Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Aug 18.
Published in final edited form as: J Vasc Surg. 2012 Dec 12;57(2):451–458. doi: 10.1016/j.jvs.2012.08.105

Prediction of Graft Patency and Mortality after Distal Revascularization and Interval Ligation for Hemodialysis Access Related Hand Ischemia

Salvatore T Scali 1, Catherine K Chang 1, Dan Raghinaru 2, Michael Daniels 2, Adam W Beck 1, Robert J Feezor 1, Scott A Berceli 1, Thomas S Huber 1
PMCID: PMC4135397  NIHMSID: NIHMS616384  PMID: 23244784

Abstract

Objective

The treatment goals for access related hand ischemia (ARHI) are to reverse symptoms and salvage the access. Many procedures have been described, but the optimal treatment strategy remains unresolved. In an effort to guide clinical decision making, this study was undertaken to document our outcomes for distal revascularization and interval ligation (DRIL) and identify predictors of bypass patency and patient mortality.

Methods

A retrospective review was performed of all patients who underwent DRIL at the University of Florida from 2002–2011. Diagnosis of ARHI was based primarily upon clinical symptoms with non-invasive studies used to corroborate in equivocal cases. Patient demographics, procedure-outcome variables and re-interventions were recorded. Bypass patency and mortality were estimated using cumulative incidence and Kaplan-Meier methodology, respectively. Cumulative incidence and Cox regression analysis were performed to determine predictors of bypass patency and mortality, respectively.

Results

134 DRILs were performed in 126 patients (age 57±12yrs (mean±SD)) following brachial artery-based access. The post-operative complication rate was 27% (19%-wound), and 30-day mortality was 2%. The wrist/brachial (WBI) and digital/brachial (DBI) indices increased 0.31±0.25 and 0.25±0.29, respectively. Symptoms resolved in 82% of patients, and 85% continued to use their access. Cumulative incidences of loss of primary and primary-assisted patency rates were 5±2%, 4±2% and 22±5%, 18±5% at 1 and 5 years, respectively with a mean follow-up of 14.8 months. Univariate predictors of primary patency failure were DRIL complications (3.3; 1.2–8.9, p=0.02), configuration other than brachiobasilic/brachiocephalic autogenous access (3.4; 1.4–8.3, p=0.009), and ≥2 prior access attempts (4.1; 1.6–10.4, p=0.004). Brachiocephalic access configuration (0.2; 0.04–0.8, p =0.02), and autogenous vein conduit (0.2; 0.06–0.58, p=0.004) were predictors of improved bypass patency. All-cause mortality was 28% and 79% at 1 and 5 years, respectively. Multi-variable predictors of mortality were age > 40 (8.3; 2.5–33.3, p =0.0004), grade 3 ischemia (2.6; 1.5–4.6, p=0.0008), complication from DRIL (2.4; 1.3–4.5, p=0.004), and smoking history (2.2; 1.3–4, p=0.007). Patients with no prior access attempts had lower predicted mortality (0.5; 0.3–0.9, p=0.02).

Conclusion

The DRIL procedure effectively improves distal perfusion and reverses the symptoms of ARHI while salvaging the access, but the long-term survival of these patients is poor. Given the poor survival, pre-operative risk stratification is critical. Patients at high risk for DRIL failure and mortality may be best served with alternate remedial procedures.

Introduction

Access-related hand ischemia (ARHI), commonly known as ‘steal syndrome’, is one of the most challenging complications to manage after hemodialysis access construction. The creation of an arteriovenous fistula results in a predictable decrease in the arterial perfusion pressure distal to the fistula that can lead to ischemia if the compensatory mechanisms are inadequate. The diagnosis of ARHI is largely clinical and can be aided in equivocal cases with noninvasive vascular laboratory studies1–3. ARHI occurs in approximately 5–20% of brachial artery-based access procedures with roughly half classified as severe and meriting some type of remedial treatment3–10. Treatment goals are to reverse the hand ischemia and preserve the access while avoiding any long-term hand disability.

A variety of remedial treatment strategies have been reported for ARHI including access ligation, banding (i.e. flow limiting strategies), bypass and proximalization of the arteriovenous anastomosis. The choice is contingent upon multiple factors including the severity of the symptoms, patient co-morbidities and the potential utility of the access itself. The distal revascularization and interval ligation procedure (DRIL) is our preferred treatment because it reverses the ischemic symptoms and salvages the access in approximately 90% of the cases11. However, several reports have raised concerns regarding the need for ligation of the brachial artery, thus leaving distal forearm and hand perfusion reliant upon a functioning brachial-brachial bypass12–14. Furthermore, the post-operative complication rates after the DRIL can range from 5–15%11, 12, 15 and primary patency at 2-years is estimated to be 70–80%8, 11, 12, 16. Given these potential issues, the decision regarding choice of remedial procedure must also consider the patient’s life-expectancy. The one year survival rate in our previous DRIL series was 70%, suggesting that the annual mortality rate for patients with ARHI may exceed the 20–23% annual rate reported in the United States for all patients on hemodialysis11, 17.

These observations and reported concerns about the DRIL procedure prompted this analysis in an effort to guide clinical decision making. This study was undertaken to evaluate our outcomes for the DRIL procedure and to identify predictors of bypass patency and patient mortality.

