There is a large and growing body of evidence related to mechanical chest compression device use. Much has been learned, but many important topics remain to be studied. This review article discusses factors important to consider during mechanical CPR to maintain survival rates and improve them. It combines detailed insight from the published literature with perspective from the author, a practicing EMS physician, a frequent user of mechanical chest compression devices and an experienced experimental and clinical trialist in the field of resuscitation. With this approach, this article defines and discusses the evidence supporting the following recommendations for high-performance use of mechanical chest compression devices:
- Focus on excellent basic life support (BLS) with manual compressions first, for at least the first two CPR cycles, before applying the device.
- Apply the device with only a brief pause in compressions; train, practice, and review data to reduce this pause to less than 10 seconds.
- Find the individual patient’s optimal chest compression point by palpating the chest compression-generated pulse, measuring end tidal carbon dioxide (ETCO2) or using ultrasound (if available).
- Promptly reposition the device if it is visually out of correct position, if ETCO2 is low, or if the palpated chest compression pulse is weak.
- Use 30:2 mode to facilitate ventilation, whether ventilating with bag/valve/mask, supraglottic device, or endotracheal tube. If continuous chest compressions, the person ventilating must only concentrate on that and give a quick breath between every nine to ten compressions.
- Pause briefly to defibrillate.
These device-oriented approaches enable you to fit chest compression devices into an overall approach to high performance CPR in ways that are both consistent with the evidence and compatible with the realities of resuscitation.
Over the last 60 years, we have tried many ways to improve cardiopulmonary resuscitation (CPR) techniques for cardiac arrest patients. One way, motivated by the observation that manual chest compressions (CC) are difficult to perform, has been to introduce manual or powered mechanical chest compression devices that seem to have potential to increase blood flow and improve outcome. Important questions should be asked about these devices. Is there a need for them? What CPR issues will be solved? Do they perform better than high quality manual CPR? Are there unintended consequences? Will their use shift the focus from patient care to device care? These questions and answers highlight the controversies about the effectiveness of these devices.
Most people will agree that manual CC are tiring and difficult to perform, especially during transport. A provider will tire quickly, and this will reduce CPR quality and potentially impact outcome. Mechanical CC devices should be able to solve this problem but, if so, why have most studies failed to show improvement in outcomes? Multiple observational studies have come up with negative results (mechanical CC was worse). Methodology in such studies makes those findings suspect due to resuscitation time bias; because the devices tend to be applied later in resuscitation attempts, the cohort of patients treated by them are already less likely to survive when the device is applied. Only studies comparing with historical controls have shown a benefit. The observation that mechanical CC solve some of the issues with manual CC without improving outcomes raises several unanswered questions about how we use the devices today.
- Mechanical CPR Devices: Where is the Science?
- Temporal Utilization Trends in Prehospital Mechanical CPR Devices
- The Science Behind Mechanical CPR Devices
Multiple randomized controlled trials (RCT) in out-of-hospital cardiac arrest (OHCA) patients found that mechanical CC devices provided no better survival to hospital discharge than manual CC (6.3*-9.4% vs 6.9*-11%, respectively, *30 days survival).1-3 Two other studies comparing AutoPulse with manual CPR found conflicting survival results [18.8% vs 6.3%, (n=133, p=0.03) and 5.8% vs 9.9% (n=1071, p=0.06), respectively]4,5 A small study (n=34) compared Vest CPR with manual and found 18% vs 6% (p=0.03) survival to 24 h, respectively.6 Do the lackluster study results reflect inherent limitations of the devices, or could better outcomes be obtained if the devices were used in a better way? How does study methodology affect study results?
Guidelines recommend against routine use of mechanical chest compression devices, instead recommending them as an “alternative in situations where high-quality manual CC is difficult or provider safety is compromised.”7 Are superiority of chest compression quality or improvement of clinical outcomes the only acceptable results to recommend mechanical devices?
