If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Risk of Major Bleeding in Patients With Atrial Fibrillation Taking Dronedarone in Combination With a Direct Acting Oral Anticoagulant (From a U.S. Claims Database)
Dronedarone may increase exposure and the risk of major bleeding when prescribed with a direct oral anticoagulant (DOAC). This retrospective cohort study examined the risk of the first occurrence of major bleeding (hospitalization or emergency room visit for gastrointestinal [GI] bleeding, intracranial hemorrhage [ICH], or bleeding at other sites) among new users of apixaban, dabigatran, and rivaroxaban in patients with AF ≥18 years (January 1, 2007 to September 30, 2017) from the United States Truven Health MarketScan claims, comparing concomitant users of dronedarone to DOAC alone users in patients with atrial fibrillation (AF). No increased risk of major bleeding was associated with use of dronedarone and apixaban (adjusted Hazard Ratio [aHR]: 0.69 [95% confidence interval [CI]: 0.40, 1.17], p = 0.16), a modestly increased risk of GI bleeding but not overall bleeding was associated with use of dronedarone and dabigatran (aHR bleeding: 1.18 [95% CI: 0.89, 1.56], p = 0.26; aHR GI bleeding: 1.40 [95% CI: 1.01, 1.93]; p = 0.04) and an increased risk of overall bleeding, driven by GI bleeding, was associated with use of dronedarone and rivaroxaban (aHR bleeding: 1.31 [95% CI: 1.01, 1.69]; p = 0.04; aHR GI bleeding: 1.39 [95% CI: 0.98, 1.95]; p = 0.06), compared to each DOAC respectively. There was no increased risk of ICH associated with combined use of dronedarone and any DOAC. Prospective analyses, preferably randomized controlled studies, are needed to further explore the risk of major bleeding with concomitant use of DOACs and CYP3A4/P-gp inhibitors such as dronedarone.
All direct oral anticoagulants (DOAC) are P-glycoprotein substrates, and apixaban and rivaroxaban in addition are metabolized by cytochrome P450 3A4 (CYP3A4).
Dronedarone, an antiarrhythmic drug with properties of Vaughn-Williams classes I-IV, primarily metabolized by the CYP P450 pathway and excreted 6% by renal route,
is an inhibitor of P-glycoprotein and a moderate inhibitor of CYP3A4. Trough plasma concentration of dabigatran was increased by 1.7-fold in a pharmacokinetic study of 150 mg dabigatran and dronedarone, however when the both the drugs were administered individually in an interval of 2 hours, the maximum dabigatran concentration (Cmax) was increased by 1.3 fold.
Drug interaction study with dabigatran etexilate and dronedarone in healthy subjects. ClinicalTrials.gov Identifier: NCT01306162. Accessed December 24, 2020. https://clinicaltrials.gov/ct2/show/NCT01306162.
Although dronedarone may increase the exposure to the DOACs via inhibition of metabolism and/or excretion, potentially leading to an increased risk of bleeding, observational data to date
Association between use of non-vitamin K oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.
Association between use of non-vitamin K oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.
This study examined the association between concomitant use of dronedarone and DOAC and the risk of major bleeding in patients with AF using a United States (US) claims database.
Methods
The study was a retrospective cohort study conducted using the Commercial Claims and the Medicare Supplemental databases of the Truven Health MarketScan claims databases for the years 2007-2017. All new users of a DOAC (apixaban, dabigatran, rivaroxaban) with a diagnosis of AF during the study period from January 1, 2007 to September 30, 2017 were identified. The analysis did not include edoxaban users due to a small sample size. The date of the first dispensing of the DOAC during the study period was defined as the index date. Patients aged ≥18 years, with continuous enrollment ≥365 days prior to the index date, with a dispensing of a DOAC, and with at least two International Classification of Diseases (ICD)-9 or ICD-10 codes for AF
Association between use of non-vitamin K oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.
(Supplemental eTable 1) between January 1, 2007 to the day preceding the index date (baseline period) were included in the study. The earliest index date corresponded to the dates of approval for each DOAC: October 19, 2010 for dabigatran, November 4, 2011 for rivaroxaban and December 28, 2012 for apixaban. Patients were excluded if they had ICD-9 or ICD-10 code for mitral stenosis, prosthetic heart valves or, hyperthyroidism or thyrotoxicity in the baseline period (Supplemental eTable 1), or they had a dispensing of warfarin within 7 days prior to or on the index date. Patients with a concomitant exposure to dronedarone on the index date were identified as concomitant users and remaining patients (i.e., DOAC alone users) were included in the comparator cohorts.
