ACLS Drugs: The Verdict

New Picture (7)

Do ACLS medications make any difference in cardiac arrest?

The American Heart Association rates drugs and interventions according to the following scheme:

Class I
Procedure/treatment or diagnostic test/assessment should be performed/administered.

Class IIa
It is reasonable to perform procedure/administer treatment or perform diagnostic test/ assessment.

Class IIb
Benefit >Risk.
Procedure/treatment or diagnostic test/assessment may be considered

Class III
Risk> Benefit.
Procedure/treatment or diagnostic test/assessment should not be performed/administered. It is not helpful and may be harmful.

Class Indeterminate
• Research just getting started• Continuing area of research• No recommendations until further research (e.g. cannot recommend for or against).

According to the 2005 AHA Guidelines for managing cardiac arrest:

Magnesium is IIa for Torsades.

Epinephrine and Amiodarone are class IIb.

Vasopressin, Atropine and Lidocaine are class Indeterminate.

In other words, according to the AHA, we have no idea if vasopressin, atropine and Lidocaine work, epi and amiodarone might work, and magnesium likely works. Not the best endorsement.

New Study

Now a new study has come out that makes the best attempt yet to answer this crucial question of whether or not ACLS cardiac arrest drugs work, as well as another question: “If the drugs aren’t doing any good, is it the drugs’ fault or perhaps the fault of poor CPR?”

Here’s the study:

Intravenous Drug Administration During Out-of-Hospital Cardiac Arrest: A Randomized Trial

It appeared is the November 25, 2009 issue of the Journal of the American Medical Association.

This was a prospective, randomized trial that took place in Oslo, Norway between May 1, 2003 and April 28, 2008. It involved patients 18 and over who suffered out-of-hospital nontraumatic cardiac arrest.

When medics arrived on scene and they opened a sealed envelope that instructed them to either start IV access and proceed with ACLS drugs or do ACLS without drugs and IV access. If ROSC (return of spontaneous circulation) occurred in the non-IV group, they were instructed to wait five minutes before starting an IV and proceeding with post-arrest care.

Medics did 3 minutes of CPR before shocking v-fib and 3 minutes between unsuccessful shocks per European guidelines.

Endotracheal intubation was the prefered airway method.

851 Patients

418 patients received IV meds, 433 did not. The primary outcome was survival to hospital discharge, while secondary outcomes such as hospital admission were also studied.

Patients were excluded if they arrested in the presence of EMS crews.

Personnel carried LifePak 12s capable of recording the quality and quantity of CPR.

Both groups had “adequate CPR” with compression and ventilation rates within the AHA guidelines. All resuscitated patient received therapeutic hypothermia.

Survival to Hospital Discharge

In the IV group 10.5% survived to hospital discharge versus 9.2% in the non-IV group.

Survival with favorable neurologic outcome was 9.8% versus 8.1% in favor of the IV patients.

ROSC was 40% in the IV group, 25% in the non-IV group.

43% IV group to 29% non-IV group in hospital admission

In patients with VT/VF as the presenting rhythm there was no difference between the groups.

In nonVF/VT, ROSC was 3 times higher in the IV group, but there was no difference in the survival rate because patients without the IV did 3 times better in the ICU than those who received the IV and ACLS drugs.

Other Results

Patients with VF/VT had 10 fold likelihood of survival.

Patients with witnessed arrest had a two fold increase in survival.

Long-terms survival odds decreased by 17% for every minute without CPR

While the IV group did slightly better than the non-IV group, statistically, it was deemed insignificant due to the sample size. The trial would have had to have included 1400 for such a difference to be considered significant.

Bottom line

ACLS drugs make no difference in long-term survival.

IV meds do not delay or affect the quality of CPR.

Patients in VFib/VT may do better without ACLS drugs.

While ACLS drugs can help return circulation to patients in PEA and asystole, they may ultimately be toxic to these patients.

Thought for the Future

Either cardiac arrests patients outside of VF/VT are gennerally not saveable or we need better ACLS drugs.

