Original Article
Doppler ultrasound-guided percutaneous thrombin injection for treating femoral pseudoaneurysms Injeção percutânea de trombina guiada por ultra-som com Doppler colorido para o tratamento de pseudo-aneurismas femorais Frederico Celestino Miranda1, Rodrigo Gobbo Garcia2, Carlos Leite Macedo Filho3, Guilherme Falleiros Mendes4, Juliano Ribeiro de Andrade5, Alexandre Maurano6, Miguel Jose Francisco Neto7, Marcelo Buarque de Gusmão Funari8
ABSTRACT Objective: To evaluate the success rate of percutaneous injection of thrombin, guided by Doppler ultrasound to treat femoral pseudoaneurysms. Methods: Twenty-three patients with femoral pseudoaneurysms were treated with ultrasound-guided thrombin injection, between September 2003 and October 2007. Pseudoaneurysm size, dose of thrombin used, result of therapy and complications were prospectively documented. Other aspects analyzed included the type of procedure that caused the pseudoaneurysm (diagnostic catheterization, angioplasty, stent placement), size of endovascular introducer, use of hemostatic device and body mass index (BMI) of patients. Results: A total of 27 injections of thrombin were performed. The mean transverse diameter was 3.5 cm. The mean dose of thrombin injected was 666.7 IU. The primary success rate ranged from 19 to 23 (83%). Reperfusion occurred in one pseudoaneurysm. The rate of secondary thrombosis was four in four (100%). No thromboembolic, infectious or allergic complications occurred. Conclusions: Ultrasound-guided percutaneous thrombin injection is the best method for treating femoral pseudoaneurysms caused by endovascular procedures. It presents high rates of success, low recurrence rates and almost no complications. Keywords: Aneurysm, false/diagnosis; Aneurysm, false /therapy; Ultrasonography, Doppler ; Thrombin; Treatment outcome
RESUMO Objetivos: Avaliar a taxa de sucesso de injeção percutânea de trombina, guiada por ultra-som com Doppler para o tratamento de
pseudo-aneurismas femorais. Métodos: Vinte e três pacientes com pseudo-aneurismas femorais foram tratados com injeção percutânea de trombina guiada por ultra-som com Doppler, entre Setembro de 2003 e Outubro de 2007. O tamanho do pseudo-aneurisma, a dose de trombina utilizada, o resultado da terapia e as complicações foram documentados prospectivamente. Outros aspectos analisados foram o tipo de procedimento que causou o pseudo-aneurisma (cateterismo diagnóstico, angioplastia, colocação do stent), o tamanho do introdutor endovascular, a utilização dos dispositivos hemostáticos e o índice de massa corpórea (IMC) do paciente. Resultados: Um total de 27 injeções de trombina foram realizadas. O diâmetro transverso médio foi de 3,5 cm. A dose média de trombina injetada foi de 666,7 UI. A taxa primária de sucesso foi de 19 em 23 pacientes (83%). A reperfusão ocorreu em um pseudo-aneurisma. A taxa de sucesso secundária foi de quatro em quatro pacientes tratados (100%). Não ocorreram complicações tromboembólicas, infecciosas ou alérgicas. Conclusões: A injeção percutânea de trombina guiada por ultra-som é o melhor método para o tratamento dos pseudo-aneurismas femorais causados por procedimentos endovasculares. Apresenta elevadas taxas de sucesso, baixas taxas de reincidência e de complicações. Descritores: Falso aneurisma/diagnóstico; Falso aneurisma/terapia; Ultra-sonografia Doppler; Trombina; Resultado de tratamento
INTRODUCTION Femoral pseudoaneurysms have a low incidence, occurring in about 2% of patients undergoing femoral catheterization. Despite their low frequency, a marked
Study carried out at the Imaging Department of Hospital Israelita Albert Einstein – HIAE, São Paulo (SP), Brazil. Graduate student at Hospital Israelita Albert Einstein – HIAE, São Paulo (SP), Brazil.
1
Radiologist of the Ultrasonography Service at the Imaging Department of Hospital Israelita Albert Einstein – HIAE, São Paulo (SP), Brazil.
2
Radiologist of the Ultrasonography Service at the Imaging Department of Hospital Israelita Albert Einstein – HIAE, São Paulo (SP), Brazil.