The index treatment episode for the concomitant users was defined from the index date ending 30 days following the end of supply for DOAC or dronedarone in case of concomitant users and at 30 days after the end of day's supply for the DOAC in case of comparator cohorts. In the concomitant users, if patients switched to another DOAC, warfarin or an anti-antiarrhythmic agent other than dronedarone, the episode ended on the date immediately prior to the dispensing date of the new ‘drug’. In the comparator cohorts, if there was a dispensing of dronedarone, another DOAC, or warfarin after the index date, the episode ended on the date immediately prior to such dispensing. The follow-up period was defined as the time from the index date until the first occurrence of bleeding or end of index treatment episode, enrollment, or study period (September 30, 2017), whichever occurred first.
The outcomes of interest included the first occurrence of overall bleeding and of each type of bleeding during the follow-up period, herein referred to as “major bleeding”, defined as a hospitalization or an emergency room (ER) visit with a primary diagnosis of gastrointestinal (GI) bleeding, intracranial hemorrhage (ICH), or bleeding at other sites (non-GI, non-ICH) identified using ICD-9 or ICD-10 diagnosis codes (Supplemental eTable 2).
Association between use of non-vitamin K oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.
At baseline the following patient demographics were collected: age on the index date, sex, geographic region, and index year. Mean Charlson comorbidity index (CCI) score calculated using the CCI calculator,
comorbidities (Supplemental eTable 3), history of procedures, ie, percutaneous coronary intervention (PCI), coronary artery bypass surgery (CABG) or bleeding prior to the index date, and concomitant medications in the 183-day period prior to and including the index identified using National Drug Codes (Supplemental eTable 4) were the covariates of interest. In each concomitant user and comparator cohort pair, a propensity score (PS) was calculated using logistic regression (all covariates of interest except the CHA2DS2-VASc score were the independent variables). PS trimming was applied excluding of the top and bottom fifth percentile in the distribution of PS in the cohorts.
Standardized differences were calculated before and after PS trimming in both cohorts. A cut-off of 0.1 was used to compare if the means and prevalence of covariates were balanced in both cohorts before and after PS trimming.
Crude incidence rates (IRs) and IRs stratified by PS quintiles per 1,000 person-years (PY) with 95% confidence intervals (CIs) of overall and each type of bleeding were calculated as the ratio of the number of bleeding events divided by the person-time at risk defined as the follow-up period. The incidence rate ratios (IRRs) for each PS quintile and a summary Mantel-Haenszel (MH) IRR with 95% CI comparing the IR of overall and each type of bleeding were calculated. Crude hazard ratios (HR) and adjusted HR (aHR) adjusting for the above-mentioned covariates and PS quintiles comparing the risk of overall and each type of bleeding in the concomitant users compared to the comparator cohort were estimated using Cox proportional hazards modeling. All data management and statistical analyses was conducted using SAS Version 9.4 (Cary, NC).
Results
After applying PS trimming, patients were identified in the concomitant user cohorts for apixaban, dabigatran, and rivaroxaban and comparator cohorts, respectively (Figure 1). Concomitant user cohorts were generally younger than comparator cohorts (Table 1). The most common comorbidities, hypertension chronic pulmonary disease, congestive heart failure, diabetes mellitus and cerebrovascular disease were generally less common in concomitant user cohorts and the CHA2DS2-VASc scores were relatively higher in the comparator cohort compared to the concomitant user cohorts. However, standardized differences after PS trimming showed that the means and prevalence of covariates of interest were very similar in both cohorts of all DOACs (data not shown).
Figure 1Flow diagram of cohort selection for each DOAC of interest.
