Treatment of Acute Iliofemoral Deep Venous Thrombosis: A Strategy of Thrombus Removal

Eur J Vasc Endovasc Surg 33, 351e360 (2007) doi:10.1016/j.ejvs.2006.11.013, available online at http://www.sciencedirect.com on Treatment of Acute Iliofemoral Deep Venous Thrombosis: A Strategy of Thrombus Removal A.J. Comerota* and D. Paolini University of Michigan, Jobst Vascular Center, The Toledo Hospital, 2109 Hughes Dr, Suite 400, Toledo, OH 43606, USA Patients with acute iliofemoral deep vein thrombosis (DVT) suffer the most severe postthrombotic sequelae. The majority of physicians treat all patients with acute DVT with anticoagulation alone, despite evidence that postthrombotic chronic venous insufficiency, leg ulceration, and venous claudication are common in patients treated only with anticoagulation. The body of evidence to date in patients with iliofemoral DVT suggests that a strategy of thrombus removal offers these patients the best long-term outcome. Unfortunately, currently published guidelines use outdated experiences to recommend against the use of techniques designed to remove thrombus, ignoring recent clinical studies showing significant benefit in patients who have thrombus eliminated. Contemporary venous thrombectomy, intrathrombus catheter-directed thrombolysis, and pharmacomechanical thrombolysis are all options that can be offered to successfully remove venous thrombus with increasing safety. The authors review evidence supporting the rationale for thrombus removal and discuss the most effective approaches for treating patients with acute iliofemoral DVT. Keywords: Deep venous thrombosis; Catheter-directed thrombolysis; Venous thrombectomy; Pharmacomechanical thrombolysis; Percutaneous mechanical thrombectomy. Introduction Anticoagulation alone is the treatment used by most physicians for patients with acute deep venous thrombosis (DVT). Unfortunately, the extent of thrombosis has little or no bearing on the management decisions made by physicians when treating acute DVT. The American College of Chest Physicians (ACCP) 6th Consensus Conference on Antithrombotic Therapy further diminished enthusiasm by physicians to adopt any strategy of thrombus removal as a result of their statements regarding operative venous thrombectomy and thrombolytic therapy. Buller et al. 1 stated that in patients with DVT, we recommend against the use of venous thrombectomy (Grade 1C). The committee goes on to say that surgical thrombectomy is commonly complicated by a recurrence of thrombus formation. Unfortunately, the authors Presented at the XX Congress of the European Society for Vascular Surgery, September 21e24, Prague, Czech Republic. *Corresponding author. A.J. Comerota, MD, FACS, Jobst Vascular Center, 2109 Hughes Dr, Suite 400, Toledo, OH 43606, USA. E-mail address: marilyn.gravett@promedica.org reference an anecdotal experience in patients treated over 40 years ago. 2,3 The follow-up of these patients was incomplete and biased. Only 50% of the patients originally treated underwent follow-up and only 25% had repeat phlebography. As most physicians know, patients who have the worst clinical outcomes, in these cases postthrombotic morbidity, represent the majority of patients who return for follow-up care. Patients who have minimal or no ongoing symptoms are less likely to seek additional longterm medical care. It is unclear how Buller et al. 1 could have overlooked a randomized trial of venous thrombectomy vs. anticoagulation with long-term follow-up. 4e6 This prospective study documented significant benefit in patients randomized to thrombus extraction. The ACCP consensus conference section stated that there is no evidence that supports the use of thrombolytic agents for the initial treatment of DVT. 1 They recommended against the routine use of catheterdirected thrombolysis (Grade 1C), and they went on to state that thrombolytic therapy be confined to patients requiring limb salvage (Grade 2C). Most physicians recognize that delivery of a thrombolytic agent into the thrombus is considerably more effective than 1078 5884/000351 + 10 $32.00/0 Ó 2006 Elsevier Ltd. All rights reserved.

352 A. J. Comerota and D. Paolini systemic administration. This observation follows the basic mechanism of thrombolysis, which occurs by way of activation of fibrin-bound plasminogen. 7 A randomized trial 8 and a vast clinical experience have demonstrated the improved effectiveness of catheterdirected intrathrombus delivery of plasminogen activators. 9e11 Despite evidence demonstrating that patients with iliofemoral venous thrombosis suffer more severe postthrombotic morbidity than patients with infrainguinal deep venous thrombosis, 12e14 the majority of physicians treat all patients with acute DVT with anticoagulation alone. A treatment approach that includes a strategy of thrombus removal and optimal anticoagulation 15 has not been adopted by most clinicians, even in patients with the most extensive form of venous thrombosis. It appears that patients with iliofemoral DVT are a clinically relevant subset of patients with acute DVT who suffer severe postthrombotic morbidity. 12e14 Following anticoagulation alone for management of iliofemoral DVT, postthrombotic chronic venous insufficiency, leg ulceration, and venous claudication are common. 12e14 It is the thesis of this report that when one considers the body of evidence to date in patients with extensive DVT, a strategy of thrombus removal is the preferred management and offers patients the best long-term outcome. Rationale for Thrombus Removal The pathophysiology of chronic venous insufficiency is ambulatory venous hypertension, defined as an elevated venous pressure during exercise. 16 Ambulatory venous pressure in the lower leg and foot generally drops to less than 50% of the standing venous pressure in normal individuals. In patients with the postthrombotic syndrome, the ambulatory venous pressure drops very little, and in individuals with persistent venous occlusion, the ambulatory venous pressure may actually rise above standing pressure. This degree of ambulatory venous hypertension often leads to the debilitating symptoms of venous claudication. The anatomic components contributing to ambulatory venous hypertension are venous valvular incompetence and luminal obstruction. Studies consistently demonstrate that the most severe postthrombotic morbidity is associated with the highest venous pressures, which occur in patients with valvular incompetence accompanied by luminal venous obstruction. 16,17 Venous obstruction is not synonymous with occlusion. Occlusion is complete obliteration of the lumen, whereas obstruction (in most patients) is relative narrowing of the lumen (Fig. 1). Unfortunately, technology has not advanced to the point that allows physicians to assess the pathophysiologic impact of partial luminal obstruction in an individual patient. Fig. 1. (A) Ascending phlebogram in a patient who had iliofemoral DVT 10 years earlier showed chronic venous disease with no evidence of obstruction, as stated on the x-ray report. Three-second maximal venous outflow performed with impedance plethysmography. A classic Linton procedure, which includes ligation of the femoral vein distal to its junction with the profunda, was performed. A cross-section of the femoral vein (B) illustrates recanalisation of the femoral vein with multiple channels showing significant luminal obstruction.

