| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Departments of Radiology (B.L.D.) and Biostatistics and Epidemiology (M.L.), Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195; the Department of Radiology, St Francis Medical Center, Peoria, Ill (F.C.); the Department of Radiology, University of California Los Angeles Medical Center (T.O.M.); the Department of Radiology, Miami Vascular Institute, Miami, Fla (G.Z.); and the Department of Radiology, Fairview Riverside Medical Center, Minneapolis, Minn (A.H.C.). Supported by MicroTherapeutics, Irvine, Calif. Received March 5, 1998; revision requested May 4; revision received January 20, 1999; accepted March 25. Address reprint requests to B.L.D.
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: In this randomized controlled trial, 77 patients with synthetic forearm loop shunts that were thrombosed were randomly assigned to undergo pharmacomechanical thrombolysis with a pulsed spray (n = 34) or a thrombolytic brush catheter (n = 43). The following findings were evaluated: declotting time, urokinase dose, procedure time, complications, and shunt patency at the first dialysis session and at 3 months. All data were collected prospectively in an unblinded manner.
RESULTS: The total amount of urokinase used, including secondary interventions, was 243,657 IU with the catheter versus 476,563 IU with the pulsed spray (P = .001). At 15 minutes, clot lysis was successful in 66% of the patients with the catheter versus in 19% with the pulsed spray (P = .001). At 30 minutes, clot lysis was successful in 98% with the catheter versus 47% with the pulsed spray (P = .001). Procedure complication rates and patency at 3 months were similar for the catheter and the pulsed-spray groups.
CONCLUSION: Use of the Cragg catheter with urokinase offered faster and more complete clot lysis than did use of the pulsed spray with urokinase. The amount of urokinase used with the catheter was half that used with the pulsed spray. Shunt patency at 3 months was similar for the two treatment methods.
Index terms: Dialysis, shunts, 91.457 • Interventional procedures, complications, 91.44, 9.454, 91.457, 91.458 • Thrombectomy, 91.1279, 91.128 • Thrombolysis, 91.1265, 91.128
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
A few years ago, pulse-spray pharmacomechanical thrombolysis with urokinase was the standard method for declotting of an arteriovenous shunt (1,2). More recently, a number of mechanical devices have been used to increase clot dissolution with or without pharmacologic thrombolysis (3–6). This study was performed to evaluate the effects of the Cragg thrombolytic brush catheter (TBC; MicroTherapeutics, Irvine, Calif) as an adjunct to pharmacologic thrombolysis with urokinase. In a randomized controlled trial, we compared the effectiveness of the Cragg catheter to that of the pulsed spray.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Randomization was conducted at the study sites by having each patient open a sealed envelope that directed enrollment into the catheter (test) or pulsed-spray (control) group. The number of catheter and pulsed-spray envelopes was equal (1:1) at the outset of the study.
Inclusion Criteria
Patients were eligible for the study if they required chronic hemodialysis via a synthetic forearm loop graft and were older than 18 years. The graft could be thrombosed for no longer than 2 weeks and had to be in place at least 4 weeks. At the beginning of the procedure, a catheter was advanced across the venous anastomosis to allow acquisition of angiographic studies that were used to determine the extent of thrombus.
Exclusion Criteria
The criteria for exclusion included the following: recent or active bleeding; known clotting disorders; a contraindication to heparin or urokinase; suspicion for graft infection on the basis of clinical examination or laboratory tests; recent (<10 days) arteriovenous shunt surgery; a native arteriovenous fistula; a graft in the upper arm or a straight synthetic graft, whether in the forearm or upper arm.
Patients
Eighty-one patients were enrolled in the study, with four patients enrolled twice. These four patients were analyzed only once, with use of data from the first enrollment. This yielded a sample of 77 unique patients. Enrollment at each of the five sites was 35%, 31%, 14%, 10%, and 9%. One site was a university-related hospital, whereas four sites were community-based postgraduate teaching centers. Personnel at all sites were active in the treatment of thrombosed and stenosed synthetic dialysis shunts at the time this study began, and all procedures were performed by interventional radiologists who were experienced in dialysis shunt intervention.
