MH900 EPIDEMIOLOGY AND STATISTICS MODULE
Section 1: MECHANICAL CHEST COMPRESSION DEVICES
SECTION 1: EPIDEMIOLOGY STUDY DESIGN
SURVIVAL FROM IN-HOSPITAL CARDIAC ARREST FOLLOWING CARDIOPULMONARY RESUSCITATION WITH A MECHANICAL CHEST COMPRESSION DEVICE VERSUS MANUAL CHEST COMPRESSIONS: A RANDOMISED CONTROLLED TRIAL
1.0 BACKGROUND
In-hospital cardiac arrest (IHCA) is associated with poor survival, or survival but with impaired neurologic function (Girotraetal, 2012). In the UK, the National Cardiac Arrest Audit (NCAA) reported 22,628 IHCA cases for ≥16 year olds for April 2011 to March 2013 (Nolan et al, 2014). The data represents approximately 60% of UK hospitals, and indicates an incidence rate of 1.6 IHCA cases per 1,000 hospital admissions. 16.9% of IHCA patients had an initial shockable rhythm (ventricular fibrillation or pulseless ventricular tachycardia) and 72.3% of IHCA patients had an initial non- shockable rhythm (asystole or pulseless electrical activity). Survival to hospital discharge rates were reported as 18.4% overall, with a 49.0% survival rate for patients with a shockable rhythm and a 10.5% survival rate for patients with a non-shockable rhythm.
Cardiopulmonary resuscitation (CPR) is one of the key links in the “chain of survival” (Cummins et al, 1991) with observational data suggesting that for every minute that CPR is delayed survival decreases by 2.3% (Larsen et al, 1993). CPR at its most basic consists of chest compressions and is preformed to keep blood flowing to vital organs and to preserve brain function.
Scientific guidelines recommend high quality chest compressions based on a depth of 5-6cm, a rate of 100-120 compressions per minute, complete recoil after each compression, and minimal interruptions (Gwinnet, Davies and Soar, 2015). Several reviews of the literature suggest that high quality chest compressions are associated with improved survival with good neurologic functioning for IHCA (Sandroni et al, 2007; Wallace et al, 2013). However, observational research indicates IHCA chest compressions to be sub-optimal and not congruent with CPR guidelines (e.g. Abella et al, 2005).
Mechanical chest compression devices have been designed as an alternative to manual chest compressions, allowing prolonged, high quality chest compressions. Results from randomised controlled trials have not found evidence for the benefit of using these devices in the out-of-hospital cardiac arrest (OHCA) population (Perkinset al, 2015a; Rubertssonet al, 2014; Wiketal, 2014), with a recent meta-analysis showing a non-significant difference for survival from OHCA with good neurologic outcome (OR 0.76, 95% CI 0.53, 1.11) (Gateset al, 2015).
Furthermore, there is only weak observational data suggesting that mechanical chest compression devices can be safely implemented in the IHCA population and improve neurologically intact survival (Bonnemeier et al, 2011); suggesting clinical equipoise with regards to the clinical benefits of using mechanical chest compression devices for IHCA.
2.0 RESEARCH QUESTION
Based on the literature review the following research question is proposed based on the PICO method (Brown et al, 2006): In adult IHCA patients with a non-shockable rhythm (P), do chest compressions given with a mechanical device (I), in comparison with manual chest compressions (C), improve survival to hospital discharge (O)?
The null hypothesis (Ho) is that there will be no difference in survival between IHCA patient's in the mechanical chest compression device group in comparison with the manual chest compressions group.
3.0 AIM AND OBJECTIVES
The aim of the study is to investigate the effect of a mechanical chest compression device on outcome in a population of IHCA patients.
The study objectives are:
1: To assess the feasibility of implementing a mechanical chest compressions device as part of the IHCA resuscitation process.
2: To evaluate safety issues relating to prolonged use of a mechanical chest compressions device for IHCA.
3: To determine the effectiveness of a mechanical chest compressions device for improving outcomes in IHCA with a non-shockable rhythm, compared with manual chest compressions.
