SEARCH

SEARCH BY CITATION

Keywords:

  • deep vein thrombosis;
  • diagnosis;
  • pulmonary embolism;
  • unprovoked

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Summary. Background: After a first unprovoked venous thromboembolism (VTE), many patients have residual pulmonary and/or lower limb vascular obstruction following completion of short-term anticoagulation. Residual vascular obstruction may complicate the diagnosis of recurrent VTE. Whether baseline imaging, conducted after completion of anticoagulation, helps in interpreting diagnostic testing in patients who subsequently have suspected recurrent VTE is unknown. Study design: The REVERSE study is a cohort study whose primary aim was to derive a clinical decision rule to guide the duration of anticoagulation after a first unprovoked VTE. All patients underwent baseline imaging after completing 5–7 months of anticoagulant therapy. We performed a post hoc randomized controlled comparison among 121 patients investigated for a suspected recurrent VTE during follow-up: the decision on recurrent VTE with or without baseline imaging was made available to two independent adjudicators. Results: The proportion of patients not classifiable for recurrent VTE was statistically significantly higher in the group with no baseline imaging than in the group with baseline imaging: one in five as compared with one in 25. The interobserver agreement between the two adjudicators was better in the group with baseline imaging than in the group with no baseline imaging: κ-values were 0.78 and 0.54, respectively. Conclusions: In patients with a first unprovoked VTE, baseline imaging at completion of anticoagulant therapy helps in interpreting diagnostic tests performed in cases of suspected recurrent VTE.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

The diagnosis of recurrent venous thromboembolism (VTE) is a frequent and challenging clinical dilemma. The usual diagnostic algorithms for deep vein thrombosis (DVT) and pulmonary embolism (PE) have limited accuracy and utility in patients with a prior episode of VTE [1]. Patients with a prior VTE are less likely to be classified as low risk by clinical decision rules, and are therefore more likely to require further diagnostic tests in order to rule out recurrent VTE [2–4]. The specificity and clinical utility of D-dimer testing is also lower in patients with a suspected recurrent VTE [5,6]. Furthermore, residual venous obstruction (RVO) is a common finding in patients with prior DVT. Approximately, 50–70% of patients with DVT will have RVO on CUS after 3 and 12 months of follow-up [7,8]. Similarly, up to 60% of patients of patient with PE will have residual perfusion defects on ventilation/perfusion (V/Q) scans performed after 5–7 months of anticoagulation therapy [9]. Residual DVT or PE complicates the diagnosis of recurrent VTE, as imaging techniques for the classification of VTE as acute or chronic have not been adequately validated.

Patients with prior unprovoked VTE frequently present with suspected recurrent VTE. In our recent study, approximately 50% of patients with unprovoked VTE who completed 5–7 months of anticoagulation therapy presented with a suspected recurrent VTE over a 2-year period following discontinuation of anticoagulation therapy, and one in four of the suspected recurrent VTEs was confirmed to be recurrent VTE [1,10].

The consequences of misdiagnosis of recurrent VTE are important. Indefinite anticoagulation is often recommended in patients with recurrent VTE [11]. Anticoagulation therapy is very effective at reducing the risk of recurrence in this patient population [12,13]. Objective diagnostic confirmation is required, as only one-third to one-fifth of patients with a clinically suspected recurrent VTE have a recurrent VTE. Furthermore, the benefits of anticoagulant therapy need to be balanced against the risk of major bleeding. The case fatality rate of a major bleed is three-fold higher than that of a recurrent VTE following anticoagulation: 11.3% vs. 3.6%, respectively [14]. Therefore, accurate diagnosis of recurrent VTE is important to avoid the unnecessary and large burden of indefinite anticoagulant therapy in patients with a false-positive diagnosis of recurrent VTE.

Baseline imaging after completion of anticoagulant therapy is controversial. Conducting baseline imaging is not only costly but also resource-intensive, and is an additional burden for patients. We have recently validated a diagnostic management approach to exclude the diagnosis of recurrent VTE that relies on comparison of imaging results obtained at the time of suspected recurrent VTE with those of baseline imaging conducted at the time of completion of anticoagulant therapy [1]. To our knowledge, no previous studies have examined whether conducting baseline imaging influences the ability to diagnostically classify patients with recurrent VTE or no recurrent VTE and the interobserver reliability of diagnostic classification. To fill this knowledge gap, we sought to assess the utility of baseline diagnostic imaging at the time of discontinuation of anticoagulant therapy in patients with suspected recurrent VTE.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Patients

