Wednesday, August 21, 2013

Kcentra Trial Review

Very quietly, an article that many EM clinicians have been waiting for was recently e-published in the journal Circulation and with no fanfare and no fancy acronym in the title. I am referring to the new [to the United States] and hot drug that has gotten a lot of press in terms of its potential for use in the emergency department that has been cited in the package insert since it first was approved by the FDA in April this past year: Kcentra.

The study design of this non-inferiority trial is very effectively laid out by the investigators, and patients with an acute major bleed with an INR greater than or equal to 2 across nearly 40 sites in the United States and Europe were assigned to treatment with either 4-factor PCC (Kcentra, although the authors do refer to it throughout the study by Beriplex, the brand name of the product outside the United States) or fresh frozen plasma (FFP). The investigators used a fairly sophisticated randomization method to ensure that treatment arms were balanced in terms of the number of patients and the type and location of the bleed. The authors aimed to determine the effects of these treatments with two primary endpoints: (1) hemostatic efficacy from the start of the infusion for a period of 24 hours following the infusion and (2) rapid reduction in INR to less than or equal to 1.3 one-half hour following completion of the infusion. The investigators also analyzed a number of secondary endpoints, including plasma levels of vitamin K dependent clotting factors, time to INR correction, proportion of patients requiring PRBC transfusion, all-cause mortality at 45 days, and safety and tolerability of the treatments. 

The dose of 4-factor PCC used in this study was based on initial INR and body weight, and is comparable to the dosing recommendations of the product, as provided here. Interestingly, the dose of FFP utilized was also based on these parameters as well, and patients only received what is typically deemed to be the standard dose of FFP in the setting of life-threatening bleed if the INR was greater than 6; otherwise a dose of 10 or 12 mL/kg was administered in patients had an INR between 2 to less than 4 and an INR between 4 to less than 6, respectively. In addition, all patients received intravenous phytonadione (vitamin K) at a dose of 5 to 10 mg as a slow infusion. 

An efficacy scale for hemostasis was developed that ranged from “excellent” to “poor/none”, with predefined laboratory and other clinical criteria related to the type of bleed being evaluated to determine if this outcome measure was met. Patients were deemed to have failed therapy and have a rating of “poor/none” if supplementary treatment was necessary to control the bleed and/or if PRBC transfusion was necessary within 24 hours of treatment. 

216 patients were randomized to treatment, with 103 patients in the 4-factor PCC arm and 109 patients in the FFP arm. Nearly half of all patients were female, with an average age of nearly 70 years. 70% of all patients resided in the United States. Median baseline INR in the 4-factor PCC arm was 3.90 versus 3.60 in the FFP arm. Nearly 65% of all bleeds were gastrointestinal and/or other non-visible bleeds while approximately 12% of all patients had intracranial hemorrhage (ICH). The majority patients were on therapy with a vitamin K antagonist for arrhythmia (57.1 and 51% respectively) with an average duration of treatment of nearly two years. The five most commonly reported comorbidities experienced by patients in the study included hypertension, atrial fibrillation, anemia, coronary artery disease, and congestive heart failure. The investigators do report that hematologic values at baseline were comparable at baseline, although this data is not presented. Nearly half of all patients did require at least one transfusion of PRBCs in both treatment arms, with an average of 1.4 and 1.2 units required for those receiving 4-factor PCC and FFP, respectively. 

In terms of hemostasis, 72.4% of patients achieved “excellent” or “good” hemostasis with 4-factor PCC compared to 65.4% of patients who received FFP, which deemed 4-factor PCC to be non-inferior to FFP (95% CI -5.8 to 19.9 [lower limit: greater than -10%]). In a subanalysis based on bleeding type, there were no statistically significant differences in hemostatic efficacy between treatments. The effect of hemostasis was more pronounced with 4-factor PCC versus FFP in those patients with visible and musculoskeletal bleeding (82.6% versus 50% at four hours) compared to gastrointestinal, intracranial, and non-visible bleeding (69.3% versus 71.1% at 24 hours). Interestingly enough, in those patients with ICH, 41.7% of patients in the 4-factor PCC arm achieved “excellent” or “good” hemostasis, while 58.3% of patients in the FFP arm reached these same endpoints.

62.2% of patients who received 4-factor PCC demonstrated rapid INR correction to less than or equal to 1.3 one-half hour after completion of the infusion versus 9.6% of patients who received FFP. The investigators do go on to further point out that one hour after the initiation of the infusion, 69% of patients reached the an INR of ≤ 1.3 with 4-factor PCC compared to none in the FFP arm, and at 24 hours after completion of the infusion, 88% of patients in the 4-factor PCC arm achieved this goal compared to 58% of patients in the FFP arm. After conducting a post-hoc analysis, the investigators determined that 4-factor PCC demonstrated an INR reduction that was superior to FFP.

The investigators also measured the plasma levels of vitamin K dependent clotting factors and proteins C and S and found higher levels of these factors in patients receiving 4-factor PCC compared to FFP at 30 minutes post-infusion. However, between 3 and 24 hours following infusion, levels of these factors in those receiving FFP resembled those levels observed in patients who received 4-factor PCC.

