- Email: [email protected]
Duplex imaging of residual venous obstruction to guide duration of therapy for lower extremity deep venous thrombosis
Duplex imaging of residual venous obstruction to guide duration of therapy for lower extremity deep venous thrombosis Elliot J.P. Stephenson, MD, and Timothy K. Liem, MD, FACS, Portland, Ore Background: Clinical trials have shown that the presence of ultrasound-identified residual venous obstruction (RVO) on follow-up scanning may be associated with an elevated risk for recurrence, thus providing a potential tool to help determine the optimal duration of anticoagulant therapy. We performed a systematic review to evaluate the clinical utility of post-treatment duplex imaging in predicting venous thromboembolism (VTE) recurrence and in adjusting duration of anticoagulation. Methods: The Ovid MEDLINE Database, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Database of Abstracts of Reviews of Effects were queried for the terms residual thrombus or obstruction, duration of therapy, deep vein thrombosis, deep venous thrombosis, DVT, venous thromboembolism, VTE, antithrombotic therapy, and anticoagulation, and 228 studies were selected for review. Six studies determined the rate of VTE recurrence on the basis of the presence or absence of RVO.
Results: Findings on venous ultrasound scans frequently remained abnormal in 38% to 80% of patients, despite at least 3 months of therapeutic anticoagulation. In evaluating for VTE recurrence, the definition of RVO varied widely in the literature. Some studies have shown an association between RVO and VTE recurrence, whereas other studies have not. Overall, the presence of RVO is a mild risk factor for recurrence (odds ratio, 1.3-2.0), but only when surveillance imaging is performed soon after the index deep venous thrombosis (3 months). Conclusions: RVO is a mild risk factor for VTE recurrence. The presence or absence of ultrasound-identified RVO has a limited role in guiding the duration of therapeutic anticoagulation. Further research is needed to evaluate its utility relative to other known risk factors for VTE recurrence. (J Vasc Surg: Venous and Lym Dis 2014; -:1-7.)
The role of duplex imaging in the initial diagnosis of deep venous thrombosis (DVT) is well established, as is the role of anticoagulation for the prevention against pulmonary embolism and DVT recurrence. The American College of Chest Physicians Evidence-Based Clinical Practice Guidelines recommend 3 months of anticoagulation therapy in patients with a proximal DVT, provoked either by surgery (grade 1B) or by a nonsurgical transient risk factor (grade 1B).1 These guidelines recommend that patients with unprovoked proximal DVT should receive at least 3 months of anticoagulation. However, these patients also should be assessed for extended-duration anticoagulation, taking into consideration the individual risk for bleeding and recurrence. Numerous clinical risk factors and laboratory parameters have been evaluated as potential markers for risk of venous thromboembolism (VTE) recurrence. These include provoked vs unprovoked DVT, thrombus location,
malignant disease, male gender, age, presence of an inherited thrombophilia, concurrent hormonal therapy, elevated body mass index, and elevated D-dimer levels. In addition, multiple studies have evaluated whether residual venous obstruction (RVO) at the time of discontinuing anticoagulation was a potential risk factor for recurrence.2,3 These are the subject of the current review. The purpose of this manuscript was to conduct a systematic review of the role of RVO or thrombosis in determining the risk for recurrent VTE and the optimal duration of anticoagulation.
From the Division of Vascular Surgery, Knight Cardiovascular Institute, Oregon Health & Science University. Author conflict of interest: none. Reprint requests: Timothy K. Liem, MD, FACS, Division of Vascular Surgery, OP-11, Oregon Health & Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97221 (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 Copyright Ó 2014 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvsv.2014.08.003
METHODS With the assistance of a research librarian, we performed a systematic review of the literature, limiting the scope of this search to studies evaluating the association between RVO of the lower extremities and recurrent VTE. The Ovid MEDLINE Database, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Database of Abstracts of Reviews of Effects were queried for the following terms: residual thrombus or obstruction, RVO, duration of therapy, deep vein thrombosis, deep venous thrombosis, DVT, venous thromboembolism, VTE, antithrombotic therapy, anticoagulation. The search algorithm identified 228 manuscripts that were selected for abstract review. Studies were excluded if they were non-English language manuscripts, if the primary focus was venous thromboprophylaxis, or if RVO and recurrent thrombosis were not evaluated. By use of these exclusion criteria, the list of manuscripts was narrowed to 41 1
2 Stephenson and Liem
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
Fig 1. Search algorithm with resulting number of manuscripts. DVT, Deep venous thrombosis.
citations. Eighteen additional articles were excluded for the following reasons: 12 were review articles, four were abstracts, and two did not specifically address recurrence. The remaining 23 citations were reviewed, and 17 additional trials were excluded. Some included only provoked DVTs. Other trials and cohorts did not evaluate recurrence on the basis of venous duplex imaging at completion of anticoagulation. Six were selected for inclusion in the final analysis (Fig 1). The remaining six studies all included data on VTE recurrence in patients with and without RVO and included patients with unprovoked DVTs with or without provoked DVTs. Two of the six were randomized trials, and the remaining were prospectively observed cohorts. The trials were reviewed to determine the following data points: date of publication, number of patients, provoked or unprovoked, duration of anticoagulation, ultrasound method for determining RVO, percentage of patients with RVO, length of follow-up, recurrence rates. The data were compiled and statistics performed by Microsoft Excel (Redmond, Wash). Two additional manuscripts were not included in the summary table.2,3 One of these studies had insufficient data regarding VTE recurrence. The second study by Prandoni et al3 evaluated a customized anticoagulation regimen, depending on the persistence or clearance of venous thrombosis on subsequent duplex scans. However, the rate of VTE recurrence was not stratified on the basis of the presence or absence of RVO. Significant insights from both of these studies are described in the results sections. Definition of RVO. Multiple definitions of RVO have been described in the literature. The first was by Prandoni et al,4 based on measurement of the venous diameter under maximal compression. RVO was considered present if the diameter of the compressed vein was >2 mm when a single
Fig 2. Definition of residual vein thrombosis (RVT) in the DACUS and extended DACUS studies. CI, Confidence interval. Reprinted with permission from Siragusa S, Malato A, Saccullo G, Iorio A, Di Ianni M, Caracciolo C, et al. Residual vein thrombosis for assessing duration of anticoagulation after unprovoked deep vein thrombosis of the lower limbs: the extended DACUS study. Am J Hematol 2011; 86:914-7; permission conveyed through Copyright Clearance Center, Inc.
