Antithrombotic therapy for vte disease

[ Evidence-Based Medicine ] Antithrombotic Therapy for VTE DiseaseCHEST Guideline and Expert Panel Report Clive Kearon, MD, PhD; Elie A. Akl, MD, MPH, PhD; Joseph Ornelas, PhD; Allen Blaivas, DO, FCCP; David Jimenez, MD, PhD, FCCP; Henri Bounameaux, MD; Menno Huisman, MD, PhD; Christopher S. King, MD, FCCP; Timothy A. Morris, MD, FCCP; Namita Sood, MD, FCCP; Scott M. Stevens, MD; Janine R. E. Vintch, MD, FCCP; Philip Wells, MD; Scott C. Woller, MD; and COL Lisa Moores, MD, FCCP BACKGROUND: We update recommendations on 12 topics that were in the 9th edition of theseguidelines, and address 3 new topics.
METHODS: We generate strong (Grade 1) and weak (Grade 2) recommendations based onhigh- (Grade A), moderate- (Grade B), and low- (Grade C) quality evidence.
RESULTS: For VTE and no cancer, as long-term anticoagulant therapy, we suggest dabigatran(Grade 2B), rivaroxaban (Grade 2B), apixaban (Grade 2B), or edoxaban (Grade 2B) overvitamin K antagonist (VKA) therapy, and suggest VKA therapy over low-molecular-weightheparin (LMWH; Grade 2C). For VTE and cancer, we suggest LMWH over VKA (Grade 2B),dabigatran (Grade 2C), rivaroxaban (Grade 2C), apixaban (Grade 2C), or edoxaban (Grade2C). We have not changed recommendations for who should stop anticoagulation at3 months or receive extended therapy. For VTE treated with anticoagulants, we recommendagainst an inferior vena cava ﬁlter (Grade 1B). For DVT, we suggest not using compressionstockings routinely to prevent PTS (Grade 2B). For subsegmental pulmonary embolism andno proximal DVT, we suggest clinical surveillance over anticoagulation with a low risk ofrecurrent VTE (Grade 2C), and anticoagulation over clinical surveillance with a high risk(Grade 2C). We suggest thrombolytic therapy for pulmonary embolism with hypotension(Grade 2B), and systemic therapy over catheter-directed thrombolysis (Grade 2C). Forrecurrent VTE on a non-LMWH anticoagulant, we suggest LMWH (Grade 2C); for recurrentVTE on LMWH, we suggest increasing the LMWH dose (Grade 2C).
CONCLUSIONS: Of 54 recommendations included in the 30 statements, 20 were strong andnone was based on high-quality evidence, highlighting the need for further research.
CHEST 2016; 149(2):315-352 KEY WORDS: antithrombotic therapy; evidence-based medicine; GRADE approach; venousthromboembolism FOR EDITORIAL COMMENT SEE PAGE 293 ABBREVIATIONS: AT9 = 9th Edition of the Antithrombotic Guideline; Lebanon; CHEST (Dr Ornelas), Glenview, IL; VA New Jersey Health AT10 = 10th Edition of the Antithrombotic Guideline; CHEST = Care System (Dr Blaivas), Newark, NJ; Hospital Ramón y Cajal and American College of Chest Physicians; CDT = catheter-directed Instituto Ramón y Cajal de Investigación Sanitaria, Universidad de thrombolysis; COI = conﬂict of interest; CTEPH = chronic thrombo- Alcala (Dr Jimenez), Madrid, Spain; University of Geneva (Dr Bou- embolic pulmonary hypertension; CTPA = CT pulmonary angiogram; nameaux), Geneva, Switzerland; Leiden University Medical Center (Dr GOC = Guidelines Oversight Committee; INR = International Huisman), Leiden, Netherlands; Virginia Commonwealth University Normalized Ratio; IVC = inferior vena cava; LMWH = low-molecular- (Dr King), Falls Church, VA; University of California (Dr Morris), San weight heparin; NOAC = non-vitamin K oral anticoagulant; PE = Diego, CA; The Ohio State University (D. Sood), Columbus, OH; pulmonary embolism; PTS = postthrombotic syndrome; RCT = ran- Intermountain Medical Center and the University of Utah (Drs Stevens domized controlled trial; UEDVT = upper extremity deep vein and Woller), Murray, UT; Harbor-UCLA Medical Center (Dr Vintch), thrombosis; US = ultrasound; VKA = vitamin K antagonist Torrance, CA; The University of Ottawa and Ottawa Hospital Research AFFILIATIONS: From McMaster University (Drs Kearon and Akl), Institute (Dr Wells), Ottawa, ON; Uniformed Services University of the Hamilton, ON; American University of Beirut (Dr Akl), Beirut, Health Sciences (Dr Moores), Bethesda, MD.
Note on Shaded Text: In this guideline, shaded text with rivaroxaban and apixaban, and is overlapped with VKA an asterisk (shading appears in PDF only) indicates therapy. See text for factors that inﬂuence choice of recommendations that are newly added or have been changed since the publication of Antithrombotic Therapyfor VTE Disease: Antithrombotic Therapy and Prevention *3. In patients with DVT of the leg or PE and cancer of Thrombosis (9th edition): American College of Chest ("cancer-associated thrombosis"), as long-term (ﬁrst Physicians Evidence-Based Clinical Practice Guidelines.
3 months) anticoagulant therapy, we suggest LMWH Recommendations that remain unchanged since that over VKA therapy (Grade 2C), dabigatran (Grade edition are not shaded. The order of our presentation of the 2C), rivaroxaban (Grade 2C), apixaban (Grade 2C), non-vitamin K oral anticoagulants (dabigatran, or edoxaban (Grade 2C).
rivaroxaban, apixaban, edoxaban) is based on thechronology of publication of the phase 3 trials in VTE Remarks: Initial parenteral anticoagulation is given treatment and should not be interpreted as the guideline before dabigatran and edoxaban, is not given before panel's order of preference for the use of these agents.
rivaroxaban and apixaban, and is overlapped with VKAtherapy. See text for factors that inﬂuence choice oftherapy.
Summary of Recommendations Choice of Long-Term (First 3 Months) and Extended *4. In patients with DVT of the leg or PE who receive (No Scheduled Stop Date) Anticoagulant extended therapy, we suggest that there is no need 1. In patients with proximal DVT or pulmonary to change the choice of anticoagulant after the ﬁrst embolism (PE), we recommend long-term (3 months) 3 months (Grade 2C).
anticoagulant therapy over no such therapy (Grade 1B).
Remarks: It may be appropriate for the choice of *2. In patients with DVT of the leg or PE and no anticoagulant to change in response to changes in the cancer, as long-term (ﬁrst 3 months) anticoagulant patient's circumstances or preferences during long-term therapy, we suggest dabigatran, rivaroxaban, apix- or extended phases of treatment.
aban, or edoxaban over vitamin K antagonist (VKA)therapy (all Grade 2B).
Duration of Anticoagulant Therapy For patients with DVT of the leg or PE and no cancer 5. In patients with a proximal DVT of the leg or who are not treated with dabigatran, rivaroxaban, PE provoked by surgery, we recommend treatment apixaban, or edoxaban, we suggest VKA therapy over with anticoagulation for 3 months over (i) treatment low-molecular weight heparin (LMWH) (Grade 2C).
of a shorter period (Grade 1B), (ii) treatment of alonger time-limited period (eg, 6, 12, or 24 months) Remarks: Initial parenteral anticoagulation is given (Grade 1B), or (iii) extended therapy (no scheduled before dabigatran and edoxaban, is not given before stop date) (Grade 1B).
6. In patients with a proximal DVT of the leg or DISCLAIMER: American College of Chest Physician guidelines are PE provoked by a nonsurgical transient risk factor, intended for general information only, are not medical advice, and donot replace professional medical care and physician advice, which we recommend treatment with anticoagulation for always should be sought for any medical condition. The complete 3 months over (i) treatment of a shorter period disclaimer for this guideline can be accessed at (Grade 1B) and (ii) treatment of a longer time-limited period (eg, 6, 12, or 24 months) (Grade 1B). We FUNDING/SUPPORT: This guideline was supported solely by internal suggest treatment with anticoagulation for 3 months funds from The American College of Chest Physicians.
ENDORSEMENTS: This guideline is endorsed by the American Asso- over extended therapy if there is a low or moderate ciation for Clinical Chemistry, the American College of Clinical bleeding risk (Grade 2B), and recommend treatment Pharmacy, the International Society for Thrombosis and Haemostasis, for 3 months over extended therapy if there is a high and the American Society of Health-System Pharmacists.
CORRESPONDENCE TO: Elie A. Akl, MD, MPH, PhD, Department of risk of bleeding (Grade 1B).
Internal Medicine, Faculty of Medicine, American University of Beirut,PO Box 11-0236, Riad-El-Solh Beirut 1107 2020, Lebanon; e-mail: Remarks: In all patients who receive extended anticoagulant therapy, the continuing use of Copyright Ó 2016 American College of Chest Physicians. Published byElsevier Inc. All rights reserved.
treatment should be reassessed at periodic intervals (eg, annually).
Evidence-Based Medicine 7. In patients with an isolated distal DVT of the leg scheduled stop date) over 3 months (Grade 1B); provoked by surgery or by a nonsurgical transient risk (ii) moderate bleeding risk (see text), we suggest factor, we suggest treatment with anticoagulation for extended anticoagulant therapy over 3 months of therapy 3 months over treatment of a shorter period (Grade 2C), (Grade 2B); or (iii) high bleeding risk (see text), we we recommend treatment with anticoagulation for suggest 3 months of anticoagulant therapy over 3 months over treatment of a longer time-limited extended therapy (no scheduled stop date) (Grade 2B).
period (eg, 6, 12, or 24 months) (Grade 1B), and we Remarks: In all patients who receive extended recommend treatment with anticoagulation for anticoagulant therapy, the continuing use of treatment 3 months over extended therapy (no scheduled stop should be reassessed at periodic intervals (eg, annually).
date) (Grade 1B).
11. In patients with DVT of the leg or PE and active Remarks: Duration of treatment of patients with isolated cancer ("cancer-associated thrombosis") and who distal DVT refers to patients in whom a decision has (i) do not have a high bleeding risk, we recommend been made to treat with anticoagulant therapy; however, extended anticoagulant therapy (no scheduled stop it is anticipated that not all patients who are diagnosed date) over 3 months of therapy (Grade 1B), or (ii) have with isolated distal DVT will be prescribed a high bleeding risk, we suggest extended anticoagulant therapy (no scheduled stop date) over 3 months of 8. In patients with an unprovoked DVT of the leg therapy (Grade 2B).
(isolated distal or proximal) or PE, we recommend Remarks: In all patients who receive extended treatment with anticoagulation for at least 3 months anticoagulant therapy, the continuing use of treatment over treatment of a shorter duration (Grade 1B), and should be reassessed at periodic intervals (eg, annually).
we recommend treatment with anticoagulation for3 months over treatment of a longer time-limitedperiod (eg, 6, 12, or 24 months) (Grade 1B).
Aspirin for Extended Treatment of VTE Remarks: After 3 months of treatment, patients with *12. In patients with an unprovoked proximal DVT unprovoked DVT of the leg or PE should be evaluated or PE who are stopping anticoagulant therapy and for the risk-beneﬁt ratio of extended therapy. Duration do not have a contraindication to aspirin, we suggest of treatment of patients with isolated distal DVT refers aspirin over no aspirin to prevent recurrent VTE to patients in whom a decision has been made to treat with anticoagulant therapy; however, it is anticipated Remarks: Because aspirin is expected to be much less that not all patients who are diagnosed with isolated effective at preventing recurrent VTE than distal DVT will be prescribed anticoagulants.
anticoagulants, we do not consider aspirin a reasonable 9. In patients with a ﬁrst VTE that is an unprovoked alternative to anticoagulant therapy in patients who proximal DVT of the leg or PE and who have a (i) low want extended therapy. However, if a patient has or moderate bleeding risk (see text), we suggest decided to stop anticoagulants, prevention of recurrent extended anticoagulant therapy (no scheduled stop VTE is one of the beneﬁts of aspirin that needs to be date) over 3 months of therapy (Grade 2B), and balanced against aspirin's risk of bleeding and (ii) high bleeding risk (see text), we recommend inconvenience. Use of aspirin should also be reevaluated 3 months of anticoagulant therapy over extended when patients stop anticoagulant therapy because therapy (no scheduled stop date) (Grade 1B).
aspirin may have been stopped when anticoagulantswere started.
Remarks: Patient sex and D-dimer level measured amonth after stopping anticoagulant therapy mayinﬂuence the decision to stop or extend anticoagulant Whether and How to Anticoagulate Isolated therapy (see text). In all patients who receive extended anticoagulant therapy, the continuing use of treatment 13. In patients with acute isolated distal DVT of the should be reassessed at periodic intervals (eg, annually).
leg and (i) without severe symptoms or risk factors for 10. In patients with a second unprovoked VTE extension (see text), we suggest serial imaging of the and who have a (i) low bleeding risk (see text), we deep veins for 2 weeks over anticoagulation (Grade 2C) recommend extended anticoagulant therapy (no or (ii) with severe symptoms or risk factors for extension (see text), we suggest anticoagulation over Whether to Anticoagulate Subsegmental PE serial imaging of the deep veins (Grade 2C).
*19. In patients with subsegmental PE (no involve- Remarks: Patients at high risk for bleeding are more ment of more proximal pulmonary arteries) and likely to beneﬁt from serial imaging. Patients who place no proximal DVT in the legs who have a (i) low a high value on avoiding the inconvenience of repeat risk for recurrent VTE (see text), we suggest clinical imaging and a low value on the inconvenience of surveillance over anticoagulation (Grade 2C) or treatment and on the potential for bleeding are likely (ii) high risk for recurrent VTE (see text), we to choose initial anticoagulation over serial imaging.
suggest anticoagulation over clinical surveillance(Grade 2C).
14. In patients with acute isolated distal DVT ofthe leg who are managed with anticoagulation, we Remarks: Ultrasound (US) imaging of the deep veins recommend using the same anticoagulation as for of both legs should be done to exclude proximal DVT.
patients with acute proximal DVT (Grade 1B).
Clinical surveillance can be supplemented by serial US 15. In patients with acute isolated distal DVT of imaging of the proximal deep veins of both legs to the leg who are managed with serial imaging, we detect evolving DVT (see text). Patients and physicians (i) recommend no anticoagulation if the thrombus are more likely to opt for clinical surveillance over does not extend (Grade 1B), (ii) suggest anticoagulation anticoagulation if there is good cardiopulmonary reserve if the thrombus extends but remains conﬁned to or a high risk of bleeding.
the distal veins (Grade 2C), and (iii) recommendanticoagulation if the thrombus extends into the Treatment of Acute PE Out of the Hospital proximal veins (Grade 1B).
*20. In patients with low-risk PE and whose home Catheter-Directed Thrombolysis for Acute circumstances are adequate, we suggest treatmentat home or early discharge over standard discharge (eg, after the ﬁrst 5 days of treatment) 16. In patients with acute proximal DVT of the leg, we suggest anticoagulant therapy alone over CDT(Grade 2C).
Remarks: Patients who are most likely to beneﬁt from Systemic Thrombolytic Therapy for PE CDT (see text), who attach a high value to prevention of 21. In patients with acute PE associated with postthrombotic syndrome (PTS), and a lower value to the hypotension (eg, systolic BP <90 mm Hg) who do not initial complexity, cost, and risk of bleeding with CDT, have a high bleeding risk, we suggest systemically are likely to choose CDT over anticoagulation alone.
administered thrombolytic therapy over no suchtherapy (Grade 2B).
