| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Diagnostic Radiology, University of Technology, Pauwelsstrasse 30, D-52057 Aachen, Germany (P.H., W.P., K.S., R.W.G.) and the Department of Diagnostic and Interventional Radiology, Klinikum Ingolstadt, Germany (D.V.). From the 1998 RSNA scientific assembly. Received July 10, 1998; revision requested August 28; revision received October 2; accepted January 19, 1999. Address reprint requests to P.H. (e-mail: haage@rad.rwth-aachen.de).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIAL AND METHODS: Fifty-seven Wallstents were placed in 50 patients with symptomatic shunt dysfunction and arm swelling due to central venous obstruction. Technical success, complication, and patency rates were evaluated.
RESULTS: Stent deployment was successful in all patients, and early rethrombosis (within 1 week) was noted in one patient (2%). Seventy-three episodes of reobstruction occurred and were treated percutaneously with angioplasty alone in 54 cases (74%). Nineteen cases (26%) necessitated additional stent placement. The 3-, 6-, 12-, and 24-month primary patency rates were 92%, 84%, 56%, and 28%, respectively. Cumulative overall stent patency was 97% after 6 and 12 months, 89% after 24 months, and 81% after 36 and 48 months.
CONCLUSION: In the treatment of brachiocephalic and subclavian venous obstruction, stent placement shows excellent technical results and helps preserve vascular access for a substantial period. Multiple repeat interventions are, however, frequently required to maintain patency.
Index terms: Dialysis, shunts • Veins, grafts and prostheses, 9461.1286, 9462.1286 • Veins, innominate, 9461.752 • Veins, stenosis or obstruction, 9461.454, 9462.454 • Veins, subclavian, 9462.752 • Veins, thrombosis, 9461.752, 9462.752 • Veins, transluminal angioplasty, 9461.1286, 9462.1286
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Since Dotter (4) reported early experimental results of stent placement in vessels in 1969, vascular stents have been used to maintain the patency of the vessel after balloon angioplasty. Vascular endoprostheses have achieved excellent results in the arterial system and are a relative recent adjunct in venous applications. Nonsurgical techniques have become a mainstay in the treatment of hemodialysis-related obstruction most frequently encountered in the venous outflow tract of the hemodialysis shunt.
Central venous stenoses or occlusions in patients undergoing hemodialysis are considered to be due to high flow states, in contrast to the low flow normally seen in veins, and occur at sites of turbulence (eg, valves or kinked vessels) (5,6). Injury during or after insertion of a temporary or tunneled dialysis access catheter placed with a jugular and subclavian route has also been recognized as a risk factor for central venous obstruction, with a reported (7) prevalence as high as 40%. The prevalence varies depending on catheter type and access and is especially high for subclavian catheters (7).
In past years, percutaneous intervention in central venous obstruction has consisted of either balloon angioplasty alone (with which modest results were obtained with respect to technical outcome and primary patency) (5,6,8–10) or additional stent placement in case of angioplasty failure.
We performed this study to determine the efficacy of stent placement as the primary treatment of brachiocephalic and subclavian vein obstruction in patients undergoing hemodialysis. Herein, we report and provide follow-up data from 50 hemodialysis patients who were initially treated with self-expanding stents after balloon dilation.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Angiographically, an obstruction was situated in the subclavian vein in 35 patients and in the brachiocephalic vein in 15. Of the 35 patients with obstruction in the subclavian vein, 25 (71%) had a high-grade stricture, and 10 (29%) had an occlusion. Of the 15 patients with obstruction in the brachiocephalic vein, 13 (87%) had a severe stenosis, and two (13%) had an occlusion. Well-pronounced collateral veins almost always were present.
A central venous lesion was defined as an indication for primary stent placement due to the reported poor outcome of percutaneous angioplasty alone, partly because of the lack of available high-pressure, large-diameter balloons that fit through the 9-F or smaller sheath recommended for treatment via the dialysis access. A self-expanding metallic stent (Wallstent; Schneider, Bülach, Switzerland) was placed in all patients admitted to our interventional unit with central venous stenosis or occlusion. Stent placement was always performed after conventional balloon dilation, after which most of the obstructed vessels showed considerable residual stenosis despite application of high balloon pressures.
