Advertisement
Full Length Article| Volume 222, P31-39, February 2023

36-month clinical outcomes of patients with venous thromboembolism: GARFIELD-VTE

Open AccessPublished:November 24, 2022DOI:https://doi.org/10.1016/j.thromres.2022.11.016

      Abstract

      Background

      Venous thromboembolism (VTE), encompassing both deep vein thrombosis (DVT) and pulmonary embolism (PE), is a leading cause of morbidity and mortality worldwide.

      Methods

      GARFIELD-VTE is a prospective, non-interventional observational study of real-world treatment practices. We aimed to capture the 36-month clinical outcomes of 10,679 patients with objectively confirmed VTE enrolled between May 2014 and January 2017 from 415 sites in 28 countries.

      Findings

      A total of 6582 (61.6 %) patients had DVT alone, 4097 (38.4 %) had PE ± DVT. At baseline, 98.1 % of patients received anticoagulation (AC) with or without other modalities of therapy. The proportion of patients on AC therapy decreased over time: 87.6 % at 3 months, 73.0 % at 6 months, 54.2 % at 12 months and 42.0 % at 36 months. At 12-months follow-up, the incidences (95 % confidence interval [CI]) of all-cause mortality, recurrent VTE and major bleeding were 6.5 (7.0–8.1), 5.4 (4.9–5.9) and 2.7 (2.4–3.0) per 100 person-years, respectively. At 36-months, these decreased to 4.4 (4.2–4.7), 3.5 (3.2–2.7) and 1.4 (1.3–1.6) per 100 person-years, respectively. Over 36-months, the rate of all-cause mortality and major bleeds were highest in patients treated with parenteral therapy (PAR) versus oral anti-coagulants (OAC) and no OAC, and the rate of recurrent VTE was highest in patients on no OAC versus those on PAR and OAC. The most frequent cause of death after 36-month follow-up was cancer (n = 565, 48.6 %), followed by cardiac (n = 94, 8.1 %), and VTE (n = 38, 3.2 %). Most recurrent VTE events were DVT alone (n = 564, 63.3 %), with the remainder PE, (n = 236, 27.3 %), or PE in combination with DVT (n = 63, 7.3 %).

      Interpretation

      GARFIELD-VTE provides a global perspective of anticoagulation patterns and highlights the accumulation of events within the first 12 months after diagnosis. These findings may help identify treatment gaps for subsequent interventions to improve patient outcomes in this patient population.

      Keywords

      1. Introduction

      Venous thromboembolism (VTE), which includes deep-vein thrombosis (DVT) and pulmonary embolism (PE), is a leading cause of morbidity and mortality worldwide [
      ISTH Steering Committe for World Thrombosis Day
      Thrombosis: a major contributor to the global disease burden.
      ]. The annual incidence of VTE is estimated at 0.75–2.69 cases per 1000 adults per year [
      • Raskob G.E.
      • et al.
      Thrombosis: a major contributor to global disease burden.
      ], and VTE is responsible for >500,000 deaths each year in the European Union [
      • Cohen A.T.
      • et al.
      Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality.
      ]. The goal of VTE treatment is to reduce the risk of the acute and long-term complications, including mortality and recurrent VTE. AC therapy is the mainstay of VTE treatment; a minority of patients may receive initial thrombolytic or other reperfusion therapies. Anticoagulation is given for at least 3 months, and treatment is extended if the risk of recurrence of VTE exceeds the risk of bleeding with continued therapy [
      • Kearon C.
      • et al.
      Antithrombotic therapy for VTE disease: CHEST guideline and expert panel report.
      ]. Contemporary VTE treatments are based largely on the results of randomized controlled trials [
      • Hokusai V.T.E.I.
      • et al.
      Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism.
      ,
      • Schulman S.
      • et al.
      Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.
      ,
      • Bauersachs R.
      • et al.
      Oral rivaroxaban for symptomatic venous thromboembolism.
      ,
      • Agnelli G.
      • et al.
      Oral apixaban for the treatment of acute venous thromboembolism.
      ]. Whilst essential for evaluating the safety and efficacy of each treatment, the highly selected patients enrolled in such trials do not represent those in everyday clinical practice. Uncertainty persists regarding the burden and patterns of treatment of VTE in the real world, and the long-term clinical outcomes of these patients.
      The Global Anticoagulant Registry in the Field-Venous Thromboembolism (GARFIELD-VTE, ClinicalTrials.gov identifier: NCT02155491) was a global, prospective, multi-centre, observational study of VTE patients treated according to local standard practices [
      • Weitz J.I.
      Global Anticoagulant Registry in the Field - Venous Thromboembolism (GARFIELD-VTE). Rationale and design.
      ]. The primary aim of the registry was to capture the initial, long-term and extended management strategies and clinical outcomes in patients with acute VTE treated in a real-world setting. Baseline characteristics of the cohort have been previously reported [
      • Ageno W.
      Characteristics of Patients with Acute Venous Thromboembolism: Perspectives from the International, Observational, GARFIELD-VTE Registry.
      ]. This study describes the clinical outcomes (including all-cause mortality, recurrent VTE, and major bleeding) after 36-months of follow-up.

      2. Methods

      2.1 Study design and participants

      GARFIELD-VTE is a non-interventional, prospective, multicentre, observational registry designed to capture real-world outcomes of patients with acute VTE. The study design for GARFIELD-VTE has been described previously [
      • Weitz J.I.
      Global Anticoagulant Registry in the Field - Venous Thromboembolism (GARFIELD-VTE). Rationale and design.
      ] and the registry was funded by an unrestricted research grant from Bayer AG. Men and women ≥18 years of age with an objectively confirmed diagnosis of VTE within 30 days of entry into the registry were eligible for inclusion. Patients with recurrent VTE must have completed treatment for the previous VTE episode. Those with superficial vein thrombosis or participating in an interventional study that dictated treatments, or for whom long-term follow up was not possible were excluded. Patients were managed according to local practices; no specific treatments, tests, or procedures were mandated by the protocol. Decisions to initiate, continue or change treatment were solely at the discretion of the treating physicians and their patients.

      2.2 Selection of study sites

      The National Coordinating Investigator identified care settings to accurately represent the management of VTE patients in their country. The contract research organization provided a list of sites that reflected these care settings before contacting a random sample of sites from each care setting. Sites that agreed to participate were recruited after a qualification telephone call. Investigators were required to complete an educational program providing guidance on patient screening, enrolment, and follow-up in the registry.

