DOAC pharmacoepidemiology
See Appendix 1, 6
The defined daily doses (DDDs) of direct oral anticoagulants established by WHO are the following: for apixaban – 10 mg (5 mg 2 times a day); rivaroxaban - 20 mg; dabigatran etexilate - 300 mg.
We calculated hypothetical dosing load for a COVID-19 patient in DDDs as recommended by the CPGs DOAC. As a result of detailed analysis of the available data in the Russian guidelines for the management of patients with COVID-19 which recommended DOAC use, we obtained the following recommended consumed doses of each DOAC to patient for a course of prophylactic treatment:
The CPG of the Ministry of Health recommended preventive use of direct oral anticoagulants in patients with COVID-19 with the following dosage regime: rivaroxaban -10 mg once daily, apixaban - 2.5 mg twice daily, dabigatran etexilate - 110 mg twice daily for prevention of deep vein thrombosis of the lower extremities and pulmonary embolism. The duration of anticoagulants use in COVID-19 out-patients was not described, however, it was recommended for up to 30-45 days depending on the patient's clinical condition and timing of recovery of motor activity. In accordance with this we calculated the recommended consumed doses of drugs in DDDs as shown in the appendix 7.
All studied CPGs were in agreement on recommending rivaroxaban, apixaban and dabigatran etexilate equally without preference for prophylaxis of deep vein thrombosis and pulmonary embolism in COVID-19 patients after discharge from hospitals in cases of concern of potential embolic complications. Out-patients with COVID-19 with persistent risk factors for deep vein thrombosis or pulmonary embolism (DVT/PE): for elderly patients, patients undergoing treatment in the intensive care unit, with active cancer, history of DVT/PE, patients with ongoing severe mobility limitation, with a D-dimer concentration in the blood 2 times the upper limit of normal and a low risk of bleeding are recommended to be treated with prophylactic doses of anticoagulants for 30 days, which can be extended to 45 days. The dabigatran dosage and mode of administration was not described in ROPNIZ -2021 (Drapkina) recommendation. In the recommendations of MGNOT-2020, NIIOZMM 2021, FAR 2020 there is information about the equal doses of rivaroxaban, apixaban and dabigatran etexilate, but there is no description of the mode of their use in patients with Covid-19. RSC- 2020 guideline does not contain information about drugs doses, indications, the dosage regimen of direct oral anticoagulants, which makes it impossible to determine the consumer dose of drugs, however, there are warnings of combination therapy with a protease, P-glycoprotein inhibitors.
The CGS of NIIOZMM-2021, MGNOT-2020, ROPNIZ-2021 and MoH-2022 contain the same information about combined prescribing anticoagulants and protease inhibitors. The use of apixaban should be avoided, or the dose (5 or 10 mg) should be reduced by 50% with simultaneous systemic therapy with P-glycoprotein and CYP450 3A4 inhibitors. Simultaneous use of dabigatran with ritonavir is not recommended in patients with impaired renal function and correction of it’s dose is necessary, monitoring of the patient for the development of bleeding. Simultaneous use of rivaroxaban with inhibitors of P-glycoprotein is contraindicated.
The lack of necessary information in the CGS of FAR-2020, NIIOZMM-2021, MGNOT-2020 does not allow us to calculate the recommended consumption of drugs from the group of DOACs.
In the recommended regimens of DOAC use in COVID-19 patients in the Russian guidelines MoH 2022; ROPNIZ 2021 (Drapkina); ROPNIZ 2021 (Livzan), we obtained equally comparable consumption rates of rivaroxaban, apixaban and dabigatran for the treatment and prevention of deep vein thrombosis and pulmonary thromboembolism in patients at high risk.
Recommended doses and calculated hypothetical DOAC dosing load are presented in
Table 4.
Gross national apixaban consumption varied through the years of 2020, 2021, and 2022 with the maximum consumption in 2021, which was nearly 2 times higher than in 2020. In 2022 national apixaban consumption decreased, but did not reach the level of 2020.
