Risk of Hospitalization With Hemorrhage Among Older Adults Taking Clarithromycin vs Azithromycin and Direct Oral Anticoagulants | Atrial Fibrillation | JAMA Internal Medicine | JAMA Network
[Skip to Navigation]
Table 1.  Baseline Patient Characteristics for the Cohort Study Comparing Clarithromycin and Azithromycin Among Patients Taking DOACs
Baseline Patient Characteristics for the Cohort Study Comparing Clarithromycin and Azithromycin Among Patients Taking DOACs
Table 2.  Thirty-Day Rate of Hemorrhage With Clarithromycin vs Azithromycin Among Patients Taking DOACs
Thirty-Day Rate of Hemorrhage With Clarithromycin vs Azithromycin Among Patients Taking DOACs
Table 3.  Baseline Characteristics of Patients Taking DOACs and Concurrent Clarithromycin With a Major Hemorrhage Event Included in a Self-controlled Case Series
Baseline Characteristics of Patients Taking DOACs and Concurrent Clarithromycin With a Major Hemorrhage Event Included in a Self-controlled Case Series
Table 4.  Rate Ratio of Hemorrhage Events With Concurrent Clarithromycin and Direct Oral Anticoagulant Use From a Self-controlled Case Series
Rate Ratio of Hemorrhage Events With Concurrent Clarithromycin and Direct Oral Anticoagulant Use From a Self-controlled Case Series
1.
Canadian Institute for Health Information.  Adverse Drug Reaction–Related Hospitalizations Among Seniors, 2006 to 2011. Canadian Institute for Health Information; 2013.
2.
Weitz  JI, Semchuk  W, Turpie  AGG,  et al.  Trends in prescribing oral anticoagulants in Canada, 2008-2014.   Clin Ther. 2015;37(11):2506-2514. doi:10.1016/j.clinthera.2015.09.008 PubMedGoogle ScholarCrossref
3.
Connolly  SJ, Ezekowitz  MD, Yusuf  S,  et al; RE-LY Steering Committee and Investigators.  Dabigatran versus warfarin in patients with atrial fibrillation.   N Engl J Med. 2009;361(12):1139-1151. doi:10.1056/NEJMoa0905561 PubMedGoogle ScholarCrossref
4.
Granger  CB, Alexander  JH, McMurray  JJV,  et al; ARISTOTLE Committees and Investigators.  Apixaban versus warfarin in patients with atrial fibrillation.   N Engl J Med. 2011;365(11):981-992. doi:10.1056/NEJMoa1107039 PubMedGoogle ScholarCrossref
5.
Patel  MR, Mahaffey  KW, Garg  J,  et al; ROCKET AF Investigators.  Rivaroxaban versus warfarin in nonvalvular atrial fibrillation.   N Engl J Med. 2011;365(10):883-891. doi:10.1056/NEJMoa1009638 PubMedGoogle ScholarCrossref
6.
Agnelli  G, Buller  HR, Cohen  A,  et al; AMPLIFY Investigators.  Oral apixaban for the treatment of acute venous thromboembolism.   N Engl J Med. 2013;369(9):799-808. doi:10.1056/NEJMoa1302507 PubMedGoogle ScholarCrossref
7.
Bauersachs  R, Berkowitz  SD, Brenner  B,  et al; EINSTEIN Investigators.  Oral rivaroxaban for symptomatic venous thromboembolism.   N Engl J Med. 2010;363(26):2499-2510. doi:10.1056/NEJMoa1007903 PubMedGoogle ScholarCrossref
8.
Büller  HR, Prins  MH, Lensin  AW,  et al; EINSTEIN–PE Investigators.  Oral rivaroxaban for the treatment of symptomatic pulmonary embolism.   N Engl J Med. 2012;366(14):1287-1297. doi:10.1056/NEJMoa1113572 PubMedGoogle ScholarCrossref
9.
Büller  HR, Décousus  H, Grosso  MA,  et al; Hokusai-VTE Investigators.  Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism.   N Engl J Med. 2013;369(15):1406-1415. doi:10.1056/NEJMoa1306638 PubMedGoogle ScholarCrossref
10.
Schulman  S, Kakkar  AK, Goldhaber  SZ,  et al; RE-COVER II Trial Investigators.  Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.   Circulation. 2014;129(7):764-772. doi:10.1161/CIRCULATIONAHA.113.004450 PubMedGoogle ScholarCrossref
11.
Schulman  S, Kearon  C, Kakkar  AK,  et al; RE-COVER Study Group.  Dabigatran versus warfarin in the treatment of acute venous thromboembolism.   N Engl J Med. 2009;361(24):2342-2352. doi:10.1056/NEJMoa0906598 PubMedGoogle ScholarCrossref
12.
Palacio  AM, Kirolos  I, Tamariz  L.  Patient values and preferences when choosing anticoagulants.   Patient Prefer Adherence. 2015;9:133-138.PubMedGoogle Scholar
13.
Skanes  AC, Healey  JS, Cairns  JA,  et al; Canadian Cardiovascular Society Atrial Fibrillation Guidelines Committee.  Focused 2012 update of the Canadian Cardiovascular Society atrial fibrillation guidelines: recommendations for stroke prevention and rate/rhythm control.   Can J Cardiol. 2012;28(2):125-136. doi:10.1016/j.cjca.2012.01.021 PubMedGoogle ScholarCrossref
14.
Thrombosis Canada. Clinical guides. Accessed August 13, 2019. http://thrombosiscanada.ca/clinicalguides/#
15.
Voukalis  C, Lip  GYH, Shantsila  E.  Drug-drug interactions of non–svitamin K oral anticoagulants.   Expert Opin Drug Metab Toxicol. 2016;12(12):1445-1461. doi:10.1080/17425255.2016.1225037 PubMedGoogle ScholarCrossref
16.
Lin  JH, Yamazaki  M.  Role of P-glycoprotein in pharmacokinetics: clinical implications.   Clin Pharmacokinet. 2003;42(1):59-98. doi:10.2165/00003088-200342010-00003 PubMedGoogle ScholarCrossref
17.
Lin  JH, Lu  AYH.  Inhibition and induction of cytochrome P450 and the clinical implications.   Clin Pharmacokinet. 1998;35(5):361-390. doi:10.2165/00003088-199835050-00003 PubMedGoogle ScholarCrossref
18.
Westphal  JF.  Macrolide-induced clinically relevant drug interactions with cytochrome P-450A (CYP) 3A4: an update focused on clarithromycin, azithromycin and dirithromycin.   Br J Clin Pharmacol. 2000;50(4):285-295. doi:10.1046/j.1365-2125.2000.00261.x PubMedGoogle ScholarCrossref
19.
Kanoh  S, Rubin  BK.  Mechanisms of action and clinical application of macrolides as immunomodulatory medications.   Clin Microbiol Rev. 2010;23(3):590-615. doi:10.1128/CMR.00078-09 PubMedGoogle ScholarCrossref
20.
Shulman  ST, Bisno  AL, Clegg  HW,  et al; Infectious Diseases Society of America.  Clinical practice guideline for the diagnosis and management of group A streptococcal pharyngitis: 2012 update by the Infectious Diseases Society of America.   Clin Infect Dis. 2012;55(10):e86-e102. Published correction appears in Clin Infect Dis. 2014;58(10):1496. doi:10.1093/cid/cis629 PubMedGoogle ScholarCrossref
21.
Stevens  DL, Bisno  AL, Chambers  HF,  et al; Infectious Diseases Society of America.  Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the Infectious Diseases Society of America.   Clin Infect Dis. 2014;59(2):e10-e52. Published correction appears in Clin Infect Dis. 2015;60(9):1448. doi:10.1093/cid/ciu296 PubMedGoogle ScholarCrossref
22.
Karlowsky  JA, Lagacé-Wiens  PRS, Low  DE, Zhanel  GG.  Annual macrolide prescription rates and the emergence of macrolide resistance among Streptococcus pneumoniae in Canada from 1995 to 2005.   Int J Antimicrob Agents. 2009;34(4):375-379. doi:10.1016/j.ijantimicag.2009.05.008 PubMedGoogle ScholarCrossref
23.
Fralick  M, Juurlink  DN, Marras  T.  Bleeding associated with coadministration of rivaroxaban and clarithromycin.   CMAJ. 2016;188(9):669-672. doi:10.1503/cmaj.150580 PubMedGoogle ScholarCrossref
24.
Liesenfeld  KH, Lehr  T, Dansirikul  C,  et al.  Population pharmacokinetic analysis of the oral thrombin inhibitor dabigatran etexilate in patients with non-valvular atrial fibrillation from the RE-LY trial.   J Thromb Haemost. 2011;9(11):2168-2175. doi:10.1111/j.1538-7836.2011.04498.x PubMedGoogle ScholarCrossref
25.
Mueck  W, Kubitza  D, Becka  M.  Co-administration of rivaroxaban with drugs that share its elimination pathways: pharmacokinetic effects in healthy subjects.   Br J Clin Pharmacol. 2013;76(3):455-466. doi:10.1111/bcp.12075 PubMedGoogle ScholarCrossref
26.
Delavenne  X, Ollier  E, Basset  T,  et al.  A semi-mechanistic absorption model to evaluate drug-drug interaction with dabigatran: application with clarithromycin.   Br J Clin Pharmacol. 2013;76(1):107-113. doi:10.1111/bcp.12055 PubMedGoogle ScholarCrossref
27.
Vranckx  P, Valgimigli  M, Heidbuchel  H.  The significance of drug-drug and drug-food interactions of oral anticoagulation.   Arrhythm Electrophysiol Rev. 2018;7(1):55-61. doi:10.15420/aer.2017.50.1 PubMedGoogle ScholarCrossref