Methods

Experimental Design

All patients undergoing creation of an upper extremity hemodialysis access between 2002 and 2011 at the University of Florida were identified. A prospectively maintained database was queried, and those who underwent remedial procedures for ARHI were further studied (126 patients including the 61 patients previously described by Huber et al.11). Approval for this study was obtained from the Institutional Review Board.

Clinical Practice

A defined and validated algorithm to optimize the use of autogenous conduit access18 was followed by all surgeons. All patients underwent non-invasive arterial and venous imaging preoperatively, and confirmatory invasive arterial and/or venous angiography was used in selected cases. Hierarchy for the access configurations followed the recommendations of the KDOQI/SVS guidelines and our previously published algorithm19, 20.

The diagnosis of ARHI was primarily a clinical assessment of the operating surgeon. Selected patients had non-invasive arterial testing, which included brachial, radial, ulnar, and digital pressures and velocity waveform analysis. These data were used to confirm ARHI diagnosis in equivocal cases. The management of patients with mild (Grade 1) ARHI was expectant, while definitive remediation was reserved for patients with moderate (Grade 2) or severe (Grade 3) ischemia (see Definitions below). Upper extremity digital subtraction arteriography was used in the majority of cases to determine if there was a correctable arterial inflow stenosis. Decision to proceed with DRIL followed a published algorithm for management of ARHI21.

Conduct of the DRIL procedure was as previously described22. The proximal anastomosis of the brachial-brachial bypass was created ≥7cm proximal to the arteriovenous anastomosis of the access. The distal anastomosis was constructed immediately distal to the access anastomosis, and the brachial artery was ligated or transected immediately proximal to the distal bypass anastomosis. The preferred conduit was greater saphenous vein (≥3mm) with alternative conduit choices based on conduit availability (arm vein>femoral vein>cadaveric vein>prosthetic graft). All attempts were made to preserve upper extremity veins for potential future access creation. Doppler insonation and palpation of the radial/ulnar pulse were used to determine technical adequacy of the bypass. Completion arteriography or intra-operative duplex were used at case completion in selected cases at the operating surgeon’s discretion. Post-operatively, patients were placed on aspirin (unless contraindicated) and evaluated with a standardized surveillance protocol. This involved upper extremity pressure/waveform analysis and duplex scanning of the brachial-brachial bypass as previously described for our lower extremity bypasses11, 23. Post-operative duplex surveillance of the DRIL occurred at 1, 3, 6, 9 and 12 months and every 6 months, thereafter. Re-intervention was based on the presence of recurrent hand symptoms, significant decrease in arterial pressures (≥ 15mmHg drop in arterial wrist pressure) and/or abnormal graft scan (mean graft velocity <50cm/s, maximum velocity ratio >3.5)24.

Definitions and Data Analysis

Co-morbidities were defined and retrospectively recorded as follows: coronary artery disease (any history of myocardial infarction [MI], angina, prior coronary intervention, or ECG changes consistent with prior MI); cerebrovascular occlusive disease (history of transient ischemic attack, stroke, and/or prior carotid endarterectomy/stent/angioplasty); congestive heart failure (chart history, New York Heart Association II or greater, diagnosis on pre-operative evaluation); chronic obstructive pulmonary disease (chart history or pre-operative pulmonary function testing consistent with the diagnosis, medication); diabetes mellitus (chart history, insulin, oral hypoglycemics); peripheral artery disease(ABI < 0.9, chart history, prior peripheral endovascular intervention or open infrainguinal reconstruction); hypertension (chart history, anti-hypertensive medications or pre-operative blood pressure ≥ 140/90mmHg); dyslipidemia (chart history, on cholesterol-lowering medications); smoking history (any prior or current smoking). Data collection also included pre-operative anti-platelet (aspirin, aspirin+dipyridamole, or clopidogrel), anti-coagulant (warfarin), and statin medication use.

DRIL bypass patency was objectively assessed with duplex ultrasound and/or selective arteriography. Patient mortality was verified by query of the Social Security Administration’s Death Master File. Pre-operative and post-operative wrist/brachial indices (WBIs) and digital/brachial indices (DBIs) were compared with a paired t test and a P value <0.05 was accepted as significant.

ARHI was categorized using the SVS reporting standards as mild (Grade 1), moderate (Grade 2), or severe (Grade 3)25. Patients categorized as Grade 2 or 3 ischemia presented with a spectrum of symptoms and physical exam findings. The predominant (most severe) symptom was recorded as the indication for the DRIL (tissue loss > motor dysfunction > rest pain > intermittent, hemodialysis-dependent rest pain > paresthesia > coolness). Grade 2 ischemia was defined by presence of hand coolness, numbness/paresthesia or intermittent hand pain while on hemodialysis. Grade 3 ischemia was defined as rest pain (including while off a hemodialysis circuit), neuromotor hand dysfunction or presence of ulceration/gangrene.