The REVERSE study was a multicenter multinational prospective cohort study whose primary aim was to develop a clinical decision rule to guide the duration of anticoagulation for unprovoked VTE. The methods have been extensively described elsewhere [10]. Briefly, consecutive unselected patients at 12 tertiary-care centers in four countries were asked to participate if they had a first objectively proven unprovoked VTE and had completed 5–7 months of anticoagulant therapy. Unprovoked VTE was defined as VTE occurring in the absence of a leg fracture, lower extremity plaster cast, immobilization for > 3 days, surgery involving a general anesthetic in the 3 months before the event, and a malignancy in the past 5 years. Patients were enrolled over a 4-year period between October 2001 and March 2006, and completed a mean of 18 months of follow-up in September 2006. Institutional research ethics board approval was obtained at all participating centers.

Diagnosis of VTE

Objective documentation of an index DVT required the presence of a non-compressible segment on a compression ultrasound (CUS) scan of the popliteal vein or a more proximal leg vein. Objective documentation of an index PE required a high-probability V/Q scan or a segmental or larger artery filling defect on a spiral computed tomography scan.

All patients underwent baseline imaging at the completion of the anticoagulant treatment. A baseline leg vein CUS scan was conducted on the leg that was symptomatic at the time of the index event. A baseline V/Q scan was performed if the patient had signs or symptoms of PE at the time of the index event. After baseline imaging was performed, patients were instructed to stop their anticoagulant treatment, and to contact study personnel if they developed symptoms of recurrent VTE during follow-up. A case report form was completed for all suspected recurrent VTEs. The clinical history, physical examination, D-dimer, date of symptom onset and type of suspected recurrent VTE (DVT, PE, or both) were recorded. Patients with suspected recurrent VTE underwent diagnostic imaging tests, the results of which were compared with those of baseline imaging. All suspected VTEs and deaths were independently adjudicated by two independent physicians.

Study analysis

For the present study, post hoc, we randomly selected a subgroup of patients who underwent diagnostic testing for suspected symptomatic recurrent VTE during follow-up. We randomized these patients’ charts into two groups: (i) baseline imaging group – the results of the baseline imaging conducted after 5–7 months of anticoagulant therapy was made available to adjudicators, as well as those of the index imaging (imaging conducted at the time of the initial first unprovoked VTE diagnosis) and the imaging performed at the time of suspected recurrent VTE; and (ii) no baseline imaging – all baseline imaging results and references to baseline imaging were removed from the patients’ charts, leaving for the adjudicators’ review only the results of the index imaging (imaging conducted at the time of the initial first unprovoked VTE diagnosis) and the imaging performed at the time of suspected recurrent VTE.

The charts from the two groups (baseline imaging, and no baseline imaging) were randomly assigned to five adjudicators. These adjudicators consisted of five thrombosis specialists from the Ottawa Hospital Thrombosis Program (M.R., M.A.F., D.S., P.W., and M.C.). Adjudicators reviewed patients’ charts that included patients’ clinical documentation, the results of imaging at the time of the index VTE, and the results of imaging at the time of suspected recurrent VTE, and finally, that included or did not include the results of baseline imaging according to randomization. Each patient’s adjudication file was reviewed by two adjudicators independently, herein identified as adjudicator 1 and adjudicator 2. The adjudicators were asked to first determine whether the patients were classifiable. Patients were not classifiable if the adjudicator felt that there was insufficient information to classify patients as having recurrent VTE or not. If patients were classifiable, the adjudicators were asked to document whether patients had recurrent VTE or no recurrent VTE.

Descriptive statistics were performed with proportions for categorical variables, and means and standard deviation for continuous variables. Our primary analysis consisted of determining the proportion classifiable in the baseline imaging group vs. the proportion in the no baseline imaging group. Our secondary analyses examined the interobserver agreement by use of κ-statistics for the determination of classifiable, recurrent VTE and no recurrent VTE, and, in the classifiable subgroup, for the determination of recurrent VTE or no recurrent VTE. We determined a priori the number of patients that needed to be included into each group in order to demonstrate a statistically significantly higher proportion of not classifiable patients in the no baseline imaging group than in the baseline imaging group. We assumed that 10% of patients in the no baseline imaging group would be not classifiable, so 120 patients would provide 80% power to detect a 9% absolute difference.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

We included 646 patients after completion of 5–7 months of anticoagulant therapy for a first unprovoked VTE in the REVERSE cohort study. The mean age of participants was 53 years (range: 18–95 years), and 48.6% were females. During a mean 18-month follow-up period (range: 1–47 months), 313 suspected recurrent VTEs occurred in 233 (36.1% of total) patients. Ninety-one of these were adjudicated as being recurrent VTEs (9.4% recurrent VTEs per year). None of the deaths during follow-up was adjudicated as being attributable to recurrent VTE.

