Peripheral Vascular Disease

Transcription

1 Peripheral Vascular Disease Safety and Effectiveness of Stent Placement for Iliofemoral Venous Outflow Obstruction Systematic Review and Meta-Analysis Mahmood K. Razavi, MD; Michael R. Jaff, DO; Larry E. Miller, PhD Downloaded from by guest on July 1, 2018 Background Endovenous recanalization of iliofemoral stenosis or occlusion with angioplasty and stent placement has been increasingly used to maintain long-term venous patency in patients with iliofemoral venous outflow obstruction. The purpose of this systematic review and meta-analysis was to determine safety and effectiveness of venous stent placement in patients with iliofemoral venous outflow obstruction. Methods and Results We searched MEDLINE and EMBASE for studies evaluating safety or effectiveness of stent placement in patients with iliofemoral venous outflow obstruction. Data were extracted by disease pathogenesis: nonthrombotic, acute thrombotic, or chronic post-thrombotic. Main outcomes included technical success, periprocedural complications, symptom relief at final follow-up, and primary/secondary patency through 5 years. A total of 37 studies reporting 45 treatment effects (nonthrombotic, 8; acute thrombotic, 19; and chronic post-thrombotic, 18) from 2869 patients (nonthrombotic, 1122; acute thrombotic, 629; and chronic post-thrombotic, 1118) were included. Technical success rates were comparable among groups, ranging from 94% to 96%. Complication rates ranged from 0.3% to 1.1% among groups for major bleeding, from 0.2% to 0.9% for pulmonary embolism, from 0.1% to 0.7% for periprocedural mortality, and from 1.0% to 6.8% for early thrombosis. Patient symptom relief data were reported inconsistently. At 1 year, primary and secondary patency were 96% and 99% for nonthrombotic, 87% and 89% for acute thrombotic, and 79% and 94% for chronic post-thrombotic. Conclusions Stent placement for iliofemoral venous outflow obstruction results in high technical success and acceptable complication rates regardless of cause of obstruction. (Circ Cardiovasc Interv. 2015;8:e DOI: / CIRCINTERVENTIONS ) Key Words: angioplasty meta-analysis pulmonary embolism stents venous thrombosis Iliofemoral venous obstruction is a common condition that has been implicated as a causal factor in patients with both post-thrombotic syndrome and nonthrombotic chronic venous insufficiency. 1 4 Venous obstruction can occur as a result of different causes, including extrinsic compression because of malignancy or anatomic variants (eg, May Thurner syndrome) and acute or chronic deep venous thrombosis (DVT). Patient symptoms are variable and largely dependent on the cause, extent of venous obstruction, and disease duration. Patients with acute DVT often present with sudden onset of lower extremity pain and swelling, whereas those with postthrombotic or long-standing compressive syndromes commonly report chronic pain, edema, and, in advanced cases, ulceration. The initial treatment for symptomatic acute DVT involves anticoagulation; however, between 23% and 60% of such patients experience post-thrombotic syndrome despite optimal anticoagulant therapy. 5 7 Chronic post-thrombotic (CPT) syndrome remains a therapeutic challenge, but a strategy of clot removal and restoration of venous outflow has shown promise in reducing its risk. 8 Endovenous treatment of iliofemoral stenosis or occlusion with angioplasty and stent placement has emerged as the procedure of choice to establish and maintain venous outflow. There is considerable variability among studies on the acute and long-term outcomes of iliofemoral venous stent placement, which complicates accurate assessment of these outcomes across different disease pathogenesis. Therefore, we performed a systematic review and meta-analysis on the safety and effectiveness of venous stent placement in patients with nonthrombotic, acute thrombotic (AT), and CPT disease pathogenesis. Given the recent development of stents specifically manufactured for placement in the venous system, a secondary goal of this study was to propose performance goals (PGs) that may be used as comparators in single-arm clinical trials of iliofemoral venous stents. Received May 13, 2015; accepted September 2, From the Heart and Vascular Center, St. Joseph Hospital, Orange, CA (M.K.R.); Fireman Vascular Center, Massachusetts General Hospital, Boston (M.R.J.); and Miller Scientific Consulting, Inc, Asheville, NC (L.E.M.). The Data Supplement is available at Correspondence to Mahmood K. Razavi, MD, Department of Clinical Trials, Heart and Vascular Center St. Joseph Hospital, Orange, CA mrazavi@pacbell.net 2015 American Heart Association, Inc. Circ Cardiovasc Interv is available at DOI: /CIRCINTERVENTIONS

2 2 Razavi et al Meta-Analysis of Venous Stents WHAT IS KNOWN With the emerging evidence in favor of catheterbased therapies for acute and chronic iliofemoral venous thrombosis, applications of venous stents are expected to grow. There are no robust clinical studies documenting long-term outcomes of stent placement for iliofemoral venous obstructions. WHAT THE STUDY ADDS This is the first meta-analysis of stent placement for the treatment of iliofemoral venous outflow obstruction. Unlike many studies of iliofemoral venous stent placement, short- and long-term patient outcomes are reported herein according to cause, ie, iliac vein compression syndrome, acute deep vein thrombosis, and chronic post-thrombotic occlusion. We propose performance goals that may be used as comparators in single-arm clinical trials of iliofemoral venous stents. Methods Literature Search This study was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses. 9 We searched MEDLINE and EMBASE for randomized or nonrandomized controlled studies and prospective or retrospective case series of stent placement for treatment of iliofemoral venous outflow obstruction by using a combination of relevant