Vascular Access–Related Distal Ischemia Requiring Intervention

Visual Abstract

Abstract

Background

Distal ischemia is a rare complication in patients undergoing placement of an arteriovenous (AV) fistula or AV graft. There are limited studies on its frequency, risk factors, clinical consequences, or feasibility of subsequent access.

Methods

A prospective vascular access database from a large academic medical center was queried retrospectively to identify 1498 patients (mean age 56±15 years, 48% female patients, 73% Black patients) undergoing placement of at least one vascular access from 2011 to 2020. For patients who developed access-related distal ischemia requiring surgical intervention, we determined the frequency of distal ischemia, clinical risk factors, and subsequent outcomes.

Results

Severe access-related distal ischemia occurred in 28 patients (1.9%; 95% confidence interval, 1.3% to 2.7%). The frequency was 0.2% for forearm AV fistulas, 0.9% for upper arm AV fistulas, 2.4% for forearm AV grafts, 2.2% for upper arm AV grafts, and 2.8% for thigh AV grafts. Risk factors independently associated with distal ischemia included female sex (odds ratio [OR], 3.64 [95% confidence interval, 1.52 to 8.72]), peripheral vascular disease (OR, 6.28 [2.84 to 13.87]), and coronary artery disease (OR, 2.37 [1.08 to 5.23]). Surgical interventions included ligation, excision, plication (banding), and other surgical procedures. Five patients developed tissue necrosis. A subsequent AV graft was placed in 13 patients, of whom only one (8%) developed distal ischemia requiring intervention.

Conclusions

Access-related distal ischemia requiring intervention was rare in this study and more common in women and patients with peripheral vascular disease or coronary artery disease. In some cases, a subsequent vascular access could be placed with a low likelihood of recurrent distal ischemia.

Introduction

Hemodialysis patients frequently undergo placement of a vascular access that can deliver blood into the extracorporeal circulation. There are two types of vascular access. An arteriovenous (AV) fistula is created by performing a direct surgical anastomosis between a native artery and vein. An AV graft is placed by interposing a synthetic hollow tube with one end inserted into an artery and the other into a vein. In either case, a portion of the blood flow that would ordinarily perfuse the distal limb is instead shunted into the AV fistula or graft. In rare cases, this may compromise perfusion of the extremity to produce symptomatic distal ischemia (also referred to as a steal syndrome).¹ A sharp decrease in the pressure across the AV anastomosis may exacerbate distal ischemia.² The range of symptoms can vary from mild (cool extremity with paresthesias) to moderate (intermittent pain aggravated by hemodialysis) to severe (persistent pain, motor weakness, or tissue necrosis).¹ The onset of distal ischemia can range from immediate (within days) to delayed (months to years). Whereas mild symptoms often improve with conservative management, severe symptoms place the patient at risk of tissue loss and require urgent surgical intervention. The most definitive intervention is excision or ligation of the vascular access, which rapidly alleviates the symptoms but results in abandonment of the vascular access and dependence on a central venous catheter (CVC). In some patients, less extreme surgical interventions are directed at either increasing the inflow of blood to the distal extremity or decreasing the outflow of blood into the vascular access, which in many cases can alleviate the distal ischemia while permitting the access to be salvaged.

There are limited (primarily retrospective) studies assessing the frequency of access-related distal ischemia, its timing, and its risk factors.^3–10 The only prospective study of this topic documented distal ischemia in 7% of patients undergoing AV fistula placement, with 4% having ischemia severe enough to require an intervention.¹⁰ There have been no prospective studies quantifying distal ischemia after AV graft placement. Several of these publications have documented the feasibility of salvaging the AV fistula or graft in a small number of patients with access-related distal ischemia by appropriate surgical interventions. Finally, the safety of placing a subsequent vascular access in patients with a previous episode of access-related distal ischemia is clinically relevant because the patient would otherwise remain catheter-dependent indefinitely. Unfortunately, there have not been any publications addressing this clinical question.

This study retrospectively analyzed a large, prospective, computerized vascular access database to quantify the frequency, timing, risk factors, clinical consequences of access-related distal ischemia requiring an intervention, and the safety of placing a subsequent vascular access.