From the 233 patients who experienced a suspected recurrent VTE, we randomly selected the charts of 121 patients. In patients with more than one suspected recurrent VTE during follow-up, only the first suspicion was eligible for randomization. Sixty-four patients were assigned to the baseline imaging group, and 57 to the no baseline imaging group.

The patients’ baseline characteristics are shown in Table 1. The proportion of men was higher in the baseline imaging group than in the no baseline imaging group (72% vs. 47%).

Table 1.   Baseline characteristics
 Baseline imaging, n = 64No baseline imaging, n = 57
  1. DVT, deep vein thrombosis; IQR, interquartile range; PE, pulmonary embolism; VTE, venous thromboembolism. *Adjudication results from original study.

Age (years), median (IQR)55 (38–62)51 (43–63)
Male gender, no. (%)46 (72)27 (47)
Index VTE, no. (%)
 Isolated PE21 (32.8)16 (28.1)
 Isolated DVT38 (59.4)32 (56.1)
 PE and DVT5 (7.8)9 (15.8)
Recurrent VTE, no. (%)*
 All24/64 (37.5)18/57 (31.6)
 Males21/46 (45.6)11/27 (40.7)
 Females3/18 (16.7)7/30 (23.3)

The proportion of patients deemed ‘not classifiable’ at the time of the suspected recurrent VTE event was statistically significantly lower in the group with baseline imaging than in the group with no baseline imaging for both adjudicators (Table 2). Approximately one in five patients was not classifiable in the group without baseline imaging, as compared with one in 25 in the group with baseline imaging. Subgroup analyses were conducted in subgroups of patients with index isolated PE, isolated DVT, combined PE and DVT, and DVT with or without PE. In each subgroup, baseline imaging resulted in a higher proportion of classifiable patients (Table 3).

Table 2.   Proportion of ‘not classifiable’ venous thromboembolism (VTE) events in patients presenting with a suspected recurrent event
 Baseline imaging, n = 64, no. (%)No baseline imaging, n = 57, no. (%)P-value
Adjudicator 13 (4.7)11 (19.3)0.012
Adjudicator 22 (3.1)13 (22.8)0.001
Table 3.   Proportion of ‘not classifiable’ venous thromboembolism (VTE) events in patients presenting with a suspected recurrent event, according to the type of index event
 Isolated PE, N = 38Isolated DVT, N = 62Combined PE and DVT, N = 21
 Baseline imaging, n = 22No baseline imaging, n = 16Baseline imaging, n = 33No baseline imaging, n = 29Baseline imaging, n = 9No baseline imaging, n = 12
  1. DVT, deep vein thrombosis ; PE, pulmonary embolism.

Adjudicator 1, no. (%)1 (4.5)2 (12.5)2 (6.1)5 (17.2)0 (0)4 (33.3)
 P = 0.36P = 0.16P = 0.054
Adjudicator 2, no. (%)2 (9.1)3 (18.8)0 (0)6 (20.7)0 (0)4 (33.3)
 P = 0.38P = 0.006P = 0.054

The overall interobserver agreements (κ-coefficient) between the two adjudicators were 0.78 (95% confidence interval [CI] 0.63–0.93) and 0.54 (95% CI 0.35–0.73) in the baseline and in the no baseline imaging groups, respectively. In the subgroup of patients deemed ‘classifiable’ for recurrent VTE, there was excellent agreement between adjudicators in the group with baseline imaging (κ-coefficient 0.93, 95% CI 0.83–1) and the group without baseline imaging (κ-coefficient 1.0, 95% CI 0.83–1).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Our study demonstrates that the availability of baseline imaging conducted after completion of anticoagulant therapy in unprovoked DVT and/or PE patients significantly improves the ability to classify suspected recurrent VTE. As a result, when baseline imaging is available, a lower proportion of patients have an inconclusive diagnosis of recurrent VTE. Baseline imaging also improves interobserver agreement for classification of suspected recurrent VTE.

