Advertisement

Prediction and prognostic importance of in-hospital major bleeding in a real-world cohort of patients with pulmonary embolism

  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Karl-Patrik Kresoja
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Department of Internal Medicine and Cardiology, Campus Virchow Klinikum (CVK), Charité – University Medicine, Berlin, Germany

    Berlin Institute of Health, Berlin, Germany

    German Cardiovascular Research Centre (DZHK), partner site Berlin, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Matthias Ebner
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Department of Internal Medicine and Cardiology, Campus Virchow Klinikum (CVK), Charité – University Medicine, Berlin, Germany

    German Cardiovascular Research Centre (DZHK), partner site Berlin, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Nina I.J. Rogge
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Clinic of Cardiology and Pneumology, Heart Center, University Medical Center Goettingen, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Carmen Sentler
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Clinic of Cardiology and Pneumology, Heart Center, University Medical Center Goettingen, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Karsten Keller
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Center for Thrombosis and Hemostasis, University Medical Center Mainz, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Lukas Hobohm
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Center for Thrombosis and Hemostasis, University Medical Center Mainz, Germany

    Center for Cardiology, Cardiology I, University Medical Center Mainz, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Gerd Hasenfuß
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Clinic of Cardiology and Pneumology, Heart Center, University Medical Center Goettingen, Germany

    German Cardiovascular Research Centre (DZHK), partner site Goettingen, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Stavros V. Konstantinides
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Center for Thrombosis and Hemostasis, University Medical Center Mainz, Germany

    Department of Cardiology, Democritus University of Thrace, Alexandroupolis, Greece
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Burkert Pieske
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Department of Internal Medicine and Cardiology, Campus Virchow Klinikum (CVK), Charité – University Medicine, Berlin, Germany

    Berlin Institute of Health, Berlin, Germany

    German Cardiovascular Research Centre (DZHK), partner site Berlin, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Mareike Lankeit
    Correspondence
    Corresponding author at: Department of Internal Medicine and Cardiology, Campus Virchow Klinikum (CVK), Charité – University Medicine Berlin, Augustenburger Platz 1, 13353 Berlin, Germany.
    Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
    Affiliations
    Department of Internal Medicine and Cardiology, Campus Virchow Klinikum (CVK), Charité – University Medicine, Berlin, Germany

    German Cardiovascular Research Centre (DZHK), partner site Berlin, Germany

    Clinic of Cardiology and Pneumology, Heart Center, University Medical Center Goettingen, Germany

    Center for Thrombosis and Hemostasis, University Medical Center Mainz, Germany
    Search for articles by this author
  • Author Footnotes
    1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
Open AccessPublished:March 15, 2019DOI:https://doi.org/10.1016/j.ijcard.2019.03.017

      Highlights

      • The VTE-BLEED score identified patients at risk for in-hospital major bleeding.
      • However, renal function and previous surgery might deserve more attention.
      • In-hospital major bleeding was a predictor of in-hospital and 1-year mortality.

      Abstract

      Background

      Assessment of bleeding risk in patients with pulmonary embolism (PE) is challenging. Recently, the VTE-BLEED score was shown to predict major bleeding. Therefore, we aimed to investigate the VTE-BLEED score and assess the prognostic impact of major bleeding in a real-world cohort of PE patients.

      Methods

      Consecutive PE patients included in a prospective single-center cohort study between 09/2008 and 11/2016 were eligible for analysis; patients treated with thrombolysis were excluded. The VTE-BLEED was calculated post-hoc; in-hospital major bleeding was defined using the ISTH definition.

      Results

      Overall, 522 patients (median age 69, IQR 56–78 years; 53% female) were included in the present analysis; major bleeding occurred in 18 (3.5%) patients. A VTE-BLEED score ≥2 points identified patients at high-risk for major bleeding (OR 3.7, 95% CI 1.1–13.0, sensitivity 83%, specificity 42%). Additionally, a GFR <30 ml/min/1.73 m2 (OR 6.0, 95% CI 1.8–19.8) and previous surgery (OR 3.6, 95% CI 1.4–9.3) were associated with major bleeding. A less frequent use of unfractionated heparin as initial treatment was associated with a decrease of major bleeding over time. Major bleeding was identified as strong predictor of in-hospital (OR 7.7, 95% CI 2.3–25.8) and 1-year mortality (HR 3.6, 95% CI 2.0–6.6), especially in normotensive patients (OR 12.1, 95% CI 3.5–43.0 and HR 6.0, 95% CI 2.9–12.6, respectively).

      Conclusions

      In a real-world cohort, the VTE-BLEED score identified PE patients at risk for in-hospital major bleeding. However, for assessment of bleeding risk, renal function and previous surgery should be considered. Major bleeding emerged as strong predictor of in-hospital and 1-year mortality.

      Abbreviations:

      AUC (area under the curve), CI (confidence interval), HR (hazard ratio), IQR (interquartile range), OR (odds ratio), PE (pulmonary embolism), ROC (receiver operating characteristic), VTE (venous thromboembolism)