keywords. The details of the MEDLINE search strategy are listed in Table I in the Data Supplement. The syntax for EMBASE was similar, but adapted as necessary. In addition, manual searches were conducted using the Directory of Open Access Journals, Google Scholar, and the reference lists of included papers and relevant meta-analyses. No date restrictions were applied to the searches. The final search was conducted March 15, Study Selection Two researchers independently selected studies for inclusion in the review. Disagreements were resolved by consensus. Titles and abstracts were initially screened to exclude review articles, commentaries, letters, case reports, and obviously irrelevant studies. Full texts of the remaining articles were retrieved and reviewed. Main inclusion criteria included studies with (1) primary diagnosis of obstructive lesions of the common femoral vein, external iliac vein, and the common iliac vein, (2) stent placement in the common femoral vein, external iliac vein, common iliac vein, and the inferior vena cava, (3) outcomes reported by acute/subacute (<30 days) or chronic ( 30 days; AT and CPT groups only), (4) sample size 10, and (5) at least 1 extractable safety or effectiveness outcome. Studies were excluded if they were published in non-english language journals; data were available only from abstracts, conference proceedings, Websites, or personal communication; most treated patients were diagnosed with malignant tumor or inferior vena cava lesion; or most patients received bilateral stents. When multiple studies included overlapping series of patients (ie, kin relationships), only the study with the largest sample size was included. Data Extraction An initial database was developed, pilot tested, and refined to maintain consistency with outcomes reported in the literature. Data were extracted from eligible peer-reviewed articles by one author and verified by another author. Data extraction discrepancies between the 2 researchers were resolved by consensus. The types of data recorded in the standardized data extraction forms included general manuscript information, patient characteristics, study characteristics, procedural details, perioperative complications, patient symptom resolution, and primary/ secondary patency. For studies where stent placement was attempted in a subset of patients (eg, >50% stenosis after angioplasty), only data specifically pertaining to patients receiving a stent were extracted. Definitions and Outcomes Main outcomes of this study included technical success, periprocedural complications (major bleeding, pulmonary embolism, death, and early thrombosis), patient symptom resolution (ie, pain, edema, and ulcer) at final follow-up, and primary and secondary patency through 5 years. When main outcomes were not explicitly reported, we used the following definitions to attempt extraction of relevant data. Technical success was defined as successful recanalization and stent deployment restoring patency to the target vessel with no major procedural complications. Major bleeding was defined as procedural/ periprocedural access site bleeding requiring transfusion or intervention within 30 days of stent placement. Thirty-day all-cause mortality and pulmonary embolism regardless of severity were recorded. Early thrombosis was defined as restenosis or occlusion of the target lesion within 30 days of the procedure. Patient symptom resolution was determined by the proportion of patients with complete resolution of pain, edema, and ulcer, respectively, at the final follow-up. Partial symptom improvement was considered a failure for analysis purposes. Patency was defined as uninterrupted venous patency (<50% diameter stenosis) without (primary) or with (secondary) a procedure or intervention directly performed on the target lesion. Patency data were extracted from text, life tables, or survival curves at annual intervals through 5 years. Many additional outcomes (eg, stented length, primary assisted patency, change in Clinical-Etiology- Anatomy-Pathophysiology class, change in Villalta scores, and change in quality of life) were initially considered for inclusion, but were discarded after pilot testing because of lack of available data. Data Analysis For patient-, study-, and procedure-related data, continuous variables are reported as mean or median, and categorical variables are reported as counts and percentages. Denominators were adjusted when appropriate to include the number of patients, limbs, or procedures. Kaplan Meier survival estimates were converted to proportions using the number enrolled, number at risk, follow-up distribution when number at risk was not reported, and the survival estimate at the respective time point. Data were extracted by disease pathogenesis: nonthrombotic, AT, or CPT. For each outcome, the pooled estimate and 95% confidence interval (CI) were calculated using a random-effects model given the a priori assumption that treatment effects among studies were heterogeneous. None of the analyses were adjusted for covariates. We used the I 2 statistic to estimate heterogeneity of treatment effects across studies with values of 25%, 50%, and 75% representing low, moderate, and high inconsistency, respectively. 10 Publication bias was visually assessed with funnel plots (not shown) and quantitatively assessed using the Egger regression test. 11 Subgroup analyses were undertaken to explore potential sources of heterogeneity in 1-year primary patency outcomes. P values were 2 sided with a significance level <0.05. All analyses were performed using Comprehensive Meta-analysis (version 2.2; Biostat, Englewood, NJ).