Table 1Baseline patient characteristics of patients from the concomitant user and comparator cohorts of the DOAC of interest identified from the Truven MarketScan database during the study period from January 1, 2007 to September 30, 2017
Patient characteristic
Apixaban
Dabigatran
Rivaroxaban
Concomitant user cohort
Comparator cohort
Concomitant user cohort
Comparator cohort
Concomitant user cohort
Comparator cohort
(n = 1,932)
(n = 51,420)
(n = 3,117)
(n = 42,312)
(n = 2,395)
(n = 57,300)
Mean age on the index date in years (± SD)
69.2 (± 11.3)
72.0 (± 12.0)
68.1 (± 10.5)
70.8 (± 11.3)
67.9 (± 10.9)
70.0 (± 12.1)
Age groups (in years)
18–19
0
0
0
2
0
3
20–29
0
26 (0.1%)
0
15 (0.0%)
2 (0.1%)
83 (0.1%)
30–39
7 (0.4%)
200 (0.4%)
5 (0.2%)
163 (0.4%)
15 (0.6%)
404 (0.7%)
40-49
62 (3.2%)
1,246 (2.4%)
91 (2.9%)
1,036 (2.4%)
62 (2.6%)
1,970 (3.4%)
50–59
313 (16.2%)
6,605 (12.8%)
562 (18.0%)
5,718 (13.5%)
453 (18.9%)
8,873 (15.5%)
60–69
638 (33.0%)
13,703 (26.6%)
1,079 (34.6%)
12,136 (28.7%)
821 (34.3%)
16,272 (28.4%)
70–79
498 (25.8%)
13,420 (26.1%)
861 (27.6%)
12,294 (29.1%)
641 (26.8%)
15,159 (26.5%)
≥ 80
414 (21.4%)
16,220 (31.5%)
519 (16.7%)
10,948 (25.9%)
401 (16.7%)
14,536 (25.4%)
Sex
Men
1,103 (57.1%)
29,892 (58.1%)
1,946 (62.4%)
26,335 (62.2%)
1,437 (60.0%)
34,870 (60.9%)
Women
829 (42.9%)
21,528 (41.9%)
1,171 (37.6%)
15,977 (37.8%)
958 (40.0%)
22430 (39.1%)
Geographic region
Northeast
377 (19.5%)
10,435 (20.3%)
430 (13.8%)
7,422 (17.5%)
377 (15.7%)
10,780 (18.8%)
Midwest
465 (24.1%)
14,797 (28.8%)
838 (26.9%)
11,547 (27.3%)
666 (27.8%)
16,897 (29.5%)
South
865 (44.8%)
19,710 (38.3%)
1,271 (40.8%)
15,405 (36.4%)
963 (40.2%)
20,366 (35.5%)
West
216 (11.2%)
6,255 (12.2%)
557 (17.9%)
7,674 (18.1%)
371 (15.5%)
8,784 (15.3%)
Unknown
9 (0.5%)
223 (0.4%)
21 (0.7%)
264 (0.6%)
18 (0.8%)
473 (0.8%)
Index year
2010
0
0
210 (6.7%)
2,100 (5.0%)
0
0
2011
0
0
1,914 (61.4%)
23,191 (54.8%)
18 (0.8%)
224 (0.4%)
2012
0
0
620 (19.9%)
9,257 (21.9%)
535 (22.3%)
7,857 (13.7%)
2013
221 (11.4%)
4,216 (8.2%)
193 (6.2%)
3,674 (8.7%)
715 (29.9%)
14,867 (25.9%)
2014
438 (22.7%)
9,875 (19.2%)
89 (2.9%)
1,807 (4.3%)
562 (23.5%)
14,937 (26.1%)
2015
486 (25.2%)
12,490 (24.3%)
32 (1.0%)
793 (1.9%)
257 (10.7%)
8,145 (14.2%)
2016
505 (26.1%)
15,072 (29.3%)
49 (1.6%)
1,220 (2.9%)
196 (8.2%)
6,855 (12.0%)
2017
282 (14.6%)
9,767 (19.0%)
10 (0.3%)
270 (0.6%)
112 (4.7%)
4,415 (7.7%)
Mean CCI score (± SD)
3.4 (± 3.0)
4.1 (± 3.3)
2.7 (± 2.5)
3.0 (± 2.7)
2.9 (± 2.7)
3.4 (± 3.0)
Mean CHA2DS2-VASc score (± SD)
4.0 (± 2.2)
4.5 (± 2.2)
3.6 (± 2.0)
4.0 (± 2.1)
3.7 (± 2.1)
4.0 (± 2.2)
Comorbidities
Hypertension
1,723 (89.2%)
46,832 (91.1%)
2,651 (85.0%)
36,128 (85.4%)
2083 (87.0%)
50,233 (87.7%)
Diabetes mellitus
662 (34.3%)
20,155 (39.2%)
975 (31.3%)
14,630 (34.6%)
757 (31.6%)
20,569 (35.9%)
Myocardial infarction
272 (14.1%)