      2.3 Data collection

      Patient data were collected using an electronic case report form (eCRF) designed by eClinicalHealth Services, Stirling, UK and submitted electronically via a secure website to the registry-coordinating centre at the Thrombosis Research Institute (TRI), London, UK. TRI was responsible for ensuring complete and accurate data collection from the medical records of enrolled patients. The GARFIELD-VTE protocol mandated that 10 % of all eCRFs are verified with source documentation, that electronic audit trails are available for all data modification, and that critical variables are subjected to additional audit. The data were extracted from the study database on 14th October 2020.

      2.4 Clinical outcomes

      The primary clinical outcomes were all-cause mortality, recurrent VTE and major bleeding. Major bleeding was defined according to the International Society of Thrombosis and Haemostasis criteria [
      • Schulman S.
      • Kearon C.
      Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients.
      ]. Non-major bleeding was defined as overt bleeding not meeting the criteria for major bleeding. Recurrent VTE consisted of both symptomatic and asymptomatic VTE and was defined as a confirmed diagnosis of a VTE event after completion of treatment for a previous VTE episode. Secondary outcomes included non-major bleeding, newly diagnosed cancer, stroke/transient ischemic attack, myocardial infarction, and hospitalization. Information was collected regarding the cause of death, reasons for hospitalization, and nature of bleeding. Cancer events diagnosed >30 days after the VTE diagnosis date were considered as cancer endpoints. Patients were characterised as having active cancer if they were diagnosed and/or receiving treatment for their cancer ≤90 days before VTE diagnosis and up to 30 days after VTE diagnosis. Patients were defined as having a history of cancer if the cancer went into remission and the patient was not receiving any cancer treatment > 90 days before the diagnosis of VTE. All other patients were classified as not having cancer. Hospitalization was defined as admission to hospital for ≥2 days. For all other outcomes, events that occurred from the day of diagnosis onwards were considered as endpoints.

      2.5 Ethics statement

      The registry is conducted in accordance with the Declaration of Helsinki and guidelines from the International Conference on Harmonisation on Good Clinical Practice (GCP) and Good Pharmaco-epidemiological Practice (GPP) and adheres to all applicable national laws and regulations. Independent ethics committee for each participating country and the hospital-based institutional review board approved the design of the registry. All patients provided written informed consent to participate. Confidentiality and anonymity of patients recruited into this registry were maintained.

      2.6 Statistical analysis

      The analyses focused on patients with 36 months of follow-up, or those who died during this period. Other than all-cause mortality, a patient can have more than one recurrence or major bleed; however, only the first occurrence of each type was considered. Patient demographics, clinical characteristics at baseline and outcomes were summarized into frequency tables (ordinal or nominal data) along with percentages of the total. Event rates are expressed in 100-person years with the associated 95 % confidence interval (CI). This study is descriptive in nature and no formal testing for significance was performed. Analyses were performed using SAS® statistical software version 9.1.3 (SAS Institute Inc., Cary, NC, USA).

      3. Results

      3.1 Patient enrolment

      A total of 11,843 patients were assessed for eligibility, of whom 10,864 were enrolled between 12th May 2014 and 4th January 2017; 979 patients were excluded because they did not meet the inclusion or exclusion criteria (n = 197), died before consent could be obtained (n = 69), or because participation was declined by the patient (n = 480) or the treating physician (n = 233). An additional 184 patients were excluded because the diagnosis of VTE was not objectively confirmed using the criteria defined by Bates et al. [
      • Bates S.M.
      • et al.
      Diagnosis of DVT: antithrombotic therapy and prevention of thrombosis, 9th ed: american College of Chest Physicians Evidence-Based Clinical Practice Guidelines.
      ] (Fig. 1). The remaining 10,679 patients were followed for 36 months and constituted the final data set for analyses.
      Fig. 1
      Fig. 1Patient enrolment flowchart.
      VTE = venous thromboembolism.

      3.2 Baseline characteristics

      The baseline characteristics have been described in detail previously [
      • Ageno W.
      • et al.
      Characteristics and Management of Patients with venous thromboembolism: the GARFIELD-VTE registry.
      ]. Briefly, a total of 6582 (61.6 %) patients had DVT alone, 4097 (38.4 %) had PE ± DVT. DVT involved the lower limb in 92.9 % of cases, with the remainder involving an upper limb (5.3 %) or the vena cava (1.8 %). Of those with lower limb DVT, 36.8 % had proximal DVT, 35.4 % had distal DVT, and 27.8 % had both proximal and distal DVT. More patients were treated in a hospital setting than in the outpatient setting (73.2 % and 26.8 %, respectively). Most patients were treated by specialists in vascular medicine (44.8 %) or internal medicine, including haematology and intensive care (43.4 %). Patients were predominantly enrolled from Europe (56.1 %), followed by Asia (17.0 %), North America/Australia (13.5 %) and Africa and the Middle East (9.8 %). The most predominant risk factors were a history of VTE (15.2 %), a history of cancer (13.3 %) recent surgery (12.5 %), hospitalization (12.1 %), or active cancer (10.2 %) (Table 1).
      Table 1Baseline characteristics.
      Variable, n (%)Overall (N = 10,679)
      VTE diagnosis
       DVT6582 (61.6 %)
       PE2457 (23.0 %)
       DVT and PE1640 (15.4 %)
      Site of DVT
       Lower limb7629 (92.9 %)
       Upper limb436 (5.3 %)
       Caval vein (inferior)100 (1.2 %)
       Caval vein (superior)49 (0.6 %)
      Type of lower limb DVT
       Proximal2780 (36.8 %)
       Distal2669 (35.4 %)
       Both2101 (27.8 %)
      Pulmonary arterial branch involved
       Main1206 (29.7 %)
       Lobar1202 (29.6 %)
       Segmental1272 (31.3 %)
       Sub-segmental387 (9.5 %)
      Care setting
       Hospital7817 (73.2 %)
       Outpatient setting2862 (26.8 %)
      Specialty
       Vascular medicine4784 (44.8 %)
       General practitioner381 (3.6 %)
       Internal medicine (haematology and intensive care)4629 (43.4 %)
       Emergency medicine275 (2.6 %)
       Cardiology605 (5.7 %)
       Missing5
      Region
       Africa and Middle East1044 (9.8 %)
       Asia1820 (17.0 %)
       Europe5990 (56.1 %)
       Latin America382 (3.6 %)
       North America/Australia1443 (13.5 %)
      Risk factors
       Prior episode of DVT and/or PE1621 (15.2 %)
       History of cancer1417 (13.3 %)
       Surgery1330 (12.5 %)
       Hospitalization1296 (12.1 %)
       Active cancer1090 (10.2 %)
       Trauma of the lower limb829 (7.8 %)
       Family history of VTE648 (6.1 %)
       Acute medical illness606 (5.7 %)
       Chronic immobilization610 (5.7 %)
       Oral contraception (females)532 (5.0 %)
       Long-haul travelling527 (4.9 %)
       Renal insufficiency391 (3.7 %)
       Recent bleed or anemia370 (3.5 %)
       Chronic heart failure338 (3.2 %)
       Known thrombophilia315 (2.9 %)
      Hormone replacement therapy (females)169 (1.6 %)
      DVT = deep vein thrombosis. MI = myocardial infarction. PE = pulmonary embolism. VTE = venous thromboembolism.