In terms of safety, treatment-related adverse events occurred in 10 patients in the 4-factor PCC arm versus 23 patients in the FFP arm. 8 patients in the 4-factor PCC arm and 4 patients in the FFP arm experienced thromboembolic adverse events, and four of these events were deemed to be related to treatment in the 4-factor PCC arm compared to 3 in the FFP arm. 32 patients who received 4-factor PCC had serious adverse events, with two patients experiencing treatment-related events (deep vein thrombosis [1] and ischemic stroke [1]). In the FFP arm, 26 patients experienced a serious adverse event, with myocardial ischemia [2], fluid overload [1], and respiratory failure [1] deemed as being related to treatment. 

At 45 days post infusion, 10 patients who received 4-factor PCC (9.7%) died compared to 5 patients in the FFP arm (4.6%). Six deaths in the 4-factor PCC arm occurred within 30 days of treatment, and all 5 in the FFP arm occurred within 30 days. The causes of death in those patients treated with 4-factor PCC were increase in ICH, worsening of cardiogenic heart failure, sudden death (deemed by the blinded safety adjudication board to be possibly related to treatment and occurring seven days after treatment), sepsis, acute renal failure, pancreatic cancer, cardiopulmonary arrest, respiratory failure, stage IV lung cancer, and myocardial infarction. 8 of the patients in the 4-factor PCC group were placed on comfort care prior to death. In those who died following treatment of FFP, none were determined to be related to treatment and four of these patients were placed on comfort care.

The authors do conclude that 4-factor PCC is a reasonable alternative to FFP in the management of acute major bleeding secondary to vitamin K antagonists.

For some, these results may seem to be a bit of a letdown after waiting for quite some time for this study to be published. Am I surprised and/or impressed by the results? Not really, for a number of reasons. 

We already know PCC does work relatively quickly (within half an hour) in reducing INR. We also know that (1) FFP will really only effectively decrease INR to 1.6; and (2) the onset of activity of FFP can range anywhere from 4 to 6 hours. So is a goal INR of less than or equal to 1.3 for this study really and truly achievable with FFP in the real world? 

In addition, although the rate of INR reduction may be important, what is more worthy is the ability for the treatment to reflect the values in the patient (i.e. achievement of hemostasis). Yes, more patients in the 4-factor PCC arm did achieve a rapid INR reduction compared to those who received FFP, but the investigators of the trial were not able to demonstrate superiority with hemostatic efficacy. The mantra of “treat the patient, not the number” can be modified in this case to “treat the patient, not only the number.”

What this study does add that not many studies have previously done is an evaluation of hemostatic efficacy with stratification of patients based on the type of bleed.  One interesting observation that can be made from this study is that for life-threatening non-visible bleeding (gastrointestinal and ICH), nearly the same rate of hemostasis was collectively achieved with 4-factor PCC and FFP at 24 hours; and 30% of patients still did not effectively achieve hemostasis. We are talking 3 out of every 10 patients here- not an insignificant proportion, especially since most of the patients in this study happened to present with life-threatening gastrointestinal bleed. These results may demonstrate that a risk-benefit analysis would need to be taken into account if considering the use of Kcentra for off-label purposes in the management of such bleeds secondary to the novel oral anticoagulants (dabigatran, rivaroxaban, and apixaban). Hopefully, we will provided with more guidance on this as more studies are conducted to evaluate the efficacy of 4-factor PCC to manage life-threatening bleeding secondary to these agents.

Sarode R, Milling TJ, Refaai MA, et al. Efficacy and safety of a four-factor prothrombin complex concentrate (4F-PCC) in patients on vitamin K antagonists presenting with major bleeding: a randomized, plasma-controlled, phase IIIb study. Circulation 2013 Aug 9. [Epub ahead of print]

Wednesday, August 14, 2013

Kcentra Administration

After the inclusion of Kcentra to the armamentarium for acute reversal of anticoagulation, a few practical issues have come up that are worth sharing.

1) Product contents:
Depending on the products previously used at a given institution (here it was Profilnine), there are several differences in the contents compared to Kcentra. One particular of note is that Kcentra contains approximately 40 units of heparin for every 500 FIX units.  Therefore a patient could receive anywhere from 200 to 400 units of heparin per dose of Kcentra.  Because of this, there is a risk of exacerbation/reactivation of HIT. In fact, Kcentra is contraindicated in patients with history of HIT (per PI).  Thus, the prospect of actually removing an item from formulary when adding an alternative may not be possible, since keeping Profilnine is necessary for patients with history of HIT. Alternatively, FEIBA contains no heparin, but may have higher relative risk of thrombosis (theoretically).

2) Administration and dilution:
Kcentra is supplied as a lyophilized powder which requires reconstitution with SWFI.   However, no further dilution is recommended. Normally not a problem for small doses (less than 50 mL volume) can be infused via syringe pump. But larger volumes require empty evacuated containers for administration. Conceptually, this may not be an issue, but practically, it is. In a hospital pharmacy, trying to find an evac bag (which are on shortage), or reconstituting the drug at the bedside (where evac bags do not exist), administration issues/delays can occur. There is no information I can find as to why it cannot be diluted further in say, 100mL of NS, but potential studies are in the works.