An insufficient result was defined as ongoing compromised shunt flow with persistent collateral veins (if observed) or a residual stenosis of at least 50%. There were three kinked stenoses and six elastic stenoses, all of which immediately collapsed after balloon deflation. The length of the obstruction ranged from less than 1 cm to 9 cm; most were short (mean length ± SD, 2.9 cm ± 1.5). Placement of two overlapping stents was required in seven patients with long stenotic segments.
Before angioplasty and stent deployment, preimplantation angiography with a digital subtraction technique was performed to study the anatomic and pathologic characteristics of the vessel after antegrade venous puncture of a shunt-leading vein or the graft itself with a 20-gauge sheath needle (Abbocath; Abbott, Sligo, Ireland). Intervention was performed with an antegrade approach via the venous outflow tract by using the Seldinger technique. Recanalization usually was attempted with a hydrophilic-coated, steerable, 0.035-inch guide wire (Terumo, Leuven, Belgium) for vessel stenosis and a straight guide wire with a movable core (Cook, Bjaeverskov, Denmark) for occlusion. After safe passage of the obstructed segment, a balloon catheter of adequate diameter, usually 8–10 mm, was advanced into the segment and dilated. Normally, balloons were dilated several times by hand, without pressure monitoring. Inflation time depended on how well the obstruction responded to dilation. In highly resistant lesions, Olbert-type balloons were preferred.
Next, a flexible, self-expanding Wallstent endoprosthesis, which is a tube woven from stainless steel alloy filaments (0.08–0.10-mm diameter), was introduced through a 9-F introducing catheter, which is flexible enough to allow implantation even in kinked vessels. Correction of the Wallstent position for exact placement was done by pulling the partially deployed stent to the affected region. Further technical properties and placement techniques have been described in detail elsewhere (11,12). Stents were dilated with a balloon of suitable size after deployment to ensure as close contact to the venous wall as possible.
Traversal of the lesion and recanalization were always attempted with a brachial approach, except when the shunt vein was severely kinked and too small for a 9-F catheter, which was the case in two patients. In these two patients, the obstruction could not be negotiated with the brachial approach, and dilation was performed with a transfemoral approach followed by stent placement with a transbrachial approach. "Through-and-through" access (13), where the transbrachially placed guide wire is pulled out through the femoral access, was performed in two patients with an extremely rigid and sharply angulated right-sided stenosis of the proximal subclavian vein.
A total of 57 stents were placed. The stent was kept as short as possible but was long enough to bridge the entire lesion with a slight overlap at its proximal and distal ends. The largest possible stent diameter was used and was dependent on lesion site. Size was adapted to the diameter of the adjacent nonobstructed vessel segment. In the 43 patients who required placement of only one stent, four 10-mm-diameter, 18 12-mm-diameter, 16 14-mm-diameter, and five 16-mm-diameter Wallstents were placed in the obstructed central vein. Two patients were treated with two 12-mm-diameter stents, one with one 12- and one 14-mm-diameter stent, one with two 16-mm-diameter stents, and three with two 14-mm-diameter stents.
Periprocedural anticoagulation was mandatory and was administered intravenously at the beginning of the intervention (5,000 IU heparin). Heparin administration was continued during hemodialysis performed immediately after the end of the procedure via the native or implant shunt or a jugular Shaldon catheter (Vygon, Aachen, Germany).
Outpatient treatment, which was preferred, consisted of subcutaneous injection of 7,500 IU of heparin three times a day for 24 hours. Inpatient treatment consisted of intravenous administration of 1,000 IU of heparin per hour for 24 hours. We recommended that the patients take 100 mg per day of acetylsalicylic acid as a chronic oral medication. No preoperative medication was administered.
