Pages Navigation Menu

Meet 'em, greet 'em, treat 'em and street 'em

SGEM#139: One Thing Leads to Another – Idarucizumab for Dabigatran Reversal?

SGEM#139: One Thing Leads to Another – Idarucizumab for Dabigatran Reversal?

Podcast Link: SGEM139
Date: December 3rd, 2015

Guest Skeptics: Dr. Ryan Radecki is Clinical Practice Lead at Kaiser Permanente North West and Clinical Assistant Professor of Emergency Medicine – The University of Texas Medical School at Houston. He has a blog called Emergency Medicine Literature of Note.  

Case: A 67-year-old man presents with a history of atrial fibrillation and hypertension. He takes metoprolol 50mg twice daily and dabigatran 150mg twice daily. He had been having mild epigastric pain for about one week and had one episode of coffee ground emesis that night. He arrives via ambulance with a blood pressure of 120/70, heart rate of 74 beats per minute and O2 saturation of 98% on room air.

Background: Dabigatran is a non-vitamin K antagonist anticoagulant that works by inhibiting thrombin. It is approved for the prevention and treatment of venous thromboembolism. It is also approved treatment of non-valvular atrial fibrillation to prevent stroke.

Life threatening bleeding can occur while taking dabigatran. Unlike warfarin that can be reversed in a variety of ways, there has been no feasible or effective method for the adjunctive treatment of these life-threatening bleeds in patients using dabigatran. The only current method of removal, hemodialysis, is impractical in the vast majority of clinical situations.

There has been no feasible reversal agent until now. Boehringer Ingleheim the maker of dabigatran has developed a monoclonal antibody fragmin called idarucizumab.

The FDA approved idarucizumab in October as the first reversal agent for dabigatran. As far as we know it has not yet been approved for use in Canada

Clinical Question: Is idarucizumab safe and effective treatment adjunct for patients taking dabigatran with serious bleeding or who require an urgent procedure?

Reference: Pollack CV et al. Idarucizumab for Dabigatran Reversal. NEJM 2015

  • Population: Patients 18 years of age or older, reported to have been taking dabigatran
    • Group A: Patients with overt, uncontrollable, or life-threatening bleeding
    • Group B: Patients who required surgery or other invasive procedure that could not be delayed for at least 8 hours.
  • Intervention: All patients received 5g of intravenous idarucizumab in the form of two 2.5g boluses 15 minutes apart.
  • Comparison: None, this was a prospective observational trial.
  • Outcome:
    • Primary Outcome: Maximum percentage reversal of the anticoagulant effect of dabigatran within four hours assessed by measurement of dilute thrombin time or ecarin clotting time.
    • Secondary Outcome: Reduction in the concentration of unbound dabigatran, restoration of hemostasis and suspected thrombotic events or deaths by 90 days.

Author’s Conclusions: “Idarucizumab completely reversed the anticoagulant effect of dabigatran within minutes.”

checklistQuality Checklist for Observational Trials:

