By James R. Roberts, MD
It may sound like medical heresy, but the decades-long use of ACLS drugs does not appear to favorably affect the outcome of out-of-hospital cardiac arrest. Their use, in fact, may be detrimental to long-term survival. Despite a growing number of studies supporting this concept, prehospital ACLS drugs are still recommended by national societies and guidelines, and they are frequently, if not universally, used in our EMS system. The number of ACLS recommended drugs has decreased over the years, specifically eschewing calcium, atropine, magnesium, lidocaine, and bicarbonate, but the current CPR guidelines from the American Heart Association (2010) still recommend the use of epinephrine/vasopressin or amiodarone as empiric therapy for cardiac arrest. EMS personnel will tell you that ACLS drugs result in an increase in return of spontaneous circulation in the field, but this is not a parameter associated with better long-term survival after hospital arrival.
This month’s column discusses two articles, both quite well done, that indicate that intravenous drug administration during out-of-hospital cardiac arrest has no proven benefit, and their use may actually be detrimental. This may seem like an outrageous contradiction to current practice, but the widespread use of ACLS drugs was adopted on theoretical grounds or animal studies without any controlled studies or verified supporting data. Apparently, it’s easier to get a drug into ACLS guidelines than it is to get it out.
Intravenous Drug Administration during Out-of-Hospital Cardiac Arrest: A Randomized Trial
Olasveengen TM, Sunde K, et al.
This study prefaces its data by emphasizing a poorly disseminated concept: Intravenous drug administration in cardiac arrest is included in ACLS guidelines despite the lack of evidence of improved outcomes. Millions of patients worldwide have received and continue to receive ACLS medications, particularly epinephrine, with little or no evidence of improved survival to hospital discharge. It is known that epinephrine will rapidly increase cerebral and coronary perfusion by redirecting peripheral blood flow, but the initial drug benefit may quickly wane, resulting in an increased concern for myocardial dysfunction and disturbed or truncated cerebral microcirculation after the cardiac arrest patient has been resuscitated. The relatively low chance of survival among patients receiving epinephrine after out-of-hospital cardiac arrest was noted more than 10 years ago. (Resuscitation 2002;54:37.)
As reported in nonrandomized studies and similar to other ACLS caveats in the literature, resuscitation drugs were never tested in any randomized controlled human studies or involved in any non-drug comparison group before they were incorporated into ACLS protocols. It may also be possible that inadequate CPR quality, often associated with stopping pristine CPR during drug administration, may also influence long-term outcome. Using data from this prospective randomized controlled trial of drug administration during out-of-hospital cardiac arrest, the authors compared outcomes for patients receiving standard ACLS interventions with patients receiving similar ACLS but not intravenous drug administration.
The study was conducted in Oslo, Norway, and such a study would not likely be allowed in the United States currently because of the strong support for ACLS drugs for cardiac arrest. Out-of-hospital resuscitation in Norway was relatively sophisticated, often run by two paramedics and an anesthesiologist. CPR was performed in those with ventricular fibrillation for three minutes prior to the administration of electrical shock, and all patients were intubated during the study. Following hospital arrival post-resuscitation, therapeutic hypothermia was instituted regardless of the initial rhythm or arrest etiology. Those with a prehospital 12-lead EKG that suggested the possible need for coronary artery intervention were directly transferred for angiographic intervention.
Patients were randomly placed into the intravenous versus no intravenous drug group just before resuscitation. Exclusion criteria included those who had an arrest witnessed by the ambulance crew, those whose resuscitation was initiated by physicians outside the ambulance team, or cardiac arrest because of asthma or anaphylactic shock. The primary outcome was survival to hospital discharge. A secondary outcome was one-year survival with a favorable neurological status.
