| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
Department of Radiology, University of Chicago, 5841 S Maryland Ave., Chicago, IL 60637.
Received April 30, 2004;
accepted after revision December 10, 2004.
Address correspondence to R. M. Peirce
(rmpeirce{at}yahoo.com).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. A retrospective review of all pharmacomechanical femoral dialysis graft declotting procedures performed in our hospital between May 11, 1993, and April 30, 2003, was performed. A total of 110 procedures were performed on 25 patients (nine males, 16 females; age range, 17-80 years; mean age, 49 years) with 30 grafts.
RESULTS. Technical success was 97.3%. Using life table analysis, 30-day postinterventional primary patency was 72%, 90-day primary patency was 46%, 180-day primary patency was 25%, and 365-day primary patency was 4%. The 30-day secondary postinterventional patency was 93%, 90-day secondary patency was 86%, 180-day secondary patency was 76%, and the 365-day secondary patency was 51%. An average of 1.96 declotting procedures were performed per year of dialysis.
CONCLUSION. Percutaneous declotting of femoral hemodialysis grafts has technical success and patency rates similar to those for percutaneous declotting in the upper extremities. Results of this series exceeded criteria established by the National Kidney Foundation in the "Dialysis Outcome Quality Initiative" for immediate patency and unassisted 3-month patency.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inherent differences exist between arm and leg grafts. These differences include variation in local anatomy and even the size, shape, and flow dynamics of the graft. At our institution, femoral PTFE grafts are placed in patients whose upper extremity access options have been exhausted. The historical infection risk, the possibility of lower extremity ischemia and amputation, and the possibility of pulmonary embolism are perils strongly considered when managing femoral grafts. The results of percutaneous declotting of femoral PTFE grafts have not, to our knowledge, been reported. Here, we report our experience of 110 percutaneous pharmacomechanical thrombolysis procedures performed in patients with PTFE femoral dialysis grafts.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
After lytic infusion, angioplasty of all stenoses was performed (Figs. 1A, 1B, and 1C). Angioplasty was continued until all stenotic segments were sufficiently dilated, all adherent thrombus was cleared, and relatively rapid blood flow within the graft was established. Arterial balloon diameters generally used were 6 cm x 2 mm, and venous balloons were 7 cm x 4 mm. Rarely, grafts demonstrating adherent thrombus within an aneurysmal segment required additional manipulation. Occlusion balloons (13 mm) with associated manual massage of the graft was used to dislodge the clot in difficult cases. Metallic stents (Wallstent, Boston Scientific) were inserted in 11 (10.0%) of 110 procedures to alleviate high-grade elastic stenoses. The stents ranged from 7 to 8 mm in width and 40 to 60 mm in length.
After the procedures, access sheaths were routinely placed and patients were dialyzed on-site (n = 79). Sheaths were removed at the dialysis center. For procedures preceding the release of these sheaths, manual pressure was applied to the puncture sites after the procedure to achieve hemostasis.
Follow-Up
All available radiologic and surgical records were retrospectively reviewed
to evaluate the duration of graft function. Discharge summaries and all other
electronically accessible records were also reviewed to identify
complications. The following parametrics were recorded: patient demographics,
details of each graft placement, individual details of radiologic
interventions, surgical revision or removal, and patient medical records and
follow-up. Because of the retrospective study design, presenting and residual
percentage of stenosis and lesion location were not available
[8]. Technical success was
defined as a successful procedure that allowed immediate dialysis without
repeat thrombosis within 24 hr. Primary patency was defined as the interval
after the initial declotting procedure (i.e., first declotting procedure in
virgin graft) until the next access thrombosis or any required vascular
interventional radiologic procedure on the graft. Secondary patency was
defined as the interval after the initial declotting procedure until surgical
revision was required or the graft was abandoned.