      3.3 Treatment patterns

      At baseline, 98.2 % of patients received anticoagulation (AC) with or without other modalities of therapy, with 17 % receiving parenteral AC only and 42.6 % receiving parenteral and oral AC. The proportion of patients receiving AC therapy decreased over time: 87.5 % at 3 months, 73.1 % at 6 months, 54.3 % at 12 months and 41.9 % at 36 months (Fig. 2). After 12-months, over half of all patients remaining on AC therapy were receiving a direct oral anticoagulant (DOAC) (12 months: 52.9 %, 24 months: 54.8 %, 36 months: 56.1 %), with a smaller proportion receiving a vitamin K antagonist (VKA) (12 months: 32.4 %, 24 months: 31.3 %, 36 months: 30.2 %).
      Fig. 2
      Fig. 2Anticoagulation treatment patterns at baseline and follow-up visits.
      AC = anticoagulant, VKA = vitamin K antagonist, DOAC = direct oral anticoagulant, LTFU = lost to follow-up.

      3.4 All-cause mortality

      Over 36 months, a total of 1163 (10.9 %) patients died. The rate of all-cause mortality was higher in the first month of follow-up than in the subsequent 11-months (10.69 [8.72–13.1] per 100 person-years vs. 6.54 [6.05–7.07] per 100 person-years, respectively). Similarly, the rate was higher in the first year (months 1–12) of follow-up (7.5 [6.97–8.06] per 100 person-years) than in the second year (months 13–24) (3.1 [2.75–3.50] per 100 person-years) or third year (months 25–36) of follow-up (2.0 (1.72–2.34) per 100 person-years) (Table 2). Fig. 3 shows the cumulative incidence of all-cause mortality over 36 months of follow-up and the corresponding number at risk. At each milestone, cancer was the leading cause of death; 42 % at 1-month, 53 % at 12-months, and 48.6 % at 36-months (Table 3). Over the 36-month follow-up period the event rate per 100 person-years for all-cause mortality was highest in patients who were on PAR compared to those on OAC or no OAC (12.09, 2.91, and 7.43 respectively) (Table 5).
      Table 2Clinical outcomes over 36-months follow-up. Rates are per 100-person year.
      Year 1 (0–12 months)Year 2 (13–24 months)Year 3 (25–36 months)Overall (0–36 months)
      EventsRate (95 % CI)Number of eventsRate (95 % CI)EventsRate (95 % CI)EventsRate (95 % CI)
      All-cause mortality7327.5 (6.97 to 8.06)2703.1 (2.75 to 3.5)1592.0 (1.72 to 2.34)11614.4 (4.15 to 4.66)
      Recurrent VTE
      Some recurrent VTEs were both PEs and DVT, therefore they are counted as both DVT events and PE events.
      5055.34 (4.89 to 5.82)2092.55 (2.23 to 2.92)1522.08 (1.77 to 2.43)8663.47 (3.24 to 3.7)
      Major bleed2582.68 (2.37 to 3.02)680.8 (0.63 to 1.01)370.48 (0.35 to 0.66)3631.4 (1.26 to 1.55)
      Non-Major bleed8218.9 (8.31 to 9.53)1431.8 (1.53 to 2.12)881.23 (1.00 to 1.52)10524.32 (4.07 to 4.59)
      New cancer
      Only cancer events that were diagnosed >30 days from VTE diagnosis date are considered cancer endpoints.
      2592.69 (2.38 to 3.03)1061.25 (1.03 to 1.51)720.93 (0.74 to 1.18)4371.69 (1.54 to 1.86)
      Stroke/TIA690.71 (0.56 to 0.9)500.58 (0.44 to 0.76)390.50 (0.36 to 0.68)1580.6 (0.52 to 0.70)
      MI700.72 (0.57 to 0.91)260.3 (0.2 to 0.44)330.42 (0.3 to 0.59)1290.49 (0.41 to 0.58)
      MI = myocardial infarction. TIA = transient ischemic attack. VTE = venous thromboembolism.
      a Some recurrent VTEs were both PEs and DVT, therefore they are counted as both DVT events and PE events.
      b Only cancer events that were diagnosed >30 days from VTE diagnosis date are considered cancer endpoints.
      Fig. 3
      Fig. 3Cumulative incidence curves with 95 % confidence intervals of primary outcomes over 36 months' follow-up.
      VTE = venous thromboembolism.
      Table 3Cause of death. The data are the occurrence and percent of patients with a cause of death during that time period.
      Cause of death1 month (N = 100)12 months (N = 732)36 months (N = 1163)
      Cancer42 (42)388 (53)565 (48.6)
      Cardiac10 (10)45 (6.1)94 (8.1)
      Bleed4 (4)32 (4.4)46 (4.0)
      VTE10 (10)29 (3.9)38 (3.2)
      Other23 (23)134 (18.3)222 (19.1)
      Unknown11 (11)104 (14.2)198 (17.0)
      Fatal bleeds include haemorrhagic stroke. VTE includes pulmonary embolism and other VTE complications.
      VTE = venous thromboembolism.

      3.5 Recurrent VTE

      Fig. 3 shows the cumulative incidence of recurrent VTE over 36 months of follow-up and the corresponding number at risk. Of the 866 events, 564 (65.1 %) were DVT, 236 (27.3 %) were PE and 63 (7.3 %) were both DVT and PE. Patients treated with no OAC had the highest rate per 100-person-years of recurrent VTE compared to those treated with OAC and PAR (6.26 vs 3.25 and 4.3, respectively) (Table 5).