28.
Zhao  Y, Hu  Z-Y.  Physiologically based pharmacokinetic modelling and in vivo [I]/Ki accurately predict P-glycoprotein-mediated drug-drug interactions with dabigatran etexilate.   Br J Pharmacol. 2014;171(4):1043-1053. doi:10.1111/bph.12533 PubMedGoogle ScholarCrossref
29.
Garonzik  S, Byon  W, Myers  E, Li  X, Marchisin  D, Murthy  B.  The effects of clarithromycin on the pharmacokinetics of apixaban in healthy volunteers: a single-sequence crossover study.   Am J Cardiovasc Drugs. 2019;19(6):561-567. doi:10.1007/s40256-019-00348-2 PubMedGoogle ScholarCrossref
30.
Gouin-Thibault  I, Delavenne  X, Blanchard  A,  et al.  Interindividual variability in dabigatran and rivaroxaban exposure: contribution of ABCB1 genetic polymorphisms and interaction with clarithromycin.   J Thromb Haemost. 2017;15(2):273-283. doi:10.1111/jth.13577 PubMedGoogle ScholarCrossref
31.
Moj  D, Maas  H, Schaeftlein  A, Hanke  N, Gómez-Mantilla  JD, Lehr  T.  A comprehensive whole-body physiologically based pharmacokinetic model of dabigatran etexilate, dabigatran and dabigatran glucuronide in healthy adults and renally impaired patients.   Clin Pharmacokinet. 2019;58(12):1577-1593. doi:10.1007/s40262-019-00776-yPubMedGoogle ScholarCrossref
32.
Stöllberger  C, Finsterer  J.  Relevance of P-glycoprotein in stroke prevention with dabigatran, rivaroxaban, and apixaban.   Herz. 2015;40(2)(suppl 2):140-145. doi:10.1007/s00059-014-4188-9 PubMedGoogle ScholarCrossref
33.
Xu  R, Ge  W, Jiang  Q.  Application of physiologically based pharmacokinetic modeling to the prediction of drug-drug and drug-disease interactions for rivaroxaban.   Eur J Clin Pharmacol. 2018;74(6):755-765. doi:10.1007/s00228-018-2430-8 PubMedGoogle ScholarCrossref
34.
Fleet  JL, Shariff  SZ, Bailey  DG,  et al.  Comparing two types of macrolide antibiotics for the purpose of assessing population-based drug interactions.   BMJ Open. 2013;3(7):e002857. doi:10.1136/bmjopen-2013-002857 PubMedGoogle Scholar
35.
ELIQUIS (axpixaban). Product monograph. Accessed September 4, 2017. http://secure.healthlinks.net.au/content/bms/pi.cfm?product=bqpeliqu11112
36.
Praxada (dabigatran etexilate). Product monograph. Accessed August 26, 2017. https://www.ema.europa.eu/docs/en_GB/document_library/EPAR__Product_Information/human/000829/WC500041059.pdf
38.
Witt  DM, Nieuwlaat  R, Clark  NP,  et al.  American Society of Hematology 2018 guidelines for management of venous thromboembolism: optimal management of anticoagulation therapy.   Blood Adv. 2018;2(22):3257-3291. doi:10.1182/bloodadvances.2018024893 PubMedGoogle ScholarCrossref
39.
Kearon  C, Akl  EA, Ornelas  J,  et al.  Antithrombotic therapy for VTE: CHEST guideline and expert panel report.   Chest. 2016;149(2):315-352. doi:10.1016/j.chest.2015.11.026 PubMedGoogle ScholarCrossref
40.
Macle  L, Cairns  J, Leblanc  K,  et al; CCS Atrial Fibrillation Guidelines Committee.  2016 Focused update of the Canadian Cardiovascular Society guidelines for the management of atrial fibrillation.   Can J Cardiol. 2016;32(10):1170-1185. doi:10.1016/j.cjca.2016.07.591 PubMedGoogle ScholarCrossref
41.
Statistics Canada. Population estimates on July 1st, by age and sex. Accessed May 1, 2018. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1710000501
42.
Levy  AR, O’Brien  BJ, Sellors  C, Grootendorst  P, Willison  D.  Coding accuracy of administrative drug claims in the Ontario Drug Benefit database.   Can J Clin Pharmacol. 2003;10(2):67-71.PubMedGoogle Scholar
43.
Benchimol  EI, Smeeth  L, Guttmann  A,  et al; RECORD Working Committee.  The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement.   PLoS Med. 2015;12(10):e1001885. doi:10.1371/journal.pmed.1001885 PubMedGoogle Scholar
44.
Wahl  PM, Rodgers  K, Schneeweiss  S,  et al.  Validation of claims-based diagnostic and procedure codes for cardiovascular and gastrointestinal serious adverse events in a commercially-insured population.   Pharmacoepidemiol Drug Saf. 2010;19(6):596-603. doi:10.1002/pds.1924 PubMedGoogle ScholarCrossref
45.
Whitaker  HJ, Farrington  CP, Spiessens  B, Musonda  P.  Tutorial in biostatistics: the self-controlled case series method.   Stat Med. 2006;25(10):1768-1797. doi:10.1002/sim.2302 PubMedGoogle ScholarCrossref
46.
Othman  F, Crooks  CJ, Card  TR.  Community acquired pneumonia incidence before and after proton pump inhibitor prescription: population based study.   BMJ. 2016;355:i5813. doi:10.1136/bmj.i5813 PubMedGoogle ScholarCrossref
47.
Douglas  IJ, Evans  SJW, Hingorani  AD,  et al.  Clopidogrel and interaction with proton pump inhibitors: comparison between cohort and within person study designs.   BMJ. 2012;345:e4388. doi:10.1136/bmj.e4388 PubMedGoogle Scholar
48.
Masclee  GMC, Valkhoff  VE, Coloma  PM,  et al.  Risk of upper gastrointestinal bleeding from different drug combinations.   Gastroenterology. 2014;147(4):784-792.e9. doi:10.1053/j.gastro.2014.06.007 PubMedGoogle ScholarCrossref
49.
Suchard  MA, Zorych  I, Simpson  SE, Schuemie  MJ, Ryan  PB, Madigan  D.  Empirical performance of the self-controlled case series design: lessons for developing a risk identification and analysis system.   Drug Saf. 2013;36(1)(suppl 1):S83-S93. doi:10.1007/s40264-013-0100-4 PubMedGoogle ScholarCrossref
50.
Abraham  NS, DeSilva  R, Richardson  P.  Derivation and validation of an algorithm to identify helicobacter pylori infected patients using administrative data: 850.   Am J Gastroenterol. 2007;102:S431. doi:10.14309/00000434-200709002-00850 Google Scholar
51.
Austin  PC.  Using the standardized difference to compare the prevalence of a binary variable between two groups in observational research.   Commun Stat. 2009;38:1228-1234. doi:10.1080/03610910902859574 Google ScholarCrossref
52.
Chang  SH, Chou  IJ, Yeh  YH,  et al.  Association between use of non–vitamin k oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.   JAMA. 2017;318(13):1250-1259. doi:10.1001/jama.2017.13883 PubMedGoogle ScholarCrossref
53.
Wang  N, Giblin  E, Rodgers  JE.  Drug interactions with non–vitamin K oral anticoagulants.   JAMA. 2018;319(8):830. doi:10.1001/jama.2017.20850 PubMedGoogle ScholarCrossref
54.
Li  Y, Dong  S, Soria-Saucedo  R.  Drug interactions with non–vitamin K oral anticoagulants.   JAMA. 2018;319(8):827-828. doi:10.1001/jama.2017.20830 PubMedGoogle ScholarCrossref
55.
Fralick  M, Juurlink  DN, Marras  T.  Bleeding associated with coadministration of rivaroxaban and clarithromycin.   CMAJ. 2016;188(9):669-672. doi:10.1503/cmaj.150580PubMedGoogle ScholarCrossref
56.
Gnoth  MJ, Buetehorn  U, Muenster  U, Schwarz  T, Sandmann  S.  In vitro and in vivo P-glycoprotein transport characteristics of rivaroxaban.   J Pharmacol Exp Ther. 2011;338(1):372-380. doi:10.1124/jpet.111.180240 PubMedGoogle ScholarCrossref
57.
Schulman  S, Beyth  RJ, Kearon  C, Levine  MN.  Hemorrhagic complications of anticoagulant and thrombolytic treatment: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition).   Chest. 2008;133(6)(suppl):257S-298S. doi:10.1378/chest.08-0674PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    EXPAND ALL
    Interaction of DOACS with clarithromycin and azithromycin
    J David Spence, M.D. | Robarts Research Institute, Western University, London, ON, Canada
    Hill et al.[1] report that the risk of bleeding was increased among persons prescribed direct-acting oral anticoagulants (DOACs) and clarithromycin or azithromycin. They suggest that blood levels of rivaroxaban and apixaban should be increased more than those of dabigatran by macrolides, because dabigatran is not metabolized by CYP3A4. However, dabigatran is affected by p-glycoprotein (Pgp); indeed that may account for the very low bioavailability of dabigatran (only 7%, compared with 50% for apixaban and 80% for rivaroxaban). Clarithromycin, in addition to inhibiting CYP3A4, is also a potent inhibitor of Pgp.[2]

