“High-Quality” CPR: Is a Mechanical Device Better than Manual CPR? Nope.
“I now had to regard the patient as dead. In spite of this, I returned immediately to the direct compression of the region of the heart.” — Friedrich Maass, 1892 (Taw 1991)
The AHA estimates that about 325,000 EMS-assessed OHCAs happen in the United States each year. The median age is 66 years. Half of these arrests are witnessed (bystander 39% and EMS provider 11%) with the other half being unwitnessed. The majority of these cardiac arrests happen at a home or residence (70%). Of those EMS-treated patients with OHCA, about a quarter have an initial shockable rhythm. Survival to discharge from hospital for adults with OHCA assessed by EMS is around 6% (AHA Statistical Update 2015).
The American Heart Association came out with updated CPR & ECC Guidelines in 2015 that included its “Chain-of-Survival.” There are five steps in the Chain-of-Survival for OHCA:
Step One — Recognition and activation of the emergency response system
- If we are going to make a difference with OHCA we need to start the chain of survival. That begins with recognizing when someone has a cardiac arrest and quickly activating EMS.
Step Two — Immediate high-quality CPR
- We have talked about this issue before on EMA. There are two components to this statement: “immediate” and “high-quality.” The immediate part was addressed by an interesting Swedish study that crowd sourced CPR. (Ringh 2015) The authors used smart phones and GPS positioning to alert trained bystanders within 500m of a presumed cardiac arrest. It increased CPR by 14%. The NNT Number-Needed-to-Text was 7. This essay focuses on the second component of this step — “high quality CPR” — and whether a machine is better than manual CPR.
Step Three — Rapid defibrillation
- This is key to survival. If you have a shockable rhythm and can defibrillate quickly the chances of survival increase significantly. The absolute decrease in survival is 7–10% for each one-minute delay to defibrillation.
Step Four — Basic and advanced emergency medical services
- This could be the topic of another essay. On this month’s EMA we review a meta-analysis from Bottiger et al that compares observational studies using physicians versus non-physicians on the emergency medical services. The classic OPALS study by Dr. Ian Stiell demonstrated that adding paramedics with advanced skills improved ROSC and survival to hospital, but did not increase survival to hospital discharge
Step Five — Advanced life support and post-arrest care
- Again, this is beyond the scope of today’s essay and includes therapeutic hypothermia and ARDS-preventative ventilator management.
The 2015 AHA guidelines emphasize a few points about high-quality CPR. This includes compressions of 100–120/minute with minimizing interruptions and a depth of between 5–6cm. Some great infographics on the top five changes in the AHA guidelines can be found at CanadiEM.org
The history of CPR is thought to go back thousands of years (Cooper 2006). Some people cite a biblical reference dating to 800 B.C. that describes mouth-to-mouth resuscitation, while others attribute early CPR to an ancient Egyptian practice of hanging the recently dead upside down and compressing the chest to force air in and out.
Mouth-to-mouth was being used on drowning victims in Paris in the 1740’s and in 1891 Friedrich Maass performed the first documented closed chest compressions on a human.
In 1903 Dr. George Crile successfully used external chest compressions to save a life. Par for the course in a world where it currently takes 17 years for 14% of evidence to reach the bedside, “external cardiac massage” took almost 60 years to be re-discovered when William Kouwenhoven reintroduced chest compressions into patient care in 1958.
In 1960 the American Heart Association developed “Cardio Pulmonary Resuscitation.” There is much more CPR history available online (http://www.mycprcertificationonline.com/info-graphics/history-of-cpr).
On EMA, we’ve previously discussed that in a dichotomous situation some CPR is probably better than no CPR. The area that Jerry Hoffman seems to get skeptical about is the quality. What is the right rate, what is the right depth and does it really matter?
There are devices that give a consistent rate and depth, and never need to take a break. This makes it possible to compare these mechanical CPR devices to manual CPR and determine if there is a difference in neurologically intact survival to hospital discharge.
Bonnes et al. reported a meta-analysis of randomized controlled trials (RCT) and observational studies comparing mechanical CPR with a variety of automated chest compression devices versus standard manual CPR.
Based on animal studies and physiological observations during in-hospital cardiac arrest scenarios, larger non-randomized observational research indicated that mechanical chest compression devices improved survival to admission leading to RCT.
The first RCT was halted early due to excess harm in the mechanical chest compression arm, but more industry-sponsored RCTs followed. Five RCTs and 15 observational studies were included and the results highlight the value of the controlled trial. While half (6/12) of the non-randomized studies indicated improved cardiac arrest survival to hospital admission, none of the RCTs demonstrated this benefit. Furthermore, for patient-centric outcomes like survival to hospital discharge and favorable neurological outcome neither RCTs nor observational designs indicated a benefit for mechanical chest compression devices.
Based on these studies, EMS systems would have no indication to invest in mechanical devices for CPR. There may (may not) be certain settings in which such mechanical devices could make sense — rural locations with limited resources, staffing and long transport times. However, until they do a properly designed RCT in this specific rural environment it would only be a hypothesis.
The capacity for non-randomized research designs (and studies with physiological, non-patient centric outcomes) to lead medicine astray is hardly without precedent.