3 3 Razavi et al Meta-Analysis of Venous Stents Results Study Selection After screening 793 studies for eligibility, 37 studies reporting 45 treatment effects (nonthrombotic, AT, 13,18,20 36 and CPT 12,14,17,30,33,34,37 48 ) from 2869 unique patients (nonthrombotic, 1122; AT, 629; and CPT, 1118) were included in the meta-analysis. The most common reasons for study exclusion were lack of relevant stent placement outcomes (19 studies), kin relationships (18 studies), sample size <10 (16 studies), combined outcome reporting in nonthrombotic and thrombotic patients (8 studies), and combined outcome reporting in AT and CPT patients (6 studies). Eight additional studies were excluded because of irrelevant topics, predominance of malignant lesions, and treatment of only inferior vena cava lesions. A flow diagram of study identification and selection is shown in Figure 1. Patient Characteristics Baseline patient characteristics among the nonthrombotic, AT, and CPT groups are presented in Table 1. In each group, most patients were women and presented with left limb symptoms. Most patients with thrombotic disease were also diagnosed with iliac vein compression syndrome. Patient characteristics were incompletely described in most studies and hence the denominators for each outcome were variable. Study Characteristics Most studies included in this meta-analysis were retrospective (nonthrombotic, 75%; AT, 84%; and CPT, 83%) and conducted at a single institution (nonthrombotic, 88%; AT, 79%; CPT, 89%). Median sample size in each included study was 38 (min max, ) for nonthrombotic, 26 (min max, ) for AT, and 33 (min max, ) for CPT. Median follow-up duration was 19 (min max, 10 50) months for nonthrombotic, 17 (min max, 6 80) months for AT, and 15 (min max, 6 46) months for CPT (Table 2). Procedural Details In general, AT was treated with catheter-directed thrombolysis (with or without thrombectomy) and followed by PTA and stent placement. Nonthrombotic and CPT were typically Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-analyses study flow diagram. AT indicates acute thrombotic; CPT, chronic post-thrombotic; and NT, nonthrombotic. treated with PTA and stent placement although adjunctive therapies were applied in some cases. Wallstents (Boston Scientific, Marlborough, MA) were used in 78% of included studies. Only 1 study (3%) used a stent specifically manufactured for venous applications in 3 (0.1%) limbs. 39 In general, more stents were placed in patients with CPT lesions than in patients with nonthrombotic and AT lesions (Table 3). Anticoagulation regimens involved an initial bolus of 4000 to 5000 IU intravenous heparin at the start of the procedure and as needed during the procedure to achieve a target activated clotting time of 280 to 300 seconds. After the procedure, low molecular weight heparin was administered until discharge. Thereafter, anticoagulation (most commonly warfarin) for 2 to 6 months with a target international normalized ratio of 2.0 to 3.0 was typical. In patients at high risk for thrombosis, anticoagulation regimens were extended to 6 to 12 months. Lifelong antiplatelet medication was routinely prescribed in some studies. Technical Success Technical success rates were comparable among groups, ranging from 94% in AT and CPT to 96% in nonthrombotic patients (Table 4). Publication bias was evident for technical success outcomes in AT patients. Perioperative Complications Major periprocedural complications were rare in each group (Table 4). Complication rates ranged from 0.3% to 1.1% among groups for major bleeding, from 0.2% to 0.9% for pulmonary embolism, and from 0.1% to 0.7% for periprocedural mortality. Descriptions of major bleeding included wound hematoma evacuation (2), infected hematoma (1), common Table 1. Patient Characteristics Variable Nonthrombotic Acute Thrombotic Chronic Post- Thrombotic No. of studies No. of patients Age, y* 41 (39 53) 54 (32 70) 43 (37 58) Female sex 58% (324/556) 68% (233/342) 58% (280/480) IVCS 100% (1109/1109) 81% (335/412) 65% (112/173) Left limb 96% (547/571) 89% (396/444) 83% (292/351) Coagulopathy 13% (29/220) 18% (25/140) 26% (212/812) CEAP class C0 C2 14% (39/273) 2% (8/396) C3 33% (91/273) 88% (45/51) 37% (147/396) C4 11% (30/273) 12% (6/51) 26% (102/396) C5 13% (36/273) 7% (29/396) C6 28% (77/273) 28% (110/396) Deep reflux 57% (147/260) 76% (144/190) Edema 61% (352/576) 90% (101/112) 62% (327/530) Pain 50% (130/260) 71% (74/104) 68% (283/414) Ulcer 27% (156/569) 25% (135/545) CEAP indicates Clinical-Etiology-Anatomy-Pathophysiology; and IVCS, iliac vein compression syndrome. *Values are median (min max).

4 4 Razavi et al Meta-Analysis of Venous Stents Table 2. Study Characteristics Study Country Study Design No. of Centers Nonthrombotic Stent Placement Attempted, n* Stent Placement Attempted, %* Treatment Period Follow-Up Duration, mo George et al 12 India R Hager et al 13 USA R Liu et al 14 China P Lou et al 15 China R Meng et al 16 China R Neglen et al 17 USA R O Sullivan et al 18 USA R Ye et al 19 China R Acute thrombotic Bjarnason et al 20 USA R Hager et al 13 USA R Holper et al 21 Germany R Husmann et al 22 Switzerland R Jackson et al 23 USA R Jeon et al 24 Korea R Kolbel et al 25 Sweden R Kwak et al 26 Korea R Matsuda et al 27 Japan R Meng et al 28 China P Mewissen et al 29 USA P Oguzkurt et al 30 Turkey R O Sullivan et al 18 USA R Park et al 31 Korea R Patel et al 32 USA R Stanley et al 33 USA R Titus et al 34 USA R Xue et al 35 China R Zhu et al 36 China P Chronic post-thrombotic Alerany et al 37 Spain R Blattler and Blattler 38 Switzerland R de Wolf et al 39 The Netherlands R Delis et al 40 USA R George et al 12 India R Kurklinsky et al 41 USA R Liu et al 14 China P Nayak et al 42 USA R Neglen et al 17 USA R Oguzkurt et al 30 Turkey R Rosales et al 43 Norway R Sang et al 44 China R Sarici et al 45 Turkey P Stanley et al 33 USA R Titus et al 34 USA R Vogel et al 46 USA R Wahlgren et al 47 Sweden R Ye et al 48 China R P indicates prospective; and R, retrospective. *Includes technical failures. Studies with values of 100 imply that stents were placed in all patients. Values are mean or median. Limbs. Data were not reported and are an estimate.

