- Email: [email protected]
Outcomes of endovascular intervention for May-Thurner syndrome
From the American Venous Forum
Outcomes of endovascular intervention for May-Thurner syndrome Eric S. Hager, MD,a Theodore Yuo, MD,a Robert Tahara, MD,b Ellen Dillavou, MD,a Georges Al-Khoury, MD,a Luke Marone, MD,a Michel Makaroun, MD,a and Rabih A. Chaer, MD,a Pittsburgh and Bradford, Pa Background: Endovascular interventions for May-Thurner syndrome (MTS) have become first-line therapy, often performed in a young patient population despite the lack of robust supportive data. This article reports on long-term outcomes from a large series of patients treated in the setting of de novo or postthrombotic presentation. Methods: A retrospective review of MTS patients treated between 2006 and 2010 was conducted at two institutions. Patients who presented with acute iliofemoral deep vein thrombosis (DVT) were treated with either catheter-directed thrombolysis (CDT) and/or pharmacomechanical thrombolysis and identified as having a venous stenosis by venogram or intravascular ultrasound (IVUS). Patients who presented with chronic venous insufficiency symptoms or recalcitrant ulceration but no DVT and evidence of MTS on duplex ultrasound, magnetic resonance venography, or computerized tomography venography were evaluated by venography. IVUS was selectively utilized. Stenting of the iliocaval junction was performed in all patients with a >50% diameter stenosis by IVUS or venogram. Results: Seventy patients with MTS underwent 77 lower extremity interventions. They were divided into two groups: postthrombotic (group 1) and de novo presentation of chronic swelling/pain or ulceration but no DVT (group 2). There were 56 extremities in group 1 and 21 extremities in group 2.
Both groups were comparable in terms of gender distribution and comorbidities, but hypercoagulable state was more common in group 1 (P [ .014), and average CEAP and Villalta scores on presentation were higher in group 2 (P < .001). There were left-sided symptoms in 40 (78%) patients in group 1 and 15 (79%) in group 2 (P [ 1.00). Female patients were more likely to have left-sided symptoms compared with male patients (odds ratio, 4.88; 95% confidence interval, 1.4915.89; P [ .014). The average stent size was significantly different among the groups (P < .027), with different types used in each group. There was one patient in group 1 who had significant periprocedural bleeding (1 unit transfused) during the CDT portion of the procedure. Mean follow up was 29.7 months in group 1 (range, 18.4-58.3 months) and 22.4 months in group 2 (range, 17.1-42 months). Complete or partial symptom relief was reported for 52 (92.9%) extremities in group 1 and 20 (95.2%) extremities in group 2 (P [ 1.00). The overall primary patency of group 1 at 36 months by life-table analysis was 91% with a secondary patency of 95%. The primary and secondary patency for group 2 was 91% at 36 months. Conclusions: Stenting of MTS has proven to be safe, efficacious, and durable for up to 36 months in both the postthrombotic patient as well as those treated for edema alone. (J Vasc Surg: Venous and Lym Dis 2013;1:270-5.)
May-Thurner syndrome (MTS) is a venous anomaly rarely encountered in contemporary vascular practice.1 In the most common variant, it is defined as left lower extremity venous hypertension from compression by the iliac artery with or without left iliofemoral deep vein thrombosis (DVT).2-4 Manifestations include left lower extremity swelling, pain, or thrombosis due to mechanical venous obstruction of the left iliac vein by the right iliac artery and the underlying vertebral body2 and occur secondary to intimal hyperplasia from the repeated trauma leading to venous narrowing due to intraluminal webs and
fibrosis.3 The incidence of MTS is extremely difficult to accurately estimate due to the high percentage of patients who are asymptomatic, and thus the denominator is unknown.3 There have been small retrospective imaging studies that have found an incidence of extrinsic iliac vein compression between 22% and 37%; however, this may be underestimated since the largest meta-analysis of 1046 patients lysed for iliofemoral DVT revealed an underlying iliac lesion in 46% of patients.3-8 Treatment paradigms have changed as experience with endovascular modalities has evolved, making traditional surgical treatment strategies for MTS essentially obsolete. While endovascular treatment with angioplasty and stenting with selective thrombolysis has become the standard for symptomatic MTS,9 the majority of the data on endovenous treatment is derived from case reports, small observational series, or inferred information from venous stents placed in other anatomic locations.10 Information on mid- and long-term follow-up is lacking in the setting of postthrombotic iliac vein compression, since other series have mainly reported on the outcomes of nonthrombotic MTS.7 The purpose of this study is to report on the mid- and long-term outcomes of patients treated with symptomatic thrombotic and nonthrombotic MTS,
From the Department of Vascular Surgery, University of Pittsburgh Medical Center, Pittsburgha; and the Bradford Medical Center, Bradford.b Author conflict of interest: none. Presented at the Twenty-fourth Annual Meeting of the American Venous Forum, Orlando, Fla, February 8-11, 2012. Reprint requests: Dr Eric S. Hager, University of Pittsburgh Medical Center, Shadyside Medical Building, 5200 Centre Ave, Ste 307, Pittsburgh, PA 15232 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the Journal policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 2213-333X/$36.00 Copyright Ó 2013 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvsv.2012.11.002
270
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume 1, Number 3
with emphasis on the durability of endovenous treatment in an effort to predict stent patency, need for reintervention, and symptom relief. METHODS Patient population. This study was approved by the Institutional Review Board at the University of Pittsburgh. All patients were identified from a prospectively maintained registry of those treated with iliac venoplasty and stenting with or without thrombolysis between 2006 and 2011 at Bradford Medical Center and the University of Pittsburgh Medical Center. The records were retrospectively reviewed for patient demographics, symptom complex, indications for treatment, periprocedural complications, and clinical outcomes. All patients had radiographic evidence of iliac vein stenosis from MTS by computerized tomography venography or duplex ultrasound, and this was confirmed with venography or intravascular ultrasound (IVUS). Radiographic criteria for the diagnosis of MTS included >50% stenosis by IVUS,11 loss of contrast density at the point of the crossing iliac artery with collateralization by venography, or clear extrinsic compression. Patients with iliac vein occlusion and failed lysis were excluded from this study. Patients were divided into two groups based on presentation; those with acute iliofemoral DVT (group 1) and those with de novo presentation of swelling and pain or recalcitrant ulceration without DVT (group 2). Group 1 was comprised of patients who were treated with thrombolysis for iliofemoral DVT with evidence of iliac vein compression after successful thrombolysis. The majority of patients (63%) had thrombosis of the femoral and iliac veins, with the rest suffering from iliofemoral and popliteal vein thrombosis. All patients in this group underwent a hypercoagulable hematologic workup. Patients were administered intravenous heparin preoperatively to maintain therapeutic partial thromboplastin times. Venous ultrasound evaluation of the lower extremities and selective computed tomography with intravenous contrast and a venous phase were used to determine proximal thrombus extension. The age of the thrombus was determined based on the onset of clinical symptoms as reported by the patient. Interventions were performed by vascular surgeons in an endovascular suite with fixed imaging (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom). Patients in group 2 included patients with recalcitrant varicosities, venous stasis ulcerations, or symptoms of venous hypertension. The majority of patients had symptoms for greater than 6 months prior to being evaluated by a vascular surgeon. All patients had a venous ultrasound reflux study to evaluate for superficial or deep venous insufficiency. Patients were treated with superficial venous ablation as indicated and were recommended medical management (compression therapy, compression dressings, leg elevation). Patents that failed to improve were screened for iliac vein pathology by duplex ultrasound, computerized tomography venography, or magnetic