    Drugs with a low bioavailability are subject to very
    large changes in blood levels with drug interactions. This is illustrated by the interaction of statins and grapefruit, a potent inhibitor of CYP3A4.[3] Simvastatin is only 5% bioavailable because of first pass metabolism by CYP3A4 in the intestinal wall. Theoretically, if CYP3A4 were completely blocked, blood levels could increase 20-fold. Indeed, the AUC of simvastatin increases 15-fold with grapefruit juice.[4] Atorvastatin is 50% bioavailable, so its AUC “only” doubles with grapefruit.[5]

    Drieier et al. described a case of rhabdomyolysis with simvastatin after only 4 days of consuming one grapefruit daily.[6] Furthermore, CYP3A4 is not only affected by grapefruit, but by many drugs;[7] that is why simvastatin should be regarded as a dangerous drug. The low bioavailability of dabigatran may be one reason why there was a drastic increase in plasma levels of dabigatran with administration of intravenous immunoglobulin. Perhaps dabigatran should also be so regarded. It has been suggested that dabigatran blood levels should be monitored.[8]

    References
    1. Hill K, Sucha E, Rhodes E, Carrier M, Garg AX, Harel Z, et al. Risk of Hospitalization With Hemorrhage Among Older Adults Taking Clarithromycin vs Azithromycin and Direct Oral Anticoagulants. JAMA Intern Med. 2020.
    2. Gessner A, König J, Fromm MF. Clinical Aspects of Transporter-Mediated Drug-Drug Interactions. Clin Pharmacol Ther. 2019;105(6):1386-94.
    3. Bailey DG, Spence JD, Munoz C, Arnold JM. Bailey DG, Spence JD, Munoz C, Arnold JM. Interaction of citrus juices with felodipine and nifedipine. Lancet 1991;337(8736):268-9. Lancet. 1991;337:268-9.
    4. Lilja JJ, Kivisto KT, Neuvonen PJ. Grapefruit juice-simvastatin interaction: effect on serum concentrations of simvastatin, simvastatin acid, and HMG-CoA reductase inhibitors. Clin Pharmacol Ther. 1998;64(5):477-83.
    5. Lilja JJ, Kivisto KT, Neuvonen PJ. Grapefruit juice increases serum concentrations of atorvastatin and has no effect on pravastatin. Clin Pharmacol Ther. 1999;66(2):118-27.
    6. Dreier JP, Endres M. Statin-associated rhabdomyolysis triggered by grapefruit consumption. Neurology. 2004;62(4):670.
    7. Dresser GK, Spence JD, Bailey DG. Pharmacokinetic-pharmacodynamic consequences and clinical relevance of cytochrome P450 3A4 inhibition. Clin Pharmacokinet. 2000;38(1):41-57.
    8. Reilly PA, Lehr T, Haertter S, Connolly SJ, Yusuf S, Eikelboom JW, et al. The effect of dabigatran plasma concentrations and patient characteristics on the frequency of ischemic stroke and major bleeding in atrial fibrillation patients: the RE-LY Trial (Randomized Evaluation of Long-Term Anticoagulation Therapy). J Am Coll Cardiol. 2014;63(4):321-8.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Original Investigation
    June 8, 2020

    Risk of Hospitalization With Hemorrhage Among Older Adults Taking Clarithromycin vs Azithromycin and Direct Oral Anticoagulants

    Author Affiliations
    • 1Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada
    • 2Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada
    • 3Division of Nephrology, Department of Medicine, Health Sciences Centre, London, Ontario, Canada
    • 4Epidemiology and Biostatistics, Western University, London, Ontario, Canada
    • 5Division of Nephrology, Department of Medicine, St Michael’s Hospital, Toronto, Canada
    • 6Division of Nephrology, Department of Medicine, The Ottawa Hospital, Ottawa, Ontario, Canada
    • 7Ottawa Hospital Research Institute, The Ottawa Hospital, Ottawa, Ontario, Canada
    JAMA Intern Med. 2020;180(8):1052-1060. doi:10.1001/jamainternmed.2020.1835
    Key Points

    Question  Is the concurrent use of clarithromycin among older adult patients taking direct oral anticoagulants associated with a higher 30-day risk of hospitalization for major hemorrhage compared with azithromycin?

    Findings  In this population-level cohort study of 24 943 older adults taking direct oral anticoagulants, clarithromycin was associated with an adjusted 1.71-fold higher rate of hospitalization (absolute risk difference, 0.34%) within 30 days for a major hemorrhage event compared with azithromycin.

    Meaning  The use of clarithromycin was associated with a high rate of hemorrhage among older adults taking direct oral anticoagulants compared with azithromycin and poses a potential drug-drug interaction.

    Abstract

    Importance  Clarithromycin is a commonly prescribed antibiotic associated with higher levels of direct oral anticoagulants (DOACs) in the blood, with the potential to increase the risk of hemorrhage.

    Objective  To assess the 30-day risk of a hospital admission with hemorrhage after coprescription of clarithromycin compared with azithromycin among older adults taking a DOAC.

    Design, Setting, and Participants  This population-based, retrospective cohort study was conducted among adults of advanced age (mean [SD] age, 77.6 [7.2] years) who were newly coprescribed clarithromycin (n = 6592) vs azithromycin (n = 18 351) while taking a DOAC (dabigatran, apixaban, or rivaroxaban) in Ontario, Canada, from June 23, 2009, to December 31, 2016. Cox proportional hazards regression was used to examine the association between hemorrhage and antibiotic use (clarithromycin vs azithromycin). Statistical analysis was performed from December 23, 2019, to March 25, 2020.

    Main Outcomes and Measures  Hospital admission with major hemorrhage (upper or lower gastrointestinal tract or intracranial). Outcomes were assessed within 30 days of a coprescription.