9,287 (18.1%)
346 (11.1%)
4,692 (11.1%)
308 (12.9%)
8,096 (14.1%)
Congestive heart failure
662 (34.3%)
22,880 (44.5%)
1063 (34.1%)
16,134 (38.1%)
778 (32.5%)
21,771 (38.0%)
Peripheral vascular disease
570 (29.5%)
18,936 (36.8%)
737 (23.6%)
11,002 (26.0%)
632 (26.4%)
17,149 (29.9%)
Cerebrovascular disease
682 (35.3%)
20,098 (39.1%)
832 (26.7%)
13,453 (31.8%)
714 (29.8%)
19,149 (33.4%)
Ischemic stroke
552 (28.6%)
16,549 (32.2%)
677 (21.7%)
10,986 (26.0%)
572 (23.9%)
15,574 (27.2%)
Transient ischemic attack
216 (11.2%)
7,111 (13.8%)
268 (8.6%)
4,464 (10.6%)
231 (9.6%)
6,477 (11.3%)
Chronic pulmonary disease
813 (42.1%)
23,091 (44.9%)
1,095 (35.1%)
15,046 (35.6%)
938 (39.2%)
23,465 (41.0%)
Anemia
690 (35.7%)
19,763 (38.4%)
736 (23.6%)
10,757 (25.4%)
673 (28.1%)
18,515 (32.3%)
Gastrointestinal ulcer disease
81 (4.2%)
2,621 (5.1%)
88 (2.8%)
1,189 (2.8%)
83 (3.5%)
2,187 (3.8%)
Coagulation defects
110 (5.7%)
3,841 (7.5%)
230 (7.4%)
3,514 (8.3%)
146 (6.1%)
4,267 (7.4%)
Chronic kidney disease (Stage)
333 (17.2%)
10,886 (21.2%)
300 (9.6%)
4,623 (10.9%)
280 (11.7%)
7,843 (13.7%)
I
14 (0.7%)
646 (1.3%)
21 (0.7%)
255 (0.6%)
14 (0.6%)
466 (0.8%)
II
67 (3.5%)
2,028 (3.9%)
55 (1.8%)
807 (1.9%)
55 (2.3%)
1,408 (2.5%)
III
231 (12.0%)
7,338 (14.3%)
172 (5.5%)
2,792 (6.6%)
180 (7.5%)
4,999 (8.7%)
IV
57 (3.0%)
2,125 (4.1%)
41 (1.3%)
691 (1.6%)
33 (1.4%)
975 (1.7%)
V
9 (0.5%)
571 (1.1%)
1 (0.0%)
95 (0.2%)
4 (0.2%)
135 (0.2%)
End-stage renal disease
21 (1.1%)
1,010 (2.0%)
10 (0.3%)
254 (0.6%)
12 (0.5%)
376 (0.7%)
Moderate to severe liver disease
7 (0.4%)
417 (0.8%)
12 (0.4%)
182 (0.4%)
5 (0.2%)
309 (0.5%)
Any malignancy, except malignant neoplasm of skin
394 (20.4%)
11,387 (22.1%)
532 (17.1%)
7,622 (18.0%)
434 (18.1%)
11,605 (20.3%)
History of recent procedures
Percutaneous coronary intervention
93 (4.8%)
2,458 (4.8%)
125 (4.0%)
1,520 (3.6%)
110 (4.6%)
2,319 (4.0%)
Coronary artery bypass surgery
45 (2.3%)
1,369 (2.7%)
63 (2.0%)
748 (1.8%)
35 (1.5%)
1,296 (2.3%)
History of bleeding any time prior to the index date
Intracranial
13 (0.7%)
462 (0.9%)
12 (0.4%)
243 (0.6%)
7 (0.3%)
356 (0.6%)
Gastrointestinal
79 (4.1%)
2,479 (4.8%)
62 (2.0%)
1,131 (2.7%)
63 (2.6%)
1,736 (3.0%)
Bleeding at other sites
62 (3.2%)
2,105 (4.1%)
78 (2.5%)
1,304 (3.1%)
57 (2.4%)
1,912 (3.3%)
Concomitant medications in the 183-day period prior to the index date
For DOAC, the use in the 183-day period prior to the index date, excluding the index date was identified. ACE = Angiotensin-converting enzyme; ARB = Angiotensin-receptor blockers; CCB = Calcium channel blocker; CCI = Charlson comorbidity index; DOAC = Direct oral anticoagulants; NSAID = Nonsteroidal anti-inflammatory drug; PPI = Proton pump inhibitor; SD = standard deviation; SNRI = Serotonin norepinephrine reuptake inhibitor; SSRI = Selective serotonin reuptake inhibitor.