Only clinical follow-up was performed, which consisted of monitoring of the dialysis protocols with respect to decreased access flow, decrease in measurement of dialysis dose, deviant urea measurements, and elevation of venous dialysis pressure. Abnormal physical findings (eg, development of arm swelling) were highly suggestive of central venous reobstruction due to stent failure. When abnormal clinical and physiologic parameters were detected, patients were referred to our interventional unit for repeat angiography. If stenosis within the stent with or without the demonstration of collateral veins or stenosis anywhere along the venous outflow tract was found, repeat intervention was indicated. Endoprosthesis shortening that uncovered the lesion was another indication for repeat intervention.
Treatment outcomes were evaluated on the basis of patency of the stent after correction of subclavian or brachiocephalic obstruction.
Primary patency of the Wallstent was diagnosed if there was no evidence of access failure related to restenosis or reocclusion by neointimal hyperplasia or thrombosis within the stent or substantial stent shortening. If shunt failure occurred owing to a condition other than stent obstruction or shortening, the stent was defined as patent. Primary patency ended when repeat intervention was necessary or when the hemodialysis fistula had to be abandoned. The interval between primary stent placement and permanent stent occlusion, including, if feasible, multiple recanalizations, was defined as cumulative stent patency. Primary and cumulative stent patency rates were calculated by means of Kaplan-Meier life-table analysis as follows: pn,n + 1 = pn x pn+1 and pn = 1 - [fn(nn - cn/2)], where p is patency, n is the total number of patients at the beginning of an interval, f is the number of failures within the indicated interval, and c is the number of patients censored within the interval.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
|
|
In 19 patients, reobstruction occurred within (n = 13) or adjacent to (n = 6) the original lesion. Proximal stent shortening necessitated stent placement in three patients, and additional deployment was necessary because of early rethrombosis within the stent in one patient. Sixteen repeat interventions required one stent per intervention for successful recanalization. To bridge the entire obstruction, we placed two additional stents in three patients and three stents in one patient at one session. A brachial approach was used in 12 of the 16 interventions, and a femoral approach was used in four. Repeat placement of one stent at two sessions due to repeated development of intimal hyperplasia was performed successfully in three patients. The interval between the two obstruction episodes ranged from 5 to 13 months (mean, 10 months).
In each of two patients, a total of three stents were deployed in two repeat intervention sessions. The venogram obtained in the first patient demonstrated thrombosis within the recently placed second stent despite intravenous anticoagulation treatment with 1,000 IU of heparin per hour. Mechanical thrombectomy with the Amplatz thrombectomy device (Microvena, Vadnais Heights, Minn) was performed, followed by placement of two overlapping Wallstents. In the latter patient, two additional stents were placed 11 months after the initial deployment. Repeat dilation was necessary 4 and 9 months after the second stent placement owing to restenosis. One month later, the patient reported to our interventional unit with arm swelling. Angiography showed occlusion of the segment with the stent. Thrombectomy and insertion of yet another stent were necessary because recanalization with a high-pressure angioplasty balloon proved to be unsuccessful.
The 3-, 6-, 12-, and 24-month primary patency rates were 92%, 84%, 56%, and 28%, respectively (Table 1). Overall stent patency was 97% after 6 and 12 months, 89% after 24 months, and 81% after 36 and 48 months (Table 2). In 35 patients, the shunt remained open until the end of the observation period or until the patient's death. The remaining patients were lost to follow-up or underwent renal transplantation.
Statistical analysis of primary and cumulative stent patency in the patients with 10-, 12-, 14-, and 16-mm-diameter Wallstents showed no statistically significant differences during the observation period. Comparison of primary and cumulative Wallstent patency rates of stents deployed on the right and left side also revealed no significant differences. Furthermore, there were no significant differences in primary and cumulative patency rates when we compared data about initial placement in the brachiocephalic vein with data about initial placement in the subclavian vein.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Abandonment and removal of the vascular access should be avoided as long as other therapeutic options are available. Surgical bypass procedures with use of polytetrafluoroethylene grafts or the saphenous vein as an initial treatment for central venous obstruction are difficult and seldom performed, because of the substantial morbidity and even mortality rates in these chronically ill patients (16). Authors (17) of earlier reports about surgical reconstruction have described modest results. Only Bhatia et al (18) reported similar success, complication, and patency rates when they retrospectively compared balloon angioplasty results with stent placement and surgical bypass results.