Treatment of Acute Iliofemoral Deep Venous Thrombosis 353 Therefore, physicians often cannot put venous obstruction into proper pathophysiologic perspective in terms of its contribution to postthrombotic morbidity. Therefore, there is widespread underappreciation regarding the importance of luminal obstruction contributing to postthrombotic morbidity. Experimental observations in canine models of acute DVT demonstrated that thrombolysis preserves endothelial function and valve competence immediately and at four weeks after therapy compared with placebo. There was less residual thrombus in veins treated with a plasminogen activator, thereby preserving the vein s structural integrity. 18,19 These experimental observations translate into improved clinical outcomes as observed in a natural history study of acute DVT in patients treated with anticoagulation. 20e22 These investigators found that patients with persistent proximal vein obstruction developed distal valvular incompetence, even when distal veins were not involved with thrombus. 20 They confirmed that the combination of venous obstruction and valve incompetence was associated with the most severe postthrombotic morbidity. 21 Spontaneous clot lysis naturally restored venous patency. Interestingly, if spontaneous lysis occurred early (within 90 days), valve function was frequently preserved. 22 It is intuitive that successful elimination of thrombus (spontaneous, mechanical, or pharmacologically accelerated) will eliminate luminal obstruction and increase the likelihood of preservation of valve function. It follows that, in patients whose thrombus is resolved (or eliminated), postthrombotic morbidity will be avoided or significantly reduced. Outcomes of a Strategy of Thrombus Removal Contemporary venous thrombectomy Advances in all methods of elimination of thrombus from the deep venous system have occurred during the past 10e15 years. Contemporary venous thrombectomy has substantially improved the early and long-term results of patients with extensive DVT compared to the initial reports. 4e6,23e32 Major technical advances between the initial and contemporary procedures are listed in Table 1. The technical details of contemporary venous thrombectomy have been recently reported. 30,33 An important randomized trial was reported by the Scandinavian investigators, comparing operative venous thrombectomy and arteriovenous fistula versus anticoagulation alone in patients with iliofemoral venous thrombosis. Peer-reviewed reports occurred Table 1. Venous thrombectomy: comparison of old and contemporary techniques Technique Old Contemporary Pretreatment phlebography/ct scan Occasionally Always Venous thrombectomy catheter No Yes Operative fluoroscopy/phlebography No Yes Correct iliac vein stenosis No Yes Arteriovenous fistula No Yes Infrainguinal thrombectomy No Yes Intraoperative thrombolysis No Yes Full post op anticoagulation Occasionally Yes Catheter-directed anticoagulation No Yes Intermittent pneumatic compression post op No Yes at 6 months, 4 5 years, 5 and 10 years 6 of follow-up. Patients randomized to venous thrombectomy demonstrated improved patency (P < 0.05), lower venous pressures (P < 0.05), less edema (P < 0.05), and fewer postthrombotic symptoms (P < 0.05) compared to anticoagulation. Interestingly, more patients undergoing venous thrombectomy preserved normal venous valve function in their femoropopliteal segment than patients treated with anticoagulation alone. Hence, elimination of iliofemoral thrombus (and proximal venous obstruction) protected the distal valves from the hemodynamic consequences of obstructed proximal veins and resultant valvular incompetence of the distal veins. 20 Catheter-directed thrombolysis The mechanism by which thrombolysis results in clot dissolution is the activation of fibrin-bound plasminogen to form the active enzyme plasmin, which dissolves clot. 7 During thrombosis, GLU-plasminogen (circulating) is converted to LYS-plasminogen (in thrombus) as a result of binding to fibrin. LYSplasminogen has more binding sites for plasminogen activators and is more efficiently activated to plasmin than GLU-plasminogen. Intrathrombus delivery naturally protects plasminogen activators from neutralization by circulating plasminogen activator inhibitor (PAI-1) and also protects the resultant active enzyme plasmin from instantaneous neutralization by circulating antiplasmins. Catheter-directed delivery of plasminogen activators into the thrombus accelerates thrombolysis, increasing the likelihood of a successful outcome. Accelerated lysis reduces the overall dose and duration of plasminogen activator infusion; therefore, it is reasonable to expect that complications also will be reduced. Numerous reports have emerged demonstrating favorable outcomes of catheter-directed thrombolysis for acute DVT. 9e11,34e38 Three large reports document approximately an 80% success rate 9e11 (Table 2).