test was performed or >3 mm when two consecutive compression measurements were performed. Siragusa et al5 described the second definition. Venous duplex scanning was used to measure the diameter of the common femoral and popliteal veins (from 12 o’clock to 6 o’clock) before and during compression, as shown in Fig 2. A compressed diameter >40% of the noncompressed diameter was defined as positive for RVO. A third definition, used by Le Gal et al,6 attempted to stratify the degree of RVO into five categories: 1. Normal compression ultrasound; 2. Minimal wall thickening (#5% of maximal vein diameter); 3. Partial resolution ($40% of the diameter of the vein was compressible); 4. Minimal resolution (<40% of the vein being compressible); 5. Stable/worsened thrombus. Last, Poli et al7 combined the definitions from Siragusa and Prandoni, using either a compressed diameter of >40% of the noncompressed venous diameter or an absolute compressed venous diameter of >2 mm to define RVO. RESULTS Brief description of studies. The six manuscripts that met inclusion criteria for review are summarized in chronologic order. The first, by Cosmi et al,8 evaluated D-dimer level and RVO in a prospective cohort of patients with a first occurrence of unprovoked DVT. RVO was defined by the Prandoni criteria. All patients received at least 6 months of oral anticoagulation (OAC), and venous duplex imaging
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
was performed at the time that OAC was discontinued. In patients with suspected recurrent DVT, duplex imaging was used to confirm the diagnosis. The RVOþ group had a higher rate of recurrence (18.2% vs 14.9%), but this was not statistically significant. Poli et al7 prospectively observed a cohort of patients with a first episode of VTE. Patients received at least 3 months of OAC for VTE occurring in association with transient risk factors (provoked). Alternatively, they received at least 6 months of OAC when the lower extremity DVT was idiopathic. At the time of OAC cessation, venous ultrasound was used to assess for RVO. Patients were considered to have RVO if they met criteria by either the Siragusa or Prandoni definition. Patients were additionally screened for D-dimer and were observed for a median of 2 years. The rate of VTE recurrence was similar between the RVOþ and RVO groups (13.3% vs 11.8%), and this difference was not statistically significant. The Duration of Anticoagulation based on Compression UltraSonography (DACUS)5 study was a multiinstitutional trial at three centers in Italy. Patients with a first episode of provoked or unprovoked DVT had a follow-up venous duplex examination 3 months after initiation of anticoagulation. The Siragusa definition was used to establish RVO. Patients with RVO were randomized either to receive an additional 9 months of OAC (total of 12 months) or to discontinue OAC after 3 months. All patients without RVO stopped OAC after 3 months. The rate of recurrence in patients with RVO and 12 months of OAC was 19.3% vs 27.2% for patients with RVO and shorter duration OAC. In patients who were RVO, the recurrence rate was an impressively low 1.5%. These authors found a significant difference in recurrence between the RVOþ and RVO groups. The PROLONG9 investigators evaluated the predictive value of RVO and D-dimer for VTE recurrence. In a multiinstitutional trial of patients with a first diagnosis of idiopathic proximal DVT, all patients received $3 months of anticoagulation. Duplex assessment for RVO (with use of the Prandoni definition) was performed at the time of
Stephenson and Liem 3
OAC discontinuation, and D-dimer testing was performed 1 month afterward. Only patients who had an abnormal D-dimer level were randomized to resumption of anticoagulation or not. Abnormal D-dimer levels were significantly associated with DVT recurrence, but residual thrombosis was not. The rates of recurrence at 18 months were 11% for RVOþ patients and 13% for RVO patients. The extended DACUS study10 enrolled patients who received 3 months of OAC for an initial diagnosis of idiopathic DVT. Residual thrombosis was defined according to the Siragusa criteria, and those who were RVO discontinued anticoagulation (33.2% of total). RVOþ patients (66.8% of total) continued OAC for a period totaling 2 years, and both groups were assessed for bleeding or recurrence at 12 months after OAC withdrawal. VTE recurrence was much more common in the RVOþ group, despite the longer term anticoagulation (10.4% vs 1.4%; relative risk, 7.4; 95% confidence interval [CI], 4.9-9.9). The REVERSE6 cohort study was a multi-institutional study enrolling patients with a first episode of unprovoked DVT who had received 5 to 7 months of OAC. A baseline duplex examination was performed, and patients were observed for an additional 18 months (mean) without further OAC. Patients with more severe thrombophilias were excluded, and the REVERSE study used the Le Gal criteria, stratifying RVO into five categories. The findings on screening duplex examination, performed after 5 to 7 months of anticoagulation, were normal in 48.7%, abnormal in 51.1%, and inconclusive in 0.2%. There was no statistically significant difference in VTE recurrence in comparing the duplex-normal patients with the RVOþ patients (14.6% vs 19.5%, respectively; P ¼ .19). Interestingly, these authors reanalyzed their data after applying the Prandoni and Siragusa definitions of RVO to their cohort. The rates of VTE recurrence were not significantly different, regardless of which criteria were applied (Prandoni criteria, 19.7% vs 14.3%; Siragusa criteria, 18.8% vs 15.1%). Trials meeting inclusion criteria are summarized in Table I. The six manuscripts included in this review were
Table I. Studies in this analysis, including the provoked vs unprovoked nature of the index thrombotic event, the duration of antithrombotic therapy, and the definition used for residual venous obstruction (RVO) Author
Date
No. of patients
Provoked or unprovoked DVT
Cosmi8 Poli7
2005 2008
400 258
Unprovoked Both
Siragusa5 DACUS Cosmi9 PROLONG Siragusa10 Extended DACUS Le Gal6 REVERSE Total
2008
258
2010
Duration of anticoagulation
Ultrasound definition
% with RVO
Prandoni Siragusa or Prandoni
56.4 41.0
Both
6 months 3 months for provoked 6 months for idiopathic 3-12 months
Siragusa
69.8
490
Unprovoked
$3 months
Prandoni
38
2011
409
Unprovoked
3 months-2 years
Siragusa
66.8
2011
452
Unprovoked
5-7 months
Le Gal
51.1
DVT, Deep venous thrombosis.
2267
50.1
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
4 Stephenson and Liem
Table II. Average follow-up, overall rate of venous thromboembolism (VTE) recurrence, and recurrence based on residual venous obstruction (RVO) status Author
Date
Follow-up
Overall recurrence, %
Cosmi8 Poli7 Siragusa5 DACUS Cosmi9 PROLONG Siragusa10 Extended DACUS Le Gal6 REVERSE Total
2005 2008 2008
22 months 6 months 2 years
16.8 12.4 16.7
41/225 (18.2) 14/105 (13.3) 42/180 (23.3)
26/174 (14.9) 18/153 (11.8) 1/78 (1.3)
Not significant Not significant Significant
2010
18 months
12.3
17/151 (11.3)
32/246 (13.0)
Not significant
2011
3 years
7.4
27/258 (10.5)
2/136 (1.5)
2011
17 months
17.1
45/231 (19.5)
32/220 (14.5)
13.8
186/1150 (16.2)
111/1007 (11.0)
significantly heterogeneous with regard to trial design, patient populations, provoked vs unprovoked DVT, duration of anticoagulation, definition of RVO, and length of follow-up. The six trials in aggregate enrolled a total of 2267 patients, with the majority having a first-time unprovoked DVT. Duration of anticoagulation ranged from 3 months to 2 years, and three different definitions of RVO were used in the six manuscripts. Despite this, the aggregate number of patients with RVO was relatively consistent across most papers, ranging from 38% to 69.8% (mean, 50.1%). Overall VTE recurrence as well as recurrence rates based on RVOþ or RVO status is summarized in Table II. When all treatment groups were evaluated, the recurrence ranged from 1.3% to 27.2%, with an average of 13.8%. On a per trial basis, RVOþ groups had recurrence rates of 10.5% to 23.3%, with a mean of 16.2%. The discrepancy between the highest observed recurrence rate (27.2%) and highest rate described in Table II (23.3%) may be explained by closer analysis of the DACUS trial, in which the RVOþ patients were randomized to receive 3 months (25 of 92; 27.2% recurrence) vs 12 months of anticoagulation (17 of 88; 19.3% recurrence). Combining these two groups of RVOþ patients on a per trial basis, the mean recurrence rate was 23.3% (42 of 180).5 Across all six trials, patients without RVO had recurrence rates ranging from 1.3% to 14.9%, with a mean of 11.0%. The DACUS and extended DACUS trials also were notable for the very low observed rate of VTE recurrence (three of 214 patients; range, 1.3%-1.5%) in patients who were RVO.5,10 This was in significant contrast to the other four studies, which documented a VTE recurrence rate ranging from 11.8% to 14.9%.6-9 The DACUS and extended DACUS trials were the main drivers of statistically significant differences between RVOþ and RVO patients. In addition, the observed rate of VTE recurrence in RVOþ patients was significantly higher in the original DACUS trial compared with the extended DACUS trial. The most likely explanation is the longer duration of
Percentage recurrence in RVOþ, n/N (%)
Percentage recurrence in RVO, n/N (%)
Difference significant
Significant Not significant
anticoagulation used in the latter study, with observed recurrence rates of 27.2%, 19.3%, and 10.5% associated with OAC durations of 3, 12, and 24 months, respectively.5 Several additional studies deserve mention for this review. In a 2006 single-cohort study, 316 patients with lower extremity DVT were treated with 3 to 6 months of OAC, and follow-up venous duplex scans were repeated at 8 weeks to 1 year after diagnosis.2 Scans were retrospectively graded according to a 4-point scale. After a mean follow-up of 2.9 years, overall recurrence rates were 10% at 2 years and 23% at 5 years. Residual thrombus was an independent risk factor for recurrence (hazard ratio [HR], 2.2), and there was a trend toward increasing recurrence with increasing degree of residual thrombus. Because VTE recurrence rates were not detailed according to presence or absence of RVO, this study was excluded. The 2009 AESOPUS11 study evaluated the efficacy of flexible-duration anticoagulation on the basis of the presence of RVO on follow-up venous duplex scans; 538 patients with a first episode of provoked or unprovoked DVT were randomized to fixed-duration anticoagulation (3 months for secondary DVT and 6 months for unprovoked DVT, regardless of RVO status) vs flexible-duration anticoagulation. In the latter group, patients underwent serial duplex scans for up to 21 months and anticoagulation was discontinued when RVO was no longer seen, following the Prandoni criteria. Patients with secondary DVT extended the duration of OAC by an average of 3.7 months, and those with unprovoked DVT extended OAC by an average of 5 months, with a maximum duration of 21 months. Patients allocated to the flexible-duration protocol had less VTE recurrence (11.9% vs 17.2%; HR, 0.64; 95% CI, 0.390.99). Despite its relevance to residual thrombosis, this study was excluded because patients in the fixed-duration group were not stratified on the basis of RVO, and patients in the flexible anticoagulation group received OAC as long as they were RVOþ. Recent meta-analyses. Three recent meta-analyses have been performed. Carrier et al12 published the first of
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
these, including 14 randomized trials or prospectively studied cohorts. A total of 4022 patients were analyzed, with slightly more than half (2905; 52%) having symptomatic unprovoked DVTs. RVO was less common in patients with provoked DVT vs unprovoked DVT (30.6% vs 48.4%). This meta-analysis demonstrated that RVO was mildly predictive of recurrent VTE when patients with provoked and unprovoked DVT were evaluated together (odds ratio [OR], 1.5; 95% CI, 1.1-2.0). However, in evaluating only patients with unprovoked DVT who stopped OAC at the time of RVO assessment, there was no significant association between RVO and recurrence (OR, 1.24; 95% CI, 0.9-1.7). The authors described significant heterogeneity between studies with regard to duration of anticoagulant therapy, provoked vs unprovoked DVT, malignant disease, and definition of RVO. A second meta-analysis by Tan et al13 included data from 11 studies, totaling 3203 patients. Studies were included if patients had a first episode of proximal DVT, if thrombus diameter was measured at least 3 months after the initial diagnosis, and if VTE recurrence rates were objectively confirmed. These authors also identified an overall modest risk for VTE recurrence in patients with RVO (OR, 2.02; 95% CI, 1.62-2.50). However, this association was heterogeneous and dependent on numerous variables. Use of the Siragusa definition for RVO was associated with an OR of 13.68 (95% CI, 6.58-28.44), the presence of malignant disease was associated with an OR of 4.55 (95% CI, 1.76-11.79), and performing the RVO assessment at 3 months had an OR of 6.43 (95% CI, 3.90-10.62). Most recently, a patient-level meta-analysis was conducted by Donadini et al14 and many of the first authors of the major RVO manuscripts, limiting their scope to patients with a first unprovoked DVT who received $3 months of anticoagulation. Individual patient-level data (2527 patients, 10 data sets) were reanalyzed, and RVO was present in 1380 patients (55.1%). Overall VTE recurrence was seen in 399 patients (15.8%). By use of multivariate Cox analysis, the following variables were found to be significant predictors of VTE recurrence:
1. 2. 3.
