Severe Hemorrhage Associated With Oral Anticoagulants
A prospective observational study of the clinical course during treatment with vitamin K antagonists or direct oral anticoagulants
; ; ; ; ; ; ; ; ; ; ; ; ; ;
Background: Few data have been published to date on outcomes after the common clinical experience of severe hemorrhage in orally anticoagulated patients.
Methods: A prospective, multicenter observational study was carried out to investigate outcomes and management in a series of consecutive patients who sustained a severe hemorrhage under treatment with vitamin K antagonists (VKA) or direct oral anticoagulant drugs (DOAC). The primary endpoint was in-hospital death up to and including day 30 after hospital admission. The secondary endpoints were the duration of bleeding, in-hospital death due to hemorrhage (as defined by the study physician examining the patient’s records), the use of antagonists, the extent of supportive measures used to stop the hemorrhage, and an assessment of causality. Consecutive patients were recruited until a predefined number of patients was reached in both groups.
Results: Among 193 patients with severe hemorrhage, 97 had been taking a VKA, and 96 had been taking a DOAC. 13.0 % (95% confidence interval [8.6; 18.5]; 25/193) of the overall group patients died in the first 30 days after hospital admission, including 17.5% ([10.6; 26.6]; 17/97) in the VKA group and 8.3% ([3.7; 15.8]; 8/96) in the DOAC group (p = 0.085). The median duration of bleeding was 19.8 hours in the VKA group and 27.8 hours in the DOAC group (p = 0.632). The in-hospital mortality due to hemorrhage was higher in the VKA group than in the DOAC group (15.5% [15/97] versus 4.2% [4/97]; p = 0.014). Only the use of prothrombin complex concentrates (PCCs) lowered the median duration of hemorrhage in the two patient groups. In 35% (68/193) of the patients, the hemorrhage was caused by an external influence, most commonly a fall.
Conclusion: The in-hospital mortality was higher among patients treated with VKA than among patients treated with DOAC, although the difference failed to reach statistical significance.
Patients with nonvalvular atrial fibrillation or venous thromboembolism require anticoagulation with a vitamin K antagonists (VKA) or direct oral anticoagulant (DOAC) in therapeutic doses (1, 2, 3, 4, 5). Meta-analyses of Phase III studies found that the efficacy of DOACs in preventing stroke or recurrent venous thrombosis was comparable to that of VKAs. Intracranial hemorrhage, fatal hemorrhage and clinically relevant non-severe hemorrhage occurred significantly less frequently in DOAC-treated patients compared to VKA-treated patients, while gastrointestinal hemorrhage was more commonly observed among patients treated with rivaroxaban, edoxaban and dabigatran compared to VKA-treated patients (6, 7). Since the conduct of the pivotal studies retrospective or single-arm prospective registries on the management and outcome of severe hemorrhage have been published, showing differences in the outcomes of patients treated with DOAC versus VKA (eTable 1, 8, 9, 10, 11, 12, 13, 14).
The management of major bleeding in patients on oral anticoagulation therapy is still challenging. Prothrombin complex concentrates (PCCs) are the antidote of choice for VKA-treated patients with severe hemorrhage. Even though specific antidotes are available, PCC is commonly used to treat severe hemorrhage in DOAC-treated patients, despite the limited evidence supporting the effectiveness of PCC in this patient population (8, 10, 11, 15, 16, 17).
There is a lack of prospective data comparing the management and outcomes in patients experiencing major bleeding during treatment with DOACs or VKAs in everyday clinical practice.
We therefore initiated the RADOA (Reversal Agent use in patients treated with Direct Oral Anticoagulants or vitamin K antagonists) registry project designed to evaluate the outcome and management of consecutive DOAC- or VKA-treated patients who were admitted to hospital with severe hemorrhages or conditions requiring urgent intervention. The registry also allows to analyze effective management strategies which can be used to control severe hemorrhage.
