| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorial |
1 From the Department of Radiology, Indiana University School of Medicine, University Hospital, Rm 0279, 550 N University Blvd, Indianapolis, IN 46202-5253 (S.O.T.); and the Department of Radiology, Clinique St Gatien, Tours, France (L.A.T.R.). Received November 2, 2000; accepted November 4. Address correspondence to S.O.T. (e-mail: streroto@iupui.edu).
Index terms: Arteries, transluminal angioplasty, 91.454 • Dialysis, shunts, 91.1282, 91.442, 91.454, 91.457 • Editorials • Grafts, interventional procedures, 91.1282, 91.454, 91.457 • Grafts, stenosis or thrombosis, 91.1282, 91.442, 91.454, 91.457 • Thrombolysis, 91.1265
In 1997, the Dialysis Outcomes Quality Initiative (DOQI) was published with much fanfare (1). One of the principal goals of the vascular access guidelines of the DOQI was to increase the number of native fistulas in the United States (guideline 29). The reason for this goal was simple: throughout most of the world, the native fistula—thought to be the preferred form of permanent access for hemodialysis—was and remains the most common form of access by far, in most countries constituting 70%–90% of permanent access (2,3). In the United States, on the other hand, native fistulas were the distinct minority of permanent accesses, with synthetic grafts predominating (4–7).
In the 3 years since the publication of the DOQI guidelines, there has been a dramatic change in the approach of the surgical community to the creation of vascular access, with widespread adoption of the guidelines and several published articles (3–8) already showing that implementation of the guidelines can result in more native fistulas and better outcomes. At the time of the publication of the DOQI guidelines, it was suggested that interventional radiologists should pay close attention to reports of interventions in native fistulas and should become familiar with angioplasty and thrombolysis of native fistulas as quickly as possible, because with the shift in surgical creation of access toward native fistulas would come a shift, albeit in a delayed fashion, toward more percutaneous interventions in native fistulas (9,10).
At Indiana University, this certainly has been the case. In 1997, 1% of the interventions performed in hemodialysis access were in native fistulas. As of this writing in 2000, 10% of the interventions have been in native fistulas. This trend has been replicated all over the United States. However, the numbers of native fistulas in the United States still remain small compared with the those in rest of the world, and thus recent American literature concerning interventions in these fistulas is relatively scarce (11–13). Incidentally, it should be noted that some of the earliest pioneering work in this area was done in the United States nearly 20 years ago by Hunter et al (14). Fortunately, our colleagues in Europe have been actively evaluating this field for some time, and in the year 2000 alone there were two relatively large series emanating from Europe that describe percutaneous management of thrombosed and failing native fistulas (15,16).
The article by Manninen et al (17) in this issue of Radiology is yet another example of contributions to this field from Europe, this time from Finland. The authors describe 103 interventions in 51 patients with failed (n = 12) or failing fistulas. Although somewhat difficult to evaluate because thrombosed and patent fistulas were not reported separately, the results appear to compare favorably with those reported in other recent series, with a 92% clinical success rate and primary patencies at 6 months and 1, 2, and 3 years of 58%, 44%, 40%, and 32% and secondary patencies of 90%, 85%, 79%, and 79%, respectively. In addition to further contributing to the growing body of literature on native fistula intervention, these authors introduce two relatively novel concepts, the first being routine use of a transarterial approach for native fistula intervention and the second being the use of brachytherapy for the prevention of restenosis in native fistulas.
With respect to the routine use of the transbrachial approach for native fistula interventions, we are reminded of a favorite Gary Larson The Far Side cartoon in which two mosquitoes are shown, one with its proboscis embedded in the skin and grossly bloated like a balloon, while the other screams "Pull out! You’ve hit an artery!" In many respects, this cartoon sums up what we and many others believe about arterial access for dialysis access interventions, whether in grafts or native fistulas. We believe that it should be avoided if at all possible. In percutaneous interventions in grafts, it is virtually never needed. On the other hand, because of the anatomy associated with native fistulas, occasionally transarterial access is needed to successfully complete interventions in native fistulas, in particular thrombolysis and nonmaturing fistulas, but in our practices it is virtually never used for failing mature fistulas.
The argument for a transvenous approach is straightforward: The dilated vein of the fistula is meant to be punctured and is larger than the brachial artery, except in immature fistulas. Veins of mature fistulas only rarely exhibit spasm, in stark contrast to the high rate of arterial spasm (a reflection of injury, in our opinion) in the present series.
The approach to failing and failed native fistulas has been described in detail, and we acknowledge that there are certainly divergent approaches. With respect to thrombosed fistulas, Turmel-Rodrigues et al (16) needed to use the transarterial approach only seven times in 56 procedures. Haage et al (15) describe routine (retrograde) puncture of the brachial artery with a 22-gauge sheath needle for diagnostic purposes, but they did not use that approach for the intervention.
Manninen et al (17) make the argument that the transbrachial approach, in their experience, facilitates crossing the anastomosis and allows better depiction of anatomy than a retrograde transvenous approach. In our experience, however, the antegrade approach is not easier or worse than the retrograde approach for traversing stenoses; we simply use the potentially less injurious approach first. Although it may be true that the arterial approach allows better depiction of the inflow arteries and anastomotic region, we do not believe this argument supports use of the artery for the intervention with larger sheaths (as large as 7 F).
