- Email: [email protected]
Percutaneous treatment of thrombosed primary arteriovenous hemodialysis access fistulae
Kidney International, Vol. 57 (2000), pp. 1169–1175
Percutaneous treatment of thrombosed primary arteriovenous hemodialysis access fistulae PATRICK HAAGE, DIERK VORWERK, JOACHIM E. WILDBERGER, WERNER PIROTH, KARL SCHU¨RMANN, and ROLF W. GU¨NTHER Department of Diagnostic Radiology, University of Technology, Aachen, and Department of Diagnostic and Interventional Radiology, Klinikum Ingolstadt, Ingolstadt, Germany
Percutaneous treatment of thrombosed primary arteriovenous hemodialysis access fistulae. Background. We reviewed the efficacy of percutaneous intervention in acute thrombotic occlusion of native arteriovenous (AV) fistulae for hemodialysis. Methods. Eight-one percutaneous procedures were performed in 54 patients presenting with a clotted native dialysis fistula. There were 60 cases of a long-segment thrombosis of the fistula. In 20 cases, a small thrombus usually caused by an underlying severe stenosis was observed. A proximal arterial occlusion was seen in one case. Treatment depended on clot size and included balloon dilation (N ⫽ 20), mechanical thrombectomy with various devices (N ⫽ 58), as well as pharmacomechanical thrombolysis (N ⫽ 3). Results. Full restoration of flow was established in 72 cases (88.9%). Early reobstruction within 14 days occurred in eight cases (11.1%). Primary patency rates after a 1-, 3-, 6-, and 12month period were 74, 63, 52, and 27%, respectively. Overall fistula patency was 75% after 3 months, 65% after 6 months, 51% after 12 months, and 22% after 24 months. Conclusions. Acute thrombotic occlusion of native AV fistulae is a major complication of hemodialysis. The results of treatment are believed to be less successful than thrombosis treatment in synthetic grafts. Our results, however, indicate the efficacy of percutaneous treatment in native fistulae, and demonstrate comparable technical results and patency rates.
Dysfunction of hemodialysis fistulae and grafts is a problem frequently encountered in hemodialysis patients. Because the number of patients with end-stage renal disease being treated by hemodialysis is rising steadily [1], the choice of an adequate treatment technique to maintain vascular access is of growing importance for the patient and his or her physician. Shunt thrombosis is a commonly observed severe complication, Key words: hemodialysis access, stenosis, thrombosis, fistula, percutaneous treatment, dialysis fistula, vascular access. Received for publication June 7, 1999 and in revised form September 1, 1999 Accepted for publication October 12, 1999
2000 by the International Society of Nephrology
contributing significantly to morbidity and hospitalization in the dialysis patient [2]. In the past, the creation and preservation of the vascular access were primarily the responsibility of the surgeon, and dialysis shunt thrombosis was regularly corrected with surgical thrombectomy. Over the last decade, vascular access complications have been increasingly approached on a multidisciplinary basis. Today, dialysis centers are often using percutaneous interventions performed by the interventional radiologist in the management of hemodialysis access-related problems. Currently, available percutaneous treatment regimens for thrombosed dialysis grafts and fistulae include mechanical thrombectomy, pharmacomechanical thrombolysis, and percutaneous infusion pharmacologic thrombolysis. According to the clinical practice guidelines for vascular access by the National Kidney Foundation [3], each institution should choose to resolve thrombosis based on the expertise of the center. Treatment of thrombosis in arteriovenous (AV) grafts is thought to offer better success rates than thrombosis correction in native fistulae [3]. However, only few data exist on the treatment of clotted native AV dialysis fistulae. The present study provides follow-up data on 81 procedures performed percutaneously to determine the efficacy of thrombolysis in acute obstruction of primary AV fistulae in the hemodialysis patient. METHODS Fifty-four patients with 81 episodes of recently thrombosed native AV fistulae were referred for percutaneous thrombectomy. As a general rule in our institution, patients with suspected shunt dysfunction are primarily referred to the interventional radiologic department for percutaneous treatment. All of the patients diagnosed with acute thrombotic occlusion were included in this study. There were 32 male and 22 female patients aged 24 to 79 years (mean 54.3 years). The type of fistula was a forearm radiocephalic Brescia-Cimino fistula in 50
1169
1170
Haage et al: Treatment of thrombosed primary AV fistulae
patients and an upper-arm brachiocephalic AV fistula in four patients. Fistula location was on the right side in 24 patients and on the left side in 30 patients. All patients had end-stage renal disease and were participating in the long-term hemodialysis program. Before the intervention, written, informed consent about the planned procedure was obtained. Treatment was performed on the same day fistula thrombosis was diagnosed. The procedure was performed as an outpatient procedure in 52 and as an inpatient procedure in 29 cases. Treatment outcomes were evaluated on the basis of technical procedural success and patency after correction of the clotted fistula. Technical success was defined as restoration of arterialized fistula flow after termination of the procedure. Primary patency was defined as maintained arterialized flow during follow-up. Primary patency ended when reintervention was necessary because of restenosis or rethrombosis. The overall fistula patency was defined as the ability to use the fistula for hemodialysis, including, if necessary and feasible, percutaneous repeat interventions. Primary and overall patency rates were calculated by the means of the Kaplan–Meier lifetable analysis. Follow-up was performed by clinical examination and by monitoring the dialysis protocols. In the case of abnormal clinical findings and/or deviant dialysis data suggesting fistula failure, repeat angiography was indicated. If rethrombosis was demonstrated, recanalization was attempted. Fistula pathology was visualized by puncturing the brachial artery with a 22-gauge sheath needle (Abbott, Sligo, Ireland) for arteriovenography. This approach was used to monitor the intervention in all procedures. Then a tourniquet was applied for passive vein congestion to select a venous puncture site by palpation, thus avoiding puncture of a clotted or stenosed segment. In selected cases, additional Doppler ultrasound was performed. The outflow vein was punctured in a retrograde fashion against the flow direction with the use of local anesthesia and a 16-gauge sheath needle (Abbott), which was exchanged for a 7-F introducer sheath (Terumo, Tokyo, Japan) after dilation of the puncture site up to 7 F. The occluded segment was passed with a 5-F multipurpose catheter (Cook, Bjaeverskov, Denmark) and a hydrophilic-coated and steerable 0.035-inch guide wire (Terumo) across the anastomosis until the arterial segment was reached. Additional antegrade distal venous puncture for treatment of an upstream venous stenosis was performed in 12 cases. In four patients with a brachiocephalic upper arm fistula, the protocol was modified by antegrade puncture close to the anastomosis. The type of treatment was chosen depending on location of the occlusion and clot size observed by angiography. In cases of a very short plug-like thrombus selectively
obstructing the AV anastomosis (N ⫽ 20), maceration of the thrombus was attempted by balloon angioplasty alone. Treatment of a long-segment thrombosis of the autologous AV fistula (N ⫽ 60) was performed by mechanical thrombectomy with various devices in 58 cases, including the Amplatz mechanical thrombectomy device (Microvena, White Bear Lake, MN, USA; N ⫽ 23), the Hydrolyser catheter (Cordis Europe, Roden, The Netherlands; N ⫽ 24), and combined use of these devices (N ⫽ 11). In selected cases, a minibasket for percutaneous removal of wall-adherent thrombi (Cook; N ⫽ 3) and a cutting balloon (Interventional Technologies, San Diego, CA, USA; N ⫽ 1) were used in addition to the previously listed devices to improve the recanalization result. In two cases, fistula occlusion was treated by percutaneous pharmacomechanical thrombolysis with 10 to 12 mg recombinant tissue plasminogen activator (rtPA) and clot maceration with a 6 mm percutaneous transluminal angioplasty (PTA) balloon. In one patient with an occlusion of the proximal radial artery, thrombolysis was performed with 15 mg rtPA and subsequent PTA. Balloon dilation was regularly performed (N ⫽ 56) to either treat an underlying stenosis or compress residual thrombus material. Balloon catheter size was usually 5 to 7 ⫻ 20 or 40 mm. Four ⫻ 20 mm balloons were chosen for angioplasty close to the AV anastomosis. No special technique was applied to prevent potential central embolization, since the 7-F introducer sheath sufficiently blocked the venous outflow for larger emboli. During the mechanical thrombectomy maneuver, the AV anastomosis was digitally compressed to avoid retrograde thrombus dislodgment. Periprocedural parenteral anticoagulation was administered regularly at the beginning of the intervention (5000 IU heparin). After the procedure was terminated, continued heparinization was recommended with 1000 IU/h for at least 24 hours for inpatients and three subcutaneous 7500 IU injections during 24 hours in outpatients. RESULTS Entrance into the thrombosed fistula was successfully performed in all 81 instances. Initial fistula salvage with restoration of flow was established in 72 of 81 (88.9%) attempts (Figs. 1 and 2). Thirty-six of the 54 patients underwent declotting once, and 10 patients were treated twice. Three or more episodes of thrombosis occurred in eight patients with up to four episodes. The estimated maximum occlusion time was 25 ⫾ 12 hours. The length of the thrombosed segment ranged from 1 to 40 cm (mean 22 ⫾ 17). Involvement of the distal radial artery was observed once. Basically, there were three groups of patients in this
Haage et al: Treatment of thrombosed primary AV fistulae
1171
Fig. 1. Digital subtraction angiography demonstrates a complete occlusion of the fistula draining vein at the arteriovenous anastomosis (arrow).
