- Email: [email protected]
Transarterial embolization for the management of hemarthrosis of the knee
European Journal of Radiology 81 (2012) 2737–2740
Contents lists available at SciVerse ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Transarterial embolization for the management of hemarthrosis of the knee P. Waldenberger a , A. Chemelli b , A. Hennerbichler c , M. Wick b , M.C. Freund b , W. Jaschke b , M. Thaler d , I.E. Chemelli-Steingruber b,∗ a
Department of Radiology, Hospital of the Sisters of Charity, Linz, Austria Department of Radiology, Innsbruck Medical University, Austria Department of Trauma Surgery, Innsbruck Medical University, Austria d Department of Orthopedic Surgery, Innsbruck Medical University, Austria b c
a r t i c l e
i n f o
Article history: Received 28 July 2011 Received in revised form 3 November 2011 Accepted 6 November 2011 Keywords: Hemarthrosis Knee Transcatheter arterial embolization Arthroplasty Arthroscopy Trauma
a b s t r a c t Background: The purpose of this retrospective study was to evaluate transarterial catheter embolization (TAE) for the management of hemarthrosis of the knee in 35 patients treated at two different hospitals. Methods: From June 1998 through January 2011, 35 patients (22 men and 13 women, mean age 57 years) underwent TAE for hemarthrosis of the knee using polyvinyl alcohol particles (PVA particles), multicurled 0.018 in. microcoils or a combination of both. In one patient a detachable microcoil was used. Hemarthrosis developed after arthroscopy in 9, after trauma in 3, after arthroplasty in 18 and after sepsis in 2 patients. 2 patients had severe gonarthrosis and in one patient hemarthrosis was due to Marfan’s syndrome. Angiographies showed abnormal and increased vasculature in 23 patients, one or more pseudoaneurysms in 8 and arterio-venous fistula in 2 patients. One patient showed both, pseudoaneurysm and hypervascularization and another one pseudoaneurysm and arterio-venous fistula. Results: Technical success was achieved in 100%. None of the patients developed procedure-related complications such as periarticular skin- or tissue necrosis, including a patient who underwent TAE of two different bleeding sources in two consecutive sessions. Clinical success was 93.4%. Two patients showed recurrent swelling of the knee, 377 and 824 days after TAE respectively. However, only one of them required secondary TAE because of abnormal and increased vasculature. Conclusions: In our view, TAE is the treatment of choice for the management of hemarthrosis of the knee. It is an effective and minimally invasive technique with very low complication rates. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Hemarthrosis of the knee may be due to various reasons including anticoagulant medication or bleeding disorders [1–3]. It may also be associated with several diseases including arthritis, collagen-, vascular- and myeloproliferative disorders, pigmented villonodular synovitis or with iatrogenic vascular complications such as aneurysm, pseudoaneurysm and arterio-venous fistula [2,4] after arthrotomy. Hemarthrosis may also occur due to hypervascularized synovial hypertrophy, which in turn may be caused by mechanical trauma of the knee or the diseases mentioned above
∗ Corresponding author at: Department of Radiology, Innsbruck Medical University, Anichstrasse 35, A-6020 Innsbruck, Austria. Tel.: +43 512 504 81843; fax: +43 512 504 22758. E-mail addresses: [email protected] (P. Waldenberger), [email protected] (A. Chemelli), [email protected] (A. Hennerbichler), [email protected] (M. Wick), [email protected] (M.C. Freund), [email protected] (W. Jaschke), [email protected] (M. Thaler), [email protected] (I.E. Chemelli-Steingruber). 0720-048X/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2011.11.016
[5–7]. On conventional angiography or MR-angiography hypervascularized synovial hypertrophy appears as a vascular blush around the knee. Hemarthrosis of the knee may resolve spontaneously. Therefore, conservative management with immobilization of the joint, application of ice packs and medication, should be the first therapeutic step unless pain and/or blood loss require immediate intervention. Patients in whom conservative management has failed or is not indicated, surgical synovectomy has been the option of choice for the treatment of knee hemarthrosis for many years [8]. Another option has been open repair such as, vessel clipping or thermoablation. However, in patients with hypervascularized synovial hypertrophy these techniques may have devastating sequelae such as uncontrollable bleeding and/or necrosis, as the hypervascularized tissue is fragile and tends to bleed quickly. Moreover, open repair is time-consuming, because the bleeding source is often difficult to localize. Several authors have reported on TAE in patients presenting with hemarthrosis of the knee. However these studies included only small patient populations [1,4,6,8–10]. The purpose of the
2738
P. Waldenberger et al. / European Journal of Radiology 81 (2012) 2737–2740
