 2.09) | 0.28 |
| Coronary artery disease | 1.38 (0.81 to 2.37) | 0.24 | 1.63 (0.90 to 2.95) | 0.10 |
| Peripheral vascular disease | 1.58 (0.81 to 3.06) | 0.17 | 1.58 (0.76 to 3.300 | 0.22 |
| Cerebrovascular disease | 0.67 (0.37 to 1.22) | 0.19 | 0.95 (0.52 to 1.73) | 0.88 |
| LVEF, per 5% increase | 1.05 (1.00 to 1.12) | 0.04 | 1.04 (0.98 to 1.10) | 0.19 |
| Obesity, BMI≥30 kg/m2 | 1.00 (0.63 to 1.59) | 0.99 | 1.13 (0.70 to 1.82) | 0.61 |
| Preoperative SBP per 10-mm Hg increase | 1.16 (1.05 to 1.28) | 0.003 | 1.19 (1.07 to 1.32) | <0.001 |
| Preoperative DBP per 10-mm Hg increase | 1.05 (0.93 to 1.20) | 0.41 | 1.14 (1.0 to 3.2) | 0.06 |
| AVF location, upper arm versus forearm | 1.06 (0.62 to 1.79) | 0.84 | 1.13 (0.67 to 1.91) | 0.65 |
| Preoperative arterial diameter per 1-mm increase | 1.43 (1.18 to 1.72) | <0.001 | 1.35 (1.11 to 1.65) | 0.002 |
| Preoperative vein diameter per 1-mm increase | 1.33 (1.09 to 1.64) | 0.006 | 1.38 (1.10 to 1.72) | 0.004 |
| Preoperative brachial blood flow per 10-ml/min increase | 1.04 (1.00 to 1.10) | 0.06 | 1.06 (1.00 to 1.13) | 0.34 |
OR, odds ratio; BMI, body mass index; SBP, systolic BP; DBP, diastolic BP.
Table 6.
Association of preoperative ultrasound parameters with AVF outcomes: Arterial and venous diameters
| AVF | <3 mm | 3–3.9 mm | 4–4.9 mm | ≥5 mm | P Value |
|---|---|---|---|---|---|
| Arterial diametera | |||||
| Total AVFs, N | 73 | 80 | 81 | 58 | |
| Failure to mature, N (%) | 30 (41%) | 35 (44%) | 26 (32%) | 10 (17%) | 0.007 |
| Assisted maturation, N (%) | 19 (26%) | 17 (21%) | 15 (18%) | 14 (24%) | |
| Unassisted maturation, N (%) | 24 (33%) | 28 (35%) | 40 (49%) | 34 (59%) | |
| Venous diameterb | |||||
| Total AVFs, N | 55 | 111 | 80 | 49 | |
| Failure to mature, N (%) | 24 (44%) | 47 (42%) | 20 (25%) | 12 (24%) | 0.02 |
| Assisted maturation, N (%) | 14 (25%) | 22 (28%) | 18 (22%) | 10 (20%) | |
| Unassisted maturation, N (%) | 17 (31%) | 42 (52%) | 42 (52%) | 27 (55%) |
Missing eight values.
Missing five values.
Table 7.
Association of preoperative ultrasound parameters with AVF outcomes: Brachial artery blood flow
| Brachial Artery Blood Flowa | <40 ml/min | 40–59 ml/min | 60–79 ml/min | ≥80 ml/min | P Value |
|---|---|---|---|---|---|
| Total AVFs, N | 94 | 72 | 65 | 67 | |
| Failure to mature, N (%) | 43 (46%) | 28 (39%) | 22 (34%) | 13 (19%) | 0.02 |
| Assisted maturation, N (%) | 20 (21%) | 15 (21%) | 13 (20%) | 16 (24%) | |
| Unassisted maturation, N (%) | 31 (33%) | 29 (40%) | 30 (46%) | 38 (57%) |
Missing two values.