The RADOA registry is a prospective, observational, non-interventional, open-label, investigator-initiated multicenter German registry, documenting the management of severe hemorrhages and/or urgent interventions in patients treated with VKA or DOAC. The aims and design of this registry have already been described elsewhere (18).
The registry does not allow to calculate the prevalence rates for the various types of hemorrhage. Furthermore, comparisons between the various treatment groups (VKA vs. DOAC) are not confirmative but explorative; both noticed and unnoticed confounders may occur.
Patient recruitment continued until the predefined sample size was reached in each group. The clinical course and the outcome were prospectively documented up to and including day 30 after hospital admission. The participating centers were hospitals with interdisciplinary teams on-call 24/7 to treat anticoagulation-associated hemorrhage in their emergency departments and intensive care units. The study protocol was reviewed and approved by all relevant institutional ethics committees (IECs). An external independent monitor performed all checks (100%) of the source data on site.
Patients with major bleeding while receiving VKA or DOAC therapy were included in the registry. The modified definition of the International Society of Thrombosis and Haemostasis (ISTH) for non-surgical patients was used for hemorrhage classification (19) (definitions of major bleeding episodes in the eBox).The characteristics of the included patients are detailed in Table 1.
In 2014, the first patient was included in the registry. The majority of patients was recruited between January 2016 and March 2018.
Since the clinical condition studied represented a medical emergency, it had to be ensured that obtaining the informed consent of the patient for inclusion in the registry did not interfere with or delay the provision of acute care. Therefore, the patients’ written informed consent was only obtained after completion of acute treatment. This procedure was approved by all ethics committees. In patients who could not provide their written consent due to their poor state of health, the consent was obtained from their legal representative. Data of patients who were unconscious or died before a legal representative had been named were also included in the registry. This procedure was expressly approved by the ethics committees to prevent significant distortion of outcomes as the result of exclusion of the most severely affected patients. The study complies with the Declaration of Helsinki.
Primary and secondary endpoints
The primary endpoint was the in-hospital mortality up to and including day 30 after admission to hospital. The secondary endpoints were:
1. The time to the cessation of bleeding after hospital admission
2. The mortality associated with major bleeding
3. The use versus non-use of reversal agents with regard to the clinical outcome
4. The level of hemostasis with regard to supportive treatments
5. The efficacy of dialysis in patients treated with dabigatran, and
6. The causality of serious adverse events with regard to anticoagulants.
Hemostasis was considered achieved when the bleeding improved instantly, when the use of red blood cell transfusions, factor concentrates and/or antidotes was no longer required and/or when the local investigator documented that the bleeding had ceased completely (see definition of hemostasis in the eBox).
Deaths were recorded as bleeding-associated or non-bleeding-associated according to the evaluation of the local investigators. Thus, the analysis of the bleeding-associated mortality rates should be interpreted with caution.
In patients with severe hemorrhage, sample-size estimation was based on the total length of the two-sided 95% confidence intervals for the incidence of the primary endpoint. These were estimated using the PASS software (NCSS, LLC. Kaysville, Utah, USA). For a sample size of 90 patients in each group (either VKA or DOAC), the total width of the confidence interval was 13.5%, assuming an incidence of the primary endpoint of approximately 10% or less. It was 12% if an endpoint incidence below 7.5% was assumed, ensuring acceptable accuracy.
The primary statistical aim was to evaluate in-hospital mortality up to and including day 30 after hospital admission with two-sided 95% confidence intervals, using the Clopper-Pearson method. All comparisons of the primary endpoint and secondary endpoints were explorative and were not adjusted for multiple testing.
The log-rank test was used to estimate and compare the cumulative incidence curves for the primary endpoint of in-hospital mortality. As a sensitivity analysis, the odds ratio (OR) of the Fisher‘s exact test was compared to the odds ratio of a multivariable logistic regression analysis, taking into account indication, sex and body weight as potential confounders.