      Keywords

      1. Introduction

      Current guidelines recommend therapeutic anticoagulation of patients with pulmonary embolism (PE) for at least 3 months. Extended anticoagulation should be considered in patients with unprovoked PE based on their individual risk-to-benefit ratio including the assessment of bleeding risk [
      • Konstantinides S.V.
      • Torbicki A.
      • Agnelli G.
      • et al.
      2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism.
      ,
      • Kearon C.
      • Akl E.A.
      • Ornelas J.
      • et al.
      Antithrombotic therapy for VTE disease: CHEST Guideline and Expert Panel report.
      ]. However, most bleeding prediction scores available (such as the HAS-BLED) were developed for patients with atrial fibrillation and are of poor value for the prediction of short- [
      • Klok F.A.
      • Niemann C.
      • Dellas C.
      • Hasenfuß G.
      • Konstantinides S.
      • Lankeit M.
      Performance of five different bleeding-prediction scores in patients with acute pulmonary embolism.
      ] and long-term [
      • Riva N.
      • Bellesini M.
      • Di Minno M.N.D.
      • et al.
      Poor predictive value of contemporary bleeding risk scores during long-term treatment of venous thromboembolism. A multicentre retrospective cohort study.
      ] bleeding complications in patients with venous thromboembolism (VTE). To address this topic, the American College of Chest Physicians (ACCP) provided a list of 18 risk factors derived from various cohort studies, and patients with at least two risk factors were considered to be at high risk for bleeding [
      • Kearon C.
      • Akl E.A.
      • Ornelas J.
      • et al.
      Antithrombotic therapy for VTE disease: CHEST Guideline and Expert Panel report.
      ]. The recently proposed VTE-BLEED score, which was developed in the dabigatran arms of the pooled RE-COVER sister studies, identified six differently weighted variables (actiVe cancer [2 points], males with uncontrolled hyperTension [1 point], anaEmia [1.5 points], history of BLeeding [1.5 points], agE ≥60 years [1.5 points] and renal Dysfunction [1.5 points]) as predictors of major bleeding in patients with VTE on stable oral anticoagulation receiving either warfarin or dabigatran [
      • Klok F.A.
      • Hösel V.
      • Clemens A.
      • et al.
      Prediction of bleeding events in patients with venous thromboembolism on stable anticoagulation treatment.
      ]. Further, the VTE-BLEED score was externally validated in the HOKUSAI-VTE study in both, patients receiving warfarin or edoxaban [
      • Klok F.A.
      • Barco S.
      • Konstantinides S.V.
      External validation of the VTE-BLEED score for predicting major bleeding in stable anticoagulated patients with venous thromboembolism.
      ]. A score of two or more points was associated with an odds ratio (OR) of 5.0 (95% confidence interval [CI], 3.5–7.1) and 4.0 (95% CI, 2.5–6.4) for bleeding complications in the pooled RE-COVER studies and the HOKUSAI-VTE study, respectively. However, a validation of the VTE-BLEED score in a real-world cohort of PE patients has not been performed thus far.
      Furthermore, early (e.g. during the in-hospital stay) major bleeding constitutes a relevant complication and challenge in the management of patients with PE with a case fatality rate of up to 20% [
      • Lecumberri R.
      • Alfonso A.
      • Jiménez D.
      • et al.
      Dynamics of case-fatalilty rates of recurrent thromboembolism and major bleeding in patients treated for venous thromboembolism.
      ]. Major bleeding was identified as a predictor of short- and midterm-mortality in the Rejestr ZATorowości płucnej w POLsce (ZATPOL) and as a predictor of 1-year mortality in the Registro Informatizado Enfermedad TromboEmbolica (RIETE) [
      • Budaj-Fidecka A.
      • Kurzyna M.
      • Fijałkowska A.
      • et al.
      In-hospital major bleeding predicts mortality in patients with pulmonary embolism: an analysis of ZATPOL Registry data.
      ,
      • Prandoni P.
      • Trujillo-Santos J.
      • Sanchez-Cantalejo E.
      • et al.
      Major bleeding as a predictor of mortality in patients with venous thromboembolism: findings from the RIETE Registry.
      ]. However, the inclusion of patients receiving thrombolytic therapy in these two studies may have led to a higher incidence of both, bleeding and fatal events.
      Therefore, the aim of our study was to validate the prognostic performance of the VTE-BLEED score in a real-world cohort of patients with PE, to identify predictors of in-hospital major bleeding and to investigate the impact of major bleeding on in-hospital and 1-year mortality.

      2. Material and methods

      2.1 Study design and patient cohort

      Consecutive patients ≥18 years with confirmed acute PE were included in a single-center prospective cohort study (Pulmonary Embolism Registry of Goettingen [PERGO]) at the University Medical Centre Goettingen, Germany. The study protocol has been described in detail before [
      • Lankeit M.
      • Jiménez D.
      • Kostrubiec M.
      • et al.
      Predictive value of the high-sensitivity troponin T assay and the simplified Pulmonary Embolism Severity Index in hemodynamically stable patients with acute pulmonary embolism, a prospective validation study.
      ,
      • Hobohm L.
      • Hellenkamp K.
      • Hasenfuß G.
      • Münzel T.
      • Konstantinides S.
      • Lankeit M.
      Comparison of risk assessment strategies for not-high-risk pulmonary embolism.
      ]. Complete baseline data on clinical, electrocardiographic, echocardiographic, radiological and laboratory parameters were obtained using a standardized case report form.
      As shown in Fig. A1 of the Supplementary material, patients were excluded for the following reasons: a) incomplete baseline data (n = 3), b) treatment with thrombolysis or inclusion in the PEITHO study (n = 93), c) inclusion in the AMPLIFY study (n = 1) d) inclusion in PERGO more than once (only the first PE event was considered for analysis; n = 7), e) treatment with surgical embolectomy or interventional approaches (n = 0), f) subsegmental PE and other acute cardiac, respiratory or inflammatory disease responsible for symptoms and hemodynamic status on admission (n = 16).
      The VTE-BLEED and the HAS-BLED [
      • Lip G.Y.H.
      • Frison L.
      • Halperin J.L.
      • Lane D.A.
      Comparative validation of a novel risk score for predicting bleeding risk in anticoagulated patients with atrial fibrillation: the HAS-BLED (Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile INR, Elderly, Drugs/Alcohol Concomitantly) score.
      ] score were calculated post-hoc for all patients, missing variables were considered to be normal.
      All patients were followed for the in-hospital stay and 1-year survival status was assessed by consulting local registration offices. The primary study outcome was in-hospital major bleeding, the secondary study outcomes were in-hospital and 1-year mortality. Major bleeding was defined as fatal and/or symptomatic bleeding in a critical area or organ and/or bleeding causing a fall in hemoglobin level of ≥2 g/dl or transfusion of ≥2 units of erythrocyte concentrates according the definition of the International Society of Thrombosis and Haemostasis (ISTH) [
      • Schulman S.
      • Kearon C.
      Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients.
      ]. All events and causes of death were independently adjudicated by two of the authors (K.-P.K. and K.K.) blinded for the results of the bleeding prediction scores using medical records and autopsy records if available or by contacting the last treating physician; disagreement was resolved by a third author (M.L.).
      The diagnostic and therapeutic management was in accordance with (at this time) current guidelines. All related decisions were left to the discretion of the treating physicians and were not influenced by the study protocol. Study results were not communicated to the clinicians and thus not used to guide patient management, or to monitor effects of treatment during the hospital stay or at any time during the 1-year follow-up period. The study protocol was conducted in accordance with the amended Declaration of Helsinki and was approved by the local independent Ethic Committee of the Medical University Goettingen, Germany; all patients gave written informed consent for participation in the study.

      2.2 Statistical analyses

      Categorical variables are expressed as absolute number or percentage and were compared using Fisher's exact test or Chi-squared test, as appropriate. Continuous variables did not follow a normal distribution when tested with the modified Kolmogorov-Smirnov test (Lilliefors test); therefore, these variables are expressed as median with the corresponding interquartile range (IQR) and were compared using the unpaired Mann-Whitney U test. Receiver operating characteristic (ROC) curve analyses were performed to determine the area under the curve (AUC) of the VTE-BLEED and HAS-BLED score with regard to major bleeding. Further, the Hosmer-Lemeshow test was used to evaluate the difference between the predicted and observed risk estimates of the VTE-BLEED and HAS-BLED score. The prognostic relevance of dichotomous/dichotomized variables with regard to the primary and secondary study outcomes was assessed using univariable logistic regression analyses; results are presented as OR with corresponding 95% CIs. Additionally, parameters univariably associated with in-hospital mortality were separately tested in combination with major bleeding in multivariable logistic regression models. Kaplan-Meier analyses were used to compare the probability of long-term survival in different subgroups; the log-rank test was used to test for differences. Cox regression analyses were performed to test the prognostic relevance of dichotomous/dichotomized variables with regard to 1-year mortality; results are presented as hazard ratios (HR) with corresponding 95% CI. To identify independent predictors of 1-year mortality, significant predictors identified in univariable Cox regression analyses were included in a multivariable model using simultaneous selection. Scores containing variables already included in the multivariable model were excluded. For analysis of temporal trends, the study period was divided in four two-year segments (09/2008 to 08/2010, 09/2010 to 08/2012, 09/2012 to 08/2014 and 09/2014 to 11/2016). Testing for temporal trends was conducted using the Cochran-Armitage trend test for dichotomous/dichotomized variables.
      A two-sided significance level of α 0.05 was defined appropriate to indicate statistical significance. Statistical analyses were performed using the SPSS software (IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0, Armonk, NY: IBM Corp.).