Data from the U.S. Cardiac Arrest Registry to Enhance Survival (CARES) reveal increased usage of mechanical CPR devices, from 22% of cases in 2015 to 27% in 2017, despite the Guidelines’ wording. Buckler et al. reported from 2013 to 2015 that a mechanical CPR device was used at least once by 42% (244 of 582) of emergency medical service agencies; for those using it, median use was 44% (interquartile range, 11.9%–59.9%).8 Manufacturers report that more than 32,000 units have been sold globally, and this is increasing. Therefore, it is a concern if the services using the devices do not focus on correct use, since wrong use may hinder survival.
This review discusses factors important to be aware of during mechanical CPR to maintain survival rates, and ideally improve them. This is based on a literature search and a perspective as a practicing EMS physician and clinical trialist with experience in trials regarding defibrillation, drug vs. no drug, CPR quality and mechanical CPR.
The literature search was done (January 26, 2021) similar to the latest Cochrane Database of Systematic Review performed by Wang PL and Brooks SC.9 They searched Cochrane Central Register of Controlled Studies (CENTRAL), MEDLINE, Embase, Science Citation Index‐Expanded (SCI‐EXPANDED), Clinicaltrials.gov, the World Health Organization International Clinical Trials Registry Platform portal and Conference Proceedings Citation Index–Science databases. In addition, experts in the field of mechanical chest compression devices and manufacturers were contacted.
The search strategy focused on types of mechanical chest compression devices, and documentation of operational issues such as optimal compression point, compression depth and rate, ventilation, evaluation of blood flow during CPR, and defibrillation when a mechanical chest compression device was used. Outcome was not the focus because that has recently been done by Wang PL and Brooks SC in their review.9 Selection criteria were randomized controlled trials (RCTs), cluster‐RCTs and quasi‐randomized studies comparing mechanical chest compressions versus manual chest compressions during CPR for patients with cardiac arrest and studies focusing on the search strategy. Nine studies were identified and approximately 13,000 patients are included.1-6,10-12
Types of Mechanical Chest Compression Devices
Independent of which device is used, what counts is how we use the device and how we manage the patient in cardiac arrest. This will be influenced by factors we are not able to control such as size and location of the patient, need for evacuation before starting CPR, and number and skill of the rescuers. Aspects of device use affect management of the patient, including how we deploy the device, and how long that task takes, how we find the optimal compression point, depth, and rate, and how these parameters influence blood flow. Deployment may delay delivery of defibrillation attempts, lengthen time intervals without organ perfusion, and, if done sub-optimally, decrease blood flow. The same may hold true for incorrect depth and rate. The goal is blood flow generated by CC, which can be influenced by all these factors.
Each compression device has its own user interface and its own way of mechanically interacting with the thorax. Differences between devices are not all readily apparent and may affect the effectiveness of the devices in practical use. Therefore, it is important that the effectiveness of each device is evaluated in well-designed clinical studies.
In principle, there are three different types of mechanical chest compression devices based on to how they operate on the chest: automated pistons (three studies),1,2,11 pneumatic vests (one study),6 and band‐like mechanisms (three studies).3-5 The automated piston delivers chest compression through a piston placed on the sternum, represented by LUCAS 2 (Stryker, Lund, Sweden), CorePuls CPR (GS Elektromedizinische Gerate, Kaufering, Germany), Lifeline ARM ACC (Defibtech, Guilford, CT, USA), Life-Stat and Thumper (Michigan Instruments, Grand Rapids, MI, USA), and Weil MCC (SunLife, Shanghai, China).
Some of these piston devices have suction cups to return the chest to its neutral position between compressions. Vest-CPR (CardioLogic Systems, Inc, USA) delivers CC through a bladder-containing vest which is pneumatically inflated and deflated.6 The band-like device delivers compressions to the chest through a load distributing band (Autopulse, Zoll. Chelmsford, MA, USA). The devices differ in how they operate on the chest and consequently generate blood flow, and they differ regarding challenges, difficulties, easiness, benefits, and limitations. The Corpuls CPR, Weil MCC, and Life-Stat are adjustable regarding rate and depth and Thumper only depth. The LUCAS 2, AutoPulse and Lifeline ARM ACC are not adjustable.