The secondary endpoint of bleeding-associated mortality was characterized by 95% confidence intervals and compared between treatment groups. In addition, competing risk models as described by Aalen-Johansen were used with a two-sided Gray test to estimate and compare cumulative incidence curves for bleeding-associated deaths and deaths due to other causes were used as competing risks.
Additional secondary exploratory analyses are shown in the Tables 1–3. Dichotomous and categorical variables were described by number and proportions and compared using a two-sided Fisher‘s exact test. Furthermore, risk differences were described using 95% score confidence intervals. Continuous and ordinal variables were described using the median (1st to 3rd quartile) and compared using two-sided Mann–Whitney U tests or Kruskal-Wallis tests. As a sensitivity analysis, a stratified Mann-Whitney U test was used to compare bleeding duration in the various patient groups according to site of bleeding (intracranial, gastrointestinal or other anatomical sites). Statistical analysis was performed using the R software, version 3.5.0 (R: Foundation for Statistical Computing, Vienna, Austria).
Between May 2014 and March 2018, altogether 194 consecutive patients with severe hemorrhage during oral anticoagulation therapy were included in the study at 10 sites in Germany. 97 patients received VKA (all received phenprocoumon) and 96 patients received DOACs. In der DOAC group, 41 patients on apixaban (43%), 46 patients on rivaroxaban (48%), 5 patients on edoxaban (5%), and 4 patients on dabigatran (4%) were treated. Another patient who was treated with phenprocoumon and apixaban (medication error) was excluded from further analysis.
The patients’ mean age was 79 years; the demographic characteristics of the two treatment groups were comparable with regard to age, sex, indication for anticoagulation, known renal disease, and severity of the condition (Table 1).
On admission, intracranial hemorrhage was more frequently observed in patients treated with VKA compared to patients receiving DOAC therapy (66.0% vs. 46.9%, p = 0.009) (Table 2), bringing about a higher rate of unconsciousness in the VKA group (24.7% vs. 8.3%, p = 0.003) (Table 1). In contrast, gastrointestinal bleeding was found less frequently in patients treated with VKA compared to patients treated with DOAC (19.6% vs. 32.3%, p = 0.050) (Table 2), resulting in a lower hemorrhagic shock rate in the VKA group (3.1% vs. 14,6%), p = 0.005) (Table 1).
Primary endpoint (in-hospital mortality)
Within the first 30 days after hospital admission, 25 of the 193 patients (13.0%, 95% confidence interval: [8.6; 18.5]) died. Of these 25 patients, 17.5% ([10.6; 26.6]; 17/97) died in the VKA group compared to 8.3% ([3.7; 15.8]; 8/96) in the DOAC group (OR = 0.43. p = 0.085) (Table 3, Figure 1). A sensitivity analysis found similar effects, taking into account indication, sex and body weight as potential confounders (OR = 0.40. p = 0.069).
1. The cessation of bleeding occurred after a median of 19.8 hours (interquartile range: 4.9–69.9) in the VKA group compared to 27.8 hours (7.6–63.4) in the DOAC group (p = 0.632).
2. The bleeding-related in-hospital mortality rates were higher in the VKA group compared to the DOAC group (15/97, 15.5% [8.9; 24.2] compared to 4/96, 4.2% [1.1; 10.3]; p = 0.014) (Table 3, eFigure 1). Fourteen of the 15 VKA-treated patients died as a result of intracranial hemorrhages and one patient died from retroperitoneal hemorrhage. In 40% (6/15) of the deceased VKA patients and 39.2% (31/79) of the surviving VKA patients the INR was >3.0 (p = 1.000). The median INR was comparable in both VKA groups (deceased VKA patients: INR 2.75. 1.66–3.15; surviving VKA patients: INR: 2.50. 1.65–4.00; p = 0.691). All 4 bleeding- related deaths in the DOAC group were caused by intracranial hemorrhage.