Kaplan-Meier analysis was used to calculate all patency data. This study pre-dates the Society of Interventional Radiology (SIR) published guidelines on reporting procedures for dialysis grafts. Therefore, much of the recommended data are not obtainable.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thirty-day postinterventional primary patency was 72%, 90-day primary patency was 46%, 180-day primary patency was 25%, and 365-day primary patency was 4% (Table 1 and Fig. 2). The 30-day secondary postinterventional patency was 93%, 90-day secondary patency was 86%, 180-day secondary patency was 76%, and the 365-day secondary patency was 51% (Table 2). The average period of unassisted graft function after the initial declotting procedure was 164 days (n = 30 grafts; range, 2-1,432 days). The average postinterventional primary assisted patency was 199 days (n = 26; range, 2-1,432 days). The average postinterventional secondary patency was 312 days (n = 17; range, 10-1,131 days). Patients in our study underwent an average of 1.96 declotting procedures per year of dialysis.
|
|
|
Complications
A 32-year-old woman experienced two separate episodes of graft infection in
temporal proximity to percutaneous intervention that required eventual
surgical revision of the graft. The first episode involved presentation of a
thigh mass in the region of the femoral loop graft 2 days after graft
declotting. On surgical exploration, a large hematoma was discovered with
associated poor tissue incorporation. This was believed to represent a
low-level infection. The graft was surgically revised, and the patient
recovered. The second incident in the same patient occurred 3 days after
percutaneous declotting when the patient developed fever, hypotension, mental
status changes, and erythema over the femoral graft site. On surgical
exploration, the graft region was grossly purulent. The infected graft was
removed successfully, and the patient eventually recovered. The same patient
also presented to the emergency department approximately 2 weeks after
successful angioplasty of her femoral graft. The patient had been receiving
successful dialysis since the angioplasty, but had developed a thigh hematoma
and overlying erythema after dialysis the preceding day. A pseudoaneurysm was
discovered, and this was revised surgically.
In addition, a 66-year-old woman experienced acute onset of left calf pain the evening after a left femoral graft declotting procedure. An emergent angiogram revealed an acute occlusion of the left popliteal artery. The patient was successfully treated with heparin and urokinase, and the symptoms resolved. The same graft in this patient required ligation 6 months later for steal phenomenon. One month after graft ligation, the patient presented with a gangrenous left toe that required amputation. No additional amputations were discovered in this group after declotting procedures.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The femoral thigh graft is anastomosed from end-to-side, typically connecting the superficial femoral artery and vein. The average femoral graft is approximately 50% longer than the arm grafts, requiring larger doses of thrombolytics. It is often used as a "last resort" in many circumstances after failure of prior accesses.
Early experiences with femoral grafts noted high infection rates. Morgan et al. [3] described their experience in 1980 using femoral vessels for dialysis access. Twenty-seven of 161 patients experienced femoral triangle infections that resulted in lower limb amputations in 22% and an overall mortality of 18%. More recent evaluations have noted less morbidity and mortality. In 2001, Coronel et al. [11] described more recent experience with femoral Thomas shunts. This study showed that the Thomas shunt offers high dialysis efficacy and provides "good permanent vascular access for HD [hemodialysis] patients." In 1995, Bhandari et al. [5] reported experience with 49 PTFE loop thigh grafts in 46 patients. In that study, all grafts functioned immediately and the 1-year graft survival was 85%. No patients required amputation in the limb of the inserted femoral graft.
The decline in femoral graft infections has not been fully characterized, but we hypothesize that it is secondary to a number of advances. Currently, most patients receive preoperative and preprocedural prophylactic antibiotics. Thorough sterilization of the femoral triangle with topical agents before surgery, declotting procedures, and dialysis has also become the standard of care. Finally, evolution of antibiotic therapy and monitoring are also probable contributors to the improved viability of femoral access. Our experience shows that infection rates among femoral grafts are comparable to rates in the upper extremity. However, one patient experienced two separate episodes of perigraft infection in temporal proximity to percutaneous declotting. Both instances required surgical intervention and IV antibiotics. This experience should serve as a reminder that femoral grafts are still at risk for infection even with the aforementioned advances in care of femoral grafts.