354 A. J. Comerota and D. Paolini Table 2. Results of catheter-directed thrombolysis with urokinase in three contemporary series: efficacy and complications Bjarnason, et al. 9 (n ¼ 77) Mewissen, Comerota, et al. 10 (n ¼ 287) et al. 11 (n ¼ 58) Efficacy Initial Success 79% 83% 84% Iliac 63% 64% 78% Femoral 40% 47% e Primary Patency at 1 yr Iliac 63% 64% 78% Femoral 40% 47% e Iliac Stent: Patency at 1 yr þstent 54% 74% 89% Stent 75% 53% 71% Complications Major Bleed 5% 11% 9% Intracranial 0% <1% 0% Bleeding Pulmonary 1% 1% 0% Embolism Fatal Pulmonary 0% 0.2% 0% Embolism Death Secondary to lysis 0% 0.4% 0% (?2%)* * Death due to multiorgan system failure 30-days post lysis, thought not related to lytic therapy. Initial success rates likely would have been higher if treatment had been restricted only to patients with acute iliofemoral DVT. However, as clinicians gained confidence with the technique, patients with more chronic thrombus were treated, resulting in lower overall success rates. In these 3 studies, 422 patients were treated with consistent rates of success and complications. Importantly, underlying iliac vein stenoses were treated with balloon angioplasty, stenting, or both to achieve unobstructed venous drainage into the vena cava, thereby reducing the risk of recurrent thrombosis. Major bleeding complications occurred in 5e10% of cases, with the majority occurring at the puncture site. Fortunately, intracranial bleeding was rare, occurring in only 3 patients in the National Venous Registry. 10 Symptomatic pulmonary embolism (PE) occurred in 1% of patients in the series reported by Bjarnason et al. 9 and the National Venous Registry 10 and fatal PE occurred in only one out of the 422 patients. Therefore, death as a result of catheter-directed thrombolysis was rare. An interesting therapeutic approach was reported by Chang et al. 36 when they used intrathrombus bolus dosing of rt-pa in 12 lower extremities of 10 patients with acute DVT. They infused rt-pa intrathrombus using the pulse-spray technique, limiting a single treatment session to 50 mg. Following the pulse-spray bolus, patients were returned to their rooms and brought back the following day for repeat phlebographic evaluation and repeat infusion if necessary. Treatment was repeated up to four times. Eleven of the 12 extremities had significant or complete lysis, and one had 50e75% lysis. Although the average total dose of rt-pa was 106 mg, bleeding complications were minor and no patient dropped their haematocrit more than 2%. This intriguing technique deserves further study to evaluate its applicability to the general population of DVT patients. The National Venous Registry 10 reported 287 patients treated in both academic and community medical centres. Sixty-six percent had acute DVT, 16% had chronic, and 19% had an acute episode superimposed upon a chronic condition. Seventy-one percent of the patients presented with iliofemoral DVT and 25% with femoropopliteal DVT. Catheter-directed thrombolysis with intrathrombus infusion of urokinase (UK) was the preferred approach. In the subgroup of patients with acute, first-time iliofemoral DVT, 65% of the patients enjoyed complete clot resolution. During follow-up, thrombus-free survival was observed in 65% at 6 months and in 60% at 12 months. There was a significant correlation (P <.001) of thrombus-free survival with the results of initial therapy. Seventy-eight percent of patients with complete clot resolution initially had patent veins at 1 year, compared to only 37% of those in whom <50% of the clot was dissolved. Interestingly, in the subgroup of patients with acute, first-time iliofemoral DVT who had initially successful thrombolysis, 96% of the veins remained patent at 1 year. In addition to sustained patency, early success directly correlated with valve function at 6 months. Sixty-two percent of patients with <50% thrombolysis had venous valvular incompetence, whereas 72% of patients who had complete lysis had normal valve function (P <.02). Patients with iliofemoral DVT treated in the National Venous Registry were compared to a contemporary cohort of patients with iliofemoral DVT treated with anticoagulation alone in the same institutions. 39 Anticoagulated patients were candidates for lytic therapy but were treated with anticoagulation alone due to physician preference. A validated qualityof-life (QOL) questionnaire was used to query patients at 16 and 22 months post treatment. Of the 98 patients evaluated, 68 were treated with catheter-directed thrombolysis and 30 treated with anticoagulation alone. Those treated with catheter-directed thrombolysis reported significantly better QOL than those treated with anticoagulation alone. 39 Quality-of-life results were directly related to the initial success of thrombolysis. Patients who had a successful lytic outcome reported a significantly better health utilities

Treatment of Acute Iliofemoral Deep Venous Thrombosis 355 index, better physical functioning, less stigma of chronic venous disease, less health distress, and fewer overall postthrombotic symptoms. Patients in whom catheter-directed thrombolysis failed had similar outcomes to patients treated with anticoagulation alone. These efficacy data combined with the observed reduction in complications with intrathrombus infusion of plasminogen activators offer a solid argument for the management of patients with iliofemoral DVT with catheter-directed thrombolysis. Subsequently, a randomized trial was performed by Elshawary et al. 40 comparing catheter-directed thrombolysis to anticoagulation alone. These authors demonstrated that catheter-directed thrombolysis offered significantly better outcomes at 6 months. Assuming patients are properly managed with anticoagulation, this six-month observation should reflect their long-term outcome. Pharmacomechanical Thrombolysis No fewer than 10 percutaneous mechanical thrombectomy techniques have been developed. There appears to be a higher incidence of embolic complications with mechanical thrombectomy. Pulse-spray pharmacomechanical thrombolysis of clotted hemodialysis grafts has demonstrated an 18% incidence of PE in patients treated with a plasminogen activator pulse-spray solution compared to a 64% incidence of PE in patients treated with a heparinised saline pulse-spray solution (P ¼.04). 