Role of Inferior Vena Cava Filter in Additionto Anticoagulation for Acute DVT or PE *22. In most patients with acute PE not associatedwith hypotension, we recommend against 17. In patients with acute DVT or PE who are treated systemically administered thrombolytic therapy with anticoagulants, we recommend against the use of an inferior vena cava (IVC) ﬁlter (Grade 1B).
*23. In selected patients with acute PE who deteri- Compression Stocking to Prevent PTS orate after starting anticoagulant therapy but have *18. In patients with acute DVT of the leg, we sug- yet to develop hypotension and who have a low gest not using compression stockings routinely to bleeding risk, we suggest systemically administered prevent PTS (Grade 2B).
thrombolytic therapy over no such therapy(Grade 2C).
Remarks: This recommendation focuses on preventionof the chronic complication of PTS and not on the Remarks: Patients with PE and without hypotension treatment of symptoms. For patients with acute or who have severe symptoms or marked cardiopulmonary chronic symptoms, a trial of graduated compression impairment should be monitored closely for stockings is often justiﬁed.
deterioration. Development of hypotension suggests that Evidence-Based Medicine thrombolytic therapy has become indicated.
mechanical and pharmacological interventions designed Cardiopulmonary deterioration (eg, symptoms, vital to lower pulmonary arterial pressure.
signs, tissue perfusion, gas exchange, cardiacbiomarkers) that has not progressed to hypotension mayalso alter the risk-beneﬁt assessment in favor of Thrombolytic Therapy in Patients With Upper thrombolytic therapy in patients initially treated with 27. In patients with acute upper extremity DVT(UEDVT) that involves the axillary or more proximal Catheter-Based Thrombus Removal for the veins, we suggest anticoagulant therapy alone over Initial Treatment of PE thrombolysis (Grade 2C).
*24. In patients with acute PE who are treated with a Remarks: Patients who (i) are most likely to beneﬁt thrombolytic agent, we suggest systemic thrombo- from thrombolysis (see text); (ii) have access to CDT; lytic therapy using a peripheral vein over CDT (iii) attach a high value to prevention of PTS; and (iv) attach a lower value to the initial complexity, cost, andrisk of bleeding with thrombolytic therapy are likely to Remarks: Patients who have a higher risk of bleeding choose thrombolytic therapy over anticoagulation alone.
with systemic thrombolytic therapy and who have accessto the expertise and resources required to do CDT 28. In patients with UEDVT who undergo are likely to choose CDT over systemic thrombolytic thrombolysis, we recommend the same intensity and duration of anticoagulant therapy as in patientswith UEDVT who do not undergo thrombolysis *25. In patients with acute PE associated with hypotension and who have (i) a high bleeding risk,(ii) failed systemic thrombolysis, or (iii) shock that islikely to cause death before systemic thrombolysis can Management of Recurrent VTE on take effect (eg, within hours), if appropriate expertise Anticoagulant Therapy and resources are available, we suggest catheter-assisted thrombus removal over no such intervention *29. In patients who have recurrent VTE on VKA therapy (in the therapeutic range) or on dabigatran,rivaroxaban, apixaban, or edoxaban (and are Remarks: Catheter-assisted thrombus removal refers believed to be compliant), we suggest switching to mechanical interventions, with or without catheter to treatment with LMWH at least temporarily Pulmonary Thromboendarterectomy for the Remarks: Recurrent VTE while on therapeutic-dose Treatment of Chronic Thromboembolic anticoagulant therapy is unusual and should prompt the Pulmonary Hypertension following assessments: (1) reevaluation of whether theretruly was a recurrent VTE; (2) evaluation of compliance *26. In selected patients with chronic thromboem- with anticoagulant therapy; and (3) consideration of bolic pulmonary hypertension (CTEPH) who are an underlying malignancy. A temporary switch to identiﬁed by an experienced thromboendarter- LMWH will usually be for at least 1 month.
ectomy team, we suggest pulmonary thromboen-darterectomy over no pulmonary *30. In patients who have recurrent VTE on long- thromboendarterectomy (Grade 2C).
term LMWH (and are believed to be compliant),we suggest increasing the dose of LMWH by about Remarks: Patients with CTEPH should be evaluated by a one-quarter to one-third (Grade 2C).
team with expertise in treatment of pulmonaryhypertension. Pulmonary thromboendarterectomy is Remarks: Recurrent VTE while on therapeutic-dose often lifesaving and life-transforming. Patients with anticoagulant therapy is unusual and should prompt the CTEPH who are not candidates for pulmonary following assessments: (1) reevaluation of whether there thromboendarterectomy may beneﬁt from other truly was a recurrent VTE; (2) evaluation of compliance with anticoagulant therapy; and (3) consideration of an particularly in relation the use of non-vitamin K oral anticoagulants (NOACs). Moreover, several VTEtreatment questions that were not addressed in the last CHEST has been developing and publishing guidelines edition have been highlighted. This article focuses on for the treatment of DVT and PE, collectively referred to new developments and ongoing controversies in the as VTE, for more than 30 years. CHEST published the treatment of VTE, updating recommendations for last (9th) edition of these guidelines in February 2012 12 topics that were included in AT9, and providing (AT9).Since then, a substantial amount of new evidence recommendations for 3 new topics. The target users of relating to the treatment of VTE has been published, this guideline are clinicians.
Study Selection, Data Abstraction, and Data Analysis Composition and Selection of Topic Panel Members The criteria for selecting the evidence were based on the Population,Intervention, Comparator, Outcome elements of the standardized The Guidelines Oversight Committee (GOC) at CHEST appointed the questions and the study design ). We followed standard editor for the guideline update. Then, the editor nominated the project processes (duplicate independent work with agreement checking and executive committee, the chair, and the remaining panelists (see disagreement resolution) for title and abstract screening, full text Acknowledgments section). The GOC approved all panelists after screening, data abstraction, and risk of bias assessment. We review of their qualiﬁcations and conﬂict of interest (COI) abstracted data on the characteristics of: study design, participants, disclosures. The 15 panelists include general internists, thrombosis intervention, control, outcomes, funding, and COI. We assessed risk specialists, pulmonologists, hematologists, and methodologists.
of bias using the Cochrane Risk of Bias Tool in randomized Throughout guideline development, panelists were required to disclose and an adapted tool for observational any potential ﬁnancial or intellectual conﬂicts of interest by topic.
When existing systematic reviews were not available or were inadequate, Financial and intellectual conﬂicts of interest were classiﬁed as we performed meta-analyses when appropriate. For each outcome of primary (more serious) or secondary (less serious) ).
interest, we calculated the risk ratios of individual studies then pooled Panelists with primary COIs were required to abstain from voting on them and assessed statistical heterogeneity using the I2 statistic. We related topic areas, but could participate in discussions provided they used a ﬁxed-effects model when pooling data from two trials, or when refrained from strong advocacy.
one of the included trials was large relative to the others. Otherwise,we used a random-effects model. We used Review Manager software Selection of Topics and Key Questions (version 5.2) to perform the meta-analyses and construct forest plots.
First, we listed all of the topic areas from AT9 and added potential new We calculated absolute effects by applying pooled relative risks to topics proposed by the panel members. Next, all panel members voted baseline risks, ideally estimated from valid prognostic observational on whether each topic should be included in the update. Finally, the data or, in the absence of the latter, from control group risks. When full panel reviewed the results of the vote and decided on the ﬁnal credible data from prognostic observational studies were not list. The panel selected a total of 15 topics: 12 "update topics" from available, we used risk estimates from control groups of RCTs AT9 and 3 "new topics." For each topic, we developed standardized included in the meta-analyses ().
questions in the Population, Intervention, Comparator, Outcomeformat ().
Assessing Quality of Evidence Based on the Grades of Recommendations, Assessment, Development,and Evaluation (GRADE) approach, quality of evidence (also known as Systematic Search certainty of evidence) is deﬁned as the extent to which our conﬁdence Systematic methods were used to search for evidence for each question.
in the effect estimate is adequate to support a recommendation.The When available, the National Library of Medicine's medical subject quality of evidence is categorized as high (A level), moderate (B level), headings keyword nomenclature was used. We searched MEDLINE or low (includes very low) (C level).The rating of the quality of via PubMed for original studies and the Cochrane Library for evidence reﬂects the strengths and limitations of the body of systematic reviews. For update topics, we searched the literature evidence and was based on the study design, risk of bias, from January 2005 to July 2014. For new topics, we searched the imprecision, inconsistency, indirectness of results, and likelihood of literature from 1946 (Medline inception) to July 2014. All searches publication bias, in addition to factors speciﬁc to observational were limited to English-language publications. We augmented Using GRADEpro software (version 3.6), we generated searches by checking reference lists of published articles and tables to summarize the judgments of the quality of the evidence and personal ﬁles, and with ongoing surveillance of the literature by the relative and absolute effects.The GRADE tables include Summary of Findings tables presented in the main text, and a moredetailed version called Evidence Proﬁles presented in the online When we identiﬁed systematic reviews, we assessed their quality according to the Assessment of Multiple Systematic Reviews tool.
recommendations to the supporting evidence.
We used those that were of highest quality and up to date as thesource of evidence. In the absence of a satisfactory systematicreview, we did our own evidence synthesis using the primary Drafting of Recommendations studies identiﬁed in AT9 and in the updated search. If the panel Following the GRADE approach, the strength of a recommendation is judged that the identiﬁed randomized controlled trials (RCTs) were deﬁned as the extent to which we can be conﬁdent that the desirable inadequate, we expanded the search to include prospective cohort effects of an intervention outweigh its undesirable effects. The strength of recommendation was categorized as strong (grade 1) or Evidence-Based Medicine weak/conditional (grade 2). In determining the strength of the recommendation (including quality of evidence and strength of recommendation, the panel considered the balance of desirable and recommendation) based on a 5-point scale derived from the GRADE undesirable consequences (typically tradeoff between recurrent VTE grid (strongly agree, weakly agree, neutral, weakly disagree, strongly and bleeding events), quality of evidence, resource implications, and Each panelist could also provide open-ended feedback on patients' average values and preferences for different outcomes and each recommendation with suggested wording edits or general management options.
remarks. To achieve consensus and be included in the ﬁnalmanuscript, The chair drafted the recommendations after the entire panel had 80% agreement (strong or weak) with a response rate of at least 75% of eligible panel members. All recommendations achieved Recommendations were then revised over a series of conference calls consensus in the ﬁrst round. We then used an iterative approach and through e-mail exchanges with the entire panel. A major aim that involved review by, and approval from, all panel members for was to ensure recommendations were speciﬁc and unambiguous.
the writing of this manuscript.
Methods for Achieving Consensus We used a modiﬁed Delphi techniqueto achieve consensus on External reviewers who were not members of the expert panel reviewed each recommendation. This technique aims to minimize group the guideline before it was published. These reviewers included content interaction bias and to maintain anonymity among respondents.
experts, a methodological expert, and a practicing clinician. The ﬁnal Using an online survey (panelists without manuscript was reviewed and approved by the CHEST GOC, the a primary COI voted their level of agreement with each CHEST Board of Regents, and the CHEST journal.
Choice of Long-Term (First 3 Months) and with VTE without cancer; and VKA may be as effective Extended (No Scheduled Stop Date) as LMWH in patients without cancer. We suggested LMWH over VKA in patients with cancer for thefollowing reasons: there is moderate-quality evidence Summary of the Evidence that LMWH was more effective than VKA in patients Phases of Anticoagulant Therapy for VTE: The need with cancer; there is a substantial rate of recurrent VTE for anticoagulant therapy in patients with proximal in patients with VTE and cancer who are treated with DVT or PE is presented in AT9.The minimum VKA; it is often harder to keep patients with cancer duration of anticoagulant therapy for DVT or PE is who are on VKA in the therapeutic range; LMWH is usually 3 months; this period of treatment is referred reliable in patients who have difﬁculty with oral therapy to as "long-term therapy."A decision to treat patients (eg, vomiting); and LMWH is easier to withhold or for longer than 3 months, which we refer to as adjust than VKA if invasive interventions are required "extended anticoagulant therapy," usually implies that or thrombocytopenia develops.
anticoagulant therapy will be continued indeﬁnitely.
One new randomized trial compared LMWH 1. In patients with proximal DVT or pulmonary (tinzaparin) with warfarin for the ﬁrst 6 months of embolism (PE), we recommend long-term (3 months) treatment in 900 cancer patients with VTE.The anticoagulant therapy over no such therapy (Grade 1B).
ﬁndings of this study are consistent with evidence inAT9 that LMWH is more effective than VKA for long- Choice of Anticoagulant for Acute and Long-Term term treatment of VTE, but that there is no difference (First 3 Months) Therapy in major bleeding or death AT9 recommendations on choice of anticoagulant Consequently, we still suggest VKA over LMWH in therapy were based on comparisons of VKA with patients without cancer, and LMWH over VKA in LMWH that were performed in the preceding two patients with cancer, and we have not changed our decand with two of the NOACs (dabigatran, assessment of the quality of evidence for either of these rivaroxthat had recently been published.
recommendations (, ).