Gross national consumption of rivaroxaban demonstrated steady downward trend through 2020, 2021 and 2022. Interestingly, in 2020 rivaroxaban consumption was equal to that of apixaban. In 2022, consumption of rivaroxaban in the Russian Federation decreased by 29.25% compared to 2021 and by 30.35% compared to 2020.
Table 6.
DDDs/а case of COVID-19 of apixaban and rivaroxaban in the Russian Federation (2020 to 2022).
Table 6.
DDDs/а case of COVID-19 of apixaban and rivaroxaban in the Russian Federation (2020 to 2022).
Name of the drug |
DDD (WHO), mg |
2020 |
2021 |
2022 |
Apixaban |
10 |
26.59 |
15.75 |
10.67 |
Rivaroxaban |
20 |
26.59 |
7.87 |
5.48 |
When studying the consumption of apixaban and rivaroxaban per patient with COVID-19, it was revealed that the maximum consumption of apixaban was in 2020, the minimum in 2022. When analyzing the results of rivaroxaban consumption, similar results were obtained: the maximum consumption was in 2020, the minimum in 2022.
In 2021, apixaban consumption decreased by 40.77% compared to 2020; in 2022, the downward trend in consumption continued: DDD decreased by another 32.25% compared to 2021.
The highest consumption of rivaroxaban was observed in 2020 - 26.59 DDD per case of COVID-19. In 2021, we found a decrease in consumption by 70.40% compared to 2020. In 2022, this trend continued and the DDD per patient of rivaroxaban decreased by 30.37% compared to 2021.
Table 7.
Estimated number of people taking a CPG recommended course of apixaban and rivaroxaban in the Russian Federation in 2020-2022 based on the assumption that only COVID-19 patients were responsible for gross national consumption.
Table 7.
Estimated number of people taking a CPG recommended course of apixaban and rivaroxaban in the Russian Federation in 2020-2022 based on the assumption that only COVID-19 patients were responsible for gross national consumption.
Name of the drug |
DDD (according to Russian clinical guidelines) |
2020 |
2021 |
2022 |
Apixaban |
15 DDD (30 days) |
5 600 000 |
11 024 000 |
7 600 000 |
22,5DDD (45 days) |
3 733 333 |
7 349 333 |
5 066 667 |
Rivaroxaban |
15 DDD (30 days) |
5 600 000 |
5 512 000 |
3 900 000 |
22,5 DDD (45 days) |
3 733 333 |
3 674 667 |
2 600 000 |
The hypothetical number of people taking the CPG-recommended course of apixaban in 2020 (if we consider a 30-day course of therapy) exceeds the number of officially registered Covid patients by 1.77 times. If we consider a 45-day course of therapy, the results practically coincide with the estimated number of Covid cases in 2020 of 3 733 333 patients.
In 2021, the hypothetical number of people taking the CPG-recommended course of apixaban was 11 024 000, which was by 4.75% more than the number of COVID-19 patients (if we consider a 30-day course of therapy). If we consider a 45-day course of therapy, the hypothetical number of people taking the CPG-recommended course of apixaban in 2021 was by 30.0% less than the number of officially registered 10 499 982 cases.
In 2022, 71.13% of patients with coronavirus infection received at least one 30-day course of apixaban therapy; If we consider a 45-day course, then 47.42% of patients received at least one course of apixaban.
The hypothetical number of people taking the CPG-recommended course of rivaroxaban in 2020 (if we consider a 30-day course of therapy) exceeds the number of COVID-19 patients by 43.58%; If we consider a 45-day course of therapy, the results obtained practically coincide with the number of COVID-19 cases in 2020 and amount to 3 733 333.
In 2021, the hypothetical number of people taking the CPG-recommended course of rivaroxaban was 5 512 000, which is 47.50% less than the number of COVID-19 patients (if we consider a 30-day course of therapy). If we consider a 45-day course of therapy, then the hypothetical number of people taking the CPG-recommended course of rivaroxaban in 2021 was 65.00% less than the number of cases.