Methods

Management of Vascular Access

The University of Alabama at Birmingham oversees the care of approximately 600 dialysis patients in metropolitan Birmingham, Alabama. Patients with advanced CKD or those dialyzing with a CVC were routinely referred to University of Alabama at Birmingham surgeons for placement of an AV fistula or graft. All patients underwent standardized preoperative sonographic vascular mapping to quantify arterial and venous diameters and exclude vascular stenosis or thrombosis.^11,12 At a minimum, placement of an AV fistula required an arterial diameter ≥2 mm, a venous diameter ≥2.5 mm, absence of stenosis or thrombosis in the draining vein, and a normal Allen test. An AV graft required a minimal venous diameter of 4 mm. The decision about the access type and its location was determined after consideration of both the preoperative mapping measurements as well as patient characteristics associated with a lower likelihood of AV fistula maturation. Most AV fistulas were placed in the upper arm (brachiocephalic or brachiobasilic), with forearm AV fistulas reserved primarily for patients at low risk of AV fistula nonmaturation (younger men with low cardiovascular comorbidity). AV grafts were placed in preference to AV fistulas if the clinicians believed an AV fistula was unlikely to mature in that patient. AV fistula anastomosis was typically 4–6 mm in length. AV graft anastomosis length was typically 4–7 mm on the arterial end and 10–15 mm on the venous end.

Patients were routinely seen postoperatively by the surgeons to evaluate access patency, potential complications, and the need for subsequent access interventions. At a minimum, patients receiving an AV fistula were scheduled for routine postoperative visits 1–2 and 4–6 weeks after access placement and more frequently if they developed a problem with the AV fistula. Patients receiving an AV graft were scheduled for a single postoperative visit 1–2 weeks after access placement but more frequently if there was a problem with the AV graft. Initial cannulation of AV fistulas was typically attempted 6–8 weeks postoperatively if the AV fistula had matured adequately, but later if the fistula required a subsequent intervention to promote its maturation. AV grafts were initially cannulated 2–4 weeks after their placement. Once an AV fistula or graft was successfully used for dialysis, subsequent surgical referrals were initiated by the dialysis nurses, rounding nephrologists, or nurse practitioners.

Distal ischemia was suspected whenever a patient developed pain or paresthesia distal to the access, a weak or absent radial pulse, decreased motor strength, or evidence of tissue necrosis. The diagnosis was confirmed in many cases by improvement of the symptoms and radial pulse after compression of the AV fistula or graft. Some patients underwent ancillary studies, including ultrasound, angiogram, or nerve conduction studies. Mild cases of distal ischemia were treated conservatively with frequent postoperative assessments and often resolved without a surgical intervention. Patients with more severe distal ischemia underwent a variety of surgical interventions, including ligation, excision, plication (banding), proximalization of arterial inflow, or distal revascularization and interval ligation (DRIL).

Two full-time vascular access coordinators acted as liaisons among the nephrologists, surgeons, and radiologists; scheduled the access procedures; and maintained a prospective, computerized vascular access database.¹³

Data Analysis

The local institutional review board approved performance of this study, and it exempted a requirement for obtaining consent forms. This study adhered to the Declaration of Helsinki. The prospective vascular access database was queried retrospectively to identify 1516 patients undergoing at least one vascular access placement during the 10-year period from January 1, 2011, to December 31, 2020. After omitting 18 patients with no postoperative visits, we were left with 1498 patients who formed the present study cohort (Figure 1). Finally, we queried the access database to identify 28 patients who developed severe access-related distal ischemia requiring an intervention. Owing to the retrospective nature of this study, we were unable to identify patients with access-related distal ischemia who did not undergo an intervention. The electronic medical record was used to ascertain the demographics (age, sex, and self-reported race) and comorbidities (diabetes, hypertension, coronary artery disease, heart failure, cerebrovascular disease, and peripheral vascular disease) for the entire cohort. There were no missing data for patient demographics or comorbidities. The electronic medical record was also used to determine for the patients who developed severe distal ischemia in their clinical presentation, type and timing of surgical intervention, and subsequent clinical outcomes. Finally, we identified a subset of patients who developed severe distal ischemia and underwent a subsequent vascular access placement.