3
Radiologist of the Ultrasonography Service at the Imaging Department of Hospital Israelita Albert Einstein – HIAE, São Paulo (SP), Brazil.
4
Radiologist of the Ultrasonography Service at the Imaging Department of Hospital Israelita Albert Einstein – HIAE, São Paulo (SP), Brazil.
5
Radiologist of the Ultrasonography Service at the Imaging Department of Hospital Israelita Albert Einstein – HIAE, São Paulo (SP), Brazil.
6
PhD; Assistant physician at the Ultrasonography Division of Instituto de Radiologia of Faculdade de Medicina da Universidade de São Paulo – INRAD-FMUSP, São Paulo (SP), Brazil.
7
PhD; Assistant physician at the Computed Tomography Division of Instituto de Radiologia of Faculdade de Medicina da Universidade de São Paulo – INRAD-FMUSP, São Paulo (SP), Brazil.
8
Corresponding author: Rodrigo Gobbo Garcia – Avenida Albert Einstein, 627/701 – Morumbi – CEP 05651-9010 – São Paulo (SP), Brasil – e-mail:
[email protected] Received on: Jan 29, 2008 – Accepted on: Oct 7, 2008
einstein. 2008; 6(4):428-33
Doppler ultrasound-guided percutaneous thrombin injection for treating femoral pseudoaneurysms
429
Objective The purpose of this study was to evaluate the success rate of percutaneous thrombin injection guided by color Doppler ultrasound (CDU) to treat femoral artery pseudoaneurysms caused by catheterization of this artery.
calcium chloride. For percutaneous injection, it was used human thrombin solution dissolved in saline solution. The procedure was carried out by interventional radiologists (AM, RGG, GFM, CLMF and JRA who had ten, seven, five, four and four years of experience in interventional radiology, respectively). The technique consisted of a ultrasound-guided percutaneous injection (HDI 5.000, Phillips, Washington, USA), using broadband transducers linear 5 to 12 MHz and convex 2 to 5 MHz by spinal anesthesia needles 22 or 25 gauge (Becton Dickinson, Franklin Lakes, NJ, USA) which were inserted inside the pseudoaneurysm with a longitudinal orientation to transducer (Figure 1), according to a technique described by Kang et al.(3).
METHODS A total of 23 patients diagnosed with pseudoaneurysm of the femoral artery were prospectively assessed. They were treated with percutaneous injection of thrombin at the Sector of the Non-Vascular Interventional Radiology at the Imaging Department of Hospital Israelita Albert Einstein (HIAE), from September 2003 to October 2007. The mean age was 73 years ± 10.1 (standard deviation), 13 men aged 53 to 90 years, mean age: 72 years; 10 women aged 52 to 81 years, mean age: 71 years (Chart 1). All patients had undergone previous endovascular procedures with catheterization of the affected femoral artery. The size and width of the pseudoaneurysm neck and the amount of thrombin needed for complete occlusion of the lesion were analyzed. It was also evaluated the type of procedure which caused the pseudoaneurysm (diagnostic catheterization, angioplasty, angioplasty with stent placement), gauge of the endovascular introducer, use or not of hemostatic device after endovascular procedure and body mass index (BMI) of patients.
Figure 1. Illustration showing the appropriate positioning of the needle inside the pseudoaneurysm regarding the transducer
rise in cases has been observed given the increased number of endovascular procedures all over the world(1). Conventional treatments for this lesion include the surgical correction or ultra-sound guided compression. In the last decade, management of pseudoaneurysms changed completely with introduction of percutaneous thrombin embolization(2-3).