Risk Model Design

The outcomes of interest were time to loss of primary or primary assisted patency, time to secondary intervention, and time to death. The distribution of mortality was estimated using the Kaplan-Meier method. The distributions of the loss of patency outcomes were summarized utilizing cumulative incidence functions due to the presence of mortality as a competing risk with occurrence of DRIL failure26, 27. Cumulative incidence regression, rather than Cox proportional hazards regression, was utilized to study predictors of primary DRIL patency, as mortality and graft failure were considered to be competing risks28. In the presence of competing risks, Kaplan Meier estimates are inaccurate as this methodology assumes all events are independent, i.e. that patients who die following DRIL are still at risk of graft failure, which is clearly false.

For mortality, all possible models with up to 5 risk factors were considered, and the model with the best measure of the relative goodness of fit (defined as lowest AIC or “an information criterion” taking into account the number of risk factors and the fit of a model) was selected. The maximum number of risk factors considered in the multivariable model was limited to 5 as the number of mortality events were only 60. No multivariable models for primary patency were considered given that there were only 14 events. Statistical analysis was performed using SAS 9.2 (Cary, North Carolina, USA) and R 2.15.0 (http://www.r-project.org/).

Results

From 2002 to 2011, 2753 access-related procedures were performed (excluding dialysis catheters), of which 1882 were new access creations. Of these new access procedures, the overall incidence of Grade 2 or 3 ARHI requiring remediation was 7.8%. The distribution of permanent access creations are noted in Figure 1, with the majority based off the brachial artery. Four patients underwent proximalization, 17 had access ligation, and 126 underwent a DRIL procedure. Of patients with an autogenous brachial-cephalic or brachial-basilic upper arm access, 8% underwent DRIL compared to 11% of those with brachial-axillary indirect autogenous or cadaveric femoral vein translocation. In contrast, only 1% with a prosthetic brachial-axillary access configuration underwent DRIL, and no patients with autogenous radial-cephalic direct wrist access required remediation for ARHI.

Figure 1.

Figure 1

Open in a new tab

Absolute numbers and proportions of new access configurations performed between 2002 to 2011.

A total of 134 DRIL operations were performed on 126 patients with an average age of 57±12 years (mean±SD). The mean total time on hemodialysis prior to presenting with ARHI requiring remediation was 26±41 months. The majority of the patients were female (59%) and had a pre-operative diagnosis of diabetes (69%). Other demographic and access-related surgical history is outlined in Table 1. Thirteen percent of the patients undergoing a DRIL had ≥ 2 prior access attempts (defined as any prior access creation operations), and 15% of patients had a prior clinical diagnosis of ARHI (with or without remediation). Table 2 highlights the distribution of pre-operative access configurations and pre-operative medication use. Sixteen percent of the DRIL patients had an autogenous brachial-axillary indirect femoral vein translocation19 and only 2% had a prosthetic brachial-axillary access. Prior to the DRIL procedure, only 10% of patients were not on any anticoagulation or anti-platelet regimen. More than half (53%) of the patients were on a statin medication at the time of their procedure.

Table 1. Patient Demographics and Comorbid Conditions.

Patient demographic information, co-morbid conditions, and prior access history (including ARHI).

Demographics N = 126
  Age(±SD) 57112
  Gender (% female) 59%
Co-morbidities
  Hypertension 95%
  Diabetes 69%
  Dyslipidemla 55%
  Coronary artery disease 49%
  Smoking 46%
  Congestive heart failure 21%
  Prior hand ischemia 15%
  Prior access attempts (≥2) 13%
Open in a new tab

Table 2. DRIL Patient Access Configurations and Pre-operative Medications.

Distribution of access configurations of patients who underwent DRIL (in %). Preoperative anti-platelet, anti-coagulation, and statin use among DRIL patients.

Access Configurations N=126
  Autogenous brachial-cephalic upper arm direct access 46%
  Autogenous brachial-basilic upper arm transposition 36%
  Autogenous/cadaveric brachial-axillary indirect femoral vein 16%
  Prosthetic brachial-axillary access 2%
Pre-operative Medication Use
  Aspirin 84%
  Statin 53%
  No anti-coagulation/anti-platelet 10%
  Clopidogrel 20%
  Warfarin 14%
Open in a new tab

Many patients presented with multiple symptoms and physical exam findings consistent with ARHI. The most severe or predominant indication for DRIL is outlined in Figure 2. Sixty nine percent of patients had Grade 3 ischemia. The average time from index access creation to DRIL was 82±153 days (Figure 3). A planned, pre-emptive DRIL (performed at time of index access procedure) was performed in 8 (6%) patients due to a prior history of ARHI (N=6) or known severe, distal forearm occlusive disease (N=2). The conduit use for the DRIL brachial-brachial artery bypass was the following: greater saphenous vein in 75% of cases, arm vein 18%, composite autogenous vein 3%, cadaveric vein or artery 3%, and prosthetic graft 1%.

Figure 2.

Figure 2

Open in a new tab

Graphical depiction of the predominant indication for DRIL procedure. A few patients underwent DRIL at time of index access creation due to prior ARHI and are not represented in this graph.

Figure 3.

Figure 3

Open in a new tab

Graphical depiction of performance of DRIL procedure in days after the index access creation.