      3. Results

      Of 642 patients included in PERGO between September 2008 and November 2016, 120 (18.7%) were excluded as described above and shown in Fig. A1 of the Supplementary material. Thus, 522 patients (median age, 69; IQR, 56–78 years; 53% female) were included in the present analysis. The baseline characteristics, comorbidities and initial findings of the study patients are presented in Table 1, left column.
      Table 1Baseline characteristics, comorbidities and initial findings of 522 study patients.
      All study patients (n = 522)Patients without major bleeding (n = 504)Patients with major bleeding (n = 18)p-Value
      Age (years)69 (56–78)69 (56–78)71 (60–78)0.402
       Age ≥ 60 years361 (69)346 (69)15 (83)0.298
      Male sex246/522 (47)240/504 (48)6 (33)0.337
      Risk factors for VTE
       Previous PE62/521 (12)61/503 (12)1 (6)0.710
       Recent trauma
      Within the past 4 weeks.
      16 (3.1)16 (3.2)0 (0.0)0.566
       Recent surgery
      Within the past 4 weeks.
      100 (19)92 (18)8 (44)0.011
       Pregnancy/peripartum period3 (0.6)3 (0.6)0 (0.0)0.900
       Contraceptive or sex hormone (replacement) therapy39 (7.5)39 (7.7)0 (0.0)0.385
       Active cancer
      Active or anti-tumor therapy within the last 6 months, or metastatic state.
      90 (17)83 (17)7 (39)0.022
       Unprovoked PE
      In the absence of a temporary or reversible risk factor such as surgery, trauma, immobilization, pregnancy/postpartum period, contraceptives or sex hormone replacement therapy according to the definition of the 2014 ESC guidelines [1].
      324 (62)315 (63)9 (50)0.283
      Comorbidities
       Arterial hypertension334 (64)321 (64)13 (72)0.619
       Male with uncontrolled arterial hypertension
      Male with initial systolic blood pressure ≥ 140 mm Hg.
      84 (16)82 (16)2 (11)0.751
       Chronic pulmonary disease
      Bronchial asthma, chronic obstructive lung disease or interstitial lung diseases.
      79/521 (15)74/503 (15)5 (28)0.059
       Chronic heart failure85 (16)82 (16)3 (17)1.000
       Coronary artery disease94 (18)88 (18)6 (33)0.085
       Previous stroke49 (10)48 (10)1 (5.6)0.482
       Diabetes mellitus87 (17)86 (17)1 (5.6)0.333
       Anemia
      Hemoglobin <13 g/dl in male and <12 g/dl in female patients.
      199 (38)188 (37)11 (61)0.038
       Known liver disease
      Defined as Child-Pugh B or C.
      15/411 (3.7)15/393 (3.8)3 (17)0.567
       Alcohol abuse15/362 (4.1)14/351 (4.0)1/11 (9.1)0.376
       Previous bleeding17/364 (4.7)16/352 (4.6)1/12 (8.3)0.442
      Laboratory values
       GFR ml/min/1.73 m272 (55–90)72 (56–91)48 (29–78)0.005
        GFR <60 ml/min/1.73 m2160/502 (32)151/486 (31)9/16 (56)0.051
        GFR <30 ml/min/1.73 m230/502 (6.0)26/486 (5.3)4/16 (25)0.011
       hsTnT ≥14 ng/l268/435 (62)256/421 (61)12/14 (86)0.233
       NT-proBNP ≥600 ng/l188/387 (49)178/375 (48)10/12 (83)0.018
      Risk stratification
       Tachycardia (HR ≥ 100 bpm)101/507 (20)97/490 (20)4/17 (24)0.757
       Systolic blood pressure <90 mm Hg18/498 (3.6)15/481 (3.1)3/17 (17.6)0.020
       Hypoxia
      Defined as oxygen saturation of <90% with or without oxygen administration.
      104/438 (24)98/424 (23)6/14 (43)0.108
       sPESI ≥1 point(s)350 (70)333 (66)17 (94)0.010
       ESC risk classes0.009
        Low risk85 (16)85 (17)0 (0.0)
        Intermediate-low risk290 (56)280 (56)10 (56)
        Intermediate-high risk123 (24)118 (23)5 (28)
        High risk23 (4.4)20 (4.0)3 (17)
      Initial treatment
       Unfractionated heparin bolus295 (57)282 (56)13 (72)0.228
       Continuous unfractionated heparin300 (57)286 (57)14 (78)0.092
       Low molecular weight heparin or fondaparinux200 (38)197 (39)3 (17)0.081
       Rivaroxaban7 (1.3)7 (1.4)0 (0.0)1.000
       Apixaban9 (1.7)9 (1.7)0 (0.0)1.000
       Other6 (1.2)5 (1.0)1 (5.6)0.789
      Abbreviations: VTE denotes venous thromboembolism; PE, pulmonary embolism; GFR, glomerular filtration rate; hsTnT, high sensitivity troponin T; NT-proBNP, N-terminal prohormone of brain natriuretic peptide; HR, heart rate; bpm, beats per minute; sPESI, simplified pulmonary embolism severity index; ESC, European Society of Cardiology.
      Bold p-values indicate significant findings (p < 0.05)
      a Within the past 4 weeks.
      b Active or anti-tumor therapy within the last 6 months, or metastatic state.
      c In the absence of a temporary or reversible risk factor such as surgery, trauma, immobilization, pregnancy/postpartum period, contraceptives or sex hormone replacement therapy according to the definition of the 2014 ESC guidelines [
      • Konstantinides S.V.
      • Torbicki A.
      • Agnelli G.
      • et al.
      2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism.
      ].
      d Male with initial systolic blood pressure ≥ 140 mm Hg.
      e Bronchial asthma, chronic obstructive lung disease or interstitial lung diseases.
      f Hemoglobin <13 g/dl in male and <12 g/dl in female patients.
      g Defined as Child-Pugh B or C.
      h Defined as oxygen saturation of <90% with or without oxygen administration.
      During the in-hospital stay, 18 patients (3.5%) had major bleeding with a median time to event of 3 (95% CI, 2–15) days; of those, two were intracranial hemorrhages. No patient died due to bleeding. Patients with major bleeding had a longer median in-hospital stay (16; 95% CI, 10–28 vs. 9; 95% CI, 5–13 days p = 0.002) and the prevalence of anemia, active cancer, recent surgery and severe renal impairment was higher compared to patients without major bleeding (Table 1, right columns).
      At the time of the major bleeding event, 12 patients were treated with continuous unfractionated heparin (UFH; 67%), three were bridged from low molecular weight heparin (LMWH) to a vitamin-K antagonist (VKA; 17%) and one patient each was treated with LMWH (5.6%), VKA (5.6%) and argatroban (5.6%). Fig. A2 of the Supplementary material shows the initial anticoagulation and the incidence of major bleeding for each two-year segment of the study period. While the number of patients treated with continuous UFH decreased over time (77%, 73%, 50% and 39%; p < 0.001), the number of patients treated with subcutaneous LMWH/fondaparinux increased (20%, 27%, 48% and 51%; p < 0.001). In parallel with the less frequent use of continuous UFH, the incidence of UFH-associated major bleedings tended to decrease (2.7%, 3.7%, 3.3% and 1.4%; p = 0.457) while the incidence of LMWH/fondaparinux-associated major bleedings remained stable at a low rate (0.9%, 0.9%, 0.0% and 0.7%; p = 0.624).