RVO Age (for every 1-year increase) Male gender
Adjusted HR 1.32 (95% CI, 1.06-1.65) 1.01 (95% CI, 1.11-1.02) 1.49 (95% CI, 1.2-1.84)
However, RVO was predictive only if ultrasound assessment was made soon after the DVT (3 months) (HR, 2.17; 95% CI, 1.11-4.25) and not predictive if the duplex scan was performed >6 months after the index thrombotic event. DISCUSSION Surveillance venous duplex imaging, after the initial detection of lower extremity DVT, may serve multiple purposes. The first of these is to identify thrombus propagation
Stephenson and Liem 5
in patients who either have failed to respond to therapy or have not been able to receive therapeutic anticoagulation. The second purpose is to establish a new baseline, should there be any future concern for recurrent VTE. Testing for this purpose usually is performed when antithrombotic therapy is being discontinued.15 The third potential role for surveillance imaging is to detect the presence of RVO as a marker for patients at elevated risk for VTE recurrence. The natural history of lower extremity DVT is one of slow resolution. In patients who receive therapeutic anticoagulation, follow-up venography performed 6 months after diagnosis demonstrates complete lysis in 38.4%, partial lysis in 54.3%, and thrombus extension in 7.3%.16 The correlation between venographic and ultrasound determination of thrombus resolution is relatively close. Observational trials performing serial ultrasound examinations on patients with lower extremity DVT demonstrate normal venous duplex findings in 38.8% at 6 months, 58.1% at 12 months, 69.3% at 2 years, and 73.8% at 3 years.3 Continued resolution of DVT also has been demonstrated in other observational ultrasound studies.17,18 In evaluating the numerous randomized trials, the prospective cohort studies, and the more recent meta-analyses in aggregate, we conclude that RVO is a mild risk factor for VTE recurrence. The meta-analyses described in the Results section agree with regard to the presence of a mild association but reach differing conclusions with regard to patients with unprovoked DVTs. Carrier et al12 concluded that RVO was not a predictor of VTE recurrence with unprovoked DVT, whereas Donadini et al14 found a weak but significant association when selecting only patients with unprovoked DVT. The latter study did have the advantage of performing a patient-level meta-analysis of original data, thus providing more compelling evidence of a true association between RVO and VTE recurrence in patients with unprovoked DVT. These meta-analyses also agreed that the association with residual thrombosis was stronger when the duplex study was performed soon after the index DVT. Despite the mild association between residual thrombus and VTE recurrence, an evaluation for RVO at a single time point (whether at 3 months or 6 months) to make longer term decisions about extended-duration anticoagulation is likely to be of limited utility. Other more potent risk factors (male gender, D-dimer level, and VTE location) likely overshadow the recurrence associated with RVO. In addition, the RVO data are hampered by the significant heterogeneity of the various RVO-related studies. The definition for RVO was not standardized, the timing of the follow-up ultrasound study was somewhat variable, and some studies included patients with provoked and unprovoked DVT. In addition, the location of the index DVT was not consistently reported throughout the manuscripts. The results of the two studies by Siragusa et al5,10 showing the strongest association have not been well replicated in other studies. This may be in part due to the definition of RVO that was used in these manuscripts. In addition, patients were treated with varying
6 Stephenson and Liem
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
Table III. Published clinical scoring systems to predict risk for recurrent venous thromboembolism (VTE), applicable to unprovoked deep venous thrombosis (DVT) only Vienna Prediction Model23 Sex Location of first VTE D-dimer after anticoagulation
DASH score24
REVERSE study25 (female patients only)
D-dimer after anticoagulation Age <50 years Sex: male Hormone use at VTE onset
Post-thrombotic signs D-dimer Body mass index $30 kg/m2 Age $65 years
DASH, D-dimer, Age, Sex, Hormonal therapy.
durations of anticoagulation, and follow-up after discontinuation of anticoagulation also was variable. Alternatively, Prandoni et al11 have used serial venous duplex evaluation to tailor therapy on the basis of clearance of thrombus. These authors demonstrated a significant reduction in VTE recurrence when using the flexibleduration anticoagulant regimen compared with fixedduration therapy. In this study, patients in the former group received a mean of 3.7 months of additional OAC if their index DVT was secondary and a mean of 5.0 months if the DVT was unprovoked. The reduction in VTE recurrence was more substantial in patients with unprovoked DVT. However, it is possible that the benefit derived may have been due to the longer term anticoagulation alone, irrespective of the RVO status. An additional question that remains to be answered is whether the benefit of flexible-duration therapy persists after discontinuation of anticoagulation. Numerous other risk factors for VTE recurrence have been evaluated as well. These include hypercoagulable conditions, provoked vs unprovoked setting, VTE location, male gender, advanced age, elevated D-dimer levels, concurrent hormonal therapy, concurrent malignant disease, and obesity (body mass index $30 kg/m2).19 The presence of inherited or acquired thrombophilias is an independent predictor for the persistence of residual thrombus.20,21 However, there is conflicting evidence regarding their association with VTE recurrence. Potential explanations for these discrepancies may be related to the type of thrombophilia in question or the duration of anticoagulation.22 The more common thrombophilias (prothrombin 20210A and factor V Leiden) may confer a slight risk for VTE recurrence, but their clinical relevance is likely to be low in comparison with other clinical and laboratory predictors. D-dimer testing was evaluated in three of the six papers in this review.7-9 Cosmi et al8 demonstrated that although RVO was not a risk for VTE recurrence, an abnormal Ddimer test result, 1 month after withdrawal of OAC, was an independent risk factor for recurrence (HR, 3.32; 95% CI, 1.78-6.75; P < .0001). Poli et al7 showed similar results with D-dimer testing 4 weeks after OAC, when D-dimer positivity was a significant predictor of recurrence (HR, 3.2; 95% CI, 1.5-6.6; P < .002). Last, Cosmi et al9 adjusted OAC on the basis of D-dimer testing. Patients who had positive D-dimer test results after completion of OAC were randomized either to resume or to remain without OAC. The recurrence risk for the patients with a normal D-dimer level was 10% vs 19% for patients
with a positive D-dimer test result (P ¼ .002). However, D-dimer-positive patients who were randomized to receive continued anticoagulation had an even lower recurrence rate of 4.9%. These studies provide relatively consistent evidence that an abnormal result of D-dimer testing, when it is performed about 4 weeks after withdrawal of anticoagulation, is a more reliable indicator for VTE recurrence. Besides thrombophilia and D-dimer testing, several other demographic, clinical, and laboratory markers have a demonstrated association with VTE recurrence, including male gender, age, presence of post-thrombotic syndrome, obesity, VTE location, and concurrent hormonal therapy. Several of these have been incorporated into various clinical scoring systems to predict risk for recurrent VTE (Vienna Prediction Model, DASH [D-dimer, Age, Sex, Hormonal therapy] score, and REVERSE study)23-25 (Table III). These prediction scores remain to be further validated and are applicable only to patients with unprovoked VTE. However, they are easy to use and are widely available through the Internet. Notably, none of these scoring systems incorporates RVO as a predictor of VTE recurrence. In addition, none of these scoring systems provides discrete guidance for the continuation of OAC after the initial course is completed. Rather, they calculate a more individualized risk for VTE recurrence that must be integrated with the patient’s preference, the estimated risk for bleeding, the patient’s comorbidities, and the cost. Although the primary focus of this review was to analyze the association between RVO and recurrent VTE, it is important to recognize the frequently encountered diagnostic dilemma posed by the patient with an established DVT who presents with recurrent or worsening symptoms of unilateral leg pain or edema. Chronic venous thrombosis usually is characterized by increased echogenicity and a contracted and thickened vein wall. However, differentiating between RVO and recurrent ipsilateral DVT can occasionally be challenging, and duplex imaging after completion of anticoagulation has been shown to improve the accuracy for diagnosis of VTE recurrence. In a large cohort study by Hamadah et al,15 patients with a first unprovoked DVT received 5 to 7 months of OAC. There were 121 patients in whom VTE recurrence was suspected. Those who received a baseline venous duplex study after completion of anticoagulation had better interobserver agreement with regard to the duplex diagnosis of recurrence compared with patients who did not receive baseline imaging.