present retrospective study was to evaluate TAE for the management of hemarthrosis of the knee in a larger population including 34 patients treated at two different hospitals. 2. Materials and methods 2.1. Patients The present study includes 35 patients (22 men and 13 women, mean age 57 years, range 15–85) who underwent TAE for the management of hemarthrosis of the knee from June 1998 through January 2011, either at a primary or at a secondary hospital with an orthopedic emphasis. In order to avoid a too heterogeneous patient population, patients who were treated with anticoagulants for a known preexisting disease or coagulation disorder were not included in this study, although anticoagulative therapy is not a contraindication for TAE. In 9 patients hemarthrosis occurred after arthroscopy. One of them required two TAE procedures for the management of two different bleeding sources 14 and 453 days after arthroscopy respectively. In 3 patients, hemarthrosis developed after trauma, in 18 after knee joint replacement and in 2 after sepsis, which occurred following the resection of a Baker’s cyst in one of them. Two patients presented with severe gonarthrosis and in one patient hemarthrosis was due to Marfan’s syndrome. Indications for TAE were acute pain and recurrent intraarticular hemorrhage causing swelling of the knee in all patients. In addition, hemorrhage was confirmed by aspiration of bloody fluid from the joint in 12 patients. Three of them were punctured once, three of them twice and 6 several times. In the remaining patients, hemarthrosis of the knee was confirmed by magnetic resonance imaging. None of the patients showed decline of hemoglobin levels. 2.2. Diagnostic work-up To identify the exact position of the bleeding area, diagnostic angiography was performed in all patients using a 5F multipurpose or a 4F vertebral catheter (Cordis, Roden, Netherlands) or a Sheperd Hook 5F (Cordis, Roden, Netherlands). After this superselective injections of contrast agent (Iodixanol, Visipaque® , GE Healthcare, Cork, Ireland) into the geniculate branches of the popliteal artery were performed using a microcatheter (Turbotracker 18, Boston Scientific, Boston, MA, USA; Progreat Microcatheter System, Terumo, Japan; Rebar-18, eV3 Inc., Plymouth, MA, USA). Abnormal and increased vasculature was observed in 23 patients (Fig. 1a–c). 8 patients showed one (n = 3) or more (n = 5) pseudoaneurysms and 2 an arterio-venous fistula. One patient showed both, pseudoaneurysm and hypervascularization and one pseudoaneurysm and arterio-venous fistula (Fig. 2a–c). The medial aspect of the knee was involved in 13 of them, the lateral in 17 and both, the medial and lateral aspects in 5 of them. The arterial feeders included 7 superior medial genicular arteries, 14 superior lateral genicular arteries, 6 inferior medial genicular arteries, 8 inferior lateral genicular arteries, 3 middle genicular arteries, 2 sural arteries and 4 descending genicular arteries. Written informed consent was obtained from all patients. The decision for TAE was made by the respective teams of trauma/orthopedic surgeons, anesthesiologists and interventional radiologists at both hospitals. Due to the retrospective nature of the study, it was not necessary to obtain a review of the study by the local ethical review board. 2.3. Technique All procedures were performed in an angiographic suite (Integris BV 3000, Philips, Eindhoven, Netherlands, Europe; Multistar,
Fig. 1. A 52 year-old male patient with severe swelling and pain after arthroscopy. Angiography 16 days after surgery shows an increase in vascularity on the medial aspect of the knee. Selective catheterization of the middle geniculate artery followed by embolization using PVA particles and microcoils. Completion angiography shows successful intervention. The postoperative follow-up was uneventful.
Siemens, Erlangen, Germany, Europe). 33 patients were treated under local anesthesia and 1 under general anesthesia. In the patient who underwent two TAE procedures, both interventions were performed under local anesthesia. In 19 patients transarterial access was obtained using a right and in 16 using a left femoral approach. Antegrade puncture of the femoral artery was performed in 30 and a cross-over procedure in 5 patients. In the patient who required two TAE procedures, the second puncture was antegrade. TAE was performed using a coaxial catheter technique by introducing a 0.018 microcatheter (Tracker-18, Fast Tracker; Target, Boston Scientific, Boston, MA, USA). Polyvinyl alcohol particles (PVA: 150–500 m, Contour® Target, Boston Scientific, Boston, MA, USA) were injected in 12 and multicurled 0.018 in. microcoils (2 mm, MWCE 18-2.0-2-hilal, COOK, Bloomington, IN, USA) using a coilpusher were inserted in 12 patients (Coilpusher 18, Target, Boston Scientific, Boston, MA, USA). A combination of both, multi-curled 0.018 microcoils and PVA particles were used in 9 patients. In one patient, 4 detachable microcoil (6 mm, DCS, COOK, Bloomington, Indiana) was implanted using a detachable device system (DLD, COOK, Bloomington, Indiana). This patient required secondary TAE, using PVA particles (150–250 m) and multi-curled 0.018 microcoils. In all patients, completion angiography was performed in order to document effective embolization of the bleeding source and to rule out embolization of non-target vessels.
P. Waldenberger et al. / European Journal of Radiology 81 (2012) 2737–2740
2739
3. Results Retrospective analysis of the post interventional course extended over a mean of 29 months (range, 0–110 months). Technical success was 100%. None of the patients developed procedure-related complications such as periarticular skin- or tissue necrosis. No signs of rebleeding such as pain, tenderness and/or swelling of the knee were apparent during hospitalization. On discharge, clinical and laboratory examinations were normal in all patients. In the patient, who presented with 2 different bleeding sources, 2 primary TAE procedures were performed in two consecutive sessions, one for pseudoaneurysm in combination with an arterio-venous fistula and the second one for hypervascularization of the knee. Both procedures were successful. Clinical success was 93.4%. Two patients showed recurrent bleeding of the knee, 377 and 824 days after TAE respectively. The first patient required secondary TAE, which was performed 453 days after the first intervention, following attempted conservative treatment by immobilization of the joint, application of ice packs and medication. In the second patient hemarthrosis resolved spontaneously. Both patients showed abnormal and increased vasculature of the knee.
