SGEM#64: Classic EM Papers (OPALS Study)

Subscribe: RSS

[display_podcast]

Date: February 28, 2014

Reference: Stiell IG et al. Advanced Cardiac Life Support in Out-of-Hospital Cardiac Arrest. NEJM 2004

Guest Skeptic: Dr. Anand Swaninathan is an assistant program director at NYU/Bellevue Hospital in the department of EM. His interests are in resident education, critical care and EBM. Swami loves debunking urban legends in EM and is a strong supporter of FOAMed.

Case Scenario: A 43-year-old man experiences a cardiac arrest on the street. You are the first provider on scene with EMS. A cardiac defibrillator is hooked up and the patient is in ventricular fibrillation. He is unsuccessfully shocked and chest compressions are started. The paramedics ask you if you want to intubate and administer medications or load for transport.

Questions: Do advanced life support techniques, specifically pharmacologic interventions, improve survival to discharge versus basic life support (rapid defibrillation and CPR) in patients with out-of-hospital cardiac arrest (OHCA)?

Background: Sudden cardiac arrest is common and, obviously, very bad. In the US, there are about 500,000 cardiac arrests each year. About half of these cardiac arrests are OHCA and the survival rate is pretty poor. The most recent survival estimates put it at 7 – 9.5% in most communities. About 10-12 years ago, the American Heart Association built the 4-step “chain-of-survival.”

Step One – Early access to emergency care
Step Two – Early CPR
Step Three – Early defibrillation

In fact, in communities with high layperson basic life support (BLS) training and automatic electronic defibrillators (AEDs) in the community, the rate of survival after OHCA is higher.

The 4^th Step in the chain, however is slightly more controversial; early advanced care. This basically means rapid access to ACLS type resuscitation skills (intubation and intravenous drug therapy). The AHA now has Five Links in the chain of survival with Step Five being Integrated post-cardiac arrest care.

ACLS has little evidence to defend it. Of course, ACLS is not a single treatment. It’s a bundle of treatments much like early goal directed therapy. It involves airway management with intubation and drug administration based on set algorithms. In spite of the lack of evidence behind it, ACLS is standard of care. Patients who have OHCA get ACLS treatment in the field (which may delay their transport). Additionally, there are a ton of providers trained in ACLS. It costs about $174-250/2 years to get certification and a number of hospitals require ACLS certification in order to practice.

If ACLS isn’t proven to help, it brings up a number of issues. Is ACLS training a waste of time and money for providers? Do the therapies in ACLS detract from critical interventions like CPR and defibrillation? Do delays in transit matter now that hospitals are doing ED ECMO? Are we harming patients with ACLS by bringing back more people with severe neurologic disabilities?

Dr. Ian Stiell

The authors of the study we are going to discuss saw these issues more than a decade ago and attempted to tackle it head on. People should be familiar with the lead author on this paper, Dr. Ian Stiell. He is arguably the most famous/cited Canadian EM researcher. If you don’t know his name you certainly know his significant contribution to the area of clinical decision instruments. These are the Ottawa Ankle and Knee Rules, Canadian C-Spine and CT Head Rules and his latest, Ottawa SAH Rules. We have covered some of these topics on past episodes of the SGEM:

SGEM#3: To X-ray or not to X-ray (Ottawa Ankle Rules)
SGEM#5: Does Johnny “kneed” an X-ray? (Ottawa Knee Rules)
SGEM#48: Thunderstruck (Subarachnoid Hemorrhage)

Reference: Stiell IG et al. Advanced Cardiac Life Support in Out-of-Hospital Cardiac Arrest. NEJM 2004

Population: All persons > 16 yo who had an OHCA and for whom resuscitation was attempted.
- Excluded: < 16, persons who were dead, patients with trauma, disorders of clearly noncardiac cause.
Intervention: Advanced-life-support program whereby paramedics were trained in inutbation, IV line placement and IV medication administration.
Control: Basic-life-support – defibrillation + CPR
Outcome:
- Primary Outcome: Survival to hospital discharge (defined as patient leaving hospital alive).
- Secondary Outcome: ROSC, admission to hospital and cerebral performance category.

Authors’ Conclusions: “The results of the OPALS study did not show any incremental benefit of introducing a full advanced-life-support program to an emergency-medical services system of optimized rapid defibrillation.”

Key Results: They recruited5,638 patients over 48 months in 17 communities and 11 hospitals. There were 1,391 in the rapid-defibillation phase (no ACLS) over 12 months and 4,247 advanced-life-support phase over 36 months. The initial cardiac rhythm was: VFib/VTach 34.5% vs. 31.5%, PEA 25.8% vs. 25.3%, and Asystole 38.8% vs. 42.0%. The medication during the ACLS phase included: Epinephrine 95.8%, Atropine 87.3%, and Lidocaine 23.6%.