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
CONCLUSIONS In patients with unprovoked DVT, ultrasound determination of RVO is mildly predictive of VTE recurrence, but only when the venous duplex scan is performed soon after the index thrombotic event (3 months). Duplex assessment for residual thrombus is of limited usefulness when it is performed to guide the duration of therapeutic anticoagulation. Much of this is related to significant study heterogeneity with nonstandardized definitions for residual obstruction and varying durations of anticoagulation and clinical follow-up. Other clinical risk factors and laboratory parameters are likely to be of greater utility in predicting risk of VTE recurrence in patients with unprovoked lower extremity DVT (male gender, elevated D-dimer level, VTE location). These other variables have been incorporated into multiple clinical scoring systems and are likely to be better predictors for VTE recurrence. However, followup venous duplex surveillance at the time of anticoagulation withdrawal still is worthwhile, potentially helping to differentiate between RVO and future recurrent ipsilateral DVT. AUTHOR CONTRIBUTIONS Conception and design: TL, ES Analysis and interpretation: TL, ES Data collection: TL, ES Writing the article: TL, ES Critical revision of the article: TL, ES Final approval of the article: TL, ES Statistical analysis: TL, ES Obtained funding: Not applicable Overall responsibility: TL REFERENCES 1. Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, et al. Antithrombotic therapy for VTE disease. Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(Suppl):e419S-94S. 2. Young L, Ockelford P, Milne D, Rolfe-Vyson V, Mckelvie S, Harper P. Post-treatment residual thrombus increases the risk of recurrent deep vein thrombosis and mortality. J Thromb Haemost 2006;4:1919-24. 3. Prandoni P, Lensing A, Prins M, Bernardi E, Marchiori A, Bagatella P, et al. Residual venous thrombosis as a predictive factor of recurrent venous thromboembolism. Ann Intern Med 2002;137:955-60. 4. Prandoni P, Cogo A, Bernardi E, Villalta S, Polistena P, Simioni P, et al. A simple approach for detection of recurrent proximal vein thrombosis. Circulation 1993;88:1730-5. 5. Siragusa S, Malato A, Anastasio R, Cigna V, Milio G, Amato C, et al. Residual vein thrombosis to establish duration of anticoagulation after a first episode of deep vein thrombosis: the Duration of Anticoagulation based on Compression UltraSonography (DACUS) study. Blood 2008;112:511-5. 6. Le Gal G, Carrier M, Kovacs M, Betancourt M, Kahn S, Wells P, et al. Residual vein obstruction as a predictor for recurrent thromboembolic events after a first unprovoked episode: data from the REVERSE cohort study. J Thromb Haemost 2011;9:1126-32. 7. Poli D, Antonucci E, Ciuti G, Abbate R, Prisco D. Combination of Ddimer, F1þ2 and residual vein obstruction as predictors of VTE recurrence in patients with first VTE episode after OAT withdrawal. J Thromb Haemost 2008;6:708-10.
Stephenson and Liem 7
8. Cosmi B, Legnani C, Cini M, Guazzaloca G, Palareti G. The role of Ddimer and residual venous obstruction in recurrence of venous thromboembolism after anticoagulation withdrawal in cancer patients. Haematologica 2005;90:713-5. 9. Cosmi B, Legnani C, Iorio A, Pengo V, Ghirarduzzi A, Testa S, et al. Residual venous obstruction, alone and in combination with D-dimer, as a risk factor for recurrence after anticoagulation withdrawal following a first idiopathic deep vein thrombosis in the PROLONG study. Eur J Vasc Endovasc Surg 2010;39:356-65. 10. Siragusa S, Malato A, Saccullo G, Iorio A, Di Ianni M, Caracciolo C, et al. Residual vein thrombosis for assessing duration of anticoagulation after unprovoked deep vein thrombosis of the lower limbs: the extended DACUS study. Am J Hematol 2011;86:914-7. 11. Prandoni P, Prins MH, Lensing AW, Ghirarduzzi A, Agena W, Imberti D, et al. Residual thrombosis on ultrasonography to guide the duration of anticoagulation in patients with deep venous thrombosis: a randomized trial. Ann Intern Med 2009;150:577-85. 12. Carrier M, Rodger MA, Wells PS, Righini M, Le Gal G. Residual vein obstruction to predict the risk of recurrent venous thromboembolism in patients with deep vein thrombosis: a systematic review and metaanalysis. J Thromb Haemost 2011;9:1119-25. 13. Tan M, Mos IC, Klok FA, Huisman MV. Residual venous thrombosis as predictive factor for recurrent venous thromboembolim in patients with proximal deep vein thrombosis: a systematic review. Br J Haematol 2011;153:168-78. 14. Donadini M, Ageno W, Antonucci E, Cosmi B, Kovacs M, Le Gal G, et al. Prognostic significance of residual venous obstruction in patients with treated unprovoked deep vein thrombosis. A patient-level metaanalysis. Thromb Haemost 2014;111:172-9. 15. Hamadah A, Alwasaidi T, Le Gal G, Carrier M, Wells PS, Scarvelis D, et al. Baseline imaging after therapy for unprovoked venous thromboembolism: a randomized controlled comparison of baseline imaging for diagnosis of suspected recurrence. J Thromb Haemost 2011;9: 2406-10. 16. Holmström M, Lindmarker P, Grangqvist S, Johnsson H, Kockner D. A 6-month venographic follow-up in 164 patients with acute deep vein thrombosis. Thromb Haemost 1997;78:803-7. 17. Piovella F, Crippa L, Barone M, Vigano D’Angelo S, Serafini S, Galli L, et al. Normalization rates of compression ultrasonography in patients with a first episode of deep vein thrombosis of the lower limbs: association with DVT recurrence and new thrombosis. Haematologica 2002;87:515-22. 18. Ageno W, Steidl L, Piantanida E, Dentali F, Mera V, Squizzato A, et al. Predictors of residual vein obstruction after deep vein thrombosis of the lower limbs: a prospective study. Thromb Res 2003;108:203-7. 19. Kyrle PA, Rosendall FR, Eichinger S. Risk assessment for recurrent venous thrombosis. Lancet 2010;376:2032-9. 20. Spiezia L, Tormene D, Pesavento R, Salmaso L, Simioni P, Prandoni P. Thrombophilia as a predictor of persistent residual vein thrombosis. Haematologica 2008;93:479-80. 21. Cosmi B, Legnani C, Pengo V, Ghirarduzzi A, Testa S, Poli D, et al. The influence of factor V Leiden and G20210A prothrombin mutation on the presence of residual vein obstruction after idiopathic deep vein thrombosis of the lower limbs. Thromb Haemost 2013;109:510-6. 22. Prandoni P, Tormene D, Spiezia L, Pesavento R, Simioni P. Duration of anticoagulation and risk of recurrent thromboembolism in carriers of factor V Leiden or prothrombin mutation. J Thromb Haemost 2008;6: 2223-4. 23. Kyrle PA, Eichinger S. Clinical scores to predict recurrence of venous thromboembolism. Thromb Haemost 2012;108:1061-4. 24. Tosetto A, Iorio A, Marcucci M, Baglin T, Cushman M, Eichinger S, et al. Predicting disease recurrence in patients with previous unprovoked venous thromboembolism: a proposed prediction score (DASH). J Thromb Haemost 2012;10:1019-25. 25. Rodger MA, Kahn SR, Wells PS, Anderson DA, Chagnon I, Le Gal G, et al. Identifying unprovoked thromboembolism patients at low risk for recurrence who can discontinue anticoagulant therapy. CMAJ 2008;179:417-26.