4. Discussion
Fig. 2. 54 year-old male patient with hemarthrosis after arthroscopy. Angiography shows a large pseudoaneurysm and an arterio-venous fistula. Transcatheter arterial embolization using four detachable microcoils to occlude the pseudoaneurysm. Completion angiography shows successful intervention. The pseudoaneurysm is completely occluded and no venous drain from the arterio-venous fistula is visualized.
2.4. Follow-up protocol The follow-up protocol included clinical examination, laboratory tests and radiography of the knee before discharge. After discharge, no specific follow-up protocol was required. However, imaging and/or clinical examinations were performed if patients showed or reported symptoms of recurrent hemorrhage. Four patients were lost to follow-up. 2.5. Definition Technical success was defined as complete cessation of hemorrhage as documented by completion angiography. Clinical success was defined as freedom of recurrence of hemorrhage within the follow-up period.
The need for elective orthopedic and trauma surgery of the knee has grown, which may be due to an increase in elderly population on the one hand and increasing sport activity on the other hand. In our series, the most common causes for hemarthrosis of the knee were arthroscopy or arthroplasty. Several authors claim that hemarthrosis following these interventions is due to the exothermic reaction of the cement, direct trauma, tourniquet application [11], and/or impingement of hypertrophic synovial tissue between prosthesis components [12–14]. However, hemarthrosis after arthrotomy may also be due to hypervascularization of hypertrophic synovial tissue, which tends to be even more fragile in patients suffering from comorbidities, such as hypertension or diabetes mellitus [8]. The link between the hypervascularization blush, seen on angiography and hypertrophic synovial tissue, was postulated by Pham et al. [9]. However, the mechanism underlying spontaneous hemarthrosis of the knee and hypertrophic synovial tissue has not yet been completely understood. Pritsch et al. stated that hypertrophic synovial tissue could be the result of previous bleeding [10]. Accumulation of blood in the joint may cause the activation and recruitment of inflammatory mediators and the onset of synovitis, which in turn may increase the vascularity of the joint. In 8 (9 with the patient with two TAEs) of our patients, hemarthrosis may have occurred as a result of this process, including the patients with hemarthrosis after trauma and sepsis, those with gonarthrosis and Marfan’s syndrome and the patient who underwent 2 primary TAE procedures. In this patient the onset of synovitis after the first TAE may have induced the hypervascularization of the knee necessitating the second intervention. Hypervascularized synovial tissue hypertrophy, which may lead to hemorrhage either spontaneously or as a result of a later bagatelle trauma, can also be induced by microtrauma caused by any knee manipulation, including diagnostic procedures such as knee punctures. Therefore, it is our policy to avoid unnecessary manipulation of the joint and to try conventional management first, unless pain, swelling and/or blood loss require immediate intervention. In the patient who finally needed secondary intervention, we attempted conventional management for 76 days. Hypervascular blush may also be observed in septic or aseptic inflammatory reactions of the knee. This was seen in two of our patients presenting with sepsis of the knee and in the two
2740
P. Waldenberger et al. / European Journal of Radiology 81 (2012) 2737–2740
patients with gonarthrosis. The association between gonarthrosis and hemarthrosis is well documented, but in most of these cases, hemarthrosis is the result of a tear of the menisci [15,16]. In the literature, there is only one report about synovial vascular abnormality causing recurrent hemarthrosis without meniscal injury in an 84 year-old man [17]. In our view, gonarthrosis may be associated with hemarthrosis, especially in those patients in whom misalignment and instability of the joint lead to microtrauma and bleeding. Medication altering blood coagulation may also induce hemorrhage [12]. Therefore, we excluded patients requiring anticoagulant medication due to preexisting coagulation disorders from this study. However, knee manipulations such as arthroscopy and joint replacement also require anticoagulant medication which may cause postinterventional hemarthrosis. This may have contributed to hemorrhage in 26 of our patients. Hemarthrosis may occur in an early period as well as in a delayed period, several months after intervention [8,12,14]. It is known that late spontaneous hemarthrosis occurs in 0.3–1% of patients who underwent knee arthroplasty [10,12]. In our patient population, there were 20 early (within 30 days postinterventionally) and 14 late hemarthroses. However, we could not find any association between a particular etiology and the time of onset of hemarthrosis requiring intervention, nor did early or late onset of hemarthrosis influence the outcome of TAE. In a recent multicentre study Dhondt et al. have shown that embolization is not always successful at the first attempt, but often when repeated [5]. In our study, recurrent hemarthrosis was present in two patients with diffuse hypervascularization at the time of primary TAE. Although MRI and clinical examinations clearly showed hemarthrosis, it resolved spontaneously in one of them. In the second patient, described above, secondary intervention was successful. Arterio-venous fistulae and pseudoaneurysms may occur due to unrecognized injury to a periarticular vessel during knee intervention. In the literature, pseudoaneurysm is reported to be the most common vascular lesion causing hemarthrosis [10,18–20]. On clinical examination, these lesions are typically undetected as they are often too small to be detected on clinical examination and may even be missed on angiography, especially in patients after arthroplasty, when they may be hidden by components of the prosthesis. Therefore, various angiographic views should be obtained. Pseudoaneurysm and arterio-venous fistula of the geniculate arteries may have the same etiology and can both be successfully treated by TAE [21]. The selection of embolic agents depends on various factors such as location and kind of lesion, material characteristics, personal experience and preference of the interventionalist and costs. In most of our patients, we used PVA particles. Contrary to calibrated particles, they have no deep penetration and accelerate thrombosis. Furthermore, they can be applied in very small vessels, where the placement of coils is difficult due to limited catheter access. In patients, in whom the placement of PVA particles alone was not efficient, microcoils were placed in addition to PVA particles, which was necessary in 9 patients. The additional use of coils after PVA embolization prevents recanalization. This is especially important in vascular territories with an extensive collateral network. Another technical detail includes our technique of “closing the back door”, which means to prevent retrograde filling by occluding all vessels potentially feeding the pseudoaneurysm or fistula prior to embolization of the proximal part of the bleeding vessel.