Fig. 2. Angiogram after mechanical thrombectomy with the Amplatz device and PTA. Recanalized fistula with excellent postprocedural flow (arrows) is shown.
study: one group with an arterial thrombosis (N ⫽ 1), one group with a short-segment thrombus (N ⫽ 20), and one group with a long-segment thrombosis (N ⫽ 60). In the 20 cases with a short thrombus and in the patient with an arterial fistula occlusion, arterialized flow was completely re-established in all of the procedures (100%). Technical failure occurred in 9 of 60 cases (15%) in the group with a long-segment thrombosis, and was considered to be due to older plug-like thrombus material in seven cases that the mechanical thrombectomy devices
were not able to break up sufficiently. In one case, the intervention had to be abandoned because of severe pain; venous perforation at the puncture site with consecutive hematoma required surgical thrombectomy in one case. The estimated amount of residual thrombus was 23% (5 to 50% range). By combining the chosen mechanical thrombectomy device (N ⫽ 58) or the thrombolytic agent (N ⫽ 2) with PTA and in four cases, an additional application of a minibasket and cutting balloon, arterialized flow was re-established in 51 cases. The results of the various mechanical devices used in this group did
1172
Haage et al: Treatment of thrombosed primary AV fistulae
Table 1. Treatment results for mechanical thrombectomy devices (N ⫽ 58)
Device Amplatz TD Hydrolyser catheter Device combination
No. of procedures 23 24 11
Technical success
3 month patency rate %
82.6 83.3 91.0
67.3 64.1 62.2
Fig. 3. Follow-up findings of the life-table analysis depicting the primary (䊐) and overall (䉬) patency rates.
not differ significantly using the two-tailed Fisher’s exact test (P ⫽ 1.000; Table 1). If percutaneous thrombectomy failed, creation of a new fistula was performed in two cases. Surgical thrombus extraction was tried in seven, but failed in three cases. A new fistula had to be created in these patients. The thrombectomy procedures in 54 fistulas were followed from 1 day to 51 months with a mean follow-up of 13 months ⫾ 14.3 months. In the early stages of this study, we were concerned about a possible reduction of blood flow from the needles in place, and therefore, postprocedural hemodialysis was performed via a Shaldon catheter in 36 cases to avoid early rethrombosis. However, there were no findings supporting our theory, and this approach was discontinued. Thirty episodes of reocclusion occurred within the observation period. Early rethrombosis within 14 days of thrombectomy was seen in eight cases (11.1%). Hypercoagulability, hypotension, and stenosis were the assumed cause of early thrombosis. Thirty-two patients had no observed fistula rethrombosis since the initial thrombectomy treatment. In addition, a total of 48 episodes of restenosis was noted, which were successfully treated by percutaneous balloon angioplasty in all instances. Taking reinterventions into consideration, the fistula remained open in 34 of 54 pa-
tients (63%) until the end of the observation period or the patient’s death. The remaining patients were lost to follow-up or underwent renal transplantation at some point in the observation period. There was no local procedure-related bleeding at the catheter site. Venous dissection necessitating stent placement occurred in three cases (3.7%). The thrombectomy procedure itself was successfully performed in these three instances, and the fistulas have remained patent for at least six months. When advancing an angioplasty catheter across the stenotic anastomosis in a patient with a forearm fistula after a first Hydrolyser run, embolization into the distal radial artery was observed. The embolus was easily retrieved by insertion of a minibasket inside a 3-F catheter (Cook) across the anastomosis to the more proximal part of the radial artery via the 7-F sheath. The embolus was dislodged into the fistula vein and subsequently removed with hydrodynamic thrombectomy. No clinical signs of pulmonary embolism with reports of chest pain, shortness of breath, and pleurisy were seen, and no other major complications were found. Kaplan–Meier survival curves are displayed in Figure 3. With the inclusion of initial failures, primary patency was calculated as 74% after 1 month (SE 5.6%), 63% after 3 months (SE 6.1%), 52% after 6 months (SE 6.3%), and 27% after 12 months (SE 6.6%). The median duration of primary access patency was 8.1 ⫾ 11.5 months. The overall fistula patency was 75% at 3 months (SE 6.6%), 65% at 6 months (SE 7.3%), 51% at 12 months (7.7%), and 22% at 24 months (SE 8.4%). DISCUSSION Early detection and treatment of complications are essential to achieve long-term access function for the hemodialysis patient. Thrombosis of dialysis grafts and fistulae represents a frequent complication [4]. Predisposing factors are an underlying stenosis or aneurysmatic vein causing flow turbulences [5, 6], as well as arterial hypotension, low cardiac output, compromised arterial inflow, and clotting disorders [7]. In the vast majority of cases, a venous stenosis is detected [6]. Surgical, as well as percutaneous, procedure results have demonstrated their effectiveness in graft recanalization. However, patency rates of percutaneous and surgical techniques are difficult to compare, as percutaneous interventions are followed beginning with the initial procedure, whereas surgical studies generally report patency starting from the time of access creation. In a retrospective comparison of surgical and percutaneous treatment for graft thrombosis, Sands et al performed 75 surgical thrombectomies and 71 percutaneous pharmacomechanical procedures and observed no significant differences in success rates, primary patencies, and rate of complications [8]. Beath-
Haage et al: Treatment of thrombosed primary AV fistulae
ard described equal initial success rates of mechanical thrombolysis compared with surgical thrombectomy [9]. Studies on surgical treatment of thrombosed native fistulae are rare. Diskin et al demonstrated a mean access survival of 244 days after surgery in 45 patients with native fistulas [10]. Wehrli, Chenevard, and Zaruba reported a success rate of 81.2% in patients with thrombosis of a newly created dialysis access [11]. Romero et al found an almost 50% cumulative patency at two years after creation of a proximal new fistula caused by late thrombosis [12]. Dapunt et al compared the results of 37 transluminal angioplasties with those of 37 operations in the treatment of stenoses and occlusions of dialysis fistulae [13]. Cumulative patency after 12 months was 31.3% for angioplasty and 19.3% for surgery. One limitation of the surgical approach is the lack of imaging guidance in the operating room to detect additional stenoses and obtain information on the often complex vessel anatomy. Conversely, performance of surgical thrombectomy in the radiologic laboratory requires a considerable amount of organization and manpower. A solely percutaneous approach is therefore a cost-reducing, attractive alternative. In recent years, percutaneous interventional techniques have been tried and established as a viable alternative means in the management of fistula dysfunction. Zaleski et al reported on 17 patients with complete thrombosis of their