      3.1 Validation of the VTE-BLEED score

      As shown in Fig. A3 of the Supplementary material, ROC analyses yielded an AUC of 0.69 (95% CI, 0.58–0.80) for the VTE-BLEED and of 0.58 (95% CI, 0.48–0.69) for the HAS-BLED score for prediction of major bleeding. Patients classified at high-risk (≥2 points) by the VTE-BLEED score (n = 305; 58%) had a major bleeding rate of 4.9% (n = 16 events) compared to 1.4% (n = 2 events) in patients classified at low-risk (n = 217; 42%). This was associated with a sensitivity of 83% (95% CI, 61–94%) and a specificity of 42% (95% CI, 38–46%). Goodness of fit was satisfactory for the VTE-BLEED score (X2 3.16, p = 0.789) while the HAS-BLED score overestimated major bleeding risk in higher risk classes (Fig. A4 of the Supplementary material). A VTE-BLEED score of ≥2 points was associated with a 3.7-fold increased OR for major bleeding during the in-hospital stay (Table 2). In comparison, the HAS-BLED score failed to identify patients at higher risk for major bleeding: 10 (3.7%) of 217 patients (42%) classified as high-risk had major bleeding compared to eight (3.2%) of 305 patients (58%) classified as low-risk. Concordantly, a HAS-BLED score >2 points was not able to predict major bleeding (OR, 1.1; 95% CI, 0.4–2.9).
      Table 2Univariable predictors of in-hospital major bleeding.
      OR (95% CI)p-Value
      Scores
      VTE-BLEED ≥2 points3.7 (1.1–13.0)0.040
      sPESI ≥1 point(s)6.8 (1.3–36.2)0.025
      Variables
      GFR <60 ml/min/1.73 m22.9 (1.1–7.9)0.038
      GFR <30 ml/min/1.73 m26.0 (1.8–19.8)0.004
      GFR −10 ml/min/1.73 m21.4 (1.1–1.7)0.003
      Previous surgery
      Within the past 4 weeks.
      3.6 (1.4–9.3)0.009
      Active cancer
      Active or anti-tumor therapy within the last 6 months, or metastatic state.
      3.2 (1.2–8.6)0.019
      Anemia
      Hemoglobin <13 g/dl in male and <12 g/dl in female patients.
      2.6 (1.0–6.9)0.048
      Abbreviations: OR denotes odds ratio; CI, confidence interval; sPESI, simplified pulmonary embolism severity index, GFR, glomerular filtration rate.
      Bold p-values indicate significant findings (p < 0.05)
      a Within the past 4 weeks.
      b Active or anti-tumor therapy within the last 6 months, or metastatic state.
      c Hemoglobin <13 g/dl in male and <12 g/dl in female patients.
      Of the variables used for calculation of the VTE-BLEED score, only anemia, active cancer and a moderately impaired renal function (GFR < 60 ml/min/1.73 m2) were identified as univariable predictors of major bleeding (Table 2). Besides those, previous surgery and a sPESI ≥1 point(s) were able to predict major bleeding and a severely impaired renal function (GFR < 30 ml/min/1.73 m2) emerged as the strongest predictor. The impact of initial anticoagulation on major bleeding events is shown and discussed in the Supplementary material.