3. The majority of VKA-treated patients received PCC (72.2%, 70/97) compared to 47.9% (46/96) of patients in the DOAC group. Vitamin K was more frequently administered to the VKA-treated patients. The frequency of use of fresh frozen plasma (FFP), tranexamic acid and desmopressin was comparable in the VKA and DOAC groups (eFigure 2). The patients in the VKA group required less red blood cell concentrates compared to the patients in the DOAC group (36.1% vs. 53.1%, p = 0.021) and nominally fewer platelet concentrates (4.1% vs. 10.4%; p = 0.104) (eTables 4 and 5, eFigure 2).
4. In the VKA group, the median duration of hemorrhage was 10.0 hours among the patients treated with PCC compared to 41.5 hours among the patients without PCC treatment (p = 0.021; with stratification by bleeding site: p = 0.040) (eTable 6a). While bleeding duration was found reduced in PCC-treated patients with CNS hemorrhage, no difference was found between treated and non-treated patients if the bleeding site was not located in the CNS (Figure 2a). DOAC-treated patients also showed a shorter median bleeding duration when they received PCC (11.0 hours vs. 33.1 hours among patients who had not received PCC, p = 0.009; with stratification by bleeding site: p = 0.005) (eTabelle 6b). All types of bleeding (CNS and non-CNS hemorrhages) showed reductions in bleeding duration after PCC treatment compared to untreated patients (Figure 2b). For all other treatments with coagulation factors or medications, no effect on bleeding duration was observed in either of the two treatment groups (eTables 7–10).
5. None of the 4 patients treated with dabigatran received hemodialysis. Two patients received idarucizumab.
6. 55 patients (28.5%; 30/97 and 25/96 of the VKA- and DOAC-treated patients, respectively) developed severe hemorrhages after falls. In 13 patients (6.7%; 5/97 and 8/96 of the VKA- and DOAC-treated patients, respectively), the bleeding was caused by other external factors (9 polytrauma, 2 suicide attempts, 1 hemorrhage after tooth extraction, 1 hemorrhage after a brawl). In the majority of patients, anticoagulation was the only identifiable factor triggering the development of major bleeding (64.8%, 125/193).
The RADOA registry provides prospective evidence obtained in an everyday clinical setting showing that the in-hospital mortality risk is nominally higher in VKA-treated patients with major bleeding compared to DOAC-treated patients with major bleeding, even though this difference did not reach the level of statistical significance. A comparison of the overall mortality among the two anticoagulation patient groups included in the RADOA registry with that of a recent retrospective Canadian cohort study including more than 2000 orally anticoagulated patients with major bleeding supports the robustness of the RADOA registry data. Even though the patients in the Canadian study received warfarin, while the patients in the RADOA registry were treated with phenprocoumon, the mortality rates among VKA-treated patients were comparable between the two study populations (17.5% and 15.2% in the RADOA registry and the Canadian study, respectively). Likewise, comparable mortality rates were found among DOAC-treated patients in the two studies (8.3% and 9.8% in the RADOA registry and the Canadian study, respectively) (10).
Another recently conducted prospective multicenter study with the same research topic analyzed the mortality rate within 30 days after hospital admission in 806 patients with major bleeding who received either VKA (76%) or DOAC (24%) therapies. Here again, the mortality rates among the VKA-treated patients were similar to the ones in the RADOA registry (18% vs. 17.5% in the RADOA registry) and significantly higher compared to those found for the DOAC-treated patients (9% vs. 8.3% in the RADOA registry) (20).
The second important insight from the RADOA registry was that PCC administration in DOAC-treated patients with major bleeding was associated with a 66% reduction in bleeding duration. Thanks to the RADOA registry, prospective data on the efficacy of PCC in this situation have become available for the first time.
The specific antidote for reversal of anticoagulation with factor Xa inhibitors, andexanet alpha, is not available in many hospitals. In contrast, PCC is available in a hospital setting (at least in Europe) and most clinicians are familiar with the use of PCC in patients experiencing major bleeding while receiving anticoagulation therapy. In addition, the cost of andexanet alfa may be too high for many hospitals.