Forty-three patients were randomly assigned to the catheter (treatment) group and 34 to the pulsed-spray (control) group. There was no difference (at the 95% confidence level) between the catheter and pulsed-spray groups, respectively, regarding age (60.2 vs 64.2 years), gender (51.2% men vs 52.9% men), or duration of thrombosis (mean of 51.2 hours and range of 6–192 hours vs mean of 52.4 hours and range of 12–192 hours). The percentage of clot seen on the preliminary angiogram was not different for the two groups (95.7% vs 91%, P = .175, Wilcoxon rank sum test). Curiously, the catheter group was 2.7 times more likely to have had a history of previous shunt intervention than was the pulsed-spray group (P = .039, 
2 test).
The definition of successful lysis was less than 20% residual clot.
Cragg Thrombolytic Brush Catheter
Two crossed 6-F sheaths (type not specified in the protocol) were placed in the loop graft near its apex, which allowed the sheaths to be pulled back so they could be positioned without overlap, if necessary. Through these sheaths, crossed 5-F multiple-side-hole infusion catheters (type not specified) were placed, and an angiogram of the clotted graft was obtained. The extent of clot was defined. Thereafter in all cases, a 1-minute clot-lacing procedure with urokinase and heparin preceded use of the catheter. A solution of 62,500 IU of urokinase (Abbokinase; Abbott Labs, North Chicago, Ill) admixed with 1,000 U heparin and mixed with saline solution to a final volume of 10 mL was pulsed in 1-mL boluses into each of the two catheters during 1 minute. If necessary, the lacing catheters were moved so that all of the arterial and venous limb clot was pulsed. At the end of the lacing procedure, the total dose of urokinase was 125,000 IU and of heparin was 2,000 U. After the lacing procedure was completed, an angiogram was obtained. The thrombolytic starting time for the catheter procedure was defined as the time when lacing with urokinase and heparin began. After lacing, the infusion catheters were removed.
The Cragg thrombolytic brush catheter consists of a 6-mm-diameter nylon brush that is integral with the distal end of a braided wire drive cable (Figure). This drive cable passes through a 6-F catheter. The device was prepared by connecting the brush cable and catheter to the motor drive unit, which ensured proper rotation of the brush. Insertion of the nylon brush catheter into the dialysis graft was facilitated by withdrawing the brush into the catheter, thereby protecting the nylon bristles. After the retracted catheter was inserted through the sheath pointing toward the arterial anastomosis, the catheter was advanced to the arterial anastomosis, and the nylon brush was exposed. If the catheter could not be adequately positioned in this manner, the brush was removed. The catheter was then positioned over a guide wire; then the wire was removed, and the brush was advanced through the appropriately positioned catheter and exposed by retracting the catheter. While several milliliters of a 30-mL solution containing 62,500 IU of urokinase and dilute contrast material was injected through the side arm, the catheter was activated for approximately 2 minutes. During this time, the catheter was slowly withdrawn through the arterial limb of the graft.
|
If less than 20% of the clot remained, the catheter procedure was considered to be successful, and secondary interventions could be pursued. If residual clot remained, catheter treatment was continued an additional 15 minutes with an additional aliquot of 62,500 IU of urokinase until less than 20% clot remained or a total lysis time of 30 minutes had passed. If less than 20% clot was noted at any time, the procedure could be stopped. Otherwise, a final angiogram was obtained at 30 minutes from the time the procedure was started. At this time, secondary interventions could be performed.
Pulse-Spray Pharmacomechanical Thrombolysis
We used the method of Valji et al (2) for pulsespray pharmacomechanical thrombolysis. Crossed 6-F sheaths (type not specified in the protocol) were placed in the loop graft near its apex, which allowed the sheaths to be pulled back if necessary so that they could be positioned without overlap. Crossed 5-F multiple-side-hole infusion catheters (type not specified) were placed through these sheaths, and an angiogram of the clotted graft was obtained, as well as confirmation of a patent venous outflow beyond the venous anastomosis. Initially, 5,000 U of heparin was administered systemically or into the clotted graft. Next, 125,000 IU of urokinase in a 5-mL saline solution was pulsed in 0.2–0.3-mL aliquots into each catheter every 30 seconds for 15 minutes, for a total dose of urokinase of 250,000 IU. An angiogram was obtained at 15 minutes, and if less than 20% residual clot was noted, secondary interventions were performed.