Submitted Feb 24, 2014; accepted Aug 13, 2014.
Results: Findings on venous ultrasound scans frequently remained abnormal in 38% to 80% of patients, despite at least 3 months of therapeutic anticoagulation. In evaluating for VTE recurrence, the definition of RVO varied widely in the literature. Some studies have shown an association between RVO and VTE recurrence, whereas other studies have not. Overall, the presence of RVO is a mild risk factor for recurrence (odds ratio, 1.3-2.0), but only when surveillance imaging is performed soon after the index deep venous thrombosis (3 months). Conclusions: RVO is a mild risk factor for VTE recurrence. The presence or absence of ultrasound-identified RVO has a limited role in guiding the duration of therapeutic anticoagulation. Further research is needed to evaluate its utility relative to other known risk factors for VTE recurrence. (J Vasc Surg: Venous and Lym Dis 2014; -:1-7.)
The role of duplex imaging in the initial diagnosis of deep venous thrombosis (DVT) is well established, as is the role of anticoagulation for the prevention against pulmonary embolism and DVT recurrence. The American College of Chest Physicians Evidence-Based Clinical Practice Guidelines recommend 3 months of anticoagulation therapy in patients with a proximal DVT, provoked either by surgery (grade 1B) or by a nonsurgical transient risk factor (grade 1B).1 These guidelines recommend that patients with unprovoked proximal DVT should receive at least 3 months of anticoagulation. However, these patients also should be assessed for extended-duration anticoagulation, taking into consideration the individual risk for bleeding and recurrence. Numerous clinical risk factors and laboratory parameters have been evaluated as potential markers for risk of venous thromboembolism (VTE) recurrence. These include provoked vs unprovoked DVT, thrombus location,
malignant disease, male gender, age, presence of an inherited thrombophilia, concurrent hormonal therapy, elevated body mass index, and elevated D-dimer levels. In addition, multiple studies have evaluated whether residual venous obstruction (RVO) at the time of discontinuing anticoagulation was a potential risk factor for recurrence.2,3 These are the subject of the current review. The purpose of this manuscript was to conduct a systematic review of the role of RVO or thrombosis in determining the risk for recurrent VTE and the optimal duration of anticoagulation.
From the Division of Vascular Surgery, Knight Cardiovascular Institute, Oregon Health & Science University. Author conflict of interest: none. Reprint requests: Timothy K. Liem, MD, FACS, Division of Vascular Surgery, OP-11, Oregon Health & Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97221 (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 Copyright Ó 2014 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvsv.2014.08.003
METHODS With the assistance of a research librarian, we performed a systematic review of the literature, limiting the scope of this search to studies evaluating the association between RVO of the lower extremities and recurrent VTE. The Ovid MEDLINE Database, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Database of Abstracts of Reviews of Effects were queried for the following terms: residual thrombus or obstruction, RVO, duration of therapy, deep vein thrombosis, deep venous thrombosis, DVT, venous thromboembolism, VTE, antithrombotic therapy, anticoagulation. The search algorithm identified 228 manuscripts that were selected for abstract review. Studies were excluded if they were non-English language manuscripts, if the primary focus was venous thromboprophylaxis, or if RVO and recurrent thrombosis were not evaluated. By use of these exclusion criteria, the list of manuscripts was narrowed to 41 1
2 Stephenson and Liem
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
Fig 1. Search algorithm with resulting number of manuscripts. DVT, Deep venous thrombosis.
citations. Eighteen additional articles were excluded for the following reasons: 12 were review articles, four were abstracts, and two did not specifically address recurrence. The remaining 23 citations were reviewed, and 17 additional trials were excluded. Some included only provoked DVTs. Other trials and cohorts did not evaluate recurrence on the basis of venous duplex imaging at completion of anticoagulation. Six were selected for inclusion in the final analysis (Fig 1). The remaining six studies all included data on VTE recurrence in patients with and without RVO and included patients with unprovoked DVTs with or without provoked DVTs. Two of the six were randomized trials, and the remaining were prospectively observed cohorts. The trials were reviewed to determine the following data points: date of publication, number of patients, provoked or unprovoked, duration of anticoagulation, ultrasound method for determining RVO, percentage of patients with RVO, length of follow-up, recurrence rates. The data were compiled and statistics performed by Microsoft Excel (Redmond, Wash). Two additional manuscripts were not included in the summary table.2,3 One of these studies had insufficient data regarding VTE recurrence. The second study by Prandoni et al3 evaluated a customized anticoagulation regimen, depending on the persistence or clearance of venous thrombosis on subsequent duplex scans. However, the rate of VTE recurrence was not stratified on the basis of the presence or absence of RVO. Significant insights from both of these studies are described in the results sections. Definition of RVO. Multiple definitions of RVO have been described in the literature. The first was by Prandoni et al,4 based on measurement of the venous diameter under maximal compression. RVO was considered present if the diameter of the compressed vein was >2 mm when a single
Fig 2. Definition of residual vein thrombosis (RVT) in the DACUS and extended DACUS studies. CI, Confidence interval. Reprinted with permission from Siragusa S, Malato A, Saccullo G, Iorio A, Di Ianni M, Caracciolo C, et al. Residual vein thrombosis for assessing duration of anticoagulation after unprovoked deep vein thrombosis of the lower limbs: the extended DACUS study. Am J Hematol 2011; 86:914-7; permission conveyed through Copyright Clearance Center, Inc.
test was performed or >3 mm when two consecutive compression measurements were performed. Siragusa et al5 described the second definition. Venous duplex scanning was used to measure the diameter of the common femoral and popliteal veins (from 12 o’clock to 6 o’clock) before and during compression, as shown in Fig 2. A compressed diameter >40% of the noncompressed diameter was defined as positive for RVO. A third definition, used by Le Gal et al,6 attempted to stratify the degree of RVO into five categories: 1. Normal compression ultrasound; 2. Minimal wall thickening (#5% of maximal vein diameter); 3. Partial resolution ($40% of the diameter of the vein was compressible); 4. Minimal resolution (<40% of the vein being compressible); 5. Stable/worsened thrombus. Last, Poli et al7 combined the definitions from Siragusa and Prandoni, using either a compressed diameter of >40% of the noncompressed venous diameter or an absolute compressed venous diameter of >2 mm to define RVO. RESULTS Brief description of studies. The six manuscripts that met inclusion criteria for review are summarized in chronologic order. The first, by Cosmi et al,8 evaluated D-dimer level and RVO in a prospective cohort of patients with a first occurrence of unprovoked DVT. RVO was defined by the Prandoni criteria. All patients received at least 6 months of oral anticoagulation (OAC), and venous duplex imaging
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
was performed at the time that OAC was discontinued. In patients with suspected recurrent DVT, duplex imaging was used to confirm the diagnosis. The RVOþ group had a higher rate of recurrence (18.2% vs 14.9%), but this was not statistically significant. Poli et al7 prospectively observed a cohort of patients with a first episode of VTE. Patients received at least 3 months of OAC for VTE occurring in association with transient risk factors (provoked). Alternatively, they received at least 6 months of OAC when the lower extremity DVT was idiopathic. At the time of OAC cessation, venous ultrasound was used to assess for RVO. Patients were considered to have RVO if they met criteria by either the Siragusa or Prandoni definition. Patients were additionally screened for D-dimer and were observed for a median of 2 years. The rate of VTE recurrence was similar between the RVOþ and RVO groups (13.3% vs 11.8%), and this difference was not statistically significant. The Duration of Anticoagulation based on Compression UltraSonography (DACUS)5 study was a multiinstitutional trial at three centers in Italy. Patients with a first episode of provoked or unprovoked DVT had a follow-up venous duplex examination 3 months after initiation of anticoagulation. The Siragusa definition was used to establish RVO. Patients with RVO were randomized either to receive an additional 9 months of OAC (total of 12 months) or to discontinue OAC after 3 months. All patients without RVO stopped OAC after 3 months. The rate of recurrence in patients with RVO and 12 months of OAC was 19.3% vs 27.2% for patients with RVO and shorter duration OAC. In patients who were RVO, the recurrence rate was an impressively low 1.5%. These authors found a significant difference in recurrence between the RVOþ and RVO groups. The PROLONG9 investigators evaluated the predictive value of RVO and D-dimer for VTE recurrence. In a multiinstitutional trial of patients with a first diagnosis of idiopathic proximal DVT, all patients received $3 months of anticoagulation. Duplex assessment for RVO (with use of the Prandoni definition) was performed at the time of
Stephenson and Liem 3