356 (18.4%)
7,386 (14.4%)
57 (1.8%)
1,121 (2.6%)
371 (15.5%)
7,236 (12.6%)
Antiplatelet agents
198 (10.2%)
5,784 (11.2%)
327 (10.5%)
3,926 (9.3%)
239 (10.0%)
5,424 (9.5%)
Heparin
31 (1.6%)
998 (1.9%)
115 (3.7%)
1,177 (2.8%)
61 (2.5%)
1,724 (3.0%)
Non-heparin anticoagulants
0
26 (0.1%)
3 (0.1%)
50 (0.1%)
1 (0.0%)
67 (0.1%)
PPIs
567 (29.3%)
14,780 (28.7%)
736 (23.6%)
9,401 (22.2%)
661 (27.6%)
14,732 (25.7%)
Glucocorticoids
460 (23.8%)
12,391 (24.1%)
629 (20.2%)
8,779 (20.7%)
563 (23.5%)
13,121 (22.9%)
NSAIDs
243 (12.6%)
5,972 (11.6%)
339 (10.9%)
4,187 (9.9%)
285 (11.9%)
7149 (12.5%)
Lipid lowering agents
1,065 (55.1%)
29,544 (57.5%)
1,758 (56.4%)
23,957 (56.6%)
1,272 (53.1%)
30,510 (53.2%)
Insulin and other glucose-lowering agents
342 (17.7%)
10,570 (20.6%)
518 (16.6%)
8,161 (19.3%)
377 (15.7%)
10,598 (18.5%)
SSRI
221 (11.4%)
6,760 (13.1%)
350 (11.2%)
5,102 (12.1%)
296 (12.4%)
7,308 (12.8%)
SNRI
92 (4.8%)
2,368 (4.6%)
139 (4.5%)
1,562 (3.7%)
101 (4.2%)
2,472 (4.3%)
Beta blockers
1,096 (56.7%)
30,240 (58.8%)
1,695 (54.4%)
23,063 (54.5%)
1,321 (55.2%)
32,318 (56.4%)
ARBs
467 (24.2%)
12,422 (24.2%)
704 (22.6%)
9,425 (22.3%)
534 (22.3%)
12,618 (22.0%)
ACE inhibitors
576 (29.8%)
16,972 (33.0%)
994 (31.9%)
14,866 (35.1%)
693 (28.9%)
18,411 (32.1%)
CCBs
688 (35.6%)
20,270 (39.4%)
1,084 (34.8%)
15,500 (36.6%)
829 (34.6%)
21,017 (36.7%)
Digoxin
144 (7.5%)
5,352 (10.4%)
436 (14.0%)
7,923 (18.7%)
236 (9.9%)
7,070 (12.3%)
Anti-fungal agents
63 (3.3%)
1,910 (3.7%)
100 (3.2%)
1,307 (3.1%)
74 (3.1%)
1,937 (3.4%)
Antibiotics
30 (1.6%)
723 (1.4%)
51 (1.6%)
678 (1.6%)
26 (1.1%)
768 (1.3%)
Rifampicin
2 (0.1%)
41 (0.1%)
2 (0.1%)
41 (0.1%)
0
61 (0.1%)
Ritonavir
0
10 (0.0%)
0
16 (0.0%)
0
13 (0.0%)
Cyclosporin
24 (1.2%)
572 (1.1%)
37 (1.2%)
372 (0.9%)
22 (0.9%)
507 (0.9%)
Mean follow-up period in Days ± SD
154.3(± 201.5)
267(± 274.9)
161.4 (± 219.5)
320.3 (± 369.5)
148.1(± 207.4)
283.4(± 323.7)
For DOAC, the use in the 183-day period prior to the index date, excluding the index date was identified.ACE = Angiotensin-converting enzyme; ARB = Angiotensin-receptor blockers; CCB = Calcium channel blocker; CCI = Charlson comorbidity index; DOAC = Direct oral anticoagulants; NSAID = Nonsteroidal anti-inflammatory drug; PPI = Proton pump inhibitor; SD = standard deviation; SNRI = Serotonin norepinephrine reuptake inhibitor; SSRI = Selective serotonin reuptake inhibitor.