Hager et al 271
renonance venography. When iliac vein compression was noted, invasive venography with IVUS was performed to assess for the extent and degree of the stenosis. Thrombolysis and stenting technique. Thrombolysis was performed on patients in group 1 as previously described.12 In brief, patients were treated under conscious sedation and local anesthesia. In the setting of iliofemoral DVT, popliteal vein access was obtained in the prone position, using ultrasound guidance. Ascending venography was performed via a 6F sheath, and a heparin bolus was administered at a dose of 100 units/kg. Pharmacomechanical thrombolysis (PMT) with either the Trellis device (Covidien, Mansfield, Mass) or AngioJet (Bayer, Leverkusen, North Rhine-Westphalia, Germany) and tissue plasminogen activator was used initially in the setting of acute DVT that was #14 days old. Catheterdirected thrombolysis (CDT) was used initially for subacute thrombus (>14 days old) or residual thrombus after PMT with either a multi-side hole catheter or ultrasound-assisted CDT with the EKOS catheter (EKOS Corporation, Ottawa, Canada). Fibrinogen levels were monitored, and the tissue plasminogen activator drip rate adjusted when the value fell below 150 mg/dL. Patients with iliac vein stenosis as determined by venography or IVUS were treated with angioplasty and stenting. In patients with obvious severe stenosis on venography, IVUS was not obtained, and stenting was performed based on venographic data. IVUS was used in equivocal cases and most patients in group 2, and it allowed for accurate measurement of the area reduction and adequate stent sizing. Patients that had an area reduction of greater than 50% in the iliac vein were stented. The stent is splayed into the inferior vena cava by 1 to 2 cm unless the lesion is more than 2 cm from the left common iliac vein orifice, and there was no evidence by IVUS of either thrombosis/fibrosis/ thickening at the orifice.10 The stent diameter was typically oversized by 10% to 20%, and the length was chosen to ensure complete lesion coverage. Follow up. Follow-up clinical and duplex ultrasound evaluations were performed within 30 days postprocedure, at 6 months, and yearly thereafter. Patients with active venous stasis ulcers were followed on a weekly basis until their wounds were healed. Postprocedure, decisions about the type and duration of anticoagulation for patients in group 1 were made in conjunction with the hematology consultant based on the clinical presentation and hypercoagulable workup. Treatment consisted of aspirin, Plavix, and/or Coumadin. Patients in group 2 were treated with aspirin and Plavix for 3 months. All patients were given graduated compression stockings to wear. The international normalizing ratio was followed to ensure proper dosing of Coumadin, with a target goal between 2 and 3. Primary end points were improvement of symptoms and freedom from stent thrombosis. Secondary end points were valve reflux in the femoral and popliteal veins and treated vein patency. These end points were chosen to determine the efficacy of lysis in preventing venous reflux and to differentiate primary stent failure versus thrombosis
272 Hager et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS July 2013
of the inflow veins. The group’s demographics and comorbidities were also compared. The patient’s initial Villalta13 and CEAP scores were recorded and compared at the most recent follow-up. Statistical analysis. Demographic and procedural result data were analyzed using Fisher’s exact test for dichotomous variables and the Kruskal-Wallis equalityof-populations rank test for age, size of the stents used during the procedures, and CEAP class. These results were reported with median and interquartile ranges. CEAP class was also analyzed after dichotomization, with scores of greater than or equal to 3 being the dividing line. Primary stent patency rates were assessed using KaplanMeier survival estimates, and comparisons between groups 1 and 2 were assessed using the log-rank test for equality of survivor functions. The threshold for significance for all statistical analyses was .05. Statistical analysis was performed with Stata IC Version 11.2 (StataCorp LP, College Station, TX). RESULTS Patient population. During the 5.5 year period, there were 70 patients identified with MTS and treated with endovenous stenting, seven of whom had bilateral iliac stents placed (Table I). These patients were divided according to symptoms, with group 1 consisting of patients diagnosed with MTS after successful lysis of iliofemoral DVT and group 2 being comprised of patients with no DVT and treated superficial venous disease but recalcitrant swelling/pain or ulceration. The average age of patients was similar between groups (group 1, 51.9 years [range, 18-73 years] vs group 2, 52.8 years [range, 16-80 years]; P ¼ .440). Of the patients treated, 43 (61.4%) were younger than 50 years old. The majority of patients suffered from left-sided symptoms in both groups (group 1, 78.4% vs group 2, 78.8%; P ¼ 1.00). Compression therapy was more commonly used preoperatively in group 2 (37.5% vs 95.2%; P < .001). Combining both groups 1 and 2, female patients were more likely to
have left-sided symptoms compared with male patients (odds ratio, 4.88; 95% confidence interval, 1.49-15.89; P ¼ .014). The initial CEAP score was significantly lower in group 1 (P < .001), with an initial score of $3 in only 29% of patients. The Villalta score was significantly lower in group 1 (P < .001), with an average score of 16.3 versus 34.7 in group 2. Procedural details. The initial procedural details and outcomes of both groups after stenting are described in Table II. Access site was predicated on indication for treatment, and therefore, the use of distal access was more common in group 1 where iliofemoral DVT was present. The popliteal approach was utilized in 85.7% of patients in group 1 (P < .001), while the common femoral or thigh femoral vein (if patent) was accessed in 90.5% of patients in group 2 (P < .001). IVUS use was used more frequently in group 2 (P < .001), as this modality was only used in group 1 when venographic findings were questionable for the presence of iliac vein stenosis. There were no patients with more than one site of iliac vein compression, nor patients with iliac vein compression by the inguinal ligament. The number of stents was therefore predicated on the length of the stenosis and was not different between the groups. There were more recorded complications in group 1, with significant bleeding (1.8%), perioperative thrombosis (5.3%), and renal insufficiency (1.8%) being more common but not achieving statistical significance. Outcomes and follow-up. The median follow-up period was 29.7 months (range, 18.4-58.3 months) in group 1, with 39% of patients having a follow-up of greater than 3 years. The median follow-up of group 2 was 22.4 months (range, 17.1-42.0 months) with 29% having follow-up of greater than 3 years (Table III). Symptom resolution or improvement occurred in 52 of 56 (92.9%) of patients in group 1 and 20 of 21 (95.7%) of patients in group 2 at the latest follow-up. The average postprocedure Villalta score improved in both groups, to 6.7 in group 1 and 19.2 in group 2, both of which achieved significance when compared with their preoperative scores
Table I. Patient demographics and comorbidities Characteristics
Group 1 (n ¼ 51)
Group 2 (n ¼ 19)
P value
Median age, years (interquartile range) Female gendera Left-sided symptomsa Diabetes Malignancy Hypertension Hypercoagulable state identified among tested patientsb Coronary artery disease Hyperlipidemia Median CEAP score (interquartile range) Initial CEAP score greater than or equal to 3c Initial Villalta score
51.9 30 40 5 2 23 13 8 16 1 16
52.8 14 15 3 0 7 0 1 6 4 21
.440 .282 1.000 .674 .982 .596 .014 .427 1.000 <.001 <.001 <.001
(41.4-63.6) (58.8%) (78.4%) (9.8%) (3.9%) (45.1%) (26.5%) (15.7%) (31.4%) (0-4) (28.6%) 16.3
(46.2-62.1) (73.7%) (78.9%) (15.8%) (0%) (36.8%) (0.0%) (5.3%) (31.6%) (4-6) (100%) 34.7
a Combining both groups 1 and 2, female patients were more likely to have left-sided symptoms compared with male patients (odds ratio, 4.88; 95% confidence interval, 1.49-15.89; P ¼ .014). b Two patients in group 1 and eight patients in group 2 were not tested for a hypercoagulable state. c CEAP scores assessed for each limb separately. Group 1, 56 limbs. Group 2, 21 limbs.