In our view, TAE is the treatment of choice for the invasive management of hemarthrosis of the knee when conservative options fail. Our study shows that it is an effective and minimally invasive technique with very low complication rates. However, hemarthrosis may resolve spontaneously and therefore, conservative management should be the first therapeutic step, so as to avoid further trauma to the knee joint by manipulation. Conflict of interest All authors declare no financial and non-financial competing interests. References [1] Given MF, Smith P, Lyon SM, Robertson D, Thomson KR. Embolization of spontaneous hemarthrosis post total knee replacement. Cardiovasc Intervent Radiol 2008;31:986–8. [2] Saksena J, Platts AD, Dowd GS. Recurrent haemarthrosis following total knee replacement. Knee 2010;17:7–14. [3] Small M, Steven MM, Freeman PA, et al. Total knee arthroplasty in haemophilic arthritis. J Bone Joint Surg Br 1983;65:163–5. [4] Karataglis D, Marlow D, Learmonth DJ. Atraumatic haemarthrosis following total knee replacement treated with selective embolisation. Acta Orthop Belg 2006;72:375–7. [5] Dhondt E, Vanhoenacker FM, D’ AO, Snoeckx A, Defreyne L. Angiographic findings and therapeutic embolization of late hemarthrosis after total joint arthroplasty. Skeletal Radiol 2009;38:31–6. [6] Katsimihas M, Robinson D, Thornton M, Langkamer VG. Therapeutic embolization of the genicular arteries for recurrent hemarthrosis after total knee arthroplasty. J Arthroplasty 2001;16:935–7. [7] Maheshwari R, Kelley SP, Langkamer VG, Loveday E. Spontaneous recurrent haemarthrosis following unicompartmental knee arthroplasty and its successful treatment by coil embolisation. Knee 2004;11:413– 5. [8] Ohdera T, Tokunaga M, Hiroshima S, Yoshimoto E, Matsuda S. Recurrent hemarthrosis after knee joint arthroplasty: etiology and treatment. J Arthroplasty 2004;19:157–61. [9] Pham TT, Bouloudian S, Moreau PE, et al. Recurrent hemarthrosis following total knee arthroplasty. Report of a case treated with arterial embolization. Joint Bone Spine 2003;70:58–60. [10] Pritsch T, Pritsch M, Halperin N. Therapeutic embolization for late hemarthrosis after total knee arthroplasty. A case report. J Bone Joint Surg Am 2003;85A:1802–4. [11] Langkamer VG. Local vascular complications after knee replacement: a review with illustrative case reports. Knee 2001;8:259–64. [12] Kindsfater K, Scott R. Recurrent hemarthrosis after total knee arthroplasty. J Arthroplasty 1995;(10 (Suppl.)):S52–5. [13] Oishi CS, Elliott ML, Colwell Jr CW. Recurrent hemarthrosis following a total knee arthroplasty. J Arthroplasty 1995;(10 (Suppl.)):S56– 8. [14] Worland RL, Jessup DE. Recurrent hemarthrosis after total knee arthroplasty. J Arthroplasty 1996;11:977–8. [15] Sasho T, Ogino S, Tsuruoka H, et al. Spontaneous recurrent hemarthrosis of the knee in the elderly: arthroscopic treatment and etiology. Arthroscopy 2008;24:1027–33. [16] Kawamura H, Ogata K, Miura H, Arizono T, Sugioka Y. Spontaneous hemarthrosis of the knee in the elderly: etiology and treatment. Arthroscopy 1994;10:171–5. [17] Blyth T, Hunter J, Madhok R, Vallance R. Subsynovial vascular abnormality causing recurrent hemarthrosis in an 84-year-old man. J Rheumatol 1995;22:552–3. [18] Becher C, Burger UL, Allenberg JR, Kaufmann GW, Thermann H. Delayed diagnosis of a pseudoaneurysm with recurrent hemarthrosis of the knee joint. Knee Surg Sports Traumatol Arthrosc 2008;16:561–4. [19] Hozack WJ, Cole PA, Gardner R, Corces A. Popliteal aneurysm after total knee arthroplasty. Case reports and review of the literature. J Arthroplasty 1990;5:301–5. [20] Woolgar JD, Reddy DS, Robbs JV. Delayed presentation of traumatic popliteal artery pseudoaneurysms: a review of seven cases. Eur J Vasc Endovasc Surg 2002;23:255–9. [21] Kalsi PS, Carrington RJ, Skinner JS. Therapeutic embolization for the treatment of recurrent hemarthrosis after total knee arthroplasty due to an arteriovenous fistula. J Arthroplasty 2007;22:1223–5.