In 2022, there was a decrease in rivaroxaban consumption compared to 2021: 36.5% of COVID-19 patients received rivaroxaban in 2022 (30-day course) or 24.3% if we consider a 45-day course of therapy.
The largest number of COVID-19 patients was registered in 2021-2022. The smallest number of patients was registered in 2020. At the same time, the highest consumption of apixaban was in 2021, the lowest - in 2020, and the largest amount of rivaroxaban was consumed in the Russian Federation in 2020, subsequently its consumption decreased and the lowest consumption was noted in 2022.
The above presented calculations mean that in 2020, each patient with covid received more than 1 course of therapy with apixaban (1.77 course of apixaban therapy (30-day)) and rivaroxaban (1.77 course of rivaroxaban therapy (30-day)), which indicates serious overuse of DOACs at the beginning of the pandemic.
In 2021, almost every patient with COVID-19 received apixaban, and every second patient was prescribed rivaroxaban.
In 2022, DOAC consumption decreased, every third patient took rivaroxaban, and about ¾ of all patients took apixaban (30-day course).
This shows a more rational and targeted use of doacs in patients with covid in the 3rd year of the pandemic, which may be the result of both, the accumulated experience and the emergence of clear algorithms for the management of patients with coronavirus infection (the CPGs).
Table 8.
Number of people taking apixaban and rivaroxaban in the Russian Federation in 2020-2022 (15 DDD, 30 days) and (22,5 DDD, 45 days).
Table 8.
Number of people taking apixaban and rivaroxaban in the Russian Federation in 2020-2022 (15 DDD, 30 days) and (22,5 DDD, 45 days).
|
2020 |
2021 |
2022 |
COVID-19 patients |
3 159 297 |
10 499 982 |
10 684 204 |
30 days |
Total consumption of Rivaroxaban and Apixaban (patients) |
11 200 000 |
16 536 000 |
11 500 000 |
45 days |
Total consumption of Rivaroxaban and Apixaban (patients) |
7 466 666 |
11 023 999 |
7 666 667 |
Results of CPGs assessment domain scores
The domain scores differed between CPGs. The domains Scope and Purpose, and Clarity of Presentation consistently received high scores across all seven CPGs. The domains Applicability, Editorial Independence and Rigour of Development received lower scores.
The domain Scope and Purpose deals with the main objective of CPGs, the health question and the population to whom the guideline is meant to apply. This domain includes 3 items (items 1–3). The range of scores for Scope and Purpose was 67%–100% (See Appendix 8 - Section assessment results SCOPE AND PURPOSE)
Most CPGs received high scores for this domain (89%–100%), while one CPG received lower score – 67 %. The CPG by the FAR (Zabolotskikh I.B., et al) obtained the lowest score for this domain. We attributed it to the guidelines having a vague and not clearly defined objective. The CPGs by ROPNIZ (Livzan M.A. et al), MGNOT (Vorobyev P.A., et al) score for Scope and Purposes. The population was clearly described and the purpose was well defined and worded here.
The domain Stakeholder Involvement is focused on assessing whether all relevant clinical professionals participated in the development of the CPG and whether the target audience who would use the CPG is specified. This domain includes three items (items 3–6). The range of scores was 31%–67% (See Appendix 8 - Section assessment results STAKEHOLDER INVOLVEMENT). The CPG by NIIOZMM (Khripun A.I., et al) received the lowest score (31%) and the CPG by ROPNIZ (Livzan M.A., et al) received the highest score – 67%.
