| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 Department of Roentgenology, Hospital Universitario Virgen de la Arrixaca,
Ctra. Madrid-Cartagena, El Palmar 30120, Murcia, Spain.
2 Department of Surgery, Hospital Universitario Virgen de la Arrixaca, Ctra.
Madrid-Murcia, El Palmar, Murcia, Spain.
3 Department of General Practice, Hospital Morales Meseguer, Murcia,
Spain.
4 Department of Statistics, University of Murcia, Spain.
Received June 5, 2004;
accepted after revision August 10, 2004.
Address correspondence to V. Vázquez
(vvs63{at}hotmail.com).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. In our experience, the use of human thrombin for the treatment of iatrogenic femoral pseudoaneurysms is highly efficient (100%), the administered dose is significantly less than with bovine thrombin, and the risk for allergy is potentially lower. At our hospital, human thrombin has replaced bovine thrombin and is the first line of treatment for an iatrogenic pseudoaneurysm.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The traditional nonoperative method for treating iatrogenic femoral pseudoaneurysms is sonogram-guided compression [3], which has recently begun to give way to percutaneous thrombin injection. Sonogram-guided compression has an efficiency of over 90% in patients who did not receive anticoagulation therapy at the time of compression. However, the success rate drops to 62–73% if anticoagulation therapy has been administered or if prolonged manipulation with large-diameter catheters has occurred [3–5]. Although the technique is safe and effective, it has considerable limitations. The process is painful and the duration of compression is lengthy, and sedation is occasionally needed, which may be contraindicated in many patients [6].
Percutaneous sonogram-guided human thrombin injection into the lumen of the pseudoaneurysm is a valuable therapeutic alternative. The technique has proven quick, safe, and comfortable for the patient and efficient in 96–100% of cases, even in patients who have received anticoagulation and antiplatelet therapy [7–21]. The most serious complication is distal arterial embolization, with a frequency of about 2% [14].
Bovine thrombin is a foreign substance and as such, may induce allergic reactions after repeated exposure [22]. In addition, patients receiving thrombin injections may develop antibodies to bovine protein, including coagulation factor V, which may cross-react with human coagulation factor V and lead to alterations in hemostasis [23, 24]. It is because of these risks that we report a series of cases using human thrombin [25].
The purpose of this article is to review our experience to determine the success rate of percutaneous sonogram-guided human thrombin injection for the treatment of iatrogenic femoral pseudoaneurysms and compare this with the results we obtained with bovine thrombin.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Exclusion criteria were local infection of the area, previous treatment with bovine thrombin to prevent possible allergic reactions, rupture of the pseudoaneurysm (these cases required emergency repair surgery), and pseudoaneurysms associated with an arteriovenous fistula. No patients were excluded.
After September 2002, pseudoaneurysms were treated with human thrombin
because it was readily available, less expensive (
180 per kit compared
with 254 for bovine thrombin), and theoretically safer than bovine
thrombin. We began using it for pseudoaneurysm repair to confirm its safety
and efficiency when compared with bovine thrombin. We treated 40
pseudoaneurysms, and we included two additional cases treated previously but
unsuccessfully with bovine thrombin. Exclusion criteria for human thrombin
were the same as for bovine thrombin. A total of 42 patients were treated with
human thrombin. One patient with a pseudoaneurysm-related arteriovenous
fistula was excluded.
Diagnosis was established in the radiology department by the same sonographers: 40 patients were diagnosed within the first 24 hr and two patients on the second day. Human thrombin injection was administered within 12–24 hr of diagnosis.
Patients were informed of the technique, its potential complications, and the different therapeutic options, and all gave their written informed consent. The use of bovine and human thrombin in our hospital for sealing pseudoaneurysms was approved by the Ethical and Research Committee.
