| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Full paper |
Department of Radiology, St George's Hospital, London, UK
Correspondence: Dr Matthew Matson, Department of Radiology, Royal London Hospital, Whitechapel Road, London E1 1BB, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodResultsDiscussionReferences |
|---|
Coagulation of aneurysms by direct percutaneous thrombin injection was described by Cope and Zeit in 1986 [2] and subsequently by Liau [3] who used bovine thrombin. A number of series have followed, with one series finding direct percutaneous thrombin injection superior to UGCR in a non-randomized retrospective comparison [4–7]. We present the results from direct percutaneous injection of Beriplast P, a fibrin adhesive (Centeon Pharmaceuticals Ltd., Marburg, Germany). The fibrin adhesive consists of a substrate, containing human fibrinogen, factor XIII and bovine aprotinin, and a "starter" containing human thrombin and calcium chloride in solution (Table 1
). The substrate and starter react within the pseudoaneurysm to produce fibrin while the blood within the pseudoaneurysm should provide further substrate for the reaction, leading to thrombosis of the pseudoaneurysm.
|
|
Materials and method |
|---|
|
TopAbstractIntroductionMaterials and methodResultsDiscussionReferences |
|---|
The contralateral femoral artery was punctured to treat the femoral artery pseudoaneurysms using balloon protection. A 6 F sheath was then inserted into the artery and a catheter was passed over the aortic bifurcation. Selective femoral angiography with oblique projections if necessary was performed to identify accurately the site of thepseudoaneurysm. The catheter was then exchanged for a balloon catheter, with the balloon left deflated across the neck of the aneurysm. The pseudoaneurysm was punctured under direct ultrasound control and the balloon inflated. Correct positioning of the balloon was suggested by cessation of pulsatile flow through the needle. A gentle injection of contrast medium through the needle confirmed its position in the pseudoaneurysm. The fibrin adhesive was injected through the needle; the fibrinogen substrate first, immediately followed by the activated thrombin. The balloon was left inflated for 10 min, during which time the guidewire was removed from the balloon catheter and heparinized saline flushed regularly through the lumen. After deflation of the balloon, ultrasound and repeat angiography were performed to confirm thrombosis of the pseudoaneurysm. If significant flow was still demonstrated, the balloon was re-inflated and further fibrin adhesive was injected. The same technique was used in the case of the brachial pseudoaneurysm, except that access was via the radial artery. After the first 20 cases, following reports of successful thrombosis of pseudoaneurysms without the use of balloon protection, subsequent patients (Group B) were treated with direct injection as the first line procedure, and without the use of a protective balloon. These procedures were performed on a radiographic couch, although ultrasound was the main modality used to direct the procedure. For this group of patients, pedal pulses were recorded before and after the procedure. Eight patients were treated this way.
Follow-up
Follow-up for all procedures was with ultrasound the following day. The patient was discharged if the pseudoaneurysm was completely thrombosed and no further routine imaging was then performed. If the pseudoaneurysm was mainly thrombosed, but a small amount of flow could be seen, the patient was managed conservatively and a further ultrasound performed 1day later. Subsequent follow-up was clinical examination by vascular surgeons or cardiologists, with re-referral if necessary.
The follow-up protocol was based upon evidenceof pseudoaneurysm recurrence following treatment with UGCR. Kumins et al [8] demonstrated that serial ultrasound scanning of patients following successful UGCR detected no further recurrences in any patient in whom the first ultrasound after UGCR was normal. Follow-up of another group of patients at our institution for6 months following thrombin injection of femoral pseudoaneurysms has confirmed this observation [9].
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodResultsDiscussionReferences |
|---|
. The mean delay between angiography and diagnostic ultrasound was 8 days (range 1–66 days) and between angiography and therapeutic embolisation it was 9.5 days (range 1–67 days). The mean volume of the pseudoaneurysms measured on ultrasound (as calculated by length x width x depth/2) was 19.4 cm3 (range 2–118 cm3).
|
Primary success after a single injection of fibrin adhesive was achieved in 16 patients (80%). One of these showed some filling of the pseudoaneurysm neck at the end of the procedure but had completely thrombosed the following day, and one continued to show some filling of the neck on ultrasound at 24 h but had thrombosed completely the following day.