Although we judged that there was no convincingevidence that the efﬁcacy of LMWH compared with We suggested VKA therapy or LMWH over the NOACs VKA differed between VTE patients without and with in AT9 because only two randomized trials had cancer, there are, nevertheless, reasons to make different compared a NOAC (da) with suggestions for the preferred anticoagulant in patients VKA therapy, and none had compared a NOAC with without and with cancer.We suggested VKA therapy long-term LMWH. In addition, at that time, there was over LMWH in patients without cancer for the following little experience using a NOAC for treatment of VTE reasons: injections are burdensome; LMWH is expensive; and a scarcity of long-term follow-up data to support there are low rates of recurrence with VKA in patients their efﬁcacy and safety. Since then, four new TABLE 1 ] Summary of Findings: LMWH vs VKA for Long-Term Treatment of Anticipated Absolute Effects Risk Difference with because of risk of bias (from 2 fewer to 2 more) Nonmetastatic Cancer (from 5 fewer to 6 more) (from 28 fewer to 35 more) because of risk of bias 11 fewer per 1,000 (from 5 fewer to 15 fewer) 28 fewer per 1000 (from 14 fewer to 39 fewer) 70 fewer per 1,000 (from 34 fewer to 98 fewer) because of imprecision 3 fewer per 1,000 (from 9 fewer to 6 more) 11 fewer per 1,000 (from 35 fewer to 26 more) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) is based on the assumedrisk in the comparison group and the relative effect of the intervention (and its 95% CI). CATCH ¼ Comparison of Acute Treatments in Cancer Haemostasis; GRADE ¼ Grades ofRecommendations, Assessment, Development, and Evaluation; LITE ¼ Long-term Innovations in Treatment program; LMWH ¼ low-molecular-weight heparin; RIETE ¼ RegistroInformatizado de Enfermedad Tromboembolica; RR ¼ risk ratio; UFH ¼ unfractionated heparin; VKA ¼ vitamin K antagonist. GRADE Working Group grades of evidence: Highquality: Further research is very unlikely to change our conﬁdence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on ourconﬁdence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our conﬁdence in the estimate ofeffect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.
aThe initial parenteral anticoagulation was similar in both arms for all except 1 study (Hull et in which patients randomized to LMWH received initially the same LWMH,whereas patients randomized to VKA initially received UFH.
bThe relative effect (RR; 95% CI) of LMWH vs VKA was assessed, and compared, in the subgroup of trials that enrolled patients without (Hull et al[LITE], Lopez-Beret et aland with (Deitcher et [ONCENOX], Hull et [LITE], Lee et [CLOT], Lee et al[CATCH], Lopez-Beret et Meyer et alcancer: Recurrent VTE: cancer RR 0.59(0.44-0.78) vs no cancer RR 0.99 (0.46-2.13); P ¼ .21 for subgroup difference. Major bleeding: cancer RR 0.96 (0.65-1.42) vs no cancer RR 0.43 (0.17 -1.17); P ¼ .14 for subgroupdifference. All-cause mortality: cancer RR 1.00 (0.88-1.33) vs no cancer RR 1.85 (0.59-5.77); P ¼ .29 for subgroup difference.
cOne study did not report deaths, which is unusual and could reﬂect selective reporting of outcomes.
dLow corresponds to patients without cancer and patients with nonmetastatic cancer. High corresponds to patients with metastatic cancer. These control event rates were derived fromthe Computerized Registry of Patients with Venous Thromboembolism (RIETE) registry, an ongoing prospective registry of consecutive patients with acute VTE) (Prandoni et al).
eNone of the studies was blinded, whereas the diagnosis of recurrent VTE has a subjective component and there could be a lower threshold for diagnosis of recurrent VTE in VKA-treated patients because switching the treatment of such patients to LMWH is widely practiced. At the same time, there is reluctance to diagnose recurrent VTE in patients whoare already on LMWH because there is no attractive alternative treatment option.
fRisk of recurrent VTE: Low corresponds to patients without cancer (3% estimate taken from recent large RCTs of acute treatment), intermediate to patients with local or recentlyresected cancer (appears to be consistent with Prandoni [particularly if low risk is increased to 4%]), and high to patients with locally advanced or distant metastatic cancer(Prandoni et algCI includes both no effect and harm with LMWH.
h95% CIs for the risk ratio for major bleeding includes a potentially clinically important increase or decrease with LMWH, and may also vary with the dose of LMWH used duringthe extended phase of therapyiRisk of bleeding: Low corresponds to patients without risk factor for bleeding (ie, > 75 years, cancer, metastatic disease; chronic renal or hepatic failure; platelet count <80,0000;requires antiplatelet therapy; history of bleeding without a reversible cause) (Prandoni et al,Byeth et alBibliography: Deitcher et al(ONCENOX), Hull et (LITE),Hull et al(LITE Home), Lee et al(CLOT), Lopaciuk et Lopez-Beret et al,Meyer et Romera et Lee et al(CATCH) Evidence-Based Medicine TABLE 2 ] Summary of Findings: Dabigatran vs VKA for Long-Term Treatment of Anticipated Absolute Effects Risk Difference with Dabigatran (95% CI) 0 fewer per 1,000 (from 6 fewer to 9 (from 5 fewer to 13 5 fewer per 1,000 (from 10 fewer to 2 The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). RE-COVER I ¼ Efﬁcacy and Safety ofDabigatran Compared to Warfarin for 6 Month Treatment of Acute Symptomatic Venous Thromboembolism; RE-COVER II ¼ Phase III Study TestingEfﬁcacy & Safety of Oral Dabigatran Etexilate vs Warfarin for 6 m Treatment for Acute Symptomatic Venous Thromboembolism. See legend forexpansion of other abbreviations and GRADE Working Group grades of evidence.
aPatients with acute VTE treated initially with LMWH or UFH.
bDabigatran 150 mg twice daily vs warfarin.
cCI includes values suggesting no effect and values suggesting either beneﬁt or harm.
dPooled analysis of Schulman et (RE-COVER I) and Schulman et (RE-COVER II) performed by Schulman et randomized trials have compared a NOAC (withor reduction for recurrent VTE with different NOACs has withoutinitial heparin therapy) with VKA therapy not been directly compared but, based on indirect (with initial heparin therapy) for the acute and long- comparisons, appears to be similar to all of the NO; term treatment of VTThe ﬁndings of these (5) the risk of bleeding with the NOACs, and particularly studies have been analyzed in a number of systematic intracranial bleeding, is less with the NOACs than with revincluding a network meta-analyIn VKA (6) based on patients with atrial addition, there is now extensive clinical experience using ﬁbrillation, GI bleeding may be higher with dabigatran, NOACs in patients with VTE and atrial ﬁbrillation. For rivaroxaban, and edoxaban than with VKA therapy, the comparison of each of the NOACs with VKA in the although this has not been seen in patients with initial and long-term treatment of VTE, current VTE; (7) based on indirect comparisons, the evidence for efﬁcacy is moderate or high quality, for risk of bleeding may be lower with apixaban than with the safety (risk of bleeding) is moderate or high quality, and other NOACsand (8) despite the lack of speciﬁc overall is moderate or high quality (, reversal agents for the NOACs, the risk that a major bleed will be fatal appears to be no higher for the NOACs thanfor VKA therapy.Based on less bleeding with In the 10th Edition of the Antithrombotic Guideline NOACs and greater convenience for patients and health- (AT10), the panel's overall assessment of the relative care providers, we now suggest that a NOAC is used in efﬁcacy and risk of bleeding with different anticoagulant preference to VKA for the initial and long-term treatment agents is that: (1) the risk reduction for recurrent VTE of VTE in patients without cancer. Factors that may with all of the NOACs appears to be similar to the risk inﬂuence which anticoagulant is chosen for initial and reduction with including in patients with long-term treatment of VTE are summarized in .
canc; (2) in patients with VTE and cancer, the risk This decision is also expected to be sensitive to patient reduction for recurrent VTE appears to be greater with preferences. The order of our presentation of the NOACs LMWH than with VKA thera; (3) the risk (dabigatran, rivaroxaban, apixaban, edoxaban) is based reduction for recurrent VTE with the NOACs compared on the chronology of publication of the phase 3 trials in to LMWH has not been assessed but, based on indirect VTE treatment and should not be interpreted as the comparisons, LMWH may be more effective that the guideline panel's order of preference for the use of these NOACs in patients with VTE and cancer(4) the risk TABLE 3 ] Summary of Findings: Rivaroxaban vs LMWH and VKA for Acute and Long-Term Treatment of Anticipated absolute effects No. of Participants Risk with LMWH and Risk difference with Rivaroxaban (95% CI) 1 fewer per 1,000 (from 6 fewer to 6 2 fewer per 1,000 (from 7 fewer to 5 8 fewer per 1,000 (from 3 fewer to 11 The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). EINSTEIN-DVT ¼ Oral Direct Factor XaInhibitor Rivaroxaban in Patients With Acute Symptomatic Deep Vein Thrombosis; EINSTEIN-PE ¼ Oral Direct Factor Xa Inhibitor Rivaroxaban in PatientsWith Acute Symptomatic Pulmonary Embolism. See legend for expansion of other abbreviations and GRADE Working Group grades of evidence.
aIncluded patients had acute, symptomatic, objectively veriﬁed proximal DVT of the legs or PE (unprovoked, 73%; cancer, 5%; previous VTE, 19%).
bRivaroxaban 20 mg daily for 6 or 12 mo after initial long-term therapy.
cCI includes values suggesting no effect and values suggesting either beneﬁt or harm.
dPooled analysis of Bauersachs et al(EINSTEIN-DVT) and Buller et (EINSTEIN-PE) performed by Prins et Bibliography: Prins et al *2. In patients with DVT of the leg or PE and no Remarks: Initial parenteral anticoagulation is given cancer, as long-term (ﬁrst 3 months) anticoagulant before dabigatran and edoxaban, is not given before therapy, we suggest dabigatran, rivaroxaban, apix- rivaroxaban and apixaban, and is overlapped with VKA aban, or edoxaban over vitamin K antagonist (VKA) therapy. See text for factors that inﬂuence choice of therapy (all Grade 2B).
For patients with DVT of the leg or PE and no cancer In patients with VTE and cancer ("cancer-associated who are not treated with dabigatran, rivaroxaban, thrombosis"), as noted earlier in this section, we still apixaban, or edoxaban, we suggest VKA therapy over suggest LMWH over VKA. In patients with VTE and LMWH (Grade 2C).
cancer who are not treated with LMWH, we do not have TABLE 4 ] Summary of Findings: Apixaban vs LMWH and VKA for Acute and Long-Term Treatment of Anticipated Absolute Effects Quality of the Evidence Risk Difference with Apixaban (95% CI) 4 fewer per 1,000 (from 9 fewer to 4 4 fewer per 1,000 (from 11 fewer to 5 13 fewer per 1,000 (from 8 fewer to 15fewer) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). AMPLIFY ¼ Apixiban for the InitialManagement of Pulmonary Embolism and Deep-Vein Thrombosis as First-Line Therapy; PE ¼ pulmonary embolism. See legend for expansion ofother abbreviations and GRADE Working Group grades of evidence.
aApixaban 10 mg twice daily for 7 days, followed by 5 mg twice daily for 6 mo.
bSubcutaneous enoxaparin, followed by warfarin.
cCI includes values suggesting no effect and values suggesting either beneﬁt or harm. Bibliography: Agnelli et al(AMPLIFY) Evidence-Based Medicine TABLE 5 ] Summary of Findings: Edoxaban vs VKA for Acute and Long-Term Treatment of VTE Anticipated Absolute Effects Risk Difference with Edoxaban (95% CI) (from 6 fewer to 10 more) 6 fewer per 1,000 (from 15 fewer to 7 more) 2 fewer per 1,000 (from 6 fewer to 3 more) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). See and legends for expansion ofabbreviations and GRADE Working Group grades of evidence.
aPatients with acute VTE who had initially received heparin.
bEdoxaban 60 mg once daily, or 30 mg once daily if patients with creatinine clearance of 30 to 50 mL/min or a body weight below 60 kg.
cCI includes values suggesting no effect and values suggesting either beneﬁt or harm.
dDeath, with PE not ruled out. Bibliography: Buller et al(Hokusai-VTE study) a preference for either an NOAC or VKA. In the absence (, ).Extended treatment with of direct comparisons between NOACs, and no convincing indirect evidence that one NOAC is superior to another, reduces recurrent VTE without being associated with we do not have a preference for one NOAC over another much bleeding ().These NOAC. Factors that may inﬂuence which anticoagulant studies provide moderate quality evidence that is chosen for initial and long-term treatment of VTE dabigatran is as effective and as safe as VKA for are summarized in . This decision is also expected extended treatment of VTE (, ) and to be sensitive to patient preferences.
provide moderate quality evidence that each of theNOACs are effective at preventing recurrent VTE *3. In patients with DVT of the leg or PE and cancer without being associated with a high risk of bleeding ("cancer-associated thrombosis"), as long-term (ﬁrst 3 months) anticoagulant therapy, we suggest LMWH In AT9, we suggested that if a decision was made to over VKA therapy (Grade 2C), dabigatran (Grade use extended treatment of VTE, the same 2C), rivaroxaban (Grade 2C), apixaban (Grade 2C), anticoagulant should be used as was used for the initial or edoxaban (Grade 2C).
treatment period. Our intention then was to indicatethat there was no obligation to switch from one Remarks: Initial parenteral anticoagulation is given before anticoagulant to a different one after 3 or 6 months of dabigatran and edoxaban, is not given before rivaroxaban treatment (eg, from LMWH to VKA in patients with and apixaban, and is overlapped with VKA therapy. See VTE and cancer). We have revised the wording of this text for factors that inﬂuence choice of therapy.
recommendation to make it clearer that we neitherencourage nor discourage use of the same Choice of anticoagulant for extended therapy (after anticoagulant for initial and extended therapy.
3 months and no scheduled stop date) Although we anticipate that the anticoagulant that was When AT9 was written, other than a comparison of low- used for initial treatment will often also be used for the and standard-intensity anticoagulant therapy,there extended therapy, if there are reasons to change the were no comparisons of different types of extended type of anticoagulant, this should be done. We also therapy. Since AT9, dabigatran has been compared with note that whereas apixaban 5 mg twice daily is used for VKA therapy for extended treatment of VTE and found long-term treatment, apixaban 2.5 mg twice daily is to be similarly effective but associated with less bleeding used for extended therapy.
TABLE 6 ] Factors That May Inﬂuence Which Anticoagulant Is Chosen for Initial and Long-Term Treatment of VTE Preferred Anticoagulant Qualifying Remarks More so if: just diagnosed, extensive VTE, metastatic cancer, very symptomatic; vomiting; on cancer chemotherapy.
Parenteral therapy to be Rivaroxaban; apixaban VKA, dabigatran, and edoxaban require initial parenteral Once daily oral therapy Rivaroxaban; edoxaban; Liver disease and NOACs contraindicated if INR raised because of liver disease; VKA difﬁcult to control and INR may not reﬂectantithrombotic effect.
Renal disease and NOACs and LMWH contraindicated with severe renal impairment. Dosing of NOACs with levels of renal impairment clearance <30 mL/min differ with the NOAC and among jurisdictions.
Coronary artery disease VKA, rivaroxaban, Coronary artery events appear to occur more often with apixaban, edoxaban dabigatran than with VKA. This has not been seen with theother NOACs, and they have demonstrated efﬁcacy forcoronary artery disease. Antiplatelet therapy should beavoided if possible in patients on anticoagulants because ofincreased bleeding.
Dyspepsia or history of GI Dabigatran increased dyspepsia. Dabigatran, rivaroxaban, and edoxaban may be associated with more GI bleeding thanVKA.
INR monitoring can help to detect problems. However, some patients may be more compliant with a NOAC because it isless complex.
Thrombolytic therapy use Greater experience with its use in patients treated with thrombolytic therapy Reversal agent needed Pregnancy or pregnancy Potential for other agents to cross the placenta Cost, coverage, licensing Varies among regions and INR ¼ International Normalized Ratio; NOAC ¼ non-vitamin K oral coagulant. See legend for expansion of other abbreviations.
*4. In patients with DVT of the leg or PE who receive "indeﬁnite"; no scheduled stopping date) therapy.
extended therapy, we suggest that there is no need to These four options were assessed in four subgroups of change the choice of anticoagulant after the ﬁrst VTE patients with different estimated risks of recurrence 3 months (Grade 2C).
after stopping anticoagulant therapy: (1) VTE provokedby surgery (a major transient risk factor; 3% recurrence Remarks: It may be appropriate for the choice of at 5 ; (2) VTE provoked by a nonsurgical anticoagulant to change in response to changes in the transient risk factor (eg, estrogen therapy, pregnancy, leg patient's circumstances or preferences during the long- injury, ﬂight of >8 h; 15% recurrence at 5 year term or extended phases of treatment.