Method
Before thrombin injection, sonography was performed at the puncture site
using a Powervision 6000 unit (Toshiba) equipped with a 6- to 11-MHz
transducer. We also used a 3.5-MHz transducer in patients presenting with an
extensive inguinal hematoma, obese patients, and cases of lobulated or
large-sized pseudoaneurysms to better visualize the site and position of the
pseudoaneurysm in relation to the artery. Sonogram diagnosis of the
pseudoaneurysm required finding a more or less rounded mass with a two-way
color flow and separate from the artery wall together with a color-flow signal
in a thin track leading from the mass to the artery (corresponding to the
pseudoaneurysm neck). The characteristics of the pseudoaneurysm and large
vessels were examined, and the surrounding soft tissues were inspected to rule
out the existence of arteriovenous fistulas. The presence of color flow was
confirmed in the artery and femoral vein as well as the presence of dorsalis
pedis and posterior tibialis pulses.
The patient's skin at the puncture site was prepared with povidone iodine, and sterile drapes were placed in the inguinal area. The sonography transducer was covered with a sterile laminated bag, and sterile gel was applied to the patient's skin to prevent infections in the clot that forms after thrombin is injected. Local anesthesia was not used on the area.
The bovine thrombin was obtained from a commercial package sold as a femoral closure device (1,000 U/mol, Duet, Vascular Solutions) and was diluted in physiologic serum with calcium chloride (1 mL = 1,000 U). The human thrombin is part of a commercial package sold as a fibrin sealant kit (500 U/mL, Tissucol Duo 1.0 mL, Baxter). Each packet contains a solution of human thrombin in calcium chloride and a fibrinogen concentrate with factor XIII and other proteins. For percutaneous injection, we used the human thrombin solution in calcium chloride, the same method as with the bovine thrombin.
The technique consisted of a percutaneous thrombin injection using a 3.5-inch, 22-gauge spinal needle (B/Braun) sonographically guided by a 6- to 11-MHz multifrequency linear-array transducer inside the pseudoaneurysm sac according to the technique described by Kang et al. [7].
The free-hand technique was used for injection [26] (Fig. 1). With this method, one hand held the sonography probe while the free hand advanced the needle toward the pseudoaneurysm in a direction perpendicular to the ultrasound beam to the transducer to identify the path of the needle. This technique allowed flexibility and changes in needle direction during the puncture.
|
The loaded needle and syringe were inserted and emptied into the pseudoaneurysm cavity under continuous sonogram guidance, and B-mode showed the end of the needle to be positioned at the periphery of the pseudoaneurysm as distally as possible to the neck (Fig. 2A, 2B, 2C, 2D). A bolus injection of approximately 0.1–0.3 mL/sec was performed (each bolus was considered an injection), and the entrance of thrombin was visualized by real-time color Doppler sonography as a turbulence of bursts of color from the needle tip. After the first injection, the flow in the lumen was monitored by Doppler sonography to detect the formation of thrombosis. When incomplete thrombosis was achieved after the first injection, two or more injections were performed with or without needle repositioning. A few patients required a second skin puncture in the same session to reposition the needle. Once the pseudoaneurysm was thrombosed, the needle was withdrawn and the presence of dorsalis pedis and posterior tibialis pulses was rechecked.
|
|
|
|
In cases of a lobulated pseudoaneurysm, we attempted first to thrombose the most distal lobe. If flow persisted, we punctured the lobe in direct communication with the neck.
Thrombosis time was noted as the time between the start of thrombin injection into the pseudoaneurysm cavity and total thrombosis of the lumen.
The method generally consists of a single puncture with a fine needle. Sedation is not usually required. After the injection of thrombin, the patients remained in bed for at least 6 hr, with pulse checks in the limb every 4 hr over the 24 hr postinjection period.
Sonography was performed at 24 hr to determine the success of the technique. If flow persisted in the pseudoaneurysm, the same technique was reattempted. Follow-ups were conducted at 1 week.
Statistical Analysis
A descriptive statistical study was done of each variable by obtaining the
distribution of absolute and relative frequencies. Characteristic parameters
such as mean, median, range, and SD were calculated for the quantitative
variables.