4.0 STUDY DESIGN
A multi-centre, parallel-group, pragmatic randomised controlled trial study design will be employed. Simple 1:1 randomisation will be utilised using a computer generated random sequence, with patients allocated to the treatment group using opaque sealed envelopes.
Due to the nature of the intervention both patients and hospital staff involved in resuscitating the patient will be unblinded. Health professionals conducting the neurologic function assessment will be blinded to the treatment allocation.
All study personnel will be blinded to the treatment allocation. During the analysis phase the two treatment groups will be identified as 0 and 1. Following completion and approval of the study manuscript, the code will be broken.
5.0 METHODS
Enrolment of patients will be 24/7. At the point of an IHCA, the hospital staff will call for the resuscitation team, whilst starting manual chest compressions which will continue until the patient is randomised. The resuscitation team will bring the mechanical chest compression device and sealed envelopes. The patient’s rhythm will be assessed and if non-shockable the patient will be enrolled. Only patients with anon-shockable rhythm will be eligible as patients with a shockable rhythm will be
defibrillated until return of spontaneous circulation (ROSC) is achieved. The next numbered sealed enveloped will be opened and the patient will be randomised to either:
Treatment arm A: mechanical chest compressions with the LUCASTM 2 device (intervention),which is designed to be a lightweight portable device, which can befitted to the patient in less than 20 seconds, providing safe and standardised chest compressions according to national scientific guidelines (LUCASTM Chest Compression System, 2015); or
Treatment arm B: manual chest compressions (control) based on Resuscitation Council (UK) 2015 guidelines for in-hospital CPR (Gwinnet, Davies and Soar, 2015).
Resuscitation will then continue until the patient achieves ROSC or the normal procedures for terminating a resuscitation attempt are followed.
This study will have a pragmatic design, whereby the benefit of the mechanical chest compression device will be assessed under real clinical practice conditions (Roland and Torgerson, 1998). This will include the amount of training given to the staff, and the conditions relating to how quickly the device takes to arrive. The results of the study will be reported in line with the CONSORT statement for pragmatic trials (Zwarenstein et al, 2008).
5.1 Bias
The study design will control for confounding, as the process of randomisation should ensure that the treatment groups are equal for any known or unknown confounders (Brookes and Ben-Shlomo, 2013).
The sealed envelope method will be used for practical reasons due to the emergency nature of a cardiac arrest, and has been used in similar studies (Rubertsson et al, 2014). Failure to conceal allocation can invalidate results (Schulzetal, 1995); however,a process will be put in place whereby the envelopes will be locked in the trolley containing the defibrillator and LUCASTM 2 device until the point of randomisation.
5.2 Population and sampling frame
The target population will be all IHCA patients with a non-shockable rhythm in England & Wales. The study population will be all IHCA patients attending participating hospitals in England & Wales. The sampling frame. will be all eligible IHCA patients aged 16 years and over, attending a participating hospital. The sampling method will be random sampling.
5.2 Setting
The setting will be a sample of hospitals in England & Wales chosen to be representative of the population.
5.3 Enrolment
Patients will be enrolled into the study if after screening they fulfil the following inclusion/exclusion criteria:
5.3.1 Inclusion criteria
• In-hospital patient in cardiac arrest
• Presenting rhythm is non-shockable
• Aged ≥16 years.
5.3.2 Exclusion criteria
• Patient admitted to hospital due to OHCA
• Patient is pregnant
• A terminal illness is present before the IHCA
• Presenting rhythm is shockable
• Resuscitation not attempted due to DNACPR order/deemed futile by senior clinician.
6.0 ETHICAL CONSIDERATIONS
Ethical approval will besought from the Local Research Ethics Committee (LREC)
6.1 Consent
Cardiac arrest is associated with immediate loss of capacity so IHCA patients will be unable to give informed consent. Ethical approval will besought to enroll eligible patients without informed consent under the Mental Capacity Act (2005) for England & Wales. Patients or their legal representative will be approached as soon as practicable after the initial emergency has passed to inform. them of their participation and request consent to continue.