The study included 851 randomized patients collected from 2003 to 2008. The rate of survival to hospital discharge was 10.5 percent for the intravenous drug and 9.2 percent for the non-drug administration group (P=0.61). Survival with a favorable neurologic outcome was likewise not statically significant, 9.8 percent for the drug treated group, 8.1 percent for the non-drug group (P=0.45). The rate of initial return of spontaneous circulation was statistically significant at 32 percent for the drug-treated group versus 21 percent for the non-drug group (P=0.001). The authors said CPR quality was comparable and within recommended guidelines for both groups. The authors also attempted to adjust for the presence or absence of ventricular fibrillation, the response time interval, witnessed versus unwitnessed arrest, or arrest in a public location. No significant difference in survival to hospital discharge was seen for any of these subgroups.
The authors concluded that those given drugs had a higher rate of short-term survival for patients who received no ACLS intravenous drug administration following outpatient cardiac arrest versus patients with standard intravenous drug administration. But no significant improvement was shown between the two groups in survival to hospital discharge or long-term survival. These two factors were inconsequential because there was an equal distribution of therapeutic hypothermia and coronary intervention. Most of the deaths were caused by persistent brain damage.
No difference was seen in short-term or long-term outcomes in patients with an initial rhythm of ventricular fibrillation or pulseless ventricular tachycardia. There was, however, a higher rate of return of spontaneous circulation in those with asystole or pulseless electrical activity with medications but no difference in long-term outcomes.
Comment: It is difficult to control for all of the variables present in any ACLS study. The downtime is always most difficult to determine, and prolonged downtime has always been associated with a poorer outcome. These authors said their study was the first attempt to scientifically evaluate the effects of intravenous drug administration on the outcome of patients with out-of-hospital cardiac arrest. They attempted to determine if removing ACLS drug therapy would improve survival to discharge. They concluded that short-term survival was higher following use of intravenous ACLS drugs, but the universally applied drug interventions did not result in statistically significant improvement in final outcome.
Unlike some other reports of a worsened final outcome with epinephrine, this study did not conclude that epinephrine was an independent predictor of a poor outcome, but the researchers said their study was consistent with other reports that found no difference in survival after implementation of standard ACLS drugs during out-of-hospital cardiac arrest. Apparently, the only drugs used were epinephrine, atropine, and amiodarone. The drugs currently suggested by the American Heart Association are only epinephrine/vasopressin or amiodarone. Epinephrine and vasopressin, both arterial vasoconstrictors, have similar results, and are interchangeable per the guidelines.
It’s always heartening to see an initial return of spontaneous circulation (ROSC), but it has been suggested that the negative resuscitation effects of epinephrine negate any short-term benefit of ROSC produced by this medication. The authors do note, like many others, that epinephrine does have substantial downsides following cardiac arrest, particularly on cerebral microcirculation.
This study is relatively good evidence to negate the previously accepted axiom that the routine use of empiric medications during prehospital cardiac arrest is helpful and not harmful. No prior prospective data show that ACLS drugs are helpful, and this study, which actually looked at long-term outcomes, indicates that epinephrine, in particular, is of no benefit to long-term survival.
One usually feels that ROSC signifies a better outcome or at least a better chance that the patient will have long-term survival. ROSC in the ED or ambulance, however, does not actually mean much. It appears to be easily accomplished with ACLS and medications. Long-term survival can’t be achieved without first restoring circulation, but current drugs used to obtain this may produce subsequent irreversible cerebral damage.
This study had a few problems, but likely the perturbations will be of no significance. The ambulance personnel were not blinded to the randomization, and the use of medications may have been dictated by EMS preference. The quality of CPR could not be universally assessed, but the authors said there was no significant difference between the CPR data that were available for analysis. The study also lacked reliable timing for drug administration, so the time from cardiac arrest to intervention was not investigated because many of the cases were unwitnessed cardiac arrest.
Prehospital Epinephrine Use and Survival among Patients with Out-of-Hospital Cardiac Arrest
Hagihara A, Hasegawa M, et al.