5 5 Razavi et al Meta-Analysis of Venous Stents Table 3. Procedural Details Study Concomitant Lesion Procedures Stent Trade Name No. of Stents per Patient* Caudal, Cephalad Stenting Boundaries Nonthrombotic George et al 12 PTA Niti, E-Luminexx 2 Common femoral, IVC Hager et al 13 PTA Wallstent, Protégé, balloon 1 Common femoral, IVC expandable stent Liu et al 14 PTA Wallstent 1 Common iliac, IVC Lou et al 15 PTA Luminexx 1 Common iliac, IVC Meng et al 16 PTA 1 Common iliac, IVC Neglen et al 17 PTA Wallstent, nitinol stent 2 Common femoral, IVC O Sullivan et al 18 PTA, CDT in 60% Wallstent, Palmaz stent 2 Common femoral, IVC Ye et al 19 PTA Wallstent 1 Common iliac, IVC Acute thrombotic Bjarnason et al 20 Thrombolysis, thrombectomy, PTA Wallstent, Gianturco 2 Common femoral, IVC Hager et al 13 Thrombolysis, PTA Wallstent, Protégé, balloon 1 Common femoral, IVC expandable stent Holper et al 21 Thrombolysis, thrombectomy Wallstent, Palmaz XXL, Symphony, 1 Common iliac S.M.A.R.T., Corinthian Husmann et al 22 Thrombolysis, thrombectomy, PTA Wallstent 1 Common iliac Jackson et al 23 Thrombolysis, thrombectomy, PTA Wallstent, nitinol stent 2 Popliteal, common iliac Jeon et al 24 Thrombolysis, thrombectomy, PTA Wallstent, Niti-S, Zilver 1 Common iliac Kolbel et al 25 Thrombolysis, PTA Wallstent, Memotherm-FLEXX, Jomed 2 Common iliac, IVC Kwak et al 26 Thrombolysis, thrombectomy Wallstent 1 Common iliac (varied), PTA (varied) Matsuda et al 27 Thrombolysis, PTA, thrombectomy Wallstent, S.M.A.R.T., Luminexx 1 Common iliac (varied) Meng et al 28 Thrombolysis, PTA Luminexx, S.M.A.R.T. 1 Common iliac Mewissen et al 29 Thrombolysis, PTA Wallstent Common femoral, IVC Oguzkurt et al 30 Thrombolysis, thrombectomy, PTA Wallstent, Protege 2 Common iliac O Sullivan et al 18 Thrombolysis, PTA Wallstent, Palmaz stent 2 Common femoral, IVC Park et al 31 Thrombolysis, PTA Wallstent, S.M.A.R.T., Zilver 1 Common iliac, IVC Patel et al 32 Thrombolysis, PTA Wallstent 1 Common iliac Stanley et al 33 Thrombolysis, PTA Wallstent Common iliac Titus et al 34 PTA Nitinol stent, stainless steel stent 2 Common femoral, common iliac Xue et al 35 Thrombolysis, PTA S.M.A.R.T., Luminexx, Maris, GPS 1 Common iliac Zhu et al 36 Thrombolysis, thrombectomy, PTA Luminexx 1 Common femoral, common iliac Chronic post-thrombotic Alerany et al 37 PTA Wallstent, Zilver Vena 3 Common femoral, IVC Blattler and Blattler 38 PTA Wallstent 1 Common iliac, IVC de Wolf et al 39 PTA, endophlebectomy in 11% Wallstent, Sinus-XL, Zilver Vena, 3 Common iliac, IVC Andrastent XL Delis et al 40 PTA Wallstent, S.M.A.R.T., Gianturco Posterior tibial vein, IVC George et al 12 PTA Niti, E-Luminexx 2 Common femoral, IVC Kurklinsky et al 41 PTA Wallstent, Protégé, S.M.A.R.T., Luminexx 3 Common femoral, IVC Liu et al 14 PTA Wallstent 1 Common iliac, IVC Nayak et al 42 PTA Wallstent, S.M.A.R.T. 1 Common femoral, IVC Neglen et al 17 PTA Wallstent, nitinol stent 2 Common femoral, IVC Oguzkurt et al 30 PTA Wallstent, Protege 2 Common iliac Rosales et al 43 PTA Wallstent Common femoral, IVC Sang et al 44 PTA Wallstent, S.M.A.R.T., Luminexx 2 Common femoral, IVC Sarici et al 45 PTA Nitinol stent 2 Common femoral, IVC Stanley et al 33 PTA, PMT in 79%, CDT in 64% Wallstent Common iliac Titus et al 34 PTA Nitinol stent, stainless steel stent 2 Common femoral, common iliac Vogel et al 46 PTA Wallstent Common iliac, IVC Wahlgren et al 47 PTA, femoral endovenectomy in Wallstent 3 Common femoral, IVC 26% Ye et al 48 PTA Wallstent, EverFlex, Lifestent 2 Common femoral, IVC CDT indicates catheter-directed thrombolysis; IVC, inferior vena cava; PMT, pharmacomechanical thrombectomy; and PTA, percutaneous transluminal angioplasty. *Mean or median.

6 6 Razavi et al Meta-Analysis of Venous Stents femoral artery puncture during sheath insertion (1), and access site bleed requiring transfusion (1). Early thrombosis was identified in 6.5% and 6.8% of AT and CPT patients, respectively, but in only 1.0% of nonthrombotic patients. Publication bias was evident for most periprocedural complication outcomes. Symptom Relief The percentage of patients reporting complete relief of venous symptoms after stent placement was not reported in most studies. Of the nonthrombotic and CPT studies reporting these data, complete symptom relief at the final follow-up visit was reported in 69% to 82% of patients for pain, 64% to 68% of patients for edema, and 71% to 81% of patients for ulcer healing. Symptom relief data were rarely reported in patients with acute DVT. Significant heterogeneity among studies was identified for relief of pain and edema (Table 4). Patency Primary patency was typically evaluated with duplex ultrasonography; however, a formal definition of primary patency was rarely provided, and independent assessment of image quality and findings was not provided (Table II in the Data Supplement). Primary patency at 1 year was 96% (95% CI, 93% 98%) for nonthrombotic, 87% (95% CI, 80% 92%) for AT, and 79% (95% CI, 76% 83%) for CPT. Significant heterogeneity and publication bias were observed in the AT group, but not in the nonthrombotic or CPT. Secondary patency at 1 Table 4. Main Outcomes by Disease Pathogenesis Pathology No. of Studies Event Rate Heterogeneity Among Studies Publication Bias Point Estimate, % 95% CI I 2, % P Value Eggar P Value Technical success Nonthrombotic Acute thrombotic Chronic post-thrombotic Major bleeding Nonthrombotic Acute thrombotic Chronic post-thrombotic <0.001 Pulmonary embolism Nonthrombotic Acute thrombotic <0.001 Chronic post-thrombotic <0.001 Periprocedural mortality Nonthrombotic Acute thrombotic <0.001 Chronic post-thrombotic Early thrombosis Nonthrombotic Acute thrombotic Chronic post-thrombotic < Complete pain relief Nonthrombotic Acute thrombotic 1 100* * Chronic post-thrombotic Complete edema relief Nonthrombotic < Acute thrombotic 1 100* * Chronic post-thrombotic < Complete ulcer healing Nonthrombotic Acute thrombotic Chronic post-thrombotic CI indicates confidence interval. *Meta-analysis not conducted because of single study contributing data. Data not calculable.