Contents lists available at SciVerse ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Transarterial embolization for the management of hemarthrosis of the knee P. Waldenberger a , A. Chemelli b , A. Hennerbichler c , M. Wick b , M.C. Freund b , W. Jaschke b , M. Thaler d , I.E. Chemelli-Steingruber b,∗ a
Department of Radiology, Hospital of the Sisters of Charity, Linz, Austria Department of Radiology, Innsbruck Medical University, Austria Department of Trauma Surgery, Innsbruck Medical University, Austria d Department of Orthopedic Surgery, Innsbruck Medical University, Austria b c
a r t i c l e
i n f o
Article history: Received 28 July 2011 Received in revised form 3 November 2011 Accepted 6 November 2011 Keywords: Hemarthrosis Knee Transcatheter arterial embolization Arthroplasty Arthroscopy Trauma
a b s t r a c t Background: The purpose of this retrospective study was to evaluate transarterial catheter embolization (TAE) for the management of hemarthrosis of the knee in 35 patients treated at two different hospitals. Methods: From June 1998 through January 2011, 35 patients (22 men and 13 women, mean age 57 years) underwent TAE for hemarthrosis of the knee using polyvinyl alcohol particles (PVA particles), multicurled 0.018 in. microcoils or a combination of both. In one patient a detachable microcoil was used. Hemarthrosis developed after arthroscopy in 9, after trauma in 3, after arthroplasty in 18 and after sepsis in 2 patients. 2 patients had severe gonarthrosis and in one patient hemarthrosis was due to Marfan’s syndrome. Angiographies showed abnormal and increased vasculature in 23 patients, one or more pseudoaneurysms in 8 and arterio-venous fistula in 2 patients. One patient showed both, pseudoaneurysm and hypervascularization and another one pseudoaneurysm and arterio-venous fistula. Results: Technical success was achieved in 100%. None of the patients developed procedure-related complications such as periarticular skin- or tissue necrosis, including a patient who underwent TAE of two different bleeding sources in two consecutive sessions. Clinical success was 93.4%. Two patients showed recurrent swelling of the knee, 377 and 824 days after TAE respectively. However, only one of them required secondary TAE because of abnormal and increased vasculature. Conclusions: In our view, TAE is the treatment of choice for the management of hemarthrosis of the knee. It is an effective and minimally invasive technique with very low complication rates. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Hemarthrosis of the knee may be due to various reasons including anticoagulant medication or bleeding disorders [1–3]. It may also be associated with several diseases including arthritis, collagen-, vascular- and myeloproliferative disorders, pigmented villonodular synovitis or with iatrogenic vascular complications such as aneurysm, pseudoaneurysm and arterio-venous fistula [2,4] after arthrotomy. Hemarthrosis may also occur due to hypervascularized synovial hypertrophy, which in turn may be caused by mechanical trauma of the knee or the diseases mentioned above
∗ Corresponding author at: Department of Radiology, Innsbruck Medical University, Anichstrasse 35, A-6020 Innsbruck, Austria. Tel.: +43 512 504 81843; fax: +43 512 504 22758. E-mail addresses: [email protected] (P. Waldenberger), [email protected] (A. Chemelli), [email protected] (A. Hennerbichler), [email protected] (M. Wick), [email protected] (M.C. Freund), [email protected] (W. Jaschke), [email protected] (M. Thaler), [email protected] (I.E. Chemelli-Steingruber). 0720-048X/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2011.11.016
[5–7]. On conventional angiography or MR-angiography hypervascularized synovial hypertrophy appears as a vascular blush around the knee. Hemarthrosis of the knee may resolve spontaneously. Therefore, conservative management with immobilization of the joint, application of ice packs and medication, should be the first therapeutic step unless pain and/or blood loss require immediate intervention. Patients in whom conservative management has failed or is not indicated, surgical synovectomy has been the option of choice for the treatment of knee hemarthrosis for many years [8]. Another option has been open repair such as, vessel clipping or thermoablation. However, in patients with hypervascularized synovial hypertrophy these techniques may have devastating sequelae such as uncontrollable bleeding and/or necrosis, as the hypervascularized tissue is fragile and tends to bleed quickly. Moreover, open repair is time-consuming, because the bleeding source is often difficult to localize. Several authors have reported on TAE in patients presenting with hemarthrosis of the knee. However these studies included only small patient populations [1,4,6,8–10]. The purpose of the
2738
P. Waldenberger et al. / European Journal of Radiology 81 (2012) 2737–2740
present retrospective study was to evaluate TAE for the management of hemarthrosis of the knee in a larger population including 34 patients treated at two different hospitals. 