OAC discontinuation, and D-dimer testing was performed 1 month afterward. Only patients who had an abnormal D-dimer level were randomized to resumption of anticoagulation or not. Abnormal D-dimer levels were significantly associated with DVT recurrence, but residual thrombosis was not. The rates of recurrence at 18 months were 11% for RVOþ patients and 13% for RVO patients. The extended DACUS study10 enrolled patients who received 3 months of OAC for an initial diagnosis of idiopathic DVT. Residual thrombosis was defined according to the Siragusa criteria, and those who were RVO discontinued anticoagulation (33.2% of total). RVOþ patients (66.8% of total) continued OAC for a period totaling 2 years, and both groups were assessed for bleeding or recurrence at 12 months after OAC withdrawal. VTE recurrence was much more common in the RVOþ group, despite the longer term anticoagulation (10.4% vs 1.4%; relative risk, 7.4; 95% confidence interval [CI], 4.9-9.9). The REVERSE6 cohort study was a multi-institutional study enrolling patients with a first episode of unprovoked DVT who had received 5 to 7 months of OAC. A baseline duplex examination was performed, and patients were observed for an additional 18 months (mean) without further OAC. Patients with more severe thrombophilias were excluded, and the REVERSE study used the Le Gal criteria, stratifying RVO into five categories. The findings on screening duplex examination, performed after 5 to 7 months of anticoagulation, were normal in 48.7%, abnormal in 51.1%, and inconclusive in 0.2%. There was no statistically significant difference in VTE recurrence in comparing the duplex-normal patients with the RVOþ patients (14.6% vs 19.5%, respectively; P ¼ .19). Interestingly, these authors reanalyzed their data after applying the Prandoni and Siragusa definitions of RVO to their cohort. The rates of VTE recurrence were not significantly different, regardless of which criteria were applied (Prandoni criteria, 19.7% vs 14.3%; Siragusa criteria, 18.8% vs 15.1%). Trials meeting inclusion criteria are summarized in Table I. The six manuscripts included in this review were
Table I. Studies in this analysis, including the provoked vs unprovoked nature of the index thrombotic event, the duration of antithrombotic therapy, and the definition used for residual venous obstruction (RVO) Author
Date
No. of patients
Provoked or unprovoked DVT
Cosmi8 Poli7
2005 2008
400 258
Unprovoked Both
Siragusa5 DACUS Cosmi9 PROLONG Siragusa10 Extended DACUS Le Gal6 REVERSE Total
2008
258
2010
Duration of anticoagulation
Ultrasound definition
% with RVO
Prandoni Siragusa or Prandoni
56.4 41.0
Both
6 months 3 months for provoked 6 months for idiopathic 3-12 months
Siragusa
69.8
490
Unprovoked
$3 months
Prandoni
38
2011
409
Unprovoked
3 months-2 years
Siragusa
66.8
2011
452
Unprovoked
5-7 months
Le Gal
51.1
DVT, Deep venous thrombosis.
2267
50.1
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
4 Stephenson and Liem
Table II. Average follow-up, overall rate of venous thromboembolism (VTE) recurrence, and recurrence based on residual venous obstruction (RVO) status Author
Date
Follow-up
Overall recurrence, %
Cosmi8 Poli7 Siragusa5 DACUS Cosmi9 PROLONG Siragusa10 Extended DACUS Le Gal6 REVERSE Total
2005 2008 2008
22 months 6 months 2 years
16.8 12.4 16.7
41/225 (18.2) 14/105 (13.3) 42/180 (23.3)
26/174 (14.9) 18/153 (11.8) 1/78 (1.3)
Not significant Not significant Significant
2010
18 months
12.3
17/151 (11.3)
32/246 (13.0)
Not significant
2011
3 years
7.4
27/258 (10.5)
2/136 (1.5)
2011
17 months
17.1
45/231 (19.5)
32/220 (14.5)
13.8
186/1150 (16.2)
111/1007 (11.0)
significantly heterogeneous with regard to trial design, patient populations, provoked vs unprovoked DVT, duration of anticoagulation, definition of RVO, and length of follow-up. The six trials in aggregate enrolled a total of 2267 patients, with the majority having a first-time unprovoked DVT. Duration of anticoagulation ranged from 3 months to 2 years, and three different definitions of RVO were used in the six manuscripts. Despite this, the aggregate number of patients with RVO was relatively consistent across most papers, ranging from 38% to 69.8% (mean, 50.1%). Overall VTE recurrence as well as recurrence rates based on RVOþ or RVO status is summarized in Table II. When all treatment groups were evaluated, the recurrence ranged from 1.3% to 27.2%, with an average of 13.8%. On a per trial basis, RVOþ groups had recurrence rates of 10.5% to 23.3%, with a mean of 16.2%. The discrepancy between the highest observed recurrence rate (27.2%) and highest rate described in Table II (23.3%) may be explained by closer analysis of the DACUS trial, in which the RVOþ patients were randomized to receive 3 months (25 of 92; 27.2% recurrence) vs 12 months of anticoagulation (17 of 88; 19.3% recurrence). Combining these two groups of RVOþ patients on a per trial basis, the mean recurrence rate was 23.3% (42 of 180).5 Across all six trials, patients without RVO had recurrence rates ranging from 1.3% to 14.9%, with a mean of 11.0%. The DACUS and extended DACUS trials also were notable for the very low observed rate of VTE recurrence (three of 214 patients; range, 1.3%-1.5%) in patients who were RVO.5,10 This was in significant contrast to the other four studies, which documented a VTE recurrence rate ranging from 11.8% to 14.9%.6-9 The DACUS and extended DACUS trials were the main drivers of statistically significant differences between RVOþ and RVO patients. In addition, the observed rate of VTE recurrence in RVOþ patients was significantly higher in the original DACUS trial compared with the extended DACUS trial. The most likely explanation is the longer duration of
Percentage recurrence in RVOþ, n/N (%)
Percentage recurrence in RVO, n/N (%)
Difference significant
Significant Not significant
anticoagulation used in the latter study, with observed recurrence rates of 27.2%, 19.3%, and 10.5% associated with OAC durations of 3, 12, and 24 months, respectively.5 Several additional studies deserve mention for this review. In a 2006 single-cohort study, 316 patients with lower extremity DVT were treated with 3 to 6 months of OAC, and follow-up venous duplex scans were repeated at 8 weeks to 1 year after diagnosis.2 Scans were retrospectively graded according to a 4-point scale. After a mean follow-up of 2.9 years, overall recurrence rates were 10% at 2 years and 23% at 5 years. Residual thrombus was an independent risk factor for recurrence (hazard ratio [HR], 2.2), and there was a trend toward increasing recurrence with increasing degree of residual thrombus. Because VTE recurrence rates were not detailed according to presence or absence of RVO, this study was excluded. The 2009 AESOPUS11 study evaluated the efficacy of flexible-duration anticoagulation on the basis of the presence of RVO on follow-up venous duplex scans; 538 patients with a first episode of provoked or unprovoked DVT were randomized to fixed-duration anticoagulation (3 months for secondary DVT and 6 months for unprovoked DVT, regardless of RVO status) vs flexible-duration anticoagulation. In the latter group, patients underwent serial duplex scans for up to 21 months and anticoagulation was discontinued when RVO was no longer seen, following the Prandoni criteria. Patients with secondary DVT extended the duration of OAC by an average of 3.7 months, and those with unprovoked DVT extended OAC by an average of 5 months, with a maximum duration of 21 months. Patients allocated to the flexible-duration protocol had less VTE recurrence (11.9% vs 17.2%; HR, 0.64; 95% CI, 0.390.99). Despite its relevance to residual thrombosis, this study was excluded because patients in the fixed-duration group were not stratified on the basis of RVO, and patients in the flexible anticoagulation group received OAC as long as they were RVOþ. Recent meta-analyses. Three recent meta-analyses have been performed. Carrier et al12 published the first of
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
these, including 14 randomized trials or prospectively studied cohorts. A total of 4022 patients were analyzed, with slightly more than half (2905; 52%) having symptomatic unprovoked DVTs. RVO was less common in patients with provoked DVT vs unprovoked DVT (30.6% vs 48.4%). This meta-analysis demonstrated that RVO was mildly predictive of recurrent VTE when patients with provoked and unprovoked DVT were evaluated together (odds ratio [OR], 1.5; 95% CI, 1.1-2.0). However, in evaluating only patients with unprovoked DVT who stopped OAC at the time of RVO assessment, there was no significant association between RVO and recurrence (OR, 1.24; 95% CI, 0.9-1.7). The authors described significant heterogeneity between studies with regard to duration of anticoagulant therapy, provoked vs unprovoked DVT, malignant disease, and definition of RVO. A second meta-analysis by Tan et al13 included data from 11 studies, totaling 3203 patients. Studies were included if patients had a first episode of proximal DVT, if thrombus diameter was measured at least 3 months after the initial diagnosis, and if VTE recurrence rates were objectively confirmed. These authors also identified an overall modest risk for VTE recurrence in patients with RVO (OR, 2.02; 95% CI, 1.62-2.50). However, this association was heterogeneous and dependent on numerous variables. Use of the Siragusa definition for RVO was associated with an OR of 13.68 (95% CI, 6.58-28.44), the presence of malignant disease was associated with an OR of 4.55 (95% CI, 1.76-11.79), and performing the RVO assessment at 3 months had an OR of 6.43 (95% CI, 3.90-10.62). Most recently, a patient-level meta-analysis was conducted by Donadini et al14 and many of the first authors of the major RVO manuscripts, limiting their scope to patients with a first unprovoked DVT who received $3 months of anticoagulation. Individual patient-level data (2527 patients, 10 data sets) were reanalyzed, and RVO was present in 1380 patients (55.1%). Overall VTE recurrence was seen in 399 patients (15.8%). By use of multivariate Cox analysis, the following variables were found to be significant predictors of VTE recurrence:
1. 2. 3.