7 7 Razavi et al Meta-Analysis of Venous Stents Figure 2. Primary patency through 5 y. Patency was estimated at annual intervals through 5 y. Different studies may contribute data to each interval because of variation in follow-up duration among studies. In-figure values represent the denominator used in the analysis at each annual interval in the order reflected in the legend (top to bottom). AT indicates acute thrombotic; CPT, chronic post-thrombotic; and NT, nonthrombotic. year was 99% (95% CI, 88% 100%) for nonthrombotic, 89% (95% CI, 76% 95%) for AT, and 94% (95% CI, 90% 96%) for CPT. Primary and secondary patency remained higher in nonthrombotic patients versus AT and CPT patients through 5 years (Figures 2 and 3). Subgroup Analyses We performed predefined subgroup analyses to identify the influence of study- and patient-related characteristics on primary patency at 1 year. Within each patient group, we identified no variable that significantly influenced primary patency rates (Table 5). However, many studies inadequately reported outcomes and, consequently, comparisons often included too few studies to draw any meaningful conclusions. Discussion To our knowledge, this is the first meta-analysis of stent placement for the treatment of iliofemoral venous outflow obstruction. Overall, results of this analysis indicate that stent placement has a high technical success rate and is effective in restoring and maintaining patency in patients with iliofemoral venous outflow obstruction because of iliac vein compression syndrome, acute DVT, or CPT occlusion. The long-term patency of stents seems to be somewhat lower in patients with thrombotic disease than in those with nonthrombotic iliac vein compression although formal statistical testing for group differences was not conducted because of the limited number of available studies. Although the degree of symptom relief was inconsistently reported, majority of patients experienced complete resolution of symptoms after the procedures. In terms of safety profile, venous stent placement is associated with a low complication rate of <1% for bleeding, PE, and mortality. It should be noted that the reported 30-day bleeding rates in the AT group were likely because of CDT and subsequent anticoagulation regimens rather than the stent placement per se as evidenced by the apparent lower bleed rate among nonthrombotic groups than among AT group (0.3% versus 1.1%). Early thrombosis seems to be related to the cause of the obstructive lesion with the highest thrombosis rate observed in the AT and CPT groups ( 6.5% and 6.8%, respectively). Factors that may impact postprocedural stent thrombosis rate include the quality of inflow and outflow through the stented segment(s), rigor of anticoagulation regimen, lesion length and location, stent diameter and length, and patient characteristics. Unfortunately the quality of the reported data was insufficient to determine the impact of these variables on early stent thrombosis. An important aspect of the current meta-analysis is that stent placement outcomes were reported separately based on distinct venous disease pathogenesis. Eijgenraam et al 49 performed a systematic review of the effect of antithrombotic therapy with venous stenting after DVT. However, the authors combined data from AT and CPT studies, included data from studies where stents were not placed in all patients, did not include nonthrombotic patients, and did not use meta-analytic techniques to synthesize main outcomes. The practice of reporting outcomes in mixed populations (eg, combining acute and chronic DVT, nonthrombotic and thrombotic causes, or stent and no stent) was frequently identified during the conduct of the current systematic review and served as a common reason for study exclusion. In accordance with reporting standards set forth by the Society of Interventional Radiology, 50 we recommend that future studies explicitly detail patient outcomes, at a minimum, according to the cause and treatment administered. Furthermore, data elements such as stent location, diameter, length, and quality of venous inflow and outflow should also be systematically recorded. We made no attempt to determine the comparative safety and effectiveness of stent placement versus no stent placement because of a paucity of available literature. In the National Venous Registry, which was a prospective multicenter registry of patients with DVT treated by CDT, a higher venous patency was observed in patients who had stents placed than in those who did not. 29 Similar results were reported by Meng et al 28 who conducted a randomized controlled trial evaluating stent placement in patients with acute DVT. Patients who underwent stent placement resulted in higher 1-year primary patency (86% versus 55%) and superior improvements in clinical (C) class according to the Clinical-Etiology- Anatomy-Pathophysiology (CEAP) classification, Venous Clinical Severity Score, and Chronic Venous Insufficiency Questionnaire scores than those with no stent. On balance, the Figure 3. Secondary patency through 5 y. Patency was estimated at annual intervals through 5 y. Different studies may contribute data to each interval because of variation in follow-up duration among studies. In-figure values represent the denominator used in the analysis at each annual interval in the order reflected in the legend (top to bottom). AT indicates acute thrombotic; CPT, chronic post-thrombotic; and NT, nonthrombotic.