The relationship between quantitative variables was determined using a regression/linear correlation analysis by calculating and contrasting Pearson's coefficient of correlation.
The association or dependence between qualitative variables was determined using an analysis of contingency tables with Pearson's chi-square test.
Comparison of groups was done using the Student's t test, as there were only two groups, and an analysis of variance.
Reference to statistical significance is made by presenting a p value for a risk of 5%; thus, any measurement with a value of p < 0.05 was considered statistically significant.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All pseudoaneurysms (100%; 42/42) treated with human thrombin thrombosed at the first attempt. Although the success rate of pseudoaneurysm repair with human thrombin was greater than with bovine thrombin, it was not a statistically significant difference (p = 0.13) (Table 1).
|
Table 2 provides a comparison of the demographics, clinical variables, and pseudoaneurysm characteristics of patients treated with the bovine and human thrombin techniques. The two groups were similar in sex distribution, age, and side of the body on which the pseudoaneurysm was located. The pseudoaneurysms were similar in number of lobes and neck length. The median volume of injected bovine thrombin was 500 U (range, 100–7,000 U). Thirty-six patients (62.6%) required a dose of 500 U or less. A significant relationship (r2 = 0.51; p < 0.001) was present between maximum diameter of the pseudoaneurysms and amount of bovine thrombin used. The median volume of injected human thrombin was less than when we used bovine thrombin (range, 50–2,500 U). Twenty-two patients (52.4%) required a dose of only 50 U, and the majority (36 patients), less than 200 U. The amount of human thrombin used was significantly less than for bovine thrombin (p = 0.009). No significant correlation was found between the maximum diameter of the pseudoaneurysms and amount of thrombin in the patients treated with human thrombin. In the pseudoaneurysms requiring a single thrombin injection, the mean thrombosis time was 2.2 sec with human thrombin and 9.7 sec with bovine thrombin.
|
In the group treated with bovine thrombin, we found a statistically significant correlation between the maximum diameter of the pseudoaneurysm and thrombosis time (r2 = 0.54; p < 0.001) and the number of injections required for thrombosis (r2 = 0.47; p < 0.001). Lobulated pseudoaneurysms required more injections of bovine thrombin (r2 = 0.37; p < 0.005). These correlations were not encountered with human thrombin.
In the group treated with human thrombin, a statistically significant relationship was found between thrombin amount and pseudoaneurysm neck width (r2 = 0.32; p < 0.05).
The use of anticoagulants did not affect the efficiency of the technique. Seventeen of the patients treated with bovine thrombin injection were on anticoagulation therapy (11 heparin, six warfarin), and 15 (88%) of 17 were treated successfully. In the group treated with human thrombin, 13 (30.9%) were on anticoagulants at the time of injection (eight heparin, five warfarin), and all were thrombosed uneventfully.
There were three minor incidents. One was a cutaneous reaction at the bovine thrombin injection site, consisting of an erythema, which disappeared a few hours later. Two patients treated with human thrombin reported local pain after the injection, which remitted without the need for analgesics.
The Doppler study performed at 1 week showed no complications or persistence of residual flow in the pseudoaneurysms.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The use of human plasma proteins as an adhesive dates back to 1909, when Bergel [29] used dry plasma to arrest bleeding during surgery. In 1915, Grey [30] used fibrin patches as a hemostatic agent in neurosurgery. The use of plasma fibrinogen mixed with thrombin as an adhesive was first reported in 1944 [31] but it was not commercialized in Europe until 1970. Since then, it has been widely used for various indications ranging from hemostasis to the transport of medicines [32–34].
Bovine thrombin is foreign to the organism and as such can induce allergic reactions after repeated exposure. For this reason, patients who have had previous exposure are advised to undergo a skin test with a thrombin solution before injection [22]. The present study excludes patients with previous exposure to bovine thrombin to avoid possible allergic reactions.