(3) unprovoked (also termed "idiopathic") VTE; notmeeting criteria for provoked by a transient risk factor Duration of Anticoagulant Therapy or by cancer (30% recurrence at 5 ; and(4) VTE associated with cancer (also termed "cancer- Summary of the Evidence associated thrombosis"; 15% annualized risk of AT9 recommendations on how long VTE should be recurrence; recurrence at 5 years not estimated because treated were based on comparisons of four durations of of high mortality from cancRecurrence risk was treatment: (1) 4 or 6 weeks; (2) 3 months; (3) longer further stratiﬁed by estimating the risk of recurrence than 3 months but still a time-limited course of therapy after: (1) an isolated distal DVT was half that after a (usually 6 or 12 months); or (4) extended (also termed proximal DVT or PEand (2) a second unprovoked Evidence-Based Medicine TABLE 7 ] Summary of Findings: Dabigatran vs VKA for Extended Treatment of ,, Anticipated Absolute Effects No. of Participants Risk Difference with Dabigatran (95% CI) All-cause mortality 1 fewer per 1,000 (from 7 fewer to 9 more) (from 3 fewer to 20 more) 8 fewer per 1,000 (from 13 fewer to 0 more) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). REMEDY ¼ Secondary Prevention of VenousThrombo Embolism. See and legends for expansion of other abbreviations and GRADE Working Group grades of evidence.
aIncluded patients had acute, symptomatic, objectively veriﬁed proximal DVT of the legs or PE.
bDabigatran 150 mg twice daily taken orally for 6 mo after an initial treatment with LMWH or IV UFH.
cWarfarin adjusted to achieve an INR of 2.0-3.0 for 6 mo after an initial treatment with LMWH or IV UFH.
dActive-Control study outcomes used from Schulman et (REMEDY).
eAllocation was concealed. Patients, providers, data collectors, and outcome adjudicators were blinded. Modiﬁed intention-to-treat analysis. 1.1% loss tofollow-up. Not stopped early for beneﬁt.
fCI includes values suggesting no effect and values suggesting either beneﬁt or harm.
gPrimary end point was composite of recurrent or fatal VTE or unexplained death. Bibliography: Schulman et al(REMEDY) proximal DVT or PE was 50% higher (1.5-fold) than patient's risk of bleeding on anticoagulant therapy as low after a ﬁrst unprovoked For the decision (no bleeding risk factors; 0.8% annualized risk of about whether to stop treatment at 3 months or to treat major bleeding), moderate (one bleeding risk factor; indeﬁnitely ("extended treatment"), we categorized a 1.6% annualized risk of major bleeding), or high (two or TABLE 8 ] Summary of Findings: Dabigatran vs Placebo for Extended Treatment of VTE,, Anticipated Absolute Effects No. of Participants Risk Difference with Dabigatran (95% CI) All-cause mortality 51 fewer per 1,000 (from 42 fewer to 55 fewer) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). RESONATE ¼ Twice-daily Oral DirectThrombin Inhibitor Dabigatran Etexilate in the Long Term Prevention of Recurrent Symptomatic VTE. See and legends for expansion of otherabbreviations and GRADE Working Group grades of evidence.
aPatients with VTE who had completed at least 3 initial mo of therapy.
bDabigatran 150 mg twice daily.
cPlacebo-control study outcomes used from Schulman et (RESONATE).
dEvent rate low in a large sample size.
eEvent rate with dabigatran was 0/681 (0%); event rate with placebo was 2/662 (0.3%); anticipated absolute effect–risk difference with dabigatran is 3fewer per 1,000 (from 11 fewer to 3 more).
fEvent rate with dabigatran was 2/681 (0.3%); event rate with placebo was 0/662 (0%); anticipated absolute effect–risk difference with dabigatran is 3more per 1,000 (from 3 fewer to 11 more). Bibliography: Schulman et al(RESONATE) TABLE 9 ] Summary of Findings: Rivaroxaban vs Placebo for Extended Treatment of Anticipated Absolute Effects Risk Difference with Risk with Placebo Rivaroxaban (95% CI) 2 fewer per 1,000 (from 3 fewer to 15 more) 57 fewer per,1000 (from 42 fewer to 64 fewer) Moderatebecause of risk The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). See and legends for expansion ofother abbreviations and GRADE Working Group grades of evidence.
aPatients who had completed 6 to 12 mo of treatment for VTE.
bRivaroxaban 20 mg daily or placebo, speciﬁc to the continued treatment study.
cCI includes values suggesting no effect and values suggesting either beneﬁt or harm.
dEvent rate with rivaroxaban was 4/598 (0.67%); event rate with placebo was 0/590 (0%); anticipated absolute effect–risk difference with rivaroxaban is 4more per 1,000 (from 1 less to 17 more). Bibliography: Bauersachs et (EINSTEIN-Extension) more bleeding risk factors; $6.5% annualized risk of Comparison of Different Time-Limited Durations of major bleeding) A VKA targeted to an Anticoagulation Since AT9: Two additional studies International Normalized Ratio (INR) of about 2.5 was have compared two time-limited durations of the anticoagulant in all studies that compared different anticoagulant therapyIn patients with a ﬁrst time-limited durations of therapy. We, therefore, unprovoked PE who had completed 6 months of VKA assumed that VKA therapy was the anticoagulant when therapy (target INR 2.5), the Extended Duration of Oral we were making our AT9 recommendations, including Anticoagulant Therapy After a First Episode of for the comparison of extended therapy with stopping Idiopathic Pulmonary Embolism: a Randomized treatment at 3 months.
Controlled Trial (PADIS) study randomized patients to TABLE 10 ] Summary of Findings: Apixaban vs Placebo for Extended Treatment of VTE Anticipated Absolute Effects No. of Participants Risk Difference with Apixaban (95% CI) All-cause mortality 9 fewer per 1,000 (from 14 fewer to 4 more) 71 fewer per 1,000 (from 59 fewer to 78 fewer) 2 fewer per 1,000 (from 4 fewer to 8 more) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). See and legends for expansion ofabbreviations and GRADE Working Group grades of evidence.
aPatients with VTE who had completed 6 to 12 mo of anticoagulation therapy.
bApixaban 2.5 mg twice-daily dose vs placebo.
cSigniﬁcantly wide CIs, including appreciable beneﬁt /harm and no effect line.
dLow number of events. Bibliography: Agnelli et a(AMPLIFY-EXT) Evidence-Based Medicine TABLE 11 ] Risk Factors for Bleeding with of metastatic disease; platelet count); (2) temporal relationships Anticoagulant Therapy and Estimated Risk (eg, interval from surgery or a previous bleeding episode); and (3) how of Major Bleeding in Low-, Moderate-, and effectively a previous cause of bleeding was corrected (eg, upper GI High-Risk categories bleeding).
cImportant for parenteral anticoagulation (eg, ﬁrst 10 d), but less important for long-term or extended anticoagulation.
dAlthough there is evidence that risk of bleeding increases with the prevalence of risk factors,the categorization scheme suggested here has not been validated. Further- Previous bleeding more, a single risk factor, when severe, will result in a high risk of bleeding(eg, major surgery within the past 2 d; severe thrombocytopenia).
eCompared with low-risk patients, moderate-risk patients are assumed to Metastatic cancer have a twofold risk and high-risk patients are assumed to have aneightfold risk of major fWe estimate that anticoagulation is associated with a 2.6-fold increase in major bleeding based on comparison of extended anticoagulation with noextended anticoagulation (Table 6 in ). The relative risk of major bleeding during the ﬁrst 3 mo of therapy may be greater that during extended VKA therapy because: (1) the intensity of anticoagulation with initial parenteral therapy may be greater that with VKA therapy; (2)anticoagulant control will be less stable during the ﬁrst 3 mo; and (3) predispositions to anticoagulant-induced bleeding may be uncovered Antiplatelet therapy during the ﬁrst 3 mo of therapy.However, studies of patientswith acute coronary syndromes do not suggest a higher than 2.6 relative Poor anticoagulant control risk of major bleeding with parenteral anticoagulation (eg, UFH, LMWH) Comorbidity and reduced functional capacity compared with control.g1.6% corresponds to the average of major bleeding with initial UFH or LMWH therapy followed by VKA therapy (Table 7 in AT9We estimated baseline risk by assuming a 2.6 relative risk of major bleeding withanticoagulation (footnote f).
hConsistent with frequency of major bleeding observed by Hull in "high- Nonsteroidal anti-inﬂammatory drug iOur estimated baseline risk of major bleeding for low-risk patients (and Categorization of Risk of adjusted up for moderate- and high-risk groups as per footnote e).
jConsistent with frequency of major bleeding during prospective studies of Estimated Absolute Risk of Major extended anticoagulation for VTE (Table 6 in AT9 another 18 months of treatment or to placebo, and then followed both groups of patients for an additional 24 months after study drug was stopped The study's ﬁndings were consistent withour recommendations in AT9; the additional 18 months Baseline risk (%) of VKA was very effective at preventing recurrent VTE but, once anticoagulation was stopped, the risk of recurrent VTE was the same in those who had been treated for 6 or for 24 months. This new information has after ﬁrst 3 mo not increased the quality of evidence for comparison of a Baseline risk (%/y) longer vs a shorter, time-limited course of anticoagulation in patients without cancer.
In patients with a ﬁrst proximal DVT or PE and active cancer who had residual DVT on US imaging after AT9 ¼ 9th Edition of the Antithrombotic Guideline.
completing 6 months of LMWH therapy, the Cancer- From AT9. Since AT9, references for bleeding with individual factors have been nonsteroidal anti-in Duration of Anticoagulation based on Compression ﬂammatory drug has been added as a risk factor; a systematic review has described the risk in VTE Ultrasonography (DACUS) study randomized patients trial patients who were randomized to no antithrombotic therapy; and to another 6 months of LMWH or to stop therapy and several recent publications have compared clinical prediction rules forbleeding in various populations.
followed patients for 12 months after they stopped bMost studies assessed risk factors for bleeding in patients who were on LMWHThe additional 6 months of LMWH reduced VKA therapy. The risk of bleeding with different anticoagulants is not recurrent VTE but, once anticoagulation was stopped, addressed in this table. The increase in bleeding associated with a riskfactor will vary with: (1) severity of the risk factor (eg, location and extent the risk of recurrent VTE was the same in those who had TABLE 12 ] Summary of Findings: 6, 12, or 24 mo vs 3 or 6 mo as Minimum Duration of Anticoagulation for , Anticipated Absolute Effects No. of Participants Risk Difference with E xtended Use (95% CI) 16 more per 1,000 (from 4 fewer to 46 more) 18 fewer per 1,000 (from 40 fewer to 8 more) (from 1 fewer to 27 more) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). DACUS ¼ Warfarin Optimal Duration ItalianPulmonary Embolism; DOTAVK ¼ Durée Optimale du Traitement AntiVitamines K; WODIT-DVT ¼ Warfarin Optimal Duration Italian Deep Vein Thrombosis;WODIT-PE ¼ Warfarin Optimal Duration Italian Pulmonary Embolism. See , and legends for expansion of other abbreviations and GRADEWorking Group grades of evidence.
aStudies vary in follow-up duration (10 mo-3 y) and in duration of time-limited VKA (3-6 mo).
bVKA as NOACs are not included.
cTiming of randomization relative to the start of treatment and length of treatment varied across studies: Pinede et and Campbell et randomizedat diagnosis; and Agnelli et Eischer et and Couturaud et alrandomized after the initial 3 mo (Agnelli et al) or 6 mo (Eischer et alCouturaud et al) of treatment to stop or continued treatment. The longer duration of treatment was 6 mo in Agnelli et (provoked PE) andPinede et 12 mo in Agnelli et al(unprovoked DVT; unprovoked PE), 24 mo in Couturaud et al,and 30 mo in Eischer et Generally, studydesign was strong. No study stopped early for beneﬁt; 3 stopped early because of slow recruitment (Campbell et Pinede et al,Eischer et al) and 1because of lack of beneﬁt (Agnelli et In 1 study (Campbell et al), 20% of VTE outcomes were not objectively conﬁrmed. Patients and caregivers wereblinded in Couturaud et al,but none of the other studies was. Adjudicators of outcomes were blinded in all but 1 study (Campbell et All studies usedeffective randomization concealment, intention-to-treat analysis, and a low unexplained dropout frequency.
dStudy populations varied across studies: Pinede et alenrolled provoked and unprovoked proximal DVT and PE; Campbell et enrolled provoked andunprovoked isolated distal DVT, proximal DVT, and PE; Agnelli et had separate randomizations for provoked PE (3 vs 6 mo) and unprovoked (3 vs 12mo); Agnelli et enrolled unprovoked proximal DVT; Eischer et alenrolled unprovoked isolated DVT, proximal DVT, and PE with high levels of factorVIII; and Couturaud et aenrolled unprovoked PE.
eCIs include both values suggesting no effect and values suggesting either beneﬁt or harm. Bibliography: Campbell et Pinede et (DOTAVK), Agnelliet (WODIT-PE Provoked and Unprovoked), Agnelli et (WODIT-DVT), Couturaud et al(PADIS-PE), Siragusa et al(DACUS), Eischer et al(AUREC-FVIII) been treated for 6 or for 12 months. In the same study, stopped) in the PADIS supports AT9 estimates all patients without residual DVT after 6 months of for the efﬁcacy of extended VKA therapy.
LMWH stopped therapy and had a low risk ofrecurrence during the next year (three episodes in 91 Since AT9, two additional studies have compared patients). This study's ﬁndings have not changed our extended NOAC therapy (daapixaban recommendations for treatment of VTE in patients with with stopping treatment (ie, placebo). These two studies, and the previous study that evaluated extendedtreatment with rivaroxaban, found that extended Evaluations of Extended Anticoagulant Therapy Since therapy with these three NOAC regimens reduced AT9: When AT9 was written, extended treatment of recurrent VTE by at least 80% and was associated with a VTE with VKA therapy had been evaluated in six modest risk of bleeding , ).
studies (mostly patients with unprovoked proximal These three studies, however, enrolled heterogeneous DVT or PEor a second episode of VTE), and populations of patients (ie, not conﬁned to unprovoked with an NOAC (rivaroxaban vs placebo) in one study of VTE) and only followed patients for 6 to 12 months, heterogeneous patients.Since AT9, no studies have which limits the implications of their ﬁndings in compared extended VKA therapy with stopping relationship to extended therapy.
anticoagulants, although the large reduction in recurrentVTE with 18 additional months of VKA therapy When considering the risks and beneﬁts of extended compared with placebo (ie, before study drug was anticoagulation in this update, the AT10 panel decided Evidence-Based Medicine to use the same estimates for the reduction in recurrent how to use D-dimer testing and a patient's sex to make VTE and the increase in bleeding with anticoagulation decisions about extended therapy in patients with a ﬁrst that we used in AT9, and that were based on VKA unprovoked VTE, we have not made recommendations therapy. Our reasoning was: (1) VKA is still widely used based on these factors.
for extended treatment of VTE; (2) we felt that there was Revised Recommendations: These are unchanged from not enough evidence of differences in efﬁcacy and AT9 with one minor exception. A qualifying remark has bleeding during extended therapy to justify separate been added to the recommendation that suggests extended recommendations for NOACs, either as a group or as therapy over stopping treatment at 3 months in patients individual agents; and (3) our recommendations about with a ﬁrst unprovoked proximal DVT or PE and a low or whether or not to use extended therapy were not moderate risk of bleeding; this remark notes that patient sensitive to assuming that there was a one-third sex and D-dimer level measured a month after stopping reduction in bleeding with extended therapy compared anticoagulant therapy may inﬂuence this treatment with the estimated risk of bleeding with extended decision. If it becomes clear that, during the extended therapy that are shown in and were used in phase of treatment, there are important differences in AT9 (eg, with a NOAC compared with VKA) the risk of recurrence or bleeding with the different (the only recommendation to change would be a strong anticoagulant agents, agent-speciﬁc recommendations instead of a weak recommendation in favor of extended for extended therapy may become justiﬁed.
therapy in patients with a second unprovoked VTE whohad a moderate risk of bleeding).