- Naloxone for acute spinal cord injury. Based upon experimental spinal cord injuries in cat models, naloxone was advocated as a readily available, non-surgical management option for acute spinal cord injury victims to reduce neurogenic hypotension and improve neurological recovery (Faden 1981). A subsequent RCT of 154 acute spinal cord injured patients treated within 14 hours with naloxone (versus placebo and versus methylprednisolone) demonstrated no improvement in sensory or motor function with the opioid antagonist (Bracken 1990).
- Antiarrhythmic medications to suppress refractory ventricular tachycardia. Encainide was labeled a safe and effective therapy for previously refractory ventricular tachycardia after a before-after study noted elimination of the dysrhythmia in 54% of patients after six months of therapy (Mason 1981). A subsequent RCT of 1498 patients evaluating encainide or flecainide versus placebo demonstrated that those receiving the antiarrhythmic had twice the risk (risk ratio 2.64, 95% CI 1.60–4.36) of cardiac death and cardiac arrest (Echts 1991).
- Lung volume reduction surgery: In the 1990’s lung surgery for severe emphysema was considered based on an observational study of 89 consecutive patients demonstrating an overall increased FEV1 and six-minute walk distance up to 36-months following the surgery (Flaherty 2001). An RCT of 139 severe emphysema patients randomized to lung reduction surgery or medical therapy demonstrated a 16% 30-day mortality in the surgical group versus 0% in the medical group (National Emphysema Treatment Trial Research Group 2001).
Many additional examples of non-randomized studies provide observations and results that differ from subsequent randomized studies (Karanicolas 2008). David Sackett once described over one hundred sub-types of bias in medical research (Sackett 1979) and the RCT is the best method available for clinical scientists to minimize the biases from observed results — both those we recognize and those we do not. Unfortunately, RCTs are expensive, time-consuming, and sometimes unethical, so they do not exist for every ED scenario faced by clinicians.
Observational studies can yield valuable proof-of-concept if they attain the strict Hill criteria noted in the Table (Tugwell 2006). Unfortunately, most observational studies do not report the components of the Hill Criteria so linking a cause-effect relationship to non-randomized studies is tenuous and usually erroneous. In general, non-randomized studies should serve only as hypothesis-generating information.
In conclusion, the current best evidence does not support the purchase of mechanical CPR devices for patient-oriented outcomes in OHCA.
Wm Kenneth Milne, MD
Adjunct Professor, Emergency Medicine
Western University, London, Ontario Canada
Faculty, Best Evidence in Emergency Medicine
Creator, Skeptics Guide to Emergency Medicine
References
Taw RL; Dr. Friedrich Maass: 100th Anniversary of “New” CPR, Clin Cardiol 1991; 14: 1000–1002. PubMed ID 1841015
Ringh M et al. Mobile-phone Dispatch of Laypersons for CPR in Out-of-Hospital Cardiac Arrest, N Engl J Med 2015; 372: 2316–2325. PubMed ID 26061836
Stiell et al. Advanced Cardiac Life Support in Out-of-Hospital Cardiac Arrest. N Engl J Med 2004; 351:647–656 August 12, 2004. PubMed ID 15306666
Cooper JA, et al. Cardiopulmonary Resuscitation: History, Current Practice, and Future Direction, Circulation 2006; 114: 2839–2849. PubMed ID 17179033
Bonnes JL et al; Manual Cardiopulmonary Resuscitation Versus CPR Including a Mechanical Chest Compression Device in Out-of-Hospital Cardiac Arrest: A Comprehensive Meta-Analysis from Randomized and Observational Studies, Ann Emerg Med 2016; 67: 349–360. PubMed ID 26607332
Flaherty KR et al; Short-term and Long-Term Outcomes After Bilateral Lung Volume Reduction Surgery: Prediction by Quantitative CT. Chest 2001; 119: 1337–1346. PubMed ID 11348937
National Emphysema Treatment Trial Research Group; Patients at High Risk of Death After Lung Volume Reduction Surgery, N Engl J Med 2001; 345: 1075–1083. PubMed ID 11596586
Faden AI et al. Opiate Antagonist Improves Neurologic Recovery After Spinal Cord Injury, Science 1981; 211: 493–494. PubMed ID 7455690
Bracken MB et al. A Randomized Controlled Trial of Methylprednisolone or Naloxone in the Treatment of Acute Spinal-Cord Injury. Results of the 2nd National Acute Spinal Cord injury Study. N Engl J Med 1990; 322: 1405–1411. PubMed ID 2278545
Mason JW, Peters FA; Antiarrhthmic Efficacy of Encainide in Patients with Refractory Recurrent Ventricular Tachycardia. Circulation 1981; 63: 670–675. PubMed ID 6780234
Echts DS et al. Mortality and Morbidity in Patients Receiving Encainide, Flecainide, or Placebo. The Cardiac Arrhythmmia Suppression Trial, N Engl J Med 1991; 324: 731–788. PubMed ID 1900101
Karanicolas PJ, Kunz R, Guyatt GH; Point: Evidence-Based Medicine Has a Sound Scientific Base. Chest 2008; 133: 1067–1071. PubMed ID 18460513
Sackett DL; Bias in Analytic Research. J Chronic Dis 1979; 32: 51–63. PubMed ID 447779
Tugwell P, Haynes B. Assessing Claims of Causation in Clinical Epidemiology: How to Do Clinical Practice Research, 3rd Edition; Haynes RB, Sackett DL, Guyatt GH, Tugwell P (eds), Lippincott Williams & Wilkins 2006, page 358.