BresciaCimino fistulas, which were treated by angioplasty and urokinase infusion [14]. Procedural success was 82% with primary, primary assisted, and secondary patency rates at 12 months of 71, 93, and 100%, respectively. Poulain et al combined a local low-dose infusion of urokinase with PTA and thromboaspiration to achieve a 12-month overall patency in 14 native fistulas of approximately 90% [15]. Twenty out of 24 patients (83%) with occluded Brescia-Cimino fistulas were successfully declotted by Overbosch et al, employing the Hydrolyser catheter [16]. The median assisted patency was 34 weeks and was significantly shorter in fistulas than in polytetrafluorolethylene (PTFE) grafts (P ⫽ 0.002). Turmel-Rodrigues et al described an 81% initial success rate using thromboaspiration and PTA in 16 patients [17]. An 81% secondary patency at one year was reported. Comparable results, however, with far less patients, have been demonstrated by various workgroups [18–23]. The reported technical success rates are similar to our own, whereas our patency rates are less than those of Zaleski et al, Poulain et al, and Turmel-Rodrigues et al, and are rather comparable to those of Overbosch et al. These results suggest that percutaneous interventions in occluded native AV fistulas are at least as effective as surgical treatment regarding technical success and patencies. Treatment results of thrombosis and associated stenosis in synthetic grafts have been summarized by Aruny et al [24]. Clinical success rates for thrombolysis or me-
1173
chanical thrombectomy range from 75 to 94% with primary patencies of 18 to 39% at 6 months. Reported 6- and 12-month secondary patencies for thrombolysis range from 62 to 80% and 57 to 69%, respectively. Trerotola et al recently reported a 95% technical success with a three-month primary patency of 39% using the newly developed Arrow-Trerotola percutaneous thrombolytic device [25]. Both the percutaneous and the surgical approach to thrombosed AV fistulas remain difficult, partly because of the pronounced organization of thrombus material in native vessels. In many centers, patients with an occluded native fistula are referred for surgical thrombectomy. However, because of anatomic reasons, the surgeon often prefers to perform a patch insertion or fistula reanastomosis instead of mechanical thrombectomy. Our results with technical success rates of almost 90% and a primary and overall fistula patency of 27% and 51% at one year, respectively, demonstrate the effectiveness of percutaneous thrombosis treatment strategies in native fistulas. It is our workgroup’s opinion that as long as less invasive comparable therapeutic options to preserve the vessel are available and feasible, abandonment of the vascular access requiring an AV neoanastomosis should be avoided, and a percutaneous approach should be considered. Generally, declotting should be performed as an outpatient procedure. In two thirds of our cases, the procedure was performed on an outpatient basis, avoiding general anesthesia and hospitalization, thus reducing the treatment risks and costs. The remaining one third of the cases were inpatients, who had been primarily admitted to the hospital because of nonfistula-related problems. We prefer to perform fistulography in native fistulas via the brachial artery for several reasons. Arterial and venous angiomorphology can be evaluated without placing a blood pressure cuff on the upper arm and inflating the cuff to a suprasystolic pressure. This procedure can be painful for the patient and lead to involuntary movements of the extremity, aggravating satisfactory shunt visualization and documentation. The blood pressure cuff has to be left up for as short as possible to avoid thrombosis. In addition, judgment of flow velocity and spontaneous filling of collaterals, indicating the hemodynamic significance of stenoses and occlusions, is impaired with the cuff inflated, especially at the AV anastomosis. If necessary, the percutaneous intervention is monitored easily from the arterial access; also, the procedural outcome can be documented after catheter removal. Complications of arterial puncture are rare. In a retrospective analysis of 454 transbrachial arteriographies, we have observed one episode (0.2%) of significant bleeding at the puncture site [26]. It is advisable to perform the intervention as soon
1174
Haage et al: Treatment of thrombosed primary AV fistulae
as possible once thrombosis is diagnosed because the occlusion period in cases with primary technical failure was considerably longer compared with the whole group (44 ⫾ 22 h vs. 25 ⫾ 12 h). Involvement of large veins such as the brachial vein substantially complicates treatment and is another important factor determining technical success [7]. Long-term shunt functionality depends on manifold factors, such as cannulation trauma, tendency for restenosis, hypercoagulability, and frequency of reobstruction episodes [27]. Reobstruction is a frequently observed complication and is more likely to occur after thrombosis has occurred for the first time [27– 29]. The treating physician must be prepared for repeat interventions, sometimes multiple, over the months and years after the initial declotting procedure. In our study, 48 episodes of restenosis and 27 episodes of reocclusion were noted in 54 hemodialysis patients, amounting to an average of 1.4 reinterventions per patient in a mean follow-up period of 13 months. In conclusion, initial percutaneous thrombectomy is very effective in the treatment of recently occluded AV fistulae with good success rates and acceptable primary and long-term patency rates comparable to other therapy regimen such as surgical thrombectomy. It has to be stressed that the decision of whether to perform a percutaneous fistula revision or a surgical intervention should rely on the expertise of the respective center regarding access dysfunction. If percutaneous thrombolysis is available, surgical revision should be reserved for failures of percutaneous techniques. The choice of the appropriate percutaneous approach depends on the size and location of the thrombus detected by angiography. A short segment thrombosis can be easily treated with PTA alone, whereas an extensive thrombosis requires a combination of mechanical devices and/or thrombolytic agents with adjunctive balloon angioplasty. According to Gray, it is unlikely that the declotting method in dialysis grafts, whether pharmacomechanical or purely mechanical, has a significant effect on short- or long-term patency [30]. Further studies have to be undertaken to offer the native fistula patient the most beneficial and durable percutaneous treatment strategy. At present, either type of method for fistula recanalization is appropriate and should be based on the experience of the interventional radiologist with the respective therapeutic modality. The decision making need not depend on the type of dialysis access presented by the patient. Our results indicate that percutaneous thrombectomy is equally successful in native and implant shunts. Effective care for the hemodialysis patient to preserve vascular access over a prolonged period of time can only be achieved on a multidisciplinary basis with the surgeon, interventional radiologist, and nephrologist closely cooperating.