Multivariable analysis demonstrated three factors predicting unassisted AVF maturation: (1) preoperative arterial diameter, (2) preoperative systolic BP, and (3) LVEF. Only two of these factors were associated with overall AVF maturation: (1) preoperative arterial diameter and (2) preoperative systolic BP (Table 8). The interactions among these factors in determining unassisted and overall AVF maturation are presented graphically in Figure 2. Among patients with a systolic BP >150 mm Hg, progressively greater arterial diameters were associated with higher rates of unassisted AVF maturation. Similarly, among patients with a systolic BP ≤150 mm Hg, larger arterial diameters were also associated with progressively greater rates in unassisted AVF maturation. Among patients with an LVEF<55%, progressively larger arterial diameters were associated with greater rates of unassisted AVF maturation. However, among patients with an LVEF≥55%, the relationship between AVF maturation and arterial diameter was less well defined.
Table 8.
Multivariable analysis of factors predicting unassisted AVF maturation and overall AVF maturation
| Variable | aOR | 95% CI | P Value |
|---|---|---|---|
| Unassisted AVF maturation | |||
| Preoperative arterial diameter per 1-mm increase | 1.50 | 1.23 to 1.83 | <0.001 |
| Preoperative SBP per 10-mm Hg increase | 1.16 | 1.05 to 1.28 | |
| LVEF per 5% increase | 1.07 | 1.01 to 1.13 | |
| Overall AVF maturation | |||
| Preoperative arterial diameter per 1-mm increase | 1.36 | 1.10 to 1.66 | 0.002 |
| Preoperative SBP per 10-mm Hg increase | 1.17 | 1.06 to 1.30 |
SBP, systolic BP.
Figure 2.
Larger arterial diameters were associated with higher rates of unassisted and overall AVF maturation in the study population, even when categorized by systolic blood pressure and left ventricular ejection fraction.
There was no significant interaction between the preoperative arterial diameter and venous diameter in determining unassisted AVF maturation (P=0.93). Finally, when the multivariable analysis was repeated after excluding the preoperative arterial diameter from the model, venous diameter was the only significant predictor of unassisted AVF maturation (adjusted odds ratio [aOR], 1.37 for each 1-mm increase in venous diameter; 95% CI, 1.11 to 1.69; P=0.004).
Sensitivity Analyses
In our initial analysis, we considered the 28 patients who had only an isolated transposition procedure prior to AVF maturation to have had unassisted AVF maturation. We performed a sensitivity analysis in which these 28 patients were reclassified as having had an assisted AVF maturation. Using this alternate definition, multivariable logistic regression analysis found that unassisted AVF maturation was associated with preoperative arterial diameter (aOR, 1.43; 95% CI, 1.16 to 1.76 per 1-mm greater diameter) and systolic BP (aOR, 1.13; 95% CI, 1.01 to 1.24 per 10 mm Hg) but not with LVEF. These odds ratios were similar to those obtained using the original definition of unassisted AVF maturation (Table 8).
Receiver operating characteristic curves assessing preoperative arterial diameter, systolic BP, and LVEF showed that each of these individual factors was mildly predictive of unassisted AVF maturation. When all three factors were considered collectively, however, the model was fairly predictive of unassisted AVF maturation, with an area under the curve of 0.69 (95% CI, 0.62 to 0.75) (Figure 3).
Figure 3.
Combining preoperative arterial diameter, left ventricular ejection fraction, and systolic blood pressure in the model was fairly predictive of unassisted AVF maturation (AUC 0.69).
Association of Preoperative Ultrasound Measurements with Postoperative AVF Measurements
Larger preoperative arterial and venous diameters were associated with greater postoperative AVF diameter and blood flow (Tables 9 and 10). In contrast, the preoperative brachial artery blood flow was only marginally associated with the postoperative AVF diameter (P=0.04) and was not associated with AVF blood flow.
Table 9.