    Results  Among the 24 943 patients (12 493 women; mean [SD] age, 77.6 [7.2] years) in the study, rivaroxaban was the most commonly prescribed DOAC (9972 patients [40.0%]), followed by apixaban (7953 [31.9%]) and dabigatran (7018 [28.1%]). Coprescribing clarithromycin vs azithromycin with a DOAC was associated with a higher risk of a hospital admission with major hemorrhage (51 of 6592 patients [0.77%] taking clarithromycin vs 79 of 18 351 patients [0.43%] taking azithromycin; adjusted hazard ratio, 1.71 [95% CI, 1.20-2.45]; absolute risk difference, 0.34%). Results were consistent in multiple additional analyses.

    Conclusions and Relevance  This study suggests that, among adults of advanced age taking a DOAC, concurrent use of clarithromycin compared with azithromycin was associated with a small but statistically significantly greater 30-day risk of hospital admission with major hemorrhage.

    Introduction

    Anticoagulant-associated hemorrhage is one of the most common adverse drug reactions requiring hospitalization among individuals of advanced age, with a 2-fold increase among those older than 75 years.1 Identification and avoidance of dangerous drug-drug interactions are associated with a significant reduction in adverse events and improvement in evidence-based prescription patterns.

    During the last decade, direct oral anticoagulants (DOACs) have supplanted traditional vitamin K antagonists as the anticoagulation drugs of choice.2 Large phase 3 trials have demonstrated noninferiority or superiority of DOACs relative to traditional anticoagulants (warfarin) for effectiveness in stroke prevention for those who have atrial fibrillation and for prevention and treatment of venous thromboembolism.3-11 Patient preferences for DOACs are based on their simplicity of use, with no need for routine bloodwork monitoring.12 As such, recent guidelines recommend DOACs as the first-line agents for the prevention of stroke in patients with nonvalvular atrial fibrillation (strong recommendation; high-quality evidence) and the treatment of venous thromboembolism.13,14Quiz Ref ID Direct oral anticoagulants have 2 predominant mechanisms of metabolism: P-glycoprotein (Pgp) cell transporters, which are involved in transcellular transportation, and the cytochrome P450 enzyme CYP3A4, which is involved in the metabolism in the human liver.15 Dabigatran etexilate mesylate requires efflux transportation by the Pgp system but is independent of the cytochrome P450 enzyme system.16 Apixaban and rivaroxaban are heavily reliant on the CYP3A4 enzyme complexes for hepatic metabolism.17

    Despite the widespread adoption of DOACs, their safety and drug interaction profile are not fully understood. Medications, such as some antibiotics, that act on these pathways have the potential to alter DOAC metabolism or excretion and change serum levels.18Quiz Ref ID Clarithromycin is a commonly prescribed macrolide antibiotic used in the treatment of respiratory infections, uncomplicated skin and soft tissue infections, nontuberculous mycobacterial infections, Helicobacter pylori eradication, and streptococcal pharyngitis.19-22 It is a potent inhibitor of CYP3A4 and Pgp. Multiple pharmacokinetic studies have demonstrated that concomitant use of apixaban, rivaroxaban, or dabigatran with clarithromycin increases serum levels of DOACs by 20% to 100% and prolongs coagulation time.23-33 In contrast, a similar and comparable macrolide class antibiotic, azithromycin, demonstrates minimal CYP3A4 and Pgp inhibition.34 Although this interaction would imply that combined use of a DOAC and clarithromycin would increase adverse bleeding events, whether this is clinically relevant remains unknown, to our knowledge. Nevertheless, based on the pharmacokinetic data, warnings about the concurrent use of strong CYP3A4 inhibitors and a heightened hemorrhagic risk are included on DOAC product monographs.35-37 In addition, recent treatment guidelines recommend DOAC dose adjustments with clarithromycin use or suggest selecting an alternative anticoagulation agent.14,38-40

    Because knowledge of the risk of bleeding with concurrent DOACs and clarithromycin is limited, we examined whether the risk of bleeding was elevated among patients taking DOACs who were treated with concurrent clarithromycin compared with azithromycin. We hypothesized that concomitant DOAC and clarithromycin use would be associated with an elevated risk of hemorrhagic events.

    Methods
    Data Sources

    We used deidentified, linked databases housed at the Institute for Clinical Sciences (ICES; see eTable 1 in the Supplement for description of databases used in this study). Demographic characteristics and vital status information were obtained from the Ontario Registered Persons Database. Medication information was obtained from the Ontario Drug Benefit Claims database. Ontario is Canada’s largest province, with more than 13 million residents.41 All citizens have access to universal public health care, with drug coverage for individuals older than 65 years. This database contains highly accurate records of all outpatient prescriptions dispensed to patients 65 years or older, with an error rate of less than 1%.42 Diagnostic and procedural information from all hospitalizations was determined using the Canadian Institute for Health Information Discharge Abstract Database. Diagnostic information from emergency department visits was determined using the National Ambulatory Care Reporting System. Information was also obtained from the Ontario Health Insurance Plan database, which contains all claims for inpatient and outpatient physician services. Whenever possible, we defined patient characteristics and outcomes using validated codes. The use of data in this project was authorized under section 45 of Ontario’s Personal Health Information Protection Act, which does not require review by a Research Ethics Board. The study used deidentified data and patient consent was waived as per the Ontario Ministry of Health. The reporting of this study follows guidelines for observational studies (eTable 2 in the Supplement).43

    Design and Setting

    The study population included all adults in Ontario, Canada, 66 years or older from June 23, 2009 (the first date that DOACs were added to the Ontario Drug Formulary), to December 31, 2016 (eFigure 1 in the Supplement). Prescription drug information is available for all adults older than 65 years in Ontario; we initiated our cohort at 66 years to allow for a 1-year look-back period for existing medications. We identified an exposed cohort of individuals who received a new prescription for a DOAC (apixaban, dabigatran, or rivaroxaban). We then identified a subset of patients who received a prescription for clarithromycin (exposure of interest) or azithromycin (active comparator; eTable 3 in the Supplement). Azithromycin is also a macrolide class antimicrobial; however, it demonstrates very weak CYP3A4 and Pgp inhibition relative to clarithromycin.18 It is a well-suited comparator for clarithromycin because it is prescribed to similar ambulatory patient populations in terms of characteristics, comorbid illnesses, medication use, cause of infection, prescribing physician, and hemorrhagic risk.34 The antibiotic dispensing date served as the study index date, and patients with prior use of other potent CYP3A4 or Pgp inhibitors during the 90-day look-back period from index (medications included conazole antifungals, tacrolimus, cyclosporine, quinines, and rifampin [eTable 4 in the Supplement]) were excluded. Clarithromycin users (exposure group) were compared with azithromycin users with follow-up for the outcome of interest of up to 30 days after index date. Discontinuaton of the DOAC drug was defined as no refill within 1.5 times the original prescription length plus 90 days. Individuals undergoing dialysis or those who had received a kidney transplant were excluded.

    Outcomes

    The study outcome was a hospital admission or emergency department visit with major hemorrhage up to 30 days after dispensing of the study antibiotic (see eTable 5 in the Supplement for outcome definitions). The following types of hemorrhage were included in the outcome of major hemorrhage: upper or lower gastrointestinal, intracerebral, subarachnoid, and other nontraumatic intracranial (sensitivity, 94%; positive predictive value, 87%).44 Hospitalizations with a diagnosis of hemorrhage were identified using International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes in the Canadian Institute for Health Information Discharge Abstract Database.

    Study Design

    We compared all patients taking DOACs who received a prescription for clarithromycin with all patients taking DOACs who received a prescription for azithromycin using a cohort study design. Potential confounders examined included demographic characteristics (age, sex, income, and place of residence), index year, comorbid illnesses (history of hemorrhage, hypertension, diabetes, stroke, atrial fibrillation, acute coronary syndrome, heart failure, coronary artery disease, coronary artery bypass grafting, percutaneous coronary intervention, peripheral vascular disease, and venous thromboembolism), health care use (numbers of hospitalizations and emergency department visits in the preceding 5 years), and medications (β-blockers, nonsteroidal anti-inflammatory drugs, proton pump inhibitors, antiplatelet agents, selective serotonin reuptake inhibitors, and statins).