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume 1, Number 3
Table II. Procedural details and initial outcomes Characteristics Preoperative use of compression stockings Approach Popliteal Femoral Other Intravenous ultrasound use Median stent size, mm (interquartile range) Self-expanding stent Protégé Wallstent Balloon-expandable stent Any postoperative complications Clinically significant bleeding Immediate thrombosis Renal insufficiency Need for blood transfusion
Group 1 Group 2 (56 limbs) (21 limbs) P value 21 (37.5%) 20 (95.2%) <.001 48 7 1 15 14
(85.7%) 0 (0.0%) <.001 (12.5%) 19 (90.5%) <.001 (1.8%) 2 (9.5%) .179 (26.8%) 19 (90.5%) <.001 (12-16) 16 (14-18) .027
50 14 36 6 8 1 3 1 1
(89.3%) 10 (47.6%) <.001 (25.0%) 10 (47.6%) .056 (64.3%) 0 (0.0%) <.001 (10.7%) 1 (4.8%) .211 (14.3%) 0 (0.0%) .099 (1.8%) 0 (0.0%) 1.000 (5.3%) 0 (0.0%) .670 (1.8%) 0 (0.0%) 1.000 (1.8%) 0 (0.0%) 1.000
(P ¼ .03). The median CEAP score was 1 in group 1 and 4 in group 2, unchanged from the preoperative assessment. Stent patency. Stent patency was similar in the followup period. Symptom relief seemed to correlate with stent patency, although there were two patients that had stent thrombosis resulting in mild symptoms only. Patients in group 1 had a primary patency rate of 91% and secondary patency of 98% at 36 months. There were five (8.9%) patients with in-stent thrombosis requiring reintervention in group 1. Three patients had rethombosis within 48 hours of the procedure, all of which were subtherapeutic on their intravenous anticoagulation. The two other patients that rethrombosed at later time points (1 week and 6 months) were on Coumadin but had subtherapeutic international normalizing ratios at the time of thrombosis. Both of these patients had factor V Leiden thrombophilia. All five of these patients had recurrence of symptoms and successfully underwent lysis of the thrombosed stent using either CDT or a combination of CDT and PMT. Of these five patients, four had Protégé (eV3 Endovascular, Plymouth, Minn) stents implanted, and one had a 14-mm balloon-expandable stent. Two had been stented with
Hager et al 273
smaller stents (12-mm Protégé) and were treated with a larger 16-mm balloon-expandable stent within the Protégé with good results. Since the time of the reintervention, only one patient has suffered from late thrombosis of the stent at 2 years. This patient elected not to have any intervention performed due to relatively mild symptoms. Patients in group 2 had a primary patency rate of 91% and secondary patency of 91% at 36 months. Group 2 had two (9.4%) stent thrombosis at 12 months. One patient had a 12-mm Protégé stent placed during the original surgery, which was most likely undersized. The patient elected not to undergo lysis due to a recent gastrointestinal bleed and was treated with anticoagulation only. The other patient’s stent was a 16-mm Wallstent that appeared to be placed too far distally, leaving proximal disease at the iliocaval junction uncovered. This patient elected not to have a reintervention due to minimal symptoms. Over the entire analysis period, there was no significant difference in patency by life-table analysis (P ¼ .612; Fig). Patients with stent failure presented with complete stent thrombosis, and there were no instances of in-stent stenosis identified in either group. DISCUSSION Stenting has become the primary treatment modality for MTS. Due to the heterogeneous symptomatology and age of patients with MTS, it is important to describe the long-term fate of the venous stents and the resultant symptom relief. The treatment of MTS patients who present with thrombosis has changed dramatically, with increased application of CDT. Catheter-based strategies have been utilized in the management of iliofemoral thrombosis, with particular emphasis on younger patients, given the high association with postthrombotic syndrome and its attendant morbidity.14 Mewissen and colleagues reported that a third of limbs treated with thrombolysis for DVT required stenting and that stented limbs had a significantly better patency than those that were not.15 Today, both the Society for Interventional Radiology and the Society of Vascular Surgery guidelines recommend iliac vein stenting in the setting of external compression.16,17 The long-term durability and efficacy of iliac venous stenting has been well described in the setting of DVT, with
Table III. Follow-up information Characteristics Median length of follow-up, months (range) Need for reintervention Symptom relief at last follow-up visit Moderate Complete Limitations with ambulation on last follow-up examination (pain/swelling) Recurrent or distal deep vein thrombosis visualized on last follow-up examination Use of compression stockings Preserved valve function CEAP class greater than or equal to 3 at last follow-up Median CEAP score (interquartile range) Villalta score
Group 1 (56 limbs)
Group 2 (21 limbs)
P value
29.7 (18.4-58.3) 5 (8.9%) 52 (92.9%) 21 (37.5%) 31 (55.4%) 17 (30.4%) 7 (12.5%) 51 (91.1%) 31 (55.4%) 29 (51.8%) 1 (0-3) 6.7
22.4 (17.1-42.0) 0 (0.0%) 20 (95.2%) 10 (47.6%) 10 (47.6%) 0 (0.0%) 1 (4.8%) 21 (100%) 1 (4.8%) 21 (100%) 4 (4-6) 19.2
NA .315 1.000 .445 .613 .004 .435 .315 <.001 <.001 <.001 <.001
274 Hager et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS July 2013
Fig. Kaplan-Meier estimates for primary stent patency.
a 79% primary patency rate at 72 months with nearly 100% primary-assisted patency.10 This demonstrated durability of venous stenting among a wide variety of patients; however, the same conclusions may not be applicable to MTS patients where ongoing external stent compression by the iliac artery is to be expected. Small case series of iliac vein stenting for MTS have been reported with greater frequency over the past decade and have advocated the use of endovascular techniques.18-20 Recently, a retrospective review, published by Titus and colleagues, identified 15 MTS patients treated by endovenous stenting and demonstrated a primary patency rate of 88%, 78%, and 78% at 6 months, 12 months, and 24 months, respectively.21 A more recent study of the long-term patency of stenting for MTS after overnight CDT and aspiration thrombectomy demonstrated an exceptional 83.3% and 90% primary and secondary patency at 1 and 5 years,22 and this finding has been seen in earlier studies at 1 year employing a similar management strategy.8,23 However, this series did not include a control group of nonthrombotic patients with MTS. The current data are concordant with previous reports of iliac vein stenting, demonstrating favorable patency rates. We describe the outcomes up to 58 months in two populations of patients with MTS: those presenting with iliofemoral DVT and found to have MTS after lysis, and those with severe symptoms associated with venous hypertension. We found both groups to have excellent primary stent patency at 30-month follow-up by life-table analysis. Patients from both groups experienced significant symptom relief that mirrored stent patency. These findings
add support to the hypothesis that iliac vein obstruction or stenosis and subsequent venous hypertension is the genesis of the majority of patients’ morbidity,8 underscoring the importance of stent patency to maintain clinical symptom relief. We did not identify age to be a risk factor for stent failure as suggested by other groups,24 although logic would dictate that the longer the life expectancy of the patient, the greater the risk of stent failure. As 61% of the patients in this study were younger than 50 years old, it is essential that these patients be surveyed routinely to assess for in-stent stenosis or thrombosis. It is interesting to note that patients in group 1 presented with a lower average CEAP score, higher Villalta score, and were less likely to wear graded compression stockings than those in group 2, which emphasizes the heterogeneity of the MTS patient population. We believe this is a result of the relative acuity of presentation and symptoms of the patients with thrombosis and the fact that many had no known antecedent venous disease. The patients represented in group 2 suffered from significant venous disease and, despite superficial vein ablation, continued to have severe symptoms as well as high Villalta scores. Stenting of the compressed venous system improved venous hypertension as is reflected by the improvement in both subjective symptoms and Villalta score. It is noteworthy, however, that in group 1, there were an increased number of patients with a CEAP score of 3 at follow-up. The majority of these patients suffered from edema, suggesting that although the majority of PTS symptoms can be mitigated, there is still significant sequelae associated with the initial thrombotic event.