During the DRIL procedure, 5.5% (N=7) had adjunctive procedures performed (subclavian angioplasty and/or stent, N=5; central vein venoplasty, N =2). Average post-operative length of stay was 4.2±4.8 days. The overall composite post-operative procedure complication rate was 27%, with the majority attributable to wound infection (Table 3). Thirty day mortality was 2% and mean follow-up is 14.8±17.6 months (median 7 months; range 0–81 months). Significant increases in mean wrist/brachial (WBI) and digital/brachial (DBI) indices were detected between the preoperative and postoperative measurements: WBI, 0.31±0.25 (P=0.02); DBI, 0.25±0.29 (P=0.03) (Figure 4). Symptoms fully resolved in 82% of patients (Figure 5), and 85% continued to use the index hemodialysis access for which the DRIL was performed at time of last follow-up.

Table 3. Morbidity and Mortality after DRIL.

30 day morbidity and mortality following DRIL.

Morbidity 27%
  Wound 19%
  Peripheral nerve 3%
  Cardiac 2%
  Gastrointestinal 2%
  Cerebrovascular 1%
30-day Mortality 2%
Open in a new tab

Figure 4.

Figure 4

Open in a new tab

Mean pre-operative (pre-op) and post-operative (post-op) wrist/brachial (WBI) and digital/brachial (DBI) indices are shown. Significant increases (P < 0.05) were noted for both indices following DRIL*. Pre-operative hemodynamic data was available in 75% (N=95) of cases, while 68% (n=86) had at least one postoperative duplex scan of the graft with WBI/FBI measurements.

Figure 5.

Figure 5

Open in a new tab

Depiction of proportion of patients with complete symptom resolution following DRIL and of patients with residual paresthesia, motor dysfunction, or tissue loss following DRIL.

Patency Results and Prediction

Cumulative incidences of loss of primary and primary-assisted patencies of the DRIL bypass (±standard error of the mean) were 5±2% and 4±2% at 1 year, and 22±5% and 18±5% at 5 years (Figure 6). Univariate predictors of primary patency failure are noted in Table 4. No multivariable analysis for DRIL patency was attempted due to the low number of patency failure events.

Figure 6.

Figure 6

Open in a new tab

Cumulative incidences of the loss of primary patency of the DRIL bypass are shown. The standard errors were <10% throughout the time interval represented. Numbers of patients at risk are shown beneath the curve.

Table 4. Uni-variate Predictors of Loss of DRIL Primary Patency.

Cumulative incidence regression1 was used to determine uni-variate associations between patient co-variates and DRIL bypass primary patency. Cumulative incidence regression was utilized due to the competing risk of patient mortality. Non-autogenous brachial-cephalic/brachial-basilic access includes configurations of autogenous indirect femoral vein translocation brachial-axillary access, cadaveric femoral artery/vein brachial-axillary access, and prosthetic brachial-axillary access.2

Predictor1 H.R. C.I. P value
≥2 prior access creations 4.1 1.6–10.4 0.004
Non autogenous brachial-cephalic/brachial-basilic access2 3.4 1.4–8.3 0.009
Complication from DRIL 3.3 1.2–8.9 0.02
Autogenous vein conduit 0.2 0.06–0.58 0.004
Autogenous brachial-cephalic upper arm direct access 0.2 0.04–0.8 0.02
Open in a new tab
1

Cumulative incidence regression

2

Includes a utogenous/cadaveric brachial-axillary indirect femoral vein translocation, prosthetic brachial-axillary access

DRIL Re-intervention and Outcome

Median time to any re-intervention (including the repeat DRIL procedures) was 9.4 months [range 0.2–16.4 months], and cumulative incidence of re-intervention was 15±6% at one year and 21±7% at 5 years. Three patients had primary-assisted patency events occurring at 1 (subclavian stent), 61 (angioplasty of proximal anastomosis), and 367 (angioplasty of distal anastomosis) days after the initial DRIL operation. Of the 11 patients who had documented brachial-brachial bypass occlusion, 3 were asymptomatic, and eight patients underwent a second DRIL procedure at a median interval of 10.2 months [range 0.2–16.4 months]. No brachial-brachial bypass underwent graft thrombectomy or thrombolysis with revision (i.e. no secondary patency events). Of the 8 patients who underwent a second DRIL procedure, 6 (75%) were patent at time of this analysis. Two patients with a patent second DRIL had mild, residual paresthesia, and one with 2 failed DRIL operations had permanent neurologic hand dysfunction secondary to ischemic monomelic neuropathy that was retrospectively determined to be present prior to the initial DRIL occlusion. No minor or major amputations resulted from DRIL thrombosis.

Mortality Prediction

All-cause mortality was 28±5% at one year and 79±6% at 5 years (Figure 7). Multivariable predictors of mortality are listed in Table 5. Predictors of mortality after DRIL were age > 40 years, Grade 3 ischemia, any complication after DRIL, and any smoking history. If patients had no prior hemodialysis access attempts before the index access creation (for which the DRIL operation was performed), improved survival was predicted.

Figure 7.

Figure 7

Open in a new tab

The Kaplan Meier curve for the patient survival is shown. The standard errors were <10% throughout the time interval represented. Numbers of patients at risk are shown beneath the curve.

Table 5. Predictors of Mortality after DRIL.

Independent predictor of all cause patient mortality after DRIL was determined using multi-variable Cox proportional hazard regression analysis.