      3.2 Influence of major bleeding on short- and long-term mortality

      Overall, 21 patients (4.0%) died during the in-hospital stay. Of those, 10 deaths were due to PE (48%), eight to cancer (38%), two to sepsis (9.5%) and one to recurrent PE (4.8%). Patients with major bleeding had a higher in-hospital mortality rate compared to patients without major bleeding (22.2% vs. 3.6%, p = 0.005) and major bleeding was associated with an elevated risk of in-hospital death (OR, 7.7; 95% CI, 2.3–25.8; p = 0.001), especially in normotensive patients (OR, 12.1; 95% CI, 3.5–43.0; p < 0.001). As shown in Table 3a , left column, active cancer was one of the strongest single predictors of in-hospital mortality. Accordingly, risk prediction scores including active cancer as a variable such as the VTE-BLEED score and the sPESI were able to predict in-hospital death, while scores not accounting for active cancer such as the HAS-BLED score (OR, 1.7; 95% CI, 0.7–4.1; p = 0.212) failed. Predictors of in-hospital mortality in 499 normotensive patients are shown in Table 3a, right column. Of note, the prognostic value of major bleeding remained independent, both, in all study and in normotensive patients, if tested with each variable associated with in-hospital mortality separately in multivariable logistic models (data not shown).
      Table 3a and b: Predictors of in-hospital and one-year mortality in all study and normotensive patients.
      In-hospital mortalityAll study patients (n = 522)Normotensive patients (n = 498)
      VariableOR (95% CI)p-ValueOR (95% CI)p-Value
      Major bleeding7.7 (2.3–25.8)0.00112.1 (3.5–43.0)<0.001
      VTE-BLEED ≥2 points7.6 (1.8–32.9)0.0076.2 (1.4–27.3)0.016
      Active cancer
      Active or anti-tumor therapy within the last 6 months, or metastatic state.
      6.5 (2.7–15.6)<0.0016.5 (2.5–17.1)<0.001
      ESC per increase of risk class2.6 (1.5–4.7)<0.0012.7 (1.2–6.0)0.015
      sPESI ≥1 point(s)10.9 (1.5–81.8)0.0209.4 (1.2–71.3)0.030
      Syncope3.3 (1.3–8.5)0.0131.7 (0.4–6.0)0.439
      NT-proBNP ≥600 ng/l4.0 (1.1–14.8)0.0346.3 (1.4–28.4)0.018
      One-year mortalityUnivariable Cox regression analysisMultivariable Cox regression analysis
      VariableHR (95% CI)p-ValueHR (95% CI)p-Value
      In-hospital major bleeding3.6 (2.0–6.6)<0.0013.2 (1.5–6.6)<0.001
      VTE-BLEED ≥2 points3.6 (2.3–5.5)<0.001
      HAS-BLED >2 points1.5 (1.1–2.1)0.013
      Active cancer
      Active or anti-tumor therapy within the last 6 months, or metastatic state.
      5.5 (3.9–7.7)<0.0019.1 (5.4–15.4)<0.001
      Age ≥60 years2.1 (1.1–3.8)0.0161.4 (0.8–2.6)0.277
      Chronic heart failure2.1 (1.5–3.1)<0.0011.5 (0.8–2.6)0.196
      Coronary artery disease1.6 (1.1–2.4)0.0111.5 (0.8–2.6)0.195
      Anemia
      Hemoglobin <13 g/dl in male and < 12 g/dl in female patients.
      2.1 (1.5–2.9)<0.0011.3 (0.8–2.1)0.195
      ESC per increase of risk class1.7 (1.4–2.1)<0.0011.9 (1.3–2.8)0.001
      sPESI ≥1 point(s)9.4 (4.6–19.1)<0.001
      Abbreviations: OR denotes odds ratio; CI, confidence interval; HR, hazard ratio; ESC, European Society of Cardiology; sPESI, simplified pulmonary embolism severity index; NT-proBNP, N-terminal prohormone of brain natriuretic peptide.
      Bold p-values indicate significant findings (p < 0.05)
      a Active or anti-tumor therapy within the last 6 months, or metastatic state.
      b Hemoglobin <13 g/dl in male and < 12 g/dl in female patients.
      One-year survival status was available for 502 patients (96%; median time of follow-up, 365; IQR, 323–365 days) and 73 patients (15%) died. Cancer was the leading cause of death (n = 35, 48%; cancer types are provided in Table A2 of the Supplementary material); further, 11 deaths were due to the initial PE (15%), nine to infectious disease (12%), seven to cardiac disease (9.6%), two each to recurrent PE (2.7%), pulmonary disease (2.7%), kidney disease (2.7%) and severe dementia (2.7%) and one to gastrointestinal bleeding (1.4%). Only two causes of death (2.7%) remained unknown. Of 18 patients with in-hospital major bleeding, as many as nine patients died during the observation period (1-year mortality rate, 50%); of those, five patients (56%) died due to cancer. As shown in Fig. 1, major bleeding was associated with a decreased probability of 1-year survival both, in patients with and without cancer. Of note, this was more pronounced in cancer patients with a 1-year case mortality rate of 100% (HR for major bleeding in cancer patients, 5.7; 95% CI, 2.5–13.0; p < 0.001). Besides active cancer and other variables shown in Table 3b, left column, major bleeding was associated with an increased risk of 1-year mortality (HR, 3.6; 95% CI, 2.0–6.6; p < 0.001); especially in normotensive patients (HR, 6.0; 95% CI, 2.9–12.6; p < 0.001). In multivariable analysis, only active cancer, increase of ESC risk classes and in-hospital major bleeding were identified as independent predictors of 1-year mortality (Table 3b, right column).
      Fig. 1
      Fig. 1Probability of 1-year survival in patients with/without active cancer and with/without in-hospital major bleeding.

      4. Discussion

      4.1 Prognostic value of the VTE-BLEED score

      The occurrence of major bleeding is the most relevant and frequent complication of therapeutic anticoagulation and associated with substantial morbidity and mortality [
      • Lecumberri R.
      • Alfonso A.
      • Jiménez D.
      • et al.
      Dynamics of case-fatalilty rates of recurrent thromboembolism and major bleeding in patients treated for venous thromboembolism.
      ,
      • Budaj-Fidecka A.
      • Kurzyna M.
      • Fijałkowska A.
      • et al.
      In-hospital major bleeding predicts mortality in patients with pulmonary embolism: an analysis of ZATPOL Registry data.
      ,
      • Prandoni P.
      • Trujillo-Santos J.
      • Sanchez-Cantalejo E.
      • et al.
      Major bleeding as a predictor of mortality in patients with venous thromboembolism: findings from the RIETE Registry.
      ,
      • Carrier M.
      • Le Gal G.
      • Wells P.S.
      • Rodger M.A.
      Systematic review, Case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism.
      ]. Since most bleeding events occur during the first weeks after treatment initiation [
      • Klok F.A.
      • Kooiman J.
      • Huisman M.V.
      • Konstantinides S.
      • Lankeit M.
      Predicting anticoagulant-related bleeding in patients with venous thromboembolism, a clinically oriented review.
      ], a score that identifies patients at high risk for in-hospital major bleeding may help to raise clinical awareness and to improve risk-adapted management such as anticoagulation quality control, blood pressure management and critical review of the concomitant use of platelet inhibitors and/or nonsteroidal anti-inflammatory drugs.
      In the present real-world cohort of patients with acute PE, we were able to demonstrate the prognostic value of the VTE-BLEED score to predict in-hospital major bleeding. A VTE-BLEED score ≥2 points was associated with a 3.7-fold increased risk for major bleeding and goodness of fit between observed and predicted rates of major bleeding was good (Fig. A4 of the Supplementary material). C-statistics and ORs of the VTE-BLEED score were comparable to those of the derivation and validation study [
      • Klok F.A.
      • Hösel V.
      • Clemens A.
      • et al.
      Prediction of bleeding events in patients with venous thromboembolism on stable anticoagulation treatment.
      ,
      • Klok F.A.
      • Barco S.
      • Konstantinides S.V.
      External validation of the VTE-BLEED score for predicting major bleeding in stable anticoagulated patients with venous thromboembolism.
      ]. Of note, compared to those, the present study included older patients with more comorbidities. Thus, as many 58% of patients were classified as high risk of bleeding based on the VTE-BLEED score compared to only 26% and 27% in the VTE-BLEED derivation [
      • Klok F.A.
      • Hösel V.
      • Clemens A.
      • et al.
      Prediction of bleeding events in patients with venous thromboembolism on stable anticoagulation treatment.
      ] and the validation [
      • Klok F.A.
      • Barco S.
      • Konstantinides S.V.
      External validation of the VTE-BLEED score for predicting major bleeding in stable anticoagulated patients with venous thromboembolism.
      ] study, respectively. In accordance, the prevalence of major bleeding of 3.5% in our cohort was higher compared to only 1.4% and 1.6% in the VTE-BLEED derivation [
      • Klok F.A.
      • Hösel V.
      • Clemens A.
      • et al.
      Prediction of bleeding events in patients with venous thromboembolism on stable anticoagulation treatment.
      ] and the validation [
      • Klok F.A.
      • Barco S.
      • Konstantinides S.V.
      External validation of the VTE-BLEED score for predicting major bleeding in stable anticoagulated patients with venous thromboembolism.
      ] study, respectively, but was comparable to those reported in other prospective registries (3.6% and 3.0%, respectively) [
      • Riva N.
      • Bellesini M.
      • Di Minno M.N.D.
      • et al.
      Poor predictive value of contemporary bleeding risk scores during long-term treatment of venous thromboembolism. A multicentre retrospective cohort study.
      ,
      • Budaj-Fidecka A.
      • Kurzyna M.
      • Fijałkowska A.
      • et al.
      In-hospital major bleeding predicts mortality in patients with pulmonary embolism: an analysis of ZATPOL Registry data.
      ,
      • Prandoni P.
      • Trujillo-Santos J.
      • Sanchez-Cantalejo E.
      • et al.
      Major bleeding as a predictor of mortality in patients with venous thromboembolism: findings from the RIETE Registry.
      ].