There is a lack of data defining when during cardiac arrest treatment is the best time to deploy the mechanical CC device. Of the RCT, the LINC study deployed the device after the first shock was delivered, while the other studies did not follow a strict rule.1 In a study where the device was started as early as possible, Olsen et al. documented by analyzing the Circulation Improving Resuscitation Care trial (CIRC) data that the odds ratio for survival to hospital discharge showed a significant benefit for mechanical‐CPR vs. manual‐CPR in the subset of patients where CC continued for at least 16.5 minutes.12
There is a rationale for recommending that deployment should not delay the first defibrillation attempt or interfere with the first few minutes of BLS care. Many of the patients who will ultimately survive cardiac arrest, initially get return of spontaneous circulation in response to the first few minutes of BLS care: defibrillation and CPR.13 Rather than deploying a CC device during those important minutes, focus instead on applying the best possible BLS with manual CPR for the first few minutes. During this time, preparation for deployment may also be done if there are enough people on scene. In most clinical situations, this approach would result in deployment of the device after two rounds of CPR.
In general, deploying the device requires an additional pause in compressions, but thereafter, device use enables compressions to be more continuous than with manual CPR. For example, an analysis of data from a subset of the patients in the LINC RCT trial found that patients randomized to the mechanical CPR group had a 36-second median pause length for device deployment but, despite that pause, had a higher chest compression fraction than patients treated with manual CPR.(Esibov 2015) In the CIRC study hands-off during the first 10 minutes of CPR (interval the device was deployed) was 1.2% higher which equates to approximately seven seconds of additional time without compressions in the mechanical group.3
It is both important and challenging to minimize the pause in CPR that occurs during device deployment. Pauses substantially shorter than those measured in the RCTs can be achieved; with quality improvement programs emphasizing training, practice and post-event debriefing based on objective data, it is possible to deploy the device quickly and effectively. Levy et al. documented a hands-off interval of median (25’th, 75’th percentile) 7 sec (4, 14) with LUCAS deployment compared to 21 sec (15, 31) before they focused on it through training.13,14
In the years since the three RCT’s were published, we have learned about how to monitor and optimize use of the devices. These improvements have not yet been tested in a randomized controlled trial. Clinically, we should pay particular attention to interrupting chest compressions for too long when applying the device and using the mechanical CPR device too early or too late in the resuscitation.
Optimal Compression Point On the Chest
During CPR, how and where on the chest we compress determines the blood flow and gas exchange. Guidelines recommend compressing on the lower half of the sternum.7 Anatomically this corresponds to below the third rib. Compressions generate blood flow by compressing the ventricles of the heart (the cardiac pump theory) or increasing the pressure in the chest (the thoracic pump theory). During relaxation, the heart / blood vessels in the chest refill with blood.
It is suggested that the optimal compression point may be 3.2 cm left of sternum center and 1.6 cm caudal to the INL.8 However, compressing that far left of sternum center may be over ribs. Compressing directly on ribs cannot be recommended because fractures will lead to sharp rib fracture ends pointing toward the heart, potentially piercing the heart during compressions.
Mechanical devices are deployed in a fixed position, and the compression point may migrate during compressions due to rib or sternal fractures that will influence the compression angle on the chest (piston-based devices). Although compressions performed in suboptimal or wrong positions may generate less blood flow and reduce intact neurologic survival, none of the RCTs focused on identifying an optimal compression point on the chest based on objective effect measures.
In addition, devices based mostly on the thoracic pump theory may fail to generate sufficient chest recoil following compression when several rib and/or sternum fractures are created. The chest is no longer an “armed” spring, and the relaxation phase of the compression-decompression cycle does not generate enough negative intrathoracic pressure to facilitate refilling of the heart.