Figure 1.

Flow chart summarizing inclusion and exclusion of potential participants. Figure 1 can be viewed in color online at www.cjasn.org.

The exposure of interest was placement of an AV fistula or graft, and the outcome of interest was the development of access-related distal ischemia requiring an intervention. We calculated the frequency of this complication for the entire study cohort, as well as its frequency for patients with each type (AV fistula or graft) and location (forearm, upper arm, or thigh) of vascular access. Because many patients underwent more than one vascular access placement, the primary analysis used unique patients. For a secondary analysis of distal ischemia associated with a specific type and location of access, only the first such access for each patient was used. For example, if a patient underwent placement of two different upper arm AV grafts, only the first one was included in the analysis.

Unadjusted logistic regression was used to determine the association of baseline demographic and clinical parameters with the development of access-related distal ischemia requiring intervention. Multiple variable logistic regression analysis was used to determine which baseline variables were independently associated with development of severe distal ischemia requiring a surgical intervention. The assumptions for the logistic model were straightforward, in that each of the variables evaluated was binary. The overall model included all the risk covariates, and some were excluded using backwards elimination (P > 0.10) to find the simplest model. The model was then repeated using forward stepwise logistic regression resulting in the same selection of variables (using P < 0.10). Separate survival curves were generated to plot the time from AV fistula or graft placement to the development of severe distal ischemia requiring an intervention. The log-rank test was used to calculate the statistical difference between the survival curves. A P value < 0.05 was considered statistically significant.

Results

During the 10-year study period, 1498 unique patients underwent one or more vascular access placements. Of these, 28 patients (or 1.9%) developed distal ischemia severe enough to require a surgical intervention. The baseline demographic characteristics of patients with or without access-related distal ischemia are summarized in Table 1. Patients developing distal ischemia requiring intervention were significantly more likely to be female (75% versus 47%) and to have diabetes (75% versus 56%), coronary artery disease (50% versus 26%), and peripheral vascular disease (50% versus 12%). The two groups were similar in age, race, and frequency of hypertension, heart failure, and cerebrovascular disease.

Table 1.

Characteristics of patients undergoing vascular access placement at the University of Alabama according to subsequent development of severe distal ischemia

Parameter	Patients with Distal Ischemia	Patients without Distal Ischemia
Total patients	28	1470
Age, yr	58±13	56±15
Black race, N (%)	23 (82)	1078 (73)
Female sex, N (%)	21 (75)	697 (47)
Diabetes, N (%)	21 (75)	830 (56)
Hypertension, N (%)	26 (93)	1336 (91)
Coronary artery disease, N (%)	14 (50)	377 (26)
Heart failure, N (%)	7 (25)	439 (30)
Cerebrovascular disease, N (%)	9 (32)	255 (17)
Peripheral vascular disease, N (%)	14 (50)	178 (12)

We then investigated the association of access-related distal ischemia requiring intervention with baseline demographic and clinical characteristics. The frequency of severe distal ischemia was 3% (21 of 718) in female patients, 7% (14 of 192) in patients with peripheral vascular disease, 12% (ten of 85) in women with peripheral vascular disease, and 21% (nine of 42) in women with peripheral vascular disease and coronary artery disease. On unadjusted logistic regression analysis, distal ischemia requiring intervention was associated with female sex (odds ratio [OR], 3.33 [95% confidence interval, 1.46 to 8.20]), diabetes (OR, 2.31 [1.02 to 5.70]), coronary artery disease (OR, 2.89 [1.33 to 6.20]), and peripheral vascular disease (OR, 7.26 [3.29 to 16.0]), but not with age, race, hypertension, heart failure, or cerebrovascular disease. Using multiple variable logistic regression analysis (backward elimination by forward selection), three factors were associated with severe access-related distal ischemia: female sex (OR, 3.64 [95% confidence interval, 1.52 to 8.72]), peripheral vascular disease (OR, 6.28 [2.84 to 13.87]), and coronary artery disease (OR, 2.37 [1.08 to 5.23]), P < 0.001 for the model. In a secondary analysis, which investigated the association of distal ischemia requiring intervention with baseline parameters in the subset of patients receiving an upper arm AV fistula or graft, distal ischemia was associated with female sex (OR, 2.95 [1.03 to 8.47]) and peripheral vascular disease (OR, 5.79 [2.26 to 14.83]), with a trend of association with an AV graft (2.33 [0.87 to 6.26]).