      4.2 Predictors of in-hospital major bleeding

      Both, in the VTE-BLEED derivation [
      • Klok F.A.
      • Hösel V.
      • Clemens A.
      • et al.
      Prediction of bleeding events in patients with venous thromboembolism on stable anticoagulation treatment.
      ] and the validation [
      • Klok F.A.
      • Barco S.
      • Konstantinides S.V.
      External validation of the VTE-BLEED score for predicting major bleeding in stable anticoagulated patients with venous thromboembolism.
      ] study, patients with cancer were underrepresented (4.4% and 2.6%, respectively). In contrast, in the present real-world cohort, as many as 17.2% of patients were diagnosed with active cancer. As expected, active cancer was associated with an increased risk for major bleeding. Accordingly, risk prediction scores including active cancer as a variable such as the VTE-BLEED score but also the sPESI were able to predict major bleeding (Table 2), while the HAS-BLED score (not accounting for active cancer) failed, as previously also shown by others [
      • Klok F.A.
      • Niemann C.
      • Dellas C.
      • Hasenfuß G.
      • Konstantinides S.
      • Lankeit M.
      Performance of five different bleeding-prediction scores in patients with acute pulmonary embolism.
      ,
      • Riva N.
      • Bellesini M.
      • Di Minno M.N.D.
      • et al.
      Poor predictive value of contemporary bleeding risk scores during long-term treatment of venous thromboembolism. A multicentre retrospective cohort study.
      ].
      The OR for major bleeding was twice as high for patients with a severely impaired renal function as compared to patients with a at least moderately impaired renal function (Table 2). As dabigatran is contraindicated in patients with severely impaired renal function (GFR < 30 ml/min/1.73 m2), those patients were excluded from the RECOVER studies [
      • Schulman S.
      • Kakkar A.K.
      • Goldhaber S.Z.
      • et al.
      Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.
      ] and therefore not accounted for in the derivation of the VTE-BLEED score. Furthermore, patients with a high bleeding risk were excluded from the RECOVER studies; thus, patients who underwent previous surgery most likely were underrepresented. In contrast, in the present real-world cohort, previous surgery was associated with major bleeding, as also reported by others [
      • Kearon C.
      • Akl E.A.
      • Ornelas J.
      • et al.
      Antithrombotic therapy for VTE disease: CHEST Guideline and Expert Panel report.
      ,
      • Hull R.D.
      • Raskob G.E.
      • Rosenbloom D.
      • et al.
      Heparin for 5 days as compared with 10 days in the initial treatment of proximal venous thrombosis.
      ,
      • Palareti G.
      • Leali N.
      • Coccheri S.
      • et al.
      Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT).
      ]. Therefore, if using the VTE-BLEED score for the assessment of bleeding risk, the prognostic impact of severe renal impairment and previous surgery might be underestimated and deserves further attention in clinical routine.

      4.3 Prognostic impact of in-hospital major bleeding on mortality

      Major bleeding was identified as strong predictor of in-hospital and 1-year mortality, especially in normotensive patients (Table 3), and remained an independent predictor in multivariable analyses. These findings are in accordance with results from ZATPOL [
      • Budaj-Fidecka A.
      • Kurzyna M.
      • Fijałkowska A.
      • et al.
      In-hospital major bleeding predicts mortality in patients with pulmonary embolism: an analysis of ZATPOL Registry data.
      ] and RIETE [
      • Prandoni P.
      • Trujillo-Santos J.
      • Sanchez-Cantalejo E.
      • et al.
      Major bleeding as a predictor of mortality in patients with venous thromboembolism: findings from the RIETE Registry.
      ]: in ZATPOL, major bleeding (defined according to the definition of the ISTH) was an independent predictor of 30-day (OR, 3.5; 95% CI, 1.5–8.0) and 90-day (OR, 2.8; 95% CI, 1.3–5.9) mortality in 1112 unselected PE patients [
      • Budaj-Fidecka A.
      • Kurzyna M.
      • Fijałkowska A.
      • et al.
      In-hospital major bleeding predicts mortality in patients with pulmonary embolism: an analysis of ZATPOL Registry data.
      ]. In 29,903 patients with VTE included in RIETE (of those, 14,590 patients [48.8%] with PE), major bleeding (defined as overt bleeding requiring transfusion of at least two units of blood cells, retroperitoneal, spinal, intracranial or fatal bleeding) was independently associated with death during (at least) 90-day follow-up (HR, 5.7; 95% CI, 5.0–6.3) [
      • Prandoni P.
      • Trujillo-Santos J.
      • Sanchez-Cantalejo E.
      • et al.
      Major bleeding as a predictor of mortality in patients with venous thromboembolism: findings from the RIETE Registry.
      ]. However, patients receiving thrombolytic therapy were not excluded in these two studies which may have led to a higher incidence of both, bleeding and fatal events.
      Interestingly, in the present real-world cohort, the case-fatality rate of bleeding was remarkable low: No patient died of bleeding during the in-hospital stay and only one patient (1.4%) during 1-year follow-up. In contrast, in 41,826 VTE patients included in RIETE, the case-fatality rate of bleeding was 19.7% (95% CI, 17.4–22.1%) during a mean duration of 7.8 months of anticoagulation (27,110 patient-years) [
      • Lecumberri R.
      • Alfonso A.
      • Jiménez D.
      • et al.
      Dynamics of case-fatalilty rates of recurrent thromboembolism and major bleeding in patients treated for venous thromboembolism.
      ]. In a meta-analysis summarizing data from 13 prospective cohort studies and 56 randomized controlled trials published until 2008 investigating 19,027 patients with VTE, the case-fatality rate of bleeding was 11.3% (95% CI, 7.5–15.9%) in the first three months of anticoagulation [
      • Lecumberri R.
      • Alfonso A.
      • Jiménez D.
      • et al.
      Dynamics of case-fatalilty rates of recurrent thromboembolism and major bleeding in patients treated for venous thromboembolism.
      ,
      • Budaj-Fidecka A.
      • Kurzyna M.
      • Fijałkowska A.
      • et al.
      In-hospital major bleeding predicts mortality in patients with pulmonary embolism: an analysis of ZATPOL Registry data.
      ,
      • Prandoni P.
      • Trujillo-Santos J.
      • Sanchez-Cantalejo E.
      • et al.
      Major bleeding as a predictor of mortality in patients with venous thromboembolism: findings from the RIETE Registry.
      ,
      • Carrier M.
      • Le Gal G.
      • Wells P.S.
      • Rodger M.A.
      Systematic review, Case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism.
      ]. Thus, although in the present study major bleeding appears to reflect the importance of underlying diseases, the prognostic value of major bleeding to predict in-hospital and 1-year mortality remained independent from comorbidities in multivariable analyses. Therefore, the present study findings do not only emphasize the prognostic importance of major bleeding but also the clinical need for bleeding risk assessment tools such as the VTE-BLEED score to predict in-hospital major bleeding and allow for an improved clinical awareness and risk-adapted anticoagulation management.
      A particularly relevant prognostic impact of major bleeding was observed in patients with cancer: no cancer patient with a major bleeding event during the in-hospital stay survived the first year after acute PE (Fig. 1). This finding is especially alarming since cancer increases both, the risk of major bleeding and of recurrent PE [
      • Prandoni P.
      • Lensing A.W.A.
      • Piccioli A.
      • et al.
      Recurrent venous thromboembolism and bleeding complications during anticoagulant treatment in patients with cancer and venous thrombosis.
      ,
      • Ay C.
      • Pabinger I.
      • Cohen A.T.
      Cancer-associated venous thromboembolism, burden, mechanisms, and management.
      ,
      • Farge D.
      • Trujillo-Santos J.
      • Debourdeau P.
      • et al.
      Fatal events in cancer patients receiving anticoagulant therapy for venous thromboembolism.
      ]. In the CATCH trial, anticoagulation with tinzaparin for 6 months compared to tinzaparin for 5 to 10 days followed by warfarin for 6 months was associated with a significant reduction of clinically relevant non-major bleedings (HR, 0.58; 95% CI, 0.40–0.84) in 900 VTE patients with cancer [
      • Lee A.Y.Y.
      • Kamphuisen P.W.
      • Meyer G.
      • et al.
      Tinzaparin vs warfarin for treatment of acute venous thromboembolism in patients with active cancer, a randomized clinical trial.
      ]. The HOKUSAI-VTE cancer trial demonstrated that edoxaban was noninferior to dalteparin with respect to the primary composite outcome of recurrent VTE or major bleeding (HR, 0.97; 95% CI, 0.70–1.36) although the rate of major bleeding was higher with edoxaban than with dalteparin (6.9% vs. 4.0%) [
      • Raskob G.E.
      • van Es N.
      • Verhamme P.
      • et al.
      Edoxaban for the treatment of cancer-associated venous thromboembolism.
      ]. Thus, the present study data support the current notion that all efforts should be made to prevent major bleeding events in PE patients with cancer. While anticoagulation with LMWH should be preferred over VKA treatment [
      • Konstantinides S.V.
      • Torbicki A.
      • Agnelli G.
      • et al.
      2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism.
      ,
      • Kearon C.
      • Akl E.A.
      • Ornelas J.
      • et al.
      Antithrombotic therapy for VTE disease: CHEST Guideline and Expert Panel report.
      ], further evidence is needed to evaluate which subgroup of PE patients with cancer has a favorable risk-to-benefit ratio for the treatment with NOACs (especially with regard to gastrointestinal bleedings).