The arguments against routine use of the brachial artery are numerous, in our opinion. First, using the doctrine of primum non nocere, we believe that one should avoid puncturing an otherwise normal artery when there are other ways to get the job done—specifically, the retrograde venous approach. It is well known that one of the causes of failure of dialysis grafts and fistulas is the development of intimal hyperplasia at the site of repeated punctures, a process exacerbated by the high flow in dialysis access circuits. It is possible, though to our knowledge never tested, that repeated trauma to the artery in this setting might also lead to stenosis. Since interventions in dialysis access circuits are repeated at various intervals throughout the life of the access, this potential for iatrogenic stenosis should be considered one of the theoretic risks of a routine transbrachial approach. In response to the concern of inciting intimal hyperplasia in the artery, the authors state that this did not occur, yet it was not specifically evaluated, and the follow-up period of the study was too short to rule out the problem. By way of comparison, it is probably safe to assume that currently, nephrologists would not consider repeated cannulation of the brachial artery for hemodialysis.
Second, dialysis access interventions, whether thrombolysis or angioplasty, are a purely outpatient procedure. Although the DOQI (1) allows 24-hour observation (guideline 21), that dispensation was made for surgery, not interventional radiology. One of the strengths of the percutaneous interventional approach is its outpatient nature. In many institutions, however, overnight observation remains the standard of care after transarterial access. Thus, routine use of the transarterial approach might increase the need for hospitalization in this patient population, something the DOQI specifically sought to avoid.
Third, the strongest argument against a routine transarterial approach is the potential for development of complications at the puncture site that are relatively unique to arterial access. Pseudoaneurysms and significant hematomas are rare when the retrograde transvenous approach is used in native fistulas. Haage et al (15) report no pseudoaneurysms or hematomas, and Turmel-Rodrigues et al (16) report a single venous pseudoaneurysm (one [1%] of 94) and no hematomas in their series; Beathard et al (11) describe no complications with a transvenous approach.
In contrast, Manninen et al (17) report a complication rate of 12% (12 of 103); all of these complications were related to the brachial arterial puncture, and four required surgical repair (two pseudoaneurysms and two hematomas). In addition, they report eight "minor" complications, including six hematomas and two pseudoaneurysms requiring compression repair. It can be argued that the latter should have been classified as major complications; the Society of Cardiovascular and Interventional Radiology quality improvement guidelines for dialysis access suggest that any complication requiring more than nominal therapy be considered major (18). The substantial numbers reported in the present series reflect the definite trauma that occurs to the artery when a 5- to 7-F introducer sheath is placed in it and further underscore our concerns about long-term complications at the puncture site.
In defense of their approach, the authors cite studies with similar complication rates, although the references we cited earlier show smaller complication rates, and the complications were less severe. Most important, the complications cited are venous complications that rarely require surgery. With application of the apples-to-apples comparisons described earlier, similarly sized recent series concerning native fistula thrombolysis report substantially fewer complications when selective brachial or "diagnostic only" brachial approaches were used: 0%–1% (11,15,16) versus 6% major complications (six of 103, applying the standards of the Society of Cardiovascular and Interventional Radiology) in the series of Manninen et al (17). If the purpose of percutaneous interventions is to keep patients from requiring surgery, then certainly the transbrachial approach failed to achieve that goal in a sizeable number of patients in this series.
With respect to radiation therapy, brachytherapy for the prevention of restenosis has received a great deal of attention recently, particularly in the cardiology literature describing its use in preventing in-stent restenosis in coronary arteries. Limited reports (19,20) of the use of brachytherapy in hemodialysis circuits have been published. The results to date have not been conclusive and are certainly not dramatic. Manninen et al (17) describe a subset of five patients in whom iridium 192 endovascular brachytherapy was used; three had restenosis at the same site within 3 months after initial angioplasty. The secondary patencies achieved in this group are identical to those achieved in the study overall. From this, the authors conclude that brachytherapy is "feasible and seems not to have acute complications."
Although this conclusion may be true, we believe these results are currently too limited to allow any firm conclusion regarding brachytherapy in native fistulas. It is not at all clear that any benefit was gained. Furthermore, emerging reports suggest that brachytherapy may be detrimental in the long term in coronary arteries (21,22), and this was also suggested by the results of an animal study of hemodialysis circuits published in 1999 (20). We believe large prospective randomized trials, with clear-cut inclusion and exclusion criteria and clear definitions of outcome variables, are needed. We further believe the temptation to use brachytherapy in this patient population should be resisted until the results of randomized trials are available.
Finally, we believe that combining thrombosed and failing fistulas should be avoided when studying native fistula interventions, just as is the case for grafts, to allow meaningful comparisons between series. Furthermore, we believe combining forearm and upper arm fistulas should be avoided because it appears outcomes may differ between these two forms of access (16,23,24).
Ideally, prospective randomized trials are the preferred way to determine the outcomes of interventions. Reporting and quality assurance standards, such as those published by the Society of Cardiovascular and Interventional Radiology (18,25), should be adhered to strictly. As the prevalence of native fistulas increases, the opportunity to perform interventional studies will increase, and radiologists should take the lead in this area. Along with this leadership position should come an absolute mandate to protect venous access sites by carefully considering the implications of those interventions. Certainly, new approaches and technology should be tested. The hemodialysis community depends on us for that.
FOOTNOTES
See also the article by Manninen et al (pp 711–718 ) in this issue.
REFERENCES
Related Article
This article has been cited by other articles:
|
|
 |
|
  S. D. Braun, H. I. Manninen, and E. T. Kaukanen Brachial Arterial Access Drs Manninen and Kaukanen respond: Radiology, September 1, 2001; 220(3): 830 - 831. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