RVO Age (for every 1-year increase) Male gender
Adjusted HR 1.32 (95% CI, 1.06-1.65) 1.01 (95% CI, 1.11-1.02) 1.49 (95% CI, 1.2-1.84)
However, RVO was predictive only if ultrasound assessment was made soon after the DVT (3 months) (HR, 2.17; 95% CI, 1.11-4.25) and not predictive if the duplex scan was performed >6 months after the index thrombotic event. DISCUSSION Surveillance venous duplex imaging, after the initial detection of lower extremity DVT, may serve multiple purposes. The first of these is to identify thrombus propagation
Stephenson and Liem 5
in patients who either have failed to respond to therapy or have not been able to receive therapeutic anticoagulation. The second purpose is to establish a new baseline, should there be any future concern for recurrent VTE. Testing for this purpose usually is performed when antithrombotic therapy is being discontinued.15 The third potential role for surveillance imaging is to detect the presence of RVO as a marker for patients at elevated risk for VTE recurrence. The natural history of lower extremity DVT is one of slow resolution. In patients who receive therapeutic anticoagulation, follow-up venography performed 6 months after diagnosis demonstrates complete lysis in 38.4%, partial lysis in 54.3%, and thrombus extension in 7.3%.16 The correlation between venographic and ultrasound determination of thrombus resolution is relatively close. Observational trials performing serial ultrasound examinations on patients with lower extremity DVT demonstrate normal venous duplex findings in 38.8% at 6 months, 58.1% at 12 months, 69.3% at 2 years, and 73.8% at 3 years.3 Continued resolution of DVT also has been demonstrated in other observational ultrasound studies.17,18 In evaluating the numerous randomized trials, the prospective cohort studies, and the more recent meta-analyses in aggregate, we conclude that RVO is a mild risk factor for VTE recurrence. The meta-analyses described in the Results section agree with regard to the presence of a mild association but reach differing conclusions with regard to patients with unprovoked DVTs. Carrier et al12 concluded that RVO was not a predictor of VTE recurrence with unprovoked DVT, whereas Donadini et al14 found a weak but significant association when selecting only patients with unprovoked DVT. The latter study did have the advantage of performing a patient-level meta-analysis of original data, thus providing more compelling evidence of a true association between RVO and VTE recurrence in patients with unprovoked DVT. These meta-analyses also agreed that the association with residual thrombosis was stronger when the duplex study was performed soon after the index DVT. Despite the mild association between residual thrombus and VTE recurrence, an evaluation for RVO at a single time point (whether at 3 months or 6 months) to make longer term decisions about extended-duration anticoagulation is likely to be of limited utility. Other more potent risk factors (male gender, D-dimer level, and VTE location) likely overshadow the recurrence associated with RVO. In addition, the RVO data are hampered by the significant heterogeneity of the various RVO-related studies. The definition for RVO was not standardized, the timing of the follow-up ultrasound study was somewhat variable, and some studies included patients with provoked and unprovoked DVT. In addition, the location of the index DVT was not consistently reported throughout the manuscripts. The results of the two studies by Siragusa et al5,10 showing the strongest association have not been well replicated in other studies. This may be in part due to the definition of RVO that was used in these manuscripts. In addition, patients were treated with varying
6 Stephenson and Liem
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
Table III. Published clinical scoring systems to predict risk for recurrent venous thromboembolism (VTE), applicable to unprovoked deep venous thrombosis (DVT) only Vienna Prediction Model23 Sex Location of first VTE D-dimer after anticoagulation
DASH score24
REVERSE study25 (female patients only)
D-dimer after anticoagulation Age <50 years Sex: male Hormone use at VTE onset
Post-thrombotic signs D-dimer Body mass index $30 kg/m2 Age $65 years
DASH, D-dimer, Age, Sex, Hormonal therapy.
durations of anticoagulation, and follow-up after discontinuation of anticoagulation also was variable. Alternatively, Prandoni et al11 have used serial venous duplex evaluation to tailor therapy on the basis of clearance of thrombus. These authors demonstrated a significant reduction in VTE recurrence when using the flexibleduration anticoagulant regimen compared with fixedduration therapy. In this study, patients in the former group received a mean of 3.7 months of additional OAC if their index DVT was secondary and a mean of 5.0 months if the DVT was unprovoked. The reduction in VTE recurrence was more substantial in patients with unprovoked DVT. However, it is possible that the benefit derived may have been due to the longer term anticoagulation alone, irrespective of the RVO status. An additional question that remains to be answered is whether the benefit of flexible-duration therapy persists after discontinuation of anticoagulation. Numerous other risk factors for VTE recurrence have been evaluated as well. These include hypercoagulable conditions, provoked vs unprovoked setting, VTE location, male gender, advanced age, elevated D-dimer levels, concurrent hormonal therapy, concurrent malignant disease, and obesity (body mass index $30 kg/m2).19 The presence of inherited or acquired thrombophilias is an independent predictor for the persistence of residual thrombus.20,21 However, there is conflicting evidence regarding their association with VTE recurrence. Potential explanations for these discrepancies may be related to the type of thrombophilia in question or the duration of anticoagulation.22 The more common thrombophilias (prothrombin 20210A and factor V Leiden) may confer a slight risk for VTE recurrence, but their clinical relevance is likely to be low in comparison with other clinical and laboratory predictors. D-dimer testing was evaluated in three of the six papers in this review.7-9 Cosmi et al8 demonstrated that although RVO was not a risk for VTE recurrence, an abnormal Ddimer test result, 1 month after withdrawal of OAC, was an independent risk factor for recurrence (HR, 3.32; 95% CI, 1.78-6.75; P < .0001). Poli et al7 showed similar results with D-dimer testing 4 weeks after OAC, when D-dimer positivity was a significant predictor of recurrence (HR, 3.2; 95% CI, 1.5-6.6; P < .002). Last, Cosmi et al9 adjusted OAC on the basis of D-dimer testing. Patients who had positive D-dimer test results after completion of OAC were randomized either to resume or to remain without OAC. The recurrence risk for the patients with a normal D-dimer level was 10% vs 19% for patients
with a positive D-dimer test result (P ¼ .002). However, D-dimer-positive patients who were randomized to receive continued anticoagulation had an even lower recurrence rate of 4.9%. These studies provide relatively consistent evidence that an abnormal result of D-dimer testing, when it is performed about 4 weeks after withdrawal of anticoagulation, is a more reliable indicator for VTE recurrence. Besides thrombophilia and D-dimer testing, several other demographic, clinical, and laboratory markers have a demonstrated association with VTE recurrence, including male gender, age, presence of post-thrombotic syndrome, obesity, VTE location, and concurrent hormonal therapy. Several of these have been incorporated into various clinical scoring systems to predict risk for recurrent VTE (Vienna Prediction Model, DASH [D-dimer, Age, Sex, Hormonal therapy] score, and REVERSE study)23-25 (Table III). These prediction scores remain to be further validated and are applicable only to patients with unprovoked VTE. However, they are easy to use and are widely available through the Internet. Notably, none of these scoring systems incorporates RVO as a predictor of VTE recurrence. In addition, none of these scoring systems provides discrete guidance for the continuation of OAC after the initial course is completed. Rather, they calculate a more individualized risk for VTE recurrence that must be integrated with the patient’s preference, the estimated risk for bleeding, the patient’s comorbidities, and the cost. Although the primary focus of this review was to analyze the association between RVO and recurrent VTE, it is important to recognize the frequently encountered diagnostic dilemma posed by the patient with an established DVT who presents with recurrent or worsening symptoms of unilateral leg pain or edema. Chronic venous thrombosis usually is characterized by increased echogenicity and a contracted and thickened vein wall. However, differentiating between RVO and recurrent ipsilateral DVT can occasionally be challenging, and duplex imaging after completion of anticoagulation has been shown to improve the accuracy for diagnosis of VTE recurrence. In a large cohort study by Hamadah et al,15 patients with a first unprovoked DVT received 5 to 7 months of OAC. There were 121 patients in whom VTE recurrence was suspected. Those who received a baseline venous duplex study after completion of anticoagulation had better interobserver agreement with regard to the duplex diagnosis of recurrence compared with patients who did not receive baseline imaging.