    Additional Analyses

    We conducted a number of additional analyses. First, we performed a self-controlled case series (SCCS), a variation of the case-control design in which all patients taking DOACs who experienced a hemorrhage (cases) would be examined for hemorrhage risk by comparing period(s) of exposure to clarithromycin vs period(s) of nonexposure45 (eFigure 2 in the Supplement). The risk of hemorrhage would be compared within the same individual. Thereby, an individual serves as their own control, limiting confounding other than for time and potential time-varying characteristics, for which additional adjustment is performed. Further strengths of the SCCS study design include it allowing for recurrent exposures and/or repeated outcome events, it is well suited to short exposure periods, and it has been previously used specifically to examine drug interactions.46-49 We identified time periods of exposure as 30 days from the dispensing date of a study antibiotic. Time periods of nonexposure were defined as all study time in which none of the study antibiotics were prescribed and the individual continued to take a DOAC. The DOAC prescription date was used as the start of follow-up, and individuals were followed up until death, DOAC discontinuation, or the maximum follow-up date. Second, we repeated all analyses using a liberal definition of hemorrhage that included any bleeding event or receipt of a blood transfusion with presentation to an emergency department or hospitalization. Third, we excluded those with a history of H pylori infection (3-year look back) because it is a common indication for clarithromycin and may predispose to gastrointestinal-related hemorrhage (identified by ICD-10 code B96.81; positive predictive value, 97.4%).50 Fourth, we repeated our models using inverse treatment probability weighting incorporating all covariates listed in Table 1, including duration of prior DOAC use. Fifth, we repeated our models examining fracture and the composite of depression and anxiety as negative outcomes. Sixth, we repeated our models in days 30 to 90 after antibiotic prescription to examine whether the association was attenuated after completion of the antibiotic course. Seventh, we repeated our models for individuals with known kidney function (using estimated glomerular filtration rate by the chronic kidney disease–epidemiology collaboration formula).

    Statistical Analysis

    Statistical analysis was performed from December 23, 2019, to March 25, 2020. For the cohort study, we used standardized differences to assess baseline characteristics by exposure status (clarithromycin vs azithromycin). Standardized differences describe differences between group mean values relative to the pooled SD and are less sensitive to large sample sizes than traditional hypothesis testing, and a significant difference is considered to be 10% or greater.51 We examined the association of clarithromycin vs azithromycin exposure and hemorrhage using Cox proportional hazards regression models. Schoenfeld residuals were examined to test the proportionality assumption. Only the first hemorrhage event was considered. Models were adjusted for variables detected to be different by a standardized difference greater than 10% between the 2 groups. To examine for effect modification by DOAC type (apixaban, dabigatran, or rivaroxaban), separate models with interaction terms were examined. For the SCCS, we used conditional Poisson regression models to determine the rate ratio of hemorrhage during clarithromycin exposure compared with nonexposure periods, adjusting for time as a continuous variable.45 Recurrent outcome events were included in the SCCS analysis. For the inverse probability treatment weighting, we calculated weights including all covariates listed in Table 1. We then used Cox proportional hazards regression models with the applied stabilized weights truncated at the first and 99th percentile. We conducted all analyses with SAS Enterprise software, version 7.1 (SAS Institute Inc). The 95% CIs that did not overlap with 1 were treated as statistically significant. All P values were from 2-sided tests and results were deemed statistically significant at P < .05.

    Results

    From a total of 24 943 unique patients taking DOACs, we identified 6592 (26.4%) who received clarithromycin and 18 351 (73.6%) who received azithromycin during the study period (Table 1). A total of 9025 patients (36.2%) were between 66 and 75 years of age, and 22 075 (88.5%) resided in urban centers. Concurrent antibiotic and DOAC use increased over time. The most common comorbidities were hypertension (21 657 [86.8%]) and diabetes (8827 [35.4%]). β-Blockers were prescribed for 14 436 patients (57.9%), and statins were prescribed for 15 840 patients (63.5%). There was little difference between the 2 groups, with the exceptions of index year of cohort entry, proton pump inhibitor use, mean daily DOAC dose, and DOAC type. Rivaroxaban was the most commonly used DOAC (9972 [40.0%]), followed by apixaban (7953 [31.9%]) and dabigatran (7018 [28.1%]). The mean (SD) daily dose of apixaban and rivaroxaban was lower among clarithromycin users than azithromycin users (apixaban, 7.41 [2.8] vs 7.49 [4.7] mg; rivaroxaban, 17.47 [3.5] vs 17.9 [6.7] mg), and the mean (SD) duration of DOAC use prior to antibiotic exposure was longer for azithromycin users than clarithromycin users (390 [0.11] vs 353 [0.11] days). Kidney function was measured for 21 673 patients (86.9%), with 8355 (33.5%) having a baseline estimated glomerular filtration rate of 60 mL/min/1.73m2 or less, and did not differ between the 2 groups.

    A total of 130 hemorrhagic events (0.52%) occurred within 30 days using the stringent outcome definition, and 308 hemorrhagic events (1.23%) occurred within 30 days using the more liberal outcome definition (Table 2). Major hemorrhage occurred in 51 of 6592 patients (0.77%) taking clarithromycin and 79 of 18 351 patients (0.43%) taking azithromycin. The crude incident rate for major hemorrhage was higher among patients taking clarithromycin compared with those taking azithromycin (95.9 [95% CI, 89.3-102.9] per 1000 person-years for clarithromycin users vs 53.1 [95% CI, 50.2-56.2] per 1000 person-years for azithromycin users). The higher rate with clarithromycin was consistent after adjustment for proton pump inhibitor use, DOAC type, and DOAC mean daily dose (hazard ratio [HR], 1.71 [95% CI, 1.20-2.45]). Neither outcome differed by DOAC type.

    In additional analyses, we identified 744 major hemorrhage events among 647 unique individuals taking DOACs who were exposed to clarithromycin in the SCCS (Table 3). A total of 69 events occurred during periods of clarithromycin use, whereas 675 occurred during period of clarithromycin nonuse. More than one-third of patients had a history of major hemorrhage, a history of atrial fibrillation, diabetes, or cardiac disease. Use of β-blockers (396 [61.2%]), proton pump inhibitors (404 [62.4%]), and statins (411 [63.5%]) was frequent. The most commonly used DOAC was rivaroxaban (276 [42.7%]), followed by dabigtran (191 [29.5%]) and apixaban (180 [27.8%]). Major hemorrhagic events were associated with concurrent clarithromycin and DOAC use compared with DOAC use alone (rate ratio, 1.44 [95% CI, 1.08-1.92]) (Table 4).

    Our findings were consistent using the more broad definition of hemorrhage in the cohort study (308 of 24 943 events [1.2%]; clarithromycin, 109 of 6592 events [1.7%]; azithromycin, 199 of 18 351 events [1.1%]), with a higher incident rate for hemorrhage with clarithromycin use (204.8 [95% CI, 191.3-219.7] vs 133.7 [95% CI, 127.0-140.8]) and in the SCCS (1760 total events; periods of clarithromycin use, 145 events; periods of clarithromycin nonuse, 1615 events; rate ratio, 1.64 [95% CI, 1.35-1.98]) and after exclusion of individuals with a history of H pylori infection (HR, 1.53 [95% CI, 1.21-1.95]) in the cohort study. Our findings were consistent in inverse probability of treatment weighting models (major hemorrhage: HR, 1.77 [95% CI, 1.20-2.59]; any hemorrhage: HR, 1.50 [95% CI, 1.16-1.93]) (see eTable 6 in the Supplement for additional analyses). Quiz Ref IDNo association was identified with an antibiotic and either negative outcome (fracture: clarithromycin, 17 of 6592 events [0.3%]; azithromycin, 65 of 18 351 events [0.4%]; adjusted HR, 0.73 [95% CI, 0.43-1.73]; and anxiety and depression: clarithromycin, 11 of 6592 events [0.2%]; azithromycin, 35 of 18 351 events [0.2%]; adjusted HR, 0.87 [95% CI, 0.44-1.71]). When we examined the hemorrhage rate in days 30 to 90, the association was attenuated (major hemorrhage: HR, 1.13 [95% CI, 0.81-1.57]; any hemorrhage: HR, 1.05 [95% CI, 0.84-1.31]). Last, the association persisted in models accounting for kidney function (major hemorrhage: HR, 1.72 [95% CI, 1.17-2.52]; any hemorrhage: HR, 1.44 [95% CI, 1.12-1.86]).