Reprint requests to Patrick Haage, M.D., Department of Diagnostic Radiology, University of Technology Aachen, Pauwelsstrasse 30, D-52057 Aachen, Germany. Email: [email protected]
REFERENCES 1. United States Renal Data System: USRDS 1998 Annual Data Report. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Disease, Bethesda, 1998 2. Windus D: Permanent vascular access: A nephrologist’s view. Am J Kidney Dis 21:457–471, 1993 3. NKF-DOQI clinical practice guidelines for vascular access. National Kidney Foundation-Dialysis Outcomes Quality Initiative. Am J Kidney Dis 30(Suppl):S174–S175, 1997 4. Zibari GB, Rohr MS, Landreneau MD, Bridges RM, Devault GA, Petty FH, Costley KJ, Brown ST, McDonald JC: Complications from permanent hemodialysis vascular access. Surgery 104:681–686, 1988 5. Valji K, Bookstein J, Roberts A, Davis G: Pharmacomechanical thrombolysis and angioplasty in the management of clotted hemodialysis grafts: Early and late clinical results. Radiology 178:243– 248, 1991 6. Schwab S, Raymond J, Saeed M, Newman G, Dennis P, Bollinger R: Prevention of hemodialysis fistula thrombosis: Early detection of venous stenoses. Kidney Int 36:707–711, 1989 7. Vorwerk D, Schu¨rmann K, Mu¨ller-Leisse C, Adam G, Bu¨cker A, Sohn M, Kierdorf H, Gu¨nther RW: Hydrodynamic thrombectomy of hemodialysis grafts and fistulae: Results of 51 procedures. Nephrol Dial Transplant 11:1058–1064, 1996 8. Sands JJ, Patel S, Plaviak DJ, Miranda CL: Pharmacomechanical thrombolysis with urokinase for treatment of thrombosed hemodialysis access grafts. ASAIO J 40:886–888, 1994 9. Beathard GA: Thrombolysis versus surgery for the treatment of thrombosed dialysis access grafts. J Am Soc Nephrol 6:1619–1624, 1995 10. Diskin CJ, Stokes TJ, Panus LW, Thomas J, Lock S: The importance of timing of surgery for hemodialysis vascular access thrombectomy. Nephron 75:233–237, 1997 11. Wehrli H, Chenevard R, Zaruba K: Surgical experiences with the arteriovenous hemodialysis shunt (1970–1988). Helv Chir Acta 56:621–627, 1989 12. Romero A, Polo JR, Garcia Morato E, Garcia Sabrido JL, Quintans A, Ferreiroa JP: Salvage of angioaccess after late thrombosis of radiocephalic fistulas for hemodialysis. Int Surg 71:122–124, 1986 13. Dapunt O, Feurstein M, Rendl KH, Prenner K: Transluminal angioplasty versus conventional operation in the treatment of haemodialysis fistula stenosis: Results from a 5-year study. Br J Surg 74:1004–1005, 1987 14. Zaleski GX, Funaki B, Kenney S, Lorenz JM, Garofalo R: Angioplasty and bolus urokinase infusion for the restoration of function in thrombosed Brescia-Cimino dialysis fistulas. J Vasc Interv Radiol 10:129–136, 1999 15. Poulain F, Raynaud A, Bourquelot P, Knight C, Rovani X, Gaux JC: Local thrombolysis and thromboaspiration in the treatment of acutely thrombosed arteriovenous hemodialysis fistulas. Cardiovasc Intervent Radiol 14:98–101, 1991 16. Overbosch EH, Pattynama PM, Aarts HJ, Schultze Kool LJ, Hermans J, Reekers JA: Occluded hemodialysis shunts: Dutch multicenter experience with the hydrolyser catheter. Radiology 201:485–488, 1996 17. Turmel-Rodrigues L, Sapoval M, Pengloan J, Billaux L, Testou D, Hauss S, Mouton A, Blanchard D: Manual thromboaspiration and dilation of thrombosed dialysis access: Mid-term results of a simple concept. J Vasc Interv Radiol 8:813–824, 1997 18. Schilling JJ, Eiser AR, Slifkin RF, Whitney JT, Neff MS: The role of thrombolysis in hemodialysis access occlusion. Am J Kidney Dis 10:92–97, 1987 19. Mangiarotti G, Canavese C, Thea A, Segoloni GP, Stratta P, Salomone M, Vercellone A: Urokinase treatment for arteriovenous fistulae declotting in dialyzed patients. Nephron 36:60–64, 1984
Haage et al: Treatment of thrombosed primary AV fistulae 20. Collier PE, Saracco GM, Young JC, Fragola JA, Contractor FM, Diamond DL: Nonoperative salvage of subcutaneous hemodialysis fistulae. Am J Nephrol 5:333–337, 1985 21. Hunter DW, So SK, Castaneda-Zuniga WR, Coleman CE, Sutherland DE, Amplatz K: Failing or thrombosed Brescia-Cimino arteriovenous dialysis fistulas: Angiographic evaluation and percutaneous transluminal angioplasty. Radiology 149:105–109, 1983 22. Pattynama PM, van Baalen J, Verburgh CA, van der Pijl JW, Schultze Kool LJ: Revascularization of occluded haemodialysis fistulae with the hydrolyser thrombectomy catheter: Description of the technique and report of six cases. Nephrol Dial Transplant 10:1224–1227, 1995 23. Goldberg JP, Contiguglia SR, Mishell JL, Klein MH: Intravenous streptokinase for thrombolysis of occluded arteriovenous access: Its use in patients undergoing hemodialysis. Arch Intern Med 145:1405–1408, 1985 24. Aruny JE, Lewis CA, Cardella JF, Cole PE, Davis A, Drooz AT, Grassi CJ, Gray RJ, Husted JW, Jones MT, McCowan TC, Meranze SG, Van Moore A, Neithamer CD, Oglevie SB, Omary RA, Patel NH, Rholl KS, Roberts AC, Sacks D, Sanchez O,
25.
26. 27. 28. 29. 30.