5. In patients with a proximal DVT of the leg orPE provoked by surgery, we recommend treatment Better Selection of Patients for Extended VTE with anticoagulation for 3 months over (i) treatment Therapy: The most common and difﬁcult decision of a shorter period (Grade 1B), (ii) treatment of a about whether to stop anticoagulants after a time- longer, time-limited period (eg, 6, 12, or 24 months) limited course or to use extended therapy is in patients (Grade 1B), or (iii) extended therapy (no scheduled with a ﬁrst unprovoked proximal DVT or PE without a stop date) (Grade 1B).
high risk of bleeding. In this subgroup of patients,patient sex and D-dimer level measured about 1 month 6. In patients with a proximal DVT of the leg or PE after stopping anticoagulant therapy can help to further provoked by a nonsurgical transient risk factor, we stratify the risk of recurrent VTE.Men have about recommend treatment with anticoagulation for a 75% higher (1.75-fold) risk of recurrence compared 3 months over (i) treatment of a shorter period with women, whereas patients with a positive D-dimer (Grade 1B) and (ii) treatment of a longer time-limited result have about double the risk of recurrence period (eg, 6, 12, or 24 months) (Grade 1B). We compared with those with a negative D-dimer, and suggest treatment with anticoagulation for 3 months the predictive value of these two factors appears to be over extended therapy if there is a low or moderate additive. The risk of recurrence in women with a bleeding risk (Grade 2B), and recommend treatment negative posttreatment D-dimer appears to be similar for 3 months over extended therapy if there is a high to the risk that we have estimated for patients with a risk of bleeding (Grade 1B).
proximal DVT or PE that was provoked by a minor Remarks: In all patients who receive extended transient risk factor (approximately 15% recurrence at anticoagulant therapy, the continuing use of treatment 5 years); consequently, the argument for extended should be reassessed at periodic intervals (eg, annually).
anticoagulation in these women is not strong, suggestingthat D-dimer testing will often inﬂuence a woman's 7. In patients with an isolated distal DVT of the leg decision. The risk of recurrence in men with a negative provoked by surgery or by a nonsurgical transient D-dimer is not much less than the overall risk of risk factor, we suggest treatment with anticoagulation recurrence that we have estimated for patients with an for 3 months over treatment of a shorter period unprovoked proximal DVT or PE (approximately (Grade 2C); we recommend treatment with 25% compared with approximately 30% recurrence at 5 anticoagulation for 3 months over treatment of a years); consequently, the argument for extended longer, time-limited period (eg, 6, 12, or 24 months) anticoagulation in these men is still substantial, (Grade 1B); and we recommend treatment with suggesting that D-dimer testing will often not inﬂuence anticoagulation for 3 months over extended therapy a male's decision. Because there is still uncertainty about (no scheduled stop date) (Grade 1B).
Remarks: Duration of treatment of patients with isolated Remarks: In all patients who receive extended distal DVT refers to patients in whom a decision has anticoagulant therapy, the continuing use of treatment been made to treat with anticoagulant therapy; however, should be reassessed at periodic intervals (eg, annually).
it is anticipated that not all patients who are diagnosed 11. In patients with DVT of the leg or PE and active with isolated distal DVT will be prescribed cancer ("cancer-associated thrombosis") and who (i) do not have a high bleeding risk, we recommend 8. In patients with an unprovoked DVT of the leg extended anticoagulant therapy (no scheduled stop (isolated distal or proximal) or PE, we recommend date) over 3 months of therapy (Grade 1B), and treatment with anticoagulation for at least 3 months (ii) have a high bleeding risk, we suggest extended over treatment of a shorter duration (Grade 1B), and anticoagulant therapy (no scheduled stop date) over we recommend treatment with anticoagulation for 3 months of therapy (Grade 2B).
3 months over treatment of a longer, time-limited Remarks: In all patients who receive extended period (eg, 6, 12, or 24 months) (Grade 1B).
anticoagulant therapy, the continuing use of treatment Remarks: After 3 months of treatment, patients with should be reassessed at periodic intervals (eg, annually).
unprovoked DVT of the leg or PE should be evaluatedfor the risk-beneﬁt ratio of extended therapy. Duration Aspirin for Extended Treatment of VTE of treatment of patients with isolated distal DVT refers Summary of the Evidence to patients in whom a decision has been made to treatwith anticoagulant therapy; however, it is anticipated AT9 did not address if there was a role for aspirin, or that not all patients who are diagnosed with isolated antiplatelet therapy generally, in the treatment of VTE.
distal DVT will be prescribed anticoagulants.
Since then, two randomized trials have compared aspirinwith placebo for the prevention of recurrent VTE in *9. In patients with a ﬁrst VTE that is an unpro- patients with a ﬁrst unprovoked proximal DVT or PE voked proximal DVT of the leg or PE and who have a who have completed 3 to 18 months of anticoagulant (i) low or moderate bleeding risk (see text), we therapyThese trials provide moderate-quality suggest extended anticoagulant therapy (no sched- evidence that extended aspirin therapy reduces recurrent uled stop date) over 3 months of therapy (Grade 2B), VTE by about one-third. In these trials, the beneﬁts of and a (ii) high bleeding risk (see text), we recom- aspirin outweighed the increase in bleeding, which was mend 3 months of anticoagulant therapy over not statistically signiﬁcant The extended therapy (no scheduled stop date) two trials enrolled patients with a ﬁrst unprovoked VTE who did not have an increased risk of bleeding; patientsfor whom these guidelines have suggested extended Remarks: Patient sex and D-dimer level measured a anticoagulant therapy. Extended anticoagulant therapy month after stopping anticoagulant therapy may is expected to reduce recurrent VTE by more than inﬂuence the decision to stop or extend anticoagulant 80% and extended NOAC therapy may be associated therapy (see text). In all patients who receive with the same risk of bleeding as aspirin.If patients extended anticoagulant therapy, the continuing use of with a ﬁrst unprovoked VTE decline extended treatment should be reassessed at periodic intervals anticoagulant therapy because they have risk factors for (eg, annually).
bleeding or because they have a lower than average riskof recurrence, the net beneﬁt of aspirin therapy is 10. In patients with a second unprovoked VTE expected to be less than in the two trials that evaluated and who have a (i) low bleeding risk (see text), we aspirin for extended treatment of VTE.
recommend extended anticoagulant therapy (noscheduled stop date) over 3 months (Grade 1B); Based on indirect comparisons, we expect the net beneﬁt (ii) moderate bleeding risk (see text), we suggest of extended anticoagulant therapy in patients with extended anticoagulant therapy over 3 months of unprovoked VTE to be substantially greater than the therapy (Grade 2B); or (iii) high bleeding risk beneﬁts of extended aspirin therapy.Consequently, we (see text), we suggest 3 months of anticoagulant do not consider aspirin a reasonable alternative to therapy over extended therapy (no scheduled stop anticoagulant therapy in patients who want extended date) (Grade 2B).
therapy. However, if a patient has decided to stop Evidence-Based Medicine TABLE 13 ] Summary of Findings: Aspirin vs Placebo for Extended Treatment of VTE Anticipated Absolute Effects Risk with Control Risk Difference with Aspirin (95% CI) 1 fewer per 1,000 (from 3 fewer to 3 more) 60 fewer per 1,000 (from 24 fewer to 89 fewer) (from 6 fewer to 29 more) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). ASPIRE ¼ Aspirin to Prevent RecurrentVenous Thromboembolism; HR ¼ hazard ratio; INSPIRE ¼ International Collaboration of Aspirin Trials for Recurrent Venous Thromboembolism;WARFASA ¼ Aspirin for the Prevention of Recurrent Venous Thromboembolism (the Warfarin and Aspirin) study. See legend for expansion of otherabbreviations and GRADE Working Group grades of evidence.
aThe Brighton et study was stopped early and included only one-third of the intended patients.
bCI includes values suggesting no effect and values suggesting either beneﬁt or harm.
cEstimate based on Simes et al(INSPIRE) of synthesis of Brighton et al(ASPIRE) and Becattini et al(WARFASA).
dEstimate taken from Douketis et Bibliography: Simes et al(INSPIRE) anticoagulants, prevention of recurrent VTE is one of Whether and How to Prescribe Anticoagulants the beneﬁts of aspirin (may also include reductions in to Patients With Isolated Distal DVT arterial thrombosis and colon cancer) that needs to be Summary of the Evidence balanced against aspirin's risk of bleeding andinconvenience. Use of aspirin should also be reevaluated AT9 discouraged routine whole-leg US examinations when patients with VTE stop anticoagulant therapy (ie, including the distal veins) in patients with suspected because aspirin may have been stopped when DVT, thereby reducing how often isolated distal DVT is anticoagulants were started (, ).
diagnosThe rationale for not routinely examiningthe distal veins in patients who have had proximal DVT *12. In patients with an unprovoked proximal DVT excluded is that: (1) other assessment may already or PE who are stopping anticoagulant therapy and indicate that isolated distal DVT is either unlikely to be do not have a contraindication to aspirin, we suggest present or unlikely to cause complications if it is present aspirin over no aspirin to prevent recurrent VTE (eg, low clinical probability of DVT, D-dimer is negative); (2) if these conditions are not met, a repeat USexamination of the proximal veins can be done after a Remarks: Because aspirin is expected to be much less week to detect possible DVT extension and the need for effective at preventing recurrent VTE than treatment; and (3) false-positive ﬁndings for DVT occur anticoagulants, we do not consider aspirin a reasonable more often with US examinations of the distal compared alternative to anticoagulant therapy in patients who want with the proximal veins.
extended therapy. However, if a patient has decided tostop anticoagulants, prevention of recurrent VTE is one of If the calf veins are imaged (usually with US) and the beneﬁts of aspirin that needs to be balanced against isolated distal DVT is diagnosed, there are two aspirin's risk of bleeding and inconvenience. Use of management options: (1) treat patients with aspirin should also be reevaluated when patients stop anticoagulant therapy or (2) do not treat patients with anticoagulant therapy because aspirin may have been anticoagulant therapy unless extension of their DVT is stopped when anticoagulants were started.
detected on a follow-up US examination (eg, after 1 and 2 weeks, or sooner if there is concern; there is no widely deep veins for 2 weeks over anticoagulation (Grade 2C), accepted protocol for surveillance US Because and (ii) with severe symptoms or risk factors for about 15% of untreated isolated distal DVT are expected extension (see text), we suggest anticoagulation over to subsequently extend into the popliteal vein and may serial imaging of the deep veins (Grade 2C).
cause PE, it is not acceptable to neither anticoagulate nor Remarks: Patients at high risk for bleeding are more do surveillance to detect thrombus extension.
likely to beneﬁt from serial imaging. Patients who place In AT9, we judged that there was high-quality evidence a high value on avoiding the inconvenience of repeat that anticoagulant therapy was effective for the imaging and a low value on the inconvenience of treatment of proximal DVT and PE, but uncertainty that treatment and on the potential for bleeding are likely to the beneﬁts of anticoagulation outweigh its risks in choose initial anticoagulation over serial imaging.
patients with isolated distal DVT because of their lower 14. In patients with acute, isolated, distal DVT of risk of progressive or recurrent VTE. We suggest the the leg who are managed with anticoagulation, we following as risk factors for extension of distal DVT recommend using the same anticoagulation as for that would favor anticoagulation over surveillance: patients with acute proximal DVT (Grade 1B).
(1) D-dimer is positive (particularly when markedlyso without an alternative reason); (2) thrombosis is 15. In patients with acute, isolated, distal DVT of extensive (eg, >5 cm in length, involves multiple veins, the leg who are managed with serial imaging, we >7 mm in maximum diameter); (3) thrombosis is (i) recommend no anticoagulation if the thrombus close to the proximal veins; (4) there is no reversible does not extend (Grade 1B), (ii) suggest anticoagulation provoking factor for DVT; (5) active cancer; (6) history if the thrombus extends but remains conﬁned to of VTE; and (7) inpatient status.We consider the distal veins (Grade 2C), and (iii) recommend thrombosis that is conﬁned to the muscular veins of the anticoagulation if the thrombus extends into the calf (ie,, soleus, gastrocnemius) to have a lower risk of proximal veins (Grade 1B).
extension than thrombosis that involves the axial(ie, true deep; peroneal, tibial) veins.Severe CDT for Acute DVT of the Leg symptoms favor anticoagulation, a high risk for bleeding() favors surveillance, and the decision to use Summary of the Evidence anticoagulation or surveillance is expected to be sensitive At the time of AT9, there was one small randomized to patient preferences. We anticipate that isolated distal trialcomparing the effect of CDT vs anticoagulant DVT that are detected using a selective approach to alone on development of PTS, and another larger whole-leg US will often satisfy criteria for initial randomized trial (Catheter-Directed Venous anticoagulation, whereas distal DVT detected by routine Thrombolysis in Acute Iliofemoral Vein Thrombosis whole-leg US often will not.
[CAVENT] Study) assessing short-term (eg, venous The updated literature search did not identify any new patency and bleeding) but not long-term (eg, PTS) randomized trials that assessed management of patients outcomes.The CAVENT Study has since reported with isolated distal DVT. Two new systematic that CDT reduced PTS, did not alter quality of life, and reviewsand a narrative reviewaddressed appears to be cost-effective treatment of isolated distal DVT. In addition to A retrospective analysis found that CDT (3649 summarizing available data, consistent with AT9, they patients) was associated with an increase in transfusion emphasize the limitations of available evidence. In the (twofold), intracranial bleeding (threefold), PE absence of substantive new evidence, the panel endorsed (1.5-fold), and vena caval ﬁlter insertion (twofold); the AT9 recommendations without revision. The long-term outcomes and PTS were not reported.
evidence supporting these recommendations remains A single-center prospective registry found that low quality because it is not based on direct comparisons US-assisted CDT in acute iliofemoral (87 patients) of the two management strategies, and ability to predict achieved high rates of venous patency, was rarely extension of distal DVT is limited.
associated with bleeding, and that only 6% of patientshad PTS at 1 y 13. In patients with acute isolated distal DVT of theleg and (i) without severe symptoms or risk factors for This new evidence has not led to a change in our extension (see text), we suggest serial imaging of the recommendation for the use of CDT in patients with Evidence-Based Medicine TABLE 14 ] Summary of Findings: Catheter-Assisted Thrombus Removal vs Anticoagulation Alone for Acute Leg DVT Anticipated Absolute Effects Risk Difference with Catheter-Assisted Thrombus Removal (95% CI) 26 fewer per 1,000 (from 43 fewer to 54 more) Moderate-Risk Population 19 fewer per 1,000 (from 34 fewer to 12 more) Moderate-Risk Population 194 more per 1,000 (from 17 fewer to 1000 more) Moderate-Risk Population 153 fewer per 1,000 (from 265 fewer to 0 more) 191 more per 1,000 (from 41 more to 386 more) The mean quality of life in the intervention groups was 0.2 higher (2.8 lower to 3 higher) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CAVENT ¼ Catheter-Directed VenousThrombolysis in Acute Iliofemoral Vein Thrombosis; EQ-5D ¼ EuroQol – 5 Dimensions; PTS ¼ postthrombotic syndrome; QoL ¼ quality of life. See and legends for expansion of other abbreviations and GRADE Working Group grades of evidence.
aCI includes values suggesting both beneﬁt and harm.
bLow number of events.
cReported deaths from Enden et al(CAVENT).
dEstimate taken from Watson et al.The 1 study included for this outcome was Enden et (CAVENT).
eBaseline risks for nonfatal recurrent VTE and for major bleeding derived from Douketis et fMost of bleeding events occur during the ﬁrst 7 d.
gThis estimate is based on the Watson et al.The 1 study included for this outcome was Enden et (CAVENT). For PTS at 6 mo, published data fromEnden et (CAVENT) provide an estimate RR of 0.93 (0.61-1.42) via Watson et al.hThis estimate is based on the ﬁndings of the VETO iFor severe PTS, assuming the same RR of 0.46 and a baseline risk of the absolute reduction is 75 fewer severe PTS per 1,000 (from 29 fewer to138 fewer) over 2 y.
jReported patency from Enden et al(CAVENT).
kOpen-label.
lDisease-speciﬁc QoL (VEINES-QOL) estimate used at 24 mo according to treatment allocation.
mGeneric QoL (EQ-5D) at 24 mo according to treatment allocation estimate is mean difference 0.04 (-0.01 to 0.17). Bibliography: Watson et alused for alloutcomes except patency and QoL; Enden et used for patency estimates; Enden et used for QoL estimates.