Gastrointestinal bleeding was the most commonly occurring major bleeding event, followed by bleeding at other sites; ICH was the least frequently occurring event in both cohorts of all DOACs (Supplemental eTable 5). The MH summary IRR of dabigatran and rivaroxaban indicated a higher incidence of major bleeding in those taking dronedarone and dabigatran or rivaroxaban concomitantly compared to those taking dabigatran or rivaroxaban alone, whereas no increased incidence of major bleeding was observed in the concomitant users of dronedarone and apixaban, (Table 2).
Table 2Number of bleeding events, crude incidence rates and incidence rates stratified by propensity score per 1,000 person-years of bleeding in the cohorts of interest
There was no increased incidence of ICH associated with the concomitant use of dronedarone and any of the DOACs (Supplemental eTable 6). There was an increased risk of GI bleeding associated with the concomitant use of dronedarone and dabigatran (IRR of dabigatran: 1.56 [95% CI: 1.14, 2.14], and rivaroxaban (IRR of rivaroxaban: 1.55 [95% CI: 1.11, 2.17]) compared to users of dabigatran or rivaroxaban, respectively (Supplemental eTable 7), and an increased incidence of bleeding at other sites (non-GI, non-ICH) associated with the concomitant use of dronedarone and rivaroxaban (IRR: 1.51 [95% CI: 1.04, 2.21]) (Supplemental eTable 8).
There was no increased risk of ICH or bleeding at other sites (non-GI, non-ICH) associated with the concomitant use of dronedarone and any of the DOACs (Table 3). Additionally, no increased risk of overall bleeding was seen with the concomitant use of dronedarone and apixaban compared to apixaban alone users (aHR: 0.69 [95% CI: 0.40, 1.17; p = 0.16]). Although there was no increased risk of overall bleeding, a statistically significantly increased risk of GI bleeding (aHR GI bleeding: 1.40 [95% CI: 1.01, 1.93; p = 0.04]) was associated with the concomitant use of dronedarone and dabigatran compared to those taking dabigatran alone. An increased risk of overall bleeding, driven by GI bleeding, was associated with the concomitant use of dronedarone and rivaroxaban compared to those taking rivaroxaban alone (aHR overall bleeding: 1.31 [95% CI: 1.01, 1.69; p = 0.04]; aHR GI bleeding: 1.39 [95% CI: 0.98, 1.95; p = 0.06]).
Table 3Crude and adjusted hazard ratios examining the risk of overall bleeding and each type of bleeding among concomitant users of dronedarone and DOAC of interest compared to the comparator cohort identified from the database during the study period
Models adjusted for all available covariates of interest (i.e., age, sex, geographic region, index year, mean CCI score, comorbidities, concomitant medications in the 183-days prior to the index date, history of procedures [PCI or CABG], and history of bleeding prior to the index date) and PS quintiles.
Bleeding at other sites excluded GI bleeding and ICH. AHR = adjusted hazard ratio; CABG = coronary artery bypass surgery; CI = confidence interval; DOAC = direct oral anticoagulants; GI = gastrointestinal; HR = hazard ratio; ICH = Intracranial hemorrhage; PCI = percutaneous coronary intervention; PS = propensity score.
1.20 (0.82, 1.76)
1.27 (0.86, 1.87)
0.23
After PS trimming.
† Models adjusted for all available covariates of interest (i.e., age, sex, geographic region, index year, mean CCI score, comorbidities, concomitant medications in the 183-days prior to the index date, history of procedures [PCI or CABG], and history of bleeding prior to the index date) and PS quintiles.