We are well aware that the typical limitations of a non-randomized observational registry also apply to the RADOA registry. In everyday clinical routine, registry data can be influenced by variations in indications. In addition, it is the severity of the hemorrhage that determines whether or not an antidote is administered. Therefore, any attempt to compare different treatments could be challenged.
On the other hand, conducting randomized studies in patients with major bleeding while on oral anticoagulation is difficult and a placebo group as a comparator would be unacceptable from an ethical point of view in this life-threatening situation.
Patients were included prospectively and consecutively in this study to achieve the highest possible data quality and minimize bias. All included patients were overseen on site by an independent external monitor. Selection bias is minimal since all consecutive patients were included in the registry study, including patients unable to give their informed consent.
In conclusion, the RADOA registry shows that in-hospital mortality is higher in VKA-treated patients with major bleeding compared to DOAC-treated patients and that PCC may be an alternative to specific antidotes in DOAC-treated patients with major bleeding.
The results indicate that there is an urgent need for further research in this clinically relevant field. Well-designed randomized trials comparing andexanet alfa with PCC for the treatment of major hemorrhages in patients on DOAC therapy could help to improve patient care in this life-threatening situation in the future.
Conflict of interest
Prof. Lindhoff-Last received lecture fees and consulting fees from Bayer, Boehringer Ingelheim, Bristol-Myers Squibb/Pfizer, Daiichi-Sankyo, Portola, CSL Behring, and Aspen. She received external funds for the conduct of a clinical contract study from Bayer and DAIICHI Sankyo, for a research project which she initiated from Bayer AG, Bristol-Myers Squibb/Pfizer, Daiichi-Sankyo, and CSL-Behring.
Prof. Konstantinides served as a paid consultant for Bayer, Daiichi-Sankyo, and Bristol-Myers Squibb/Pfizer. He received lecture fees from Bayer, Boehringer Ingelheim and Bristol-Myers Squibb/Pfizer. He received external funds for research projects from Bayer and Boehringer Ingelheim.
PD Dr. Grottke served as a paid consultant for Bayer, Boehringer Ingelheim, CSL Behring, Octapharma, Sanofi, Shire, Pfizer, and Portola. He received lecture fees and reimbursement of travel expenses from Boehringer Ingelheim, Portola, CSL Behring, Octapharma, Sanofi, and Shire. In research projects, he was responsible for external funds from Bayer, Boehringer Ingelheim, CSL Behring, Octapharma, Novo Nordisk, Nycomed, and Portola.
Prof. Nowak-Goettl served as a paid consultant for Bayer Vital and Boehringer Ingelheim.
Prof. von Heymann served as a paid consultant for Daiichi Sankyo and Grünenthal. He received lecture fees and reimbursement of travel expenses from Bayer, Biotest, Bristol-Myers Squibb, CSL Behring, Daiichi Sankyo, Grünenthal, Leo Pharma, Mitsubishi Pharma, NovoNordisk, and Pfizer. In the conduct of a clinical contract study, he was responsible for external funds from Daiichi Sankyo.
PD Dr. Birschmann received reimbursement of travel expenses from Bristol-Myers Squibb/Pfizer and lecture fees from Bristol-Myers Squibb and Bayer Vital.
PD Dr. Beyer-Westendorf served as a paid consultant for Bayer, Daiichi Sankyo and Portola. He received reimbursement of travel expenses and lecture fees from Bayer, Daiichi Sankyo and Portola. He received external funds for research projects and contract studies from Bayer, Daiichi Sankyo, Portola, Pfizer, and Boehringer Ingelheim.
Prof. Schellong received lecture fees and consulting fees from Bristol-Myers Squibb, Boehringer Ingelheim, Bayer, Daiichi-Sankyo, and Aspen.
Prof. Meybohm received reimbursement of travel expenses and lecture fees from CSL Behring. He received fees for the conduct of clinical contract studies from B. Braun Melsungen, CSL Behring, Fresenius Kabi, and Vifor Pharma.