Association of preoperative ultrasound parameters with postoperative AVF measurements: Arterial and venous diameters
| Preoperative Vessel Diameter | <3 mm | 3–3.9 mm | 4–4.9 mm | ≥5 mm | P Value |
|---|---|---|---|---|---|
| Arterial diameter | |||||
| Total AVFs, N (%) | 73 (25%) | 80 (27%) | 81 (28%) | 58 (20%) | |
| Postoperative diameter, mm, mean±SD | 6.1±1.8 | 7.5±2.3 | 8.2±2.5 | 7.8±2.0 | <0.001 |
| Postoperative blood flow, ml/min, median [IQR] | 546 [195–761] | 536 [362–878] | 815 [537–1190] | 838 [624–1167] | <0.001 |
| Venous diameter | |||||
| Total AVFs, N (%) | 55 (19%) | 111 (38%) | 80 (27%) | 49 (17%) | |
| Postoperative diameter, mm, mean±SD | 6.7±1.9 | 7.0±2.4 | 7.7±1.8 | 9.0±2.4 | <0.001 |
| Postoperative blood flow, ml/min, median [IQR] | 556 [279–804] | 550 [266–873] | 833 [598–1099] | 809 [531–1106] | <0.001 |
Table 10.
Association of preoperative ultrasound parameters with postoperative AVF measurements: Brachial artery blood flow
| Brachial Artery Blood Flow | <40 ml/min | 40–59 ml/min | 60–79 ml/min | ≥80 ml/min | P Value |
|---|---|---|---|---|---|
| Total AVFs, N (%) | 94 (32%) | 72 (24%) | 65 (22%) | 67 (22%) | |
| Postoperative diameter, mm, mean±SD | 7.0±2.1 | 8.2±2.5 | 7.5±2.4 | 7.3±2.1 | 0.04 |
| Postoperative blood flow, ml/min, median [IQR] | 643 [356–897] | 708 [343–988] | 683 [483–1091] | 680 [495–1060] | 0.52 |
Discussion
This study assessed the predictive value of preoperative vascular ultrasound and hemodynamic measurements for AVF maturation, with several noteworthy findings. First, unassisted and overall AVF maturation had a linear association with the preoperative arterial diameter and did not correspond to a single threshold value. Second, using multivariable analysis, AVF maturation was more strongly associated with the preoperative arterial diameter than the preoperative venous diameter at the site of the anastomosis. Finally, in the multivariable model, preoperative systolic BP and LVEF were also predictive of AVF maturation. These observations have important implications for vascular access in patients on hemodialysis and suggest that further study is needed to evaluate the role of the preoperative arterial diameter and hemodynamic parameters in vascular access planning.
As currently used, preoperative vascular mapping ultrasound increases the rates of AVF placement in the United States20 without concomitant increases in AVF maturation.9,21–24 Existing vascular access guidelines propose relatively low minimum arterial and venous diameters for AVF creation, with the implicit assumption that, after these thresholds have been met, further increases in vessel diameter will not improve AVF maturation outcomes.1 These diameter thresholds are largely on the basis of a single 1998 European study that enrolled 35 patients receiving forearm AVFs exclusively, and they are not consistent with current United States practice.5
The limitations inherent in using the current thresholds are perhaps best reflected by the dramatic evolution in clinical practice over the last decade. In the United States, there has been a steady rise in placement of AVFs preferentially in the upper arm rather than in the forearm, representing a substantial departure from the 2006 Kidney Disease Outcomes Quality Initiative vascular access guidelines. For example, 46% of AVFs in the Dialysis Access Consortium study were placed in the upper arm.2 Just 7 years later, the HFM study reported that 76% of AVFs were placed in the upper arm.3 Likewise, a recent analysis from the Dialysis Outcomes and Practice Pattern Study found that the proportion of AVFs placed in the upper arm in the United States has risen steadily from 30% to 68% over the past 20 years.25,26 Among our patient cohort, 72% of AVFs were placed in the upper arm. This dramatic increase in upper arm AVFs is in keeping with the idea that AVFs created with larger vessels have superior maturation rates, one of the major findings of our study.