The strengths of our study include: (i) the use of actual patient data that provide a real-world combination of suspected recurrent events – that is, we included patients with all types of index event (isolated DVT, isolated PE, or DVT and PE), all patterns of residual disease on baseline imaging (isolated residual pulmonary vascular obstruction, isolated RVO, or both), and all possible types of suspected recurrent VTE (recurrent VTE presenting as isolated recurrent DVT, recurrent isolated PE, or recurrent DVT and PE); (ii) the fact all of our patient charts had index VTE imaging reports available, which is not always the case in real-world practice – despite this information, baseline imaging was of incremental benefit in classifying patients; and (iii) the randomization, which limited selection bias and provided a control group.

There are limitations to our study. First, baseline imaging was restricted to the index symptomatic VTE site. However, it is well known that 50% of patients with symptomatic DVT have associated asymptomatic perfusion defects on V/Q scan when performed. Similarly, asymptomatic DVT is found in nearly 50% of patients with acute PE [15]. Therefore, we probably underestimated the proportion of residual obstruction in the pulmonary and lower extremity vasculature, and hence a larger proportion of patients would have had documented residual disease, which may have to have led to an overestimation of recurrent disease in our study patients. Second, this was a retrospective comparison analysis, and the results may not be applicable to daily clinical practice, because physicians may have altered their diagnostic pursuit when faced with the uncertainty of classification that we have documented in this study. Nonetheless, when clinicians are confronted with this common clinical scenario, comparison with index or baseline imaging results is always retrospective in nature, and suspected recurrent VTE imaging results are often also retrospectively reviewed by other healthcare clinicians who are subsequently involved in a patient’s care. Third, only patients with a first unprovoked VTE were included in our study. Therefore, our results do not apply to patients with prior multiple recurrent VTEs or provoked VTE. Nonetheless, recurrent VTE in patients with multiple prior VTEs is more likely to be non-classifiable, given the higher likelihood of residual vascular obstruction. Fourth, although D-dimer results were made available to adjudicators in both groups if they were available, they were not systematically collected for all patients at the time of suspected recurrent VTE. If D-dimer results had been available for all patients, this may have improved overall classifiability and perhaps altered our study results. However, although evidence suggests that a negative D-dimer result safely rules out VTE in patients with a prior VTE, the clinical utility of the test is halved in this group: in a post hoc analysis of a diagnostic management outcome study, only 15.9% of patients with a previous history of VTE had PE ruled out on the basis of the combination of a non-high clinical probabillity of PE and a negative D-dimer result, as compared with 32.7% of patients with no previous episode [5]. Finally, whether baseline imaging is cost-effective or not deserves further attention. However, given that 50% of patients present with suspected recurrent VTE during the first 2 years after anticoagulant therapy discontinuation [1], it would be useful in a large proportion of patients after completion of therapy for a first unprovoked VTE, and as such, in our view, is likely to be cost-effective when one considers the costs and consequences of subsequent misdiagnosis, including lifelong anticoagulation in false-positives.

In conclusion, conducting baseline imaging after completion of anticoagulant therapy should be strongly considered; baseline imaging appears to increase diagnostic certainty in the common group of patients with unprovoked VTE who subsequently present with suspected recurrent VTE, and therefore probably provides a more solid basis on which to make a decision to indefinitely resume lifelong anticoagulation.

Addendum

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

M. Rodger, G. Le Gal, A. Hamadah, T. Alwasaidi, M. Kovacs, and P. Wells: conception and design; M. Rodger: administrative support; M. Kovacs and M. Rodger: grant funding; A. Hamadah, T. Alwasaidi, G. Le Gal, M. Carrier, and M. Rodger: data analysis; A. Hamadah, T. Alwasaidi, G. Le Gal, M. Carrier, and M. Rodger: drafting of the paper; M. Kovacs, M. Rodger, M. Carrier, and G. Le Gal: critical revision for important intellectual content. All authors were responsible for data acquisition and final approval of the manuscript. M. Kovacs and M. Rodger are co-senior authors. G. Le Gal and M. Rodger had full access to all data in the study, and take responsibility for the integrity of the data and the accuracy of the data analysis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

M. Carrier is a recipient of a Canadian Institute for Health Research RCT mentoring award. O. Wells is a recipient of a Canada Research Chair. M. Rodger was the recipient of the Maureen Andrew New Investigator Award and holds a Career Investigator Award from the Heart and Stroke Foundation of Canada and a University of Ottawa, Faculty of Medicine Research Chair in Venous Thrombosis and Thrombophilia.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