If residual clot of more than 20% remained, a second dose of 250,000 IU of urokinase was administered in an identical manner during 15 minutes. If less than 20% clot was noted at any time, the procedure could be stopped; otherwise, a final angiogram was obtained at 30 minutes from the time the procedure was started.
Secondary Interventions
If thrombolysis with either the catheter or pulsed spray was successful (less than 20% residual clot), secondary procedures were performed that were directed at correcting an underlying stenosis. Such interventions included percutaneous transluminal angioplasty, atherectomy, stent placement, or dislodgment of focal residual clot by means of balloon maceration. The location and nature of the stenosis were noted.
If thrombolysis was unsuccessful (more than 20% residual clot) at the end of 30 minutes, the physician was permitted to proceed with secondary interventions for clot removal. These included the use of angioplasty balloons, compliant balloons, clot aspiration, and additional pulsed-spray therapy. The Cragg catheter, however, could not be used for secondary therapy. Once clot had been successfully removed, interventions directed at treating an underlying stenosis were performed as described previously.
Patient Follow-up
Complications related to the procedure were recorded and treated as necessary. All patients were evaluated for perforation of the artery, graft, or vein; signs of hemorrhage, hematoma, or arterial embolization; and allergic reaction to contrast media. Clinical signs and symptoms of pulmonary embolization were assessed, including shortness of breath, chest pain, prolonged hypoxia unrelated to sedation on the basis of transcutaneous oxygen saturation, sudden tachycardia, and tachypnea. During the study, infection of a graft or death of any patient was reported.
All patients were followed up at the initial dialysis session after thrombolytic treatment of the graft. Thereafter, patients were followed up for graft patency at 3 months. Data regarding successful dialysis, venous dialysis pressures, and shunt pulse were collected from the dialysis centers.
Statistical Analysis
All data were reviewed by an author (M.L.). Statistical tests included Wilcoxon rank sum test, 2 test, and the Fisher exact test, as indicated.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
2 test).
Procedural success without complication after all interventions was 93% (38 of 41 procedures) for the catheter group and 97% (32 of 33 procedures) for the pulsed-spray group. This difference was not significant (P = .404, 2 test). Two major complications (5%) were reported in the catheter group. One patient experienced symptomatic arterial embolization and required urokinase infusion overnight, and a second patient was admitted to the hospital for observation of transient shortness of breath that was believed to be due to pulmonary embolization. Because the patient's condition improved quickly without specific therapy for pulmonary embolization, no diagnostic studies were performed (such as ventilation perfusion scintigraphy or pulmonary angiography); the possibility of pulmonary embolization was based on clinical assessment alone. One major complication (vein perforation) (3%) was noted in the pulsed-spray group. The difference in incidence of major complications between the catheter and pulsed-spray groups was not significant (P = .689, 2 test).
Minor complications were noted in 12% (five of 41) of the catheter procedures and 6% (two of 33) of the pulsed-spray procedures that were often not directly related to use of the catheter or pulsed spray. Statistical analysis showed no significant difference in minor complications between the two groups (P = .370, 2 test). Asymptomatic arterial embolization was associated with the precatheter-lacing procedure in three patients (7%) in the catheter group and one patient (2%) in the pulsed-spray group.
Shunt patency at 3 months for patients with a successful first dialysis was 44% (15 of 34 patients) in the catheter group compared with 58% (15 of 26 patients) in the pulsed-spray group. This difference was not significant (P = .297, 2 test). Despite the lack of significant differences between the two groups, these patency rates are of clinical interest regarding the use of thrombolytic techniques for obtaining secondary patency in synthetic loop dialysis grafts (see Discussion).
