Use of Direct Oral Anticoagulants in People Living with HIV: A Single-Center Experience

Direct oral anticoagulants (DOACs), also termed non–vitamin K antagonist oral anticoagulants (NOACs), have been specifically developed to overcome various challenges posed by vitamin K antagonists, and have positively enhanced the outlook regarding anticoagulant-driven prevention of stroke and thromboembolic events.[1] [2] [3] Indeed, more than 70% of human immunodeficiency virus (HIV)-negative patients requiring anticoagulant therapy are now treated with these drugs.[1] [2] [3] Conversely, published experiences on the use of NOACs in people living with HIV are limited to case reports, small case series, and analyses of theoretically anticipated drug–drug interactions (DDIs) with antiretrovirals (reviewed in West et al and Cattaneo et al[4] [5]). Here, we sought to assess the use of NOACs in HIV-infected patients from our Infectious Disease Clinic.

The database of our clinic, with nearly 2,700 HIV-positive patients on active follow-up, was retrospectively investigated to search for patients starting antithrombotic therapy after January 2012, or starting therapy on any date but modifying therapy after 2012, which was the year of the marketing of the first NOAC in Italy. The risk of DDIs between antiretroviral and antithrombotic therapies was scored using the University of Liverpool HIV Drug Interactions checker (https://www.hiv-druginteractions.org/checker) which is as follows: “red flag” (do not co-administer), “orange/yellow flag” (potential DDI), or “green flag” (no DDI expected). Approval by Ethics Committee was deemed unnecessary because, under Italian law, such approvals are required only in the hypothesis of prospective clinical trials on medical products for clinical use and not for retrospective, observational studies (decree 211/2003). Written informed consent for medical procedures/interventions performed for routine treatment purposes was collected for each patient.

Out of the 76 HIV-infected patients, 50 patients on antithrombotic therapy from our database fulfilling the inclusion criteria were further considered and clinical characteristics are given in [Table 1]. All patients except one had started anticoagulant therapy after being diagnosed with HIV. The anticoagulant therapy was associated with a significant increase in hemoglobin concentrations (from 137 ± 24 to 143 ± 22 g/L, p = 0.036) with no effects on serum creatinine (from 1.0 ± 0.4 to 1.1 ± 0.4 mg/dL, p = 0.409) or on alanine aminotransferase (from 35 ± 31 to 30 ± 39 IU/L, p = 0.513). Only 14% of the patients were administered NOACs, with the large majority still being treated with a vitamin K antagonist (66% with warfarin and 16% with acenocoumarol), despite the fact that these regimens resulted in 58% of the cases at risk of potential DDI (orange flag). We subsequently proposed a hypothetical scenario in which all patients were ideally shifted to NOACs by checking their DDIs with the current antiretroviral regimens. We found that all patients could have been theoretically shifted to the novel anticoagulant therapy. Remarkably, this shift, if applied, would have been resulted in a further reduction from 58% to 42% in the orange flag DDIs ([Table 1]).

Table 1 Demographic and clinical characteristics of the 50 HIV-infected patients on oral antithrombotic therapies identified from our database

Characteristics	Current situation	Hypothetical NOAC-based scenario
Patients, n	50
Age, y	66 ± 13
Body weight, kg	77 ± 12
CD4 cell count, cells/µL	629 ± 256
Patients with HIV RNA > 50 copies/mL	1
Antiretroviral therapy	30% abacavir-based 36% tenofovir-based 10% lamivudine-based 24% NRTI free 38% boosted PI-based[a] 30% NNRTI-based[a] 64% INI-based[a]
Days of antiretroviral therapy, mean ± SD	3,923 ± 2,015
Antithrombotic therapy	82% vitamin K antagonists 14% NOACs 4% others[b]	100% NOACs
Indications for antithrombotic therapy	52% atrial fibrillation 24% DVT 10% PE 10% DVT + PE 4% valve replacement
Days of antithrombotic therapy, mean ± SD	2,624 ± 2,328
Liverpool score for potential DDIs	0% red flag 58% orange flag 42% green flag	0% red flag 42% orange flag 58% green flag

Abbreviations: DDI, drug–drug interactions; DVT, deep vein thrombosis; HIV, human immunodeficiency virus; INI, integrase inhibitors; NNRTI, non-nucleoside reverse transcriptase inhibitors; NOACs, non–vitamin K antagonist oral anticoagulants; NRTI, nucleoside reverse transcriptase inhibitors; PE, pulmonary embolism; PI, protease inhibitors; RNA, ribonucleic acid; SD, standard deviation.

^a The sum is > 100% because patients are treated with different combinations including more than one antiretroviral class.

^b One patient each on fondaparinux and enoxaparin.