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume 1, Number 3
Also noteworthy was the finding that the majority of patients with right-sided MTS were male. This finding may be due to the shape of the male pelvis. Although numerous articles have described the left-sided and female predominance of MTS,10 there has been no correlation between right-sided MTS and male gender. We postulate that the conical shape of the male pelvis may predispose the right iliac vein to compression by the iliac artery, although it is possible that MTS may have been overdiagnosed in some of these patients. One limitation of this study is the inability to identify the entire pool of patients who suffer from MTS, as some of them may have been treated conservatively. As such, it is difficult to make conclusions on all patients with MTS, and statistical analysis is limited by the lack of knowledge of the denominator. Further limitations of this study are those inherent to retrospective database review and include reporting and selection bias during data collection as well as missing data secondary to loss of follow-up. In addition, some patients were lost to follow-up, which makes the long-term data somewhat limited. Another potential limitation was the variability in stent type and size used, which was not uniform early on in our experience. Nevertheless, to our knowledge, this study represents one of the larger reports on the endovascular treatment and long-term outcomes of patients with MTS comparing thrombotic and nonthrombotic patients. These data support the durability and efficacy of endovenous stenting among symptomatic patients with MTS with thrombotic or nonthrombotic presentation. On long-term follow-up, stent patency does not seem to be affected by the ongoing extrinsic compression by the adjacent iliac artery. AUTHOR CONTRIBUTIONS Conception and design: EH, RC Analysis and interpretation: TY, EH, RC Data collection: EH, TY Writing the article: EH, RC, TY Critical revision of the article: ED, GAK, LM, MM Final approval of the article: RC, MM Statistical analysis: TY Obtained funding: Not applicable Overall responsibility: RC
5.
6.
7. 8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21.
REFERENCES 22. 1. Seidensticker D, Wilcox J, Gagne P. Treatment of May-Thurner syndrome with catheter-directed thrombolysis and stent placement, complicated by heparin-induced thrombocytopenia. Cardiovasc Surg 1998;6:607-13. 2. Suwanabol PA, Tefera G, Schwarze ML. Syndromes associated with the deep veins: phlegmasia Cerulea Dolens, May-Thurner Syndrome, and Nutcracker Syndrome. Perspect Vasc Surg Endovasc Ther 2010;22: 223-30. 3. Kibbe MR, Ujiki M, Goodwin AL, Eskandari M, Yao J, Matsumura J. Iliac vein compression in an asymptomatic patient population. J Vasc Surg 2004;39:937-43. 4. Murphy EH, Davis CM, Journeycake JM, DeMuth RP, Arko FR. Symptomatic iliofemoral DVT after onset of oral contraceptive use in
23.
24.
Hager et al 275
women with previously undiagnosed May-Thurner syndrome. J Vasc Surg 2009;49:697-703. Wolpert LM, Rahmani O, Stein B, Gallagher JJ, Drezner AD. Magnetic resonance venography in the diagnosis and management of May-Thurner syndrome. Vasc Endovasc Surg 2002;36:51-7. Vedantham S, Thorpe PE, Cardella JF, Grassi CJ, Patel NH, Ferral H, et al. Quality improvement guidelines for the treatment of lower extremity deep vein thrombosis with use of endovascular thrombus removal. J Vasc Interv Radiol 2006;17:435-7. Taheri SA, Williams J, Powell S. Iliocaval compression syndrome. Am J Surg 1987;154:169-72. O’Sullivan GJ, Semba CP, Bittner CA, Kee ST, Razavi MK, Sze DY, et al. Endovascular management of iliac vein compression (MayThurner) syndrome. J Vasc Interv Radiol 2000;11:823-36. Dhillon RK, Stead LG. Acute deep vein thrombosis due to MayThurner syndrome. Am J Emerg Med 2010;28:254. Neglén P, Hollis KC, Oliver 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;46:979-90. Oguzkurt L, Ozkan U, Tercan F, Koc Z. Ultrasonographic diagnosis of iliac vein compression (May-Thurner) syndrome. Diagn Interv Radiol 2007;13:152-5. Jeyabalan G, Marone L, Rhee R, Hirsch S, Makaroun MS, Cho JS, et al. Inflow thrombosis does not adversely affect thrombolysis outcomes of symptomatic iliofemoral deep vein thrombosis. J Vasc Surg 2011;54:448-53. Villalta S, Bagatella P, Piccioli A, Lensing A, Prins M, Prandoni P. Assessment of validity and reproducibility of a clinical scale for the postthrombotic syndrome. Haemostasis 1994;24:158a. Goldenberg NA, Durham JD, Knapp-Clevenger R, MancoJohnson MJ. A thrombolytic regimen for high-risk deep venous thrombosis may substantially reduce the risk of post-thrombotic syndrome in children. Blood 2007;110:45-53. 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;211:39-49. Vedantham S, Millward SF, Cardella JF, Hofmann LV, Razavi MK, Grassi CJ, et al. Society of Interventional Radiology position statement: treatment of acute iliofemoral deep vein thrombosis with use of adjunctive catheter-directed intrathrombus thrombolysis. J Vasc Interv Radiol 2006;17:613-6. Meissner MH, Gloviczki P, Comerota AJ, Dalsing MC, Eklof BG, Gillespie DL, et al. Early thrombus removal strategies for acute venous thrombosis: clinical practice guidelines of the Society for Vascular Surgery and the American Venous Forum. J Vasc Surg 2012;55:1449-62. Binkert CA, Schoch E, Stuckmann G, Largiader J, Wigger P, Schoepke W, et al. Treatment of pelvic venous spur (May-Thurner syndrome) with self-expanding metallic endoprostheses. Cardiovasc Intervent Radiol 1998;21:22-6. Heniford BT, Senler SO, Olsofka JM, Carrillo EH, Bergamini TM. May-Thurner syndrome: management by endovascular surgical techniques. Ann Vasc Surg 1998;12:482-6. Baron HC, Shams J, Wayne M. Iliac vein compression syndrome: a new method of treatment. Am Surg 2000;66:653-5. Titus JM, Moise MA, Bena J, Lyden SP, Clair DG. Iliofemoral stenting for venous occlusive disease. J Vasc Surg 2011;53:706-12. Jeon UB, Chung JW, Jae HJ, Kim HC, Kim SJ, Ha J, et al. MayThurner Syndrome complicated by acute iliofemoral vein thrombosis: helical CT venography for evaluation of long-term stent patency and changes in the iliac vein. AJR 2010;195:751-7. Lamont JP, Pearl GJ, Patetsios P, Warner MT, Gable DR, Garrett W, et al. Prospective evaluation of endoluminal venous stents in the treatment of the May-Thurner syndrome. Ann Vasc Surg 2002;16:61-4. Knipp BS, Ferguson E, Williams DM, Dasika NJ, Cwikiel W, Henke PK, Wakefield TW. Factors associated with outcome after interventional treatment of symptomatic iliac vein compression syndrome. J Vasc Surg 2007;46:743-9.