2. Materials and methods 2.1. Patients The present study includes 35 patients (22 men and 13 women, mean age 57 years, range 15–85) who underwent TAE for the management of hemarthrosis of the knee from June 1998 through January 2011, either at a primary or at a secondary hospital with an orthopedic emphasis. In order to avoid a too heterogeneous patient population, patients who were treated with anticoagulants for a known preexisting disease or coagulation disorder were not included in this study, although anticoagulative therapy is not a contraindication for TAE. In 9 patients hemarthrosis occurred after arthroscopy. One of them required two TAE procedures for the management of two different bleeding sources 14 and 453 days after arthroscopy respectively. In 3 patients, hemarthrosis developed after trauma, in 18 after knee joint replacement and in 2 after sepsis, which occurred following the resection of a Baker’s cyst in one of them. Two patients presented with severe gonarthrosis and in one patient hemarthrosis was due to Marfan’s syndrome. Indications for TAE were acute pain and recurrent intraarticular hemorrhage causing swelling of the knee in all patients. In addition, hemorrhage was confirmed by aspiration of bloody fluid from the joint in 12 patients. Three of them were punctured once, three of them twice and 6 several times. In the remaining patients, hemarthrosis of the knee was confirmed by magnetic resonance imaging. None of the patients showed decline of hemoglobin levels. 2.2. Diagnostic work-up To identify the exact position of the bleeding area, diagnostic angiography was performed in all patients using a 5F multipurpose or a 4F vertebral catheter (Cordis, Roden, Netherlands) or a Sheperd Hook 5F (Cordis, Roden, Netherlands). After this superselective injections of contrast agent (Iodixanol, Visipaque® , GE Healthcare, Cork, Ireland) into the geniculate branches of the popliteal artery were performed using a microcatheter (Turbotracker 18, Boston Scientific, Boston, MA, USA; Progreat Microcatheter System, Terumo, Japan; Rebar-18, eV3 Inc., Plymouth, MA, USA). Abnormal and increased vasculature was observed in 23 patients (Fig. 1a–c). 8 patients showed one (n = 3) or more (n = 5) pseudoaneurysms and 2 an arterio-venous fistula. One patient showed both, pseudoaneurysm and hypervascularization and one pseudoaneurysm and arterio-venous fistula (Fig. 2a–c). The medial aspect of the knee was involved in 13 of them, the lateral in 17 and both, the medial and lateral aspects in 5 of them. The arterial feeders included 7 superior medial genicular arteries, 14 superior lateral genicular arteries, 6 inferior medial genicular arteries, 8 inferior lateral genicular arteries, 3 middle genicular arteries, 2 sural arteries and 4 descending genicular arteries. Written informed consent was obtained from all patients. The decision for TAE was made by the respective teams of trauma/orthopedic surgeons, anesthesiologists and interventional radiologists at both hospitals. Due to the retrospective nature of the study, it was not necessary to obtain a review of the study by the local ethical review board. 2.3. Technique All procedures were performed in an angiographic suite (Integris BV 3000, Philips, Eindhoven, Netherlands, Europe; Multistar,
Fig. 1. A 52 year-old male patient with severe swelling and pain after arthroscopy. Angiography 16 days after surgery shows an increase in vascularity on the medial aspect of the knee. Selective catheterization of the middle geniculate artery followed by embolization using PVA particles and microcoils. Completion angiography shows successful intervention. The postoperative follow-up was uneventful.
Siemens, Erlangen, Germany, Europe). 33 patients were treated under local anesthesia and 1 under general anesthesia. In the patient who underwent two TAE procedures, both interventions were performed under local anesthesia. In 19 patients transarterial access was obtained using a right and in 16 using a left femoral approach. Antegrade puncture of the femoral artery was performed in 30 and a cross-over procedure in 5 patients. In the patient who required two TAE procedures, the second puncture was antegrade. TAE was performed using a coaxial catheter technique by introducing a 0.018 microcatheter (Tracker-18, Fast Tracker; Target, Boston Scientific, Boston, MA, USA). Polyvinyl alcohol particles (PVA: 150–500 m, Contour® Target, Boston Scientific, Boston, MA, USA) were injected in 12 and multicurled 0.018 in. microcoils (2 mm, MWCE 18-2.0-2-hilal, COOK, Bloomington, IN, USA) using a coilpusher were inserted in 12 patients (Coilpusher 18, Target, Boston Scientific, Boston, MA, USA). A combination of both, multi-curled 0.018 microcoils and PVA particles were used in 9 patients. In one patient, 4 detachable microcoil (6 mm, DCS, COOK, Bloomington, Indiana) was implanted using a detachable device system (DLD, COOK, Bloomington, Indiana). This patient required secondary TAE, using PVA particles (150–250 m) and multi-curled 0.018 microcoils. In all patients, completion angiography was performed in order to document effective embolization of the bleeding source and to rule out embolization of non-target vessels.
P. Waldenberger et al. / European Journal of Radiology 81 (2012) 2737–2740
2739
3. Results Retrospective analysis of the post interventional course extended over a mean of 29 months (range, 0–110 months). Technical success was 100%. None of the patients developed procedure-related complications such as periarticular skin- or tissue necrosis. No signs of rebleeding such as pain, tenderness and/or swelling of the knee were apparent during hospitalization. On discharge, clinical and laboratory examinations were normal in all patients. In the patient, who presented with 2 different bleeding sources, 2 primary TAE procedures were performed in two consecutive sessions, one for pseudoaneurysm in combination with an arterio-venous fistula and the second one for hypervascularization of the knee. Both procedures were successful. Clinical success was 93.4%. Two patients showed recurrent bleeding of the knee, 377 and 824 days after TAE respectively. The first patient required secondary TAE, which was performed 453 days after the first intervention, following attempted conservative treatment by immobilization of the joint, application of ice packs and medication. In the second patient hemarthrosis resolved spontaneously. Both patients showed abnormal and increased vasculature of the knee.