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In this study, we compared two pharmacomechanical techniques for treating thrombosed synthetic dialysis shunts, both of which involve use of urokinase. We found that the catheter method clears clot faster and more completely than does the pulsed-spray method. Furthermore, this greater efficiency of clot lysis with the catheter was achieved with use of approximately half the dose of urokinase that was used with the pulsed spray. We believe that the catheter method is more effective than the pulsed-spray method because it optimizes fragmentation of the clot. This potentially exposes a greater surface area of clot to the enzymatic effects of urokinase, but the role of pharmacologic clot dissolution remains ill defined for both the pulsed-spray and catheter methods.
Although the catheter reduced declotting time and the dose of urokinase used for thrombolysis, we did not observe any reduction in procedure-related complications. There was one case of clinical pulmonary embolization in the catheter group, which was not surprising. In fact, acute pulmonary emboli occur with most declotting procedures including the pulsed-spray method. Pulmonary embolization appeared on 35%–59% of pulmonary perfusion scans obtained after pulse-spray pharmacomechanical thrombolysis (9,10) and was as high as 91% before the pulsed-spray method in an animal model (10). Clearly, a lot of graft thrombus is not being dissolved with the pulsed-spray technique during the initial declotting phase. The risk for pulmonary embolization with the catheter also exists. Perhaps no method or device can completely prevent pulmonary embolization during declotting of an arteriovenous shunt. Even when the intent is to extract thrombus (either surgically or with catheter techniques), there is always the risk that sudden restoration of flow in the shunt may flush clot into the pulmonary arterial system.
Another complication seen in both groups in this study was arterial embolization. In four of the five occurrences, this was the result of the lacing procedure rather than of the pulsed-spray or catheter procedure. Reflux of thrombus into the arterial system was likely a result of insertion of catheters through clot and across the arterial anastomosis, pulsed-spray lysis, or both. Only one patient (catheter group) developed symptoms of arterial embolization and underwent thrombolytic therapy.
We did not perform a cost comparison, but we recognize that this is an important topic for further analysis from the perspective of both the health care provider and the physician. In the future, the selection of the most appropriate method for declotting of dialysis shunts may well be dominated by cost considerations, since many clinical studies have revealed that the success for any mechanical method is 80%–95%, and no device offers superior safety. In a simple analysis of cost, declotting method A can be compared with declotting method B by comparing costs for urokinase, catheters, angiography room time, personnel time, and so forth. But the issue of cost for dialysis shunt declotting goes beyond an assessment of these line items. In part, this is because the cost of these devices is decreasing, as many new catheters compete in clinical practice. Comparison of costs is complex, and the assumptions upon which cost analyses are made change frequently. For these reasons, and the fact that our study was designed to evaluate efficacy not cost issues, we did not attempt to formulate a comparison of costs between the methods.
We evaluated patency and found no difference between the catheter and pulsed-spray methods. After successful first dialysis, the 3-month shunt patency was 44% for the catheter group and 58% for the pulsed-spray group. These patency rates are comparable to those reported in most other series (3–8). Patency after declotting is correlated with recurrence of stenosis and independent of the method used for declotting. Although declotting is successful in 80%–95% of cases, the outlook for long-term patency is poor. We await advances in dialysis shunt patency that prevent recurrent neointimal stenosis and perhaps even inhibit primary shunt stenosis. Until these breakthroughs are attained, however, patency results may be meager after successful declotting of dialysis shunts achieved with either percutaneous or surgical methods.
Is the Cragg catheter the final answer for percutaneous declotting of thrombosed dialysis shunts? A few years ago, the interventional radiology community seemed content to accept pulse-spray pharmacomechanical thrombolysis as a standard of reference. Now, with new declotting methods reported almost monthly, it may be many years before we agree on a method that provides the best approach. On the basis of results in our study, we are confident that the Cragg thrombolytic brush catheter offers more efficient clot lysis, in less time with use of a reduced dose of urokinase, than does pulse-spray pharmacomechanical thrombolysis. In this regard, the Cragg catheter is a step forward.
|
Footnotes |
|---|
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Castañeda, R. Li, J. Patel, J. R. DeBord, and J. L. Swischuk Comparison of Three Mechanical Thrombus Removal Devices in Thrombosed Canine Iliac Arteries Radiology, April 1, 2001; 219(1): 153 - 156. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