    Discussion

    In a retrospective cohort study of patients taking DOACs, we found that the 30-day rate of hemorrhage requiring hospitalization or an emergency department visit after dispensing of clarithromycin was higher relative to azithromycin. Furthermore, the hemorrhage rate was similarly elevated when comparing periods of clarithromycin use with periods of nonuse within the same individual. These findings were consistent when a more broad-based definition for hemorrhage was used that included receipt of a blood transfusion, after excluding individuals with a history of H pylori infection, using alternative methods of controlling for confounding (inverse probability of treatment weighting), and limited to patients with known kidney function. No association was evident using negative controls or in the 30- to 90-day follow-up period after antibiotic administration. Our results suggest that the coadministration of clarithromycin and DOACs poses a small but significant drug-drug interaction and a higher clinical 30-day rate of hemorrhage.

    To date, limited clinical evidence exists on the use of clarithromycin with DOACs. Fralick et al23 reported a single case in which a patient taking rivaroxaban experienced spontaneous intracranial and pulmonary hemorrhages after being started on clarithromycin. The patient’s anti-Xa level, measured more than 30 hours after the last reported dose of rivaroxaban, was 537 μg/L (normal 24-hour trough level, 8-150 μg/L), suggesting significantly elevated serum levels at the time of bleeding. Chang et al52 evaluated the clinical risk of bleeding when DOACs were combined with other medications. With 4770 major bleeding events seen in the 91 330 patients who were taking a DOAC and followed up for 1 year, they found a paradoxical decreased adjusted incidence rate of bleeding in patients who received clarithromycin or erythromycin. The investigators postulated that this lower bleeding rate was the result of a decrease in gastrointestinal bleeding, secondary to clarithromycin use in the treatment of H pylori peptic ulcers, which may outweigh the increased risk of bleeding from elevated DOAC levels. However, this possibility was untested, discrepant with the existing pharmacokinetic evidence and potentially due to residual confounding, leading to further clinical uncertainty.53

    With regard to the clinical significance of our study, to our knowledge, this is the largest study to date examining clinically relevant bleeding with concomitant use of DOACs and clarithromycin. We used 2 different but complementary study designs that demonstrated consistency. Quiz Ref IDFrom a clinical perspective, the risk of major hemorrhage observed was less than 1.0% overall with clarithromycin use, with an absolute difference of 0.34% (roughly 1 in 300 exposures) between clarithromycin and azithromycin. Thereby, an individual’s hemorrhage risk, indication for anticoagulation, and availability of a suitable antibiotic substitute need to be carefully considered. In scenarios in which DOAC and clarithromycin are concurrently administered, our findings suggest a potential role for monitoring DOAC levels to prevent supratherapeutic levels.

    Direct oral anticoagulant levels appear to be consistently increased with concurrent clarithromycin use based on pharmacokinetic and pharmacodynamic studies.15,24-26,29-33,36,37,53-56 For dabigatran, when combined with clarithromycin, the areas under the plasma concentration–time curves (AUCs) of dabigatran increased from 49% to 100%, and the peak serum concentrations of dabigatran increased from 60% to 80%. In the case of apixaban, increases were seen in both its AUC (60%) and peak serum concentration (30%). With rivaroxaban, a similar trend was seen with increases in both AUCs (50%-94%) and peak serum concentrations (40%-92%) with concomitant use of clarithromycin. Of the 3 DOACs evaluated, apixaban and rivaroxaban are dependent on CYP3A4 metabolism and should appear to pose more risk with clarithromycin exposure. We specifically examined for effect modification in our models and did not detect any difference in bleeding risk by DOAC. The inability to detect differences by DOAC types suggests that either there is no increased risk of hemorrhage with all 3 DOACs examined or, alternatively, there was an inability to detect differences owing to limited sample sizes.

    Limitations

    Our findings should be interpreted with the limitations of our study in mind. First, our population was composed exclusively of patients older than 66 years. Differences may exist between our patient cohort and a younger population. However, older patients are at a high risk for bleeding events given comorbidities, polypharmacy, and increased risk of falls.57 Second, while our sample of patients exposed to DOACs and clarithromycin or azithromycin was quite large, the number of observed bleeding events in both groups was relatively small. Third, we did not examine for dosage adjustments in either DOACs or concurrent antibiotics at the time of prescription. Fourth, we excluded only potent CYP3A4 or Pgp inhibitors; other drugs with less potent inhibition may have been used. Quiz Ref IDFifth, unmeasured confounding may have occurred. Sixth, we identified patients based on prescription filling and are unable to make inferences on patient adherence to the medication. It was assumed that patients completed their full course of antibiotics and took their medications as prescribed, which may not have been true in all cases.

    Conclusions

    Among a large cohort of patients of advanced age (>66 years old) taking DOACs who were dispensed clarithromycin, there was a higher rate of hemorrhage requiring hospitalization compared with either azithromycin or periods of no clarithromycin use. Thus, the concurrent use of clarithromycin and DOACs poses a significant drug-drug interaction. Clinicians need to consider the risk of hemorrhage, the indication and microbial susceptibility of the infection being treated, and whether viable alternatives (either anticoagulant or antimicrobial) are readily available.

    Back to top
    Article Information

    Accepted for Publication: April 11, 2020.

    Corresponding Author: Manish M. Sood, MD, Ottawa Hospital Research Institute, The Ottawa Hospital, 1053 Carling Ave, PO Box 693, Civic Campus, 2-014 Administrative Services Building, Ottawa, ON K1Y 4E9, Canada (msood@toh.on.ca).

    Published Online: June 8, 2020. doi:10.1001/jamainternmed.2020.1835

    Author Contributions: Dr Sood had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Hill, Garg, Sood.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Hill, Hundemer, Sood.

    Critical revision of the manuscript for important intellectual content: Sucha, Rhodes, Carrier, Garg, Harel, Hundemer, Clark, Knoll, McArthur, Sood.

    Statistical analysis: Sucha, Hundemer.

    Obtained funding: Sood.

    Administrative, technical, or material support: Rhodes, Carrier, Harel, McArthur, Sood.

    Supervision: Knoll, Sood.

    Conflict of Interest Disclosures: Dr Carrier reported receiving grants and personal fees from Leo Pharma, BMS, and Pfizer; and personal fees from Servier, Bayer, and Sanofi outside the submitted work. Dr Sood reported receiving grants from the Heart and Stroke Foundation of Canada during the conduct of the study and speaker fees from AstraZeneca. No other disclosures were reported.

    Funding/Support: This work was supported by a grant-in-aid from the Heart and Stroke Foundation of Canada. This study was supported by the Institute for Clinical Sciences (ICES), which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. Core funding for ICES Ottawa is provided by University of Ottawa and The Ottawa Hospital Research Institute. The research was conducted by members of the ICES Kidney, Dialysis and Transplantation team. ICES is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. Dr Sood is supported by the Jindal Research Chair for the Prevention of Kidney Disease. Dr Garg is supported by the Dr Adam Linton Chair in Kidney Health Analytics and a Clinician Investigator Award from the Canadian Institutes of Health Research.

    Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Disclaimer: Parts of this material are based on data and/or information compiled and provided by the Canadian Institute for Health Information. However, the analyses, conclusions, opinions and statements expressed in the material are those of the authors, and not necessarily those of the Canadian Institute for Health Information. The opinions, results and conclusions are those of the authors and are independent from the funding sources. No endorsement by ICES, Western University, University of Ottawa, Ottawa Hospital Research Institute, Heart and Stroke Foundation of Canada, or the Ontario Ministry of Health and Long-Term Care is intended or should be inferred.

    Additional Contributions: We thank IMS Brogan Inc for use of their Drug Information Database.

    Additional Information: This study was completed at the ICES Western and Ottawa sites.