This study was funded by the Canadian Institutes of Health Research (Grant no. MOP 64319) and BioMerieux (through an unrestricted research grant). BioMerieux had no role in the design and conduct of the study, the collection, management, analysis and interpretation of the data, or the preparation, review or approval of the manuscript. The Canadian Institutes of Health Research reviewed the design of the study but had no role in the conduct of the study, the collection, management, analysis and interpretation of the data, or the preparation, review or approval of the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  • 1
    Le Gal G, Kovacs MJ, Carrier M, Do K, Kahn SR, Wells PS, Anderson DA, Chagnon I, Solymoss S, Crowther M, Righini M, Perrier A, White RH, Vickars L, Rodger M. Validation of a diagnostic approach to exclude recurrent venous thromboembolism. J Thromb Haemost 2009; 7: 7529.
  • 2
    Le Gal G, Righini M, Roy PM, Sanchez O, Aujesky D, Bounameaux H, Perrier A. Prediction of pulmonary embolism in the emergency department: the revised Geneva score. Ann Intern Med 2006; 144: 16571.
  • 3
    Wells PS, Anderson DR, Rodger MA, Ginsberg JS, Kearon C, Gent M, Turpie AGG, Bormanis J, Weitz J, Chamberlain M, Bowie D, Barnes D, Hirsh J. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the model’s utility with the SimpliRED D-dimer. Thromb Haemost 2000; 83: 41620.
  • 4
    Wells PS, Anderson DR, Rodger M, Forgie MA, Kearon C, Dreyer JF, Kovacs G, Mitchell M, Lewandowski B, Kovacs MJ. Evaluation of D-dimer in the diagnosis of suspected deep vein thrombosis. N Engl J Med 2003; 349: 122735.
  • 5
    Le Gal G, Righini M, Roy PM, Sanchez O, Aujesky D, Perrier A, Bounameaux H. Value of D-dimer testing for the exclusion of pulmonary embolism in patients with previous venous thromboembolism. Arch Intern Med 2006; 166: 17680.
  • 6
    Rathbun SW, Whitsett TL, Raskob GE. Negative D-dimer result to exclude recurrent deep venous thrombosis: a management trial. Ann Intern Med 2004; 141: 83945.
  • 7
    Kearon C. Natural history of venous thromboembolism. Circulation 2003; 107: I-22I-30.
  • 8
    Siragusa S, Malato A, Anastasio R, Cigna V, Milio G, Amato C, Bellisi M, Attanzio MT, Cormaci O, Pellegrino M, Dolce A, Casuccio A, Bajardi G, Mariani G. 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: 51115.
  • 9
    Nijkeuter M, Hovens MM, Davidson BL, Huisman MV. Resolution of thromboemboli in patients with acute pulmonary embolism: a systematic review. Chest 2006; 129: 1927.
  • 10
    Rodger MA, Kahn SR, Wells PS, Anderson DA, Chagnon I, Le Gal G, Solymoss S, Crowther M, Perrier A, White R, Vickars L, Ramsay T, Betancourt MT, Kovacs MJ. Identifying unprovoked thromboembolism patients at low risk for recurrence who can discontinue anticoagulant therapy. CMAJ 2008; 179: 41726.
  • 11
    Kearon C, Kahn SR, Agnelli G, Goldhaber SZ, Raskob GE, Comerota AJ. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians evidence-based clinical practice guidelines (8th edn). Chest 2008; 133(Suppl.): 454545.
  • 12
    Kearon C, Gent M, Hirsh J, Weitz J, Kovacs MJ, Anderson DR, Turpie AGG, Green D, Ginsberg JS, Wells PS, MacKinnon B, Julian JA. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med 1999; 340: 9017.
  • 13
    Agnelli G, Prandoni P, Santamaria MG, Bagatella P, Iorio A, Bazzan M, Moia M, Guazzaloca G, Bertoldi A, Tomasi C, Scannapieco G, Ageno W. Three months versus one year of oral anticoagulant therapy for idiopathic deep venous thrombosis. Warfarin optimal duration Italian trial investigators. N Engl J Med 2001; 345: 1659.
  • 14
    Carrier M, Le Gal G, Wells PS, Rodger MA. Systematic review: case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism. Ann Intern Med 2010; 152: 57889.
  • 15
    Le Gal G, Righini M, Sanchez O, Roy PM, Baba-Ahmed M, Perrier A, Bounameaux H. A positive compression ultrasonography of the lower limb veins is highly predictive of pulmonary embolism on computed tomography in suspected patients. Thromb Haemost 2006; 95: 9636.