DVT. Although the quality of the evidence has CDT have iliofemoral DVT, symptoms for <14 days, improved, the overall quality is still low because of very good functional status, life expectancy of $1 year, and a serious imprecision. Unchanged from AT9, we propose low risk of bleeding and , ).
that the patients who are most likely to beneﬁt from Because the balance of risks and beneﬁts with CDT is TABLE 15 ] Risk Factors for Bleeding With, and thrombolytic therapy in 5 clinical trials, the following factors were indepen- Contraindications to Use of, Thrombolytic dently associated with moderate or severe bleeding: older age (OR, 1.04 per Therapy (Both Systemic and Locally year); black (OR, 1.4); female (OR, 1.5); hypertension (OR, 1.2); lower weight (OR, 0.99 per kg).We estimate that systemic thrombolytic therapy isassociated with relative risk of major bleeding of 3.5 within 35 d Major Contraindicatio (RR, approximately 7 for intracranial bleeding); about three-quarters of theexcess of major bleeds with thrombolytic therapy occur in the ﬁrst 24 h Structural intracranial disease Previous intracranial hemorrhage Ischemic stroke within 3 mo uncertain, we consider that anticoagulant therapy alone is an acceptable alternative to CDT in all patients with Recent brain or spinal surgery acute DVT who do not have impending venous Recent head trauma with fracture or brain injury Bleeding diathesis 16. In patients with acute proximal DVT of the leg, we suggest anticoagulant therapy alone over CDT Systolic BP >180 Diastolic BP >110 Recent bleeding (nonintracranial) Remarks: Patients who are most likely to beneﬁt from CDT (see text), who attach a high value to prevention of Recent invasive procedure PTS, and a lower value to the initial complexity, cost, Ischemic stroke more than 3 mo previously and risk of bleeding with CDT, are likely to choose CDT Anticoagulated (eg, VKA therapy) over anticoagulation alone.
Traumatic cardiopulmonary resuscitation Pericarditis or pericardial ﬂuid Role of IVC Filter in Addition to Diabetic retinopathy Anticoagulation for Acute DVT or PE Summary of the Evidence Low body weight (eg, <60 kg) Our recommendation in AT9 was primarily based on ﬁndings of the Prevention du Risque d'Embolie Pulmonaire par Interruption Cave (PREPIC) randomizedtrialwhich showed that placement of a permanent See and legends for expansion of abbreviations and GRADE IVC ﬁlter increased DVT, decreased PE, and did not Working Group grades of evidence.
aThe presence of major contraindications usually precludes use of inﬂuence VTE (DVT and PE combined) or mortality. Since thrombolytic therapy; consequently, these factors have not been well then, several registries have suggested that IVC ﬁlters can studied as risk factors for bleeding associated with thrombolytic therapy.
Patients with 1 or more major contraindication are usually considered to reduce early mortality in patients with acute VTE, although be "high risk for bleeding with thrombolytic therapy." The factors listed in this evidence has been questionedThe recently this table are consistent with other recommendations for the use of published PREPIC 2 randomized trial found that thrombolytic therapy in patients with PE.bRisk factors for bleeding during anticoagulant therapy that are noted placement of an IVC ﬁlter for 3 months did not reduce in that are not included in this table are also likely to be relative recurrent PE, including fatal PE, in anticoagulated patients contraindications to thrombolytic therapy. The increase in bleeding associ- with PE and DVT who had additional risk factors for ated with a risk factor will vary with: (1) severity of the risk factor (eg, extent oftrauma or recent surgery) and (2) temporal relationships (eg, interval from recurrent VTE , This new evidence surgery or a previous bleeding episode; believed to decrease markedly after is consistent with our recommendations in AT9. However, approximately 2 wk). Risk factors for bleeding at critical sites (eg, intracranial, because it is uncertain if there is bene intraocular) or noncompressible sites are stronger contraindications for ﬁt to placement of an thrombolytic therapy. Depending on the nature, severity, temporality, and IVC ﬁlter in anticoagulated patients with severe PE (eg, number of relative contraindications, patients may be considered "high risk of with hypotension), and this is done by some experts, our bleeding with thrombolytic therapy" or "non-high risk for thrombolytictherapy.
recommendation against insertion of an IVC " Patients with no risk factors, 1-2 minor risk factors (eg, female and black race) are usually considered "low risk of bleeding with thrombolytic patients with acute PE who are anticoagulated may not therapy." Among 32,000 Medicare patients ($65 y) with myocardial infrac- apply to this select subgroup of patients.
tion who were treated with thrombolytic therapy, the following factorswere independently associated with intracranial haemorrhage: age $75 y Although the PREPIC 2 study has improved the quality of (OR, 1.6); black (OR, 1.6); female (OR, 1.4); previous stroke (OR, 1.5);systolic BP $160 mm Hg (OR, 1.8); women #65 kg or men # 80 kg (OR, 1.5); evidence for this recommendation, overall quality is still INR >4 (OR, 2.2The rate of intracranial hemorrhage increased from moderate because of imprecision ).
0.7% with 0 or 1 of these risk factors, to 4.1% with $5 risk factors. Among The AT10 panel decided against combining the results of 32,000 patients with myocardial infraction who were treated with Evidence-Based Medicine TABLE 16 ] Summary of Findings: Temporary IVC Filter vs No Temporary IVC Filter in Addition to Anticoagulation for Acute DVT or PE, Anticipated Absolute Effects Risk With No Temporary IVC Filter in Addition to Risk Difference with Temporary IVC Filter (95% CI) 15 more per 1,000 (from 24 fewer to 96 15 more per 1,000 (from 7 fewer to 104 10 fewer per 1,000 (from 34 fewer to 49 The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). IVC ¼ inferior vena cava. See and legends for expansion of other abbreviations and GRADE Working Group grades of evidence.
aAll patients received full-dose anticoagulant therapy according to guidelines for at least 6 mo.
bFilter removal was attempted in 164 patients and successful for 153 (93.3%).
cCI includes values suggesting no effect and values suggesting either beneﬁt or harm.
dSmall number of events. Bibliography: Mismetti et (PREPIC 2) the PREPIC and PREPIC 2 studies because of differences The same study found that routine use of graduated in the type of ﬁlter used, the duration of ﬁlter placement, compression stockings did not reduce leg pain during and differences in the length of follow-up.
the 3 months after DVT diagnosis , This ﬁnding, however, does not mean that 17. In patients with acute DVT or PE who are treated graduated compression stockings will not reduce acute with anticoagulants, we recommend against the use of symptoms of DVT or chronic symptoms in those who an IVC ﬁlter (Grade 1B).
have developed PTS.
Compression Stocking to Prevent PTS *18. In patients with acute DVT of the leg, we sug-gest not using compression stockings routinely to Summary of the Evidence prevent PTS (Grade 2B).
AT9 suggested routine use of graduated compressionstockings for 2 years after DVT to reduce the risk of PTS.
Remarks: This recommendation focuses on prevention That recommendation was mainly based on ﬁndings of two of the chronic complication of PTS and not on the small, single-center, randomized trials in which patients treatment of symptoms. For patients with acute or and study personnel were not blinded to stocking use (no chronic symptoms, a trial of graduated compression placebo stocking).The quality of the evidence was stockings is often justiﬁed.
moderate because of risk of bias resulting from a lack ofblinding of an outcome (PTS) that has a large subjective Whether to Treat Subsegmental PE component and because of serious imprecision of thecombined ﬁndings of the two trials , ).
Summary of the Evidence Since AT9, a much larger multicenter, placebo-controlled Subsegmental PE refers to PE that is conﬁned to the trial at low risk of bias found that routine use of graduated subsegmental pulmonary arteries. Whether these compression stockings did not reduce PTS or have other patients should be treated, a question that was not important beneﬁts.Based on this trial, we now suggest addressed in AT9, has grown in importance because that graduated compression stockings not be used improvements in CT pulmonary angiography have routinely to prevent PTS and consider the quality to the increased how often subsegmental PE is diagnosed evidence to be moderate ).
(ie, from approximately 5% to more than 10% of TABLE 17 ] Summary of Findings: Elastic Compression Stockings vs No Elastic Compression Stockings to Prevent Anticipated Absolute Effects Risk with No Elastic Risk Difference with Elastic Compression Stockings Compression Stockings (95% CI) (from 67 fewer to 86 more) 34 fewer per 1,000 (from 97 fewer to 65 more) The mean acute leg pain The mean acute leg pain in in the control groups the intervention groups was 1.13 leg pain was 0.26 higher (0.03 severity assessed lower to 0.55 higher) on an 11-pointnumerical painrating The mean QoL in the intervention groups was0.12 lower (1.11 lower to0.86 higher) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). ECS ¼ elastic compression stockings; SF-36 ¼ Short Form 36. See and legends for expansion of other abbreviations and GRADE Working Group grades of evidence.
aFor included studies, number of PTS events as assessed by Villalta's criteria bLow number of events.
cThere were 3 studies originally included for this outcome (Brandjes et Prandoni et al,and Kahn et [SOX]). There was very high heterogeneityamong the 3 studies, I2 ¼ 92% (P < .01). The pooled effect of the 3 studies was RR, 0.63 (0.35-1.13). Yet, because of the high risk of bias associated withBrandjes et and Prandoni et al,it was decided to focus on the estimate of the low-risk trial, Kahn et (SOX), which is used here.
dThis estimate is based on the ﬁndings of the VETO study.
eCI includes values suggesting no effect and values suggesting either beneﬁt or harm.
fThere were 3 studies originally included for this outcome (Brandjes et al,Prandoni et and Kahn et [SOX]). The pooled effect of the 3 studieswas RR, 0.91 (0.65-1.27). Yet, because of the high risk of bias associated with Brandjes et aland Prandoni et al,it was decided to focus on the estimateof the low-risk trial, Kahn et al(SOX), which is used here.
gThis estimate is the mean of 2 estimates derived from 2 studies: 12.4% probable/deﬁnite and 29.1% conﬁrmed VTE.
hWide CI that includes no effect.
iEstimate derived from Kahn et jEstimate based on VEINES-QOL score improvement of 5.8 points (SD, 7.5) for active ECS vs 5.9 (SD, 7.1) for placebo ECS.
kSF-36 physical component score improved by 8.4 points (SD, 13.6) for active ECS vs 9.9 (SD, 13.2) for placebo ECS (difference between groups of -1.53points, 95% CI, -3.44 to 0.39; P ¼ .12). Bibliography: Kahn et al(SOX) for PTS and recurrent VTE; Kahn et alfor acute leg pain PE).There is uncertainty whether these patients Our literature search did not identify any randomized should be anticoagulated for two reasons. First, because trials in patients with subsegmental PE. There is, the abnormalities are small, a diagnosis of subsegmental however, high-quality evidence for the efﬁcacy and PE is more likely to be a false-positive ﬁnding than a safety of anticoagulant therapy in patients with larger diagnosis of PE in the segmental or more proximal PE, and this is expected to apply similarly to patients pulmonary arSecond, because a true with subsegmental PE.Whether the risk of progressive subsegmental PE is likely to have arisen from a small or recurrent VTE is high enough to justify DVT, the risk of progressive or recurrent VTE without anticoagulation in patients with subsegmental PE is anticoagulation is expected to be lower than in patients uncertain.There were no episodes of recurrent with a larger PE.
VTE in retrospective reports that included about 60 Evidence-Based Medicine patients with subsegmental PE and no proximal DVT favors no anticoagulant therapy. The decision to and who were not anticoagulated.However, in anticoagulate or not is also expected to be sensitive to another retrospective analysis, patients with patient preferences. Patients who are not anticoagulated subsegmental PE appeared to have a similar risk of should be told to return for reevaluation if symptoms recurrent VTE during 3 months of anticoagulant persist or worsen.
therapy as patients with larger PE, and a higher risk than The evidence supporting our recommendations is low in patients who were suspected of having PE but had PE quality because of indirectness and because there is limited ability to predict which patients will have VTEcomplications without anticoagulation.
The AT10 panel endorsed that, if no anticoagulanttherapy is an option, patients with subsegmental PE *19. In patients with subsegmental PE (no should have bilateral US examinations to exclude involvement of more proximal pulmonary arteries) proximal DVT of the DVT should also be and no proximal DVT in the legs who have a (i) excluded in other high-risk locations, such as in upper low risk for recurrent VTE (see text), we suggest extremities with central venous catheters. If DVT is clinical surveillance over anticoagulation (Grade detected, patients require anticoagulation. If DVT is not 2C), and (ii) high risk for recurrent VTE (see text), detected, there is uncertainty whether patients should be we suggest anticoagulation over clinical surveillance anticoagulated. If a decision is made not to anticoagulate, there is the option of doing one or morefollow-up US examinations of the legs to detect (and Remarks: US imaging of the deep veins of both legs then treat) evolving proximal DVT.Serial testing should be done to exclude proximal DVT. Clinical for proximal DVT has been shown to be a safe surveillance can be supplemented by serial US imaging management strategy in patients with suspected PE who of the proximal deep veins of both legs to detect evolving have nondiagnostic ventilation-perfusion scans, many of DVT (see text). Patients and physicians are more likely whom are expected to have subsegmental PE.
to opt for clinical surveillance over anticoagulation if We suggest that a diagnosis of subsegmental PE is more there is good cardiopulmonary reserve or a high risk of likely to be correct (ie, a true positive) if: (1) the CT pulmonary angiogram is of high quality with goodopaciﬁcation of the distal pulmonary arteries; (2) there Treatment of Acute PE Out of the Hospital are multiple intraluminal defects; (3) defects involvemore proximal subsegmental arteries (ie, are larger); Summary of the Evidence (4) defects are seen on more than one image; (5) defects Our recommendation in AT9 was based on: (1) two are surrounded by contrast rather than appearing to be trials that randomized patients with acute PE to receive adherent to the pulmonary artery walls; (6) defects are LMWH for only 3 days in the hospitalor entirely at seen on more than one projection; (7) patients are homecompared with being treated with LMWH in symptomatic, as opposed to PE being an incidental the hospital for a longer period; (2) 15 observational ﬁnding; (8) there is a high clinical pretest probability for studies, 9 of which were prospective, that evaluated PE; and (9) D-dimer level is elevated, particularly if the treatment of acute PE out of the hospitaland increase is marked and otherwise unexplained.