A tertiary analysis calculated the frequency of severe distal ischemia requiring intervention with different types and locations of vascular access, which was 0.2% for forearm AV fistulas, 0.9% for upper arm AV fistulas, 2.4% for forearm AV grafts, 2.2% for upper arm AV grafts, and 2.8% for thigh AV grafts (Table 2).

Table 2.

Frequency of access-related distal ischemia requiring an intervention

Group	Total N Patients	N with Distal Ischemia	Percent with Distal Ischemia, %	95% CI
All patients	1498	28	1.9	1.3 to 2.7
Patients with forearm AV fistula	392	1	0.2	0 to 1.4
Patients with upper arm AV fistula	668	6	0.9	0.4 to 1.9
Patients with forearm AV graft	84	2	2.4	0.4 to 7.9
Patients with upper arm AV graft	670	15	2.2	1.3 to 3.6
Patients with thigh AV graft	142	4	2.8	0.9 to 6.8

CI, confidence interval; AV, arteriovenous.

The 28 patients who developed severe access-related distal ischemia requiring intervention included 21 patients with an AV graft and seven with an AV fistula. The time interval from access placement to intervention for distal ischemia was compared in patients receiving an AV fistula or graft. The median time (interquartile range) was substantially shorter for AV grafts than for AV fistulas (19 [6–72] versus 383 [42–1191] days, P = 0.02). The presenting symptom was pain distal to the access (all patients) and variable other symptoms, including paresthesia (seven patients), coolness (seven patients), or motor weakness (12 patients). Five patients manifested evidence of tissue damage (ischemic ulcer or gangrene of digits). Among the 21 patients who developed severe distal ischemia after AV graft placement, a surgical intervention occurred in eight patients (38%) within 1 week, 12 patients (57%) within 1 month, and 15 patients (71%) within 3 months. The type of surgical intervention was an excision in seven patients, ligation in eight, and some type of surgical revision (to preserve access patency) in eight. Distal ischemia persisted after plication in three patients, who then required access ligation. Clinical outcomes after the surgical intervention for distal ischemia were variable. Whereas 21 patients had complete or partial resolution of their symptoms (pain, paresthesias, and motor weakness), three required an amputation (finger, toe, above-the-knee amputation). Four patients had persistent pain and paresthesia lasting for months to years despite surgical intervention, requiring chronic intake of narcotics or gabapentin.

Thirteen of the 28 patients with access-related distal ischemia requiring intervention (46%) subsequently underwent placement of an AV graft at a median time of 168 days (interquartile range, 91–256) after ligation or excision of their previous access. Notably, most of the subsequent accesses used more proximal (axillary or femoral) arteries than the ones used for the previous access that had produced distal ischemia. In three of these patients, the first access used the brachial artery, and the second one used the axillary artery of the same extremity. Only one of the 13 patients (8%) receiving a subsequent AV graft developed severe distal ischemia, and it required ligation at 6 weeks. In the other 12 patients, patency of the subsequent AV graft was maintained for periods ranging from 3 months to 5 years without developing symptoms of distal ischemia.

Discussion

This study highlights several important features of severe distal ischemia related to a vascular access. It is a relatively rare complication, which occurs more frequently and at an earlier time with AV grafts versus fistulas. Distal ischemia is more common in women and in patients with peripheral vascular disease or coronary artery disease. Surgical interventions usually, but not always, alleviate the ischemic symptoms. Finally, many patients with a prior access-related distal ischemia were able to receive a subsequent vascular access without recurrence of distal ischemia.