Submitted Aug 22, 2012; accepted Nov 25, 2012.
Outcomes of endovascular intervention for May-Thurner syndrome Eric S. Hager, MD,a Theodore Yuo, MD,a Robert Tahara, MD,b Ellen Dillavou, MD,a Georges Al-Khoury, MD,a Luke Marone, MD,a Michel Makaroun, MD,a and Rabih A. Chaer, MD,a Pittsburgh and Bradford, Pa Background: Endovascular interventions for May-Thurner syndrome (MTS) have become first-line therapy, often performed in a young patient population despite the lack of robust supportive data. This article reports on long-term outcomes from a large series of patients treated in the setting of de novo or postthrombotic presentation. Methods: A retrospective review of MTS patients treated between 2006 and 2010 was conducted at two institutions. Patients who presented with acute iliofemoral deep vein thrombosis (DVT) were treated with either catheter-directed thrombolysis (CDT) and/or pharmacomechanical thrombolysis and identified as having a venous stenosis by venogram or intravascular ultrasound (IVUS). Patients who presented with chronic venous insufficiency symptoms or recalcitrant ulceration but no DVT and evidence of MTS on duplex ultrasound, magnetic resonance venography, or computerized tomography venography were evaluated by venography. IVUS was selectively utilized. Stenting of the iliocaval junction was performed in all patients with a >50% diameter stenosis by IVUS or venogram. Results: Seventy patients with MTS underwent 77 lower extremity interventions. They were divided into two groups: postthrombotic (group 1) and de novo presentation of chronic swelling/pain or ulceration but no DVT (group 2). There were 56 extremities in group 1 and 21 extremities in group 2.
Both groups were comparable in terms of gender distribution and comorbidities, but hypercoagulable state was more common in group 1 (P [ .014), and average CEAP and Villalta scores on presentation were higher in group 2 (P < .001). There were left-sided symptoms in 40 (78%) patients in group 1 and 15 (79%) in group 2 (P [ 1.00). Female patients were more likely to have left-sided symptoms compared with male patients (odds ratio, 4.88; 95% confidence interval, 1.4915.89; P [ .014). The average stent size was significantly different among the groups (P < .027), with different types used in each group. There was one patient in group 1 who had significant periprocedural bleeding (1 unit transfused) during the CDT portion of the procedure. Mean follow up was 29.7 months in group 1 (range, 18.4-58.3 months) and 22.4 months in group 2 (range, 17.1-42 months). Complete or partial symptom relief was reported for 52 (92.9%) extremities in group 1 and 20 (95.2%) extremities in group 2 (P [ 1.00). The overall primary patency of group 1 at 36 months by life-table analysis was 91% with a secondary patency of 95%. The primary and secondary patency for group 2 was 91% at 36 months. Conclusions: Stenting of MTS has proven to be safe, efficacious, and durable for up to 36 months in both the postthrombotic patient as well as those treated for edema alone. (J Vasc Surg: Venous and Lym Dis 2013;1:270-5.)
May-Thurner syndrome (MTS) is a venous anomaly rarely encountered in contemporary vascular practice.1 In the most common variant, it is defined as left lower extremity venous hypertension from compression by the iliac artery with or without left iliofemoral deep vein thrombosis (DVT).2-4 Manifestations include left lower extremity swelling, pain, or thrombosis due to mechanical venous obstruction of the left iliac vein by the right iliac artery and the underlying vertebral body2 and occur secondary to intimal hyperplasia from the repeated trauma leading to venous narrowing due to intraluminal webs and
fibrosis.3 The incidence of MTS is extremely difficult to accurately estimate due to the high percentage of patients who are asymptomatic, and thus the denominator is unknown.3 There have been small retrospective imaging studies that have found an incidence of extrinsic iliac vein compression between 22% and 37%; however, this may be underestimated since the largest meta-analysis of 1046 patients lysed for iliofemoral DVT revealed an underlying iliac lesion in 46% of patients.3-8 Treatment paradigms have changed as experience with endovascular modalities has evolved, making traditional surgical treatment strategies for MTS essentially obsolete. While endovascular treatment with angioplasty and stenting with selective thrombolysis has become the standard for symptomatic MTS,9 the majority of the data on endovenous treatment is derived from case reports, small observational series, or inferred information from venous stents placed in other anatomic locations.10 Information on mid- and long-term follow-up is lacking in the setting of postthrombotic iliac vein compression, since other series have mainly reported on the outcomes of nonthrombotic MTS.7 The purpose of this study is to report on the mid- and long-term outcomes of patients treated with symptomatic thrombotic and nonthrombotic MTS,
From the Department of Vascular Surgery, University of Pittsburgh Medical Center, Pittsburgha; and the Bradford Medical Center, Bradford.b Author conflict of interest: none. Presented at the Twenty-fourth Annual Meeting of the American Venous Forum, Orlando, Fla, February 8-11, 2012. Reprint requests: Dr Eric S. Hager, University of Pittsburgh Medical Center, Shadyside Medical Building, 5200 Centre Ave, Ste 307, Pittsburgh, PA 15232 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the Journal policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 2213-333X/$36.00 Copyright Ó 2013 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvsv.2012.11.002
270
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume 1, Number 3
with emphasis on the durability of endovenous treatment in an effort to predict stent patency, need for reintervention, and symptom relief. METHODS Patient population. This study was approved by the Institutional Review Board at the University of Pittsburgh. All patients were identified from a prospectively maintained registry of those treated with iliac venoplasty and stenting with or without thrombolysis between 2006 and 2011 at Bradford Medical Center and the University of Pittsburgh Medical Center. The records were retrospectively reviewed for patient demographics, symptom complex, indications for treatment, periprocedural complications, and clinical outcomes. All patients had radiographic evidence of iliac vein stenosis from MTS by computerized tomography venography or duplex ultrasound, and this was confirmed with venography or intravascular ultrasound (IVUS). Radiographic criteria for the diagnosis of MTS included >50% stenosis by IVUS,11 loss of contrast density at the point of the crossing iliac artery with collateralization by venography, or clear extrinsic compression. Patients with iliac vein occlusion and failed lysis were excluded from this study. Patients were divided into two groups based on presentation; those with acute iliofemoral DVT (group 1) and those with de novo presentation of swelling and pain or recalcitrant ulceration without DVT (group 2). Group 1 was comprised of patients who were treated with thrombolysis for iliofemoral DVT with evidence of iliac vein compression after successful thrombolysis. The majority of patients (63%) had thrombosis of the femoral and iliac veins, with the rest suffering from iliofemoral and popliteal vein thrombosis. All patients in this group underwent a hypercoagulable hematologic workup. Patients were administered intravenous heparin preoperatively to maintain therapeutic partial thromboplastin times. Venous ultrasound evaluation of the lower extremities and selective computed tomography with intravenous contrast and a venous phase were used to determine proximal thrombus extension. The age of the thrombus was determined based on the onset of clinical symptoms as reported by the patient. Interventions were performed by vascular surgeons in an endovascular suite with fixed imaging (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom). Patients in group 2 included patients with recalcitrant varicosities, venous stasis ulcerations, or symptoms of venous hypertension. The majority of patients had symptoms for greater than 6 months prior to being evaluated by a vascular surgeon. All patients had a venous ultrasound reflux study to evaluate for superficial or deep venous insufficiency. Patients were treated with superficial venous ablation as indicated and were recommended medical management (compression therapy, compression dressings, leg elevation). Patents that failed to improve were screened for iliac vein pathology by duplex ultrasound, computerized tomography venography, or magnetic