Patients will be free to withdraw their consent from the study at anytime after regaining ROSC, and reasons for withdrawal will be collected and reported. A request will be made asking the patient’s permission to use the data obtained prior to withdrawal and to include data for the primary outcome measure. If permission is granted the patient will be included in the final analysis; if not, all data from that patient will be destroyed.
6.2 Safety
Monthly Data relating to patient safety will be provided to an independent Data and Safety Monitoring Committee (DSMC), and the DSMC can request further analyses. A pre-specified interim analysis will be conducted by an independent blinded Statistician. On the basis of an analysis the DSMC may pause or stop the study based on:
• A significant difference in benefit between the treatment groups for the primary outcome measure is found, based on a pre-defined threshold and allowing the study to be stopped early
• A group difference relating to patient safety is found in the analysis indicating harm as a result of the LUCASTM 2 device
• Results from other studies showing a statistically significant benefit / harm as a result of the LUCASTM 2 device.
6.3 Good Clinical Practice
The study will be carried out in accordance with principles of the MRC Good Clinical Practice (GCP) Guidelines.
7.0 MEASURES AND OUTCOMES
Protocols and training will be put in place to standardise collecting of measures across sites.
7.1 Baseline data
Baseline data collected will include patient identifiable information, event data / clinical information, patient characteristics, IHCA aetiology, and the hospital department / unit that the resuscitation attempt took place in.
7.2 Primary outcome measure
The primary outcome measure will be survival to hospital discharge which is a binary endpoint of yes/no
7.3 Secondary outcome measures
The main secondary outcome measures will be ROSC which is a binary end point of yes/no, and neurologic function which will be measured using the Cerebral Performance Category (CPC) Scale. CPC scores will be categorised into good neurologic function (CPC score 1 to 2) and poor neurologic function (CPC score 3 to 5) giving a binary endpoint. The CPC is well validated in the cardiac arrest population and its use is recommended by the Utstein guidelines (Perkinset al, 2015b).
Further secondary outcome measures will be time to ROSC and standardised data will be collected relating to adverse physical effects from chest compressions (e.g. fractures).
8.0 STATISTCIAL MEASURES
8.1 Treatment effect
In practice an initial pilot study could be conducted as a preliminary investigation to determine a likely treatment effect and sample size calculation, as well as to assess feasibility and safety. However, due to a lack of published research in the IHCA population, and for the purposes of this study design a treatment effect estimate of a 4% increase in survival to hospital discharge in the mechanical chest compression treatment groupshall be considered.
8.2 Sample size calculation
A sample size calculation shows that to detect a treatment effect of a 4% increase in survival (10.5% vs.14.5%), with alpha (α) set at 0.05 and beta (β) set at 0.8 gives a sample size requirement of 2,144 patients with 1,072 patients in each treatment group (Sampsize Calculator, 2003). There is nofollow- up for the primary outcome so an adjustment for follow-up is not required.
8.2 Data analysis
Descriptive statistics for the data will be produced and a comparison between groups for baseline demographic and clinical characteristics will be conducted, analysing whether there were any statistically significant differences between the treatment groups at baseline.
A main intention to treat analysis will be conducted. For the primary and main secondary outcomes, logistic regression analysis will be used to produce odds ratios and 95% confidence intervals. Crude odds ratios will be produced as well as adjusted odds ratios (adjusted for age, sex, ethnicity, initial rhythm, aetiology, hospital department / unit, time to mechanical chest compressions, and time to ROSC). A two-sided significance level of 0.05 will be used for all statistical analyses.
REFERENCES
Abella, B.S., et al. (2005). Quality of cardiopulmonary resuscitation during in-hospital cardiac arrest. Journal of the American Medical Association, 293(3),pp. 305-310.
Bonnemeier, H., et al. (2011). Continuous mechanical chest compression during in-hospital cardiopulmonary resuscitation of patients with pulseless electrical activity. Resuscitation, 82, pp. 155- 159.