Chart 1. Demographic data Age (n = 23), mean ± sd [max.-min.] 73 ± 10.1 [90-52] BMI (n = 20), mean ± sd [max.-min.] 27.8 ± 4.5 [34.7-17.6] Sex (n = 23) Male Female
n (%) 13 (56.5) 10 (43.5)
BMI = body mass index; sd = standard deviation
In the procedure of percutaneous injection to treat pseudoaneurysms it was used purified human thrombin obtained from a commercial kit of sealant fibrin (Beriplast P®, ZLB Behring AG, Marburg, Germany). Each unit contains two sets: a bottle of lyophilized product containing fibrinogen and coagulation factor XIII, and a bottle of aprotinin solution; a bottle of lyophilized thrombin (400 to 600 U/ml) and a bottle of
Under local anesthesia, the injection was carried out slowly and in small doses of 100 to 300 IU (0.1 to 0.3 ml/ sec), and they were visualized as a turbulence inside the pseudoaneurysm. After injection, the flow in the lumen was monitored by Doppler ultrasound to detect the formation of clots. With confirmation of occlusion of the pseudoaneurysm, the needle was removed and the adjacent vessels were checked. All patients were instructed to remain bedridden until undergoing the control ultrasound examination, 12 to 24 hours after the procedure. The dose of thrombin, therapy result and complications were prospectively recorded. For analysis, the data was stored in an electronic database built in statistical package (SPSS® Statistical Package for Social Sciences, version 15.0, for Windows®). The interval or ratio variables were expressed by the
einstein. 2008; 6(4):428-33
430
Miranda FC, Garcia RG, Macedo Filho CL, Mendes GF, Andrade JR, Maurano A, Francisco Neto MJ, Funari MBG
mean ± standard deviation, and the Mann-Whitney non-parametric test was used for comparison. Nominal variables were expressed by proportions and the χ² test or the Fisher’s exact test were used for comparison, according to indication. The association between quantitative variables was established by the Pearson’s (non-parametric) correlation.
RESULTS A total of 27 thrombin injections were carried out in 23 patients studied. The pseudoaneurysm was in the common femoral artery in all cases. The mean transverse diameter of pseudoaneurysms was 3.5 cm. The neck mean measurement was 4.2 mm. The mean dose of thrombin injected was 666.7 IU ± 384.8 (1.200 to 100 IU). The mean BMI was 27.3 (normal range: 18.5 to 25; patients with a BMI above 30 are considered obese). Most patients underwent catheterization and angioplasty. Eleven patients received drug-eluting stents. Seven patients used some type of hemostatic device at the end of the endovascular procedure (Table 1). The size of the endovascular introducer used in endovascular procedures resulting in pseudoaneurysms varied between 6 to 8 F (Chart 2). Table 1. Data related to the procedure Data Procedure (n = 22) Catheterization Catheterization + Angioplasty Introducer (n = 21) 6F > 6F Coated stent (n = 21) Drug-eluting stent Bare metal stent/NO Hemostatic device (n = 21) Yes No
n (%) 9 (40.9) 13 (59.1) 15 (71.4) 6 (28.6) 11 (52.4) 10 (47.6) 5 (23.8) 16 (76.2)
Chart 2. Data related to pseudoaneurysm and intervention Larger measurement (mm) (n = 22), mean ± sd [max.-min] Neck measurement (mm) (n = 16), mean ± sd [max.-min] Thrombin dose (IU) (n = 15), mean ± sd [max.-min] Occlusion rate – 1st attempt (n = 23) Complete Partial Complication (n = 23) Yes No
einstein. 2008; 6(4):428-33
35.0 ± 18.7 [90.0-12.0] 4.2 ± 1.5 [7-3] 666.7 ± 384.8 [1.200-100] n (%) 19 (82.6) 4 (17.4) 0 (0) 23 (100)
The primary therapeutic success (complete thrombosis after a single injection) was reached in 19 of 23 (83%) pseudoaneurysms. Ten minutes after the initial injection of thrombin, residual blood flow (indicating partial thrombosis of the lobe) was present in three pseudoaneurysms. Reperfusion of a previously thrombosed lobe occurred in one pseudoaneurysm from 6 to 24 hours after treatment. In these four cases a second injection was necessary. The secondary success rate was four out of four (100%). No thromboembolic, infectious or allergic complications occurred (Chart 3). Chart 3. Main results 27 thrombin injections were performed Mean transverse diameter of pseudoaneurysms was 3.5 cm Mean dose of thrombin injected was 666.7 IU Primary success rate was 19 out of 23 (83%) Reperfusion in one pseudoaneurysm Secondary success rate was four out of four (100%) No thromboembolic, infectious or allergic complication occurred
There was a trend to statistical significance in the comparison between the diameter of the introducer and the size of pseudoaneurysm (p = 0.055; MannWhitney). There was no statistical significance between the size of the neck and the diameter of introducer; between the hemostatic device and the size of aneurysm; between the size of introducer and occlusion in the first attempt; and between the size of pseudoaneurysm and occlusion in the first attempt.