A large series of patients with access-related distal ischemia have been sparse,^3–10 and they have reported an incidence ranging between 2% and 7%, with a somewhat higher frequency in patients receiving an AV graft (Table 3). Of these, the only prospective study of distal ischemia was reported from the Hemodialysis Fistula Maturation Study, which enrolled 602 patients undergoing placement of an AV fistula.¹⁰ The frequency of distal ischemia in this study was 7%, but only 4% of the 602 patients required a surgical intervention.

Table 3.

Large series of distal ischemia after vascular access placement

Reference	Study Design	N Vascular Accesses	Frequency of Distal Ischemia	Risk Factors	Timing of Distal Ischemia
Morsy et al.,3 1998	Retrospective	110 AV fistulas 299 AV grafts	AV graft—4.3% AV fistula—1.8%	Peripheral vascular disease
Goff et al.,4 2000	Retrospective	1800 AV fistulas or grafts	1.7%
Davidson et al.,5 2003	Retrospective	162 AV fistulas 163 AV grafts	6.2%	Female, diabetes, AV graft
Lazarides et al.,6 2003	Retrospective	619 AV fistulas 343 AV grafts	2.9% with severe distal ischemia AV graft—5.2% Brachiocephalic AV fistula—3.4%		AV graft—2 d AV fistula—165 d
Scheltinga et al.,7 2009	Systemic review	Analysis of 21 studies			Acute—88% AV grafts Chronic—91% AV fistulas
Gupta et al.,8 2011	Retrospective	922	7.6% (severe)	Female, diabetes, peripheral vascular disease, coronary artery disease
Rocha et al.,9 2012	Retrospective	324 (all proximal AV fistulas)	8%	Female, diabetes
Huber et al.,10 2016	Prospective	602 AV fistulas	7% (4% severe)	Female, diabetes, coronary artery disease	63% >30 d
Allon (current study)	Retrospective (prospective access database)	1498	1.9% (severe)	Female, diabetes, peripheral vascular disease	AV grafts—19 d AV fistulas—383 d

Severe distal ischemia refers to patients requiring an intervention. AV, arteriovenous.

The published studies have consistently identified female sex, diabetes, peripheral vascular disease, and coronary artery disease as risk factors for developing distal ischemia after vascular access placement.^3,5,8–10 It is likely that diabetes does not directly lead to distal ischemia but is rather a surrogate marker for severe vascular disease, which in turn contributes to distal ischemia after vascular access placement.

Only a minority of the studies reported on the timing of access-related distal ischemia. In a report of 962 patients undergoing access surgery, of whom 28 patients developed distal ischemia, the median time to onset of symptoms was 2 days in patients receiving an AV graft, as compared with 165 days in those with an AV fistula.⁶ A prospective study of 602 patients receiving an AV fistula found that in 63% of cases, the distal ischemia occurred more than 30 days after AV fistula creation.¹⁰ In this study, the median time to distal ischemia was 19 days for patients receiving an AV graft, as compared with 383 days for those receiving an AV fistula. Finally, a systematic analysis of 21 studies reported that when the onset of distal ischemia was acute (<1 month), 88% of the patients had an AV graft; in contrast, when it was chronic (>1 month), 91% of cases had an AV fistula.⁷ The discrepancy in timing is likely related to the time course of the increase in blood flow through the access. The increase is immediate in patients receiving an AV graft, whereas it is slowly progressive over 6 weeks in patients receiving an AV fistula.¹⁴

The definitive solution to severe distal ischemia is to ligate (or excise) the vascular access. This maneuver immediately increases blood flow to the distal extremity, thereby alleviating the ischemia. Unfortunately, it also sacrifices the AV fistula or graft, thereby consigning the patient to dialysis with a CVC, with all its associated complications, including catheter-related bloodstream infections, frequent thrombosis, and central vein stenosis.¹⁵ Less extreme approaches have been devised to alleviate distal ischemia while maintaining patency of the vascular access. One approach is plication (or banding), in which flow to the AV fistula or graft is reduced, thereby increasing blood flow to the hand.^16,17 This approach can be tricky because excessive banding results in access thrombosis, whereas insufficient banding permits the distal ischemia to persist. Surgical options designed to increase distal flow include the distal revascularization and interval ligation (DRIL) procedure, revision using distal inflow, and proximalization of arterial inflow. In this study, 20 of 28 patients (71%) required ligation or excision, whereas in the remaining patients, other surgical approaches relieved the distal ischemia while preserving vascular access patency.