3) INR recheck:
It seems that the critical INR recheck 15 min after the end of the infusion is simply not being done. Partially as a result of poor education for all parties involved (pharmacy, nursing, PA, physician, lab), and partially because of poor communication at the time of the order to ensure that, "no this isnt a duplicate order, we need another INR 15min after the infusion."

Lastly, the still "unpublished" data comparing Kcentra to FFP which led to the FDA approval is still... unpublished. I am growing more and more concerned as to why, and fearing there is more to the significant 6.1% increase in the incidence of death in the Kcentra arm vs FFP...

More to come on the evolution of the Kcentra, PCC, FEIBA, Novoseven saga.

Thursday, August 1, 2013

Paging Goldilocks to the ER: Acute Pain Management in the Emergency Department, Part I

You have a patient who presents to the emergency department with what she describes as "excruciating pain" that radiates to the lower back. The pain is sudden in onset and when you ask her to rate the pain on a scale from 1 to 10, with 10 being the worst pain, the patient responds emphatically with "15."


And so you decide to order morphine 4 mg IV push as a one-time dose and conduct further workup. The patient is now calm and subdued, relieved to have the ED physician finally understand and empathize enough to order an intravenous pain medication. After all, isn't that the expectation of patients when they do present to the ED?

Until 15 minutes later. You hear screams coming from the patient's room, and the nurse comes hurrying over and tells you that the patient is writhing and screaming in pain, rating it as a 12 on the pain scale. You reevaluate the patient and she tells you, "Doc, that medication did NOTHING for me. It was like 'water' in the wind. I am still in pain!" And so you go ahead and order another dose of morphine 4 mg IV push.

Another 15 minutes pass by, and you reevaluate the patient, asking her to rate her pain. She looks up at you and states, "Doc, I am still in pain. I am trying to not make a fuss about it, but I cannot help it. It's now 9 out of 10. Do you have anything stronger that you can give me? Please? That second dose did not help all that much."

Hearing this, you inform the nurse to administer a dose of hydromorphone 1 mg IV push to the patient. After 20 minutes, you reevaluate the patient, and ask her about her pain. She states, "That medication that was just given to me finally did the trick. I am in no pain. Why didn't you give that to me from the beginning?"

Why not, indeed?

This is all too common of a scenario that we encounter in the emergency department. Acute pain management in the emergency department is a beast. I often find two extremes. We may underdose patients with opioid analgesics, similar to the scenario described above, that it takes at least three or more attempts to get the pain under control. The other extreme is that we end up overshooting with the initial dose and/or with "dose-stacking" of medications, patients may experience oxygen desaturation and respiratory depression, often requiring measures and agents (such as naloxone) to reverse these effects.

Much of it has to do with a matter of convenience...or rather, inconvenience. Intravenous opioid analgesics such as morphine and hydromorphone are stocked in automated dispensing machines (ADMs) in the emergency department. With an attempt to streamline and minimize the different concentrations of these agents in these machines to prevent medication errors, some emergency departments may stock a single concentration of each of these agents. Oftentimes, you will find the following concentrations of morphine and hydromorphone respectively stocked in ADMs:


Where's the "Easy" button?

With these prefilled carpujects containing these medications, it makes it all too easy for clinicians to become reliant on these concentrations as initial doses for our patients.

The truth is, this is entirely not appropriate. Believe it or not, these medications, in both children and adults, are actually dosed based on weight. Here are the doses:


So, unless your patient happens to weigh in the neighborhood of 40 kg, that initial dose of 4 mg of IV morphine is probably not going to cut it for most of your adult patients.

In addition, there is a dosing conversion between IV morphine and IV hydromorphone that can be surmised from the table above:

10 mg morphine IV = 1.5 mg hydromorphone IV

To make it a little easier to think about, this can be rounded to the following:

7 mg morphine IV = 1 mg hydromorphone IV

In the instances where I have had appropriate weight-based initial doses of these agents being ordered by EM physicians and residents, I have had nurses come to me in a panic and state, "Can you believe Dr. [insert last name here] ordered 8 mg of morphine?! That's way too much!!" While I calmly inform them that this dose is indeed appropriate, given the patient's weight, and putting it into the perspective of the dose conversion and the fact that it is nearly equivalent to a dose of hydromorphone 1 mg IV (of which there happens to be no qualms of ordering or administering), there still exists some fear of administering such a "high dose" of morphine.

So the question now is:
How can we better optimize pain control for our patients in the emergency department? In other words, can we find a happy medium for our patients when it comes to acute pain management, as Goldilocks did in the classic tale?

The answer:
Look out for Part II, coming in the next post, which will consist of:
  • An evaluation of various studies that have investigated different methods of acute pain management using opioid analgesics for patients in the emergency department.
  • A discussion of the ways we can apply and incorporate the findings of these studies to our practice.

Featured Post

Giapreza - A Closer Look at the Pharmacology of Ang2

Angiotensin-II (Ang2) is now an FDA approved vasopressor. With this new addition to the available options, experts are combing over the ...