Early clinical applications of percutaneously placed stents were described by Zollikofer et al (19) in 1988. Today, interventional radiologic measures are the primary treatments of choice, and percutaneous transluminal angioplasty has been advocated (5,8,13,20) as an effective, safe, and comparatively inexpensive procedure with a high technical success rate. Central venous obstruction, however, has a tendency to recur earlier than does peripheral venous stenosis after percutaneous transluminal angioplasty (21,22). Severe elastic recoil after balloon dilation in central venous lesions is thought to be the underlying factor for the unsatisfactory long-term results (23). Kinked stenoses due to elongation of veins also respond well to balloon inflation but often collapse immediately after deflation. Kovalik et al (21) reported a 100% recurrence rate with a mean time to recurrence of 2.9 months following percutaneous transluminal angioplasty of elastic stenoses (ie, 
50% improvement in the diameter of the lesion after angioplasty). In patients with nonelastic stenoses, defined as lesions with more than 50% reconstitution of the vessel lumen after percutaneous transluminal angioplasty, the mean time to recurrence was 7.6 months, with a recurrence rate of 81%. Glanz et al (9) found patency rates of 50% at 6 months, 30% at 1 year, and 0% at 2 years after subclavian vein dilation; their initial success rate was 77%. Beathard (8) reported a 6-month patency rate of 29% and a 1-year patency rate of 0% after percutaneous transluminal angioplasty for central venous stenosis. Better results were reported by Lund et al (24), who reported a patency rate of 26% after a mean follow-up of 7 months after percutaneous transluminal angioplasty.
Because even longer primary and cumulative patency rates are being strived for, additional tools, including stents and, in some instances atherectomy, were sought to achieve this goal. Vascular endoprostheses have been successfully used in the arterial system and consequently were added to the basic therapeutic adjuncts in the treatment of hemodialysis-associated venous obstructions.
Stents are particularly helpful in the treatment of elastic and kinked stenoses. Furthermore, metallic endoprostheses are used for sealing circumscribed perforations and dissections and for reestablishing patency of chronic venous occlusions (25). Quinn et al (10) treated 40 patients with 50 endoluminal prostheses, using the Gianturco stent whenever possible (n = 44). The primary patency rates in their trial were 67%, 11%, and 11% at 2, 6, and 12 months, respectively. Their cumulative 1-year patency rate was 78%. They stated that no conclusions about the role of endovascular stent placement versus balloon angioplasty for the treatment of central venous stenoses and occlusions could be made on the basis of these intermediate patency results. Vesely et al (26) reported primary patency rates of 42% after 6 months and 25% after 1 year. The cumulative patency rate in 20 patients undergoing hemodialysis who were treated with Wallstents was 64% at 6 months, 56% at 12 months, and 22% at 24 months. Kovalik et al (21) reported a mean primary patency duration of 9.8 months in patients with elastic lesions and 4.2 months in patients with nonelastic lesions; they concluded that nonelastic lesions do not respond to intraluminal stent placement in central venous obstruction. Gray et al (27) placed Wallstents in 32 central veins and 24 peripheral veins, with a technical success rate of 96%. The primary patency rate was 46% after 6 months and 20% after 12 months. With repeat interventions, the cumulative patency rate increased to 76% at 6 months and 33% at 12 months (27). Mickley et al (16) implanted 15 Wallstents in 14 patients undergoing hemodialysis. Results of life-table analysis revealed primary patency rates of 70% and 50% after 1 and 2 years, respectively. The cumulative secondary patency rates were 100% and 85% at 12 and 24 months, respectively. Shoenfeld et al (28) reported a primary patency rate of 68% and a cumulative assisted patency rate of 93% at 17-month follow-up. The overall patency rate in their study is similar to that in ours, although the primary patency rates we found were lower (56% at 12 months and 28% at 24 months) and were more comparable to those in the study of Mickley et al (16).