The contribution of venous diameter to AVF maturation evident on univariable analysis in this study disappeared on multivariable analysis, whereas the arterial diameter remained a key predictor of both unassisted and overall AVF maturation. This finding contradicts conventional wisdom, which regards the preoperative venous diameter as the most influential factor for AVF maturation. However, previous studies looking at preoperative ultrasound measurements associated with AVF maturation have not always included multivariable analysis. In addition, some United States centers, especially those using venography for preoperative mapping, omit preoperative arterial measurements altogether.27–31 We suspect that the significant association of venous diameter with unassisted maturation on univariable analysis reflects the correlation between preoperative arterial and venous diameters. In essence, a larger venous diameter is a surrogate marker for a larger arterial diameter but by itself, is not the best predictor of AVF maturation. Thus, we suggest that measurements of arterial diameter be routinely incorporated into preoperative ultrasound mapping prior to AVF placement.
A higher LVEF was associated with greater unassisted AVF maturation, and systolic BP was associated with both unassisted and overall maturation. The latter finding is consistent with a previous study by Feldman et al.,32 which showed that mean arterial pressure of 85 mm Hg or higher immediately prior to AVF creation was associated with higher rates of maturation. Although our study suggests that higher systolic BPs may improve AVF outcomes, multicenter prospective studies are needed to confirm this finding before changing clinical practice. Furthermore, the safety of “permissive hypertension” in patients on dialysis with maturing AVFs has not been established, and it may potentially lead to adverse cardiovascular outcomes. These hemodynamic factors help to determine arterial inflow, which is the primary force driving venous remodeling following AVF creation. Interestingly, the preoperative brachial artery blood flow was not itself associated with AVF maturation, suggesting that the systolic BP and ejection fraction affect AVF hemodynamics in ways not captured by the brachial artery blood flow. In addition, there is compelling evidence that AVF maturation is influenced by arterial function, a parameter not typically evaluated under the current standard of care.33
We recognize that adopting larger arterial diameters to determine suitability for AVF creation would have far-reaching clinical implications. If, for example, the minimal arterial diameter was raised from 2 to 4 mm, 52% (153 of 300) of the patients in this study would not have been eligible for an AVF at their current locations. Because arterial diameters are larger in the upper arm than in the forearm, stipulating a larger arterial diameter would predictably increase AVF placement in the upper arm. Although such a change contradicts current vascular access guidelines,1 it mirrors the aforementioned trends being observed in the United States.26
The reason for the shift toward placement of upper arm AVFs in the United States is likely twofold. First, forearm AVFs have a higher nonmaturation rate than those created in the upper arm.34–38 Second, Medicare reimbursement for outpatient hemodialysis is linked in part to the percentage of patients dialyzing via an AVF, providing a significant incentive to create functional and durable AVFs.26 Although the potential disadvantages of upper arm AVFs, such as limiting future vascular access sites or potential for cardiac steal, are important considerations, they must be carefully weighed against the greater likelihood of achieving a mature, functional AVF and decreasing reliance on CVCs. In this regard, a recent commentary challenged the rationale for preferring forearm AVFs indiscriminately in all patients on hemodialysis.39
A second consequence if larger arterial diameters were to be adopted for AVF creation is that AVF creation might be precluded in patients not meeting such a threshold, even in the brachial artery. If the minimal preoperative arterial diameter was raised to 4 mm in this study cohort, then 36% of our patients with upper arm AVFs would be precluded from receiving any AVF. In practical terms, this strategy would likely result in more patients receiving an arteriovenous graft but might also concurrently decrease the proportion who are CVC dependent by increasing the probability of AVF maturation. It would also mean that more patients receiving an AVF would likely achieve unassisted maturation, with the accompanying benefits of shorter catheter dependence, greater secondary AVF patency, and fewer interventions to maintain long-term patency for hemodialysis. However, should the results of this study be validated by large, multicenter studies, we would still recommend exercising caution in applying a vascular access planning strategy with the potential to limit future vascular access sites to younger patients on hemodialysis with a greater life expectancy.
The strengths of this study include its large study population; availability of complete preoperative and postoperative ultrasound data and clinical outcomes for the vast majority of patients; use of standardized preoperative and postoperative ultrasound protocols; and analysis of vascular diameters, systolic BP, and LVEF as continuous variables, permitting nuanced interpretation of their associations with unassisted AVF maturation.