    References
    1.
    Canadian Institute for Health Information.  Adverse Drug Reaction–Related Hospitalizations Among Seniors, 2006 to 2011. Canadian Institute for Health Information; 2013.
    2.
    Weitz  JI, Semchuk  W, Turpie  AGG,  et al.  Trends in prescribing oral anticoagulants in Canada, 2008-2014.   Clin Ther. 2015;37(11):2506-2514. doi:10.1016/j.clinthera.2015.09.008 PubMedGoogle ScholarCrossref
    3.
    Connolly  SJ, Ezekowitz  MD, Yusuf  S,  et al; RE-LY Steering Committee and Investigators.  Dabigatran versus warfarin in patients with atrial fibrillation.   N Engl J Med. 2009;361(12):1139-1151. doi:10.1056/NEJMoa0905561 PubMedGoogle ScholarCrossref
    4.
    Granger  CB, Alexander  JH, McMurray  JJV,  et al; ARISTOTLE Committees and Investigators.  Apixaban versus warfarin in patients with atrial fibrillation.   N Engl J Med. 2011;365(11):981-992. doi:10.1056/NEJMoa1107039 PubMedGoogle ScholarCrossref
    5.
    Patel  MR, Mahaffey  KW, Garg  J,  et al; ROCKET AF Investigators.  Rivaroxaban versus warfarin in nonvalvular atrial fibrillation.   N Engl J Med. 2011;365(10):883-891. doi:10.1056/NEJMoa1009638 PubMedGoogle ScholarCrossref
    6.
    Agnelli  G, Buller  HR, Cohen  A,  et al; AMPLIFY Investigators.  Oral apixaban for the treatment of acute venous thromboembolism.   N Engl J Med. 2013;369(9):799-808. doi:10.1056/NEJMoa1302507 PubMedGoogle ScholarCrossref
    7.
    Bauersachs  R, Berkowitz  SD, Brenner  B,  et al; EINSTEIN Investigators.  Oral rivaroxaban for symptomatic venous thromboembolism.   N Engl J Med. 2010;363(26):2499-2510. doi:10.1056/NEJMoa1007903 PubMedGoogle ScholarCrossref
    8.
    Büller  HR, Prins  MH, Lensin  AW,  et al; EINSTEIN–PE Investigators.  Oral rivaroxaban for the treatment of symptomatic pulmonary embolism.   N Engl J Med. 2012;366(14):1287-1297. doi:10.1056/NEJMoa1113572 PubMedGoogle ScholarCrossref
    9.
    Büller  HR, Décousus  H, Grosso  MA,  et al; Hokusai-VTE Investigators.  Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism.   N Engl J Med. 2013;369(15):1406-1415. doi:10.1056/NEJMoa1306638 PubMedGoogle ScholarCrossref
    10.
    Schulman  S, Kakkar  AK, Goldhaber  SZ,  et al; RE-COVER II Trial Investigators.  Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.   Circulation. 2014;129(7):764-772. doi:10.1161/CIRCULATIONAHA.113.004450 PubMedGoogle ScholarCrossref
    11.
    Schulman  S, Kearon  C, Kakkar  AK,  et al; RE-COVER Study Group.  Dabigatran versus warfarin in the treatment of acute venous thromboembolism.   N Engl J Med. 2009;361(24):2342-2352. doi:10.1056/NEJMoa0906598 PubMedGoogle ScholarCrossref
    12.
    Palacio  AM, Kirolos  I, Tamariz  L.  Patient values and preferences when choosing anticoagulants.   Patient Prefer Adherence. 2015;9:133-138.PubMedGoogle Scholar
    13.
    Skanes  AC, Healey  JS, Cairns  JA,  et al; Canadian Cardiovascular Society Atrial Fibrillation Guidelines Committee.  Focused 2012 update of the Canadian Cardiovascular Society atrial fibrillation guidelines: recommendations for stroke prevention and rate/rhythm control.   Can J Cardiol. 2012;28(2):125-136. doi:10.1016/j.cjca.2012.01.021 PubMedGoogle ScholarCrossref
    14.
    Thrombosis Canada. Clinical guides. Accessed August 13, 2019. http://thrombosiscanada.ca/clinicalguides/#
    15.
    Voukalis  C, Lip  GYH, Shantsila  E.  Drug-drug interactions of non–svitamin K oral anticoagulants.   Expert Opin Drug Metab Toxicol. 2016;12(12):1445-1461. doi:10.1080/17425255.2016.1225037 PubMedGoogle ScholarCrossref
    16.
    Lin  JH, Yamazaki  M.  Role of P-glycoprotein in pharmacokinetics: clinical implications.   Clin Pharmacokinet. 2003;42(1):59-98. doi:10.2165/00003088-200342010-00003 PubMedGoogle ScholarCrossref
    17.
    Lin  JH, Lu  AYH.  Inhibition and induction of cytochrome P450 and the clinical implications.   Clin Pharmacokinet. 1998;35(5):361-390. doi:10.2165/00003088-199835050-00003 PubMedGoogle ScholarCrossref
    18.
    Westphal  JF.  Macrolide-induced clinically relevant drug interactions with cytochrome P-450A (CYP) 3A4: an update focused on clarithromycin, azithromycin and dirithromycin.   Br J Clin Pharmacol. 2000;50(4):285-295. doi:10.1046/j.1365-2125.2000.00261.x PubMedGoogle ScholarCrossref
    19.
    Kanoh  S, Rubin  BK.  Mechanisms of action and clinical application of macrolides as immunomodulatory medications.   Clin Microbiol Rev. 2010;23(3):590-615. doi:10.1128/CMR.00078-09 PubMedGoogle ScholarCrossref
    20.
    Shulman  ST, Bisno  AL, Clegg  HW,  et al; Infectious Diseases Society of America.  Clinical practice guideline for the diagnosis and management of group A streptococcal pharyngitis: 2012 update by the Infectious Diseases Society of America.   Clin Infect Dis. 2012;55(10):e86-e102. Published correction appears in Clin Infect Dis. 2014;58(10):1496. doi:10.1093/cid/cis629 PubMedGoogle ScholarCrossref
    21.
    Stevens  DL, Bisno  AL, Chambers  HF,  et al; Infectious Diseases Society of America.  Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the Infectious Diseases Society of America.   Clin Infect Dis. 2014;59(2):e10-e52. Published correction appears in Clin Infect Dis. 2015;60(9):1448. doi:10.1093/cid/ciu296 PubMedGoogle ScholarCrossref
    22.
    Karlowsky  JA, Lagacé-Wiens  PRS, Low  DE, Zhanel  GG.  Annual macrolide prescription rates and the emergence of macrolide resistance among Streptococcus pneumoniae in Canada from 1995 to 2005.   Int J Antimicrob Agents. 2009;34(4):375-379. doi:10.1016/j.ijantimicag.2009.05.008 PubMedGoogle ScholarCrossref
    23.
    Fralick  M, Juurlink  DN, Marras  T.  Bleeding associated with coadministration of rivaroxaban and clarithromycin.   CMAJ. 2016;188(9):669-672. doi:10.1503/cmaj.150580 PubMedGoogle ScholarCrossref
    24.
    Liesenfeld  KH, Lehr  T, Dansirikul  C,  et al.  Population pharmacokinetic analysis of the oral thrombin inhibitor dabigatran etexilate in patients with non-valvular atrial fibrillation from the RE-LY trial.   J Thromb Haemost. 2011;9(11):2168-2175. doi:10.1111/j.1538-7836.2011.04498.x PubMedGoogle ScholarCrossref
    25.
    Mueck  W, Kubitza  D, Becka  M.  Co-administration of rivaroxaban with drugs that share its elimination pathways: pharmacokinetic effects in healthy subjects.   Br J Clin Pharmacol. 2013;76(3):455-466. doi:10.1111/bcp.12075 PubMedGoogle ScholarCrossref
    26.
    Delavenne  X, Ollier  E, Basset  T,  et al.  A semi-mechanistic absorption model to evaluate drug-drug interaction with dabigatran: application with clarithromycin.   Br J Clin Pharmacol. 2013;76(1):107-113. doi:10.1111/bcp.12055 PubMedGoogle ScholarCrossref
    27.
    Vranckx  P, Valgimigli  M, Heidbuchel  H.  The significance of drug-drug and drug-food interactions of oral anticoagulation.   Arrhythm Electrophysiol Rev. 2018;7(1):55-61. doi:10.15420/aer.2017.50.1 PubMedGoogle ScholarCrossref
    28.
    Zhao  Y, Hu  Z-Y.  Physiologically based pharmacokinetic modelling and in vivo [I]/Ki accurately predict P-glycoprotein-mediated drug-drug interactions with dabigatran etexilate.   Br J Pharmacol. 2014;171(4):1043-1053. doi:10.1111/bph.12533 PubMedGoogle ScholarCrossref