Prof. Greinacher served as a paid consultant for Boehringer Ingelheim, Bristol-Myers Squibb, Bayer Healthcare, and Portola. He received fees for expert opinions from Aspen, Boehringer Ingelheim, MSD, Bristol-Myers Squibb, and Portola. He received fees for clinical contract studies from Portola and Ergomed.
Prof. Herrmann, Dr. Lindau, Dr. Lucks, Dr. Zydek, and Dr. Sümnig declare no conflict of interest.
Manuscript received on 1 November 2019, revised version accepted on 9 March 2020
Translated from the original German by Ralf Thoene, MD.
Prof. Dr. med. Edelgard Lindhoff-Last
CardioAngiologisches Centrum Bethanien (CCB)
CCB Gerinnungszentrum und CCB Studienzentrum
Im Prüfling 23
60389 Frankfurt, Germany
Cite this as:
Lindhoff-Last E, Herrmann E, Lindau S, et al.: Severe hemorrhage associated with oral anticoagulants—a prospective observational study of the clinical course during treatment with vitamin K antagonists or direct oral anticoagulants.
Dtsch Arztebl Int 2020; 117: 312–9. DOI: 10.3238/arztebl.2020.0312
eFigures, eBox, eTables:
Coagulation Center and Research Center, Cardioangiology Center Bethanien Hospital (CCB), Frankfurt/Main: Prof. Dr. med. Edelgard Lindhoff-Last
Institute of Biostatistics und Mathematical Modelling, Goethe University, Frankfurt: Prof. Dr. rer. nat. Eva Herrmann
Department of Anesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Frankfurt, Frankfurt/Main: Dr. med. Simone Lindau
Center for Thrombosis and Hemostasis (CTH), University Medical Center Mainz, Mainz: Prof. Dr. med. Stavros Konstantinides
Department of Anesthesiology, RWTH Aachen University Hospital, Aachen: PD Dr. med. Oliver Grottke
Department of Clinical Chemistry, Thrombosis and Hemostasis Unit, University Hospital of Kiel and Lübeck, Kiel: Prof. Dr. med. Ulrike Nowak-Goettl
Research Center of the Cardioangiology Center Bethanien Hospital (CCB), Frankfurt/Main: Dr. phil. nat. Jessica Lucks, Dr. med. Barbara Zydek
Department of Anesthesiology, Intensive Care and Emergency Medicine and Pain Therapy, Vivantes Klinikum im Friedrichshain, Berlin: Prof. Dr. med. Christian von Heymann
Institute for Laboratory and Transfusion Medicine, Heart and Diabetes Center, Ruhr University Bochum, Bochum: PD Dr. med. Dr. rer. nat. Ingvild Birschmann
Department of Immunology and Transfusion medicine, University Medicine Greifswald, Greifswald: Dr. med. Ariane Sümnig
Thrombosis Research Unit, Center for Internal Medicine 1, Department of Hematology, University Hospital Dresden, Dresden, Germany, and Department of Haematology and Oncology, Kings College London, England: PD Dr. med. Jan Beyer-Westendorf
Medical Department 2, Municipal Hospital Dresden-Friedrichstadt, Dresden: Prof. Dr. med. Sebastian Schellong
Department of Anesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Frankfurt, Frankfurt am Main, and Department of Anesthesiology, University Hospital Würzburg, Würzburg: Prof. Dr. med. Patrick Meybohm
Department of Immunology and Transfusion Medicine, University Medicine Greifswald, Greifswald: Prof. Dr. med. Andreas Greinacher
|1.||Patel MR, Mahaffey KW, Garg J, et al.: Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011; 365: 883–91 CrossRef MEDLINE|
|2.||Connolly SJ, Ezekowitz MD, Yusuf S, et al.: Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009; 361: 1139–51 CrossRef MEDLINE|
|3.||Granger CB, Alexander JH, McMurray JJ, et al.: Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011; 365: 981–92 CrossRef MEDLINE|