Predictor1 H.R. C.I. P value
Age >40 years 8.3 2.5–33.3 0.0004
Grade 3 ischemia 2.6 1.5–4.6 0.0008
Complication from DRIL 2.4 1.3–4.5 0.004
Smoking history (past/current) 2.2 1.3–4 0.007
No prior access procedures 0.5 0.3–0.9 0.02
Open in a new tab
1

Cox proportional hazards regression

Discussion

The results of this study highlight the safety and efficacy of the DRIL procedure for management of ARHI and elucidate risk factors that impact bypass patency and predict patient mortality. The benefits of the DRIL in relieving the symptoms of ARHI while simultaneously preserving the access for hemodialysis are substantial, as greater than 80% of patients had complete symptom relief and were still utilizing their access at time of last follow-up. These findings are corroborated by the demonstrated hemodynamic benefits, with significant improvements in both WBI and DBI after DRIL.

The hemodynamic basis for the DRIL procedure is the low resistance arterial bypass that overcomes the high resistance collateral circulation and the ligation that prevents retrograde flow from the distal vessels through the fistula. Interestingly, the components that afford the hemodynamic advantage (i.e. arterial bypass, ligation) have been also cited as limitations of this procedure. Concerns have been raised of the possibility of catastrophic outcomes with DRIL thrombosis due to the bypass-dependent hand perfusion. The patency rates demonstrated in our series are quite good, with a 5-year primary-assisted patency rate > 80%. Furthermore, of the 11 patients in our series who had DRIL thrombosis, none underwent amputation and only one had permanent hand dysfunction that predated the index DRIL procedure. Thus, the concerns for catastrophic complication with DRIL thrombosis may be unfounded. Unfortunately, the observed hemodynamic benefits of the DRIL procedure failed to relieve all of the precipitating neurological complaints as a small subset of patients had persistent paresthesias. We hypothesize that these patients likely had a severe, irreversible ischemic nerve injury prior to the DRIL29. It has been our anecdotal impression that this irreversible nerve injury can occur fairly quickly in terms of the time elapsed from access creation, and suggests importance in the timing of DRIL.

Several univariate associations with primary patency failure were found in this analysis. As noted above, it was impossible to evaluate the independent influence of each factor on DRIL patency due to the small number of events. However, these associations still merit discussion. The observations that 2 or more prior access creations and access configurations other than autogenous brachial-cephalic or brachial-basilic upper arm access were predictive of decreased DRIL patency may be due to several factors. The majority of these patients had a brachial-axillary indirect access configuration with either translocated (autogenous or cadaveric) femoral vein or, rarely, PTFE. This likely selected for a subgroup of patients dependent on hemodialysis for a longer duration and/or one with multiple prior failures, potentially due to the presence of more severe forearm arterial occlusive disease or poor vein graft remodeling.

The use of autogenous conduit was found to be protective of primary graft patency. The protective association of this factor is consistent with reports of its impact on lower extremity bypass patency30. We noted this association some time ago, and since 2009, we have largely abandoned the use of cadaveric conduit. Similar to lower extremity bypasses, the greater saphenous vein from the thigh is our preferred conduit, with ≥ 3 mm as our diameter criteria for a suitable vein18. We are reluctant to harvest the saphenous vein below the knee in patients with significant peripheral vascular disease, due to concerns with wound healing. In patients that do not have suitable saphenous vein, we have used the cephalic or basilic vein in select cases, attempting to balance the benefit of preserving the current access against the loss of a future access option.

The DRIL procedure has been our preferred choice for ARHI given our favorable results. However, there are several other alternatives that may play a role in specific scenarios. For patients that lack suitable autogenous conduit, proximalization of the arterial inflow (PAI) with PTFE may be a reasonable alternative to ligation31, 32. Others have also described a variant of the “proximalization” approach by complete mobilization of the venous outflow track (i.e. the entire cephalic vein from the antecubital fossa to the shoulder) and translocating this in a looped configuration with the anastomosis based on the proximal brachial artery to obviate the use of a prosthetic conduit10, 31. Our own experience with the PAI procedure has been limited, and we echo conclusions by Zanow et al.31 that it is likely ineffective in patients with severe tissue loss. We have not performed any revision with distal inflow (i.e. RUDI) procedures, but have been struck by the lower incidence of ARHI in access procedures performed using the proximal radial artery as opposed to the brachial artery at the antecubital fossa33. Our enthusiasm for the flow-limiting approaches is tempered by the inconsistent reports in the literature and the requisite, tenuous balance between adequate distal perfusion and sufficient access flow to sustain effective dialysis. However, the various flow limiting strategies may be effective for patients that have very high flow rates, particularly those with cardiac dysfunction34.

The management of ARHI really begins during the initial evaluation prior to the index access procedure. All available strategies should be employed to reduce this adverse outcome including originating the arteriovenous anastomosis as distal on the arterial tree as possible, and confirming absence of an arterial inflow stenosis. We follow the KDOQI/SVS guidelines35 and our own well-defined algorithm to optimize the use of autogenous conduit for access creation18 that prioritize radial artery over brachial artery inflow. Patients are evaluated with both arterial and venous non-invasive duplex imaging in contrast to many practices that just focus on the quality of the vein. In addition, for patients estimated to be at particularly high risk for ARHI, (such as those with a previous history of hand ischemia or documented evidence of severe forearm occlusive disease and cases using large diameter, compliant conduit such as translocated autogenous femoral vein), we have performed simultaneous access creation and DRIL procedures. In other high risk patients, we have obtained a pre-operative saphenous vein survey to identify a suitable conduit for the DRIL prior to the index access creation.