41 Since the haemodialysis graft is in direct communication with the venous circulation, it appears that results in these patients should be similar to patients with proximal acute DVT. However, one would expect observations to be magnified when treating thromboses in the much larger veins. In an experimental model comparing mechanical, pharmacomechanical, and pharmacologic thrombolysis, Greenberg and associates 42 reported findings consistent with Kinney et al. 41 Greenberg et al. 42 demonstrated that pulse-spray mechanical thrombectomy was associated with the largest number and greatest size of distal emboli. Embolic particles diminished in number and size and the speed of lysis increased when a plasminogen activator (UK) was added to the pulse-spray solution. Catheter-directed thrombolysis alone was associated with the slowest rate of reperfusion but also the fewest number of distal emboli. Generally speaking, pharmacomechanical thrombectomy alone most often is inadequate. Hemolytic complications of rheolytic mechanical thrombectomy are common and occasionally can result in anemia and renal dysfunction. Several percutaneous mechanical thrombectomy devices have recently been used either alone or in conjunction with thrombolytic therapy for acute iliofemoral DVT. A number of devices deserve further mention. The Amplatz device utilizes a rotational blender-like impeller design rotating at 100,000 to 150,000 rpm to aspirate and re-circulate macerated thrombus. Removal of thrombus is reported at 75e83% in lower extremity acute DVT with 6 month patency of 77%. 43,44 Although transient proceduralrelated desaturation resolved shortly after catheter deactivation, there were no occurrences of clinically significant PE. 44 Rotational design was also adopted by the Arrow- Trerotola device, consisting of 4 helically arranged nitinol wires driven by a handheld battery-powered engine rotating the catheter head at 3000 rpm. Thrombus is macerated and re-circulated. In vitro studies have shown 99% thrombus removal by weight with no remaining thrombus within the synthetic silicon grafts. 45 When used clinically in addition to thrombolytic therapy and angioplasty with stents, technical and clinical success was reported in 100% of patients with a 16-month clinical success of 92%. 46 Concerns with valve and intimal damage, although justified, have not been reported. 46 The AngioJet device has an entirely different design, using a high-velocity saline jet (350e450 km/hr) to produce a zone of negative pressure ( 760 mmhg) around the catheter tip. Theoretical advantages of this design include less vessel trauma and the ability to aspirate thrombus particles through the exhaust port. In a study without adjunctive preprocedural thrombolytic therapy, Kasirajan et al. 47 reported that half of the patients treated had 50% of their thrombus removed. Patency was restored in 77% of those patients with 50% thrombus removal. Improved results have been observed in a similar study in which the AngioJet was used without preprocedural thrombolytics. 48 Sixty-five percent of patients had complete thrombus removal while partial thrombus removal was observed in the remaining 35%. 48 A new device recently released for segmental and controlled pharmacomechanical thrombolysis is the reengineered Trellis Ò catheter (Bacchus Vascular, Santa Clara, CA). This is a hybrid catheter which isolates the thrombosed vein segment between two occluding balloons. The plasminogen activator is infused into the thrombus between the occluding balloons. A dispersion wire is inserted into the catheter, resulting in the catheter shaft assuming a spiral configuration that, when activated, spins at 1500 rpm. After 15e 20 minutes, the liquefied and particulate thrombus is aspirated. Phlebographic evaluation of the result is

356 A. J. Comerota and D. Paolini Fig. 2. (A, B) Ascending phlebogram of a patient presenting with phlegmasia cerulea dolens of his left leg. Phlebogram shows extensive venous thrombosis of the femoral vein inclusive of the iliac vein which extends to the junction with the vena cava. Not shown is thrombus in the posterior tibial and popliteal veins. The patient was treated with segmental pharmacomechanical thrombolysis. (C) The Trellis catheter (Bacchus Vascular, Santa Clara, CA) is in place with balloons inflated. Segmental pharmacomechanical thrombolysis successfully lysed clot (D) and restored patency to the iliofemoral venous system. The posterior tibial and popliteal thrombosis was treated with ultrasound-accelerated thrombolysis using the EKOS LysUS system (EKOS Corp, Bothell, WA). (E) The EKOS catheter was inserted into the posterior tibial vein at the ankle and advanced into the popliteal vein. (FeH) Completion ascending phlebography demonstrates complete clot lysis from the posterior tibial vein to the vena cava. The patient s leg normalized and he became asymptomatic. performed before treating additional segments of thrombosed vein. The advantages of such a device are its ability to incorporate mechanical and pharmacologic therapies, even in patients with a contraindication to thrombolytic therapy, since the infusate is aspirated, and the rapidity with which treatment can be achieved. The rationales behind the design of this catheter are 1) rapidly resolving thrombus during a short course of treatment, 2) limiting or avoiding systemic thrombolysis by reducing exposure to plasminogen activators as a result of aspirating liquefied thrombus and infused plasminogen activator, and 3) preventing PE by proximal balloon occlusion. A clinical trial designed to evaluate the success and complication rate of this technique is currently underway.