No statistical difference difference in survival to hospital discharge:

rapid defibrillation (no ACLS) 5.0% vs. ACLS 5.1%

Primary Outcome: Survival to hospital discharge
- 5.0% vs. 5.1% (absolute change 0.1% p 0.83)
Secondary Outcomes:
- ROSC 12.9% vs. 18.0% (absolute change 5.1% p < 0.001)
- Admission to hospital 10.9% vs. 14.6% (absolute change 3.7% p < 0.001)
- Survivors’ Cerebral-performance category level 1 – 78.3% vs. 66.8% (p 0.73)
- Survivors’ median Health Utility Index at one year 0.84 vs. 0.79 (p 0.67)

Critical Appraisal Toolkit for Cohort Studies:

Did the review ask a clear question? – Yes. The issue here was whether ACLS management increased patient survival to hospital discharge
Appropriate method to answer their question? – Yes. Ideally, a randomized control trial would have been better but not possible. This was a “before-after” multicenter, cohort trial (before and after ACLS was instituted) with rapid defibrillation only for 12 months and advanced-life-support for 36 months. The methods were very good and specifically cite following Ustein-style guidelines for reporting the data about cardiac arrests. Doing a RCT would not be feasible. You would have to provide ACLS to half of patients (which is believed to be the standard of care) and BLS to the other half. Ethically you cannot do trials that may involve harm. Evidence based medicine has a hierarchy of evidence. A before/after trial is less robust than a RCT. EBM also has limitations and this well-done cohort trial identifies some of these limitations. Sometimes a lower form of evidence will be the “best” evidence we can get.
Cohort recruited in an acceptable way? – Yes. All OHCA in the Ottawa enchantment area were included for analysis.
Was the exposure accurately measures to minimize bias?– Yes. Objective measures were used. The outcomes are ones that both doctors and patients would care about. In fact if there was any bias it was probably in favour the intervention (ACLS). Given the fact that there was no difference in the primary outcome strengthens the conclusion.
Was the outcome accurately measured to minimize bias? – There could be no blinding for the patients or doctors in this type of study design. However, the abstractors who are collecting data can be blinded to outcomes. It’s not clear if that occurred. Lack of blinding for outcome should not have impacted on the mortality data. Either the patient was or was not alive at discharge. Secondary outcome of CPC also validated in past.
All important confounding factors identified? – Unsure. There is a problem of multiple interventions with the OPALS study. Incorporating ACLS into OHCA treatment involves multiple interventions including drugs (epi, lidoc, atropine), fluids and procedures (intubation). Any of these could individually improve outcomes or worsen outcomes but we have to regard them as a bundle (much like EGDT).
Follow-up complete and long enough? – Yes. ROSC, admit to hospital, survive to discharge, cerebral-performance category at discharge and quality of life at one year
How precise are the results? – Yes. The confidence intervals were pretty small.
Do you believe the results? – Yes
Can you apply them to your local population? – Unsure. This study was not ED focus but rather on the pre-hospital setting. However, many systems have MDs working in the field with EMS. It would be important for those individuals to be aware of this study. Also we do not work in isolation but in a continuum of care from pre-hospital, emergency department and then in-patient units (ICU, CCU, trauma). These results apply to the pre-hospital setting in Ontario, Canada. We cannot necessarily extrapolate them to other EMS systems. With regards to the ED setting, there have not been any trials showing benefit of ACLS medications for in-hospital cardiac arrest. In addition, there have been some observational studies suggesting ACLS medications could cause harm. (Dr. David Newman The NNT)
Do the results fit with other available evidence? – Yes. There was another study published five years later from Norway published (JAMA 2009). It was a RCT of 851 adults with out of hospital non-traumatic cardiac arrests. Patients were randomized to ACLS with or without intravenous drugs. Like OPALS, they showed improved ROSC and admit to hospital but not benefit for discharge from hospital neurologically intact or alive at one year.

Dr. Anand Swaminathan

Does this Change What You Do? No because it’s now 10 years later and ACLS is still the standard of care treatment. However, OPALS is one paper in a group that questions the quality of ACLS care and there should be readdressing of this algorithm. Also, we do not work in the pre-hospital setting. We should focus on good CPR and early defibrillation.

Bottom Line: Addition of an advanced life support algorithm to BLS management did not increase the survival to hospital discharge for patients with out of hospital cardiac arrest.

Case Resolution: You tell the paramedics to support the patient with bag mask ventilation and rapidly transport to the hospital.

This concludes the first ever SGEM Classic Paper critical review. Let me know what you think. Should we do more podcasts on historic/practice changing papers? If so, what papers would you like the SGEM to put under the skeptical lens? Send me your suggestions to TheSGEM@gmail.com with “classic” in the subject heading.

Keener Contest: Last weeks winner was Dr. Mark McAllister. He knew that PDS or polydioxanone sutures take greater than 90 days for complete absorption. Listen to this weeks episode of the SGEM for the Keener Kontest? If you know the answer send an email to TheSGEM@gmail.com with “keener” in the subject line. The first person will receive a skeptical prize.