DISCUSSION A pseudoaneurysm is a self-contained rupture of the arterial wall, with total or partial laceration of its layers(4). The pulsating blood leaks to the perivascular space and it is contained by the middle layer, adventitial layer or by adjacent tissues, thus forming the pseudoaneurysm wall(5). Pseudoaneurysms can occur in four circumstances(4): after catheterization; at the site of anastomosis of a synthetic prosthesis with the native artery (aorto-femoral prosthesis); in trauma; and after infections (mycotic pseudoaneurysm). Post-catheterization pseudoaneurysms are one of the most common vascular complications of cardiac and peripheral angiographic procedures(6). The incidence of femoral pseudoaneurysms after diagnostic catheterization ranges from approximately 0.05 to 2%(7). When the coronary or peripheral intervention is carried out, the incidence increases to 6%(7). Since 1996, the number of peripheral interventions doubled, thus increasing the number of pseudoaneurysms(8).
Doppler ultrasound-guided percutaneous thrombin injection for treating femoral pseudoaneurysms
The risk factors comprise large vascular introducers (larger than 6 F), multiple attempts for arterial puncture during the endovascular access, inappropriate compression after removal of the introducer and inadequate hemostasis or anticoagulation, including that induced by heparin or warfarin therapy(6,8-9). Several patients and procedural factors may contribute to the formation of pseudoaneurysms. Among those, some stand out, such as increased incidence of femoral artery pseudoaneurysms when the site of puncture is not the common femoral artery, but the deep or superficial femoral artery or the external iliac artery(1). It is believed that accuracy of the first puncture, physician’s experience and duration of compression of the site of puncture after removal of introducer may be important factors in the development of pseudoaneurysms. In this series it was noticed a trend to statistical significance between the gauge of introducers and the size of pseudoaneurysms, with larger lesions being caused by larger introducers. Pain or swelling in the inguinal region after catheterization is the most common presentation of a pseudoaneurysm. Additionally, a large pseudoaneurysm may also lead to compression of the nerves and vessels with associated neuropathy, venous thrombosis, claudication or rarely critical ischemia of the limb(4). In the physical examination it is possible to observe a pulsating palpable mass or a systolic murmur. The preferred diagnostic procedure is duplex ultrasound with a linear transducer of 5 to 7 Mhz. Color Doppler improves the ultrasound diagnostic accuracy through identification of a pulsating flow in the sac(10). Duplex ultrasound provides 94% of sensitivity and 97% of specificity for identification of pseudoaneurysms(4,7). The typical appearance of a pseudoaneurysm in the mode B ultrasound (image mode without Doppler) is a pulsating hypoechogenic sacular image (Figure 2). In the color Doppler, it is possible to observe a whirled flow in the lumen (Figure 3). A neck connects the pseudoaneurysm to the vessel, usually the common femoral artery (Figure 4). When the pulsed Doppler is positioned in the neck, a “to-and-fro” pattern is obtained, translating the multiple directions of the intralesional flow (Figure 5). In the Doppler ultrasound study it must be assessed the pseudoaneurysm geometry and position (Figure 6) in relation to the femoral neurovascular bundle, with identification of the number, size and depth of the lobes. The volume of pseudoaneurysm (length x height x width x 0.523) and its neck dimensions must be measured. Venous structures must also be evaluated to rule out the presence of deep venous thrombosis secondary to prolonged compression in the inguinal region or pressure from an expanding hematoma. The examination must include assessment of the middle and
431
Figure 2. Pulsating hypoechogenic sacular image
Figure 3. Whirled flow in the lumen
Figure 4. Neck
distal segments of the external iliac artery, common femoral artery and the proximal portion of the superficial and deep femoral artery. Small (< 2.0 cm) asymptomatic pseudoaneurysms can be followed up with no need of specific treatment. However, the presence of pain must be seen as einstein. 2008; 6(4):428-33
432
Miranda FC, Garcia RG, Macedo Filho CL, Mendes GF, Andrade JR, Maurano A, Francisco Neto MJ, Funari MBG
Figure 5. “To-and-fro” pattern
Figure 6. Pseudoaneurysm in the right femoral artery
an indication to treatment. The most catastrophic complication of pseudoaneurysms is rupture(4). The risk of spontaneous rupture of pseudoaneurysms is related to their size (lesions > 3 cm present a higher risk of rupture), the presence of neurological symptoms or severe pain, active infection at the puncture site and expanding hematoma or continuous growth of pseudoaneurysms. Although most postcatheterization pseudoaneurysms are sterile, infection of a pseudoaneurysm significantly increases the risk of rupture and of septic embolism. einstein. 2008; 6(4):428-33