8 8 Razavi et al Meta-Analysis of Venous Stents Table 5. Subgroup Analysis of Study-Related Factors on Primary Patency at 1 Year Comparison* Nonthrombotic Acute Thrombotic Chronic Post-Thrombotic No. of Studies Point Estimate (95% CI) No. of Studies Point Estimate (95% CI) No. of Studies Point Estimate (95% CI) Overall 5 96 (93 98) (80 92) (76 83) No. of centers Single center 4 96 (94 98) (79 93) (76 82) Multicenter 1 90 (66 97) 4 86 (72 94) 2 81 (67 91) Sample size > (94 98) 7 84 (73 92) 5 81 (76 84) (73 97) 9 90 (83 94) 7 77 (70 83) Study location USA 3 95 (90 97) 7 83 (69 92) 4 80 (76 84) OUS 2 98 (93 99) 9 90 (85 93) 8 78 (73 83) Study design Prospective 1 95 (72 99) 3 85 (66 94) 1 83 (37 98) Retrospective 4 96 (92 98) (80 93) (76 82) Median treatment period > (78 97) 7 90 (85 93) 7 81 (76 85) (94 98) 9 83 (72 91) 5 78 (73 83) Caudal stent boundary CFV Yes 3 95 (90 97) 6 83 (67 92) 8 79 (75 82) No 2 98 (93 99) (85 93) 4 82 (70 90) IVC Yes 5 96 (93 98) 5 81 (64 91) 9 79 (76 82) No (85 93) 3 81 (69 89) Patient age, y, > (56 99) 8 91 (86 94) 1 77 (68 85) (66 97) 2 85 (59 96) 6 79 (72 85) Female sex proportion, %, > (66 97) 5 90 (81 95) 3 75 (59 86) (56 99) 5 90 (85 94) 4 80 (74 85) CFV indicates common femoral vein; CI, confidence interval; IVC, inferior vena cava; and OUS, outside United States. *Within each patient group, no comparison achieved a P<0.05. Heterogeneity among studies: I 2 =2%, P=0.39; publication bias: Eggar P=0.37. Heterogeneity among studies: I 2 =58%, P=0.002; publication bias: Eggar P= Heterogeneity among studies: I 2 =0%, P=0.79; publication bias: Eggar P=0.82. Values for comparisons represent the median for all studies. Data not reported in all studies. study design was questionable considering the unexplained imbalance in sample size between groups (n=45 versus n=29) and the paucity of reported data. No other prospective randomized or nonrandomized trials of iliofemoral venous stent placement have been conducted. Primary patency at 1 year is a standard effectiveness end point in clinical trials of endovascular treatment of femoropopliteal arterial lesions and in ongoing studies of iliofemoral venous stents. Given the lack of a commonly accepted comparator for venous stents and because most available studies are small, retrospective, single-site evaluations, the pooled data from this meta-analysis represent the highest quality data available to use for PG development in clinical trials of iliofemoral venous stent placement. The typical approach to developing a PG is to identify the lower or upper 95% confidence limit of the estimate (depending on outcome) and then add or subtract a margin from that value. For safety outcomes, we recommend taking a more conservative approach for identifying a PG for venous stent trials by using the appropriate 95% confidence limit with no additional margin applied. However, because most studies were retrospective with no formal definitions for patency, primary and secondary patency estimates are likely overestimated in comparison with prospective trials with imaging core laboratory patency assessments using objective criteria. Therefore, in the case of primary/secondary patency, we propose subtracting a margin of 5% to 10% from the lower 95% confidence limits identified in this meta-analysis to establish a PG. Specifically, the lower 95% confidence limit of primary patency outcomes at 1 year in a clinical trial of venous stents must exceed the PG to

9 9 Razavi et al Meta-Analysis of Venous Stents demonstrate reasonable evidence of effectiveness. For primary patency at 1 year, the corresponding PG ranges, depending on choice of margin, would be 83% to 88% for nonthrombotic, 70% to 75% for AT, and 66% to 71% for CPT groups. It is important to discuss several inherent limitations considering that the estimates were primarily derived from retrospective case series. In the absence of available patient-level data, development of a PG is constrained by the quantity and quality of peer-reviewed literature. Unreported confounding factors such as patient and procedural characteristics, medical history, disease severity, physician experience, and periprocedural anticoagulant regimens may diminish the interpretability of outcomes. We attempted to extract all possible confounders and perform subgroup analyses to determine their influence on primary patency. Unfortunately, the number of available studies with complete data were inadequate for some comparisons. Despite the known limitations of the current review, the data herein represent the most comprehensive synthesis of data available for iliofemoral venous stent placement and thus a reasonable starting point for PG development. As data from prospective venous stent clinical trials become available, we anticipate that the choice of study end points and outcome comparators will be refined over time. This meta-analysis is associated with several issues that may influence interpretation. We did not attempt to analyze treatment effects by stent type because these data were not reported. Of the stents deployed in included studies, 99.9% were developed for nonvenous applications; stents manufactured specifically to accommodate venous anatomy are currently in clinical trials because there is no on-label stent approved specifically for this indication in the United States. The rationale for stent placement (routine versus 50% residual thrombus after angioplasty) was inadequately described in most studies and was further confounded because many retrospective studies included only patients treated with a stent. There was significant variability in the consistency and thoroughness of patient characteristics and outcomes reporting among studies. Finally, this meta-analysis was composed primarily of retrospective, single-site, level IV studies. Strengths of this meta-analysis are structured data extraction methodology, exclusion of multiple publications reporting on common patients, outcomes reporting by distinct venous disease pathogenesis, and inclusion of all relevant studies, including those where only a subset of patients underwent attempted