Hager et al 271
renonance venography. When iliac vein compression was noted, invasive venography with IVUS was performed to assess for the extent and degree of the stenosis. Thrombolysis and stenting technique. Thrombolysis was performed on patients in group 1 as previously described.12 In brief, patients were treated under conscious sedation and local anesthesia. In the setting of iliofemoral DVT, popliteal vein access was obtained in the prone position, using ultrasound guidance. Ascending venography was performed via a 6F sheath, and a heparin bolus was administered at a dose of 100 units/kg. Pharmacomechanical thrombolysis (PMT) with either the Trellis device (Covidien, Mansfield, Mass) or AngioJet (Bayer, Leverkusen, North Rhine-Westphalia, Germany) and tissue plasminogen activator was used initially in the setting of acute DVT that was #14 days old. Catheterdirected thrombolysis (CDT) was used initially for subacute thrombus (>14 days old) or residual thrombus after PMT with either a multi-side hole catheter or ultrasound-assisted CDT with the EKOS catheter (EKOS Corporation, Ottawa, Canada). Fibrinogen levels were monitored, and the tissue plasminogen activator drip rate adjusted when the value fell below 150 mg/dL. Patients with iliac vein stenosis as determined by venography or IVUS were treated with angioplasty and stenting. In patients with obvious severe stenosis on venography, IVUS was not obtained, and stenting was performed based on venographic data. IVUS was used in equivocal cases and most patients in group 2, and it allowed for accurate measurement of the area reduction and adequate stent sizing. Patients that had an area reduction of greater than 50% in the iliac vein were stented. The stent is splayed into the inferior vena cava by 1 to 2 cm unless the lesion is more than 2 cm from the left common iliac vein orifice, and there was no evidence by IVUS of either thrombosis/fibrosis/ thickening at the orifice.10 The stent diameter was typically oversized by 10% to 20%, and the length was chosen to ensure complete lesion coverage. Follow up. Follow-up clinical and duplex ultrasound evaluations were performed within 30 days postprocedure, at 6 months, and yearly thereafter. Patients with active venous stasis ulcers were followed on a weekly basis until their wounds were healed. Postprocedure, decisions about the type and duration of anticoagulation for patients in group 1 were made in conjunction with the hematology consultant based on the clinical presentation and hypercoagulable workup. Treatment consisted of aspirin, Plavix, and/or Coumadin. Patients in group 2 were treated with aspirin and Plavix for 3 months. All patients were given graduated compression stockings to wear. The international normalizing ratio was followed to ensure proper dosing of Coumadin, with a target goal between 2 and 3. Primary end points were improvement of symptoms and freedom from stent thrombosis. Secondary end points were valve reflux in the femoral and popliteal veins and treated vein patency. These end points were chosen to determine the efficacy of lysis in preventing venous reflux and to differentiate primary stent failure versus thrombosis
272 Hager et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS July 2013
of the inflow veins. The group’s demographics and comorbidities were also compared. The patient’s initial Villalta13 and CEAP scores were recorded and compared at the most recent follow-up. Statistical analysis. Demographic and procedural result data were analyzed using Fisher’s exact test for dichotomous variables and the Kruskal-Wallis equalityof-populations rank test for age, size of the stents used during the procedures, and CEAP class. These results were reported with median and interquartile ranges. CEAP class was also analyzed after dichotomization, with scores of greater than or equal to 3 being the dividing line. Primary stent patency rates were assessed using KaplanMeier survival estimates, and comparisons between groups 1 and 2 were assessed using the log-rank test for equality of survivor functions. The threshold for significance for all statistical analyses was .05. Statistical analysis was performed with Stata IC Version 11.2 (StataCorp LP, College Station, TX). RESULTS Patient population. During the 5.5 year period, there were 70 patients identified with MTS and treated with endovenous stenting, seven of whom had bilateral iliac stents placed (Table I). These patients were divided according to symptoms, with group 1 consisting of patients diagnosed with MTS after successful lysis of iliofemoral DVT and group 2 being comprised of patients with no DVT and treated superficial venous disease but recalcitrant swelling/pain or ulceration. The average age of patients was similar between groups (group 1, 51.9 years [range, 18-73 years] vs group 2, 52.8 years [range, 16-80 years]; P ¼ .440). Of the patients treated, 43 (61.4%) were younger than 50 years old. The majority of patients suffered from left-sided symptoms in both groups (group 1, 78.4% vs group 2, 78.8%; P ¼ 1.00). Compression therapy was more commonly used preoperatively in group 2 (37.5% vs 95.2%; P < .001). Combining both groups 1 and 2, female patients were more likely to
have left-sided symptoms compared with male patients (odds ratio, 4.88; 95% confidence interval, 1.49-15.89; P ¼ .014). The initial CEAP score was significantly lower in group 1 (P < .001), with an initial score of $3 in only 29% of patients. The Villalta score was significantly lower in group 1 (P < .001), with an average score of 16.3 versus 34.7 in group 2. Procedural details. The initial procedural details and outcomes of both groups after stenting are described in Table II. Access site was predicated on indication for treatment, and therefore, the use of distal access was more common in group 1 where iliofemoral DVT was present. The popliteal approach was utilized in 85.7% of patients in group 1 (P < .001), while the common femoral or thigh femoral vein (if patent) was accessed in 90.5% of patients in group 2 (P < .001). IVUS use was used more frequently in group 2 (P < .001), as this modality was only used in group 1 when venographic findings were questionable for the presence of iliac vein stenosis. There were no patients with more than one site of iliac vein compression, nor patients with iliac vein compression by the inguinal ligament. The number of stents was therefore predicated on the length of the stenosis and was not different between the groups. There were more recorded complications in group 1, with significant bleeding (1.8%), perioperative thrombosis (5.3%), and renal insufficiency (1.8%) being more common but not achieving statistical significance. Outcomes and follow-up. The median follow-up period was 29.7 months (range, 18.4-58.3 months) in group 1, with 39% of patients having a follow-up of greater than 3 years. The median follow-up of group 2 was 22.4 months (range, 17.1-42.0 months) with 29% having follow-up of greater than 3 years (Table III). Symptom resolution or improvement occurred in 52 of 56 (92.9%) of patients in group 1 and 20 of 21 (95.7%) of patients in group 2 at the latest follow-up. The average postprocedure Villalta score improved in both groups, to 6.7 in group 1 and 19.2 in group 2, both of which achieved significance when compared with their preoperative scores
Table I. Patient demographics and comorbidities Characteristics
Group 1 (n ¼ 51)
Group 2 (n ¼ 19)
P value
Median age, years (interquartile range) Female gendera Left-sided symptomsa Diabetes Malignancy Hypertension Hypercoagulable state identified among tested patientsb Coronary artery disease Hyperlipidemia Median CEAP score (interquartile range) Initial CEAP score greater than or equal to 3c Initial Villalta score
51.9 30 40 5 2 23 13 8 16 1 16
52.8 14 15 3 0 7 0 1 6 4 21
.440 .282 1.000 .674 .982 .596 .014 .427 1.000 <.001 <.001 <.001
(41.4-63.6) (58.8%) (78.4%) (9.8%) (3.9%) (45.1%) (26.5%) (15.7%) (31.4%) (0-4) (28.6%) 16.3
(46.2-62.1) (73.7%) (78.9%) (15.8%) (0%) (36.8%) (0.0%) (5.3%) (31.6%) (4-6) (100%) 34.7
a Combining both groups 1 and 2, female patients were more likely to have left-sided symptoms compared with male patients (odds ratio, 4.88; 95% confidence interval, 1.49-15.89; P ¼ .014). b Two patients in group 1 and eight patients in group 2 were not tested for a hypercoagulable state. c CEAP scores assessed for each limb separately. Group 1, 56 limbs. Group 2, 21 limbs.