Until the beginning of the 1990’s, the only treatment available for pseudoaneurysms was surgery. Since then, several other types of treatment have emerged, such as the ultrasound-guided compression and endovascular correction with a coated stent. Surgical management is still very important in pseudoaneurysms originating from vascular anastomosis or mycotic pseudoaneurysms. However, surgery is rarely needed to treat the postcatheterization pseudoaneurysms. Surgical intervention in inguinal pseudoaneurysms cause significant risks of infection in a population of patients frequently affected by several cardiovascular comorbidities(4). In 1991, Fellmeth et al. described a safe and non-invasive method to treat pseudoaneurysms, the ultrasound-guided compression(11). This method has a successful rate of 75 to 98%. By using the ultrasound to assess and identify the pseudoaneurysm lobe and neck, the ultrasonographic transducer is positioned and pressure is applied to compress the lobe and neck while the flow in the native artery is allowed. Through direct ultrasonographic assessment it is possible to confirm the absence of flow in the pseudoaneurysm. Compression is generally used in ten-minute cycles. Obesity, size, depth, number of lobes and concurrent anticoagulation may limit the success rate of therapy. In patients on anticoagulants, the success rate is between 30 to 73%(7). The main limiting factor is the time necessary to induce and maintain thrombosis. This is a very painful procedure for the patient and analgesia is sometimes required. It is also a very extenuating examination for the physician performing it because it is difficult to maintain the compression at the correct spot for a prolonged period of time(12). More rare complications include vasovagal reactions, rupture of pseudoaneurysms, skin necrosis and deep venous thrombosis(4,13). Treatment of pseudoaneurysms by ultrasoundguided compression has been successfully replaced by percutaneous obliteration with thrombin, becoming the treatment of choice for post-catheterization pseudoaneurysms(4,14). Percutaneous coagulation was reported for the first time in 1986, by Cope and Zeit(15), and it was later adjusted to be guided by ultrasound. This method is based on the fact that thrombin is important in the conversion of fibrinogen into fibrin(16). Therefore, a fibrin clot is instantly formed, even in the presence of antiplatelet or anticoagulation therapy. Success rates vary between 91 and 100%(7,14). Even in patients treated with antiplatelet and antithrombolytic therapies, the success rate is higher than in those treated only with compression(12,14). The data obtained here do not point to any statistically significant association between the efficacy of pseudoaneurysm thrombosis and the gauge of introducers or size of their necks, taking into account
Doppler ultrasound-guided percutaneous thrombin injection for treating femoral pseudoaneurysms
the success of primary thrombosis (i.e, obtained after a single thrombin injection). Larger aneurysms and with larger necks were thrombosed with the same efficacy as smaller lesions, which reinforces the great applicability of the method for correcting the lesions originated from therapeutic percutaneous angiographic procedures requiring larger endovascular instruments. The complication rates described in the literature vary between 0.5 and 1.3%(7). The most severe complications are arterial thromboembolic events, development of deep venous thrombosis (if thrombin is inadvertently injected in the vein) and pulmonary embolism(15,17). Most of these complications are due to technical errors with involuntary injections in the vascular lumen. Thrombin injections in the pseudoaneurysm neck lead to higher risk of thromboembolic complications(15). In the present series, it was not observed any type of complication. Thrombin treatment must be performed only in patients who developed pseudoaneurysms secondary to catheterization. The presence of spontaneous pseudoaneurysms must lead to the suspect of mycotic etiology, of which treatment of choice is surgical correction. The size of pseudoaneurysm is not a contraindication per se. However, if a pseudoaneurym is large enough to cause skin necrosis or nerve or blood vessel compression, then surgery must be the method of choice. After interruption of the flow in the pseudoaneurysm, distal pulses and limb perfusion must be carefully monitored for 12 hours. It is recommended to avoid extenuating activities for 24 hours. Absolute bed rest is recommended for one to three hours, as well as ultrasonographic control in 24 hours. A follow-up CDU is carried out 24 to 72 hours after the procedure. Recurrent pseudoaneurysms can be treated with subsequent injections not compromising their efficacy.