Except for the studies undertaken by Quinn et al (10), Shoenfeld et al (28), and Mickley et al (16), stent deployment was always performed after failed angioplasty or because of recurrent stenosis but not as the initial therapy. This may be a contributing factor in the shorter cumulative patency rates in these studies compared with those for primary stent implantation, including the rates in our study.
With regard to the patency of the central venous stent, there is no evident difference with respect to the type of dialysis graft (ie, the synthetic bridge graft and the autologous arteriovenous graft). It is well known that native fistulas offer superior long-term patency rates. The observation time in patients with an autologous graft might therefore be longer; thus, follow-up data about ipsilateral central venous stent placement may be obtainable over a longer period.
Our results indicate that primary Wallstent placement for central venous obstruction is useful in the salvage and maintenance of hemodialysis access patency. Technical success rates are better than those of balloon angioplasty alone. Repeat intervention for the treatment of reobstruction can often be performed with balloon dilation alone. Twenty-six percent of reobstruction episodes necessitated additional stent placement owing to unsatisfactory results of percutaneous transluminal angioplasty. In all but four cases, the stent was in place and could be traversed and dilated easily, which led to excellent patency rates. Atherectomy has not been used as a primary therapy for central venous obstructions, mainly because of the sharp angles and thin vessel walls in this region. Conversely, atherectomy devices such as the Simpson atherectomy device (Mallinckrodt, St Louis, Mo) are effective tools for debulking neointimal tissue in case of stent reobstruction.
The choice of a suitable endoprosthesis for central veins depends on several factors. The stent should be flexible enough to be used in curved and tortuous vessels. To avoid stent dislocation and central embolization, a self-adjusting stent is advantageous because especially chronic venous occlusions may undergo progressive luminal enlargement after stent placement (14). Also, the size of the delivery system should be moderate. For these reasons, the self-expanding Wallstent is the preferred stent at many institutions, including ours. We believe that stents with available diameters of 10–16 mm should be used for central veins to maintain good stent contact with the adjacent vessel wall.
Bridging of side branches should be avoided whenever possible. Because there is a gradual increase in venous diameter at the planned stent site, we use a Wallstent that matches the size of the larger-diameter central vein. Stent length should be kept as short as possible, covering the entire lesion with a slight overlap. To avoid stent infection, stent placement should be delayed if there are signs of active infection in the patient. We did not administer prophylactic antibiotics during the procedure, and there were no cases of sepsis after deployment, such as were described by Quinn et al (29). There are no data to help prove that periprocedural application of heparin offers a substantial prolongation of acute patency after central venous stent implantation. Anticoagulants, however, have been used by some investigators to reduce chances of early rethrombosis in central venous interventions (9,13) and are recommended at least for the interval between the beginning of the interventional procedure and end of dialysis.
In conclusion, stents evidently do not decrease the likelihood of recurrence in central venous obstructions but prolong patency intervals and allow for easy repeat intervention. Restenosis or reocclusion represent an ongoing process, with no apparent regression over time. The interventional radiologist must be prepared for repeat, sometimes multiple, interventions during the years after stent placement. To minimize intimal hyperplasia, safe stent fixation with an adequate stent size and the maintenance of as small a stent overlap as possible are advisable.