    29.
    Garonzik  S, Byon  W, Myers  E, Li  X, Marchisin  D, Murthy  B.  The effects of clarithromycin on the pharmacokinetics of apixaban in healthy volunteers: a single-sequence crossover study.   Am J Cardiovasc Drugs. 2019;19(6):561-567. doi:10.1007/s40256-019-00348-2 PubMedGoogle ScholarCrossref
    30.
    Gouin-Thibault  I, Delavenne  X, Blanchard  A,  et al.  Interindividual variability in dabigatran and rivaroxaban exposure: contribution of ABCB1 genetic polymorphisms and interaction with clarithromycin.   J Thromb Haemost. 2017;15(2):273-283. doi:10.1111/jth.13577 PubMedGoogle ScholarCrossref
    31.
    Moj  D, Maas  H, Schaeftlein  A, Hanke  N, Gómez-Mantilla  JD, Lehr  T.  A comprehensive whole-body physiologically based pharmacokinetic model of dabigatran etexilate, dabigatran and dabigatran glucuronide in healthy adults and renally impaired patients.   Clin Pharmacokinet. 2019;58(12):1577-1593. doi:10.1007/s40262-019-00776-yPubMedGoogle ScholarCrossref
    32.
    Stöllberger  C, Finsterer  J.  Relevance of P-glycoprotein in stroke prevention with dabigatran, rivaroxaban, and apixaban.   Herz. 2015;40(2)(suppl 2):140-145. doi:10.1007/s00059-014-4188-9 PubMedGoogle ScholarCrossref
    33.
    Xu  R, Ge  W, Jiang  Q.  Application of physiologically based pharmacokinetic modeling to the prediction of drug-drug and drug-disease interactions for rivaroxaban.   Eur J Clin Pharmacol. 2018;74(6):755-765. doi:10.1007/s00228-018-2430-8 PubMedGoogle ScholarCrossref
    34.
    Fleet  JL, Shariff  SZ, Bailey  DG,  et al.  Comparing two types of macrolide antibiotics for the purpose of assessing population-based drug interactions.   BMJ Open. 2013;3(7):e002857. doi:10.1136/bmjopen-2013-002857 PubMedGoogle Scholar
    35.
    ELIQUIS (axpixaban). Product monograph. Accessed September 4, 2017. http://secure.healthlinks.net.au/content/bms/pi.cfm?product=bqpeliqu11112
    36.
    Praxada (dabigatran etexilate). Product monograph. Accessed August 26, 2017. https://www.ema.europa.eu/docs/en_GB/document_library/EPAR__Product_Information/human/000829/WC500041059.pdf
    38.
    Witt  DM, Nieuwlaat  R, Clark  NP,  et al.  American Society of Hematology 2018 guidelines for management of venous thromboembolism: optimal management of anticoagulation therapy.   Blood Adv. 2018;2(22):3257-3291. doi:10.1182/bloodadvances.2018024893 PubMedGoogle ScholarCrossref
    39.
    Kearon  C, Akl  EA, Ornelas  J,  et al.  Antithrombotic therapy for VTE: CHEST guideline and expert panel report.   Chest. 2016;149(2):315-352. doi:10.1016/j.chest.2015.11.026 PubMedGoogle ScholarCrossref
    40.
    Macle  L, Cairns  J, Leblanc  K,  et al; CCS Atrial Fibrillation Guidelines Committee.  2016 Focused update of the Canadian Cardiovascular Society guidelines for the management of atrial fibrillation.   Can J Cardiol. 2016;32(10):1170-1185. doi:10.1016/j.cjca.2016.07.591 PubMedGoogle ScholarCrossref
    41.
    Statistics Canada. Population estimates on July 1st, by age and sex. Accessed May 1, 2018. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1710000501
    42.
    Levy  AR, O’Brien  BJ, Sellors  C, Grootendorst  P, Willison  D.  Coding accuracy of administrative drug claims in the Ontario Drug Benefit database.   Can J Clin Pharmacol. 2003;10(2):67-71.PubMedGoogle Scholar
    43.
    Benchimol  EI, Smeeth  L, Guttmann  A,  et al; RECORD Working Committee.  The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement.   PLoS Med. 2015;12(10):e1001885. doi:10.1371/journal.pmed.1001885 PubMedGoogle Scholar
    44.
    Wahl  PM, Rodgers  K, Schneeweiss  S,  et al.  Validation of claims-based diagnostic and procedure codes for cardiovascular and gastrointestinal serious adverse events in a commercially-insured population.   Pharmacoepidemiol Drug Saf. 2010;19(6):596-603. doi:10.1002/pds.1924 PubMedGoogle ScholarCrossref
    45.
    Whitaker  HJ, Farrington  CP, Spiessens  B, Musonda  P.  Tutorial in biostatistics: the self-controlled case series method.   Stat Med. 2006;25(10):1768-1797. doi:10.1002/sim.2302 PubMedGoogle ScholarCrossref
    46.
    Othman  F, Crooks  CJ, Card  TR.  Community acquired pneumonia incidence before and after proton pump inhibitor prescription: population based study.   BMJ. 2016;355:i5813. doi:10.1136/bmj.i5813 PubMedGoogle ScholarCrossref
    47.
    Douglas  IJ, Evans  SJW, Hingorani  AD,  et al.  Clopidogrel and interaction with proton pump inhibitors: comparison between cohort and within person study designs.   BMJ. 2012;345:e4388. doi:10.1136/bmj.e4388 PubMedGoogle Scholar
    48.
    Masclee  GMC, Valkhoff  VE, Coloma  PM,  et al.  Risk of upper gastrointestinal bleeding from different drug combinations.   Gastroenterology. 2014;147(4):784-792.e9. doi:10.1053/j.gastro.2014.06.007 PubMedGoogle ScholarCrossref
    49.
    Suchard  MA, Zorych  I, Simpson  SE, Schuemie  MJ, Ryan  PB, Madigan  D.  Empirical performance of the self-controlled case series design: lessons for developing a risk identification and analysis system.   Drug Saf. 2013;36(1)(suppl 1):S83-S93. doi:10.1007/s40264-013-0100-4 PubMedGoogle ScholarCrossref
    50.
    Abraham  NS, DeSilva  R, Richardson  P.  Derivation and validation of an algorithm to identify helicobacter pylori infected patients using administrative data: 850.   Am J Gastroenterol. 2007;102:S431. doi:10.14309/00000434-200709002-00850 Google Scholar
    51.
    Austin  PC.  Using the standardized difference to compare the prevalence of a binary variable between two groups in observational research.   Commun Stat. 2009;38:1228-1234. doi:10.1080/03610910902859574 Google ScholarCrossref
    52.
    Chang  SH, Chou  IJ, Yeh  YH,  et al.  Association between use of non–vitamin k oral anticoagulants with and without concurrent medications and risk of major bleeding in nonvalvular atrial fibrillation.   JAMA. 2017;318(13):1250-1259. doi:10.1001/jama.2017.13883 PubMedGoogle ScholarCrossref
    53.
    Wang  N, Giblin  E, Rodgers  JE.  Drug interactions with non–vitamin K oral anticoagulants.   JAMA. 2018;319(8):830. doi:10.1001/jama.2017.20850 PubMedGoogle ScholarCrossref
    54.
    Li  Y, Dong  S, Soria-Saucedo  R.  Drug interactions with non–vitamin K oral anticoagulants.   JAMA. 2018;319(8):827-828. doi:10.1001/jama.2017.20830 PubMedGoogle ScholarCrossref
    55.
    Fralick  M, Juurlink  DN, Marras  T.  Bleeding associated with coadministration of rivaroxaban and clarithromycin.   CMAJ. 2016;188(9):669-672. doi:10.1503/cmaj.150580PubMedGoogle ScholarCrossref
    56.
    Gnoth  MJ, Buetehorn  U, Muenster  U, Schwarz  T, Sandmann  S.  In vitro and in vivo P-glycoprotein transport characteristics of rivaroxaban.   J Pharmacol Exp Ther. 2011;338(1):372-380. doi:10.1124/jpet.111.180240 PubMedGoogle ScholarCrossref
    57.
    Schulman  S, Beyth  RJ, Kearon  C, Levine  MN.  Hemorrhagic complications of anticoagulant and thrombolytic treatment: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition).   Chest. 2008;133(6)(suppl):257S-298S. doi:10.1378/chest.08-0674PubMedGoogle ScholarCrossref
    ×