      3.6 Major bleeding

      Fig. 3 shows the cumulative incidence of major bleeding over 36 months of follow-up and the corresponding number at risk. The most common sites of major bleeding were the upper and lower GI tract (15.4 % and 16.2 %, respectively). Haemorrhagic stroke accounted for 20 (5.5 %) major bleeds (Table 4). Patients treated with OAC had the lowest rate per 100 person-years of major bleeding compared to PAR and no OAC (1.17 vs 2.67 and 2.16, respectively) (Table 5).
      Table 4Site of major bleeding. The data are the occurrence and percent of patients with a major bleed during that time period.
      Site of major bleed12 months (N = 258)36 months (N = 363)
      Intra-spinal4 (1.6)5 (1.4)
      Intra-ocular/retinal8 (3.1)12 (3.3)
      GI upper41 (15.9)56 (15.4)
      GI lower40 (15.5)59 (16.2)
      Macroscopic haematuria13 (5)19 (5.2)
      Metro-menorrhagia124 (15.1)137 (13)
      Intra-articular3 (1.2)8 (2.2)
      Intra-muscular (no compartment syndrome)7 (2.7)7 (1.9)
      Intra-muscular (with compartment syndrome)6 (2.3)6 (1.6)
      Epistaxis3 (1.2)5 (1.4)
      Gingival3 (1.2)3 (0.8)
      Intra-peritoneal5 (1.9)7 (1.9)
      Retro-peritoneal7 (2.7)12 (3.3)
      Puncture site2 (0.8)4 (1.1)
      Skin (ecchymosis other than instrument site)3 (1.2)4 (1.1)
      Haemoptysis9 (3.5)9 (2.5)
      Haemopericardium4 (1.6)4 (1.1)
      Haemothorax4 (1.6)5 (1.4)
      Haemorrhagic stroke10 (3.9)20 (5.5)
      Other38 (14.7)63 (17.3)
      Unknown16 (6.2)19 (5.2)
      GI = gastrointestinal.
      Table 5Unadjusted events and event rates from 0 to 36 months according to treatment.
      EndpointOverall (0–36 months) (n = 10,387)PAR (n = 1788)OAC (n = 8410)No AC (n = 189)
      Number of eventsRate per 100 person-yearsNumber of eventsRate per 100 person-yearsNumber of eventsRate per 100 person-yearsNumber of eventsRate per 100 person-years
      All-cause mortality11184.3445812.096282.91327.43
      Cancer
      Only cancer events that were diagnosed >30 days from VTE diagnosis date are considered cancer endpoints.
      4301.7992.693221.5292.14
      Bleeding13245.672226.3910755.52277.14
      Non-major bleed10424.391413.968814.46205.14
      Major bleed3541.4982.672471.1792.16
      Stroke/TIA1540.6280.741160.54102.37
      Recurrent VTE
      Some recurrent VTEs were both PEs and DVT, therefore they are counted as both DVT events and PE events.
      8433.451524.36663.25256.26
      Hospital admission235910.7155218.2217509.385716.42
      MI1280.5230.61970.4581.89
      DVT event6052.441113.094752.28194.67
      PE event3041.2541.462431.1571.67
      Patients with missing treatment information were excluded from this analysis.
      DVT = deep vein thrombosis, MI = myocardial infarction, (O)AC = (oral) anticoagulant, PAR = parenteral therapy, TIA = transient ischemic attack, VTE = venous thromboembolism.
      a Only cancer events that were diagnosed >30 days from VTE diagnosis date are considered cancer endpoints.
      b Some recurrent VTEs were both PEs and DVT, therefore they are counted as both DVT events and PE events.

      3.7 Other events

      There were 1052 non-major bleeding events, equivalent to a rate of 4.32 (4.07–4.59) per 100 person-years. The most common sites of non-major bleeding were epistaxis (17 %), macroscopic haematuria (13.9 %) and lower GI tract bleeding (12 %) (Appendix Table A1).
      A total of 437 patients developed cancer (diagnosed >30 days after VTE diagnosis), at a rate of 1.69 (1.54–1.86) per 100 person-years. The rates of stroke/transient ischemic attacks (TIA) and myocardial infarctions (MI) were low; 0.6 (0.52–0.70) per 100 person-years and 0.49 (0.41–0.58) per 100 person-years, respectively. Patients treated with OAC had the lowest rate per 100 person-years of stroke/TIA (0.54), MI (0.45), DVT (2,28), and PE events (1.15) compared to patients treated with PAR (0.74, 0.61, 3.09, and 1.46, respectively) and no OAC (2.34, 1.89, 4.67, and 1.67, respectively) (Table 5).