In addition, patients receiving thrombin injections may develop antibodies to bovine protein, including coagulation factor V, which may cross-react with human coagulation factor V and lead to alterations in hemostasis [23, 24]. In a recent study, Krüger et al. [21] maintained that such changes in the coagulation parameters and the risk for allergic complications were due to the probable passage of thrombin to the blood flow.
To reduce the potential risks for allergy, it was decided to use human thrombin in a group of patients. Few studies have been published on the use of human thrombin for the treatment of pseudoaneurysms. In a series of 114 patients, Paulson et al. [14] used human thrombin in 13 patients but did not recommend its use, since, according to author, it was just as efficient as bovine thrombin but more expensive. Elford et al. [35] recently published a series a 14 patients treated successfully with human thrombin injection at a greater concentration (1,000 U/mL) than used in our study (500 U/mL). Neither report performed a comparative study with bovine thrombin.
In the present study, Tissucol Duo 1.0 mL was used as a fibrin adhesive. The exhaustive system of safety mechanisms in the manufacturing of the product [36, 37] includes a strict selection of donors using the repeated-donor technique and infectious-marker tests for each individual donation. In June 2001, German authorities made a pronouncement in favor of fibrin adhesives, assuming that the risk/benefit ratio was positive for such products.
We found no significant differences in efficiency between the two groups, although all pseudoaneurysms treated with human thrombin thrombosed successfully. Moreover, this occurred at the first attempt in all the cases (42/42), compared with just 51/56 in the bovine thrombin group. We also found no significant differences in thrombosis time between the two techniques, but there was a clear tendency for pseudoaneurysms to thrombose more quickly with human thrombin. This, together with the fact that the dose of human thrombin necessary for pseudoaneurysm thrombosis is significantly lower than for bovine thrombin, suggests that human thrombin is probably more potent than bovine thrombin.
In Sheiman and Brophy's series [19], all pseudoaneurysms were treated with 0.5–1 mL (500–1,000 U) of bovine thrombin, regardless of size, because if any flow persisted after injection, they waited for a spontaneous resolution in 10 min, without the need to inject more thrombin. They conclude in their series that the thrombin, trapped inside the thrombus, remains active and continues to be effective in thrombus formation. In our study, as in the study by Paulson et al. [14], we injected the thrombin in boluses of 0.1–0.3 mL/sec (each bolus is considered an injection) using continuous monitoring to check the presence of flow in the lumen and to reinject until total thrombosis is achieved, which means that the amount of injected thrombin varies from case to case.
In the group treated with bovine thrombin, a significant relationship was found between maximum diameter of the pseudoaneurysms and dose. This relationship was not found with human thrombin because the doses used were generally very small, regardless of the size of the pseudoaneurysm. Paradoxically, two cases receiving high doses of human thrombin (1,000 U) had small-sized pseudoaneurysms (25 mm and 35 mm), but with short, wide necks (the aneurysmatic sac was practically joined to the artery). We do not know why; however, it may have been that a short, wide neck allowed a greater percentage of injected thrombin to pass directly into the systemic circulation, leaving a smaller amount of effective local thrombin. La Perna et al. [13] also reported that pseudoaneurysms with short, wide necks had greater difficulty in closing. In such cases, Elford et al. [35] used a balloon occlusion device across the neck during thrombin injection to prevent the passage of thrombin to the blood flow and the possibility of a distal embolism. We did not use it and had no distal embolism complications. In a recent study, Sheiman and Mastromatteo [38] reported that the flow velocities of wider pseudoaneurysm necks were higher than with narrower ones. The same authors relate the width of the pseudoaneurysm neck to the larger doses required (more than 1,000 U). They also verify that it was an indirect indicator of an arterial wall hole after arteriotomy; this arterial wall deficit was at least 8 mm in size in all the cases with thrombin failure.
In the bovine thrombin group, we also found statistically significant relationships between maximum diameter of the pseudoaneurysm and number of injections and thrombosis time, which shows that larger pseudoaneurysms tend to require more injections, thus increasing thrombosis time. This relationship was not found for human thrombin.