‡ Bleeding at other sites excluded GI bleeding and ICH. AHR = adjusted hazard ratio; CABG = coronary artery bypass surgery; CI = confidence interval; DOAC = direct oral anticoagulants; GI = gastrointestinal; HR = hazard ratio; ICH = Intracranial hemorrhage; PCI = percutaneous coronary intervention; PS = propensity score.
Sensitivity analyses were conducted to examine the impact on the association (Supplemental eTable 9). The results of all sensitivity analyses were comparable to the results from the main analysis and remained consistent across all sensitivity analyses. Of note, when a cut-off was used after the first 183 days, the data suggested a trend towards an increased overall bleeding risk in the first 6 months of the concomitant use of dronedarone and dabigatran compared to the use of dabigatran alone (aHR: 1.26; [95% CI: 0.91, 1.75; p = 0.16]) and a statistically significantly increased risk of overall bleeding observed in the first 6 months of the concomitant use of dronedarone and rivaroxaban compared to the use of rivaroxaban alone (aHR: 1.42 [95% CI: 1.06, 1.91; p = 0.02]). However, no such trend was observed in the period starting after Day 183 for either of these DOACs.
Discussion
This study observed that there was no increased risk of ICH or bleeding at other sites (non-GI, non-ICH) associated with the concomitant use of dronedarone and any of the DOACs. While there was no increased risk of major bleeding associated with the concomitant use of dronedarone and apixaban, a modestly increased risk of GI bleeding was associated with the concomitant use of dronedarone and dabigatran; and an increased risk of overall bleeding, driven by GI bleeding, was associated with the concomitant use of dronedarone and rivaroxaban compared to those taking these DOACs alone.
In previous trials, no association has been reported between increased exposure of the individual DOACs and the risk of ICH.
Data from the present study is in line with these findings and is reassuring as ICH is the most feared complication of anticoagulation. The increased risk of GI bleeding with dabigatran and rivaroxaban seen in this analysis are also consistent with previous reports where use of rivaroxaban once daily and dabigatran 150 mg twice daily (bid) have been associated with an increased risk of GI bleeding in meta-analysis, whereas apixaban and dabigatran 110 mg bid have not, indicating that DOAC type, and exposure (dose, dosing) contribute to bleeding risk.
Dronedarone is an inhibitor of P-glycoprotein and a moderate inhibitor of CYP3A4 and while the findings in this study is likely due in part to the differences in exposure of the DOACs upon concomitant use with dronedarone, the once daily dosing of rivaroxaban (half-life of rivaroxaban 5-9 hours, vs a half-life of 12 hours of apixaban bid) could also contribute to increased exposure with higher systemic peak concentrations. The findings from the present study differed from previous studies that reported no risk of increased bleeding in the concomitant users of dronedarone and DOACs.
Association between use of non-vitamin K oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.
Use of doses outside the approved doses of 20/15 mg for rivaroxaban for AF in the US could explain some of differences seen, as two of these studies were conducted in Taiwanese populations.
Association between use of non-vitamin K oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.
No studies, including the present, could fully assess the effect of dose of each DOAC while concomitantly taking dronedarone, on the risk of bleeding, and data cannot be compared between studies.
To control for confounding, Cox models were adjusted for the presence of chronic kidney disease (CKD) and moderate/severe liver disease as these patients may have a higher risk of bleeding and drug-drug interactions, and the overall findings were consistent across all the sensitivity analyses that were conducted during this study, even after excluding patients with CKD. Sensitivity analysis also showed that the risk of GI-bleeding was increased in the first 6 months after the initiation of dronedarone and rivaroxaban, with a similar trend was observed among users of dronedarone and dabigatran; but not with either drug after the first 6 months. The findings on the early risk of GI-bleeding are well known from previous reports where GI bleeding events occur early
Non-vitamin K oral anticoagulants are non-inferior for stroke prevention but cause fewer major bleedings than well-managed warfarin: a retrospective register study.