Dialysis patients are known to have poor long-term survival in comparison to the general population, as documented by the United States Renal Data System (USRDS). In fact, the USRDS reported in 2010 a strikingly high 20% annual mortality for all hemodialysis patients36. Our data suggest that patients with ARHI may represent a population at even higher risk, with a one-year mortality of nearly 30% after their DRIL procedure. We do not believe the DRIL procedure contributes to this higher mortality, but rather that patients with ARHI have increased number and severity of comorbid conditions, including longer dependence on hemodialysis, that likely lead to poorer survival. Our analysis demonstrated that an age > 40, any smoking history, and Grade 3 ischemia were all independently predictive of increased mortality. Unfortunately, none of these identified predictors can be modified to reduce risk in the pre-operative setting. This novel survival finding may be better applied in the selection of patients for DRIL or patient counseling of risks and benefits.

The study has several limitations that merit further discussion. The retrospective nature of our data collection likely leads to an underestimation of the incidence of hand ischemia and precludes an accurate assessment of the disease severity of both the hand ischemia and the underlying patient comorbidities. It is conceivable that several of our access patients developed ARHI and had their remedial procedures performed at outside institutions, although this is unlikely given our access referral practice. In addition, there was an inherent selection bias that impacts both which access and which remedial procedure were performed. This is evident in the lack of a comparative alternative remedial strategy to DRIL in our series. Complication from DRIL was identified as a risk factor for primary patency failure and mortality but, unfortunately, cannot be determined pre-operatively to guide decision making. The development of a prediction model for complication after DRIL may be important in defining a subset of patients with higher concurrent risk of peri-operative complication, DRIL failure, and all-cause mortality. This is the subject of a future analysis and may further identify patients who would be best served with alternative remedial operations for management of ARHI.

In conclusion, the DRIL procedure effectively improves distal perfusion and reverses the symptoms of ARHI while salvaging the access. Avoidance of non-autogenous conduit use is important to achieve good outcomes. All-cause mortality after DRIL is high and given the high mortality of this patient population, pre-operative risk stratification is critical for optimal utilization of this remedial strategy. Patients at high risk for DRIL failure and mortality may be best served with alternate remedial procedures.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Presented at the 40th Annual Symposium for the Society for Clinical Vascular Surgery, Las Vegas, Nevada, March, 14th, 2012