First, one limitation of this study is its retrospective nature. Second, the reported analyses were limited to a single-center population with a large black population, and the results may not extrapolate to hemodialysis centers with different racial composition. Validation of our observations at other dialysis centers would strengthen the confidence in these findings. Third, preoperative echocardiogram and postoperative ultrasound data were not available for all patients, and although 81% of available echocardiograms were done within 1 year preceding AVF creation, they may not reflect the patient’s hemodynamic state at the time of surgery. Fourth, sonographic markers of arterial function, such as flow-mediated dilation or nitroglycerin-mediated dilation, are not routinely performed on preoperative vascular mapping ultrasounds. These tests, if done, might offer a more precise assessment of arterial function and could offer further insight into the predictive value of preoperative systolic BP or LVEF for AVF maturation.33 Fifth, we had a limited number of surgeons creating AVFs and thus, cannot account for variations in practice patterns or individual surgeon training and experience, which are associated with AVF maturation and survival.40,41 Finally, measurements of the diameter throughout the entire course of the outflow vein and measurements of central vein patency were not available with routine preoperative vascular mapping reports. These measurements may also affect AVF maturation, but their influence cannot be determined by this study.
In summary, up to 60% of new AVFs fail to mature adequately to be used for hemodialysis.2,23,42–44 Approximately 50% of those that mature require at least one surgical or percutaneous intervention to assist maturation.21,22,45,46 Both AVF nonmaturation and assisted maturation result in prolonged reliance on CVCs, exposing patients to the risk of catheter-related complications.47–49 Compared with AVFs with unassisted maturation, those with assisted maturation have shortened secondary survival and require more frequent interventions to maintain patency following maturation.21,22 If preoperative ultrasound can more accurately identify those patients who are most likely to experience successful AVF maturation, there could be a significant reduction in patient morbidity and cost related to multiple invasive procedures or abandoned accesses.
Our study suggests that (1) the preoperative arterial diameter should be evaluated as part of routine vascular access planning; (2) the venous diameter is less predictive of unassisted or overall AVF maturation when the arterial diameter is taken into consideration; (3) the preoperative arterial diameter may be an under-recognized predictor of AVF maturation outcomes; and (4) preoperative systolic BP and LVEF also predict unassisted AVF maturation. Larger arterial diameters might be associated with improved AVF maturation rates, but this retrospective study cannot establish a causal relationship. Validation with additional prospective, multicenter studies is needed prior to implementing widespread change in clinical practice.
Disclosures
Dr. Allon reports being a consultant for CorMedix. Dr. Lee reports other from Proteon Therapeutics, other from Boston Scientific, other from Merck, and grants from the National Institutes of Health outside the submitted work. Dr. Robbin reports grants and personal fees from Philips Ultrasound outside the submitted work. All remaining authors have nothing to disclose.
Funding
Dr. Farrington reports National Institutes of Health NIH T32 training grant DK007545-30 during the conduct of the study.
Acknowledgments
Dr. Farrington, Dr. Allon, and Dr. Robbin designed the study; Dr. Farrington collected the data; Dr. Allon, Dr. Barker-Finkel, and Dr. Farrington analyzed the data; Dr. Allon and Dr. Farrington made the figures and tables; Dr. Allon and Dr. Farrington drafted the paper; Dr. Allon, Dr. Barker-Finkel, Dr. Farrington, Dr. Lee, and Dr. Robbin revised the paper; and Dr. Allon, Dr. Barker-Finkel, Dr. Farrington, Dr. Lee, and Dr. Robbin approved the final version of the manuscript.
Footnotes
Published online ahead of print. Publication date available at www.jasn.org.