4. Discussion
Fig. 2. 54 year-old male patient with hemarthrosis after arthroscopy. Angiography shows a large pseudoaneurysm and an arterio-venous fistula. Transcatheter arterial embolization using four detachable microcoils to occlude the pseudoaneurysm. Completion angiography shows successful intervention. The pseudoaneurysm is completely occluded and no venous drain from the arterio-venous fistula is visualized.
2.4. Follow-up protocol The follow-up protocol included clinical examination, laboratory tests and radiography of the knee before discharge. After discharge, no specific follow-up protocol was required. However, imaging and/or clinical examinations were performed if patients showed or reported symptoms of recurrent hemorrhage. Four patients were lost to follow-up. 2.5. Definition Technical success was defined as complete cessation of hemorrhage as documented by completion angiography. Clinical success was defined as freedom of recurrence of hemorrhage within the follow-up period.
The need for elective orthopedic and trauma surgery of the knee has grown, which may be due to an increase in elderly population on the one hand and increasing sport activity on the other hand. In our series, the most common causes for hemarthrosis of the knee were arthroscopy or arthroplasty. Several authors claim that hemarthrosis following these interventions is due to the exothermic reaction of the cement, direct trauma, tourniquet application [11], and/or impingement of hypertrophic synovial tissue between prosthesis components [12–14]. However, hemarthrosis after arthrotomy may also be due to hypervascularization of hypertrophic synovial tissue, which tends to be even more fragile in patients suffering from comorbidities, such as hypertension or diabetes mellitus [8]. The link between the hypervascularization blush, seen on angiography and hypertrophic synovial tissue, was postulated by Pham et al. [9]. However, the mechanism underlying spontaneous hemarthrosis of the knee and hypertrophic synovial tissue has not yet been completely understood. Pritsch et al. stated that hypertrophic synovial tissue could be the result of previous bleeding [10]. Accumulation of blood in the joint may cause the activation and recruitment of inflammatory mediators and the onset of synovitis, which in turn may increase the vascularity of the joint. In 8 (9 with the patient with two TAEs) of our patients, hemarthrosis may have occurred as a result of this process, including the patients with hemarthrosis after trauma and sepsis, those with gonarthrosis and Marfan’s syndrome and the patient who underwent 2 primary TAE procedures. In this patient the onset of synovitis after the first TAE may have induced the hypervascularization of the knee necessitating the second intervention. Hypervascularized synovial tissue hypertrophy, which may lead to hemorrhage either spontaneously or as a result of a later bagatelle trauma, can also be induced by microtrauma caused by any knee manipulation, including diagnostic procedures such as knee punctures. Therefore, it is our policy to avoid unnecessary manipulation of the joint and to try conventional management first, unless pain, swelling and/or blood loss require immediate intervention. In the patient who finally needed secondary intervention, we attempted conventional management for 76 days. Hypervascular blush may also be observed in septic or aseptic inflammatory reactions of the knee. This was seen in two of our patients presenting with sepsis of the knee and in the two
2740
P. Waldenberger et al. / European Journal of Radiology 81 (2012) 2737–2740
patients with gonarthrosis. The association between gonarthrosis and hemarthrosis is well documented, but in most of these cases, hemarthrosis is the result of a tear of the menisci [15,16]. In the literature, there is only one report about synovial vascular abnormality causing recurrent hemarthrosis without meniscal injury in an 84 year-old man [17]. In our view, gonarthrosis may be associated with hemarthrosis, especially in those patients in whom misalignment and instability of the joint lead to microtrauma and bleeding. Medication altering blood coagulation may also induce hemorrhage [12]. Therefore, we excluded patients requiring anticoagulant medication due to preexisting coagulation disorders from this study. However, knee manipulations such as arthroscopy and joint replacement also require anticoagulant medication which may cause postinterventional hemarthrosis. This may have contributed to hemorrhage in 26 of our patients. Hemarthrosis may occur in an early period as well as in a delayed period, several months after intervention [8,12,14]. It is known that late spontaneous hemarthrosis occurs in 0.3–1% of patients who underwent knee arthroplasty [10,12]. In our patient population, there were 20 early (within 30 days postinterventionally) and 14 late hemarthroses. However, we could not find any association between a particular etiology and the time of onset of hemarthrosis requiring intervention, nor did early or late onset of hemarthrosis influence the outcome of TAE. In a recent multicentre study Dhondt et al. have shown that embolization is not always successful at the first attempt, but often when repeated [5]. In our study, recurrent hemarthrosis was present in two patients with diffuse hypervascularization at the time of primary TAE. Although MRI and clinical examinations clearly showed hemarthrosis, it resolved spontaneously in one of them. In the second patient, described above, secondary intervention was successful. Arterio-venous fistulae and pseudoaneurysms may occur due to unrecognized injury to a periarticular vessel during knee intervention. In the literature, pseudoaneurysm is reported to be the most common vascular lesion causing hemarthrosis [10,18–20]. On clinical examination, these lesions are typically undetected as they are often too small to be detected on clinical examination and may even be missed on angiography, especially in patients after arthroplasty, when they may be hidden by components of the prosthesis. Therefore, various angiographic views should be obtained. Pseudoaneurysm and arterio-venous fistula of the geniculate arteries may have the same etiology and can both be successfully treated by TAE [21]. The selection of embolic agents depends on various factors such as location and kind of lesion, material characteristics, personal experience and preference of the interventionalist and costs. In most of our patients, we used PVA particles. Contrary to calibrated particles, they have no deep penetration and accelerate thrombosis. Furthermore, they can be applied in very small vessels, where the placement of coils is difficult due to limited catheter access. In patients, in whom the placement of PVA particles alone was not efficient, microcoils were placed in addition to PVA particles, which was necessary in 9 patients. The additional use of coils after PVA embolization prevents recanalization. This is especially important in vascular territories with an extensive collateral network. Another technical detail includes our technique of “closing the back door”, which means to prevent retrograde filling by occluding all vessels potentially feeding the pseudoaneurysm or fistula prior to embolization of the proximal part of the bleeding vessel.