Our experience shows that femoral grafts function adequately and at least as well as upper extremity grafts. Femoral grafts also carry advantages in some patients. Leg grafts are longer than the typical arm graft, which facilitates cannulation at dialysis. Flow is increased in the femoral grafts, which may aid in prolonging patency [12]. Many of our patients prefer the privacy associated with the femoral location of their access. Finally, femoral grafts appear durable. Postsurgical primary patency of femoral grafts is reported at 72% for 6 months [6]. This compares favorably with the average primary patency of upper extremity grafts, reported at 58% for 6 months [13]. In our study group, 16 (53.3%) of 30 grafts were used for more than 3 years. One patient had a graft for 12 years. Three additional patients had a functioning graft for over 9 years. Many of the remaining 13 grafts were being used at the time of the study, so the number will be higher with future follow-up. Four patients died during the study period. All four had functioning lower extremity grafts at the time of death. No patient deaths were directly attributed to the dialysis graft.
Stents were inserted in patients with elastic stenoses that were refractory to angioplasty. The relatively high use of stents in this series suggests that this type of lesion may be more common in the lower extremity and pelvis than in the upper extremity. The only reported series of stent deployment for hemodialysis-related stenoses in the pelvis and legs showed effective treatment of venous outflow stenoses in thigh grafts [14]. These metallic stents are important adjuncts in the treatment of elastic or recurrent stenoses that do not respond to standard angioplasty.
The technical success of percutaneous declotting is excellent at 97.3% (107/110) and is comparable to success in the upper extremities. In our experience, both urokinase and t-PA worked equally well. We currently favor t-PA because of its lower cost and equivalent efficacy. Declotting of femoral grafts produced patency rates similar to those found in the upper extremities. Percutaneous techniques yield 90-day patencies ranging from 30% to 40% in the upper extremities [15-17].
To date, no one has reported patency outcomes with femoral grafts after percutaneous declotting. Current standards and guidelines of postinterventional patency are described in the "Dialysis Outcome Quality Initiative" (DOQI) published by the National Kidney Foundation [18]. DOQI outlines 40% unassisted patency at 3 months and 85% immediate patency rate as reasonable goals after declotting. The procedural success, as recommended by the SIR, defines anatomic success as less than 30% residual stenosis and at least one indicator of hemodynamic or clinical success [19]. Our results of 97% initial success rate and 3-month unassisted postprocedural rate of 46% exceed the National Kidney Foundation expectations. Our results also compare favorably with previously published experiences with upper extremity grafts. In addition, complication rates in our study seemed to be similar to those for declotting PTFE grafts in the upper extremities. Our experience has yielded no elevated incidence of symptomatic deep venous thrombosis or pulmonary embolism.
The limitations of this study inherently include its retrospective design. The small number of patients overall likely led to some observational bias when evaluating the outcomes. Despite the relatively large number of procedures, the reentry of patients could bias our results. Evaluation for postprocedural complications such as small hematomas and arterial emboli was not routinely and meticulously performed. Our study pre-dates the SIR published guidelines for reporting on the declot procedures of the dialysis grafts. Therefore, much of those data were not available retrospectively.
For many of these patients, femoral loop grafts are the final arteriovenous shunt for hemodialysis after failed upper extremity access. As the final option for hemodialysis, these grafts were closely monitored. Patients were referred to our clinic if they had high venous pressures during dialysis or exhibited signs of poor circulation, suggesting graft dysfunction. Patients were also referred to us when experiencing symptoms of graft insufficiency or had leg swelling. Graft patency was routinely maintained rather than recovered.
The recent literature and published DOQI recommendations suggest that a more aggressive search for and treatment of stenosis are warranted [20]. This philosophy could account for the outstanding durability noted in our test group and may provide insight to improved management of all PTFE grafts. It is likely that the number of femoral grafts placed will grow with increasing surgical demand and comfort placing them. Our data indicate that percutaneous declotting technical success and patency in femoral grafts are comparable to that in the upper extremities. The femoral graft should be considered a viable alternative in all appropriate patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
= primary patency;
= secondary patency.