Direct Oral Anticoagulants Versus Phenprocoumon: Mortality, Thromboembolism and Major Hemorrhage

A Systematic Review and Meta-Analysis of Routine Data Studies

Background: In the randomized controlled trials (RCTs) of the non–vitamin K dependent oral anticoagulants (NOACs) that led to their approval, the vitamin K antagonist (VKA) warfarin was used as a comparator drug. The efficacy and safety of NOACs compared to phenprocoumon, the VKA predominantly used in Germany, are unknown, as it has not been studied in RCTs.

Methods: For this systematic review and meta-analysis (registration: CRD42024619047), we systematically searched 3 databases for studies based on routine data that yielded matched or adjusted results for overall mortality, thromboembolism, and major hemorrhage in anticoagulant-naïve patients who were treated with either phenprocoumon or a NOAC.

Results: Seven studies based on German routine data were identified, covering a total of 1 842 015 anticoagulant-naïve patients. In a pooled analysis of all NOACs (apixaban, dabigatran, edoxaban, and rivaroxaban), these were associated with a notably higher risk of mortality (hazard ratio [HR] 1.14, 95% confidence interval [1.00; 1.30]), a higher risk of thromboembolic events (HR 1.08, [1.01; 1.15]), and a lower risk of major hemorrhages (HR 0.80, [0.72; 0.90]) than phenprocoumon. Among individual NOACs, rivaroxaban was associated with a notably higher risk of mortality than phenprocoumon; the other NOACs also displayed a higher risk of mortality than phenprocoumon, but the differences were not statistically significant.

Conclusion: In this meta-analysis of NOACs versus phenprocoumon, the former were found in a pooled analysis to be associated with a higher overall risk of mortality and a higher risk of thromboembolic events on the one hand, but a lower risk of major hemorrhages on the other hand. Warfarin was the comparator drug in all of the clinical trials that led to the approval of NOAC. The findings of this meta-analysis cast doubt on the benefit of the preferential use of NOACs in countries where phenprocoumon is the standard VKA.

Cite this as: Engelbertz C, Reinecke H, Köppe J: Direct oral anticoagulants versus phenprocoumon: Mortality, thromboembolism and major hemorrhage. A systematic review and meta-analysis of routine data studies. Dtsch Arztebl Int 2026; 123: 377–84. DOI: 10.3238/arztebl.m2026.0064

Thromboembolic events are the main cause of ischemic heart disease and stroke. Oral anticoagulation therapy is recommended for prophylaxis in various guidelines on atrial fibrillation (1, 2, 3). Currently, the non-vitamin K dependent oral anticoagulants (NOACs) apixaban, dabigatran, edoxaban, and rivaroxaban are preferred over vitamin K antagonists (VKAs).

In the pivotal randomized controlled trials (RCTs) for stroke prophylaxis in patients with atrial fibrillation, all NOACs were compared with warfarin (4, 5, 6, 7). Warfarin is the most commonly employed VKA worldwide (8), used, for example, in the USA, the UK, and China. The VKA predominantly used in Germany, phenprocoumon, was not tested in the RCTs (9). Phenprocoumon and warfarin differ widely in their pharmacological properties, e.g., plasma half-life and metabolism (10). The small number of RCTs comparing VKAs focus on prothrombin activity or time in therapeutic range (TTR), with inconclusive results (e1, e2, eSupplement Table 1). Observational studies from Denmark, Brazil, and Austria have shown an advantage of phenprocoumon over warfarin regarding quality of anticoagulation (e3, e4, e5, eSupplement Table 1), reporting a higher TTR in patients taking phenprocoumon than in those on warfarin.

Table

Characteristics of studies included in the main analysis

International guidelines all recommend NOACs over VKAs, making no distinction between the different VKAs (1, 2). This generalized recommendation is not evidence-based. More precisely, only three RCTs have examined whether NOACs are also non-inferior or superior to phenprocoumon, with inconsistent results (e6, e7, e8, eSupplement Table 1); two other RCTs have not yet published their results (e9, e10, eSupplement Table 1).

Several routine data studies have compared phenprocoumon with different NOACs (11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23). Routine data studies can supplement the findings of RCTs and can serve as an important enhancement to the assessment of the effectiveness and safety of drugs in daily life, but their results should always be interpreted with caution (24, 25). Due to the absence of RCTs comparing NOACs with phenprocoumon, we performed a meta-analysis using the available routine data studies that compare treatment of anticoagulation-naïve patients with phenprocoumon or one of the four NOACs, with the aim of addressing the paucity of comparative data between phenprocoumon and NOACs.