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume 1, Number 3
Table II. Procedural details and initial outcomes Characteristics Preoperative use of compression stockings Approach Popliteal Femoral Other Intravenous ultrasound use Median stent size, mm (interquartile range) Self-expanding stent Protégé Wallstent Balloon-expandable stent Any postoperative complications Clinically significant bleeding Immediate thrombosis Renal insufficiency Need for blood transfusion
Group 1 Group 2 (56 limbs) (21 limbs) P value 21 (37.5%) 20 (95.2%) <.001 48 7 1 15 14
(85.7%) 0 (0.0%) <.001 (12.5%) 19 (90.5%) <.001 (1.8%) 2 (9.5%) .179 (26.8%) 19 (90.5%) <.001 (12-16) 16 (14-18) .027
50 14 36 6 8 1 3 1 1
(89.3%) 10 (47.6%) <.001 (25.0%) 10 (47.6%) .056 (64.3%) 0 (0.0%) <.001 (10.7%) 1 (4.8%) .211 (14.3%) 0 (0.0%) .099 (1.8%) 0 (0.0%) 1.000 (5.3%) 0 (0.0%) .670 (1.8%) 0 (0.0%) 1.000 (1.8%) 0 (0.0%) 1.000
(P ¼ .03). The median CEAP score was 1 in group 1 and 4 in group 2, unchanged from the preoperative assessment. Stent patency. Stent patency was similar in the followup period. Symptom relief seemed to correlate with stent patency, although there were two patients that had stent thrombosis resulting in mild symptoms only. Patients in group 1 had a primary patency rate of 91% and secondary patency of 98% at 36 months. There were five (8.9%) patients with in-stent thrombosis requiring reintervention in group 1. Three patients had rethombosis within 48 hours of the procedure, all of which were subtherapeutic on their intravenous anticoagulation. The two other patients that rethrombosed at later time points (1 week and 6 months) were on Coumadin but had subtherapeutic international normalizing ratios at the time of thrombosis. Both of these patients had factor V Leiden thrombophilia. All five of these patients had recurrence of symptoms and successfully underwent lysis of the thrombosed stent using either CDT or a combination of CDT and PMT. Of these five patients, four had Protégé (eV3 Endovascular, Plymouth, Minn) stents implanted, and one had a 14-mm balloon-expandable stent. Two had been stented with
Hager et al 273
smaller stents (12-mm Protégé) and were treated with a larger 16-mm balloon-expandable stent within the Protégé with good results. Since the time of the reintervention, only one patient has suffered from late thrombosis of the stent at 2 years. This patient elected not to have any intervention performed due to relatively mild symptoms. Patients in group 2 had a primary patency rate of 91% and secondary patency of 91% at 36 months. Group 2 had two (9.4%) stent thrombosis at 12 months. One patient had a 12-mm Protégé stent placed during the original surgery, which was most likely undersized. The patient elected not to undergo lysis due to a recent gastrointestinal bleed and was treated with anticoagulation only. The other patient’s stent was a 16-mm Wallstent that appeared to be placed too far distally, leaving proximal disease at the iliocaval junction uncovered. This patient elected not to have a reintervention due to minimal symptoms. Over the entire analysis period, there was no significant difference in patency by life-table analysis (P ¼ .612; Fig). Patients with stent failure presented with complete stent thrombosis, and there were no instances of in-stent stenosis identified in either group. DISCUSSION Stenting has become the primary treatment modality for MTS. Due to the heterogeneous symptomatology and age of patients with MTS, it is important to describe the long-term fate of the venous stents and the resultant symptom relief. The treatment of MTS patients who present with thrombosis has changed dramatically, with increased application of CDT. Catheter-based strategies have been utilized in the management of iliofemoral thrombosis, with particular emphasis on younger patients, given the high association with postthrombotic syndrome and its attendant morbidity.14 Mewissen and colleagues reported that a third of limbs treated with thrombolysis for DVT required stenting and that stented limbs had a significantly better patency than those that were not.15 Today, both the Society for Interventional Radiology and the Society of Vascular Surgery guidelines recommend iliac vein stenting in the setting of external compression.16,17 The long-term durability and efficacy of iliac venous stenting has been well described in the setting of DVT, with
Table III. Follow-up information Characteristics Median length of follow-up, months (range) Need for reintervention Symptom relief at last follow-up visit Moderate Complete Limitations with ambulation on last follow-up examination (pain/swelling) Recurrent or distal deep vein thrombosis visualized on last follow-up examination Use of compression stockings Preserved valve function CEAP class greater than or equal to 3 at last follow-up Median CEAP score (interquartile range) Villalta score
Group 1 (56 limbs)
Group 2 (21 limbs)
P value
29.7 (18.4-58.3) 5 (8.9%) 52 (92.9%) 21 (37.5%) 31 (55.4%) 17 (30.4%) 7 (12.5%) 51 (91.1%) 31 (55.4%) 29 (51.8%) 1 (0-3) 6.7
22.4 (17.1-42.0) 0 (0.0%) 20 (95.2%) 10 (47.6%) 10 (47.6%) 0 (0.0%) 1 (4.8%) 21 (100%) 1 (4.8%) 21 (100%) 4 (4-6) 19.2
NA .315 1.000 .445 .613 .004 .435 .315 <.001 <.001 <.001 <.001
274 Hager et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS July 2013
Fig. Kaplan-Meier estimates for primary stent patency.
a 79% primary patency rate at 72 months with nearly 100% primary-assisted patency.10 This demonstrated durability of venous stenting among a wide variety of patients; however, the same conclusions may not be applicable to MTS patients where ongoing external stent compression by the iliac artery is to be expected. Small case series of iliac vein stenting for MTS have been reported with greater frequency over the past decade and have advocated the use of endovascular techniques.18-20 Recently, a retrospective review, published by Titus and colleagues, identified 15 MTS patients treated by endovenous stenting and demonstrated a primary patency rate of 88%, 78%, and 78% at 6 months, 12 months, and 24 months, respectively.21 A more recent study of the long-term patency of stenting for MTS after overnight CDT and aspiration thrombectomy demonstrated an exceptional 83.3% and 90% primary and secondary patency at 1 and 5 years,22 and this finding has been seen in earlier studies at 1 year employing a similar management strategy.8,23 However, this series did not include a control group of nonthrombotic patients with MTS. The current data are concordant with previous reports of iliac vein stenting, demonstrating favorable patency rates. We describe the outcomes up to 58 months in two populations of patients with MTS: those presenting with iliofemoral DVT and found to have MTS after lysis, and those with severe symptoms associated with venous hypertension. We found both groups to have excellent primary stent patency at 30-month follow-up by life-table analysis. Patients from both groups experienced significant symptom relief that mirrored stent patency. These findings
add support to the hypothesis that iliac vein obstruction or stenosis and subsequent venous hypertension is the genesis of the majority of patients’ morbidity,8 underscoring the importance of stent patency to maintain clinical symptom relief. We did not identify age to be a risk factor for stent failure as suggested by other groups,24 although logic would dictate that the longer the life expectancy of the patient, the greater the risk of stent failure. As 61% of the patients in this study were younger than 50 years old, it is essential that these patients be surveyed routinely to assess for in-stent stenosis or thrombosis. It is interesting to note that patients in group 1 presented with a lower average CEAP score, higher Villalta score, and were less likely to wear graded compression stockings than those in group 2, which emphasizes the heterogeneity of the MTS patient population. We believe this is a result of the relative acuity of presentation and symptoms of the patients with thrombosis and the fact that many had no known antecedent venous disease. The patients represented in group 2 suffered from significant venous disease and, despite superficial vein ablation, continued to have severe symptoms as well as high Villalta scores. Stenting of the compressed venous system improved venous hypertension as is reflected by the improvement in both subjective symptoms and Villalta score. It is noteworthy, however, that in group 1, there were an increased number of patients with a CEAP score of 3 at follow-up. The majority of these patients suffered from edema, suggesting that although the majority of PTS symptoms can be mitigated, there is still significant sequelae associated with the initial thrombotic event.