      4. Discussion

      This observational study yielded some interesting findings. Firstly, the highest incidence of outcome events took place within the first 12 months. The incidence of all-cause mortality, recurrent VTE and major bleeding decreased after 36 months of follow-up as compared to the incidence rates at 24 and 12 months; the highest incidence of these events taking place was in the first 12 months, particularly the first month. The rates of new cancer, stroke/TIA and MI were also higher at 12 months than at 24 or 36 months. These findings are consistent with previous reports [
      • Cohen A.T.
      • et al.
      Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality.
      ,
      • Ageno W.
      • et al.
      Characteristics and Management of Patients with venous thromboembolism: the GARFIELD-VTE registry.
      ,
      • Cohen A.T.
      • G.A
      • Bauersachs R.
      The management of acute venous thromboembolism in clinical practice. Results from the European PREFER in VTE Registry.
      ].
      When comparing these findings to other registries such as the PREFER in VTE, which followed patients with acute VTE for a 12-month period, similar incidences of events such as all-cause mortality, recurrent VTE, major bleeding, non-major bleeding, and stroke/TIA were observed after this time point. Understandably, fewer patients in the PREFER in VTE registry were treated with DOACs across all countries since DOACs at that time were only launched in DACH regions, but when excluding patients from non-DOAC launched countries, a similar trend was observed over 12 months, where there was a large reduction in the proportion of patients treated with heparin and a steep increase in the proportion of patients treated with DOACs [
      • Cohen A.T.
      • G.A
      • Bauersachs R.
      The management of acute venous thromboembolism in clinical practice. Results from the European PREFER in VTE Registry.
      ]. In the RIETE registry, where follow-up data was limited to 90 days, similar trends in events such as major bleeding and VTE recurrence were also seen [].
      What this shows is that GARFIELD-VTE is complementary to and builds upon the current data that is available on clinical outcomes in VTE, highlighting that the highest incidence of events occurs in the first 12 months. These findings emphasise the clinical importance of providing optimal treatment soon after diagnosis.
      Approximately one-third of major bleeds at 36 months were localized to the upper or lower GI tract. This finding is consistent with what was seen at 12 months and supports the concept that the GI tract is the most common site of bleeding in patients receiving AC [
      • Ahmed M.
      Blood thinners and gastrointestinal endoscopy.
      ]. Almost two thirds of all episodes of recurrent VTE presented as DVTs alone, with the remainder being PE with or without associated DVT. It should also be noted that a small proportion of patients experienced a second major bleed or recurrent VTE event, which could have potentially affected their treatment regimen and/or risk of other events, however any further analysis of this was beyond the scope of the study.
      The proportion of patients who died from VTE or bleeding in this registry was small (<11 % in total). It should be noted, however, that there may be patients in whom PE as a cause of death could not be ruled out, such as patients with sudden death. In many randomized trials of VTE treatment, such deaths were classified as PE deaths. Some of the risk factors of VTE could have been identified pre-diagnosis. This study highlights the importance of preventative strategies such as conducting a robust risk assessment before elective and emergency surgery admissions and appropriate thromboprophylaxis for high-risk patients [
      • Rebecca Caroline Barker B.B.F.
      Paul Marval Venous thromboembolism: risks and prevention.
      ].
      When comparing event rates of patients on different treatments, those treated with OAC had a lower rate of all-cause mortality, major bleeding, and recurrent VTE compared to patients treated with PAR or no OAC. Although these findings are consistent with that observed in long-term safety trials of OACs [
      • Sardar P.
      • C.S
      • Mukherjee D.
      Efficacy and safety of new oral anticoagulants for extended treatment of venous thromboembolism: systematic review and meta-analyses of randomized controlled trials.
      ] they have not been reflected in such a large global observational registry and provide supportive real-world evidence of the long-term safety of OAC treatment in patients with VTE. It should also be noted that typically those that are treated with PAR would be cancer patients, so this could explain the higher rate of all-cause mortality.
      VTE is a multifaceted disease and shares many risk factors with ischemic stroke and myocardial infarction, such as hypertension, obesity, dyslipidaemia, diabetes and immobilization [
      • Glynn R.J.
      • Rosner B.
      Comparison of risk factors for the competing risks of coronary heart disease, stroke, and venous thromboembolism.
      ]. This association was first suggested in 2003, where the presence of a carotid artery plaque was shown to increase the risk of VTE by more than two fold [
      • Prandoni P.
      • et al.
      An association between atherosclerosis and venous thrombosis.
      ]. Although the observed incidence of arterial events was relatively low in GARFIELD-VTE, evidence suggests that VTE overlaps in etiology and pathophysiology with coronary artery disease, and clinicians should always consider the same cardiovascular risk reduction strategies for the prevention of VTE [
      • Piazza G.
      • Goldhaber Samuel Z.
      Venous thromboembolism and atherothrombosis an integrated approach.
      ].
      Both in this registry and in others, mortality rates were higher than those in randomized clinical trials (RCTs), likely reflecting the stringent inclusion and exclusion criteria for patient selection in the RCTs [
      • Hokusai V.T.E.I.
      • et al.
      Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism.
      ,
      • Schulman S.
      • et al.
      Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.
      ,
      • Bauersachs R.
      • et al.
      Oral rivaroxaban for symptomatic venous thromboembolism.
      ,
      • Agnelli G.
      • et al.
      Oral apixaban for the treatment of acute venous thromboembolism.
      ] and the smaller proportion of patients with cancer.
      Many of the strengths of the GARFIELD-VTE registry have been previously published, however some points worth reinterring is that the GARFIELD-VTE protocol mandated that 10 % of all eCRFs are verified with source documentation to ensure a high quality of data. This registry also required investigators to be educated in advance about methods of patient screening, enrolment, and appropriate follow-up procedures.
      Due to the observational nature of registries, GARFIELD-VTE is subject to limitations, including collection of non-randomized data, incomplete data collection and the lack of centrally adjudicated outcomes. The presence of missing data is not surprising given the random selection of sites, some of which have minimal clinical research experience, however, the proportion of missing data is small. It should also be considered that the treatment duration of anticoagulants and ultimately, patient outcomes may be influenced by whether the patients has any transient or persistent risk factors. Current guidelines recommend anticoagulation treatment for 3 months after VTE caused by transient risk factors, and longer if caused by persistent risk factors. A separate analysis, comparing 1-year outcomes in patients with transient provoking factors and unprovoked factors in the GARFIELD-VTE registry has already been published and showed that event rates of recurrent VTE and major bleeding were comparable between groups [
      • Ageno W.
      • F.A
      • Haas S.
      • Weitz J.I.
      • Goldhaber S.Z.
      • Turpie A.G.G.
      • Goto S.
      • Angchaisuksiri P.
      • Dalsgaard Nielsen J.
      • Kayani G.
      • Schellong S.
      • Bounameaux H.
      • Mantovani L.G.
      • Prandoni P.
      • Kakkar A.K.
      Provoked versus unprovoked venous thromboembolism: findings from GARFIELD-VTE.
      ], however this comparison was not conducted in this analysis.
      Despite these limitations, it is important to note that prospective registries, compared to retrospective cohort or chart review studies are usually considered to be of higher quality evidence. The key reasons for this may be that when a study is designed prospectively, clear research questions have already been defined and recall bias is limited, which is a feature typically associated with retrospective studies such as case control studies. Unlike retrospective studies using electronic health records, the data are recorded according to physician and study nurse knowledge of the patient plus the patient record, not just the mapping of patient codes such as ICD-10 or CPT codes. The prospective nature of this study also means that it was possible to estimate the population at risk of key outcomes such as recurrent VTE, major bleeding and all-cause mortality [
      • Sedgwick P.
      Prospective cohort studies: advantages and disadvantages.
      ] .
      In summary, this study provides a global perspective of patient outcomes 36 months after the diagnosis of VTE and provides a framework for future VTE studies. The high rates of death, recurrent VTE and bleeding in the first year after VTE diagnosis, particularly in the first month, highlights the importance of swift and effective VTE treatment in newly diagnosed patients. When comparing outcomes of difference treatments, the low event rates of mortality, major bleeding, and recurrent VTE seen in patients treated with OAC versus PAR and no OAC further supports their long-term safety in a real-word setting.

      Funding

      This work was supported by the Thrombosis Research Institute (London, UK).