In our view, TAE is the treatment of choice for the invasive management of hemarthrosis of the knee when conservative options fail. Our study shows that it is an effective and minimally invasive technique with very low complication rates. However, hemarthrosis may resolve spontaneously and therefore, conservative management should be the first therapeutic step, so as to avoid further trauma to the knee joint by manipulation. Conflict of interest All authors declare no financial and non-financial competing interests. References [1] Given MF, Smith P, Lyon SM, Robertson D, Thomson KR. Embolization of spontaneous hemarthrosis post total knee replacement. Cardiovasc Intervent Radiol 2008;31:986–8. [2] Saksena J, Platts AD, Dowd GS. Recurrent haemarthrosis following total knee replacement. Knee 2010;17:7–14. [3] Small M, Steven MM, Freeman PA, et al. Total knee arthroplasty in haemophilic arthritis. J Bone Joint Surg Br 1983;65:163–5. [4] Karataglis D, Marlow D, Learmonth DJ. Atraumatic haemarthrosis following total knee replacement treated with selective embolisation. Acta Orthop Belg 2006;72:375–7. [5] Dhondt E, Vanhoenacker FM, D’ AO, Snoeckx A, Defreyne L. Angiographic findings and therapeutic embolization of late hemarthrosis after total joint arthroplasty. Skeletal Radiol 2009;38:31–6. [6] Katsimihas M, Robinson D, Thornton M, Langkamer VG. Therapeutic embolization of the genicular arteries for recurrent hemarthrosis after total knee arthroplasty. J Arthroplasty 2001;16:935–7. [7] Maheshwari R, Kelley SP, Langkamer VG, Loveday E. Spontaneous recurrent haemarthrosis following unicompartmental knee arthroplasty and its successful treatment by coil embolisation. Knee 2004;11:413– 5. [8] Ohdera T, Tokunaga M, Hiroshima S, Yoshimoto E, Matsuda S. Recurrent hemarthrosis after knee joint arthroplasty: etiology and treatment. J Arthroplasty 2004;19:157–61. [9] Pham TT, Bouloudian S, Moreau PE, et al. Recurrent hemarthrosis following total knee arthroplasty. Report of a case treated with arterial embolization. Joint Bone Spine 2003;70:58–60. [10] Pritsch T, Pritsch M, Halperin N. Therapeutic embolization for late hemarthrosis after total knee arthroplasty. A case report. J Bone Joint Surg Am 2003;85A:1802–4. [11] Langkamer VG. Local vascular complications after knee replacement: a review with illustrative case reports. Knee 2001;8:259–64. [12] Kindsfater K, Scott R. Recurrent hemarthrosis after total knee arthroplasty. J Arthroplasty 1995;(10 (Suppl.)):S52–5. [13] Oishi CS, Elliott ML, Colwell Jr CW. Recurrent hemarthrosis following a total knee arthroplasty. J Arthroplasty 1995;(10 (Suppl.)):S56– 8. [14] Worland RL, Jessup DE. Recurrent hemarthrosis after total knee arthroplasty. J Arthroplasty 1996;11:977–8. [15] Sasho T, Ogino S, Tsuruoka H, et al. Spontaneous recurrent hemarthrosis of the knee in the elderly: arthroscopic treatment and etiology. Arthroscopy 2008;24:1027–33. [16] Kawamura H, Ogata K, Miura H, Arizono T, Sugioka Y. Spontaneous hemarthrosis of the knee in the elderly: etiology and treatment. Arthroscopy 1994;10:171–5. [17] Blyth T, Hunter J, Madhok R, Vallance R. Subsynovial vascular abnormality causing recurrent hemarthrosis in an 84-year-old man. J Rheumatol 1995;22:552–3. [18] Becher C, Burger UL, Allenberg JR, Kaufmann GW, Thermann H. Delayed diagnosis of a pseudoaneurysm with recurrent hemarthrosis of the knee joint. Knee Surg Sports Traumatol Arthrosc 2008;16:561–4. [19] Hozack WJ, Cole PA, Gardner R, Corces A. Popliteal aneurysm after total knee arthroplasty. Case reports and review of the literature. J Arthroplasty 1990;5:301–5. [20] Woolgar JD, Reddy DS, Robbs JV. Delayed presentation of traumatic popliteal artery pseudoaneurysms: a review of seven cases. Eur J Vasc Endovasc Surg 2002;23:255–9. [21] Kalsi PS, Carrington RJ, Skinner JS. Therapeutic embolization for the treatment of recurrent hemarthrosis after total knee arthroplasty due to an arteriovenous fistula. J Arthroplasty 2007;22:1223–5.