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume 1, Number 3
Also noteworthy was the finding that the majority of patients with right-sided MTS were male. This finding may be due to the shape of the male pelvis. Although numerous articles have described the left-sided and female predominance of MTS,10 there has been no correlation between right-sided MTS and male gender. We postulate that the conical shape of the male pelvis may predispose the right iliac vein to compression by the iliac artery, although it is possible that MTS may have been overdiagnosed in some of these patients. One limitation of this study is the inability to identify the entire pool of patients who suffer from MTS, as some of them may have been treated conservatively. As such, it is difficult to make conclusions on all patients with MTS, and statistical analysis is limited by the lack of knowledge of the denominator. Further limitations of this study are those inherent to retrospective database review and include reporting and selection bias during data collection as well as missing data secondary to loss of follow-up. In addition, some patients were lost to follow-up, which makes the long-term data somewhat limited. Another potential limitation was the variability in stent type and size used, which was not uniform early on in our experience. Nevertheless, to our knowledge, this study represents one of the larger reports on the endovascular treatment and long-term outcomes of patients with MTS comparing thrombotic and nonthrombotic patients. These data support the durability and efficacy of endovenous stenting among symptomatic patients with MTS with thrombotic or nonthrombotic presentation. On long-term follow-up, stent patency does not seem to be affected by the ongoing extrinsic compression by the adjacent iliac artery. AUTHOR CONTRIBUTIONS Conception and design: EH, RC Analysis and interpretation: TY, EH, RC Data collection: EH, TY Writing the article: EH, RC, TY Critical revision of the article: ED, GAK, LM, MM Final approval of the article: RC, MM Statistical analysis: TY Obtained funding: Not applicable Overall responsibility: RC
5.
6.
7. 8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21.
REFERENCES 22. 1. Seidensticker D, Wilcox J, Gagne P. Treatment of May-Thurner syndrome with catheter-directed thrombolysis and stent placement, complicated by heparin-induced thrombocytopenia. Cardiovasc Surg 1998;6:607-13. 2. Suwanabol PA, Tefera G, Schwarze ML. Syndromes associated with the deep veins: phlegmasia Cerulea Dolens, May-Thurner Syndrome, and Nutcracker Syndrome. Perspect Vasc Surg Endovasc Ther 2010;22: 223-30. 3. Kibbe MR, Ujiki M, Goodwin AL, Eskandari M, Yao J, Matsumura J. Iliac vein compression in an asymptomatic patient population. J Vasc Surg 2004;39:937-43. 4. Murphy EH, Davis CM, Journeycake JM, DeMuth RP, Arko FR. Symptomatic iliofemoral DVT after onset of oral contraceptive use in
23.
24.
Hager et al 275
women with previously undiagnosed May-Thurner syndrome. J Vasc Surg 2009;49:697-703. Wolpert LM, Rahmani O, Stein B, Gallagher JJ, Drezner AD. Magnetic resonance venography in the diagnosis and management of May-Thurner syndrome. Vasc Endovasc Surg 2002;36:51-7. Vedantham S, Thorpe PE, Cardella JF, Grassi CJ, Patel NH, Ferral H, et al. Quality improvement guidelines for the treatment of lower extremity deep vein thrombosis with use of endovascular thrombus removal. J Vasc Interv Radiol 2006;17:435-7. Taheri SA, Williams J, Powell S. Iliocaval compression syndrome. Am J Surg 1987;154:169-72. O’Sullivan GJ, Semba CP, Bittner CA, Kee ST, Razavi MK, Sze DY, et al. Endovascular management of iliac vein compression (MayThurner) syndrome. J Vasc Interv Radiol 2000;11:823-36. Dhillon RK, Stead LG. Acute deep vein thrombosis due to MayThurner syndrome. Am J Emerg Med 2010;28:254. Neglén P, Hollis KC, Oliver 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;46:979-90. Oguzkurt L, Ozkan U, Tercan F, Koc Z. Ultrasonographic diagnosis of iliac vein compression (May-Thurner) syndrome. Diagn Interv Radiol 2007;13:152-5. Jeyabalan G, Marone L, Rhee R, Hirsch S, Makaroun MS, Cho JS, et al. Inflow thrombosis does not adversely affect thrombolysis outcomes of symptomatic iliofemoral deep vein thrombosis. J Vasc Surg 2011;54:448-53. Villalta S, Bagatella P, Piccioli A, Lensing A, Prins M, Prandoni P. Assessment of validity and reproducibility of a clinical scale for the postthrombotic syndrome. Haemostasis 1994;24:158a. Goldenberg NA, Durham JD, Knapp-Clevenger R, MancoJohnson MJ. A thrombolytic regimen for high-risk deep venous thrombosis may substantially reduce the risk of post-thrombotic syndrome in children. Blood 2007;110:45-53. 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;211:39-49. Vedantham S, Millward SF, Cardella JF, Hofmann LV, Razavi MK, Grassi CJ, et al. Society of Interventional Radiology position statement: treatment of acute iliofemoral deep vein thrombosis with use of adjunctive catheter-directed intrathrombus thrombolysis. J Vasc Interv Radiol 2006;17:613-6. Meissner MH, Gloviczki P, Comerota AJ, Dalsing MC, Eklof BG, Gillespie DL, et al. Early thrombus removal strategies for acute venous thrombosis: clinical practice guidelines of the Society for Vascular Surgery and the American Venous Forum. J Vasc Surg 2012;55:1449-62. Binkert CA, Schoch E, Stuckmann G, Largiader J, Wigger P, Schoepke W, et al. Treatment of pelvic venous spur (May-Thurner syndrome) with self-expanding metallic endoprostheses. Cardiovasc Intervent Radiol 1998;21:22-6. Heniford BT, Senler SO, Olsofka JM, Carrillo EH, Bergamini TM. May-Thurner syndrome: management by endovascular surgical techniques. Ann Vasc Surg 1998;12:482-6. Baron HC, Shams J, Wayne M. Iliac vein compression syndrome: a new method of treatment. Am Surg 2000;66:653-5. Titus JM, Moise MA, Bena J, Lyden SP, Clair DG. Iliofemoral stenting for venous occlusive disease. J Vasc Surg 2011;53:706-12. Jeon UB, Chung JW, Jae HJ, Kim HC, Kim SJ, Ha J, et al. MayThurner Syndrome complicated by acute iliofemoral vein thrombosis: helical CT venography for evaluation of long-term stent patency and changes in the iliac vein. AJR 2010;195:751-7. Lamont JP, Pearl GJ, Patetsios P, Warner MT, Gable DR, Garrett W, et al. Prospective evaluation of endoluminal venous stents in the treatment of the May-Thurner syndrome. Ann Vasc Surg 2002;16:61-4. Knipp BS, Ferguson E, Williams DM, Dasika NJ, Cwikiel W, Henke PK, Wakefield TW. Factors associated with outcome after interventional treatment of symptomatic iliac vein compression syndrome. J Vasc Surg 2007;46:743-9.
Submitted Aug 22, 2012; accepted Nov 25, 2012.