1175
Silverstein MI, Singh H, Swan TL, Towbin RB, Trerotola SO, Bakal CW: Quality improvement guidelines for percutaneous management of the thrombosed or dysfunctional dialysis access. J Vasc Interv Radiol 10:491–498, 1999 Trerotola SO, Vesely TM, Lund GB, Soulen MC, Ehrman KO, Cardella JF: Treatment of thrombosed hemodialysis access grafts: Arrow-Trerotola percutaneous thrombolytic device versus pulsespray thrombolysis. Radiology 206:403–414, 1998 Bu¨cker A, Vorwerk D, Gu¨nther RW: Transbrachial fine-needle arteriography with special focus on haemodialysis shunt imaging. Nephrol Dial Transplant 10:838–841, 1995 Brotman DN, Fandos L, Faust GR, Doscher W, Cohen JR: Hemodialysis graft salvage. J Am Coll Surg 178:431–434, 1994 Etheredge EE, Haid SD, Maeser MN, Sicard GA, Anderson CB: Salvage operations for malfunctioning polytetrafluoroethylene hemodialysis access grafts. Surgery 94:464–470, 1983 Gaylord GM, Taber TE: Long-term hemodialysis access salvage: Problems and challenges for nephrologists and interventional radiologists. J Vasc Interv Radiol 4:103–107, 1993 Gray RJ: Percutaneous intervention for permanent hemodialysis access: A review. J Vasc Interv Radiol 8:313–327, 1997
Percutaneous treatment of thrombosed primary arteriovenous hemodialysis access fistulae PATRICK HAAGE, DIERK VORWERK, JOACHIM E. WILDBERGER, WERNER PIROTH, KARL SCHU¨RMANN, and ROLF W. GU¨NTHER Department of Diagnostic Radiology, University of Technology, Aachen, and Department of Diagnostic and Interventional Radiology, Klinikum Ingolstadt, Ingolstadt, Germany
Percutaneous treatment of thrombosed primary arteriovenous hemodialysis access fistulae. Background. We reviewed the efficacy of percutaneous intervention in acute thrombotic occlusion of native arteriovenous (AV) fistulae for hemodialysis. Methods. Eight-one percutaneous procedures were performed in 54 patients presenting with a clotted native dialysis fistula. There were 60 cases of a long-segment thrombosis of the fistula. In 20 cases, a small thrombus usually caused by an underlying severe stenosis was observed. A proximal arterial occlusion was seen in one case. Treatment depended on clot size and included balloon dilation (N ⫽ 20), mechanical thrombectomy with various devices (N ⫽ 58), as well as pharmacomechanical thrombolysis (N ⫽ 3). Results. Full restoration of flow was established in 72 cases (88.9%). Early reobstruction within 14 days occurred in eight cases (11.1%). Primary patency rates after a 1-, 3-, 6-, and 12month period were 74, 63, 52, and 27%, respectively. Overall fistula patency was 75% after 3 months, 65% after 6 months, 51% after 12 months, and 22% after 24 months. Conclusions. Acute thrombotic occlusion of native AV fistulae is a major complication of hemodialysis. The results of treatment are believed to be less successful than thrombosis treatment in synthetic grafts. Our results, however, indicate the efficacy of percutaneous treatment in native fistulae, and demonstrate comparable technical results and patency rates.
Dysfunction of hemodialysis fistulae and grafts is a problem frequently encountered in hemodialysis patients. Because the number of patients with end-stage renal disease being treated by hemodialysis is rising steadily [1], the choice of an adequate treatment technique to maintain vascular access is of growing importance for the patient and his or her physician. Shunt thrombosis is a commonly observed severe complication, Key words: hemodialysis access, stenosis, thrombosis, fistula, percutaneous treatment, dialysis fistula, vascular access. Received for publication June 7, 1999 and in revised form September 1, 1999 Accepted for publication October 12, 1999
2000 by the International Society of Nephrology
contributing significantly to morbidity and hospitalization in the dialysis patient [2]. In the past, the creation and preservation of the vascular access were primarily the responsibility of the surgeon, and dialysis shunt thrombosis was regularly corrected with surgical thrombectomy. Over the last decade, vascular access complications have been increasingly approached on a multidisciplinary basis. Today, dialysis centers are often using percutaneous interventions performed by the interventional radiologist in the management of hemodialysis access-related problems. Currently, available percutaneous treatment regimens for thrombosed dialysis grafts and fistulae include mechanical thrombectomy, pharmacomechanical thrombolysis, and percutaneous infusion pharmacologic thrombolysis. According to the clinical practice guidelines for vascular access by the National Kidney Foundation [3], each institution should choose to resolve thrombosis based on the expertise of the center. Treatment of thrombosis in arteriovenous (AV) grafts is thought to offer better success rates than thrombosis correction in native fistulae [3]. However, only few data exist on the treatment of clotted native AV dialysis fistulae. The present study provides follow-up data on 81 procedures performed percutaneously to determine the efficacy of thrombolysis in acute obstruction of primary AV fistulae in the hemodialysis patient. METHODS Fifty-four patients with 81 episodes of recently thrombosed native AV fistulae were referred for percutaneous thrombectomy. As a general rule in our institution, patients with suspected shunt dysfunction are primarily referred to the interventional radiologic department for percutaneous treatment. All of the patients diagnosed with acute thrombotic occlusion were included in this study. There were 32 male and 22 female patients aged 24 to 79 years (mean 54.3 years). The type of fistula was a forearm radiocephalic Brescia-Cimino fistula in 50
1169
1170
Haage et al: Treatment of thrombosed primary AV fistulae
patients and an upper-arm brachiocephalic AV fistula in four patients. Fistula location was on the right side in 24 patients and on the left side in 30 patients. All patients had end-stage renal disease and were participating in the long-term hemodialysis program. Before the intervention, written, informed consent about the planned procedure was obtained. Treatment was performed on the same day fistula thrombosis was diagnosed. The procedure was performed as an outpatient procedure in 52 and as an inpatient procedure in 29 cases. Treatment outcomes were evaluated on the basis of technical procedural success and patency after correction of the clotted fistula. Technical success was defined as restoration of arterialized fistula flow after termination of the procedure. Primary patency was defined as maintained arterialized flow during follow-up. Primary patency ended when reintervention was necessary because of restenosis or rethrombosis. The overall fistula patency was defined as the ability to use the fistula for hemodialysis, including, if necessary and feasible, percutaneous repeat interventions. Primary and overall patency rates were calculated by the means of the Kaplan–Meier lifetable analysis. Follow-up was performed by clinical examination and by monitoring the dialysis protocols. In the case of abnormal clinical findings and/or deviant dialysis data suggesting fistula failure, repeat angiography was indicated. If rethrombosis was demonstrated, recanalization was attempted. Fistula pathology was visualized by puncturing the brachial artery with a 22-gauge sheath needle (Abbott, Sligo, Ireland) for arteriovenography. This approach was used to monitor the intervention in all procedures. Then a tourniquet was applied for passive vein congestion to select a venous puncture site by palpation, thus avoiding puncture of a clotted or stenosed segment. In selected cases, additional Doppler ultrasound was performed. The outflow vein was punctured in a retrograde fashion against the flow direction with the use of local anesthesia and a 16-gauge sheath needle (Abbott), which was exchanged for a 7-F introducer sheath (Terumo, Tokyo, Japan) after dilation of the puncture site up to 7 F. The occluded segment was passed with a 5-F multipurpose catheter (Cook, Bjaeverskov, Denmark) and a hydrophilic-coated and steerable 0.035-inch guide wire (Terumo) across the anastomosis until the arterial segment was reached. Additional antegrade distal venous puncture for treatment of an upstream venous stenosis was performed in 12 cases. In four patients with a brachiocephalic upper arm fistula, the protocol was modified by antegrade puncture close to the anastomosis. The type of treatment was chosen depending on location of the occlusion and clot size observed by angiography. In cases of a very short plug-like thrombus selectively
obstructing the AV anastomosis (N ⫽ 20), maceration of the thrombus was attempted by balloon angioplasty alone. Treatment of a long-segment thrombosis of the autologous AV fistula (N ⫽ 60) was performed by mechanical thrombectomy with various devices in 58 cases, including the Amplatz mechanical thrombectomy device (Microvena, White Bear Lake, MN, USA; N ⫽ 23), the Hydrolyser catheter (Cordis Europe, Roden, The Netherlands; N ⫽ 24), and combined use of these devices (N ⫽ 11). In selected cases, a minibasket for percutaneous removal of wall-adherent thrombi (Cook; N ⫽ 3) and a cutting balloon (Interventional Technologies, San Diego, CA, USA; N ⫽ 1) were used in addition to the previously listed devices to improve the recanalization result. In two cases, fistula occlusion was treated by percutaneous pharmacomechanical thrombolysis with 10 to 12 mg recombinant tissue plasminogen activator (rtPA) and clot maceration with a 6 mm percutaneous transluminal angioplasty (PTA) balloon. In one patient with an occlusion of the proximal radial artery, thrombolysis was performed with 15 mg rtPA and subsequent PTA. Balloon dilation was regularly performed (N ⫽ 56) to either treat an underlying stenosis or compress residual thrombus material. Balloon catheter size was usually 5 to 7 ⫻ 20 or 40 mm. Four ⫻ 20 mm balloons were chosen for angioplasty close to the AV anastomosis. No special technique was applied to prevent potential central embolization, since the 7-F introducer sheath sufficiently blocked the venous outflow for larger emboli. During the mechanical thrombectomy maneuver, the AV anastomosis was digitally compressed to avoid retrograde thrombus dislodgment. Periprocedural parenteral anticoagulation was administered regularly at the beginning of the intervention (5000 IU heparin). After the procedure was terminated, continued heparinization was recommended with 1000 IU/h for at least 24 hours for inpatients and three subcutaneous 7500 IU injections during 24 hours in outpatients. RESULTS Entrance into the thrombosed fistula was successfully performed in all 81 instances. Initial fistula salvage with restoration of flow was established in 72 of 81 (88.9%) attempts (Figs. 1 and 2). Thirty-six of the 54 patients underwent declotting once, and 10 patients were treated twice. Three or more episodes of thrombosis occurred in eight patients with up to four episodes. The estimated maximum occlusion time was 25 ⫾ 12 hours. The length of the thrombosed segment ranged from 1 to 40 cm (mean 22 ⫾ 17). Involvement of the distal radial artery was observed once. Basically, there were three groups of patients in this
Haage et al: Treatment of thrombosed primary AV fistulae
1171
Fig. 1. Digital subtraction angiography demonstrates a complete occlusion of the fistula draining vein at the arteriovenous anastomosis (arrow).