Methods

A detailed description of the methods and data analysis is given in the eSupplement. A literature search was performed in MEDLINE via PubMed, in the Web of Science, and in Scopus. The search terms, search strategy, and all retrieved studies are presented in eSupplement Figure 1 and eSupplement Tables 2 and 3.

We included studies based on retrospective routine data analyzing anticoagulation-naïve patients aged 18 years or older who had started treatment with either one of the four NOACs apixaban, dabigatran, edoxaban, and rivaroxaban or with phenprocoumon. All studies published adjusted results.

The pooled effects comparing NOACs with VKA were estimated using a random‑effects meta‑analysis to account for between‑study heterogeneity.

Results

The literature search identified 149 articles (eSupplement Table 3). After screening the titles and abstracts, 132 articles were excluded (eSupplement Figure 1). The full texts of the remaining 17 articles were examined in detail. Six of these publications were excluded due to violation of inclusion criteria. Six publications used the same database with substantial overlaps of study period, indications, and study design (11, 13, 14, 15, 16, 17), and another two studies were based on the same cohort (22, 23). Of these six studies with identical data source and two with the same cohort, we selected from the former only those with the most complete data (13, 16, 17) and from the latter the study with the standard NOAC dose (23); the remaining four studies (11, 14, 15, 22) were excluded. This ensured the independence of the studies and avoided double counting of patients and endpoints, which would otherwise have led to overweighting of study results. The final analysis comprised seven German routine data studies (12, 13, 16, 17, 19, 21, 23), one of which we ourselves published. The characteristics of the included studies are presented in the Table and in eSupplement Table 4.

Figure 1

Forest plots of the random-effects and fixed-effect meta-analyses for overall mortality

Study population

A total of 1 842 015 anticoagulation-naïve patients were included: 729 851 (39.6%) treated with phenprocoumon and 1 112 164 (60.4%) with NOACs (apixaban: n = 342 388, dabigatran: n = 114 908, edoxaban: n = 83 225, rivaroxaban: n = 585 188). The patient characteristics at baseline and the findings regarding the analyzed endpoints are presented in eSupplement Tables 5 and 6.

Endpoint overall mortality

Apixaban, dabigatran, and rivaroxaban were each represented by four studies in the analysis, three reporting results from propensity-score matched (PSM) cohorts and one reporting adjusted hazard ratios (HRs) from Cox regression analysis of the whole cohort. For edoxaban, three studies were included, one reporting PSM-adjusted HRs, the other two reporting results from Cox regression analyses. Apixaban (HR 1.22, 95% confidence interval [0.74; 2.01]), dabigatran (HR 1.04 [0.98; 1.11]), and edoxaban (HR 1.15 [0.47; 2.82]) showed a tendency toward higher overall mortality, but without statistical significance (Figure 1). Rivaroxaban (HR 1.22 [1.06; 1.40]) and the pooled analysis of all NOACs (HR 1.14 [1.00; 1.30]) were associated with a noticeably higher risk of death than phenprocoumon.

All studies except for one (17) reported event rates, incidence rates, or 1-year mortality rates. The 1-year mortality rates ranged from 4.6% to 5.3% for patients treated with phenprocoumon and 5.0% to 7.9% for those treated with NOACs (12). The event or incidence rates ranged from 3.0 to 8.2 events/100 patient-years for phenprocoumon-treated patients and from 2.4 to 9.5 events/100 patient-years for NOAC-treated patients (13, 16, 19, 21, 23; eSupplement Table 6). Details for all endpoints are presented in eSupplement Table 6.

Endpoint thromboembolic events

Apixaban, dabigatran, and rivaroxaban were each represented by five studies; for edoxaban, four studies were identified (Figure 2). The risk of thromboembolic events was noticeably elevated for all NOACs (HR 1.08 [1.01; 1.15]). Apixaban (HR 1.16 [0.97; 1.39]) and edoxaban (HR 1.11 [0.93; 1.33]) showed a tendency toward higher risk compared with phenprocoumon and a noticeably higher risk in the sensitivity analysis using the fixed-effect model. The results were inconclusive for dabigatran (HR 1.10 [0.85; 1.42]) and rivaroxaban (HR 1.01 [0.98; 1.05]).

Endpoint major hemorrhage

Apixaban, dabigatran, and rivaroxaban were each represented by five studies; for edoxaban, four studies were included (Figure 3). Apixaban (HR 0.69 [0.50; 0.94]), dabigatran (HR 0.75 [0.57; 0.99]), and all NOACs combined (HR 0.80 [0.72; 0.90]) had a much lower risk of major bleeding events than phenprocoumon. Edoxaban showed a tendency towards lower bleeding risk (HR 0.81 [0.58; 1.15]). No difference in bleeding risk was observed for rivaroxaban (HR 0.99 [0.89; 1.11]).

Figure 3

Forest plots of the random-effects and fixed-effect meta-analyses for major hemorrhages

Sensitivity analyses

Firstly, we performed a fixed-effect meta-analysis for all endpoints (Figures 1–3), showing an elevated risk of mortality for the individual NOACs as well as all NOACs and a higher risk of thromboembolic events for apixaban and edoxaban. The risk of bleeding was lower for apixaban, dabigatran, edoxaban, and all NOACs, supporting the results of the random-effects meta-analyses.

When including the results of the three distinct thromboembolic endpoints reported separately by Paschke et al. (19), edoxaban showed a tendency toward lower risk of thromboembolic events than phenprocoumon (eSupplement Figure 2).

Figure 2

Forest plots of the random-effects and fixed-effect meta-analyses for thromboembolic events

Finally, two sensitivity analyses for low-dose and normal-dose NOACs (16, 22, 23)—including one study that was not considered in the main analysis (22)—revealed the same trends as the main analysis, with the exception of normal-dose dabigatran, which had a tendency towards lower overall mortality than phenprocoumon (eSupplement Table 7).

Risk of bias

We found a moderate risk of bias for the included studies using the ROBINS-I tool, due to confounding and deviation from intended interventions. All studies applied statistical methods, e.g., PSM or regression analysis, to reduce the impact of confounders (eSupplement Figure 3). Nevertheless, a higher risk of bias is inherent in meta-analyses of routine data studies.

Discussion

The present study is the first meta-analysis of routine data studies comparing NOACs with phenprocoumon in anticoagulation-naïve patients. We found an elevated risk of overall mortality of at least 15% for all three factor Xa inhibitors compared with phenprocoumon, although only for rivaroxaban was the the higher risk of overall mortality statistically significant. Furthermore, apixaban and edoxaban showed a tendency towards a higher risk of thromboembolic events. The risk of major bleeding was reduced for apixaban, dabigatran, and edoxaban, although the elevated risk of overall mortality for edoxaban did not attain statistical significance. The 95% prediction intervals of the random-effects meta-analyses, illustrating the range of the predicted true treatment effect in a new study, support the observed trends. Therefore, this meta-analysis casts doubt on whether the NOACs are at least non-inferior to phenprocoumon regarding mortality and the prevention of thromboembolic events.

Compared with warfarin, the pivotal trials reported a significant reduction in overall mortality for apixaban and low-dose edoxaban, although the latter is not approved for use in Germany, and a tendency toward reduced risk of ischemic stroke for NOACs; also with regard to bleeding events, NOACs were mostly superior (4, 5, 6, 7). These findings were supported by meta-analyses of observational studies (eSupplement Figure 4, 27, 28, 29, 30), but are in stark contrast to our meta-analysis using phenprocoumon instead of warfarin as the comparator drug. Solely the reduced risk of major hemorrhage with NOACs was consistent regardless of the VKA used.

Two of the studies evaluated (16, 23) showed a tendency towards lower mortality risk for apixaban, edoxaban, and dabigatran. All of the other studies we analyzed revealed a higher overall mortality risk for the various NOACs. Paschke et al. (19) reported higher mortality for apixaban (74.45 vs. 64.26/1000 patient-years [PY]), dabigatran (63.49 vs. 52.34/1000 PY), and rivaroxaban (75.02 vs. 57.79/1000 PY) than for phenprocoumon, but lower rates for edoxaban (30.62 vs. 53.58/1000 PY). These differences may be caused by the more heterogeneous patient cohorts in routine data studies than in RCTs. The studies analyzed here used data from five different sources (two statutory health insurance funds, two companies collecting data predominantly from company health insurance funds, one nationwide database) which differ in patient characteristics (age, sex, socio-economic features) and regional availability (31). Moreover, there was notable variation of about 15% to 30% in the prevalence of severe comorbidities, e.g., cancer, in the included studies.

Other German and Austrian routine data studies also point to higher overall mortality risk with NOACs than with phenprocoumon. Mueller et al. (18) found greater mortality risk for NOACs than for phenprocoumon in PSM cohorts (HR 1.22 [1.17; 1.28]). Preinreich et al. (20) found a significantly higher risk of mortality for apixaban (HR 1.27 [1.18; 1.37]) and rivaroxaban (HR 1.24 [1.15; 1.32]) in an Austrian study.

The primary indication for prescribing oral anticoagulation is the prevention of thromboembolic events in patients with atrial fibrillation without increasing the bleeding risk. In daily clinical practice, effectiveness and safety may differ from the RCT results. Subanalyses of the pivotal trials by geographic region showed that no significant difference between NOACs and warfarin was found for stroke/systemic embolism (risk ratio [RR] 0.97 [0.85; 1.11]) in European countries (32). In Western Europe, warfarin even appeared to be more favorable than apixaban (RR 1.20 [0.71; 2.03]) or edoxaban (RR 1.47 [0.89; 2.43]), resulting in only marginal differences in avoidable events per 1000 patients. This low efficacy of NOACs was attributed to better control of the international normalized ratio (INR) and a higher TTR in patients treated with VKA (32).

Few RCTs have used phenprocoumon as comparator drug (e6, e7, e8). The AXADIA-AFNET 8 study (n = 97 patients, all on hemodialysis) found no difference in efficacy and safety between apixaban and phenprocoumon (e6). A small study (n = 16) with patients on left ventricular assist device therapy was ended prematurely due to excess thromboembolic events with dabigatran (e7). Finally, a Dutch RCT (n = 1330) showed that switching from acenocoumarol or phenprocoumon to an NOAC was associated with more thromboembolic events (HR 1.26 [0.60; 2.61]) and more bleeding events (HR 1.69 [1.23; 2.32]) (e8).

The more favorable outcomes with phenprocoumon may be explained by its two- to threefold longer half-life than warfarin, leading to more stable plasma levels and a higher TTR (10, e3, e4, e5). Regular medical contacts by patients treated with VKA—mainly for INR monitoring—may also have had a positive incidental effect due to timely identification of health issues. Furthermore, an important role may also have been played by the phenomenon, frequently observed in clinical routine, of non-indicated dose reduction of NOACs deviating from guideline recommendations (33).

Strengths and limitations

The present meta-analysis was based on a large sample of anticoagulation-naïve patients, reducing the bias that might be found with anticoagulant-experienced patients, e.g., in terms of bleeding events. All studies reported adjusted or matched effect estimates. These inclusion criteria were selected to reduce confounding by indication. The endpoints were clinically relevant events, diminishing the probability of missed events in the health insurance databases. In particular, documentation of the endpoint overall mortality, the only definitive endpoint, is 100% complete, as the probability of missed events is extremely low for the health insurance funds.

The studies included used exclusively data from German health insurance funds, so double counting of patients cannot be ruled out. This applies especially to the studies that used data provided by a company (13, 16, 17). These studies gave no information as to which health insurance funds had provided their claims data to the company.

Routine data studies are usually more heterogeneous, and quality of analysis depends also on the statistical methods used. Although the studies we analyzed reported adjusted or matched estimates, potential unmeasured residual confounding or confounding by indication cannot be ruled out. The lack of information on patient adherence and persistence, laboratory values, causes of death, physician adherence to dosing according to indication, and serious comorbidities, e.g., cancer, that might have influenced decisions on the type of anticoagulation may contribute to residual confounding. Furthermore, most studies did not provide information on exact NOAC dosage.

The included studies used different outcome definitions for thromboembolic and major bleeding events. Some studies defined hemorrhagic stroke as part of a composite endpoint, “thromboembolic events.” To clearly distinguish between thromboembolic and bleeding events, we only considered results of thromboembolic events without hemorrhagic stroke.

Finally, this meta-analysis is based on a low number of studies, leading to a higher risk of bias.

Conclusion and clinical perspective

This meta-analysis of routine data studies questions the assumption of superior or at least non-inferior treatment outcomes for NOACs compared with phenprocoumon. The risk of overall mortality was noticeably higher for rivaroxaban. Both apixaban and edoxaban showed a tendency towards higher risk of overall mortality and of thromboembolic events than phenprocoumon. While the reduced risk of major hemorrhage with apixaban and dabigatran remains a relevant clinical advantage, this benefit must be weighed carefully against the trend towards increased overall mortality and augmented thromboembolic risk.

However, the risk of bias in this meta-analysis on routine data studies is inherently higher than in meta-analyses of RCTs.

The general preference for NOACs over phenprocoumon should, therefore, be reconsidered—particularly if the high costs of NOACs are taken into account: In the year 2023 alone, the costs of NOACs amounted to over € 2.5 billion in Germany (34). Thus, with a view to patient welfare and costs, a RCT comparing NOACs with phenprocoumon is urgently needed.

Data sharing
The data used in this systematic review and meta-analysis were extracted from published studies available elsewhere. All processed data are presented in this article and the supplements.

Conflict of interest statement

HR has received speaker honoraria from NeoVasc, Corvia, MedUpdate, StreamedUp, and NovoNordisk. He has acted as a consultant for Daiichi Sankyo, Pfizer, DiaPlan, and Pluristem, receiving in part financial compensation for his consulting activities. His department at University Hospital Münster has taken part or is still taking part in multicenter trials conducted by BARD, BMS/Pfizer, BIOTRONIK, and NovoNordisk, receiving patient fees or financial compensation.

JK has received research funding, sponsored by Stryker, from the German Society of Trauma Surgery.

CE has received an honorarium for manuscript writing from Forum Sanitas.

Manuscript submitted on 15 September 2025, revised version received on 14 April 2026

Corresponding author
Dr. rer. nat. Christiane Engelbertz, PhD

References

1. Van Gelder IC, Rienstra M, Bunting KV, et al..: 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J 2024; 45: 3314–414

2. Steffel J, Verhamme P, Potpara TS, et al..: The 2018 European Heart Rhythm Association Practical Guide on the use of non-vitamin K antagonist oret al. anticoagulants in patients with atrial fibrillation. Eur Heart J 2018; 39: 1330–93

3. Deutsche Gesellschaft für Kardiologie: S3-Leitlinie Vorhofflimmern – Version 1. https://register.awmf.org/assets/guidelines/019-014l_S3_Vorhofflimmern_2025-07.pdf (last accessed on 19 May 2026).

4. Connolly SJ, Ezekowitz MD, Yusuf S, et al.: Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009; 361: 1139–51

5. Patel MR, Mahaffey KW, Garg J, et al.: Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011; 365: 883–91

6. Granger CB, Alexander JH, McMurray JJ, et al.: Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011; 365: 981–92

7. Giugliano RP, Ruff CT, Braunwald E, et al.: Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2013; 369: 2093–104

8. Pirmohamed M: Warfarin: Almost 60 years old and still causing problems. Br J Clin Pharmacol 2006; 62: 509–11

9. Le Heuzey JY, Ammentorp B, Darius H, et al.: Differences among western European countries in anticoagulation management of atrial fibrillation. Data from the PREFER IN AF registry. Thromb Haemost 2014; 111: 833–41

10. Stehle S, Kirchheiner J, Lazar A, Fuhr U: Pharmacogenetics of oral anticoagulants: A basis for dose individualization. Clin Pharmacokinet 2008; 47: 565–94

11. Bonnemeier H, Huelsebeck M, Kloss S: Comparative effectiveness of rivaroxaban versus a vitamin K antagonist in patients with renal impairment treated for non-valvular atrial fibrillation in Germany—a retrospective cohort study. Int J Cardiol Heart Vasc 2019; 23: 100367

12. Engelbertz C, Marschall U, Feld J, et al.: Apixaban, edoxaban and rivaroxaban but not dabigatran are associated with higher mortality compared to vitamin-K antagonists: A retrospective German claims data analysis. J Intern Med 2024; 296: 362–76

13. Hohmann C, Lutz M, Vignali S, et al.: Clinical outcomes in patients receiving edoxaban or phenprocoumon for prevention of stroke in atrial fibrillation: A German real-world cohort study. Thromb J 2022; 20: 37

14. Hohmann C, Hohnloser SH, Jacob J, Walker J, Baldus S, Pfister R: Non-vitamin K oral anticoagulants in comparison to phenprocoumon in geriatric and non-geriatric patients with non-valvular atrial fibrillation. Thromb Haemost 2019; 119: 971–80

15. Hohnloser SH, Basic E, Nabauer M: Comparative risk of major bleeding with new oral anticoagulants (NOACs) and phenprocoumon in patients with atrial fibrillation: A post-marketing surveillance study. Clin Res Cardiol 2017; 106: 618–28

16. Hohnloser SH, Basic E, Hohmann C, Nabauer M: Effectiveness and safety of non-vitamin K oral anticoagulants in comparison to phenprocoumon: Data from 61,000 patients with atrial fibrillation. Thromb Haemost 2018; 118: 526–38

17. Kreutz R, Kloss S, Enders D, et al.: Comparative effectiveness of factor Xa non-vitamin K antagonist oral anticoagulants versus phenprocoumon in patients with non-valvular atrial fibrillation. Int J Cardiol 2024; 404: 131894

18. Mueller S, Groth A, Spitzer SG, Schramm A, Pfaff A, Maywald U: Real-world effectiveness and safety of oral anticoagulation strategies in atrial fibrillation: A cohort study based on a German claims dataset. Pragmat Obs Res 2018; 9: 1–10

19. Paschke LM, Klimke K, Altiner A, von Stillfried D, Schulz M: Comparing stroke prevention therapy of direct oral anticoagulants and vitamin K antagonists in patients with atrial fibrillation: A nationwide retrospective observational study. BMC Med 2020; 18: 254

20. Preinreich J, Sheikh Rezaei S, Mittlböck M, Greisenegger S, Reichardt B, Wolzt M: Oral anticoagulant therapy and outcome in patients with stroke. A retrospective nation-wide cohort study in Austria 2012–2017. Pharmacoepidemiol Drug Saf 2021; 30: 1332–8

21. Ujeyl M, Köster I, Wille H, et al.: Comparative risks of bleeding, ischemic stroke and mortality with direct oral anticoagulants versus phenprocoumon in patients with atrial fibrillation. Eur J Clin Pharmacol 2018; 74: 1317–25

22. Warkentin L, Hueber S, Deiters B, Klohn F, Kühlein T: Vitamin-K-antagonist phenprocoumon versus low-dose direct oral anticoagulants (DOACs) in patients with atrial fibrillation: A real-world analysis of German claims data. Thromb J 2022; 20: 31

23. Warkentin L, Klohn F, Deiters B, Kühlein T, Hueber S: Vitamin-K-antagonist phenprocoumon versus direct oral anticoagulants in patients with atrial fibrillation: A real-world analysis of German claims data. BMJ Open 2023; 13: e063490

24. Sherman RE, Anderson SA, Dal Pan GJ, et al.: Real-world evidence—what is it and what can it tell us? N Engl J Med 2016; 375: 2293–7

25. Camm AJ, Fox KAA: Strengths and weaknesses of ‚real-world‘ studies involving non-vitamin K antagonist oral anticoagulants. Open Heart 2018; 5: e000788

26. Sterne JA, Hernán MA, Reeves BC, et al.: ROBINS-I: A tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016; 355: i4919.

27. Chan YH, Chen SW, Chan CY, Chao TF: Comparative safety and effectiveness of non-vitamin K oral anticoagulants versus warfarin in patients with non-valvular atrial fibrillation: A network meta-analysis. J Formos Med Assoc 2024; 123: 578–86

28. Deitelzweig S, Bergrath E, di Fusco M, et al.: Real-world evidence comparing oral anticoagulants in non-valvular atrial fibrillation: A systematic review and network meta-analysis. Future Cardiol 2022; 18: 393–405

29. Coleman CI, Briere JB, Fauchier L, et al..: Meta-analysis of real-world evidence comparing non-vitamin K antagonist oral anticoagulants with vitamin K antagonists for the treatment of patients with non-valvular atrial fibrillation. J Mark Access Health Policy 2019; 7: 1574541

30. Ntaios G, Papavasileiou V, Makaritsis K, Vemmos K, Michel P, Lip GYH: Real-world setting comparison of nonvitamin-K antagonist oral anticoagulants versus vitamin-K antagonists for stroke prevention in atrial fibrillation: A systematic review and meta-analysis. Stroke 2017; 48: 2494–503

31. Kreis K, Neubauer S, Klora M, Lange A, Zeidler J: Status and perspectives of claims data analyses in Germany—a systematic review. Health Policy 2016; 120: 213–26

32. Gómez-Outes A, Terleira-Fernández AI, Calvo-Rojas G, Suárez-Gea ML, Vargas-Castrillón E: Direct oral anticoagulants for stroke prevention in patients with atrial fibrillation: Meta-analysis by geographic region with a focus on European patients. Br J Clin Pharmacol 2016; 82: 633–44

33. Camm AJ, Cools F, Virdone S, et al.: Mortality in patients with atrial fibrillation receiving nonrecommended doses of direct oral anticoagulants. J Am Coll Cardiol 2020; 76: 1425–36

34. Mühlbauer B, Ludwig WD: Arzneiverordnungen 2023 im Überblick. In: Ludwig WD, Mühlbauer B, Seifert R (eds.): Arzneiverordnungs-Report 2024. Springer Berlin, Heidelberg, 2025; 11–15

e1. Rodman T, Pastor BH, Resnick M. Phenprocoumon, diphenadione, warfarin and bishydroxycoumarin: A comparative study. Am J Med Sci. 1964; 247: 655–660

e2. van Leeuwen Y, Gebuis EPA, van der Meer FJM, Rosendaal FR. Late breaking clinical trial: randomised comparison of the quality of anticoagulant therapy with two vitamin K antagonists: warfarin versus phenprocoumon. J Thromb Haemost. 2009; 7: 1202. LB-MO-001.

e3. Jensen CF, Christensen TD, Maegaard M, Hasenkam JM. Quality of oral anticoagulant therapy in patients who perform self management: warfarin versus phenprocoumon. J Thromb Thrombolysis 2009; 28: 276–281

e4. Leiria TL, Pellanda L, Miglioranza MH, Sant‘anna RT, Becker LS, Magalhães E, Lima GG. [Warfarin and phenprocoumon: experience of an outpatient anticoagulation clinic]. Arq Bras Cardiol 2010; 94: 41–45

e5. Schlöglhofer T, Marschütz A, Combs P, Stonebraker C, Lupo S, Jeevanandam V, Riebandt J, Schima H, Zimpfer D, Meehan K. Quality of Anticoagulation With Phenprocoumon and Warfarin in Left Ventricular Assist Device Patients: A Multicenter Study. ASAIO J 2023; 69: 595–601

e6. Reinecke H, Engelbertz C, Bauersachs R, et al.: A Randomized Controlled Trial Comparing Apixaban With the Vitamin K Antagonist Phenprocoumon in Patients on Chronic Hemodialysis: The AXADIA-AFNET 8 Study. Circulation 2023; 147: 296–309

e7. Andreas M, Moayedifar R, Wieselthaler G, et al.: Increased Thromboembolic Events With Dabigatran Compared With Vitamin K Antagonism in Left Ventricular Assist Device Patients: A Randomized Controlled Pilot Trial. Circ Heart Fail 2017; 10: e003709

e8. Joosten LPT, van Doorn S, van de Ven PM, et al.: Safety of Switching From a Vitamin K Antagonist to a Non-Vitamin K Antagonist Oral Anticoagulant in Frail Older Patients With Atrial Fibrillation: Results of the FRAIL-AF Randomized Controlled Trial. Circulation 2024; 149: 279–289

e9. Ferner M, Wachtlin D, Konrad T, et al.: Rationale and design of the RE-LATED AF--AFNET 7 trial: REsolution of Left atrial-Appendage Thrombus--Effects of Dabigatran in patients with Atrial Fibrillation. Clin Res Cardiol 2016; 105: 29–36

e10. Riesinger L, Strobl C, Leistner DM, et al.: Apixaban versus PhenpRocoumon: Oral AntiCoagulation plus antiplatelet tHerapy in patients with Acute Coronary Syndrome and Atrial Fibrillation (APPROACH-ACS-AF): Rationale and design of the prospective randomized parallel-group, open-label, blinded-endpoint, superiority, multicenter-trial of a triple therapy versus a dual therapy in patients with Atrial Fibrillation and Acute Coronary Syndrome undergoing coronary stenting. Int J Cardiol Heart Vasc 2021; 35: 100810