References

  • 1.Goff CD, Sato DT, Bloch PH, DeMasi RJ, Gregory RT, Gayle RG, Parent FN, Meier GH, Wheeler JR. Steal syndrome complicating hemodialysis access procedures: Can it be predicted? Ann Vasc Surg. 2000;14:138–144. doi: 10.1007/s100169910025. [DOI] [PubMed] [Google Scholar]
  • 2.Hurton S, Embil JM, Reda A, Smallwood S, Wall C, Thomson L, Zacharias J, Dascal M, Trepman E, Koulack J. Upper extremity complications in patients with chronic renal failure receiving haemodialysis. J Ren Care. 2010;36:203–211. doi: 10.1111/j.1755-6686.2010.00197.x. [DOI] [PubMed] [Google Scholar]
  • 3.Tordoir JH, Dammers R, van der Sande FM. Upper extremity ischemia and hemodialysis vascular access. Eur J Vasc Endovasc Surg. 2004;27:1–5. doi: 10.1016/j.ejvs.2003.10.007. [DOI] [PubMed] [Google Scholar]
  • 4.Papasavas PK, Reifsnyder T, Birdas TJ, Caushaj PF, Leers S. Prediction of arteriovenous access steal syndrome utilizing digital pressure measurements. Vasc Endovascular Surg. 2003;37:179–184. doi: 10.1177/153857440303700304. [DOI] [PubMed] [Google Scholar]
  • 5.Keuter XH, Kessels AG, de Haan MH, van der Sande FM, Tordoir JH. Prospective evaluation of ischemia in brachial-basilic and forearm prosthetic arteriovenous fistulas for hemodialysis. Eur J Vasc Endovasc Surg. 2008;35:619–624. doi: 10.1016/j.ejvs.2007.11.004. [DOI] [PubMed] [Google Scholar]
  • 6.Lazarides MK, Staramos DN, Kopadis G, Maltezos C, Tzilalis VD, Georgiadis GS. Onset of arterial 'steal' following proximal angioaccess: Immediate and delayed types. Nephrol Dial Transplant. 2003;18:2387–2390. doi: 10.1093/ndt/gfg346. [DOI] [PubMed] [Google Scholar]
  • 7.Morsy AH, Kulbaski M, Chen C, Isiklar H, Lumsden AB. Incidence and characteristics of patients with hand ischemia after a hemodialysis access procedure. J Surg Res. 1998;74:8–10. doi: 10.1006/jsre.1997.5206. [DOI] [PubMed] [Google Scholar]
  • 8.Schanzer H, Schwartz M, Harrington E, Haimov M. Treatment of ischemia due to"steal" by arteriovenous fistula with distal artery ligation and revascularization. J Vasc Surg. 1988;7:770–773. doi: 10.1067/mva.1988.avs0070770. [DOI] [PubMed] [Google Scholar]
  • 9.Schanzer H, Eisenberg D. Management of steal syndrome resulting from dialysis access. Semin Vasc Surg. 2004;17:45–49. doi: 10.1053/j.semvascsurg.2003.11.003. [DOI] [PubMed] [Google Scholar]
  • 10.Gupta N, Yuo TH, Konig Gt, Dillavou E, Leers SA, Chaer RA, Cho JS, Makaroun MS. Treatment strategies of arterial steal after arteriovenous access. J Vasc Surg. 2011;54:162–167. doi: 10.1016/j.jvs.2010.10.134. [DOI] [PubMed] [Google Scholar]
  • 11.Huber TS, Brown MP, Seeger JM, Lee WA. Midterm outcome after the distal revascularization and interval ligation (dril) procedure. J Vasc Surg. 2008;48:926–932. doi: 10.1016/j.jvs.2008.05.028. discussion 932-923. [DOI] [PubMed] [Google Scholar]
  • 12.Knox RC, Berman SS, Hughes JD, Gentile AT, Mills JL. Distal revascularization-interval ligation: A durable and effective treatment for ischemic steal syndrome after hemodialysis access. J Vasc Surg. 2002;36:250–255. doi: 10.1067/mva.2002.125025. discussion 256. [DOI] [PubMed] [Google Scholar]
  • 13.Balaji S, Evans JM, Roberts DE, Gibbons CP. Treatment of steal syndrome complicating a proximal arteriovenous bridge graft fistula by simple distal artery ligation without revascularization using intraoperative pressure measurements. Ann Vasc Surg. 2003;17:320–322. doi: 10.1007/s10016-001-0262-x. [DOI] [PubMed] [Google Scholar]
  • 14.Minion DJ, Moore E, Endean E. Revision using distal inflow: A novel approach to dialysis-associated steal syndrome. Ann Vasc Surg. 2005;19:625–628. doi: 10.1007/s10016-005-5827-7. [DOI] [PubMed] [Google Scholar]
  • 15.Diehl L, Johansen K, Watson J. Operative management of distal ischemia complicating upper extremity dialysis access. Am J Surg. 2003;186:17–19. doi: 10.1016/s0002-9610(03)00118-1. [DOI] [PubMed] [Google Scholar]
  • 16.Korzets A, Kantarovsky A, Lehmann J, Sachs D, Gershkovitz R, Hasdan G, Vits M, Portnoy I, Korzets Z. The"dril" procedure--a neglected way to treat the"steal" syndrome of the hemodialysed patient. Isr Med Assoc J. 2003;5:782–785. [PubMed] [Google Scholar]
  • 17.Goodkin DA, Bragg-Gresham JL, Koenig KG, Wolfe RA, Akiba T, Andreucci VE, Saito A, Rayner HC, Kurokawa K, Port FK, Held PJ, Young EW. Association of comorbid conditions and mortality in hemodialysis patients in europe, japan, and the united states: The dialysis outcomes and practice patterns study (dopps) J Am Soc Nephrol. 2003;14:3270–3277. doi: 10.1097/01.asn.0000100127.54107.57. [DOI] [PubMed] [Google Scholar]
  • 18.Huber TS, Ozaki CK, Flynn TC, Lee WA, Berceli SA, Hirneise CM, Carlton LM, Carter JW, Ross EA, Seeger JM. Prospective validation of an algorithm to maximize native arteriovenous fistulae for chronic hemodialysis access. J Vasc Surg. 2002;36:452–459. doi: 10.1067/mva.2002.127342. [DOI] [PubMed] [Google Scholar]