Treatment of Acute Iliofemoral Deep Venous Thrombosis 357 Fig. 2. (continued) The emission of ultrasound waves from an infusion catheter delivering the plasminogen activator is an interesting new adjunct to catheter-directed thrombolysis. Reports have emerged indicating that an infusion catheter with ultrasound transducers built into the infusion end of the catheter accelerates thrombolysis. 49e52 Ultrasound enhancement of the fibrinolytic activity of tissue plasminogen activator has been observed in vitro. 53e55 This effect is the result of clot fibrin fragmentation, increasing exposed fibrin surface and thereby allowing greater tpa binding to fibrin due to its larger available surface area. 56,57 In vivo models 58 and clinical trials 59 are now underway to assess the potential value of ultrasound enhancement of thrombolysis for the management of acute DVT. A patient presenting with phlegmasia cerulea dolens two days after laparotomy (Fig. 2) illustrates the advantage of using segmental pharmacomechanical thrombolysis and ultrasound-accelerated catheterdirected thrombolysis in order to shorten treatment duration and limit exposure of the thrombolytic agent, yet maximizing the chance of a successful result. Thrombolysis is effective in restoring venous patency and has become safer with the direct intrathrombus infusion and adjunctive mechanical techniques. As technology continues to improve and as direct clot-dissolving pharmacologic agents are developed, 60,61 lytic infusion times should considerably shorten and complication rates will be significantly reduced. Recommendations for Management of Patients with Iliofemoral DVT The general treatment strategy for patients with acute iliofemoral DVT is summarized in the algorithm illustrated in Fig. 3. Few patients need to be treated as an emergency; however, all patients are immediately anticoagulated, placed in snug, long-leg, multilayered compression bandages extending from the base of the toes to the upper thigh, and treated with leg elevation. 62 Patients are encouraged to ambulate following anticoagulation. When not ambulating, patients are in bed with their legs elevated. A rapid computed tomography (CT) scan with contrast is performed of the chest, abdomen, and pelvis to evaluate for PE and other pathology, which may be associated with their aggressive degree of thrombosis (Fig. 4). Patients who are physically active are considered for a strategy of thrombus removal. Generally, for patients who are minimally active or have an expected lifespan of less than 2 years, anticoagulation with compression is recommended. Patients who have a free-floating thrombus in their vena cava usually have a vena caval filter inserted prior to catheterdirected thrombolysis. Patients considered for a strategy of thrombus removal who have a contraindication to thrombolysis are treated either with a contemporary venous thrombectomy 30 or segmental pharmacomechanical thrombolysis using the Trellis Ò catheter. Patients who have no

358 A. J. Comerota and D. Paolini Fig. 3. Algorithm illustrating the general treatment strategy for patients with iliofemoral DVT. contraindication to thrombolysis are offered catheterdirected thrombolysis, usually with the addition of pharmacomechanical techniques. Once the thrombus is removed, patients are evaluated for an underlying venous lesion, which should be corrected to provide unobstructed venous drainage into the vena cava, thereby reducing the risk of recurrent thrombosis. Subsequently, early and long-term therapeutic anticoagulation is important to avoid recurrent thrombosis. Intermittent pneumatic compression units are applied while patients are in bed during their hospitalization. Therapeutic long-term anticoagulation is important to avoid recurrent thrombosis. Since these patients have demonstrated an aggressive propensity for thrombosis compared to the majority of DVT patients, we tend to treat these patients with a longer duration of anticoagulation, at least one year with most patients treated much longer. These patients generally have a low risk of bleeding complications but face serious morbidity if rethrombosis occurs. When we consider terminating anticoagulation, we obtain a venous duplex examination to evaluate for residual venous thrombus, 63 and we check for evidence of thrombus activity by attaining D-dimer levels. 64 If the venous Fig. 4. CT of the chest, abdomen, and pelvis is part of the routine evaluation of patients with iliofemoral and vena caval thrombosis. The initial CT scan of the chest (A) demonstrates an asymptomatic pulmonary embolism (arrow). An additional cut (B) demonstrates mediastinal lymphadenopathy (top arrow) which is compressing the trachea posteriorly (bottom arrow). Further images of the abdomen demonstrated additional retroperitoneal lymphadenopathy which subsequently proved to be lymphoma. duplex shows persistent venous thrombus and/or the D-dimer level is elevated, anticoagulation is continued. If an underlying aetiology for the patient s extensive venous thrombosis cannot be identified, a thrombophilia evaluation is performed. The majority of these patients do not have an identifiable cause (e.g., metastatic or extensive cancer) for their aggressive thrombosis; therefore, the majority undergo a thrombophilia evaluation. Discussion Postthrombotic morbidity following iliofemoral DVT is certain. 12e14 The more extensive the thrombosis