(3) longstanding experience treating DVT withoutadmission to a hospital. Since AT9, no further In addition to whether or not patients truly have randomized trials have evaluated out-of-hospital subsegmental PE, we consider the following to be risk treatment of acute PE. Several additional prospective factors for recurrent or progressive VTE if patients are and retrospective observational studies have reported not anticoagulated—patients who: are hospitalized or ﬁndings consistent with earlier reports, and the ﬁndings have reduced mobility for another reason; have active of all of these studies have been included in recent meta- cancer (particularly if metastatic or being treated with analyses that have addressed treatment of acute PE out chemotherapy); or have no reversible risk factor for VTE of the hospital.
such as recent surgery. Furthermore, a lowcardiopulmonary reserve or marked symptoms that Studies that evaluated NOACs for the acute treatment of cannot be attributed to another condition favor PE did not report the proportion of patients who were anticoagulant therapy, whereas a high risk of bleeding treated entirely out of hospital, but it is probable that this was uncommon. Treatment of acute PE with a dying from PE and the highest risk of bleeding obtain NOAC that does not require initial heparin therapy the least net beneﬁt from thrombolytic therapy and are (eg, rivaroxaban, apixaban) facilitates treatment without likely to be harmed.
hospital admission. Consistent with AT9, we suggest Evidence for the Use of Thrombolytic Therapy in that patients who satisfy all of the following criteria are Patients With Acute PE: AT9 recommendations for the suitable for treatment of acute PE out of the hospital: use of thrombolytic therapy in acute PE were based on (1) clinically stable with good cardiopulmonary reserve; low-quality evidence.At that time, only about 800 (2) no contraindications such as recent bleeding, severe patients with acute PE had been randomized to receive renal or liver disease, or severe thrombocytopenia thrombolytic therapy or anticoagulant therapy alone and, (ie, <70,000/mm3); (3) expected to be compliant with consequently, estimates of efﬁcacy, safety, and overall treatment; and (4) the patient feels well enough to be mortality were very imprecise. In addition, the trials that treated at home. Clinical decision rules such as the enrolled these 800 patients had a high risk of bias and Pulmonary Embolism Severity Index (PESI), either the there was a strong suspicion that there was selective original form with score <85 or the simpliﬁed form with reporting of studies that favored thrombolytic therapy score of 0, can help to identify low-risk patients who are (ie, publication bias). Randomized trials have clearly suitable for treatment at home.However, we established that thrombolytic therapy increases bleeding consider clinical prediction rules as aids to decision- in patients with acute myocardial infarction,but that making and do not require patients to have a predeﬁned evidence was indirect when applied to patients with PE.
score (eg, low-risk PESI score) to be considered fortreatment at home. Similarly, although we do not Since AT9, two additional small, randomized trials suggest the need for routine assessment in patients with and a much larger trialhave evaluated systemic acute PE, we agree that the presence of right ventricular thrombolytic therapy in about 1,200 patients with acute dysfunction or increased cardiac biomarker levels should PE. The ﬁndings of these new studies have been discourage treatment out of the hospitThe combined with those of earlier studies in a number of quality of the evidence for treatment of acute PE at meta-analyses.These new data, by reducing home remains moderate because of marked imprecision.
imprecision for estimates of efﬁcacy and safety and the The updated recommendation has been modiﬁed to overall risk of bias, have increased the quality of the state that appropriately selected patients may be treated evidence from low to moderate for recommendations entirely at home, rather than just be discharged early.
about the use of systemic thrombolytic therapy in acutePE (, ).
*20. In patients with low-risk PE and whose homecircumstances are adequate, we suggest treatment at Most of the new evidence comes from the Pulmonary home or early discharge over standard discharge Embolism Thrombolysis trial, which randomized 1,006 (eg, after the ﬁrst 5 days of treatment) (Grade 2B).
patients with PE and right ventricular dysfunction totenecteplase and heparin or to heparin therapy alone(with placeThe most notable ﬁndings of this Systemic Thrombolytic Therapy for PE study were that thrombolytic therapy prevented Summary of the Evidence cardiovascular collapse but increased major (includingintracranial) bleeding; these beneﬁts and harms were It has long been established that systemic thrombolytic ﬁnely balanced, with no convincing net beneﬁt from therapy accelerates resolution of PE as evidenced by thrombolytic therapy. An additional ﬁnding was that more rapid lowering of pulmonary artery pressure, "rescue thrombolytic therapy" appeared to be of beneﬁt increases in arterial oxygenation, and resolution of in patients who developed cardiovascular collapse after perfusion scan defects, and that this therapy increases initially being treated with anticoagulant therapy alone.
bleeding.The net mortality beneﬁt of thrombolytictherapy in patients with acute PE, however, has been Management Implication of the Updated Evidence: uncertain and depends on an individual patient's The improved quality of evidence has not resulted in baseline (ie, without thrombolytic therapy) risk of dying substantial changes to our recommendations because: from acute PE and risk of bleeding. Patients with the (1) the new data support that the beneﬁts of systemic highest risk of dying from PE and the lowest risk of thrombolytic therapy in patients without hypotension, bleeding obtain the greatest net beneﬁt from including those with right ventricular dysfunction or an thrombolytic therapy. Patients with the lowest risk of increase in cardiac biomarkers ("intermediate-risk PE"), Evidence-Based Medicine TABLE 18 ] Summary of Findings: Systemic Thrombolytic Therapy vs Anticoagulation Alone for Acute PE Anticipated Absolute Effects Risk Difference with Systemic Thrombolytic Therapy (95% CI) 18 fewer per 1,000 (from 5 fewer to 26 fewer) 18 fewer per 1,000 (from 8 fewer to 24 fewer) 54 more per 1,000 (from 29 more to 87 more) (from 2 more to 21 more) The basis for the assumed risk (eg, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) isbased on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). See and legends for expansion ofabbreviations and GRADE Working Group grades of evidence.
aLow number of events.
bEstimate from Chatterjee et al.Other estimates from meta-analyses on this topic include Dong et OR, 0.89 (0.45-1.78); Cao et RR, 0.64(0.29-1.40); Marti et al:OR, 0.59 (0.36-0.96); Nakamura et RR, 0.72 (0.39-1.31); Chatterjee et (intermediate-risk PE only): OR, 0.46(0.25-0.92); Marti et al(intermediate-risk PE only): OR, 0.42 (0.17-1.03).
cMajority (83%) of participants in Chatterjee et alwere "moderate" risk.
dEstimate from Chatterjee et Other estimates from meta-analyses on this topic include Dong et OR, 0.63 (0.33-1.20); Cao et al:RR 0.44(0.19-1.05); Marti et al:OR, 0.50 (0.27-0.94); Nakamura et al:RR, 0.60 (0.21-1.69).
eEstimate from Chatterjee et Other estimates from meta-analyses on this topic include Dong et OR, 1.61 (0.91-2.86); Cao et al: RR, 1.16(0.51-2.60); Marti et al:OR, 2.91 (1.95-4.36); Nakamura et al:RR, 2.07 (0.58-7.35).
fEstimate from Chatterjee et al.Bibliography: Chatterjee et are largely offset by the increase in bleeding; and PE Without Hypotension: Consistent with AT9, we (2) among patients without hypotension, it is still not recommend that most patients with acute PE who do possible to conﬁdently identify those who will derive net not have hypotension are not treated with thrombolytic beneﬁt from this therapy.
therapy. However, patients with PE withouthypotension include a broad spectrum of presentations.
PE With Hypotension: Consistent with AT9, we suggest At the mild end of the spectrum are those who have that patients with acute PE with hypotension (ie, systolic minimal symptoms and minimal cardiopulmonary BP <90 mm Hg for 15 min) and without high bleeding impairment. As noted in the section "Setting for initial risk () are treated with thrombolytic therapy.
anticoagulation for PE," many of these patients can be The more severe and persistent the hypotension, and the treated entirely at home or can be discharged after a more marked the associated features of shock and brief admission. At the severe end of the spectrum are myocardial dysfunction or damage, the more compelling those with severe symptoms and more marked the indication for systemic thrombolytic therapy.
cardiopulmonary impairment (even though systolic Conversely, if hypotension is transient or less marked, BP is >90 mm Hg). In addition to clinical features not associated with features of shock or myocardial of cardiopulmonary impairment (eg, heart rate, BP, dysfunction, and if there are risk factors for bleeding, respiratory rate, jugular venous pressure, tissue physicians and patients are likely to initially choose hypoperfusion, pulse oximetry), they may have anticoagulant therapy without thrombolytic therapy. If evidence of right ventricular dysfunction on their CT thrombolytic therapy is not used and hypotension pulmonary angiogram or on echocardiography, or persists or becomes more marked, or clinical features of evidence of myocardial damage as reﬂected by shock or myocardial damage develop or worsen, increases in cardiac biomarkers (eg, troponins, brain thrombolytic therapy may then be used.
We suggest that patients without hypotension who are impairment should be monitored closely for at the severe end of the spectrum be treated with deterioration. Development of hypotension suggests that aggressive anticoagulation and other supportive thrombolytic therapy has become indicated.
measures, and not with thrombolytic therapy. These Cardiopulmonary deterioration (eg, symptoms, vital patients need to be closely monitored to ensure that signs, tissue perfusion, gas exchange, cardiac deteriorations are detected. Development of hypotension biomarkers) that has not progressed to hypotension may suggests that thrombolytic therapy has become also alter the risk-beneﬁt assessment in favor of indicated. Deterioration that has not resulted in thrombolytic therapy in patients initially treated with hypotension may also prompt the use of thrombolytic therapy. For example, there may be a progressive Catheter-Based Thrombus Removal for the increase in heart rate, a decrease in systolic BP Initial Treatment of PE (which remains >90 mm Hg), an increase in jugularvenous pressure, worsening gas exchange, signs of Summary of the Evidence shock (eg, cold sweaty skin, reduced urine output, Interventional catheter-based treatments for acute PE confusion), progressive right heart dysfunction on include delivery of CDT if there is not a high risk of echocardiography, or an increase in cardiac biomarkers.
bleeding, or catheter-based treatment without We do not propose that echocardiography or cardiac thrombolytic therapy if there is a high risk of bleeding.
biomarkers are measured routinely in all patients withPE, or in all patients with a non–low-risk PESI CDT: The most important limitation of systemic assessmenThis is because, when measured thrombolytic therapy is that it increases bleeding, routinely, the results of these assessments do not have including intracranial bleeding. CDT, because it uses a clear therapeutic implications. For example, we do not lower dose of thrombolytic drug (eg, about one-third), recommend thrombolytic therapy routinely for patients is expected to cause less bleeding at remote sites without hypotension who have right ventricular (eg, intracranial, CDT, however, may be dysfunction and an increase in cardiac biomarkers.
as or more effective than systemic thrombolytic therapy However, we encourage assessment of right ventricular for two reasons: (1) it achieves a high local concentration function by echocardiography and/or measurement of of thrombolytic drug by infusing drug directly into cardiac biomarkers if, following clinical assessment, the PE and (2) thrombus fragmentation resulting from there is uncertainty about whether patients require more placement of the infusion catheter in the thrombus or intensive monitoring or should receive thrombolytic additional maneuvers, or an increase in thrombus permeability from US delivered via the catheter, mayenhance endogenous or pharmacologic thrombolysis.
21. In patients with acute PE associated with Thrombolytic therapy is usually infused over many hypotension (eg, systolic BP < 90 mm Hg) who do not hours or overnight. In emergent situations, systemic have a high bleeding risk, we suggest systemically thrombolytic therapy can be given while CDT is being administered thrombolytic therapy over no such arranged, and active thrombus fragmentation and therapy (Grade 2B).
aspiration (see below) can be combined with CDT.
*22. In most patients with acute PE not associated A single randomized trial of 59 patients found that, with hypotension, we recommend against systemi- compared with anticoagulation alone, US-assisted CDT cally administered thrombolytic therapy (Grade 1B).
improved right ventricular function at 24 h.Observational studies also suggest that CDT is effective *23. In selected patients with acute PE who at removing thrombus, lowering pulmonary arterial deteriorate after starting anticoagulant therapy but pressure, and improving right ventricular function have yet to develop hypotension and who have a low without being associated with a high risk of bleeding risk, we suggest systemically administered bleedinMost of these studies are small (fewer thrombolytic therapy over no such therapy than 30 patients) and retrospective, although a recent prospective registry of 101 patients and a prospectivecohort study of 150 patients also support the efﬁcacy of Remarks: Patients with PE and without hypotension CDT.Whereas there was no major bleeding in the who have severe symptoms or marked cardiopulmonary registry, there were 15 episodes in the cohort study Evidence-Based Medicine (10%; no intracranial or fatal bleeds). An older expertise and resources are available, we suggest randomized trial of 34 patients with massive PE found catheter-assisted thrombus removal over no such that infusion of recombinant tissue plasminogen intervention (Grade 2C).
activator into a pulmonary artery as opposed to aperipheral vein did not accelerate thrombolysis, but Remarks: Catheter-assisted thrombus removal refers to caused more frequent bleeding at the catheter insertion mechanical interventions, with or without catheter sitNo randomized trials or observational studies have compared contemporary CDT with systemicthrombolytic therapy. For patients who requirethrombolytic therapy and do not have a high risk of Pulmonary Thromboendarterectomy in for the bleeding, the AT10 panel favored systemic thrombolytic Treatment of Chronic Thromboembolic therapy over CDT because, compared with Pulmonary Hypertension anticoagulation alone, there is a higher quality ofevidence in support of systemic thrombolytic therapy Summary of the Evidence than for CDT.
The AT9 recommendation was based on case series thathave shown marked improvements in cardiopulmonary Catheter-Based Thrombus Removal Without status after thromboendarterectomy in patients with Thrombolytic Therapy: Catheter-based mechanical chronic thromboembolic pulmonary hypertension techniques for thrombus removal involve thrombus (CTEPH).Although additional case series have fragmentation using various types of catheters, some of been reported, the quality of the evidence for which are designed speciﬁcally for this purpos thromboendarterectomy in patients with CTEPH has Fragmentation results in distal displacement of not improved.The AT10 panel decided, thrombus, with or without suctioning and removal of however, that our previous recommendation for some thrombus through the catheter. Mechanical thromboendarterectomy in selected patients with methods alone are used when thrombus removal is CTEPH was too restrictive and could contribute to indicated but there is a high risk of bleeding that suboptimal evaluation and treatment of patients with precludes thrombolytic therapy. No randomized trial or CTEPH. For example, because of improvements in prospective cohort studies have evaluated catheter-based surgical technique, it is now often possible to remove thrombus removal of PE without thrombolytic therapy.
organized thrombi from peripheral pulmonary arteries.
Evidence for the use of CDT compared with In patients with inoperable CTEPH or persistent anticoagulation alone, CDT compared with systemic pulmonary hypertension after pulmonary thrombolytic therapy, and catheter-based treatment thromboendarterectomy, there is new evidence from a without thrombolytic therapy is of low quality and our randomized trial that pulmonary vasodilator therapy recommendations are weak.
may be of beneﬁt.For these reasons, we no longeridentify central disease as a selection factor for *24. In patients with acute PE who are treated with a thromboendarterectomy in patients with CTEPH, and thrombolytic agent, we suggest systemic thrombo- we emphasize that patients with CTEPH should be lytic therapy using a peripheral vein over CDT assessed by a team with expertise in the evaluation and management of pulmonary hy Remarks: Patients who have a higher risk of bleeding *26. In selected patients with chronic thromboem- with systemic thrombolytic therapy, and who have bolic pulmonary hypertension (CTEPH) who are access to the expertise and resources required to do identiﬁed by an experienced thromboendarter- CDT, are likely to choose CDT over systemic ectomy team, we suggest pulmonary thromboen- darterectomy over no pulmonarythromboendarterectomy (Grade 2C).