Fig. 2. Angiogram after mechanical thrombectomy with the Amplatz device and PTA. Recanalized fistula with excellent postprocedural flow (arrows) is shown.
study: one group with an arterial thrombosis (N ⫽ 1), one group with a short-segment thrombus (N ⫽ 20), and one group with a long-segment thrombosis (N ⫽ 60). In the 20 cases with a short thrombus and in the patient with an arterial fistula occlusion, arterialized flow was completely re-established in all of the procedures (100%). Technical failure occurred in 9 of 60 cases (15%) in the group with a long-segment thrombosis, and was considered to be due to older plug-like thrombus material in seven cases that the mechanical thrombectomy devices
were not able to break up sufficiently. In one case, the intervention had to be abandoned because of severe pain; venous perforation at the puncture site with consecutive hematoma required surgical thrombectomy in one case. The estimated amount of residual thrombus was 23% (5 to 50% range). By combining the chosen mechanical thrombectomy device (N ⫽ 58) or the thrombolytic agent (N ⫽ 2) with PTA and in four cases, an additional application of a minibasket and cutting balloon, arterialized flow was re-established in 51 cases. The results of the various mechanical devices used in this group did
1172
Haage et al: Treatment of thrombosed primary AV fistulae
Table 1. Treatment results for mechanical thrombectomy devices (N ⫽ 58)
Device Amplatz TD Hydrolyser catheter Device combination
No. of procedures 23 24 11
Technical success
3 month patency rate %
82.6 83.3 91.0
67.3 64.1 62.2
Fig. 3. Follow-up findings of the life-table analysis depicting the primary (䊐) and overall (䉬) patency rates.
not differ significantly using the two-tailed Fisher’s exact test (P ⫽ 1.000; Table 1). If percutaneous thrombectomy failed, creation of a new fistula was performed in two cases. Surgical thrombus extraction was tried in seven, but failed in three cases. A new fistula had to be created in these patients. The thrombectomy procedures in 54 fistulas were followed from 1 day to 51 months with a mean follow-up of 13 months ⫾ 14.3 months. In the early stages of this study, we were concerned about a possible reduction of blood flow from the needles in place, and therefore, postprocedural hemodialysis was performed via a Shaldon catheter in 36 cases to avoid early rethrombosis. However, there were no findings supporting our theory, and this approach was discontinued. Thirty episodes of reocclusion occurred within the observation period. Early rethrombosis within 14 days of thrombectomy was seen in eight cases (11.1%). Hypercoagulability, hypotension, and stenosis were the assumed cause of early thrombosis. Thirty-two patients had no observed fistula rethrombosis since the initial thrombectomy treatment. In addition, a total of 48 episodes of restenosis was noted, which were successfully treated by percutaneous balloon angioplasty in all instances. Taking reinterventions into consideration, the fistula remained open in 34 of 54 pa-
tients (63%) until the end of the observation period or the patient’s death. The remaining patients were lost to follow-up or underwent renal transplantation at some point in the observation period. There was no local procedure-related bleeding at the catheter site. Venous dissection necessitating stent placement occurred in three cases (3.7%). The thrombectomy procedure itself was successfully performed in these three instances, and the fistulas have remained patent for at least six months. When advancing an angioplasty catheter across the stenotic anastomosis in a patient with a forearm fistula after a first Hydrolyser run, embolization into the distal radial artery was observed. The embolus was easily retrieved by insertion of a minibasket inside a 3-F catheter (Cook) across the anastomosis to the more proximal part of the radial artery via the 7-F sheath. The embolus was dislodged into the fistula vein and subsequently removed with hydrodynamic thrombectomy. No clinical signs of pulmonary embolism with reports of chest pain, shortness of breath, and pleurisy were seen, and no other major complications were found. Kaplan–Meier survival curves are displayed in Figure 3. With the inclusion of initial failures, primary patency was calculated as 74% after 1 month (SE 5.6%), 63% after 3 months (SE 6.1%), 52% after 6 months (SE 6.3%), and 27% after 12 months (SE 6.6%). The median duration of primary access patency was 8.1 ⫾ 11.5 months. The overall fistula patency was 75% at 3 months (SE 6.6%), 65% at 6 months (SE 7.3%), 51% at 12 months (7.7%), and 22% at 24 months (SE 8.4%). DISCUSSION Early detection and treatment of complications are essential to achieve long-term access function for the hemodialysis patient. Thrombosis of dialysis grafts and fistulae represents a frequent complication [4]. Predisposing factors are an underlying stenosis or aneurysmatic vein causing flow turbulences [5, 6], as well as arterial hypotension, low cardiac output, compromised arterial inflow, and clotting disorders [7]. In the vast majority of cases, a venous stenosis is detected [6]. Surgical, as well as percutaneous, procedure results have demonstrated their effectiveness in graft recanalization. However, patency rates of percutaneous and surgical techniques are difficult to compare, as percutaneous interventions are followed beginning with the initial procedure, whereas surgical studies generally report patency starting from the time of access creation. In a retrospective comparison of surgical and percutaneous treatment for graft thrombosis, Sands et al performed 75 surgical thrombectomies and 71 percutaneous pharmacomechanical procedures and observed no significant differences in success rates, primary patencies, and rate of complications [8]. Beath-
Haage et al: Treatment of thrombosed primary AV fistulae
ard described equal initial success rates of mechanical thrombolysis compared with surgical thrombectomy [9]. Studies on surgical treatment of thrombosed native fistulae are rare. Diskin et al demonstrated a mean access survival of 244 days after surgery in 45 patients with native fistulas [10]. Wehrli, Chenevard, and Zaruba reported a success rate of 81.2% in patients with thrombosis of a newly created dialysis access [11]. Romero et al found an almost 50% cumulative patency at two years after creation of a proximal new fistula caused by late thrombosis [12]. Dapunt et al compared the results of 37 transluminal angioplasties with those of 37 operations in the treatment of stenoses and occlusions of dialysis fistulae [13]. Cumulative patency after 12 months was 31.3% for angioplasty and 19.3% for surgery. One limitation of the surgical approach is the lack of imaging guidance in the operating room to detect additional stenoses and obtain information on the often complex vessel anatomy. Conversely, performance of surgical thrombectomy in the radiologic laboratory requires a considerable amount of organization and manpower. A solely percutaneous approach is therefore a cost-reducing, attractive alternative. In recent years, percutaneous interventional techniques have been tried and established as a viable alternative means in the management of fistula dysfunction. Zaleski et al reported on 17 patients with complete thrombosis of their