Treatment of Acute Iliofemoral Deep Venous Thrombosis 359 and the more active the patient, the more severe the postthrombotic sequelae. Patients who are active and those with an anticipated survival of 2 or more years should be offered a strategy of thrombus removal. Long-term benefits of a strategy of restoring venous patency have been documented in a randomized trial of venous thrombectomy. 4e6 Likewise, randomized trials of systemic thrombolytic therapy show that when it is successful, patients have fewer postthrombotic sequelae and improved venous function compared to patients treated with anticoagulation alone or when lytic therapy is unsuccessful. 65 Experimental observations 18,19 and clinical studies have documented improved venous function 22,65 and improved QOL 39 following successful elimination of the acute venous thrombus. Currently, the key clinical question is not whether patients will benefit, but whether methods to achieve thrombus removal will be successful, and can be applied safely. Improvements in operative technique as well as catheter-based technology now document higher rates of success and complication rates which are acceptable. As clinicians become more experienced with these techniques, technology continues to improve, and new pharmacologic agents are developed that lyse thrombus more rapidly with fewer bleeding complications, the application of these principles should expand. A national randomized trial is being planned, which should definitively answer the question as to whether a strategy of thrombus removal is more effective than anticoagulation alone in reducing postthrombotic sequelae. Until that trial is completed, the available information strongly supports offering a strategy of thrombus removal for patients with acute, symptomatic iliofemoral DVT. References 1BULLER HR, AGNELLI G, HULL RD, HYERS TM, PRINS MH, RASKOB GE. Antithrombotic therapy for venous thromboembolic disease: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004 Sep;126(3 Suppl):401Se428S. 2HALLER JA, ABRAMS BL. Use of thrombectomy in the treatment of acute iliofemoral venous thrombosis in forty-five patients. Ann Surg 1963 Oct;158:561e569. 3LANSING AM, DAVIS WM. Five-year follow-up study of iliofemoral venous thrombectomy. Ann Surg 1968 Oct;168(4):620e628. 4PLATE G, EINARSSON E, OHLIN P, JENSEN R, QVARFORDT P, EKLOF B. Thrombectomy with temporary arteriovenous fistula: the treatment of choice in acute iliofemoral venous thrombosis. J Vasc Surg 1984 Nov;1(6):867e876. 5PLATE G, AKESSON H, EINARSSON E, OHLIN P, EKLOF B. Long-term results of venous thrombectomy combined with a temporary arterio-venous fistula. Eur J Vasc Surg 1990 Oct;4(5):483e489. 6 PLATE G, EKLOF B, NORGREN L, OHLIN P, DAHLSTROM JA. Venous thrombectomy for iliofemoral vein thrombosise10-year results of a prospective randomised study. Eur J Vasc Endovasc Surg 1997 Dec;14(5):367e374. 7 ALKJAERSIG N, FLETCHER AP, SHERRY S. The mechanism of clot dissolution by plasmin. J Clin Invest 1959 Jul;38(7):1086e1095. 8 BERRIDGE DC, GREGSON RH, HOPKINSON BR, MAKIN GS. Randomized trial of intra-arterial recombinant tissue plasminogen activator, intravenous recombinant tissue plasminogen activator and intra-arterial streptokinase in peripheral arterial thrombolysis. Br J Surg 1991 Aug;78(8):988e995. 9 BJARNASON H, KRUSE JR, ASINGER DA, NAZARIAN GK, DIETZ Jr CA, CALDWELL MD et al. Iliofemoral deep venous thrombosis: safety and efficacy outcome during 5 years of catheter-directed thrombolytic therapy. J Vasc Interv Radiol 1997 May;8(3):405e418. 10 MEWISSEN MW, SEABROOK GR, MEISSNER MH, CYNAMON J, LABROPOULOS N, HAUGHTON SH. Catheter-directed thrombolysis for lower extremity deep venous thrombosis: report of a national multicenter registry. Radiology 1999 Apr;211(1):39e49. 11 COMEROTA AJ, KAGAN SA. Catheter-directed thrombolysis for the treatment of acute iliofemoral deep venous thrombosis. Phlebology 2001;15:149e155. 12 O DONNELL Jr TF, BROWSE NL, BURNAND KG, THOMAS ML. The socioeconomic effects of an iliofemoral venous thrombosis. J Surg Res 1977 May;22(5):483e488. 13 AKESSON H, BRUDIN L, DAHLSTROM JA, EKLOF B, OHLIN P, PLATE G. Venous function assessed during a 5 year period after acute ilio-femoral venous thrombosis treated with anticoagulation. Eur J Vasc Surg 1990 Feb;4(1):43e48. 14 DELIS KT, BOUNTOUROGLOU D, MANSFIELD AO. Venous claudication in iliofemoral thrombosis: long-term effects on venous hemodynamics, clinical status, and quality of life. Ann Surg 2004 Jan; 239(1):118e126. 15 COMEROTA AJ, ALDRIDGE SC, COHEN G, BALL DS, PLISKIN M, WHITE JV. A strategy of aggressive regional therapy for acute iliofemoral venous thrombosis with contemporary venous thrombectomy or catheter-directed thrombolysis. J Vasc Surg 1994 Aug;20(2):244e254. 16 SHULL KC, NICOLAIDES AN, FERNANDES E FERNANDES J, MILES C, HORNER J, NEEDHAM T et al. Significance of popliteal reflux in relation to ambulatory venous pressure and ulceration. Arch Surg 1979 Nov;114(11):1304e1306. 17 JOHNSON BF, MANZO RA, BERGELIN RO, STRANDNESS Jr DE. Relationship between changes in the deep venous system and the development of the postthrombotic syndrome after an acute episode of lower limb deep vein thrombosis: a one- to six-year followup. J Vasc Surg 1995 Feb;21(2):307e312. 18 CHO JS, MARTELLI E, MOZES G, MILLER VM, GLOVICZKI P. Effects of thrombolysis and venous thrombectomy on valvular competence, thrombogenicity, venous wall morphology, and function. J Vasc Surg 1998 Nov;28(5):787e799. 19 RHODES JM, CHO JS, GLOVICZKI P, MOZES G, ROLLE R, MILLER VM. Thrombolysis for experimental deep venous thrombosis maintains valvular competence and vasoreactivity. J Vasc Surg 2000 Jun;31(6):1193e1205. 20 KILLEWICH LA, BEDFORD GR, BEACH KW, STRANDNESS Jr DE. Spontaneous lysis of deep venous thrombi: rate and outcome. J Vasc Surg 1989 Jan;9(1):89e97. 21 MARKEL A, MANZO RA, BERGELIN RO, STRANDNESS Jr DE. Valvular reflux after deep vein thrombosis: incidence and time of occurrence. J Vasc Surg 1992 Feb;15(2):377e382. 22 MEISSNER MH, MANZO RA, BERGELIN RO, MARKEL A, STRANDNESS Jr DE. Deep venous insufficiency: the relationship between lysis and subsequent reflux. JVascSurg1993 Oct;18(4):596e605. 23 JUHAN C, ALIMI Y, DI MP, HARTUNG O. Surgical venous thrombectomy. Cardiovasc Surg 1999 Oct;7(6):586e590. 24 TORNGREN S, SWEDENBORGJ. Thrombectomy and temporary arteriovenous fistula for ilio-femoral venous thrombosis. Int Angiol 1988 Jan;7(1):14e18. 25 EKLOF B, KISTNER RL. Is there a role for thrombectomy in iliofemoral venous thrombosis? Semin Vasc Surg 1996 Mar;9(1):34e45.