stent placement provided stent-specific outcomes were reported. Conclusions Stent placement for iliofemoral venous obstruction results in high technical success and low complication rates regardless of disease pathogenesis. Avoidance of early thrombosis and maintenance of long-term patency remain therapeutic challenges with stent placement in patients with acute DVT and CPT syndrome. We propose a mechanism for PG comparisons so that future trials of stents designed specifically for this indication may be implemented. Acknowledgments We thank Teresa Nelson, MS, for statistical assistance. Sources of Funding Dr Miller was compensated by Veniti, Inc, for statistical analysis and interpretation. Neither Dr Razavi nor M.R. Jaff were compensated for their contributions to this work. Disclosures Dr Razavi is a coprinciple investigator for the VIRTUS: An Evaluation of the Veniti Vici Venous Stent System in Patients With Chronic Iliofemoral Venous Outflow Obstruction trial and an advisor to Abbott Vascular, Bard Peripheral, Boston Scientific, Covidien/ Medtronic, and Veniti. Dr Miller is a consultant to Veniti, Inc. M.R. Jaff is a noncompensated advisor to Abbott Vascular, Boston Scientific, Cordis Corporation, Medtronic Vascular; is the chair of the DSMB for BiO2/Novella; is a board member of Vascular Interventional Advances Physicians, a 501 (c) 3 not-for-profit education and research organization and the Intersocietal Accreditation Commission. References 1. Raju S, Fredericks RK. Late hemodynamic sequelae of deep venous thrombosis. J Vasc Surg. 1986;4: doi: / (86) Raju S, Neglen P. High prevalence of nonthrombotic iliac vein lesions in chronic venous disease: a permissive role in pathogenicity. J Vasc Surg 2006;44: , discussion Narayan A, Eng J, Carmi L, McGrane S, Ahmed M, Sharrett AR, Streiff M, Coresh J, Powe N, Hong K. Iliac vein compression as risk factor for left- versus right-sided deep venous thrombosis: case-control study. Radiology. 2012;265: doi: /radiol Taheri SA, Williams J, Powell S, Cullen J, Peer R, Nowakowski P, Boman L, Pisano S. Iliocaval compression syndrome. Am J Surg. 1987;154: Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S, Carta M, Cattelan AM, Polistena P, Bernardi E, Prins MH. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med. 1996;125: Prandoni P, Kahn SR. Post-thrombotic syndrome: prevalence, prognostication and need for progress. Br J Haematol. 2009;145: doi: /j x. 7. Ashrani AA, Heit JA. Incidence and cost burden of post-thrombotic syndrome. J Thromb Thrombolysis. 2009;28: doi: / s Enden T, Haig Y, Kløw NE, Slagsvold CE, Sandvik L, Ghanima W, Hafsahl G, Holme PA, Holmen LO, Njaastad AM, Sandbæk G, Sandset PM; CaVenT Study Group. Long-term outcome after additional catheterdirected thrombolysis versus standard treatment for acute iliofemoral deep vein thrombosis (the CaVenT study): a randomised controlled trial. Lancet. 2012;379: doi: /S (11) Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Intern Med. 2009;151:W65 W Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327: doi: / bmj Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315: George R, Verma H, Ram B, Tripathi R. The effect of deep venous stenting on healing of lower limb venous ulcers. Eur J Vasc Endovasc Surg. 2014;48: doi: /j.ejvs Hager ES, Yuo T, Tahara R, Dillavou E, Al-Khoury G, Marone L, Makaroun M, Chaer RA. Outcomes of endovascular intervention for May-Thurner syndrome. J Vasc Surg 2013;1: doi: /j. jvsv Liu Z, Gao N, Shen L, Yang J, Zhu Y, Li Z, Si Y. Endovascular treatment for symptomatic iliac vein compression syndrome: a prospective consecutive series of 48 patients. Ann Vasc Surg. 2014;28: doi: /j. avsg Lou WS, Gu JP, He X, Chen L, Su HB, Chen GP, Song JH, Wang T. Endovascular treatment for iliac vein compression syndrome: a comparison between the presence and absence of secondary thrombosis. Korean J Radiol. 2009;10: doi: /kjr

10 10 Razavi et al Meta-Analysis of Venous Stents 16. Meng QY, Li XQ, Qian AM, Sang HF, Rong JJ, Zhu LW. Endovascular treatment of iliac vein compression syndrome. Chin Med J (Engl). 2011;124: Neglén P, Hollis KC, Olivier J, Raju S. Stenting of the venous outflow in chronic venous disease: long-term stent-related outcome, clinical, and hemodynamic result. J Vasc Surg. 2007;46: doi: /j. jvs O Sullivan GJ, Semba CP, Bittner CA, Kee ST, Razavi MK, Sze DY, Dake MD. Endovascular management of iliac vein compression (May-Thurner) syndrome. J Vasc Interv Radiol. 2000;11: Ye K, Lu X, Li W, Huang Y, Huang X, Lu M, Jiang M. Long-term outcomes of stent placement for symptomatic nonthrombotic iliac vein compression lesions in chronic venous disease. J Vasc Interv Radiol. 2012;23: doi: /j.jvir Bjarnason H, Kruse JR, Asinger DA, Nazarian GK, Dietz CA Jr, Caldwell MD, Key NS, Hirsch AT, Hunter DW. Iliofemoral deep venous thrombosis: safety and efficacy outcome during 5 years of catheter-directed thrombolytic therapy. J Vasc Interv Radiol. 1997;8: Hölper P, Kotelis D, Attigah N, Hyhlik-Dürr A, Böckler D. Longterm results after surgical thrombectomy and simultaneous stenting for symptomatic iliofemoral venous thrombosis. Eur J Vasc Endovasc Surg. 2010;39: doi: /j.ejvs Husmann MJ, Heller G, Kalka C, Savolainen H, Do DD, Schmidli J, Baumgartner I. Stenting of common iliac vein obstructions combined with regional thrombolysis and thrombectomy in acute deep vein thrombosis. Eur J Vasc Endovasc Surg. 2007;34: doi: /j. ejvs Jackson LS, Wang XJ, Dudrick SJ, Gersten GD. Catheter-directed thrombolysis and/or thrombectomy with selective endovascular stenting as alternatives to systemic anticoagulation for treatment of acute deep vein thrombosis. Am J Surg. 2005;190: doi: /j. amjsurg Jeon UB, Chung JW, Jae HJ, Kim HC, Kim SJ, Ha J, Park JH. May- Thurner syndrome complicated by acute iliofemoral vein thrombosis: helical CT venography for evaluation of long-term stent patency and changes in the iliac vein. AJR Am J Roentgenol. 2010;195: doi: / AJR Kölbel T, Lindh M, Holst J, Uher P, Eriksson KF, Sonesson B, Malina M, Ivancev K. Extensive acute deep vein thrombosis of the iliocaval segment: midterm results of thrombolysis and stent placement. J Vasc Interv Radiol. 