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
CONCLUSIONS In patients with unprovoked DVT, ultrasound determination of RVO is mildly predictive of VTE recurrence, but only when the venous duplex scan is performed soon after the index thrombotic event (3 months). Duplex assessment for residual thrombus is of limited usefulness when it is performed to guide the duration of therapeutic anticoagulation. Much of this is related to significant study heterogeneity with nonstandardized definitions for residual obstruction and varying durations of anticoagulation and clinical follow-up. Other clinical risk factors and laboratory parameters are likely to be of greater utility in predicting risk of VTE recurrence in patients with unprovoked lower extremity DVT (male gender, elevated D-dimer level, VTE location). These other variables have been incorporated into multiple clinical scoring systems and are likely to be better predictors for VTE recurrence. However, followup venous duplex surveillance at the time of anticoagulation withdrawal still is worthwhile, potentially helping to differentiate between RVO and future recurrent ipsilateral DVT. AUTHOR CONTRIBUTIONS Conception and design: TL, ES Analysis and interpretation: TL, ES Data collection: TL, ES Writing the article: TL, ES Critical revision of the article: TL, ES Final approval of the article: TL, ES Statistical analysis: TL, ES Obtained funding: Not applicable Overall responsibility: TL REFERENCES 1. Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, et al. Antithrombotic therapy for VTE disease. Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(Suppl):e419S-94S. 2. Young L, Ockelford P, Milne D, Rolfe-Vyson V, Mckelvie S, Harper P. Post-treatment residual thrombus increases the risk of recurrent deep vein thrombosis and mortality. J Thromb Haemost 2006;4:1919-24. 3. Prandoni P, Lensing A, Prins M, Bernardi E, Marchiori A, Bagatella P, et al. Residual venous thrombosis as a predictive factor of recurrent venous thromboembolism. Ann Intern Med 2002;137:955-60. 4. Prandoni P, Cogo A, Bernardi E, Villalta S, Polistena P, Simioni P, et al. A simple approach for detection of recurrent proximal vein thrombosis. Circulation 1993;88:1730-5. 5. Siragusa S, Malato A, Anastasio R, Cigna V, Milio G, Amato C, et al. Residual vein thrombosis to establish duration of anticoagulation after a first episode of deep vein thrombosis: the Duration of Anticoagulation based on Compression UltraSonography (DACUS) study. Blood 2008;112:511-5. 6. Le Gal G, Carrier M, Kovacs M, Betancourt M, Kahn S, Wells P, et al. Residual vein obstruction as a predictor for recurrent thromboembolic events after a first unprovoked episode: data from the REVERSE cohort study. J Thromb Haemost 2011;9:1126-32. 7. Poli D, Antonucci E, Ciuti G, Abbate R, Prisco D. Combination of Ddimer, F1þ2 and residual vein obstruction as predictors of VTE recurrence in patients with first VTE episode after OAT withdrawal. J Thromb Haemost 2008;6:708-10.
Stephenson and Liem 7
8. Cosmi B, Legnani C, Cini M, Guazzaloca G, Palareti G. The role of Ddimer and residual venous obstruction in recurrence of venous thromboembolism after anticoagulation withdrawal in cancer patients. Haematologica 2005;90:713-5. 9. Cosmi B, Legnani C, Iorio A, Pengo V, Ghirarduzzi A, Testa S, et al. Residual venous obstruction, alone and in combination with D-dimer, as a risk factor for recurrence after anticoagulation withdrawal following a first idiopathic deep vein thrombosis in the PROLONG study. Eur J Vasc Endovasc Surg 2010;39:356-65. 10. Siragusa S, Malato A, Saccullo G, Iorio A, Di Ianni M, Caracciolo C, et al. Residual vein thrombosis for assessing duration of anticoagulation after unprovoked deep vein thrombosis of the lower limbs: the extended DACUS study. Am J Hematol 2011;86:914-7. 11. Prandoni P, Prins MH, Lensing AW, Ghirarduzzi A, Agena W, Imberti D, et al. Residual thrombosis on ultrasonography to guide the duration of anticoagulation in patients with deep venous thrombosis: a randomized trial. Ann Intern Med 2009;150:577-85. 12. Carrier M, Rodger MA, Wells PS, Righini M, Le Gal G. Residual vein obstruction to predict the risk of recurrent venous thromboembolism in patients with deep vein thrombosis: a systematic review and metaanalysis. J Thromb Haemost 2011;9:1119-25. 13. Tan M, Mos IC, Klok FA, Huisman MV. Residual venous thrombosis as predictive factor for recurrent venous thromboembolim in patients with proximal deep vein thrombosis: a systematic review. Br J Haematol 2011;153:168-78. 14. Donadini M, Ageno W, Antonucci E, Cosmi B, Kovacs M, Le Gal G, et al. Prognostic significance of residual venous obstruction in patients with treated unprovoked deep vein thrombosis. A patient-level metaanalysis. Thromb Haemost 2014;111:172-9. 15. Hamadah A, Alwasaidi T, Le Gal G, Carrier M, Wells PS, Scarvelis D, et al. Baseline imaging after therapy for unprovoked venous thromboembolism: a randomized controlled comparison of baseline imaging for diagnosis of suspected recurrence. J Thromb Haemost 2011;9: 2406-10. 16. Holmström M, Lindmarker P, Grangqvist S, Johnsson H, Kockner D. A 6-month venographic follow-up in 164 patients with acute deep vein thrombosis. Thromb Haemost 1997;78:803-7. 17. Piovella F, Crippa L, Barone M, Vigano D’Angelo S, Serafini S, Galli L, et al. Normalization rates of compression ultrasonography in patients with a first episode of deep vein thrombosis of the lower limbs: association with DVT recurrence and new thrombosis. Haematologica 2002;87:515-22. 18. Ageno W, Steidl L, Piantanida E, Dentali F, Mera V, Squizzato A, et al. Predictors of residual vein obstruction after deep vein thrombosis of the lower limbs: a prospective study. Thromb Res 2003;108:203-7. 19. Kyrle PA, Rosendall FR, Eichinger S. Risk assessment for recurrent venous thrombosis. Lancet 2010;376:2032-9. 20. Spiezia L, Tormene D, Pesavento R, Salmaso L, Simioni P, Prandoni P. Thrombophilia as a predictor of persistent residual vein thrombosis. Haematologica 2008;93:479-80. 21. Cosmi B, Legnani C, Pengo V, Ghirarduzzi A, Testa S, Poli D, et al. The influence of factor V Leiden and G20210A prothrombin mutation on the presence of residual vein obstruction after idiopathic deep vein thrombosis of the lower limbs. Thromb Haemost 2013;109:510-6. 22. Prandoni P, Tormene D, Spiezia L, Pesavento R, Simioni P. Duration of anticoagulation and risk of recurrent thromboembolism in carriers of factor V Leiden or prothrombin mutation. J Thromb Haemost 2008;6: 2223-4. 23. Kyrle PA, Eichinger S. Clinical scores to predict recurrence of venous thromboembolism. Thromb Haemost 2012;108:1061-4. 24. Tosetto A, Iorio A, Marcucci M, Baglin T, Cushman M, Eichinger S, et al. Predicting disease recurrence in patients with previous unprovoked venous thromboembolism: a proposed prediction score (DASH). J Thromb Haemost 2012;10:1019-25. 25. Rodger MA, Kahn SR, Wells PS, Anderson DA, Chagnon I, Le Gal G, et al. Identifying unprovoked thromboembolism patients at low risk for recurrence who can discontinue anticoagulant therapy. CMAJ 2008;179:417-26.
Submitted Feb 24, 2014; accepted Aug 13, 2014.