360 A. J. Comerota and D. Paolini 26 ENDRYS J, EKLOF B, NEGLEN P, ZYKA I, PEREGRIN J. Percutaneous balloon occlusion of surgical arteriovenous fistulae following venous thrombectomy. Cardiovasc Intervent Radiol 1989 Jul;12(4):226e229. 27 NEGLEN P, AL-HASSAN HK, ENDRYS J, NAZZAL MM, CHRISTENSON JT, EKLOF B. Iliofemoral venous thrombectomy followed by percutaneous closure of the temporary arteriovenous fistula. Surgery 1991 Sep;110(3):493e499. 28 EKLOF B. Arteriovenous fistulas as an adjunct to venous surgery. Semin Vasc Surg 2000 Mar;13(1):20e26. 29 EKLOF B, RUTHERFORD RB. Surgical thrombectomy for acute deep vein thrombosis. In: RUTHERFORD RB, ed. Vascular surgery. 6th ed. Philadelphia: Elsevier Saunders; 2005:2188e2198. 30 COMEROTA AJ, GALE SS. Technique of contemporary iliofemoral and infrainguinal venous thrombectomy. J Vasc Surg 2006 Jan; 43(1):185e191. 31 SCHWARZBACH MH, SCHUMACHER H, BOCKLER D, FURSTENBERGER S, THOMAS F, SEELOS R et al. Surgical thrombectomy followed by intraoperative endovascular reconstruction for symptomatic ilio-femoral venous thrombosis. Eur J Vasc Endovasc Surg 2005 Jan;29(1):58e66. 32 BLATTLER W, HELLER G, LARGIADER J, SAVOLAINEN H, GLOOR B, SCHMIDLI J. Combined regional thrombolysis and surgical thrombectomy for treatment of iliofemoral vein thrombosis. J Vasc Surg 2004 Oct;40(4):620e625. 33 COMEROTA AJ, GALE SS. Operative venous thrombectomy. In: BERGAN JJ, ed. The Vein Book. San Diego: Elsevier; 2006:405e416. 34 VERHAEGHE R, STOCKX L, LACROIX H, VERMYLEN J, BAERT AL. Catheter-directed lysis of iliofemoral vein thrombosis with use of rt-pa. Eur Radiol 1997;7(7):996e1001. 35 SHORTELL CK, OURIEL K. Thrombolysis in acute peripheral arterial occlusion: predictors of immediate success. Ann Vasc Surg 1994 Jan;8(1):59e65. 36 CHANG R, CANNON III RO, CHEN CC, DOPPMAN JL, SHAWKER TH, MAYO DJ et al. Daily catheter-directed single dosing of t-pa in treatment of acute deep venous thrombosis of the lower extremity. J Vasc Interv Radiol 2001 Feb;12(2):247e252. 37 CASTANEDA F, LI R, YOUNG K, SWISCHUK JL, SMOUSE B, BRADY T. Catheter-directed thrombolysis in deep venous thrombosis with use of reteplase: immediate results and complications from a pilot study. J Vasc Interv Radiol 2002 Jun;13(6):577e580. 38 SILLESEN H, JUST S, JORGENSEN M, BAEKGAARD N. Catheter-directed thrombolysis for treatment of ilio-femoral deep venous thrombosis is durable, preserves venous valve function and may prevent chronic venous insufficiency. Eur J Vasc Endovasc Surg 2005. epub. 39 COMEROTA AJ, THROM RC, MATHIAS SD, HAUGHTON S, MEWISSEN M. Catheter-directed thrombolysis for iliofemoral deep venous thrombosis improves health-related quality of life. J Vasc Surg 2000 Jul;32(1):130e137. 40 ELSHARAWY M, ELZAYAT E. Early results of thrombolysis vs anticoagulation in iliofemoral venous thrombosis. A randomised clinical trial. Eur J Vasc Endovasc Surg 2002 Sep;24(3):209e214. 41 KINNEY TB, VALJI K, ROSE SC, YEUNG DD, OGLEVIE SB, ROBERTS AC et al. Pulmonary embolism from pulse-spray pharmacomechanical thrombolysis of clotted hemodialysis grafts: urokinase versus heparinized saline. J Vasc Interv Radiol 2000 Oct;11(9):1143e1152. 42 GREENBERG RK, OURIEL K, SRIVASTAVA S, SHORTELL C, IVANCEV K, WALDMAN D et al. Mechanical versus chemical thrombolysis: an in vitro differentiation of thrombolytic mechanisms. J Vasc Interv Radiol 2000 Feb;11(2 Pt 1):199e205. 43 GANDINI R, MASPES F, SODANI G, MASALA S, ASSEGNATI G, SIMONETTI G. Percutaneous ilio-caval thrombectomy with the Amplatz device: preliminary results. Eur Radiol 1999;9(5):951e958. 44 DELOMEZ M, BEREGI JP, WILLOTEAUX S, BAUCHART JJ, JANNE DB, ASSEMAN P et al. Mechanical thrombectomy in patients with deep venous thrombosis. Cardiovasc Intervent Radiol 2001 Jan; 24(1):42e48. 45 WILDBERGER JE, HAAGE P, BOVELANDER J, PFEFFER J, WEISS C, VORWERK D et al. Percutaneous venous thrombectomy using the Arrow-Trerotola percutaneous thrombolytic device (PTD) with temporary caval filtration: in vitro investigations. Cardiovasc Intervent Radiol 2005 Mar;28(2):221e227. 