2007;18: doi: /j.jvir Kwak HS, Han YM, Lee YS, Jin GY, Chung GH. Stents in common iliac vein obstruction with acute ipsilateral deep venous thrombosis: early and late results. J Vasc Interv Radiol. 2005;16: doi: /01. RVI Matsuda A, Yamada N, Ogihara Y, Tsuji A, Ota S, Ishikura K, Nakamura M, Ito M. Early and long-term outcomes of venous stent implantation for iliac venous stenosis after catheter-directed thrombolysis for acute deep vein thrombosis. Circ J. 2014;78: Meng QY, Li XQ, Jiang K, Qian AM, Sang HF, Rong JJ, Duan PF, Zhu LW. Stenting of iliac vein obstruction following catheter-directed thrombolysis in lower extremity deep vein thrombosis. Chin Med J (Engl). 2013;126: 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;211: doi: /radiology r99ap Oguzkurt L, Tercan F, Ozkan U, Gulcan O. Iliac vein compression syndrome: outcome of endovascular treatment with long-term follow-up. Eur J Radiol. 2008;68: doi: /j.ejrad Park JY, Ahn JH, Jeon YS, Cho SG, Kim JY, Hong KC. Iliac vein stenting as a durable option for residual stenosis after catheter-directed thrombolysis and angioplasty of iliofemoral deep vein thrombosis secondary to May-Thurner syndrome. Phlebology. 2014;29: doi: / Patel NH, Stookey KR, Ketcham DB, Cragg AH. Endovascular management of acute extensive iliofemoral deep venous thrombosis caused by May-Thurner syndrome. J Vasc Interv Radiol. 2000;11: Stanley GA, Murphy EH, Plummer MM, Chung J, Modrall JG, Arko FR. Midterm results of percutaneous endovascular treatment for acute and chronic deep venous thrombosis. J Vasc Surg: Venous and Lym Dis 2013;1: Titus JM, Moise MA, Bena J, Lyden SP, Clair DG. Iliofemoral stenting for venous occlusive disease. J Vasc Surg. 2011;53: doi: /j. jvs Xue GH, Huang XZ, Ye M, Liang W, Zhang H, Zhang JW, Zhang BG. Catheter-directed thrombolysis and stenting in the treatment of iliac vein compression syndrome with acute iliofemoral deep vein thrombosis: outcome and follow-up. Ann Vasc Surg. 2014;28: doi: /j. avsg Zhu QH, Zhou CY, Chen Y, Wang J, Mo HY, Luo MH, Huang W, Yu XF. Percutaneous manual aspiration thrombectomy followed by stenting for iliac vein compression syndrome with secondary acute isolated iliofemoral deep vein thrombosis: a prospective study of single-session endovascular protocol. Eur J Vasc Endovasc Surg. 2014;47: doi: /j. ejvs Alerany MB, Lamoca LM, Ortega MR, Rivas IL, Desboeufs RZ, Kiuri SS. Endovascular treatment of iliofemoral chronic post-thrombotic venous flow obstruction. J Vasc Surg 2014;2: Blättler W, Blättler IK. Relief of obstructive pelvic venous symptoms with endoluminal stenting. J Vasc Surg. 1999;29: de Wolf M, Arnoldussen C, Grommes J, Hsien SG, Nelemans PJ, de Hann MW, de Graaf R, Wittens CH. Minimally invasive treatment of chronic iliofemoral venous occlusive disease. J Vasc Surg: Venous and Lym Dis 2013;1: Delis KT, Bjarnason H, Wennberg PW, Rooke TW, Gloviczki P. Successful iliac vein and inferior vena cava stenting ameliorates venous claudication and improves venous outflow, calf muscle pump function, and clinical status in post-thrombotic syndrome. Ann Surg. 2007;245: doi: /01.sla Kurklinsky AK, Bjarnason H, Friese JL, Wysokinski WE, McBane RD, Misselt A, Moller SM, Gloviczki P. Outcomes of venoplasty with stent placement for chronic thrombosis of the iliac and femoral veins: single-center experience. J Vasc Interv Radiol. 2012;23: doi: /j.jvir Nayak L, Hildebolt CF, Vedantham S. Postthrombotic syndrome: feasibility of a strategy of imaging-guided endovascular intervention. J Vasc Interv Radiol. 2012;23: doi: /j.jvir Rosales A, Sandbaek G, Jørgensen JJ. Stenting for chronic post-thrombotic vena cava and iliofemoral venous occlusions: mid-term patency and clinical outcome. Eur J Vasc Endovasc Surg. 2010;40: doi: /j.ejvs Sang H, Li X, Qian A, Meng Q. Outcome of endovascular treatment in postthrombotic syndrome. Ann Vasc Surg. 2014;28: doi: /j.avsg Sarici IS, Yanar F, Agcaoglu O, Ucar A, Poyanli A, Cakir S, Aksoy SM, Kurtoglu M. Our early experience with iliofemoral vein stenting in patients with post-thrombotic syndrome. Phlebology. 2014;29: doi: / Vogel D, Comerota AJ, Al-Jabouri M, Assi ZI. Common femoral endovenectomy with iliocaval endoluminal recanalization improves symptoms and quality of life in patients with postthrombotic iliofemoral obstruction. J Vasc Surg. 2012;55: doi: /j.jvs Wahlgren CM, Wahlberg E, Olofsson P. Endovascular treatment in postthrombotic syndrome. Vasc Endovascular Surg. 2010;44: doi: / Ye K, Lu X, Jiang M, Yang X, Li W, Huang Y, Huang X, Lu M. Technical details and clinical outcomes of transpopliteal venous stent placement for postthrombotic chronic total occlusion of the iliofemoral vein. J Vasc Interv Radiol. 2014;25: doi: /j.jvir Eijgenraam P, ten Cate H, ten Cate-Hoek AJ. Venous stenting after deep venous thrombosis and antithrombotic therapy: a systematic review. Rev Vasc Med 2014;2: doi: /j.rvm Vedantham S, Grassi CJ, Ferral H, Patel NH, Thorpe PE, Antonacci VP, Janne d Othée BM, Hofmann LV, Cardella JF, Kundu S, Lewis CA, Schwartzberg MS, Min RJ, Sacks D; Technology Assessment Committe of the Society of Interventional Radiology. Reporting standards for endovascular treatment of lower extremity deep vein thrombosis. J Vasc Interv Radiol. 2006;17: doi: /01. RVI c2.

11 Safety and Effectiveness of Stent Placement for Iliofemoral Venous Outflow Obstruction: Systematic Review and Meta-Analysis Mahmood K. Razavi, Michael R. Jaff and Larry E. Miller Circ Cardiovasc Interv. 2015;8:e doi: /CIRCINTERVENTIONS Circulation: Cardiovascular Interventions is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX Copyright 2015 American Heart Association, Inc. All rights reserved. Print ISSN: Online ISSN: The online version of this article, along with updated information and services, is located on the World Wide Web at: Data Supplement (unedited) at: Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation: Cardiovascular Interventions can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: Subscriptions: Information about subscribing to Circulation: Cardiovascular Interventions is online at: