Internal and Emergency Medicine https://doi.org/10.1007/s11739-018-1843-9

DEBATE

Has time come for the use of direct oral anticoagulants in the extended prophylaxis of venous thromboembolism in acutely ill medical patients? No Francesco Marongiu1 · Doris Barcellona1  Received: 18 February 2018 / Accepted: 21 March 2018 © SIMI 2018

Abstract Acutely ill hospitalized medical patients are at high risk of venous thromboembolism (VTE). Although thromboprophylaxis in these patients is recommended since 2004 by the American College of Chest Physicians, it is widely underused. The doubt as to whether or not to treat patients at high VTE risk after hospital discharge came from the knowledge that this risk may persist after the hospital admission period. Two meta-analyses comparing extended- versus short-duration prophylaxis are published. The results demonstrate an unfavorable balance between VTE prevention and incidence of major bleeding in patients assigned to extended-duration thromboprophylaxis. Only in the APEX study, betrixaban, a direct inhibitor of factor Xa, shows similar efficacy and safety compared to enoxaparin. However, while it is very promising, oral anticoagulant phase III studies and post-marketing registers are lacking. Moreover, betrixaban has a long half-life, an excretion in the gut by means of P-glycoprotein, and the lack of an antidote. These characteristics and the meta-analysis results prompt us to answer no to the extended thromboprophylaxis in hospitalized medical patients, at least now. Keywords  Direct oral anticoagulants · Venous thromboembolism · Extended thromboprophylaxis It is known that sudden death due to venous thromboembolism (VTE) may occur during hospitalization [1] in a percentage ranging from 5 to 15% [2]. Overall, medical patients are the most vulnerable ones, so that in the past few years, an effort to minimize high VTE risk has been made by means of randomized controlled trials (RCTs). In particular, these RCTs were planned to compare enoxaparin, dalteparin and fondaparinux versus placebo in hospitalized medical patients [3–5], showing that these pharmacological interventions are able to reduce more than half the incidence of VTE. These impressive results prompted the American College of Chest Physicians (ACCP) to recommend since 2004 anticoagulant thromboprophylaxis for acutely ill hospitalized medical patients with congestive heart failure, severe respiratory disease, or confined to bed with one or more risk factors such as active cancer, previous

* Doris Barcellona [email protected] 1



Department of Medical Sciences and Public Health, University of Cagliari, Cagliari, Italy

VTE, sepsis, acute neurologic disease, or inflammatory bowel disease [6]. However, prophylaxis with anticoagulants is reported to be widely underused. In 2010, the global Endorse survey evaluated data of 37,356 patients across 32 countries demonstrating that only a minority of at risk hospitalized medical patients receive ACCP-recommended prophylaxis [7]. This is thought to be due to the lack of precise VTE risk assessment. In 2012, ACCP [8] proposed the use of a score aimed at focussing the attention on high risk patients: the Padua Prediction Score (PPS) [9]. PPS was the result of a study carried out in Padua and published in 2010. This was a prospective cohort study that enrolled 1180 consecutive patients admitted to an Internal Medicine ward over a 2-year period. They were classified as having a high or low risk of VTE on the basis of a predefined risk assessment model (RAM). The aim was to achieve a hazard ratio of VTE considering high risk patients submitted to in-hospital thromboprophylaxis versus those who did not. Four hundred and sixty-nine patients (39.7%) were found to be at high risk of thrombosis. VTE occurred in 2.2% of patients treated with thromboprophylaxis, while this figure was 11.0% in those who did not (HR of VTE, 0.13; 95% CI 0.04–0.40).

13

Vol.:(0123456789)



Only two patients (0.3%) at low risk developed VTE. This study for the first time demonstrates that a thromboprophylaxis in hospitalized VTE high risk patients can significantly reduce thromboembolic events paying a low risk of bleeding (1.6%). In 2014, PPS had an external validation by a multicentre study addressed to hospitalized medical patients at risk of VTE: the Geneva Risk Score [10]. A total of 1478 hospitalized medical patients entered the study. A percentage of 43% of them did not receive thromboprophylaxis. VTE occurred in 3.2% of patients at high risk score, while only 0.6% of patients at low risk score suffered from VTE, which occurred in 3.5% of patients without prophylaxis versus 0.8% of those with anti-thrombotic prophylaxis. These figures are similar to those obtained using PPS: 3.5% (95% CI 2.3–5.3%) in 714 high risk vs 1.1% (95% CI 0.6–2.3%) in 764 low risk patients. The authors conclude that both the Geneva Score and PPS are able to identify patients at high VTE risk, and that the two scores are quite comparable overall for the identification of low risk patients who do not necessitate an anti-thrombotic prophylaxis. In 2016, Rafizadeh et al. [11] published a retrospective study confirming that PPS, as another prediction score (IMPROVE), could reduce the inappropriate anti-thrombotic prophylaxis in low risk patients. Recently, a high PPS score is found to be associated with a hypercoagulative state, studied by means of the Clot Waveform Analysis, in hospitalized medical patients before receiving anti-thrombotic prophylaxis [12]. In these patients, the velocity and the acceleration of the clot formation detected in vitro by a simple coagulative test (aPTT) is significantly higher than in patients at low VTE risk. If it is evident that anti-thrombotic prophylaxis is of value in avoiding a large number of VTE episodes during hospitalization, one wonders if it could be important to extend this prophylaxis at home or at residences for elderly care. It is known that the risk of pulmonary embolism (PE) and deep vein thrombosis (DVT) may persist after hospital discharge in acutely ill medical patients [13]. However, no studies have clearly supported the routine use of extended-duration antithrombotic prophylaxis in this setting. In 2012, ACCP recommend avoidance of anti-thrombotic prophylaxis once a patient is discharged from the hospital [8]. This recommendation, although weak (2C), is based on the fact that studies dealing with this aspect are observational, and that there is a lack of solid patient-end point. In other words, there is uncertainty on whether thromboprophylaxis is of benefit in this setting. In 2017, two meta-analyses were published on possible extended anti-thrombotic prophylaxis in hospitalized medical ill patients [14, 15]. In the first one, Dentali et al. define as efficacy outcomes the prevention of symptomatic DVT, PE, and VTE-related mortality. Principal safety outcome was major bleeding. Four RCTs comparing extended-duration prophylaxis (24–47 days) versus short-duration prophylaxis

13

Internal and Emergency Medicine

(6–14 days) were included for a total of 28,105 hospitalized medical patients. RCTs included in this meta-analysis are the following: EXCLAIM [16], ADOPT [17], MAGELLAN [18] and APEX [19]. Pooled results show that extended thromboprophylaxis reduces the risk of both DVT (OR 0.50, 95% CI 0.28–0.88) and VTE (OR 0.54, 95% CI 0.29–0.99), while the reduction of PE risk did not reach statistical significance. However, the risk of major bleeding is higher in patients treated with thromboprophylaxis (OR 2.09, 95% CI 1.33–3.29). The risk benefit analysis shows that the Number Needed to Treat (NNT) is 339 and 239 for DVT and VTE, respectively, while the Number Needed to Harm (NNH) for major bleeding is 247. The high OR value for major bleeding and the value of NNH close to that of NNT for VTE induced the authors to conclude that thromboprophylaxis is not to be considered in medical ill patients beyond the period of hospitalization. The second meta-analysis considered the same studies analyzed by Dentali et al. Results are similar confirming a correct search of RCTs dealing with this topic. Among the four RCTs considered in both metaanalyses cited above, the only one that shows a non significant difference between the experimental anticoagulant and the comparator is the APEX study. This study confers a heterogeneity to the final random effect model lowering the OR obtained for major bleeding. The APEX study was planned to compare subcutaneous enoxaparin (40 mg once daily) for 10 ± 4 days plus oral betrixaban placebo for 35–42 days or subcutaneous enoxaparin placebo for 10 ± 4 days plus oral betrixaban (at a loading dose of 160 mg for the first dose and then 80 mg once daily) for 35–42 days. Patients with severe renal failure were treated with 50% of the prespecified dose (i.e., 20 mg of enoxaparin or a loading dose of 80 mg of betrixaban and then 40 mg once daily). Patients who were treated with a concomitant P-glycoprotein inhibitor received a reduced dose of betrixaban (40 mg once daily). A total of 7513 patients were randomized. Three groups of patients were analyzed: (a) with high d-dimer (cohort 1), (b) with high d-dimer plus those aged 75 years or older (cohort 2), and (c) the overall population cohort. In cohort 1, the primary efficacy outcome (asymptomatic and symptomatic proximal or distal DVT, symptomatic nonfatal PE, or death from VTE) occurs in 6.9% of patients treated with betrixaban and 8.5% receiving enoxaparin (RR 0.81; 95% CI 0.65–1.00). These rates become 5.6 and 7.1%, (RR 0.80; 95% CI 0.66–0.98; p = 0.03) in cohort 2 and 5.3 and 7.0% (RR 0.76; 95% CI 0.63–0.92; p = 0.006) in the overall population, respectively. In the overall population, major bleeding is 0.7% in betrixaban group and 0.6% in the enoxaparin group (RR 1.19; 95% CI 0.67–2.12). Betrixaban, the latest available DOAC, is a direct activated factor X inhibitor (Por-tola Pharmaceuticals Inc. USA) derived from 1,N-(5-chloropyridin-2-yl)-2-(4-(N,N-dimethylcar-

Internal and Emergency Medicine

bamimidoyl)-benzamido)benzamide, which is an anthranilamide-based compound. Betrixaban, at a dose of 80 mg daily, reaches a peak plasma concentration within 3–4 h (in healthy volunteers). The oral bioavailability is 34%, and is affected by fatty food, which can reduce the peak concentration by 50%. Betrixaban is excreted in the gut by about 85% by means of a P-glycoprotein efflux pump. It is minimally metabolized (< 1%) by CYP450 enzymes, and does not induce or inhibit this cytochrome. Betrixaban is characterized by having a low renal clearance (5–7%) [20]. However, the half-life is 37 h, while pharmacodynamic half-life is ∼ 20 h. Therefore, a stable anticoagulation may be reached. If we compare the characteristics of betrixaban with those of the other DOACs, it appears clear that it could be the drug of choice for patients with a glomerular filtration rate less than 30 ml/min, a cut-off now considered a contraindication to the use of these new oral anticoagulants. However, if this point may be of paramount importance on one hand, its metabolic characteristics may potentially be problematic on the other. In fact, its use could be dangerous if bleeding occurs, or urgent surgery is required in the absence of an antidote. Again, we still do not know if an accumulation of this drug might ensue, especially after long periods of treatment. Finally, since P-glycoprotein has an important role in determining its efflux, drugs that inhibit or enhance this pump system might significantly interfere in inducing wide variations of the betrixaban plasma concentration. Up to now, other than the phase III APEX study, only phase II dose-finding studies have been carried out: VTE prevention in total knee replacement (Expert) and stroke prevention in atrial fibrillation (Explore Xa). However, we still do not have enough information on possible adverse events. In other words, betrixaban seems to be a very promising oral anticoagulant dedicated to critical patients, since the risk of bleeding is generally low, and it has the best risk to benefit profile as compared to previous extended treatment trials using other DOACs. However, we think that phase III studies and post-marketing registers on both DVT and atrial fibrillation are required before completely accepting betrixaban among the different anticoagulant options. In the end, what is our answer to the question “Has time come for the use of direct oral anticoagulants in the extended prophylaxis of venous thromboembolism in acutely ill medical patients”? Considering the unfavorable balance between NNT and NNH expressed by the two meta-analysis quoted above, and the lack of phase III studies and post-marketing registers, our answer is, at least now, no.

Compliance with ethical standards  Conflict of interest  The authors declare that they have no conflict of interest.

Statement of human and animal rights  All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent  Informed consent was not required for this type of study.