CONCLUSIONS Doppler ultrasound-guided percutaneous thrombin injection seems to be the best method for the treatment of femoral artery pseudoaneurysms. It presents high success rates, low recurrence and low complication rates. By using the complete thrombosis of the lesion as an indicator, the series result had a 100% success rate. Only a minority of patients underwent more than one embolization session. Continuous monitoring with color Doppler allowed the use of small amounts of thrombin
433
and consequently the rapid and effective occlusion of lesions without any type of complication.
REFERENCES 1. Gabriel M, Pawlaczyk K, Waliszewski K, Krasiński Z, Majewski W. Location of femoral artery puncture site and the risk of postcatheterization pseudoaneurysm formation. Int J Cardiol. 2007;120(2):167-71. 2. Ghersin E, Karram T, Gaitini D, Ofer A, Nitecki S, Schwarz H, et al. Percutaneous ultrasonographically guided thrombin injection of iatrogenic pseudoaneurysms in unusual sites. J Ultrasound Med. 2003;22(8):809-16. 3. Kang SS, Labropoulos N, Mansour MA, Baker WH. Percutaneous ultrasound guided thrombin injection: a new method for treating postcatheterization femoral pseudoaneurysms. J Vasc Surg. 1998;27(6):1032-8. 4. Saad NE, Saad WE, Davies MG, Waldman DL, Fultz PJ, Rubens DJ. Pseudoaneurysms and the role of minimally invasive techniques in their management. Radiographics. 2005;25 Suppl 1:S173-89. 5. Padidar AM, Kee ST, Razavi MK. Treatment of femoral artery pseudoaneurysms using ultrasound-guided thrombin injection. Tech Vasc Interv Radiol. 2003;6(2):96-102. 6. Knight CG, Healy DA, Thomas RL. Femoral artery pseudoaneurysms: risk factors, prevalence, and treatment options. Ann Vasc Surg. 2003;17(5):503-8. 7. Webber GW, Jang J, Gustavson S, Olin JW. Contemporary management of postcatheterization pseudoaneurysms. Circulation. 2007;115(20):2666-74. 8. Morgan R, Belli AM. Current treatment methods for postcatheterization pseudoaneurysms. J Vasc Interv Radiol. 2003;14(6):697-710. 9. Maleux G, Hendrickx S, Vaninbroukx J, Lacroix H, Thijs M, Desmet W, et al. Percutaneous injection of human thrombin to treat iatrogenic femoral pseudoaneurysms: short and midterm ultrasound follow-up. Eur Radiol. 2003;13(1):209-12. 10. Demirbas O, Guven A, Batyraliev T. Management of 28 consecutive iatrogenic femoral pseudoaneurysms with ultrasound-guided compression. Heart Vessels. 2005;20(3):91-4. 11. Fellmeth BD, Roberts AC, Bookstein JJ, Freischlag JA, Forsythe JR, Buckner NK, et al. Postangiographic femoral artery injuries: non surgical repair with US-guided compression. Radiology. 1991;178(3): 671-5. 12. Krueger K, Zaehringer M, Strohe D, Stuetzer H, Boecker J, Lackner K. Postcatheterization pseudoaneurysm: results of US-guided percutaneous thrombin injection in 240 patients. Radiology. 2005;236(3):1104-10. 13. Lenartova M, Tak T. Iatrogenic pseudoaneurysm of femoral artery: case report and literature review. Clin Med Res. 2003;1(3):243-7. 14. Dasyam AK, Middleton WD, Teefey SA. Development of nonobstructive intraarterial thrombi after injection of thrombin into pseudoaneurysms. AJR Am J Roentgenol. 2006;186(2):401-5. 15. Bhat R, Chakraverty S. Femoral artery thrombosis following percutaneous treatment with thrombin injection of a femoral artery pseudoaneurysm: a case report. Cardiovasc Intervent Radiol. 2007;30(4):789-92. 16. Vázquez V, Reus M, Piñero A, Abellán D, Canteras M, Espinosa de Rueda M, et al. Human thrombin for treatment of pseudoaneurysms: comparison of bovine and human thrombin sonogram-guided injection. AJR Am J Roentgenol. 2005;184(5):1665-71. 17. Sheiman RG, Mastromatteo M. Iatrogenic femoral pseudoaneurysms that are unresponsive to percutaneous thrombin injection: potential causes. AJR Am J Roentgenol. 2003;181(5):1301-4.
einstein. 2008; 6(4):428-33