Brookes, S.T. and Ben-Shlomo, Y. (2013). Epidemiological Concepts. In: Ben-Shlomo, Y., Brookes, S.T. and Hickman, M. eds. Epidemiology, Evidence-Based Medicine and Public Health: Lecture Notes. 6th Ed. Chichester: Wiley-Blackwell, pp. 20-25.
Brown, P., et al. (2006). How to formulate research recommendations. British Medical Journal, 333, pp. 804-806.
Cummins, R.O., et al. (1991). Improving survival from sudden cardiac arrest: the "Chain of Survival" concept. A statement for health professionals from the Advanced Cardiac Life Support Subcommittee and the Emergency Cardiac Care Committee, American Heart Association. Circulation, 83(5), pp. 1832- 1847.
Gates, S., et al. (2015). Mechanical chest compression for out of hospital cardiac arrest: Systematic review and meta-analysis. Resuscitation, 94, pp. 91-97.
Girotra, S., et al. (2012). Trends in Survival after In-Hospital Cardiac Arrest. New England Journal of Medicine, 367(20), pp. 1912-1920.
Gwinnet, C., Davies, R. and Soar, J. (2015). Resuscitation Council (UK) Guidelines 2015: In-Hospital Resuscitation.https://www.resus.org.uk/resuscitation-guidelines/in-hospital-resuscitation/
Larsen, M.P., et al. (1993). Predicting survival from out-of-hospital cardiac arrest: a graphic model. Annals of Emergency Medicine, 22(11),pp. 1652-1658.
LUCASTM 2 Chest Compression System. (2015). LUCASTM CPR, http://www.lucas-
cpr.com/en/lucas_cpr/lucas_cpr
Nolan, J.P., et al. (2014). Incidence and outcome of in-hospital cardiac arrest in the United Kingdom National Cardiac Arrest Audit. Resuscitation, 85, pp. 987-992.
Perkins G.D., et al. (2015a). Mechanical versus manual chest compression for out-of-hospital cardiac arrest (PARAMEDIC): a pragmatic, cluster randomised controlled trial. Lancet, 385, pp. 947-955.
Perkins G.D., et al. (2015b). Cardiac arrest and cardiopulmonary resuscitation outcome reports: update of the Utstein resuscitation registry templates for out-of-hospital cardiac arrest: A statement for healthcare professionals from a taskforce of the international liaison committee on resuscitation. Resuscitation, 96, pp. 328-340.
Roland, M. and Torgerson, D.J. (1998). Understanding controlled trials: what are pragmatic trials? British Medical Journal, 316(285), doi:http://dx.doi.org/10.1136/bmj.316.7127.285.
Rubertsson, S., et al. (2014). Mechanical chest compressions and simultaneous defibrillation vs conventional cardiopulmonary resuscitation in out-of-hospital cardiac arrest: the LINC randomized trial. Journal of the American Medical Association, 311(1), pp. 53-61.
Sandroni, C., et al. (2007). In-hospital cardiac arrest: incidence, prognosis and possible measures to improve survival. Intensive Care Medicine, 33, pp. 237-245.
Schulz, K.F., et al. (1995). Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. Journal of the American Medical Association, 273(5), pp. 408-412.
Wallace, S.K., Abella, B.S. and Becker, L.B. (2013). Quantifying the effect of cardiopulmonary resuscitation quality on cardiac arrest outcome. A systematic review and meta-analysis.Circulation: Cardiovascular Quality and Outcomes, 6(2),pp. 148-156.
Wik, L., et al. (2014). Manual vs. integrated automatic load-distributing band CPR with equal survival after out of hospital cardiac arrest. The randomized CIRC trial. Resuscitation, 85, pp. 741-748.
Zwarenstein, M., et al, (2008). Improving the reporting of pragmatic trials: an extension of the CONSORT statement. British Medical Journal, 337(a2390), doi:10.1136/bmj.a2390.