Our results suggest that initial Wallstent deployment is effective in the treatment of central venous stenosis or occlusion and shows improved long-term patency rates as compared with those of other therapeutic modalities (eg, percutaneous balloon angioplasty alone). Continued technical improvements in stent design and thrombolytic devices are necessary to specify definite treatment guidelines. The aim of future stent modifications, such as pharmacologically manipulated stents, biodegradable stents, and radiolabeled stents, will be to prevent restenosis due to intimal hyperplasia. At present, we propose liberal use of stents only in central venous lesions. Early detection and treatment of complications are mandatory to preserve the functionality of the dialysis access graft during a prolonged period.
|
Footnotes |
|---|
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Yevzlin and A. Asif Stent Placement in Hemodialysis Access: Historical Lessons, the State of the Art and Future Directions Clin. J. Am. Soc. Nephrol., May 1, 2009; 4(5): 996 - 1008. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Warshauer, R. K. Archer, C. H. Selzman, H. S. Tamaddon, and M. A. Mauro Case 115: Aortic Pseudoaneurysm from Penetrating Superior Vena Cava Stent Radiology, June 1, 2007; 243(3): 901 - 904. [Full Text] [PDF]
|
|
|
|
|
|
R. D. Levit, R. M. Cohen, A. Kwak, R. D. Shlansky-Goldberg, T. W. I. Clark, A. A. Patel, S. W. Stavropoulos, J. I. Mondschein, J. A. Solomon, C. M. Tuite, et al. Asymptomatic Central Venous Stenosis in Hemodialysis Patients Radiology, March 1, 2006; 238(3): 1051 - 1056. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Ruiz, E. Gutierrez, A. Martinez, E. Hernandez, J. M. Alcazar, J. C. Herrero, and M. Praga Unilateral pleural effusions associated with stenoses of left brachiocephalic veins in haemodialysis patients Nephrol. Dial. Transplant., June 1, 2005; 20(6): 1257 - 1259. [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Surowiec, A. J. Fegley, W. J. Tanski, N. Sivamurthy, K. A. Illig, D. E. Lee, D. L. Waldman, R. M. Green, and M. G. Davies Endovascular Management of Central Venous Stenoses in the Hemodialysis Patient: Results of Percutaneous Therapy Vascular and Endovascular Surgery, July 1, 2004; 38(4): 349 - 354. [Abstract] [PDF]
|
|
|
|
|
|
L. Turmel-Rodrigues Stenosis and thrombosis in haemodialysis fistulae and grafts: the radiologist's point of view Nephrol. Dial. Transplant., February 1, 2004; 19(2): 306 - 308. [Full Text] [PDF]
|
|
|
|
|
|
V. Mickley Surgical alternatives to central venous catheters in chronic renal replacement therapy Nephrol. Dial. Transplant., June 1, 2003; 18(6): 1045 - 1051. [Full Text] [PDF]
|
|
|
|
 |
|
 P. Faries, N. J. Morrissey, V. Teodorescu, E. C. Gravereaux, J. A. Burks JR, A. Carroccio, K. C. Kent, L. H. Hollier, and M. L. Marin Recent Advances in Peripheral Angioplasty and Stenting Angiology, November 1, 2002; 53(6): 617 - 626. [Abstract] [PDF]
|
|
|
|
|
|
V. Mickley Central venous catheters: many questions, few answers Nephrol. Dial. Transplant., August 1, 2002; 17(8): 1368 - 1373. [Full Text] [PDF]
|
|
|
|
|
|
L. Turmel-Rodrigues, J. Pengloan, S. Baudin, D. Testou, M. Abaza, G. Dahdah, A. Mouton, and D. Blanchard Treatment of stenosis and thrombosis in haemodialysis fistulas and grafts by interventional radiology Nephrol. Dial. Transplant., December 1, 2000; 15(12): 2029 - 2036. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Turmel-Rodrigues, P. Bourquelot, A. Raynaud, M. Sapoval, P. Haage, and D. Vorwerk Primary Stent Placement in Hemodialysis-related Central Venous Stenoses: The Dangers of a Potential "Radiologic Dictatorship" Drs Haage and Vorwerk respond: Radiology, November 1, 2000; 217(2): 600 - 602. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
) and cumulative (
) Wallstent patency rates obtained with life-table analysis.