      4.4 Limitations

      Some potential limitations deserve consideration: First, patients were included at a single tertiary referral university center. Thus, the present study findings might be affected by local standards and resources and not generalizable to other settings. Second, therapeutic decision making was left to the discretion of the treating physicians. Thus, the type of initial anticoagulation and measures for anticoagulation quality control were not standardized. Third, although the prevalence of major bleeding in the present study was comparable to other prospective registries [
      • Riva N.
      • Bellesini M.
      • Di Minno M.N.D.
      • et al.
      Poor predictive value of contemporary bleeding risk scores during long-term treatment of venous thromboembolism. A multicentre retrospective cohort study.
      ,
      • Budaj-Fidecka A.
      • Kurzyna M.
      • Fijałkowska A.
      • et al.
      In-hospital major bleeding predicts mortality in patients with pulmonary embolism: an analysis of ZATPOL Registry data.
      ,
      • Prandoni P.
      • Trujillo-Santos J.
      • Sanchez-Cantalejo E.
      • et al.
      Major bleeding as a predictor of mortality in patients with venous thromboembolism: findings from the RIETE Registry.
      ], the low number of major bleeding events precluded the performance of multivariable analyses and resulted in wide CIs of corresponding statistical findings.

      5. Conclusion

      In conclusion, in the present real-world cohort of patients with acute PE, the VTE-BLEED score was able to identify patients at risk for in-hospital major bleeding. However, in the assessment of bleeding risk in clinical routine, renal function and previous surgery deserve further consideration. Furthermore, in-hospital major bleeding was associated with an increased risk of in-hospital and 1-year mortality, especially in normotensive patients, emphasizing the need for reliable bleeding risk assessment tools such as the VTE-BLEED score to predict in-hospital major bleeding and allow for an improved clinical awareness and risk-adapted anticoagulation management.
      Fig. A1
      Fig. A1Flow chart of study patients. Abbreviations: PE denotes pulmonary embolism; PEITHO, Pulmonary Embolism Thrombolysis trial; AMPLIFY, Apixaban for the Initial Management of Pulmonary Embolism and Deep-Vein Thrombosis as First-Line Therapy; PERGO, Pulmonary Embolism Registry Goettingen.
      Fig. A2
      Fig. A2Trends in initial anticoagulation and incidence of in-hospital major bleeding. Abbreviations: UFH denotes unfractionated heparin; LMWH, low molecular weight heparin.
      Fig. A3
      Fig. A3Prognostic performance of the VTE-BLEED and HAS-BLED score for prediction of in-hospital major bleeding.
      Fig. A4
      Fig. A4Observed and predicted rate of in-hospital major bleeding according the (A) VTE-BLEED and (B) HAS-BLED score.

      Conflicts of interest

      None of the authors reports a relationship with industry and other relevant entities – financial or otherwise – that might pose a conflict of interest in connection with the submitted article. The following authors report financial activities outside the submitted work:
      • Karl-Patrik Kresoja reports having received lecture honoraria from Bristol-Myers Squibb.
      • Matthias Ebner reports no relationships that could be construed as a conflict of interest.
      • Nina I. J. Rogge reports no relationships that could be construed as a conflict of interest.
      • Carmen Sentler reports no relationships that could be construed as a conflict of interest.
      • Karsten Keller reports no relationships that could be construed as a conflict of interest.
      • Lukas Hobohm reports having received lecture honoraria from MSD.
      • Gerd Hasenfuß reports having received consultancy and lecture honoraria from AstraZeneca, Berlin Chemie, Corvia, Impulse Dynamics, Novartis, Servier and Vifor Pharma; and editor honoraria from Springer International Publishing AG.
      • Stavros V. Konstantinides reports having received consultancy and lecture honoraria from Bayer, Boehringer Ingelheim, Daiichi-Sankyo, MSD and Pfizer-Bristol-Myers Squibb; and institutional grants from Actelion, Bayer, Boehringer Ingelheim, Daiichi-Sankyo and Pfizer-Bristol-Myers Squibb.
      • Burkert Pieske reports having received consultancy and lecture honoraria from Bayer, Daiichi Sankyo, MSD, Novartis, Sanofi-Aventis, Stealth Peptides and Vifor Pharma; and editor honoraria from the Journal of the American College of Cardiology.
      • Mareike Lankeit reports having received consultancy and lecture honoraria from Actelion, Bayer, Daiichi-Sankyo, MSD and Pfizer-Bristol-Myers Squibb.