BresciaCimino fistulas, which were treated by angioplasty and urokinase infusion [14]. Procedural success was 82% with primary, primary assisted, and secondary patency rates at 12 months of 71, 93, and 100%, respectively. Poulain et al combined a local low-dose infusion of urokinase with PTA and thromboaspiration to achieve a 12-month overall patency in 14 native fistulas of approximately 90% [15]. Twenty out of 24 patients (83%) with occluded Brescia-Cimino fistulas were successfully declotted by Overbosch et al, employing the Hydrolyser catheter [16]. The median assisted patency was 34 weeks and was significantly shorter in fistulas than in polytetrafluorolethylene (PTFE) grafts (P ⫽ 0.002). Turmel-Rodrigues et al described an 81% initial success rate using thromboaspiration and PTA in 16 patients [17]. An 81% secondary patency at one year was reported. Comparable results, however, with far less patients, have been demonstrated by various workgroups [18–23]. The reported technical success rates are similar to our own, whereas our patency rates are less than those of Zaleski et al, Poulain et al, and Turmel-Rodrigues et al, and are rather comparable to those of Overbosch et al. These results suggest that percutaneous interventions in occluded native AV fistulas are at least as effective as surgical treatment regarding technical success and patencies. Treatment results of thrombosis and associated stenosis in synthetic grafts have been summarized by Aruny et al [24]. Clinical success rates for thrombolysis or me-
1173
chanical thrombectomy range from 75 to 94% with primary patencies of 18 to 39% at 6 months. Reported 6- and 12-month secondary patencies for thrombolysis range from 62 to 80% and 57 to 69%, respectively. Trerotola et al recently reported a 95% technical success with a three-month primary patency of 39% using the newly developed Arrow-Trerotola percutaneous thrombolytic device [25]. Both the percutaneous and the surgical approach to thrombosed AV fistulas remain difficult, partly because of the pronounced organization of thrombus material in native vessels. In many centers, patients with an occluded native fistula are referred for surgical thrombectomy. However, because of anatomic reasons, the surgeon often prefers to perform a patch insertion or fistula reanastomosis instead of mechanical thrombectomy. Our results with technical success rates of almost 90% and a primary and overall fistula patency of 27% and 51% at one year, respectively, demonstrate the effectiveness of percutaneous thrombosis treatment strategies in native fistulas. It is our workgroup’s opinion that as long as less invasive comparable therapeutic options to preserve the vessel are available and feasible, abandonment of the vascular access requiring an AV neoanastomosis should be avoided, and a percutaneous approach should be considered. Generally, declotting should be performed as an outpatient procedure. In two thirds of our cases, the procedure was performed on an outpatient basis, avoiding general anesthesia and hospitalization, thus reducing the treatment risks and costs. The remaining one third of the cases were inpatients, who had been primarily admitted to the hospital because of nonfistula-related problems. We prefer to perform fistulography in native fistulas via the brachial artery for several reasons. Arterial and venous angiomorphology can be evaluated without placing a blood pressure cuff on the upper arm and inflating the cuff to a suprasystolic pressure. This procedure can be painful for the patient and lead to involuntary movements of the extremity, aggravating satisfactory shunt visualization and documentation. The blood pressure cuff has to be left up for as short as possible to avoid thrombosis. In addition, judgment of flow velocity and spontaneous filling of collaterals, indicating the hemodynamic significance of stenoses and occlusions, is impaired with the cuff inflated, especially at the AV anastomosis. If necessary, the percutaneous intervention is monitored easily from the arterial access; also, the procedural outcome can be documented after catheter removal. Complications of arterial puncture are rare. In a retrospective analysis of 454 transbrachial arteriographies, we have observed one episode (0.2%) of significant bleeding at the puncture site [26]. It is advisable to perform the intervention as soon
1174
Haage et al: Treatment of thrombosed primary AV fistulae
as possible once thrombosis is diagnosed because the occlusion period in cases with primary technical failure was considerably longer compared with the whole group (44 ⫾ 22 h vs. 25 ⫾ 12 h). Involvement of large veins such as the brachial vein substantially complicates treatment and is another important factor determining technical success [7]. Long-term shunt functionality depends on manifold factors, such as cannulation trauma, tendency for restenosis, hypercoagulability, and frequency of reobstruction episodes [27]. Reobstruction is a frequently observed complication and is more likely to occur after thrombosis has occurred for the first time [27– 29]. The treating physician must be prepared for repeat interventions, sometimes multiple, over the months and years after the initial declotting procedure. In our study, 48 episodes of restenosis and 27 episodes of reocclusion were noted in 54 hemodialysis patients, amounting to an average of 1.4 reinterventions per patient in a mean follow-up period of 13 months. In conclusion, initial percutaneous thrombectomy is very effective in the treatment of recently occluded AV fistulae with good success rates and acceptable primary and long-term patency rates comparable to other therapy regimen such as surgical thrombectomy. It has to be stressed that the decision of whether to perform a percutaneous fistula revision or a surgical intervention should rely on the expertise of the respective center regarding access dysfunction. If percutaneous thrombolysis is available, surgical revision should be reserved for failures of percutaneous techniques. The choice of the appropriate percutaneous approach depends on the size and location of the thrombus detected by angiography. A short segment thrombosis can be easily treated with PTA alone, whereas an extensive thrombosis requires a combination of mechanical devices and/or thrombolytic agents with adjunctive balloon angioplasty. According to Gray, it is unlikely that the declotting method in dialysis grafts, whether pharmacomechanical or purely mechanical, has a significant effect on short- or long-term patency [30]. Further studies have to be undertaken to offer the native fistula patient the most beneficial and durable percutaneous treatment strategy. At present, either type of method for fistula recanalization is appropriate and should be based on the experience of the interventional radiologist with the respective therapeutic modality. The decision making need not depend on the type of dialysis access presented by the patient. Our results indicate that percutaneous thrombectomy is equally successful in native and implant shunts. Effective care for the hemodialysis patient to preserve vascular access over a prolonged period of time can only be achieved on a multidisciplinary basis with the surgeon, interventional radiologist, and nephrologist closely cooperating.