  • 19.Huber TS, Hirneise CM, Lee WA, Flynn TC, Seeger JM. Outcome after autogenous brachial-axillary translocated superficial femoropopliteal vein hemodialysis access. J Vasc Surg. 2004;40:311–318. doi: 10.1016/j.jvs.2004.04.018. [DOI] [PubMed] [Google Scholar]
  • 20.Sidawy AN, Spergel LM, Besarab A, Allon M, Jennings WC, Padberg FT, Jr, Murad MH, Montori VM, O'Hare AM, Calligaro KD, Macsata RA, Lumsden AB, Ascher E. The society for vascular surgery: Clinical practice guidelines for the surgical placement and maintenance of arteriovenous hemodialysis access. J Vasc Surg. 2008;48:2S–25S. doi: 10.1016/j.jvs.2008.08.042. [DOI] [PubMed] [Google Scholar]
  • 21.Scali ST, Huber TS. Treatment strategies for access-related hand ischemia. Semin Vasc Surg. 2011;24:128–136. doi: 10.1053/j.semvascsurg.2011.05.012. [DOI] [PubMed] [Google Scholar]
  • 22.Berman SS, Gentile AT, Glickman MH, Mills JL, Hurwitz RL, Westerband A, Marek JM, Hunter GC, McEnroe CS, Fogle MA, Stokes GK. Distal revascularization-interval ligation for limb salvage and maintenance of dialysis access in ischemic steal syndrome. J Vasc Surg. 1997;26:393–402. doi: 10.1016/s0741-5214(97)70032-6. discussion 402-394. [DOI] [PubMed] [Google Scholar]
  • 23.Seeger JM, Pretus HA, Carlton LC, Flynn TC, Ozaki CK, Huber TS. Potential predictors of outcome in patients with tissue loss who undergo infrainguinal vein bypass grafting. J Vasc Surg. 1999;30:427–435. doi: 10.1016/s0741-5214(99)70069-8. [DOI] [PubMed] [Google Scholar]
  • 24.Gibson KD, Caps MT, Gillen D, Bergelin RO, Primozich J, Strandness DE., Jr Identification of factors predictive of lower extremity vein graft thrombosis. J Vasc Surg. 2001;33:24–31. doi: 10.1067/mva.2001.112214. [DOI] [PubMed] [Google Scholar]
  • 25.Sidawy AN, Gray R, Besarab A, Henry M, Ascher E, Silva M, Jr, Miller A, Scher L, Trerotola S, Gregory RT, Rutherford RB, Kent KC. Recommended standards for reports dealing with arteriovenous hemodialysis accesses. J Vasc Surg. 2002;35:603–610. doi: 10.1067/mva.2002.122025. [DOI] [PubMed] [Google Scholar]
  • 26.Andersen PK, Geskus RB, de Witte T, Putter H. Competing risks in epidemiology: Possibilities and pitfalls. Int J Epidemiol. 2012 doi: 10.1093/ije/dyr213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Putter H, Fiocco M, Geskus RB. Tutorial in biostatistics: Competing risks and multi-state models. Stat Med. 2007;26:2389–2430. doi: 10.1002/sim.2712. [DOI] [PubMed] [Google Scholar]
  • 28.Kim HT. Cumulative incidence in competing risks data and competing risks regression analysis. Clin Cancer Res. 2007;13:559–565. doi: 10.1158/1078-0432.CCR-06-1210. [DOI] [PubMed] [Google Scholar]
  • 29.Thermann F, Kornhuber M. Ischemic monomelic neuropathy: A rare but important complication after hemodialysis access placement--a review. J Vasc Access. 2011;12:113–119. doi: 10.5301/jva.2011.6365. [DOI] [PubMed] [Google Scholar]
  • 30.Conte MS, Bandyk DF, Clowes AW, Moneta GL, Namini H, Seely L. Risk factors, medical therapies and perioperative events in limb salvage surgery: Observations from the prevent iii multicenter trial. J Vasc Surg. 2005;42:456–464. doi: 10.1016/j.jvs.2005.05.001. discussion 464-455. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Zanow J, Kruger U, Scholz H. Proximalization of the arterial inflow: A new technique to treat access-related ischemia. J Vasc Surg. 2006;43:1216–1221. doi: 10.1016/j.jvs.2006.01.025. discussion 1221. [DOI] [PubMed] [Google Scholar]
  • 32.Zanow J, Petzold K, Petzold M, Krueger U, Scholz H. Flow reduction in high-flow arteriovenous access using intraoperative flow monitoring. J Vasc Surg. 2006;44:1273–1278. doi: 10.1016/j.jvs.2006.08.010. [DOI] [PubMed] [Google Scholar]
  • 33.Jennings WC. Creating arteriovenous fistulas in 132 consecutive patients: Exploiting the proximal radial artery arteriovenous fistula: Reliable, safe, and simple forearm and upper arm hemodialysis access. Arch Surg. 2006;141:27–32. doi: 10.1001/archsurg.141.1.27. discussion 32. [DOI] [PubMed] [Google Scholar]
  • 34.Shemesh D, Goldin I, Olsha O. Banding between dialysis puncture sites to treat severe ischemic steal syndrome in low flow autogenous arteriovenous access. J Vasc Surg. 2010;52:495–498. doi: 10.1016/j.jvs.2010.01.096. [DOI] [PubMed] [Google Scholar]
  • 35.K/doqi clinical practice guidelines for chronic kidney disease: Evaluation, classification, and stratification. Am J Kidney Dis. 2002;39:S1–S266. [PubMed] [Google Scholar]
  • 36.Collins AJ, Foley RN, Herzog C, Chavers B, Gilbertson D, Ishani A, Kasiske B, Liu J, Mau LW, McBean M, Murray A, St Peter W, Guo H, Gustafson S, Li Q, Li S, Peng Y, Qiu Y, Roberts T, Skeans M, Snyder J, Solid C, Wang C, Weinhandl E, Zaun D, Arko C, Chen SC, Dalleska F, Daniels F, Dunning S, Ebben J, Frazier E, Hanzlik C, Johnson R, Sheets D, Wang X, Forrest B, Constantini E, Everson S, Eggers P, Agodoa L. Us renal data system 2010 annual data report. Am J Kidney Dis. 2011;57(A8):e1–526. doi: 10.1053/j.ajkd.2010.10.007. [DOI] [PubMed] [Google Scholar]

RESOURCES