      Acknowledgements

      None.

      References

        • Konstantinides S.V.
        • Torbicki A.
        • Agnelli G.
        • et al.
        2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism.
        Eur. Heart J. 2014; 35 (3069a-3069k): 3033-3069
        • Kearon C.
        • Akl E.A.
        • Ornelas J.
        • et al.
        Antithrombotic therapy for VTE disease: CHEST Guideline and Expert Panel report.
        Chest. 2016; 149: 315-352
        • Klok F.A.
        • Niemann C.
        • Dellas C.
        • Hasenfuß G.
        • Konstantinides S.
        • Lankeit M.
        Performance of five different bleeding-prediction scores in patients with acute pulmonary embolism.
        J. Thromb. Thrombolysis. 2016; 41: 312-320
        • Riva N.
        • Bellesini M.
        • Di Minno M.N.D.
        • et al.
        Poor predictive value of contemporary bleeding risk scores during long-term treatment of venous thromboembolism. A multicentre retrospective cohort study.
        Thromb. Haemost. 2014; 112: 511-521
        • Klok F.A.
        • Hösel V.
        • Clemens A.
        • et al.
        Prediction of bleeding events in patients with venous thromboembolism on stable anticoagulation treatment.
        Eur. Respir. J. 2016; 48: 1369-1376
        • Klok F.A.
        • Barco S.
        • Konstantinides S.V.
        External validation of the VTE-BLEED score for predicting major bleeding in stable anticoagulated patients with venous thromboembolism.
        Thromb. Haemost. 2017; 117: 1164-1170
        • Lecumberri R.
        • Alfonso A.
        • Jiménez D.
        • et al.
        Dynamics of case-fatalilty rates of recurrent thromboembolism and major bleeding in patients treated for venous thromboembolism.
        Thromb. Haemost. 2013; 110: 834-843
        • Budaj-Fidecka A.
        • Kurzyna M.
        • Fijałkowska A.
        • et al.
        In-hospital major bleeding predicts mortality in patients with pulmonary embolism: an analysis of ZATPOL Registry data.
        Int. J. Cardiol. 2013; 168: 3543-3549
        • Prandoni P.
        • Trujillo-Santos J.
        • Sanchez-Cantalejo E.
        • et al.
        Major bleeding as a predictor of mortality in patients with venous thromboembolism: findings from the RIETE Registry.
        Journal of thrombosis and haemostasis: JTH. 2010; 8: 2575-2577
        • Lankeit M.
        • Jiménez D.
        • Kostrubiec M.
        • et al.
        Predictive value of the high-sensitivity troponin T assay and the simplified Pulmonary Embolism Severity Index in hemodynamically stable patients with acute pulmonary embolism, a prospective validation study.
        Circulation. 2011; 124: 2716-2724
        • Hobohm L.
        • Hellenkamp K.
        • Hasenfuß G.
        • Münzel T.
        • Konstantinides S.
        • Lankeit M.
        Comparison of risk assessment strategies for not-high-risk pulmonary embolism.
        Eur. Respir. J. 2016; 47: 1170-1178
        • Lip G.Y.H.
        • Frison L.
        • Halperin J.L.
        • Lane D.A.
        Comparative validation of a novel risk score for predicting bleeding risk in anticoagulated patients with atrial fibrillation: the HAS-BLED (Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile INR, Elderly, Drugs/Alcohol Concomitantly) score.
        J. Am. Coll. Cardiol. 2011; 57: 173-180
        • Schulman S.
        • Kearon C.
        Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients.
        J. Thromb. Haemost.: JTH. 2005; 3: 692-694
        • Carrier M.
        • Le Gal G.
        • Wells P.S.
        • Rodger M.A.
        Systematic review, Case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism.
        Ann. Intern. Med. 2010; 152: 578-589
        • Klok F.A.
        • Kooiman J.
        • Huisman M.V.
        • Konstantinides S.
        • Lankeit M.
        Predicting anticoagulant-related bleeding in patients with venous thromboembolism, a clinically oriented review.
        Eur. Respir. J. 2015; 45: 201-210
        • Schulman S.
        • Kakkar A.K.
        • Goldhaber S.Z.
        • et al.
        Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.
        Circulation. 2014; 129: 764-772
        • Hull R.D.
        • Raskob G.E.
        • Rosenbloom D.
        • et al.
        Heparin for 5 days as compared with 10 days in the initial treatment of proximal venous thrombosis.
        N. Engl. J. Med. 1990; 322: 1260-1264
        • Palareti G.
        • Leali N.
        • Coccheri S.
        • et al.
        Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT).
        Italian Study on Complications of Oral Anticoagulant Therapy. Lancet (London, England). 1996; 348: 423-428
        • Prandoni P.
        • Lensing A.W.A.
        • Piccioli A.
        • et al.
        Recurrent venous thromboembolism and bleeding complications during anticoagulant treatment in patients with cancer and venous thrombosis.
        Blood. 2002; 100: 3484-3488
        • Ay C.
        • Pabinger I.
        • Cohen A.T.
        Cancer-associated venous thromboembolism, burden, mechanisms, and management.
        Thromb. Haemost. 2017; 117: 219-230
        • Farge D.
        • Trujillo-Santos J.
        • Debourdeau P.
        • et al.
        Fatal events in cancer patients receiving anticoagulant therapy for venous thromboembolism.
        Medicine. 2015; 94e1235
        • Lee A.Y.Y.
        • Kamphuisen P.W.
        • Meyer G.
        • et al.
        Tinzaparin vs warfarin for treatment of acute venous thromboembolism in patients with active cancer, a randomized clinical trial.
        JAMA. 2015; 314: 677-686
        • Raskob G.E.
        • van Es N.
        • Verhamme P.
        • et al.
        Edoxaban for the treatment of cancer-associated venous thromboembolism.
        N. Engl. J. Med. 2018; 378: 615-624

      Linked Article

      • Assessment of in-hospital bleeding risk in pulmonary embolism: What's the score? And what do I do with it?
        International Journal of CardiologyVol. 290
        • Preview
          The evaluation of bleeding risk is an imperative step when initiating anticoagulant therapy in the setting of venous thrombo-embolic disease. The risk of recurrent venous thrombo-embolism (VTE) in the absence of anticoagulation should be weighed against the risk of bleeding events that may occur under anticoagulant therapy. This may lead to vena cava filter insertion in patients with acute pulmonary embolism (PE) and absolute contraindications to anticoagulation, or conversely, to initiation of long-term therapy beyond 3 to 6 months in patients at high risk of recurrence combined with low bleeding risk [1].
        • Full-Text
        • PDF