  1. Did the study address a clearly focused issue? Yes and No. To measure the percentage of dabigatran reversal, a study needs no more than healthy volunteers in a controlled environment. These patients were recruited in a clinical setting more suited to providing patient-oriented outcomes, rather than pharmacokinetic data.
  2. Did the authors use an appropriate method to answer their question? Yes. In answering their primary outcome, percentage of dabigatran reversal, they use surrogate measures of thrombin time and ecarin clotting time. These surrogates do seem to be correlated to unbound dabigatran, however.
  3. Was the cohort recruited in an acceptable way? No and Yes. The cohort was not recruited in an appropriate fashion with regards to their primary outcome. Regarding secondary clinical outcomes, the recruitment was reasonable and performed in an appropriate acute setting. However, there is no comparator with which to interpret any of the clinically relevant outcomes. The study protocol states each patient acts as their own control, but this is only relevant regarding pharmacokinetic data, and assumes the post-treatment baseline reflects the patients’ normal baseline pre-dabigatran initiation. Patients in Group A, enrolled with supposedly overt or life-threatening bleeding necessitating emergency reversal, required packed cell transfusion only approximately half the time. Lastly, the study does not report the number of patients screened for enrollment nor the reasons regarding their lack of inclusion, resulting in an inability to assess the magnitude of selection bias.
  4. Was the exposure accurately measured to minimize bias? No. Patients were enrolled if there was a report of dabigatran use, but the design of the study made it infeasible to monitor exposure. Then, patients were then excluded from efficacy analysis if their initial coagulation surrogates were normal. Of those enrolled, nearly a quarter had normal thrombin time or ecarin clotting time.
  5. Was the outcome accurately measured to minimize bias? Yes and No. The primary outcome of dabigatran reversal was measured via surrogate, and these are reliably sampled in the relevant population. Assessments of clinical outcomes were subjectively surveyed.
  6. Have the authors identified all important confounding factors? No. The authors do not report the presence or absence of other anti-platelet or anti-coagulants in patients taking dabigatran, although, it is probably reasonable to assume these medications should not have been concomitantly prescribed.
  7. Was the follow up of subjects complete enough? Yes. Follow-up of at least one month is a reasonable timeframe to capture potentially treatment-associated events.
  8. How precise are the results/is the estimate of risk? The measurement of coagulation reversal is reliably precise, and samples were taken frequently throughout the first 24 hours following treatment.
  9. Do you believe the results? Yes and No. The measurement of thrombin time, ecarin clotting time, and unbound dabigatran are reliable and believable. The clinically relevant patient-oriented outcomes and hemostasis outcomes are poorly described and cannot be interpreted in context.
  10. Can the results be applied to the local population? Unsure. It is unclear whether the results of this study can be reliably applied in clinical practice. It is likely idarucizumab binds a substantial portion of circulating dabigatran. The coagulation parameters in this study generally normalize, with some exceptions. There is little evidence, however, tying the use of idarucizumab and normalization of these coagulation parameters to clinically relevant cessation in bleeding.
  11. Do the results of this study fit with other available evidence? Yes. Regarding the primary outcome of pharmacokinetic and surrogate laboratory data. These are consistent with prior healthy volunteer data (Glund et al Lancet 2015). There is a paucity of relevant clinical data, which is not surprising given the stage of development of

Key Results: This is a preliminary report on an ongoing cohort study expected to recruit up to 300 patients at 400 centers in 38 countries. This report includes 90 patients with a median age of 76.5 years with 96% of patients being on dabigatran for atrial fibrillation.

  • Group A (Bleeding):
    • 51 patients
    • 18 intracranial hemorrhage, 20 gastrointestinal bleeds, 9 traumas and 11 other causes. This adds up to 58 rather than 51 because patients may have had more than one type of bleeding.
  • Group B (Surgery/Procedures):
    • 39 patients
    • Indication for surgery or urgent invasive procedure varied but the most common (8/39) was bone fractures.

Twenty-two patients were excluded from analysis because at study entry they had a dilute thrombin times that were within the normal range. This left only 68 patients in total with abnormal results, which were the focus of their analysis.

Primary outcome: Median maximum percentage reversal of the anticoagulant effect of dabigatran, assessed by both the dilute thrombin time and ecarin clotting time, within four hours was 100% (95% CI, 100% to 100%).

  • Secondary Outcomes:
    • Reduction in the concentration of unbound dabigatran: Of patients with dabigatran levels tracked, the unbound levels dropped to a level that produces little or no anticoagulant effect in all but one patient. A handful of patients in Group B had markedly elevated dabigatran levels, and these patients never entirely normalized. Dabigatran levels dropped to their nadir in these patients after completion of the infusion, but began rising again within an hour.
    • Restoration of hemostasis: Only 38 of the 51 patients in Group A could be evaluated for ongoing bleeding. The median time to reach clinical hemostasis was 11.4 hours. There were 36 out of 39 patients in Group B who underwent an urgent invasive procedure. Normal operative hemostasis was reported in 92% (33/36).
    • Deaths by 90 Days: There were 18 deaths in total with 9 in each of the two groups. Half of the deaths occurred over a week after enrolment.
    • Thrombotic Events: There were a total of five thrombotic events up to 90 days.
      • DVT/PE day 2, DVT/PE and left atrial thrombus day 9, DVT day 7, NSTEMI day 13 and ischemic CVA day 26

Screen Shot 2015-04-25 at 3.11.12 PM

It seems more and more patients are anti-coagulated these days with non-vitamin K antagonists. Some of these people are going to bleed and we need information on how to best treat them in the emergency department.