Previous publications have not reported on whether patients who developed distal ischemia with a vascular access were at higher risk of distal ischemia after a subsequent access. This question was addressed in this study, which identified 13 patients undergoing a subsequent AV graft placement. Only one of the 13 patients (8%) developed distal ischemia with the new AV graft. Of note, in most cases, the subsequent AV graft used a more proximal artery (axillary or femoral). Presumably, this ensured adequate blood flow to the distal extremity despite having a portion of it diverted to the AV graft. Although caution is indicated in generalizing from such a small sample, our experience suggests that a subsequent AV access may be safely placed in selected patients with previous access-related distal ischemia.

This study has several strengths. Because the vascular access data were collected prospectively, we are confident that they are comprehensive, accurate, and minimize potential bias. Second, the large number of patients with new vascular access placement during the 10-year study period provides a more precise estimate of the incidence of access-related distal ischemia requiring intervention, including the incidence associated with both AV fistulas and grafts at different anatomic locations. Third, to our knowledge, this is the only study to address the outcomes of subsequent accesses placed after a prior episode of access-related distal ischemia requiring intervention.

Our study also has some limitations. First, it was a single-center study, and the results may not generalize to all dialysis centers. Second, we were able to identify only the subset of patients with distal ischemia sufficiently severe to warrant a surgical intervention because our database did not capture patients with milder forms of access-related distal ischemia that improved without a surgical intervention. The only prospective study of access-related distal ischemia (limited to AV fistulas) suggests that milder cases may represent about half of all such cases.¹⁰ Third, some patients may have been lost to follow-up because of early death or transfer to an outside dialysis unit. To the extent that the diagnosis may have been missed in such patients (ascertainment bias), our retrospective analysis may have underestimated the true incidence of access-related distal ischemia.

Distal ischemia requiring surgical intervention is a rare complication of AV fistula and graft placement, which is more common in women and in patients with peripheral vascular disease or coronary artery disease. In most cases, surgical intervention resolves or alleviates distal ischemia. A subsequent vascular access placement is feasible in selected patients and may be associated with a low risk of recurrent distal ischemia.

Disclosures

M. Allon is the Editor-in-Chief of Kidney360. G.R. Cutter reports consultancy for AI therapeutics, AMO, AstraZeneca, Avexis, Bristol Meyers Squibb/Celgene, CSL Behring, DSMBs: Applied Therapeutics, Horizon, Immunic, Karuna, Kezar, Mapi, Merck, Mitsubishis, NHLBI (Protocol Review Committee), Novartis, Opko Biologics, Prothena, Reata, Regeneron, Sanofi-Aventis, Teva, University of Pennsylvania, University of Texas Southwestern, and Visioneering Technologies, Inc.; ownership interest in Pythagoras, Inc., a private consulting company; advisory or leadership role as a JASN Statistical editor; role on Multiple Sclerosis Journal Editorial Board and Multiple Sclerosis and Related Diseases Editorial Board; and Consulting or Advisory Boards for Alexion, Antisense Therapeutics, Avotres, Biogen, Clene Nanomedicine, Clinical Trial Solutions LLC, Entelexo Biotherapeutics, Inc., Genentech, Genzyme, GW Pharmaceuticals, Hoya Corporation, Immunic, Immunosis Pty Ltd, Klein-Buendel Incorporated, Linical, Merck/Serono, Novartis, Perception Neurosciences, Protalix Biotherapeutics, Regeneron, Roche, and SAB Biotherapeutics. The remaining author has nothing to disclose.

Funding

M. Allon: National Center on Minority Health and Health Disparities (1R01MD013818-01).

Author Contributions

Conceptualization: Michael Allon.

Data curation: Michael Allon.

Formal analysis: Michael Allon, Gary R. Cutter.

Writing – original draft: Michael Allon.

Writing – review & editing: Gary R. Cutter, Carlton J. Young.

Data Sharing Statement

Anonymized data created for the study are or will be available in a persistent repository on publication.