      Declaration of competing interest

      The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Alexander G. G. Turpie: Honoraria from Bayer Pharma AG, Janssen. Sylvia Haas: Honoraria from Bayer Pharma AG, Bristol Myers Squibb, Daiichi-Sankyo, Pfizer, Portola, Sanofi. Walter Ageno: Honoraria from Boehringer Ingelheim, Bayer Pharma AG, Bristol Myers Squibb, Pfizer, Daiichi-Sankyo, Portola, Aspen, Sanofi. Research support from Bayer Pharma AG. Jeffrey I. Weitz: Research support from Canadian Institutes of Health Research, Heart and Stroke Foundation, and the Canadian Fund for Innovation. Honoraria from Alnylam, Anthos, Bayer Pharma AG, Boehringer-Ingelheim, Bristol Myers Squibb, Daiichi-Sankyo, Ionis, Janssen, Merck, Novartis Pfizer, PhaseBio, and Servier. Samuel Z. Goldhaber: Research Support from Bayer Pharma AG, Boehringer-Ingelheim, BMS, BTG EKOS, Daiichi, Janssen, NHLBI, Thrombosis Research Institute. Consultancy fees from Bayer Pharma AG, Boehringer-Ingelheim. Shinya Goto: Research funding from Ono, Bristol Myers Squibb, Sanofi, and Pfizer. Personal fees from Thrombosis Research Institute and the American Heart Association. Renato D. Lopes: Research grants and personal fees from Bristol-Myers Squibb and Pfizer, personal fees from Boehringer Ingelheim and Bayer AG, research grants from Amgen Inc, GlaxoSmithKline, Medtronic PLC, and Sanofi Aventis.
      Chern-En Chiang: Honoraria from Astrazeneca, Boehringer Ingelheim, Daiichi-Sankyo, MSD, Novartis, Pfizer, and Sanofi. Harry Gibbs: Personal fees from Pfizer, Bayer, Boehringer Ingelheim. Peter Verhamme: Research support and honoraria from Bayer Healthcare, Boehringer, Daiichi-Sankyo, Anthos Pharmaceuticals, Portola, Leo-Pharma, BMS and Pfizer. Hugo ten Cate: research support from Bayer, consultant for Alveron. Shareholder Coagulation Profile.
      Second affiliation is Center of Thrombosis and Hemostasis (CTH), Gutenberg University Medical Center, Mainz, Germany. Juan Muntaner: Speaker fees from Bayer Pharma Latin America. Sebastian Schellong: Speaker fees from Bayer Pharma AG, Boehringer-Ingelheim, Bristol Meyer Squibb, Daiichi-Sankyo, Sanofi Aventis and Pfizer. Consultancy fees from Bayer Pharma AG, Boehringer-Ingelheim, Daiichi-Sankyo, Sanofi Aventis, Aspen and Pfizer. Paolo Prandoni: Personal fees from Bayer Pharma AG, Pfizer, Daiichi-Sankyo and Sanofi. Professor Ajay K Kakkar: Research grants from Bayer Pharma AG and Sanofi. Personal fees from Anthos Therapeutics, Bayer Pharma AG, Sanofi S.A, and Alfredo E. Farjat, Henri Bounameaux, Pantep Angchaisuksiri, Gloria Kayani, Eric Tse: None.

      Acknowledgements

      Manuscript drafting assistance was provided by Shakirah Chowdhury (Thrombosis Research Institute, London, UK). SAS programming support was provided by Uma Maheshwari (Thrombosis Research Institute, London, UK).

      Appendix A

      Appendix Table A1Site of Non-major bleeds. The data are the occurrence and percent of patients with a Non-major bleed during that time period.
      Site of major bleed12 months (N = 821)36 months (N = 1053)
      Intra-spinal0 (0.0)0 (0.0)
      Intra-ocular/retinal0 (0.0)0 (0.0)
      GI upper25 (3)32 (3)
      GI lower96 (11.7)126 (12)
      Macroscopic haematuria114 (13.9)143 (13.6)
      Abnormal uterine (metorrhagia)35 (4.3)43 (4.1)
      Menorrhagia89 (10.8)94 (8.9)
      Intra-articular0 (0.0)0 (0.0)
      Intra-muscular (no compartment syndrome)15 (1.8)21 (2)
      Intra-muscular (with compartment syndrome)0 (0.0)0 (0.0)
      Epistaxis140 (17.1)179 (17)
      Gingival44 (5.4)54 (5.1)
      Intra peritoneal1 (0.1)2 (0.2)
      Retro-peritoneal0 (0.0)0 (0.0)
      Puncture site20 (2.4)26 (2.5)
      Skin (ecchymosis other than puncture site)86 (10.5)121 (11.5)
      Haemoptysis34 (4.1)48 (4.6)
      Haemopericardium0 (0.0)0 (0.0)
      Haemothorax2 (0.2)3 (0.3)
      Haemorrhagic stroke0 (0.0)0 (0.0)
      Other96 (11.7)135 (12.8)
      Unknown9 (1.1)9 (0.9)