46 LEE KH, HAN H, LEE KJ, YOON CS, KIM SH, WON JY et al. Mechanical thrombectomy of acute iliofemoral deep vein thrombosis with use of an Arrow-Trerotola percutaneous thrombectomy device. J Vasc Interv Radiol 2006 Mar;17(3):487e495. 47 KASIRAJAN K, GRAY B, OURIEL K. Percutaneous AngioJet thrombectomy in the management of extensive deep venous thrombosis. J Vasc Interv Radiol 2001 Feb;12(2):179e185. 48 BUSH RL, LIN PH, BATES JT, MUREEBE L, ZHOU W, LUMSDEN AB. Pharmacomechanical thrombectomy for treatment of symptomatic lower extremity deep venous thrombosis: safety and feasibility study. J Vasc Surg 2004 Nov;40(5):965e970. 49 STEFFEN W, FISHBEIN MC, LUO H, LEE DY, NITA H, CUMBERLAND DC et al. High intensity, low frequency catheter-delivered ultrasound dissolution of occlusive coronary artery thrombi: an in vitro and in vivo study. J Am Coll Cardiol 1994 Nov 15;24(6):1571e1579. 50 ROSENSCHEIN U, GAUL G, ERBEL R, AMANN F, VELASGUEZ D, STOERGER H et al. Percutaneous transluminal therapy of occluded saphenous vein grafts: can the challenge be met with ultrasound thrombolysis? Circulation 1999 Jan 5;99(1):26e29. 51 TACHIBANA K, TACHIBANA S. Ultrasound energy for enhancement of fibrinolysis and drug delievery: special emphasis on the use of a transducer-tipped ultrasound system. In: SIEGEL RJ, ed. Ultrasound angioplasty. Boston: Kluwer; 1996:121e133. 52 TACHIBANA K, TACHIBANA S. Prototype therapeutic ultrasound emitting catheter for accelerating thrombolysis. J Ultrasound Med 1997 Aug;16(8):529e535. 53 TRUBESTEIN G, ENGEL C, ETZEL F, SOBBE A, CREMER H, STUMPFF U. Thrombolysis by ultrasound. Clin Sci Mol Med Suppl 1976 Dec; 3:697se698s. 54 ARIANI M, FISHBEIN MC, CHAE JS, SADEGHI H, MICHAEL AD, DUBIN SB et al. Dissolution of peripheral arterial thrombi by ultrasound. Circulation 1991 Oct;84(4):1680e1688. 55 ROSENSCHEIN U, BERNSTEIN JJ, DISEGNI E, KAPLINSKY E, BERNHEIM J, ROZENZSAJN LA. Experimental ultrasonic angioplasty: disruption of atherosclerotic plaques and thrombi in vitro and arterial recanalization in vivo. J Am Coll Cardiol 1990 Mar 1;15(3):711e717. 56 LAUER CG, BURGE R, TANG DB, BASS BG, GOMEZ ER, ALVING BM. Effect of ultrasound on tissue-type plasminogen activatorinduced thrombolysis. Circulation 1992 Oct;86(4):1257e1264. 57 HONG AS, CHAE JS, DUBIN SB, LEE S, FISHBEIN MC, SIEGEL RJ. Ultrasonic clot disruption: an in vitro study. Am Heart J 1990 Aug; 120(2):418e422. 58 DROBINSKI G, BRISSET D, PHILIPPE F, KREMER D, LAURIAN C, MONTALESCOT G et al. Effects of ultrasound energy on total peripheral artery occlusions: initial angiographic and angioscopic results. J Interv Cardiol 1993 Jun;6(2):157e163. 59 ATAR S, LUO H, NAGAI T, SIEGEL RJ. Ultrasonic thrombolysis: catheter-delivered and transcutaneous applications. Eur J Ultrasound 1999 Mar;9(1):39e54. 60 OURIEL K, CYNAMON J, WEAVER FA, DARDIK H, AKERS D, BLEBEA J et al. A phase I trial of alfimeprase for peripheral arterial thrombolysis. J Vasc Interv Radiol 2005 Aug;16(8):1075e1083. 61 DEITCHER SR, TOOMBS CF. Non-clinical and clinical characterization of a novel acting thrombolytic: alfimeprase. Pathophysiol Haemost Thromb 2005;34(4e5):215e220. 62 PARTSCH H. Immediate ambulation and leg compression in the treatment of deep vein thrombosis. Dis Mon 2005 Feb;51(2e3): 135e140. 63 PRANDONI P. Risk factors of recurrent venous thromboembolism: the role of residual vein thrombosis. Pathophysiol Haemost Thromb 2003 Sep 20;33(5e6):351e353. 64 WELLS PS, ANDERSON DR, RODGER M, FORGIE M, KEARON C, DREYER J et al. Evaluation of D-dimer in the diagnosis of suspected deep-vein thrombosis. N Engl J Med 2003 Sep 25; 349(13):1227e1235. 65 COMEROTA AJ, ALDRIDGE SE. Thrombolytic therapy for acute deep vein thrombosis. Semin Vasc Surg 1992;5:76e84. Accepted 7 November 2006 Available online 11 December 2006