|4.||Einstein investigators, Bauersachs R, Berkowitz SD, et al.: Oral rivaroxaban for symptomatic venous thromboembolism. N Engl J Med 2010; 363: 2499–510 CrossRef MEDLINE|
|5.||Giugliano RP, Ruff CT, Braunwald E, et al.: Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2013; 369: 2093–104 CrossRef MEDLINE|
|6.||van Es N, Coppens M, Schulman S, Middeldorp S, Buller HR: Direct oral anticoagulants compared with vitamin K antagonists for acute venous thromboembolism: evidence from phase 3 trials. Blood 2014; 124: 1968–75 CrossRef MEDLINE|
|7.||Ruff CT, Giugliano RP, Braunwald E, et al.: Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet 2014; 383: 955–62 CrossRef MEDLINE|
|8.||Majeed A, Agren A, Holmstrom M, et al.: Management of rivaroxaban- or apixaban-associated major bleeding with prothrombin complex concentrates: a cohort study. Blood 2017; 130: 1706–12 CrossRef MEDLINE|
|9.||Halvorsen S, Ghanima W, Fride Tvete I, et al.: A nationwide registry study to compare bleeding rates in patients with atrial fibrillation being prescribed oral anticoagulants. Eur Heart J Cardiovasc Pharmacother 2017; 3: 28–36 CrossRef MEDLINE PubMed Central|
|10.||Xu Y, Schulman S, Dowlatshahi D, et al.: Direct oral anticoagulant- or warfarin-related major bleeding: characteristics, reversal strategies, and outcomes from a multicenter observational study. Chest 2017; 152: 81–91 CrossRef MEDLINE|
|11.||Albaladejo P, Samama CM, Sie P, et al.: Management of severe bleeding in patients treated with direct oral anticoagulants: an observational registry analysis. Anesthesiology 2017; 127: 111–20 CrossRef MEDLINE|
|12.||Green L, Tan J, Morris JK, et al.: A three-year prospective study of the presentation and clinical outcomes of major bleeding episodes associated with oral anticoagulant use in the UK (ORANGE study). Haematologica 2018; 103: 738–45 CrossRef MEDLINE PubMed Central|
|13.||Kurogi R, Nishimura K, Nakai M, et al.: Comparing intracerebral hemorrhages associated with direct oral anticoagulants or warfarin. Neurology 2018; 90: e1143-9 CrossRef MEDLINE PubMed Central|
|14.||Green L, Tan J, Antoniou S, et al.: Haematological management of major bleeding associated with direct oral anticoagulants – UK experience. Br J Haematol 2019; 185: 514–22 CrossRef MEDLINE|
|15.||Piran S, Khatib R, Schulman S, et al.: Management of direct factor Xa inhibitor-related major bleeding with prothrombin complex concentrate: a meta-analysis. Blood Adv 2019; 3: 158–67 CrossRef MEDLINE PubMed Central|
|16.||Steffel J, Verhamme P, Potpara TS, et al.: The 2018 European Heart Rhythm Association practical guide on the use of non-vitamin K antagonist oral anticoagulants in patients with atrial fibrillation. Eur Heart J 2018; 39: 1330–93 CrossRef MEDLINE|
|17.||Dager WE, Roberts AJ, Nishijima DK: Effect of low and moderate dose FEIBA to reverse major bleeding in patients on direct oral anticoagulants. Thromb Res 2019; 173: 71–6 CrossRef MEDLINE|
|18.||Lindhoff-Last E: Direct oral anticoagulants (DOAC) – management of emergency situations. Hämostaseologie 2017; 37: 257–66 CrossRef|
|19.||Schulman S, Kearon C, Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis: Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005; 3: 692–4 CrossRef MEDLINE|
|20.||Becattini C, Franco L, Beyer-Westendorf J, et al.: Major bleeding with vitamin K antagonists or direct oral anticoagulants in real-life. Int J Cardiol 2017; 227: 261–6 CrossRef MEDLINE|