Patients with HIV and atrial fibrillation or thromboembolic disease have a higher stroke risk, with guidelines strongly recommending anticoagulation in the majority of cases.[6] [7] However, consensus guidelines have suggested that the coadministration of NOACs with antiretroviral drugs should be considered with caution, mainly because of the risk of potential DDIs.[8] In fact, some DDIs involving apixaban, dabigatran, or rivaroxaban have recently been reported when shifting from ritonavir to cobicistat-based (or vice versa) antiretroviral regimens.[5] On the other hand, it must be considered that clinically relevant DDIs have been reported also between antiretrovirals and warfarin or acenocoumarol, two drugs characterized by a narrow therapeutic index and difficult management in several clinical conditions, such as renal insufficiency, polypharmacy, and hypoalbuminemia, compared with NOACs.[9] [10] [11] [12] [13] [14] Therefore, the choice of leaving HIV patients on vitamin K antagonists may not automatically translate in fewer DDIs compared with NOACs, as documented in the present study. Indeed, we found that the large majority of HIV-infected patients from our database were treated with warfarin and, most importantly, this was associated with a more than 50% risk of “orange flag” DDIs. By designing a hypothetical scenario, we found that 100% of patients could theoretically be shifted to NOAC, by selecting an optimal drug, with a further reduction in the orange flag DDIs. This has also become possible thanks to the recent marketing of edoxaban, an NOAC characterized by less variable pharmacokinetics and lower propensity for DDIs compared with other NOACs. Edoxaban is only minimally metabolized by cytochrome P450 3A4.[15] In support, Serrao et al have recently reported the successful use of the full dose of edoxaban in three HIV-infected patients with optimal tolerability during a 6- to 10-month follow-up.[16] It is, therefore, likely that the widespread use of edoxaban could further increase the use of NOACs in people living with HIV, irrespective of the antiretroviral regimens adopted.

Given, the retrospective, observational design of the present study, we were not able to assess if the changes in the anticoagulation therapy might impact the clinical outcomes of our patients. Nevertheless, some small case series are available reporting the successful use of NOACs, selected individually and eventually used at reduced doses, in HIV patients also if treated with ritonavir- or cobicistat-boosting antiretroviral therapies.[5]

In conclusion, an underuse of NOACs was found in HIV-infected patients from our clinics, probably linked to an unwarranted fear for DDIs. A rationale selection of the best antithrombotic agent for each antiretroviral therapy would improve the quality of life of people living with HIV.

Authors' Contributions

All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.

Publication History

Article published online:
23 December 2020

• References

• 1 Lutsey PL, Walker RF, MacLehose RF, Alonso A, Adam TJ, Zakai NA. Direct oral anticoagulants and warfarin for venous thromboembolism treatment: trends from 2012 to 2017. Res Pract Thromb Haemost 2019; 3 (04) 668-673
• 2 Lippi G, Gosselin R, Favaloro EJ. Current and emerging direct oral anticoagulants: state-of-the-art. Semin Thromb Hemost 2019; 45 (05) 490-501
• 3 Lippi G, Mattiuzzi C, Cervellin G, Favaloro EJ. Direct oral anticoagulants: analysis of worldwide use and popularity using Google trends. Ann Transl Med 2017; 5 (16) 322
• 4 West TA, Perram J, Holloway CJ. Use of direct oral anticoagulants for treatment of atrial fibrillation in patients with HIV: a review. Curr Opin HIV AIDS 2017; 12 (06) 554-560
• 5 Cattaneo D, Cossu MV, Rizzardini G. Pharmacokinetic drug evaluation of ritonavir (versus cobicistat) as adjunctive therapy in the treatment of HIV. Expert Opin Drug Metab Toxicol 2019; 15 (11) 927-935
• 6 Pastori D, Mezzaroma I, Pignatelli P, Violi F, Lip GYH. Atrial fibrillation and human immunodeficiency virus type-1 infection: a systematic review. Implications for anticoagulant and antiarrhythmic therapy. Br J Clin Pharmacol 2019; 85 (03) 508-515
• 7 Jackson BS, Pretorius E. Pathological clotting and deep vein thrombosis in patients with HIV. Semin Thromb Hemost 2019; 45 (02) 132-140
• 8 Steffel J, Verhamme P, Potpara TS. et al; ESC Scientific Document Group. 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 (16) 1330-1393
• 9 Tseng AL, Luetkehoelter J, Walmsley SL. Increase in international normalized ratio after switching from atazanavir/ritonavir to darunavir/cobicistat in a patient on warfarin: boosters are not always equal. AIDS 2017; 31 (01) 175-176
• 10 Good BL, Gomes DC, Fulco PP. An unexpected interaction between warfarin and cobicistat-boosted elvitegravir. AIDS 2015; 29 (08) 985-986
• 11 Esterly JS, Darin KM, Gerzenshtein L, Othman F, Postelnick MJ, Scarsi KK. Clinical implications of antiretroviral drug interactions with warfarin: a case-control study. J Antimicrob Chemother 2013; 68 (06) 1360-1363
• 12 Liedtke MD, Vanguri A, Rathbun RC. A probable interaction between warfarin and the antiretroviral TRIO study regimen. Ann Pharmacother 2012; 46 (11) e34
• 13 Welzen ME, van den Berk GE, Hamers RL, Burger DM. Interaction between antiretroviral drugs and acenocoumarol. Antivir Ther 2011; 16 (02) 249-252
• 14 Liedtke MD, Rathbun RC. Drug interactions with antiretrovirals and warfarin. Expert Opin Drug Saf 2010; 9 (02) 215-223
• 15 Aspromonte N, Colivicchi F. Clinical relevance of pharmacokinetic and pharmacodynamic properties of edoxaban when treating patients with atrial fibrillation and heart failure. Expert Opin Drug Metab Toxicol 2017; 13 (01) 113-122
• 16 Serrao A, Santoro C, Foà R, Chistolini A. Use of edoxaban for the treatment of venous thromboembolism in HIV-infected patients. HIV Med 2020; 21 (03) e7