Reprint requests to Patrick Haage, M.D., Department of Diagnostic Radiology, University of Technology Aachen, Pauwelsstrasse 30, D-52057 Aachen, Germany. Email: [email protected]
REFERENCES 1. United States Renal Data System: USRDS 1998 Annual Data Report. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Disease, Bethesda, 1998 2. Windus D: Permanent vascular access: A nephrologist’s view. Am J Kidney Dis 21:457–471, 1993 3. NKF-DOQI clinical practice guidelines for vascular access. National Kidney Foundation-Dialysis Outcomes Quality Initiative. Am J Kidney Dis 30(Suppl):S174–S175, 1997 4. Zibari GB, Rohr MS, Landreneau MD, Bridges RM, Devault GA, Petty FH, Costley KJ, Brown ST, McDonald JC: Complications from permanent hemodialysis vascular access. Surgery 104:681–686, 1988 5. Valji K, Bookstein J, Roberts A, Davis G: Pharmacomechanical thrombolysis and angioplasty in the management of clotted hemodialysis grafts: Early and late clinical results. Radiology 178:243– 248, 1991 6. Schwab S, Raymond J, Saeed M, Newman G, Dennis P, Bollinger R: Prevention of hemodialysis fistula thrombosis: Early detection of venous stenoses. Kidney Int 36:707–711, 1989 7. Vorwerk D, Schu¨rmann K, Mu¨ller-Leisse C, Adam G, Bu¨cker A, Sohn M, Kierdorf H, Gu¨nther RW: Hydrodynamic thrombectomy of hemodialysis grafts and fistulae: Results of 51 procedures. Nephrol Dial Transplant 11:1058–1064, 1996 8. Sands JJ, Patel S, Plaviak DJ, Miranda CL: Pharmacomechanical thrombolysis with urokinase for treatment of thrombosed hemodialysis access grafts. ASAIO J 40:886–888, 1994 9. Beathard GA: Thrombolysis versus surgery for the treatment of thrombosed dialysis access grafts. J Am Soc Nephrol 6:1619–1624, 1995 10. Diskin CJ, Stokes TJ, Panus LW, Thomas J, Lock S: The importance of timing of surgery for hemodialysis vascular access thrombectomy. Nephron 75:233–237, 1997 11. Wehrli H, Chenevard R, Zaruba K: Surgical experiences with the arteriovenous hemodialysis shunt (1970–1988). Helv Chir Acta 56:621–627, 1989 12. Romero A, Polo JR, Garcia Morato E, Garcia Sabrido JL, Quintans A, Ferreiroa JP: Salvage of angioaccess after late thrombosis of radiocephalic fistulas for hemodialysis. Int Surg 71:122–124, 1986 13. Dapunt O, Feurstein M, Rendl KH, Prenner K: Transluminal angioplasty versus conventional operation in the treatment of haemodialysis fistula stenosis: Results from a 5-year study. Br J Surg 74:1004–1005, 1987 14. Zaleski GX, Funaki B, Kenney S, Lorenz JM, Garofalo R: Angioplasty and bolus urokinase infusion for the restoration of function in thrombosed Brescia-Cimino dialysis fistulas. J Vasc Interv Radiol 10:129–136, 1999 15. Poulain F, Raynaud A, Bourquelot P, Knight C, Rovani X, Gaux JC: Local thrombolysis and thromboaspiration in the treatment of acutely thrombosed arteriovenous hemodialysis fistulas. Cardiovasc Intervent Radiol 14:98–101, 1991 16. Overbosch EH, Pattynama PM, Aarts HJ, Schultze Kool LJ, Hermans J, Reekers JA: Occluded hemodialysis shunts: Dutch multicenter experience with the hydrolyser catheter. Radiology 201:485–488, 1996 17. Turmel-Rodrigues L, Sapoval M, Pengloan J, Billaux L, Testou D, Hauss S, Mouton A, Blanchard D: Manual thromboaspiration and dilation of thrombosed dialysis access: Mid-term results of a simple concept. J Vasc Interv Radiol 8:813–824, 1997 18. Schilling JJ, Eiser AR, Slifkin RF, Whitney JT, Neff MS: The role of thrombolysis in hemodialysis access occlusion. Am J Kidney Dis 10:92–97, 1987 19. Mangiarotti G, Canavese C, Thea A, Segoloni GP, Stratta P, Salomone M, Vercellone A: Urokinase treatment for arteriovenous fistulae declotting in dialyzed patients. Nephron 36:60–64, 1984
Haage et al: Treatment of thrombosed primary AV fistulae 20. Collier PE, Saracco GM, Young JC, Fragola JA, Contractor FM, Diamond DL: Nonoperative salvage of subcutaneous hemodialysis fistulae. Am J Nephrol 5:333–337, 1985 21. Hunter DW, So SK, Castaneda-Zuniga WR, Coleman CE, Sutherland DE, Amplatz K: Failing or thrombosed Brescia-Cimino arteriovenous dialysis fistulas: Angiographic evaluation and percutaneous transluminal angioplasty. Radiology 149:105–109, 1983 22. Pattynama PM, van Baalen J, Verburgh CA, van der Pijl JW, Schultze Kool LJ: Revascularization of occluded haemodialysis fistulae with the hydrolyser thrombectomy catheter: Description of the technique and report of six cases. Nephrol Dial Transplant 10:1224–1227, 1995 23. Goldberg JP, Contiguglia SR, Mishell JL, Klein MH: Intravenous streptokinase for thrombolysis of occluded arteriovenous access: Its use in patients undergoing hemodialysis. Arch Intern Med 145:1405–1408, 1985 24. Aruny JE, Lewis CA, Cardella JF, Cole PE, Davis A, Drooz AT, Grassi CJ, Gray RJ, Husted JW, Jones MT, McCowan TC, Meranze SG, Van Moore A, Neithamer CD, Oglevie SB, Omary RA, Patel NH, Rholl KS, Roberts AC, Sacks D, Sanchez O,
25.
26. 27. 28. 29. 30.
1175
Silverstein MI, Singh H, Swan TL, Towbin RB, Trerotola SO, Bakal CW: Quality improvement guidelines for percutaneous management of the thrombosed or dysfunctional dialysis access. J Vasc Interv Radiol 10:491–498, 1999 Trerotola SO, Vesely TM, Lund GB, Soulen MC, Ehrman KO, Cardella JF: Treatment of thrombosed hemodialysis access grafts: Arrow-Trerotola percutaneous thrombolytic device versus pulsespray thrombolysis. Radiology 206:403–414, 1998 Bu¨cker A, Vorwerk D, Gu¨nther RW: Transbrachial fine-needle arteriography with special focus on haemodialysis shunt imaging. Nephrol Dial Transplant 10:838–841, 1995 Brotman DN, Fandos L, Faust GR, Doscher W, Cohen JR: Hemodialysis graft salvage. J Am Coll Surg 178:431–434, 1994 Etheredge EE, Haid SD, Maeser MN, Sicard GA, Anderson CB: Salvage operations for malfunctioning polytetrafluoroethylene hemodialysis access grafts. Surgery 94:464–470, 1983 Gaylord GM, Taber TE: Long-term hemodialysis access salvage: Problems and challenges for nephrologists and interventional radiologists. J Vasc Interv Radiol 4:103–107, 1993 Gray RJ: Percutaneous intervention for permanent hemodialysis access: A review. J Vasc Interv Radiol 8:313–327, 1997