References
- 1.KDOQI; National Kidney Foundation : KDOQI clinical practice guidelines and clinical practice recommendations for anemia in chronic kidney disease [published correction appears in Am J Kidney Dis 48: 518, 2006]. Am J Kidney Dis 47[Suppl 3]: S11–S145, 2006. [DOI] [PubMed] [Google Scholar]
- 2.Dember LM, Beck GJ, Allon M, Delmez JA, Dixon BS, Greenberg A, et al.; Dialysis Access Consortium Study Group: Effect of clopidogrel on early failure of arteriovenous fistulas for hemodialysis: A randomized controlled trial. JAMA 299: 2164–2171, 2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Dember LM, Imrey PB, Beck GJ, Cheung AK, Himmelfarb J, Huber TS, et al.; Hemodialysis Fistula Maturation Study Group: Objectives and design of the hemodialysis fistula maturation study. Am J Kidney Dis 63: 104–112, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Robbin ML, Gallichio MH, Deierhoi MH, Young CJ, Weber TM, Allon M: US vascular mapping before hemodialysis access placement. Radiology 217: 83–88, 2000. [DOI] [PubMed] [Google Scholar]
- 5.Malovrh M: Non-invasive evaluation of vessels by duplex sonography prior to construction of arteriovenous fistulas for haemodialysis. Nephrol Dial Transplant 13: 125–129, 1998. [DOI] [PubMed] [Google Scholar]
- 6.Silva MB Jr., Hobson RW 2nd, Pappas PJ, Jamil Z, Araki CT, Goldberg MC, et al.: A strategy for increasing use of autogenous hemodialysis access procedures: Impact of preoperative noninvasive evaluation. J Vasc Surg 27: 302–307, 1998 [DOI] [PubMed] [Google Scholar]
- 7.Sorace AG, Robbin ML, Umphrey H, Abts CA, Berry JL, Lockhart ME, et al.: Ultrasound measurement of brachial artery elasticity prior to hemodialysis access placement: A pilot study. J Ultrasound Med 31: 1581–1588, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Allon M, Lockhart ME, Lilly RZ, Gallichio MH, Young CJ, Barker J, et al.: Effect of preoperative sonographic mapping on vascular access outcomes in hemodialysis patients. Kidney Int 60: 2013–2020, 2001. [DOI] [PubMed] [Google Scholar]
- 9.Kosa SD, Al-Jaishi AA, Moist L, Lok CE: Preoperative vascular access evaluation for haemodialysis patients. Cochrane Database Syst Rev 9: CD007013, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Umphrey HR, Lockhart ME, Abts CA, Robbin ML: Dialysis grafts and fistulae: Planning and assessment. Ultrasound Clin 6: 477–490, 2011 [Google Scholar]
- 11.Haładaj R, Wysiadecki G, Dudkiewicz Z, Polguj M, Topol M: The high origin of the radial artery (brachioradial artery): Its anatomical variations, clinical significance, and contribution to the blood supply of the hand. BioMed Res Int 2018: 1520929, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Robbin ML, Greene T, Cheung AK, Allon M, Berceli SA, Kaufman JS, et al.; Hemodialysis Fistula Maturation Study Group: Arteriovenous fistula development in the first 6 weeks after creation. Radiology 279: 620–629, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Robbin ML, Chamberlain NE, Lockhart ME, Gallichio MH, Young CJ, Deierhoi MH, et al.: Hemodialysis arteriovenous fistula maturity: US evaluation. Radiology 225: 59–64, 2002. [DOI] [PubMed] [Google Scholar]
- 14.Lee T, Magill M, Burke SK, Blair AT, Robbin ML, Allon M: Comparison of postoperative ultrasound criteria to predict unassisted use of arteriovenous fistulas for hemodialysis. J Vasc Access 19: 167–171, 2018 [DOI] [PubMed] [Google Scholar]
- 15.Farrington CA, Robbin ML, Lee T, Barker-Finkel J, Allon M: Postoperative ultrasound, unassisted maturation, and subsequent primary patency of arteriovenous fistulas. Clin J Am Soc Nephrol 13: 1364–1372, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Allon M, Bailey R, Ballard R, Deierhoi MH, Hamrick K, Oser R, et al.: A multidisciplinary approach to hemodialysis access: Prospective evaluation. Kidney Int 53: 473–479, 1998. [DOI] [PubMed] [Google Scholar]