This article is similar to the previous one in evaluating the association between epinephrine use before hospital arrival and short- and long-term mortality in patients with cardiac arrest. It was done in Japan, and analyzed data from more than 400,000 patients during a three-year period. The EMS time-to-arrival from first call averaged about seven minutes, and the average patient age was 72 years. The authors could not control for a variety of parameters, but they could assess the effectiveness of epinephrine use during CPR. Similar to the previous reports, they concluded that epinephrine is not beneficial for one-month survival, even though there was a significant improved likelihood of achieving ROSC in those given the drug. Epinephrine was given in this study for all causes of cardiac arrest after verifying the absence of a pulse in the carotid artery. The parameters studied were the return of spontaneous circulation before hospital arrival, survival at one-month post-arrest, survival with good or moderate cerebral performance, and survival with no, mild, or moderate CNS adverse outcomes.
Like other studies, ROSC before hospital arrival was observed more commonly in patients given epinephrine than in the non-epinephrine group (18.5% versus 5.7%, P< 0.001). Importantly, in all of the other parameters studied, the use of epinephrine resulted in a statistically significant decrease in patients with one-month survival, those with survival with mild to moderate cerebral performance, and those who survived with no and mild neurological disability. Epinephrine administration also had a negative effect in all final outcome parameters studied. The negative association between prehospital epinephrine and long-term outcome variables suggests that there was a decreased chance of ultimate survival and good functional outcome despite an increased chance of ROSC with epinephrine.
Comment: This is another study indicating that epinephrine might make medical providers feel better during the initial evaluation and treatment of cardiac arrest, but worse in the long run because it has no proven value. That by itself is not a catastrophic effect, but epinephrine was actually associated with worse neurological outcome. Per the AHA, no drug has been proven to increase survival rates in cardiac arrest. Although downsides and lack of benefit of drugs are mentioned in the 2010 CPR guidelines, one wonders why the American Heart Association has not focused on these studies, and amended its recommendation that still includes empiric epinephrine/vasopressin or amiodarone.
I remember giving very high doses of epinephrine with a better initial ROSC more than 20 years ago. It was not uncommon to give 5 mg/5 amps of epinephrine to someone in asystole, and see at least an initial return of circulation in many. When this was followed up, it did not have any beneficial effect on long-term survival and was subsequently abandoned. Likewise, any benefit from bicarbonate, lidocaine, atropine, calcium, and glucocorticoids has not been demonstrated.
These two articles, and a number of others, are unsettling. It appears that there are significant data suggesting that the use of epinephrine in cardiac arrest is detrimental, even though drugs may result in a transient pulse and a heartbeat. A worsened final outcome, despite an initial benefit, may be reversed with some yet to be suggested interventions provided post-arrest. Perhaps some cerebral vasodilator may prove useful in derailing the subsequent adverse effects of epinephrine used in resuscitation. Given the tremendous time, effort, and expense required for patients resuscitated from cardiac arrest, a good neurological outcome is still in the single percentage numbers. Overall, only about 10 percent of prehospital cardiac arrest patients are ever discharged, many of whom are markedly impaired neurologically.
Time-to-treatment is probably the most sensitive predictor of a good outcome. There is at least a three-to-five percent drop in survival rate for each minute of untreated VF. The best scenario is for someone to have VT/VF, immediate CPR, only a few minutes for EMS response time, and rapid defibrillation.
I remember a malpractice case where an emergency physician was found to have committed malpractice because he delayed epinephrine use in an in-house cardiac arrest. Having to respond from the ED to in-hospital arrests is bad enough, but that was an absurd verdict then. Now data indicate that the outcome would likely have been better had no epinephrine been used.
Whether recent information will change the American Heart Association recommendations is questionable, but one can certainly not be faulted for failing to routinely administer epinephrine in patients who have had lack of ROSC from paramedic interventions. Patients who arrive in asystole are probably best left alone unless there are extenuating circumstances, such as hypothermia, drowning, young child, or lack of any ACLS interventions. It does no good to resuscitate an 80-year-old nursing home patient who hangs on for three to five days while the family is offered false hope, and many thousands of dollars are spent on useless interventions with no hope of viable long-term outcome.