      A full list of GARFIELD-VTE Principal Investigators

      Ab Loualidi, Abdurrahim Colak, Abraham Bezuidenhout, Abu Abdool-Carrim, Addala Azeddine, Adriaan Beyers, Adriaan Dees, Ahmed Mohamed, Ahmet Aksoy, Akihiko Abiko, Akinori Watanabe, Alan Krichell, Alberto Alfredo Fernandez, Alberto Tosetto, Alexey Khotuntsov, Alisha Oropallo, Alison Slocombe, Allan Kelly, Amanda Clark, Amr Gad, Amy Arouni, Andor Schmidt, Andrea Berni, Andres Javier Kleiban, Andrew Machowski, Andrey Kazakov, Angel Galvez, Ann Lockman, Anna Falanga, Anoop Chauhan, Antoni Riera-Mestre, Antonino Mazzone, Armando D'Angelo, Artur Herdy, Atsushi Kato, Ayman Abd Elhamid Ebrahim Mahmoud Salem, Azlan Husin, Barbara Erdelyi, Barry Jacobson, Beatrice Amann-Vesti, Bektas Battaloglu, Benedicte Wilson, Benilde Cosmi, Bergmann Jean Francois, Berremeli Toufek, Beverley Hunt, Bhavesh Natha, Bisher Mustafa, Bonnie Chi Shan Kho, Boulon Carine, Brian Zidel, Brisot Dominique, Brousse Christophe, Bruno Trimarco, Canhua Luo, Carlos Alberto Cuneo, Carlos Jerjes Sanchez Diaz, Carsten Schwencke, Cas Cader, Celal Yavuz, Cesar Javier Zaidman, Charles Lunn, Chau-Chung Wu, Cheng Hock Toh, Chern-En Chiang, Chevrier Elisa, Chien-Hsun Hsia, Chien-Lung Huang, Chi-Hang Kevin Kwok, Chih-Cheng Wu, Chi-Hung Huang, Chris Ward, Christian Opitz, Christina Jeanneret-Gris, Chung Yin Ha, Chun-Yao Huang, Claude Luyeye Bidi, Clifford Smith, Cornelia Brauer, Corrado Lodigiani, Couturaud Francis, Cynthia Wu, Daniel Staub, Daniel Theodoro, Daniela Poli, David-Riesco Acevedo, David Adler, David Jimenez, David Keeling, David Scott, Davide Imberti, Desmond Creagh, Desmurs-Clavel Helene, Dirk Hagemann, Dirk Le Roux, Dirk Skowasch, Dmitry Belenky, Dmitry Dorokhov, Dmitry Petrov, Dmitry Zateyshchikov, Domenico Prisco, Dorthe Møller, Dusan Kucera, Ehab M. Esheiba, Elizaveta Panchenko, Elkouri Dominique, Emre Dogan, Emre Kubat, Enrique Diaz Diaz, Eric Wai Choi Tse, Erik Yeo, Erman Hashas, Ernst Grochenig, Eros Tiraferri, Erwin Blessing, Escande Orthlieb Michèle, Esther Usandizaga, Ettore Porreca, Fabian Ferroni, Falvo Nicolas, Félix Ayala-Paredes, Firas Koura, Fitjerald Henry, Franco Cosmi, Frans Erdkamp, Gadel Kamalov, Garcia-Bragado Dalmau, Garrigues Damien, Garry Klein, Gaurand Shah, Geert Hollanders, Geno Merli, Georg Plassmann, George Platt, Germain Poirier, German Sokurenko, Ghassan Haddad, Gholam Ali, Giancarlo Agnelli, Gin Gin Gan, Grace Kaye-Eddie, Gregoire Le Gal, Gregory Allen, Guillermo Antonio Llamas Esperón, Guillot Jean-Paul, Hagen Gerofke, Hallah Elali, Hana Burianova, Hans-Juergen Ohler, Haofu Wang, Harald Darius, Harinder S. Gogia, Harry Striekwold, Harry Gibbs, Hatice Hasanoglu, Hatice Turker, Hendrik Franow, Henri Bounameaux, Herbert De Raedt, Herman Schroe, Hesham Salah ElDin, Hesham Zidan, Hiroaki Nakamura, Ho Young Kim, Holger Lawall, Hong Zhu, Hongyan Tian, Ho-Young Yhim, Hugo ten Cate, Hun Gyu Hwang, Hyeok Shim, Igor Kim, Igor Libov, Igor Sonkin, Igor Suchkov, Ik-Chan Song, Ilker Kiris, Ilya Staroverov, Irene Looi, Isabel M De La Azuela Tenorio, Ismail Savas, Ivan Gordeev, Ivo Podpera, Jae Hoon Lee, Jameela Sathar, James Welker, Jan Beyer-Westendorf, Jan Kvasnicka, Jan Vanwelden, JangYong Kim, Jaromira Svobodova, Jaspal Gujral, Javier Marino, Javier Tristan Galvar, Jeannine Kassis, Jen-Yuan Kuo, Jhih-Yuan Shih, JiHyun Kwon, Jin Hyun Joh, Jin Hyun Park, Jin Seok Kim, Jinghua Yang, Jiri Krupicka, Jiri Lastuvka, Jiri Pumprla, Jiri Vesely, Joan Carlos Souto, João Antônio Correa, Johan Duchateau, John Perry Fletcher, Jorge del Toro, Jorge del Toro, Jorge Guillermo Chavez Paez, Jørn Nielsen, Jose Dalmo Araujo Filho, Jose Saraiva, Jose Antonio Diaz Peromingo, Jose Gomez Lara, Jose Luis Fedele, Jose Maria Surinach, Joseph Chacko, Juan Antonio Muntaner, Juan Carlos Álvarez Benitez, Juan Moreno Hoyos Abril, Julian Humphrey, Julio Bono, Junji Kanda, Juree Boondumrongsagoon, Kai Hang Yiu, Kanchana Chansung, Karin Boomars, Kate Burbury, Katsuhiro Kondo, Kemal Karaarslan, Kensuke Takeuchi, Knut Kroeger, Konstantin Zrazhevskiy, Koscál Svatopluk, Kou-Gi Shyu, Kristel Vandenbosch, Kuan-Cheng Chang, Kuan-Ming Chiu, Kubina Jean-Manuel, Kwan Jing Wern, Kwo-Chang Ueng, Lalita Norasetthada, Laure Binet, Lee Ping Chew, Lei Zhang, Leone Maria Cristina, Lidwine Tick, Lilia Beatriz Schiavi, Lily Lee Lee Wong, Lohana Borges, Louis Botha, Luc Capiau, Luc Timmermans, Luciano Eduardo López, Luigi Ria, Luis Manuel hernandez Blasco, Luis Alberto Guzman, Luis Flota Cervera, Mahe Isabelle, Manuel Monreal Bosch, Manuel de los Rios Ibarra, Manuel Núñez Fernandez, Marc Carrier, Marcelo Raul Barrionuevo, Marco Antonio Alcocer Gamba, Marco Cattaneo, Marco Moia, Margaret Bowers, Mariam Chetanachan, Mario Alberto Berli, Mark Fixley, Markus Faghih, Markus Stuecker, Marlin Schul, Martin Banyai, Martin Koretzky, Martin Myriam, Mary Elizabeth Gaffney, Masao Hirano, Masashi Kanemoto, Mashio Nakamura, Mersel Tahar, Messas Emmanuel, Michael Kovacs, Michael Leahy, Michael Levy, Michael Munch, Michael Olsen, Michel De Pauw, Michel Gustin, Michiel Van Betsbrugge, Mikhail Boyarkin, Miroslav Homza, Modise Koto, Mohamed Abdool-Gaffar, Mohamed Ayman Fakhry Nagib, Mohamed El- Dessoki, Mohamed Khan, Monniaty Mohamed, Moo Hyun Kim, Moon-Hee Lee, Mosaad Soliman, Mostafa Shawky Ahmed, Mostafa Soliman Abd el Bary, Moustafa A. Moustafa, Muhammad Hameed, Muhip Kanko, Mujibur Majumder, Nadezhda Zubareva, Nicola Mumoli, Nik Azim Nik Abdullah, Nisa Makruasi, Nishen Paruk, Nonglak Kanitsap, Norberto Duda, Nordiana Nordin, Ole Nyvad, Olga Barbarash, Orcun Gurbuz, Oscar Gomez Vilamajo, Oscar Nandayapa Flores, Ozcan Gur, Oztekin Oto, Pablo Javier Marchena, Pantep Angchaisuksiri, Patrick Carroll, Pavel Lang, Peter MacCallum, Peter Baron von Bilderling, Peter Blombery, Peter Verhamme, Petr Jansky, Peuch Bernadette, Philippe De Vleeschauwer, Philippe Hainaut, Piera Maria Ferrini, Piriyaporn Iamsai, Ponchaux Christian, Pongtep Viboonjuntra, Ponlapat Rojnuckarin, Prahlad Ho, Pramook Mutirangura, Rachel Wells, Rafael Martinez, Raimundo Tirado Miranda, Ralf Kroening, Rapule Ratsela, Raquel Lopez Reyes, Raul Franco Diaz de Leon, Raymond Siu Ming Wong, Raz Alikhan, Reinhold Jerwan-Keim, Remedios Otero, Renate Murena-Schmidt, Reto Canevascini, Richard Ferkl, Richard White, Rika Van Herreweghe, Rita Santoro, Robert Klamroth, Robert Mendes, Robert Prosecky, Roberto Cappelli, Rudolf Spacek, Rupesh Singh, Sam Griffin, Sang Hoon Na, Sanjeev Chunilal, Saskia Middeldorp, Satoshi Nakazawa, Sebastian Schellong, See Guan Toh, Seinturier Christophe, Selim Isbir, Selma Raymundo, Seng Kiat Ting, Serge Motte, Serir Ozkan Aktogu, Servaas Donders, Seung Ick Cha, Seung-Hyun Nam, Sevestre-Pietri Marie-Antoinette, Shaun Maasdorp, Shenghua Sun, Shenming Wang, Sherif Mohamed Essameldin, Sherif Mohamed Sholkamy, Shintaro Kuki, Shinya Goto, Shuichi Yoshida, Shunzo Matsuoka, Simon McRae, Simon Watt, Siriwimon Patanasing, Siwe-Nana Jean-Léopold, Somchai Wongkhantee, Soo-Mee Bang, Sophie Testa, Stanislav Zemek, Steffen Behrens, Stephan Dominique, Stuart Mellor, Suaran Singh Gurcharan Singh, Sudip Datta, Sunee Chayangsu, Susan Solymoss, Tamara Everington, Tarek Ahmed Adel Abdel-Azim, Tawatchai Suwanban, Taylan Adademir, Terence Hart, Terriat Béatrice, Thifhelimbilu Luvhengo, Thomas Horacek, Thomas Zeller, Tim Boussy, Tim Reynolds, Tina Biss, Ting-Hsing Chao, Tomas Smith Casabella, Tomoya Onodera, Tontanai Numbenjapon, Victor Gerdes, Vladimir Cech, Vladimir Krasavin, Vladimir Tolstikhin, W.A. Bax, Wagih Fawzy Abdel Malek, Wai Khoon Ho, Walter Ageno, Walter Pharr, Weihong Jiang, Wei-Hsiang Lin, Weihua Zhang, Wei-Kung Tseng, Wen-Ter Lai, Wilfried De Backer, Wilhelm Haverkamp, Winston Yoshida, Wolfgang Korte, Won Il Choi, Yang-Ki Kim, Yasuhiro Tanabe, Yasushi Ohnuma, Yeung-Chul Mun, Yohan Balthazar, Yong Park, Yoshisato Shibata, Yuriy Burov, Yuriy Subbotin, Zdenek Coufal, Zhenwen Yang, Zhicheng Jing, Zhicheng Jing, Zhongqi Yang.