Two patients required supplemental procedures. One patient, in whom UGCR before embolisation had been unsuccessful, had some filling of the pseudoaneurysm neck on follow-up ultrasound after embolisation. This thrombosed following 10 min of further UGCR. One patient had incomplete thrombosis of the pseudoaneurysm following the first injection, but a second treatment the next day resulted in complete thrombosis. Overall, therefore, 18 pseudoaneurysms (90%) were successfully treated.
Group B
All patients received fibrin adhesive in a dose of 1–6 ml (mean 2 ml). All pseudoaneurysms were completely thrombosed on initial ultrasound after a single treatment. Pedal pulses were unchanged before and after each procedure.
Complications
In the first group of patients treated with an occlusion balloon (Group A), 2 (10%) of the 20 patients experienced primary failure, both of whom required surgical repair of the pseudoaneurysms. One patient who had one unsuccessful treatment that was followed by an apparently successful treatment the following day, but the pseudoaneurysm was seen to be patent on ultrasound 1 day after the second procedure. The other patient had three treatments on successive days but the aneurysm was not completely thrombosed at the end of the procedure or at first follow-up. This patient developed cellulitis, with pain, redness and induration of the soft tissues around the puncture sites, which was treated with oral antibiotics. There was concern that the pseudoaneurysm itself was infected, and that this would make surgery more complicated. However, the pseudoaneurysm did not appear infected at surgery, a pinhole sized defect in the artery was easily closed with primary suture and subsequent cultures were negative.
One late secondary failure was seen in this group. This was in the patient whose initial treatment was unsuccessful but who had a second treatment with fibrin adhesive the following day, following which ultrasound demonstrated thrombosis of the pseudoaneurysm. She was discharged but presented 1 month later with a symptomatic recurrence of the pseudoaneurysm.
Minor complications occurred in two patients in Group A. In one patient, a small amount of fresh thrombus was seen around the tip of theballoon catheter during the procedure. Heparinized saline flushes were continued, and at the end of the procedure, following successful thrombosis of the pseudoaneurysm, an angiogram of the distal lower limb vessels was normal. In another patient, a small amount of non-occlusive thrombus was seen in the profunda femoris artery at the end of the procedure. Neither patient had any associated clinical symptoms.
No complications occurred in the eight patients treated without balloon occlusion (Group B). No change in pedal pulses was observed.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodResultsDiscussionReferences |
|---|
UGCR has emerged as the first line treatment of pseudoaneurysms since the description by Fellmeth in 1991 [15]. The success rate is 70–100%, with a complication rate of about 3%, including a 1% risk of rupture [16]. The two main disadvantages of this technique in everyday practice are the demand on the operator's time and pain for the patient. Compression times of up to 210 min have been used [8], but this would not be feasible in many ultrasound departments within the UK. Digital compression applied after ultrasound identification and characterization of the pseudoaneurysm has been demonstrated to shorten compression time [17] and may help relieve pressure on an ultrasound department. Other groups have shown that mechanical compression devices placed without image guidance have similar efficacy to those placed with ultrasound guidance [18]. Our policy has been to apply compression for a limited time only.
Symptomatic patients with pseudoaneurysms and associated bruising are generally very tender in the groin and find UGCR uncomfortable. Although we have never resorted to administering general anaesthesia to perform UGCR, as described in Fellmeth's series [15], we regularly use opiate analgesia. Despite this, there is a significant group of patients who are unable to tolerate UGCR or in whom UGCR is unsuccessful. Our success rate of 50% for UGCR is at the lower end of reported success rates, but is in line with others [1], and most likely reflects our policy of a maximum compression time of 30 min owing to constraints on operator's time. In this respect, it is probably more representative of everyday practice than some of the initial series that described and promoted UGCR.
The management of patients in whom UGCR has been unsuccessful has traditionally relied on open surgery. We treated some patients with iatrogenic pseudoaneurysms by direct percutaneous injection of coils, with a protective balloon placed across the pseudoaneurysm neck [19]. One of the disadvantages of this technique is that it leaves a palpable subcutaneous lump of coils. The other potential pitfall is that introduction of foreign material may act as a nidus for infection.