- 17.United States Renal Data System : 2019 USRDS Annual Data Report: Epidemiology of Kidney Disease in the United States, Bethesda, MD, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2019 [Google Scholar]
- 18.Al-Balas A, Lee T, Young CJ, Barker-Finkel J, Allon M: Predictors of initiation for predialysis arteriovenous fistula. Clin J Am Soc Nephrol 11: 1802–1808, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lee T, Thamer M, Zhang Y, Zhang Q, Allon M: Outcomes of elderly patients after predialysis vascular access creation. J Am Soc Nephrol 26: 3133–3140, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lynch JR, Wasse H, Armistead NC, McClellan WM: Achieving the goal of the Fistula First breakthrough initiative for prevalent maintenance hemodialysis patients. Am J Kidney Dis 57: 78–89, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Harms JC, Rangarajan S, Young CJ, Barker-Finkel J, Allon M: Outcomes of arteriovenous fistulas and grafts with or without intervention before successful use. J Vasc Surg 64: 155–162, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lee T, Ullah A, Allon M, Succop P, El-Khatib M, Munda R, et al.: Decreased cumulative access survival in arteriovenous fistulas requiring interventions to promote maturation. Clin J Am Soc Nephrol 6: 575–581, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Allon M, Robbin ML: Increasing arteriovenous fistulas in hemodialysis patients: Problems and solutions. Kidney Int 62: 1109–1124, 2002. [DOI] [PubMed] [Google Scholar]
- 24.Allon M: Current management of vascular access. Clin J Am Soc Nephrol 2: 786–800, 2007. [DOI] [PubMed] [Google Scholar]
- 25.Chan C, Ochoa CJ, Katz SG: Prognostic factors for arteriovenous fistula maturation. Ann Vasc Surg 49: 273–276, 2018. [DOI] [PubMed] [Google Scholar]
- 26.Pisoni RL, Zepel L, Fluck R, Lok CE, Kawanishi H, Süleymanlar G, et al.: International differences in the location and use of arteriovenous accesses created for hemodialysis: Results from the dialysis outcomes and practice patterns study (DOPPS). Am J Kidney Dis 71: 469–478, 2018. [DOI] [PubMed] [Google Scholar]
- 27.Bashar K, Clarke-Moloney M, Burke PE, Kavanagh EG, Walsh SR: The role of venous diameter in predicting arteriovenous fistula maturation: When not to expect an AVF to mature according to pre-operative vein diameter measurements? A best evidence topic. Int J Surg 15: 95–99, 2015. [DOI] [PubMed] [Google Scholar]
- 28.Lauvao LS, Ihnat DM, Goshima KR, Chavez L, Gruessner AC, Mills JL Sr.: Vein diameter is the major predictor of fistula maturation. J Vasc Surg 49: 1499–1504, 2009. [DOI] [PubMed] [Google Scholar]
- 29.Mendes RR, Farber MA, Marston WA, Dinwiddie LC, Keagy BA, Burnham SJ: Prediction of wrist arteriovenous fistula maturation with preoperative vein mapping with ultrasonography. J Vasc Surg 36: 460–463, 2002. [DOI] [PubMed] [Google Scholar]
- 30.Lockhart ME, Robbin ML, Fineberg NS, Wells CG, Allon M: Cephalic vein measurement before forearm fistula creation: Does use of a tourniquet to meet the venous diameter threshold increase the number of usable fistulas? J Ultrasound Med 25: 1541–1545, 2006. [DOI] [PubMed] [Google Scholar]
- 31.Dageforde LA, Harms KA, Feurer ID, Shaffer D: Increased minimum vein diameter on preoperative mapping with duplex ultrasound is associated with arteriovenous fistula maturation and secondary patency. J Vasc Surg 61: 170–176, 2015. [DOI] [PubMed] [Google Scholar]
- 32.Feldman HI, Joffe M, Rosas SE, Burns JE, Knauss J, Brayman K: Predictors of successful arteriovenous fistula maturation. Am J Kidney Dis 42: 1000–1012, 2003. [DOI] [PubMed] [Google Scholar]