      References

        • ISTH Steering Committe for World Thrombosis Day
        Thrombosis: a major contributor to the global disease burden.
        J. Thromb. Haemost. 2014; 12: 1580-1590
        • Raskob G.E.
        • et al.
        Thrombosis: a major contributor to global disease burden.
        Arterioscler. Thromb. Vasc. Biol. 2014; 34: 2363-2371
        • Cohen A.T.
        • et al.
        Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality.
        Thromb. Haemost. 2007; 98: 756-764
        • Kearon C.
        • et al.
        Antithrombotic therapy for VTE disease: CHEST guideline and expert panel report.
        Chest. 2016; 149: 315-352
        • Hokusai V.T.E.I.
        • et al.
        Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism.
        N. Engl. J. Med. 2013; 369: 1406-1415
        • Schulman S.
        • et al.
        Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.
        Circulation. 2014; 129: 764-772
        • Bauersachs R.
        • et al.
        Oral rivaroxaban for symptomatic venous thromboembolism.
        N. Engl. J. Med. 2010; 363: 2499-2510
        • Agnelli G.
        • et al.
        Oral apixaban for the treatment of acute venous thromboembolism.
        N. Engl. J. Med. 2013; 369: 799-808
        • Weitz J.I.
        Global Anticoagulant Registry in the Field - Venous Thromboembolism (GARFIELD-VTE). Rationale and design.
        Thromb. Haemost. 2016; 116: 1172-1179
        • Ageno W.
        Characteristics of Patients with Acute Venous Thromboembolism: Perspectives from the International, Observational, GARFIELD-VTE Registry.
        2018 (In preparation)
        • Schulman S.
        • Kearon C.
        Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients.
        J. Thromb. Haemost. 2005; 3: 692-694
        • Bates S.M.
        • et al.
        Diagnosis of DVT: antithrombotic therapy and prevention of thrombosis, 9th ed: american College of Chest Physicians Evidence-Based Clinical Practice Guidelines.
        Chest. 2012; 141: e351S-e418S
        • Ageno W.
        • et al.
        Characteristics and Management of Patients with venous thromboembolism: the GARFIELD-VTE registry.
        Thromb. Haemost. 2019; 119: 319-327
        • Cohen A.T.
        • G.A
        • Bauersachs R.
        The management of acute venous thromboembolism in clinical practice. Results from the European PREFER in VTE Registry.
        Thromb. Haemost. 2017; 11: 1326-1337
        • Registry
        R. Major bleeding and VTE recurrences.
        (Available from:)
        • Ahmed M.
        Blood thinners and gastrointestinal endoscopy.
        World J. Gastrointest. Endosc. 2016; 17: 584-590
        • Rebecca Caroline Barker B.B.F.
        Paul Marval Venous thromboembolism: risks and prevention.
        Contin. Educ. Anaesth. Crit. Care Pain. 2011; 11 (BMedSci BMBS FRCA): 18-23
        • Sardar P.
        • C.S
        • Mukherjee D.
        Efficacy and safety of new oral anticoagulants for extended treatment of venous thromboembolism: systematic review and meta-analyses of randomized controlled trials.
        Drugs. 2013; 73: 1171-1182
        • Glynn R.J.
        • Rosner B.
        Comparison of risk factors for the competing risks of coronary heart disease, stroke, and venous thromboembolism.
        Am. J. Epidemiol. 2005; 162: 975-982
        • Prandoni P.
        • et al.
        An association between atherosclerosis and venous thrombosis.
        N. Engl. J. Med. 2003; 348: 1435-1441
        • Piazza G.
        • Goldhaber Samuel Z.
        Venous thromboembolism and atherothrombosis an integrated approach.
        Circulation. 2010; 121: 2146-2150
        • Ageno W.
        • F.A
        • Haas S.
        • Weitz J.I.
        • Goldhaber S.Z.
        • Turpie A.G.G.
        • Goto S.
        • Angchaisuksiri P.
        • Dalsgaard Nielsen J.
        • Kayani G.
        • Schellong S.
        • Bounameaux H.
        • Mantovani L.G.
        • Prandoni P.
        • Kakkar A.K.
        Provoked versus unprovoked venous thromboembolism: findings from GARFIELD-VTE.
        Res. Pract. Thromb. Haemost. 2021; 20: 326-341
        • Sedgwick P.
        Prospective cohort studies: advantages and disadvantages.
        BMJ [Br. Med. J.]. 2013; 347f6726