Personal Comment:

This study squares with my observations. I have gotten ROSC many times in nonVF/VT patients, and that ROSC almost always comes shortly after I have given them IV epi. (Thus the phrase “That’s the epi talking.” )

Many times (in the old days) when I started working a patient in the house and gave epi down the tube, I would get nothing, but then once removed to the ambulance for transport (as everyone was transported then), I would get a peripheral IV, and give epi and suddenly have ROSC.

Earlier this week I had my first cardiac arrest of the new year, 80-year old man collapses in front of his wife. First responders start CPR. Patient is apneic with no pulse on my arrival. But the monitor shows a sinus brady at 40. I intubate the patient with an initial ETCO2 of 20. With IV epi, the ETCO2 almost immediately goes up to 70 — an indication of ROSC. It then stabilizes in the 35-40 range. We get a bounding pulse with a BP of 110/60. We start the hypothermia protocol. 10 minutes later, I see the ETCO2 start to steadily drop — all the way down to 20. BP is 58/30. Start the dopamine, right away the ETCO2 goes back to 40.* Bounding pulses again. We arrive at the ED with a BP of 120/70. I’m feeling that this could be a save.

An hour later, they are doing CPR on the patient again. When I check back, they tell me he is up in the ICU. He is septic and they have been having a hard time maintaining his BP, giving him 5 liters of saline and using multiple pressors.

Two days later he is in the obits.

Another one bites the dust.

With the exception of patients suspected to be victims of respiratory arrest, all my asystole/PEA ROSC patients have died in the ED or ICU.

The few true cardiac arrest survivors (nonrespiratory induced) I have had in my career have almost all been patients in their 50s and 60s who collapsed in a public place and recieved early CPR, and who (finding them to be in VFIB) I shocked on arrival.

* For an explanation of ETC02 in cardiac arrest see:

10 Things Every Paramedic Should Know About Capnography

New Picture


  • Michael Dulitz EMT-B says:

    What type of change is this going to have on ACLS in the foreseeable future?

  • medicscribe says:

    The new American Heart ACLS guidelines should be coming out this October. I am going to be posting on this topic in the next week. No one knows what the guidelines will be yet, but we do know there will be significant changes. My post will include a link to some of the scientific reviewer work sheets that are just now being posted that may give a hint at what the changes might be.


  • totwtytr says:

    “What type of change is this going to have on ACLS in the foreseeable future?”

    Likely none. It’s incredibly rare for a single study to have any effect on practice. It will take at least one more study that reproduces the results using the same parameters before the committees that decide these things will consider changes.

    The study is more confirmatory than ground breaking. We’ve known for more than ten years that generally only patients in VF/VT survive cardiac arrest. There are exceptions, but they are just that. We’ve suspected for at least ten years that most ACLS drugs are, not to be too blunt, useless. If the patient doesn’t have ROSC with CPR and electricity, their survival is unlikely.

    Remember that survival is defined as the person leaving the hospital neurologically intact. Just restoring a heart beat is pretty useless if the patient has no mental function.

    I expect more emphasis on CPR and early defibrillation in the next ACLS standards. I’d guess that we might see therapeutic hypothermia in the protocols. I don’t know what drugs we might see added, or removed for that matter.

    • medicscribe says:

      I guess I would agree that they are likely to keep things somewhat the same with regard to ACLS drugs in cardiac arrest. From reading the AHA worksheet below, their view (while not final yet) seems to be that short-term gain is better than no improvement regardless of no difference in long-term outcome.

      As I mentioned I will be addressing the 2010 AHA guidelines and process in a post later this week.

      Here is an example of an AHA worksheet on the issue of vasopressors and cardiac arrest in which they mention the study cited in my post:

      AHA Vasopressor Worksheet

      “Question: In adult patients in cardiac arrest (asystole, pulseless electrical activity, pulseless VT and VF) (prehospital [OHCA], in-hospital [IHCA]) (P), does the use of vasopressors (epinephrine, norepinephrine, others) or combination of vasopressors (I) compared with not using drugs (or a standard drug regimen) (C), improve outcomes (eg. ROSC, survival) (O).