- 33.Shiu YT, Litovsky SH, Cheung AK, Pike DB, Tey JC, Zhang Y, et al.: Preoperative vascular medial fibrosis and arteriovenous fistula development. Clin J Am Soc Nephrol 11: 1615–1623, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Miller PE, Tolwani A, Luscy CP, Deierhoi MH, Bailey R, Redden DT, et al.: Predictors of adequacy of arteriovenous fistulas in hemodialysis patients. Kidney Int 56: 275–280, 1999. [DOI] [PubMed] [Google Scholar]
- 35.Miller CD, Robbin ML, Allon M: Gender differences in outcomes of arteriovenous fistulas in hemodialysis patients. Kidney Int 63: 346–352, 2003. [DOI] [PubMed] [Google Scholar]
- 36.Ernandez T, Saudan P, Berney T, Merminod T, Bednarkiewicz M, Martin PY: Risk factors for early failure of native arteriovenous fistulas. Nephron Clin Pract 101: c39–c44, 2005. [DOI] [PubMed] [Google Scholar]
- 37.Peterson WJ, Barker J, Allon M: Disparities in fistula maturation persist despite preoperative vascular mapping. Clin J Am Soc Nephrol 3: 437–441, 2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Farber A, Imrey PB, Huber TS, Kaufman JM, Kraiss LW, Larive B, et al.; HFM Study Group: Multiple preoperative and intraoperative factors predict early fistula thrombosis in the Hemodialysis Fistula Maturation Study. J Vasc Surg 63: 163–170.e6, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Allon M: Vascular access for hemodialysis patients: New data should guide decision making. Clin J Am Soc Nephrol 14: 954–961, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Saran R, Elder SJ, Goodkin DA, Akiba T, Ethier J, Rayner HC, et al.: Enhanced training in vascular access creation predicts arteriovenous fistula placement and patency in hemodialysis patients: Results from the dialysis outcomes and practice patterns study. Ann Surg 247: 885–891, 2008. [DOI] [PubMed] [Google Scholar]
- 41.Shahinian VB, Zhang X, Tilea AM, He K, Schaubel DE, Wu W, et al.: Surgeon characteristics and dialysis vascular access outcomes in the United States: A retrospective cohort study. Am J Kidney Dis 75: 158–166, 2020. [DOI] [PubMed] [Google Scholar]
- 42.Dixon BS: Why don’t fistulas mature? Kidney Int 70: 1413–1422, 2006. [DOI] [PubMed] [Google Scholar]
- 43.Hu H, Patel S, Hanisch JJ, Santana JM, Hashimoto T, Bai H, et al.: Future research directions to improve fistula maturation and reduce access failure. Semin Vasc Surg 29: 153–171, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Roy-Chaudhury P, Spergel LM, Besarab A, Asif A, Ravani P: Biology of arteriovenous fistula failure. J Nephrol 20: 150–163, 2007. [PubMed] [Google Scholar]
- 45.Falk A: Maintenance and salvage of arteriovenous fistulas. J Vasc Interv Radiol 17: 807–813, 2006. [DOI] [PubMed] [Google Scholar]
- 46.Lee T, Barker J, Allon M: Comparison of survival of upper arm arteriovenous fistulas and grafts after failed forearm fistula. J Am Soc Nephrol 18: 1936–1941, 2007. [DOI] [PubMed] [Google Scholar]
- 47.Woodside KJ, Bell S, Mukhopadhyay P, Repeck KJ, Robinson IT, Eckard AR, et al.: Arteriovenous fistula maturation in prevalent hemodialysis patients in the United States: A national study. Am J Kidney Dis 71: 793–801, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Lee T, Qian J, Thamer M, Allon M: Tradeoffs in vascular access selection in elderly patients initiating hemodialysis with a catheter. Am J Kidney Dis 72: 509–518, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Shingarev R, Barker-Finkel J, Allon M: Natural history of tunneled dialysis catheters placed for hemodialysis initiation. J Vasc Interv Radiol 24: 1289–1294, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]