The use of a liquid agent to effect thrombosis overcomes these problems. In 1986, Cope and Zeit [2] used thrombin to treat three pseudoaneurysms of the iliac, femoral and peroneal arteries and an aneurysm of the hepatic artery. We were concerned that any liquid agent would be more likely to escape from the pseudoaneurysm and enter the distal circulation. It was hoped that the use of a protective balloon might reduce this risk, as well as providing stasis of blood within the pseudoaneurysm, which might aid thrombosis. However, we changed our practice following reports of use of thrombin to treat pseudoaneurysms without the use of a protective balloon and without encountering the complication of distal embolisation. It is interesting to note that no complications occurred in Group B, although this may be related to a learning curve.
The majority of reports of embolisation of pseudoaneurysms have used human thrombin alone. Loose and Haslam [20] used fibrin adhesive to thrombose 13 pseudoaneurysms with good results. The product contains bovine aprotinin, raising the question of transmission of prion disease, although no cases have been reported despite the products use in over 1 million instances. This particular product is no longer on the market, although other proprietary brands are available. There is a theoretical argument that the addition of the fibrinogen substrate, as well as cofactors such as calcium and factor XIII, may be advantageous in ensuring development of a fibrin mesh within the pseudoaneurysm. There is some evidence that fibrin adhesive is superior to thrombin alone in facilitating puncture closure following dialysis graft intervention [21]. Currently, however, there is no evidence showing superiority in the treatment of pseudoaneurysms of fibrin adhesive over thrombin alone obtained from these proprietary preparations or that obtained from pooled blood products. The ideal solution for obtaining thrombin may be autologous extraction from venesected blood, which overcomes any fears of transmission of infection and this is now being pursued at our institution [9].
It is interesting to note that 21 of the 27 (78%) groin punctures were inferior to the common femoral artery (Table 2
). Low puncture has been associated with a higher incidence of complications [22]. The femoral sheath surrounding the common femoral vessels may help tamponade any haematoma from the common femoral artery, but is not present below the common femoral artery bifurcation. The superficial femoral artery and profunda femoris arteries are relatively unsupported; the femoral head does not act as a posterior support against which manual compression can be applied. Both these vessels are smaller than the common femoral artery, so that the ratioof puncture hole size to vessel size is increased. Correct puncture technique may therefore reduce the number of pseudoaneurysms requiring treatment.
The incidence of complications with this technique is acceptable compared with UGCR. The fresh clots seen in the distal vessels in two cases were only identified because routine post-procedural angiography was performed, since neither patient had any symptoms. The balloon occlusion technique was used in both cases, which may have been the mechanism for the clot formation rather than the thrombin injection. There has been one report of distal thrombosis in a brachial artery following thrombin injection into a brachial pseudoaneurysm [23]. The reported overall complication rate is very low, and there appears to be no evidence to support the use of a protective balloon. By not using a protective balloon, the technique is transferred away from the angiography suite towards a wholly ultrasound guided procedure that may be performed quickly with small caliber needles, causing minimum discomfort to the patient.
There was clinical concern that the pseudoaneurysm had become infected as a complication of the procedures in the patient who developed local cellulitis following three injections of thrombin. Surgical repair of an infected pseudoaneurysm is more complex than simple repair, possibly requiring bypass with graft material kept away from the infected site. In this case the pseudoaneurysm was not infected and surgical repair was straightforward, but particular care should be given to aseptic technique when performing thrombin injection.
Although this appears to be a safe and effective technique, patient selection is important. Patients with very large or expanding aneurysms, mycotic aneurysms and aneurysms with a wide neck are better served by surgical treatment. Similarly, patients with a large haematoma in association with a false aneurysm, in whom the skin is tense and ischaemic or necrosed, require surgical repair of the pseudoaneurysm and evacuation of the haematoma, ideally with input from a plastic surgeon [24].
Received for publication November 8, 2000. Revision received April 9, 2001. Accepted for publication April 19, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R Vaidhyanath and K S Blanshard Insertion of a covered stent for treatment of a popliteal artery pseudoaneurysm following total knee arthroplasty Br. J. Radiol., March 1, 2003; 76(903): 195 - 198. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