We assigned relatively low score (31%) to the NIIOZMM (Khripun A.I., et al) for the domain Stakeholder Involvement as it did not include any specifications about the target audience. The experts pointed out that epidemiologists and medical statisticians should have been included in the development of CPG of NIIOZMM (Khripun A.I., et al)
The ROPNIZ CPG (Livzan M.A., et al) received the highest score (67%) for the domain Stakeholder Involvement (See Appendix 8 - Section assessment results STAKEHOLDER INVOLVEMENT). The variety of specialists were involved in the development of these guidelines, which was better than in all other identified CPGs. A similar score (65%) for this domain received in the ROPNIZ CPGs (Drapkina O.M., et al) and MGNOT (Vorobyev P.A., et al) (See Appendix 8 - Section assessment results STAKEHOLDER INVOLVEMENT). These three CPGs received similar comments.
The domain Rigour of Development deals with the actual methods used to compose the CPGs. It is the largest domain in the AGREE II and consists of eight items (items 7–14). The range of scores for this domain was 0%–49% (See Appendix 8 - Section assessment results RIGOUR OF DEVELOPMENT). Rigour and Development had the largest variation between the highest and the lowest scores, as it is the biggest and arguably the most important domain in the AGREE II. The scores for all CPGs varied.
The lowest score for the domain Rigour and Development we attributed to the ROPNIZ (Livzan M.A., et al) CPG by (See Appendix 8 - Section assessment results RIGOUR OF DEVELOPMENT).
This CPG lacked a procedure of evidence search, strengths and limitations of this evidence and most other items of this domain. The only item in the Rigour and Development that received a score above 1 from all three author-experts was the item 14, which deals with a procedure for updating the guideline, which is described in the guideline.
The highest score (49%) for the domain Rigour and Development was attributed to the ROPNIZ (Drapkina O.M., et al) CPG (Table_. Section assessment results RIGOUR OF DEVELOPMENT). Even though this CPG had a high score, there were comments for this domain: the ROPNIZ (Drapkina O.M., et al) CPG did not include the methods for formulating the recommendations and aren’nt clearly described, opinion of external experts and the update procedure is not described. It did not include a method for reviewing the strengths and limitations of the evidence (such as GRADE), or a Delphi method for formulating the recommendations.
The fourth domain Clarity of Presentation consists of three items (items 15–17). Clarity of presentation is focused on the way the CPG is written, specifically its language, format, and structure. This domain consistently received a high score in each of the CPGs (96%–100%) (See Appendix 8 - Section assessment results CLARITY OF PRESENTATION). All CPGs scored high (96-100%). The comments stated that the recommendations are clear and unambiguous (See Appendix 8 - Section assessment results CLARITY OF PRESENTATION).
The domain Applicability includes four items (items 18–21). Applicability mostly assesses the way that the CPG can be used in the society, including crucial factors such as assessment of cost, barriers and facilitators of implementation, as well as possible tactics for uptake of the recommendations. Overall, Applicability received a low score from the experts throughout all studied CPGs (24%–50%) (See Appendix 8 - Section assessment results APPLICABILITY).
The lowest score (24%) for the Applicability domain we attributed to the MGNOT (Vorobyev P.A., et al). Еhere was not enough information about the implementation factors in the CPG. On the other hand, the ROPNIZ (Drapkina O.M., et al) and its assumption received the highest score for the domain Applicability (See Appendix 8 -Section assessment results APPLICABILITY).
The last domain in the AGREE II is called Editorial Independence and it consists of two items (items 22–23). This domain score reflects whether the authors of a CPG could have had any conflict of interest in the matter of creating the CPG. This domain received the lowest ratings from all experts. The range of the scores was 0%–50%.
The lowest score of 0% was obtained for the ROPNIZ (Drapkina O.M., et al) CPG (See Appendix 8 - Section assessment results EDITORIAL INDEPENDENCE). because no information about the potential conflict of interest of the CPG authors was provided.
The highest score was received in the CPG for the diagnosis and treatment of circulatory diseases in the context of the COVID-19 pandemic (both received 50%; See Appendix 8 -Section assessment results EDITORIAL INDEPENDENCE).