References 1. Nichols L, Chew B (2012) Causes of sudden unexpected death of adult hospital patients. J Hosp Med 7:706–708 2. Francis CW (2007) Clinical practice. Prophylaxis for thromboembolism in hospitalized medical patients. N Engl J Med 356:1438–1444 3. Samama MM, Cohen AT, Darmon JY et al (1999) A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients. Prophylaxis in Medical Patients with Enoxaparin Study Group. N Engl J Med 341:793–800 4. Leizorovicz A, Cohen AT, Turpie AG et al (2004) Randomized, placebo-controlled trial of dalteparin for the prevention of venous thromboembolism in acutely ill medical patients. Circulation 110:874–879 5. Cohen AT, Davidson BL, Gallus AS et al (2006) Efficacy and safety of fondaparinux for the prevention of venous thromboembolism in older acute medical patients: randomised placebo controlled trial. Br Med J 332:325–329 6. Geerts WH, Pineo GF, Heit JA et al (2004) Prevention of venous thromboembolism: the seventh ACCP conference on antithrombotic and thrombolytic therapy. Chest 126(3 Suppl):338S–400S 7. Bergmann JF, Cohen AT, Tapson VF, ENDORSE Investigators et al (2010) Venous thromboembolism risk and prophylaxis in hospitalized medically ill patients. The ENDORSE Global Survey. Thromb Haemost 103:736–748 8. Kahn SR, Lim W, Dunn AS et  al (2012) Prevention of VTE in nonsurgical patients: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 141(2 Suppl):e195S–e226S 9. Barbar S, Noventa F, Rossetto V et al (2010) A risk assessment model for the identification of hospitalized medical patients at risk for venous thromboembolism: the Padua Prediction Score. J Thromb Haemost 8:2450–2457 10. Nendaz M, Spirk D, Kucher N et al (2014) Multicentre validation of the Geneva Risk Score for hospitalized medical patients at risk of venous thromboembolism Explicit ASsessment of Thromboembolic RIsk and Prophylaxis for Medical PATients in SwitzErland (ESTIMATE). Thromb Haemost 111:531–538 11. Rafizadeh R, Ricky D, Turgeon RD et al (2016) Characterization of venous thromboembolism risk in medical inpatients using different clinical risk assessment models. Can J Hosp Pharm 69:454–459 12. Ruberto MF, Marongiu F, Mandas A et al (2018) The venous thromboembolic risk and the clot wave analysis: a useful relationship? Clin Chem Lab Med 56:448–453 13. Amin AN, Varker H, Princic N et al (2012) Duration of venous thromboembolism risk across a continuum in medically ill hospitalized patients. J Hosp Med 7:231–238 14. Dentali F, Mumoli N, Prisco D et al (2017) Efficacy and safety of extended thromboprophylaxis for medically ill patients A

13

meta-analysis of randomised controlled trials. Thromb Haemost 117:606–617 15. Liew AYL, Siavash Piran S, Eikelboom JW et al (2017) Extendedduration versus short-duration pharmacological thromboprophylaxis in acutely ill hospitalized medical patients: a systematic review and meta-analysis of randomized controlled trials. J Thromb Thrombolysis 43:291–301 1 6. Hull RD, Schellong SM, Tapson VF et al (2010) Extended-duration venous thromboembolism prophylaxis in acutely ill medical patients with recently reduced mobility: a randomized trial. EXCLAIM (Extended Prophylaxis for Venous ThromboEmbolism in Acutely ill Medical Patients with Prolonged Immobilization) study. Ann Intern Med 153:8–18

13

Internal and Emergency Medicine 17. Goldhaber SZ, Leizorovicz A, Kakkar AK, ADOPT Trial Investigators et  al (2011) Apixaban versus enoxaparin for thromboprophylaxis in medically ill patients. N Engl J Med 365:2167–2177 18. Cohen AT, Spiro TE, Buller HR et al (2013) Rivaroxaban for thromboprophylaxis in acutely ill medical patients. N Engl J Med 368:513–523 19. Cohen AT, Harrington RA, Goldhaber SZ, APEX Investigators et al (2016) Extended thromboprophylaxis with betrixaban in acutely ill medical patients. N Engl J Med 375:534–544 20. Chan NC, Bhagirath V, Eikelboom JW (2015) Profile of betrixaban and its potential in the prevention and treatment of venous thromboembolism. Vasc Health Risk Manag 11:343–351