*25. In patients with acute PE associated with hy-potension and who have (i) a high bleeding risk, Remarks: Patients with CTEPH should be evaluated by a (ii) failed systemic thrombolysis, or (iii) shock that is team with expertise in treatment of pulmonary likely to cause death before systemic thrombolysis hypertension. Pulmonary thromboendarterectomy is can take effect (eg, within hours), if appropriate often lifesaving and life-transforming. Patients with CTEPH who are not candidates for pulmonary with UEDVT who do not undergo thrombolysis thromboendarterectomy may beneﬁt from other mechanical and pharmacological interventions designedto lower pulmonary arterial pressure.
Management of Recurrent VTE onAnticoagulant Therapy Thrombolytic Therapy in Patients With Upper Summary of Evidence There are no randomized trials or prospective cohort Summary of the Evidence studies that have evaluated management of patients withrecurrent VTE on anticoagulant therapy. Consequently, The AT9 recommendation was based on: (1) mostly management is based on low-quality evidence and an retrospective observational studies suggesting that assessment of the probable reason for the recurrence.
thrombolysis could improve short- and long-term Risk factors for recurrent VTE while on anticoagulant venous patency, but a lack of data about whether therapy can be divided into two broad categories: thrombolysis reduced PTS of the arm; (2) occasional (1) treatment factors and (2) the patient's intrinsic risk reports of bleeding in patients with UEDVT who were of recurrence. How a new event should be treated will treated with thrombolysis, and clear evidence that depend on the reason(s) for recurrence.
thrombolysis increases bleeding in other settings; and(3) recognition that, compared to anticoagulation alone, Treatment Factors: The risk of recurrent VTE decreases thrombolytic therapy is complex and We rapidly after starting anticoagulant therapy, with a much suggest that thrombolysis is most likely to be of beneﬁt higher risk during the ﬁrst week (or month) compared in patients who meet the following criteria: severe with the second week (or month).A recurrence symptoms; thrombus involving most of the subclavian soon after starting therapy can generally be managed by vein and the axillary vein; symptoms for <14 days; good a time-limited (eg, 1 month) period of more aggressive functional status; life expectancy of $1 year; and low anticoagulant intensity (eg, switching from an oral agent risk for bleeding. We also suggested CDT over systemic back to LMWH, an increase in LMWH dose). Other thrombolysis to reduce the dose of thrombolytic drug treatment factors that are associated with recurrent VTE and the risk of bleeding. There is new moderate quality and will suggest speciﬁc approaches to management evidence that CDT can reduce PTS of the leg include: (1) was LMWH being used; (2) was the patient (, ) and that systemic thrombolysis adherent; (3) was VKA subtherapeutic; (4) was increases bleeding in patients with acute PE,and anticoagulant therapy prescribed correctly; (5) was the low-quality evidence that CDT can accelerate breakdown patient taking an NOAC and a drug that reduced of acute PE.This evidence has indirect bearing on anticoagulant effect; and (6) had anticoagulant dose thrombolysis in patients with UEDVT, but it has not been reduced (drugs other than VKA)? changed the overall quality of the evidence or our There is moderate-quality evidence that LMWH is more recommendations for use of thrombolysis in these patients.
effective than VKA therapy in patients with VTE and 27. In patients with acute upper extremity DVT cancer. A switch to full-dose LMWH, therefore, is often (UEDVT) that involves the axillary or more proximal made if there has been an unexplained recurrent VTE on veins, we suggest anticoagulant therapy alone over VKA therapy or an NOAC. If the recurrence happened thrombolysis (Grade 2C).
on LMWH, the dose of LMWH can be increased. If thedose of LMWH was previously reduced (eg, by 25% after Remarks: Patients who (i) are most likely to beneﬁt 1 month of treatment), it is usually increased to the from thrombolysis (see text); (ii) have access to CDT; previous level. If the patient was receiving full-dose (iii) attach a high value to prevention of PTS; and LMWH, the dose may be increased by about 25%. In (iv) attach a lower value to the initial complexity, cost, practice, the increase in dose is often inﬂuenced by the and risk of bleeding with thrombolytic therapy are likely LMWH preﬁlled syringe dose options that are available.
to choose thrombolytic therapy over anticoagulation Once-daily LMWH may also be switched to a twice- daily regimen, particularly if two injections are required 28. In patients with UEDVT who undergo to deliver the increase in LMWH dose. Treatment thrombolysis, we recommend the same intensity adherence, including compliance, can be difﬁcult to and duration of anticoagulant therapy as in patients assess; for example, symptoms of a recurrent DVT may Evidence-Based Medicine encourage medication adherence and a return of truly was a recurrent VTE; (2) evaluation of compliance coagulation results to the "therapeutic range." with anticoagulant therapy; and (3) consideration of anunderlying malignancy.
Patient Factors: The most important intrinsic riskfactor for recurrent VTE while on anticoagulant therapy is active cancer, with an unexplained There is substantial new evidence since AT9 about how recurrence often pointing to yet-to-be-diagnosed to treat VTE. This evidence led the panel to change disease. Antiphospholipid syndrome is also associated many of the AT9 recommendations that are included in with recurrent VTE, either because of associated this update, and has strengthened the evidence quality hypercoagulability or because a lupus anticoagulant has that underlies others that are unchanged. We now led to underdosing of VKA because of spurious increases suggest the use of NOACs over VKA for the treatment in INR results. Anticoagulated patients may be taking of VTE in patients without cancer. Although we still medications that increase the risk of thrombosis such as suggest LMWH as the preferred long-term treatment estrogens or cancer chemotherapy, in which case these for VTE and cancer, we no longer suggest VKA over treatments may be withdrawn.
NOACs in these patients. Although we note factors in A retrospective observational study found an acceptable individual patients that may favor selection of one risk of recurrence (8.6%) and major bleeding (1.4%) NOAC over another in patients without or with cancer, during 3 months of follow-up in 70 cancer patients or may favor selection of either a NOAC or VKA in with recurrent VTE while on anticoagulant therapy patients with cancer, we have not expressed an overall who either switched from VKA therapy to LMWH preference for one NOAC over another, or for either (23 patients) or had their LMWH dose increased by a NOAC or VKA in patients with cancer, because: about 25% (47 patients).If there is no reversible (1) there are no direct comparisons of different reason for recurrent VTE while on anticoagulant NOACs; (2) NOACs have not been compared with therapy, and anticoagulant intensity cannot be increased VKA in a broad spectrum of patients with VTE and because of risk of bleeding, a vena caval ﬁlter can be cancer; and (3) indirect comparisons have not shown inserted to prevent PE.However, it is not known if convincingly different outcomes with different insertion of a ﬁlter in these circumstances is worthwhile, NOACs. Another notable change in AT10 is that, and the AT10 panel consider this an option of last resort.
based on a new low risk of bias study, we now suggestthat graduated compression stocking are not routinely *29. In patients who have recurrent VTE on VKA used to prevent PTS. Recommendations that are therapy (in the therapeutic range) or on dabigatran, unchanged but are now supported by better evidence rivaroxaban, apixaban, or edoxaban (and are believed include: (1) discouragement of IVC ﬁlter use in to be compliant), we suggest switching to treatment anticoagulated patients; (2) encouragement of indeﬁnite with LMWH at least temporarily (Grade 2C).
anticoagulant therapy after a ﬁrst unprovoked PE;and (3) discouragement of thrombolytic therapy in Remarks: Recurrent VTE while on therapeutic-dose PE patients who are not hypotensive and are not anticoagulant therapy is unusual and should prompt the deteriorating on anticoagulation.
following assessments: (1) reevaluation of whether theretruly was a recurrent VTE; (2) evaluation of compliance Of the 54 recommendations that are included in the 30statements in this update, 20 (38%) are strong with anticoagulant therapy; and (3) consideration of an recommendations (Grade 1) and none is based on high- underlying malignancy. A temporary switch to LMWH quality (Grade A) evidence. The absence of high-quality will usually be for at least 1 month.
evidence highlights the need for further research to *30. In patients who have recurrent VTE on long- guide VTE treatment decisions. As new evidence term LMWH (and are believed to be compliant), we becomes available, these guidelines will need to be suggest increasing the dose of LMWH by about one- updated. Goals of our group and CHEST include quarter to one-third (Grade 2C).
transition to continually updated "living guidelines."The modular format of this update is designed to Remarks: Recurrent VTE while on therapeutic-dose facilitate this development, with individual topics and anticoagulant therapy is unusual and should prompt the questions being addressed as new evidence becomes following assessments: (1) reevaluation of whether there available. We will also facilitate implementation of our recommendations into practice by developing new and Enterprises for VTE talks. T. M. and C. S. K. have received honorariafrom Chest Enterprises for VTE Prep Courses. T. M.'s institution convenient ways to disseminate our recommendations.
has received grant funding (no salary support) from Portola This will enable achievement of another of our goals— Pharmaceuticals for the Acute Medically Ill VTE Prevention With reduction in the burden of VTE in individual patients Extended Duration Betrixaban Study (APEX) related to extendedprophylaxis against VTE with betrixaban. T. M.'s institution received and in the general population.
grant support from Bayer Pharmaceuticals for a research projectconcerning the etiology of chronic thromboembolic pulmonary hypertension. He has also authored textbook chapters related tothrombolytic interventions in patients with acute PE and pulmonary Author contributions: C. K. was the chair of the panel. C. K., E. A., thromboendarterectomy in chronic thromboembolic pulmonary A. B., J. O., D. J., and L .M. were executive committee members of the hypertension. S. M. S.'s and S. C. W.'s institution has received grant panel. C. K. and N. S. were the topic editors for "Treatment of Acute funding (no salary support) from the Canadian Institutes of Health Pulmonary Embolism Out of Hospital". C. K. and D. J. were the topic for the D-dimer Optimal Duration Study Phase II (DODS- editors for "Pulmonary Thromboendarterectomy in the Treatment of Extension), from Washington University via the National Institutes Chronic Thromboembolic Pulmonary Hypertension". E. K. and A. B.
of Health (Genetic Informatics Trial), Bayer related to VTE were the topic editors for "Compression Stocking to Prevent Post- (EINSTEIN studies), and from Bristol-Myers Squibb related to Thrombotic Syndrome". E. K. and A. B. were the topic editors for apixaban for the Secondary Prevention of Thromboembolism "Thrombolytic Therapy in Patients with Upper Extremity Deep Vein (Apixaban for the Secondary prevention of Thromboembolism: A Thrombosis". D. J. and C. S. K. were the topic editors for "Management prospective Randomized Outcome pilot study among patients with of Recurrent Venous Thromboembolism on Anticoagulant Therapy".
the AntiphosPholipid Syndrome). J. R. E. V.'s institution has received H. B. and N. S. were the topic editors for "Whether and How to grant funding (no salary support) from Bristol-Myers Squibb for Anticoagulate Patients with Isolated Distal Deep Vein Thrombosis".
evaluating the role of apixaban for long-term treatment of VTE. P. W.
M. H. and H. B. were the topic editors for "Catheter-Directed is a coinvestigator on a grant regarding the treatment of subsegmental Thrombolysis for Acute Deep Vein Thrombosis of the Leg". M. H. and PE. He has authored several studies and grants related to the long- J. V. were the topic editors for "Duration of Anticoagulant Therapy".
term and extended anticoagulation (using vitamin K antagonists and L. M. and C. S. K. were the topic editors for "Whether to Anticoagulate the direct oral anticoagulants). P. W. has received grant funding from Subsegmental Pulmonary Embolism". S. S., T. M. and P. W. were the Bristol-Myers Squibb and has received honoraria for talks from topic editors for "Catheter-Based Thrombus Removal for the Initial Bayer. E. A. A., H. B., C. K., P. W., and S. C. W. participated in the Treatment of Pulmonary Embolism". S.W. and T. M. were the topic last edition of the CHEST Antithrombotic Therapy for VTE Disease editors for "Choice of Long-Term (First 3 Months) and Extended Guidelines (AT9). None declared (A. B., J. O., N. S.).
(No Scheduled Stop Date) Anticoagulant". S. S., S. W. and J. V. werethe topic editors for "Systemic Thrombolytic Therapy for Pulmonary Role of sponsors: This study was funded in total by internal funds Embolism". L. M. and P. W. were the topic editors for "Aspirin from the American College of Chest Physicians.
for Extended Treatment of Venous Thromboembolism". L. M. and Dedication: All of the authors would like to acknowledge the C. S. K. were the topic editors for "Role of Inferior Vena Caval contributions of previous authors of the CHEST Antithrombotic Filter in Addition to Anticoagulation in Patients with Acute Deep Vein Thrombosis or Pulmonary Embolism". E. A. and J. O. weremethodologists for the panel. A. B. was the GOC liaison to the panel.
Additional information: The e-Tables and e-Figures can be found L. M. was an overall guideline editor.
in the Supplemental Materials section of the online article.
Financial/nonﬁnancial disclosures: The authors have reported toCHEST the following: In the past 3 years, E. A. A. was an author on a number of systematic reviews on anticoagulation in patients with cancer. H. B. has received compensation for participation on advisory committees with speaking engagements sponsored by Sanoﬁ-Aventis, Bayer Healthcare, and Daiichi-Sankyo. His institution has received grant funding (no salary support) from Daiichi-Sankyo for studying VTE treatment. He has also served as a coauthor of original studies using rivaroxaban (EINSTEIN, EINSTEIN Pulmonary Embolism [PE]) and edoxaban (Hokusai-VTE study). M. H. has received grant funding and has delivered talks related to long-term and extendedanticoagulation and treatment of subsegmental PE. He has also authored several papers related to long-term and extended anticoagulation, treatment of subsegmental PE, and compressionstocking in preventing postthrombotic syndrome. D. J.'s institution has received grant funding (no salary support) from Instituto de salud Carlos III, Sociedad Española de Neumología y Cirugía Torácica, and NeumoMadrid for studying PE. He was a member of Steering Committee of the Pulmonary Embolism Thrombosis Study (PEITHO), a principal investigator of an original study related to the role of the inferior vena cava ﬁlter in addition to anticoagulation in patients with acute DVT or PE and has participated in the derivation of scores for identiﬁcation of low-risk PE. He has delivered talksrelated to treatment of acute PE. C. K. has been compensated for speaking engagements sponsored by Boehringer Ingelheim and Bayer Healthcare related to VTE therapy. His institution has received grant funding (no salary support) from the National Institutes of Healthrelated to the topic of catheter-assisted thrombus removal in patients with leg DVT. He has also published many studies related to long- term anticoagulation and compression stockings in preventing postthrombotic syndrome. L. M. has frequently lectured on the duration of long-term anticoagulation and is a coauthor on several risk-stratiﬁcation papers. She has received honoraria from CHEST Evidence-Based Medicine Evidence-Based Medicine Evidence-Based Medicine Evidence-Based Medicine

Source: http://www.forotromboprofilaxis.com.uy/moduloPublicaciones/noti_3/archivosAdjuntos/ACCP_VTE_2016_GPCChest_688.pdf

dentsplyimplants.nl

Clinical & Attachment-retained restorationsASTRA TECH Implant System™ EV The ASTRA TECH Implant System EV is designed for ease of use and versatility in providing treatment solutions for your implant patients. The foundation of this evolutionary system remains the unique ASTRA TECH Implant System BioManagement Complex, which has been proven to predictably provide long-term marginal bone maintenance and esthetic results.

Slide

ATTRACTING MINING INVESTMENT n Parivodić WHY IS NATIONAL MINING ESSENTIAL Minerals are essential to every industry and every aspect of life In Europe, we need over 3 billion tons of mineral raw materials every year and demand is likely to increase to a least 4 billion tons in the medium term About 70% of EU manufacturing depends on minerals