      Here is an excerpt:

      1. Any Vasopressor vs. Placebo:
      The most significant work related to this topic has just recently been published (Olasveengen, 2009, 2222). In this prospective study, the authors compare outcomes related to groups receiving intravenous drug administration vs. no intravenous drug administration for out-of-hospital cardiac arrest. They demonstrate improved short term outcomes in the intravenous group (ROSC, survival to hospital admission, survival to ICU admission) when compared to the no intravenous group. However, no significant differences are noted between the groups with regards to survival to hospital discharge or good neurologic outcomes. The individual contribution of vasopressors (epinephrine) to these survival data is not elucidated.

      There is one prospective randomized control trial comparing high dose (HDE) and standard dose epinephrine (SDE) to placebo for the treatment of cardiac arrest (Woodhouse, 1995, 243). In the Woodhouse trial, patients were blindly randomized to receive HDE (10mg) or placebo (saline) during the first 5-10 minutes of arrest, followed by 1mg aliquots of epinephrine as per standard ACLS protocols . As there were many protocol violations in which 1mg epinephrine doses were given in place of the blinded drugs (likely due to practioners fear of the patient receiving placebo), a third study arm was created post-hoc to include this group for comparison. The conclusions state that there were no differences in immediate survival or survival to hospital discharge
      between treatment groups. However, the study is methodologically flawed and is of overall poor quality. At the conclusion of the study, the patients had not been systematically randomized to a treatment groups, their treatment by clinicians was not truly blinded, and the groups were not treated equally.

      There is a retrospective review comparing VF patients treated by EMS personnel authorized to administer epinephrine during ACLS to another group of VF patients treated by EMS personnel unauthorized to use epinephrine (Herlitz, 1995, 195) (N=1203). Those patients with sustained VF after 3 shocks and then treated with epinephrine were more likely to achieve ROSC and survival to admission. There was no difference in rate of hospital discharge in these patients (12/206 vs. 17/145). This conclusion was similar in patients who converted from VF to PEA or asystole (9/246 vs. 6/671). The LOE is fair, with poor confounder control.

      One other study exists which examines patient populations before and after the introduction of epinephrine for use by the EMS system in Taiwan (Ong, 2007, 635) (N=1296). The conclusions demonstrated no improvement in ROSC, survival to hospital admission, or survival to discharge with the use of epinephrine. However, the study has methodologic issues including untreated
      patients in the treatment arm and is ranked LOE3-poor due to poor confounder control and group selection.

      In summary, the data on which to draw conclusions related to the use of vasopressors vs. placebo during cardiopulmonary arrest is limited. The most recent and best performed study demonstrates significant improvement in short term survival statistics when using ACLS drugs, including epinephrine, but no improvement to hospital discharge or neurologic outcomes.”

  • So if they don’t work then why do we give them? My limited perspective of 5 years in EMS tells me that electricity and immediacy is the foundation of cardiac survival.

    Further why I am constantly told your iv percentage is this, iv this iv that!

    Why aren’t my reports reviewed for correctness of assessment and finding? Why can’t I find out about patients once they are in the hospital?

    Don’t know if this all ties together but that is where I am at.

  • medicscribe says:

    Great questions.

    First on the IV percentages, that’s better than some places. One place around here can tell you your driving score, the percentage of signatures you get on your billing and your attendence record, but nothing about your patient care.

    Why can’t you find out about how your patient made out? This is so important. QA has to tie the run form to the patient outcome. As a clinical coordinator now, I am working hard to let people know the outcomes of their patients as I have access to that information now at my hospital and can give it out in the QA process.

    QA around here is done by the sponsor hospital (each ALS service in our state must by regulation be sponsored by a hospital), but I would like to see it moved to recieving hospitals. You bring a patient to a hospital, that hospital provides the QA.

    As far as ACLS drugs, like many things in medicine, many treatments started as conjecture, but now are being scrutinized by evidence-based medicine. I have often thought the reason ACLS drugs didn’t work in the outcome that matters most – discharge with neurological function– is that maybe it wasn’t the fault of the drugs, but poor CPR. New evidence is showing that is not the case.

    My guess is that they will stay as long as they don’t show real harm until something better comes along. I guess we’ll find out in October.

    Thanks again for the comments,


Leave a Reply

Your email address will not be published. Required fields are marked *