In summary, this retrospective observational study reported that no increased risk of ICH or bleeding at other sites (non-GI, non-ICH) was associated with the concomitant use of dronedarone and any of the DOACs, and no increased risk of overall bleeding was associated with the concomitant use of dronedarone and apixaban. However, use of dronedarone was associated with a risk of GI-bleeding with concomitant treatment with rivaroxaban or dabigatran, with an increased risk in the first 6 months after initiating treatment; but no increased risk after the first 6 months. While not desirable, the bleeding events at extracranial sites have seldom been reported as fatal; hence the potential impact of these events on patients is typically less than a stroke-related death or disability and can also unveil an early sign of a malignancy, paradoxically leading to early surgery and cure.
Analysis of these observational claims data carry several limitations. 1. Use of ICD codes may have resulted in misclassification bias, although likely non-differential across both cohorts. 2. There were no data on the severity of the bleeding event, although it is reasonable to assume that an ICH event is potentially life-threatening. Bleeding events captured during hospitalization and ER visits, could range from severe GI bleeding, requiring transfusion and surgical intervention, to blood in feces with no decrease in hemoglobin, due to hemorrhoids. 3. There is no information on the timing of drug intake for DOACs and dronedarone, important as exposure of dabigatran is lower when taken two hours apart. 4. The use of drugs available over the counter (NSAIDs and aspirin) may have been different in both cohorts and may not have been captured consistently 5. The risk of bleeding in patients starting dronedarone after the initiation of a DOAC was not assessed due to small sample size. 6. The effect of the dose of DOAC on the risk of bleeding was not assessed due to lack of availability of information on dosage. Additionally, residual confounding due to unmeasured and unavailable confounders cannot be ruled out, e.g., alcohol use, medications such as antiepileptic drugs, St John's wort, and laboratory data on kidney and liver function was not availbale. Since the dose of DOACs, concomitant drugs, and abnormal renal function are independently associated with a risk of bleeding, regardless of dronedarone exposure, the findings should be interpreted with these limitations in mind.
To conclude, this study highlights the need to increase awareness among healthcare professionals of the potentially increased risk of bleeding that may be associated when starting concomitant use of some DOACs and CYP3A4/P-gp inhibitors such as dronedarone, and the need to follow recommendations and closely monitor these patients after treatment initiation.
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.
Prospective analyses, preferably randomized controlled studies, are needed to further explore the risk of major bleeding with concomitant use of dronedarone and DOACs.
Acknowledgement
Editorial assistance/medical writing support was provided by Dr. Ubedul Hoda (PhD) and Archana Rai, Sanofi India, Hyderabad. We thank Wanda Stipek, PharmD, BCPS, (Sanofi) for coordinating the development, facilitating author discussions, and critical review of this manuscript. We would also like to thank Michael D Ezekowitz II, MBChB, Dphil from Mainline Health, Thomas Jefferson University, Villanova, PA, USA for his valuable contribution in data interpretation and manuscript review.
Disclosures
Rania Boiron, and Mattias Wieloch are employees of Sanofi. Sampada K. Gandhi was an employee of Sanofi during the conduct of the study. Mattias Wieloch owns Sanofi shares and has an investment portfolio which at times may include shares of Sanofi or other pharmaceutical companies. James A. Reiffel received grants and personal fees from Medtronic Inc, Janssen/Johnson&Johnson, and Sanofi, and personal fees from, Acesion, Amerin, Correvio, and Roivant.
Drug interaction study with dabigatran etexilate and dronedarone in healthy subjects. ClinicalTrials.gov Identifier: NCT01306162. Accessed December 24, 2020. https://clinicaltrials.gov/ct2/show/NCT01306162.
Association between use of non-vitamin K oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.
Non-vitamin K oral anticoagulants are non-inferior for stroke prevention but cause fewer major bleedings than well-managed warfarin: a retrospective register study.
The 2018 European heart rhythm association practical guide on the use of non-vitamin k antagonist oral anticoagulants in patients with atrial fibrillation.
For DOAC, the use in the 183-day period prior to the index date, excluding the index date was identified.ACE = Angiotensin-converting enzyme; ARB = Angiotensin-receptor blockers; CCB = Calcium channel blocker; CCI = Charlson comorbidity index; DOAC = Direct oral anticoagulants; NSAID = Nonsteroidal anti-inflammatory drug; PPI = Proton pump inhibitor; SD = standard deviation; SNRI = Serotonin norepinephrine reuptake inhibitor; SSRI = Selective serotonin reuptake inhibitor.
00772-4&id=gr1.jpg){kind=link}