When the pseudoaneurysms were lobulated in studies from Bloom [39] and Krüger et al. [21], thrombin was injected into the lobe nearest the artery. They assumed that with occlusion of the proximal lobe, the lobes located more distally would also occlude, and the risk of injection of the lobe nearest the native vessel is the same as with injection into any simple pseudoaneurysm. In the present study, as in others [14, 19], the most distal lobe was punctured first, and if it did not thrombose completely, the procedure was repeated on the lobe nearest the artery. Good results were obtained with this method. Unlike for other authors [9, 19, 21], the success of the technique was not influenced by pseudoaneurysm lobulation. The three failures in our series corresponded to simple pseudoaneurysms. The incidence of failure in the complex pseudoaneurysms was not higher. The group treated with bovine thrombin had a statistically significant relationship between the number of lobes and number of injections necessary for thrombosis.
One patient with a pseudoaneurysm-related arteriovenous fistula was excluded. Little mention has been made of such cases in the literature. Kang et al. [10] report five cases of pseudoaneurysms associated with arteriovenous fistulas. In two of them, thrombosis of the pseudoaneurysm failed, but no serious complications occurred. Kang et al. [40] reported the death of a patient after thrombin injection due to the thrombus passing into the vein and leading to a pulmonary thromboembolism. In our study, as in the study by Piñero et al. [41], we opted for conservative treatment. It was considered that the high flow associated with the arteriovenous fistula would have instigated successful thrombosis with thrombin.
Complications with this technique are uncommon. The most serious complication is distal arterial embolization, with a frequency of about 2% [14] in the series cited using bovine thrombin. No major complication occurred in our study. There were three minor incidents: one case with the use of bovine thrombin, presenting with a cutaneous erythema; and two with human thrombin reporting pain at the puncture site, which remitted a few hours later without treatment.
This study has two notable limitations: the absence of a control group and the fact that the study spans different time periods. However, this is not a comparative study of two techniques with random allocation of treatment groups, but simply a nonrandom observational study to evaluate the efficiency and safety of bovine and human thrombin injections in the treatment of iatrogenic pseudoaneurysms and to draw operative conclusions. Although the two groups were not managed at the same time, the differences encountered cannot be attributed to different techniques or different operators. The only minor influence may be the greater experience with the second group (human thrombin), as the study with bovine thrombin was performed previously.
Besides homologous fibrin adhesives, the treatment of pseudoaneurysms has occasionally included injections of autologous human thrombin [42]. This is obtained by treatment of the patient's own blood in a sterile medium. The aim of its use is to reduce the possibilities of anaphylaxis and, especially, prion contagion. Diffusion of this thrombin is limited by the complicated technique for obtaining it and the need for multidisciplinary teams to be available at all times.
In conclusion, although the efficiency of human thrombin has not proven considerably superior to bovine thrombin, we believe that the use of human thrombin is advisable in the treatment of iatrogenic femoral pseudoaneurysms because of the lower risk of allergy. Moreover, the use of significantly lower doses, together with a lower concentration, may reduce the risk for complications.
Acknowledgments
Our thanks to Manuel Canteras Jordana, professor of statistics at the
University of Murcia (Spain), for performing the statistical study.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Luedde, U. Krumsdorf, J. Zehelein, B. Ivandic, T. Dengler, H. A. Katus, and C. Tiefenbacher Treatment of Iatrogenic Femoral Pseudoaneurysm by Ultrasound-Guided Compression Therapy and Thrombin Injection Angiology, September 1, 2007; 58(4): 435 - 439. [Abstract] [PDF]
|
|
|
|
 |
|
 G. W. Webber, J. Jang, S. Gustavson, and J. W. Olin Contemporary Management of Postcatheterization Pseudoaneurysms Circulation, May 22, 2007; 115(20): 2666 - 2674. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