However, this is a disappointing publication regarding the safety and efficacy of idarucizumab for the reversal of anticoagulation in the setting of dabigatran use. It is hard to think of a good justification for the publication of these interim results.

They suggest this data reflect important insights into the clinical outcomes of patients treated with idarucizumab. However, they acknowledge a major limitation of their study was the lack of a control group.

The authors state it was unethical to perform a placebo-controlled study in the setting of life-threatening bleeding risk; this stance assumes the untested therapy is actually clinically effective and safe in a real-world setting without ever being tested.

While most reported coagulation parameters normalized within minutes, and most unbound dabigatran levels dropped to a level below detection, the median investigator-reported to hemostasis was not reported for over 11 hours in patients with overt bleeding.

It is not clear whether this reflects a lack of true clinical efficacy, or whether confounders related to individual patient condition resulted in the long delay to hemostasis. Regardless, the publication of interim results seems unwarranted given the content presented.

Such a presentation might be justified based on a safety evaluation, given the fast-track approval of this antidote, but, a cohort size of 90 is almost certainly too small to detect any pattern of adverse events.

The report of peri-procedural hemostasis is also questionable, particularly considering the procedure list. While a handful of significant surgical cases are included, several interventions are typically minimally invasive laproscopic procedures, and the list of “surgeries” also includes cutaneous abscess I&D, lumbar puncture, placement dialysis catheters, and placement of ureteral stent. No reliable conclusion can truly be drawn regarding the clinical effectiveness of idarucizumab for procedural hemostasis for urgent procedures.

Another concern regarding the trial was the 22 patients who received the drug and were subsequently found to have normal clotting levels. This means almost 25% of patients receiving a drug that was not necessary and exposed them to potential risk of adverse reactions.

Without a routinely available, rapidly feasible method for detection of therapeutic dabigatran levels, the widespread clinical usage could be fraught with waste. Especially considering the expected cost of each dose of idarucizumab is ~$3,400 USD.

Finally, it should be mention this trial was funded by the manufacturer and sponsor, Boehringer Ingelheim who also had a role in the study design. Many authors, as well as the author of the accompanying editorial, report relevant financial conflicts of interest.

Many trials are sponsored by the pharmaceutical industry. This does not make the results wrong but it probably should make us more skeptical of the results.

Dr. Ryan Radecki

Dr. Ryan Radecki

Comment on authors conclusion compared to SGEM Conclusion: We agree idarucizumab rapidly normalizes abnormal coagulation parameters associated with dabigatran use, but its suspected clinical effectiveness and safety cannot be confirmed at this time.

SGEM Bottom Line: Idarucizumab is here (USA) and probably works but its patient oriented efficacy and safety are still pending.

Case Resolution: You establish two large bore intravenous lines and group and cross him for four units of blood. Laboratory studies are sent off stat and gastroenterology is called to the emergency department. He is taken to the procedure suite and found to have a bleeding duodenal ulcer.

Clinical Application: It is too early to know how to clinically apply this preliminary research.

What do I tell my patient? You are having a serious bleed. It is made worse because of the blood thinner you take to prevent strokes. We are getting some blood for you. The gastroenterologist is on their way to take you to the operating room. If you become unstable I can give you a new drug just approved that seems effective and seems to works very quickly. However, we don’t have much information on how well it works and there may be safety issues we have not yet identified.

Keener Kontest: Last weeks’ winner was Garreth Debiegun an Assistant Clinical Professor Maine Medical Center & Tufts University School of Medicine. Garreth knew it is harder to get an effective regional nerve block for a more proximal hip fracture compared to a more distal hip fracture because the hip is innervated by the femoral nerve and the obturator nerve but also by a few others (superior gluteal and quadratis femoris).  Given this fractures up in and at the joint capsule are only partially blocked by a 3-in-1 (femoral, lateral femoral cutaneous & obturator) there is likely to be some residual pain from the non-blocked nerves.

Listen to the podcast for the question this week. Send your answer to with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.

FOAM logoOther FOAMed resources:

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.