In practice, the optimal compression point varies from patient to patient. Magnetic resonance imaging (MRI) of patients with cardiac disease documented that the left ventricular outflow tract, aortic valve /- root are located under the sternum in the inter nipple line (INL) in 46% of patients and that the ventricles are located there in only 2%.15 Computed tomography (CT) demonstrated the ventricles were frequently (in 99%) beneath the 4th-6th rib and sometimes (in 36%) extended to beneath the 2nd-4th rib.
If compressions were performed at the INL, the ascending aorta, aortic root, or left ventricular outflow tract would be compressed in 80% of the patients, presumably impeding the desired blood flow. Cha et al found that compressing the lower end of the sternum, provided higher arterial pressures (peak systolic: 114±51 vs 95±42 mmHg, p=0.01; mean: 56±27 vs 50±23 mmHg, p=0.01) and ETCO2 (11.0±6.7 vs 9.6±6.9 mmHg, p=0.02) than standard compressions (lower half of the sternum).16 They recommended compression at the sterno-xiphoid junction.16
When using transesophageal echocardiography (TEE) during CPR, Hwang et al. found significant narrowing of the aorta root or left ventricular outflow tract. The area of maximal compression (left ventricle significantly compressed), was identified at the aorta/aorta valve in 20 patients (59%), left ventricular outflow tract in 14 (41%), and within 2 cm of the aorta valve in 79% of the patients.17 The left ventricular stroke volume correlated with area of maximal compression location (R2=0.165, p=0.017), with higher stroke volumes for this area locations closer to the ventricle.17
Taking all of this evidence together leads to a recommendation that the compression point to start with should be on the left part of the sternum (but on sternum) and caudal to the INL but cranial to the xiphoid.
The practical clinical question then is how to find the optimal compression point for a particular patient. The most obvious technique is to palpate for CC-generated pulse in the groin. If the pulse is absent or weak, change the compression point. Many services routinely measure ETCO2, and that measurement can be used to find the best CC point by searching for the highest ETCO2. Prehospital arterial lines are rarely inserted but pressure measured there can similarly be used. TEE can also be used but is for ED treatment.
During mechanical CC, adjusting the CC point is cumbersome; the device must be paused, repositioned, and restarted with a new evaluation of the effect. The same must be done during manual CC, but it is easier and faster. As providers of CPR, we need a device with which the CC point can be easily adjusted by sliding the part of the device touching the chest lateral, caudal or cranial, mimicking the flexibility of manual CC.
All RCT’s followed the European Resuscitation Council (ERC)/ American Heart Association (AHA) guidelines, which recommend compression depth of approximately 2 inches (5 cm), avoiding excessive depth (> 2.4 inches, 6 cm), and allowing full chest-wall recoil between compressions by avoiding leaning on the chest.7 Stiell et. al. documented in a retrospective analysis of 9,136 adult out-of-hospital cardiac arrest patients that received manual CPR, the adjusted odds ratio for survival to discharge was 1.04 (95% CI, 1.00-1.08) for each 5 mm increase in compression depth, 1.45 (95% CI, 1.20-1.76) for cases with >38mm compression depth, and 1.05 (95% CI, 1.03-1.08) for each 10% change in minutes with CCs in depth range.18 The highest survival was found between 40.3mm and 55.3mm with no difference between men and women.18 The same group found that adequate compression depth, per guidelines, was often not provided, and that CC rate and depth were inversely related (p<0.001); 53% of cases with CC rate >120 min-1 had depth <38.8mm.
Optimal compression depth is uncertain, and it is unclear whether compression depth should be recommended in absolute depth or relative to chest height of the patient. Some information regarding the force applied to the sternum/chest by CC devices and the degree of displacement achieved for different body sizes, sex, and age has been reported; compressing 5.3cm required 219 to 568 N of force, and that depth corresponded to a 20-28% anterior posterior (AP) compression of the chest, where the smallest chest receives more relative AP displacement.19
Some mechanical CPR devices have built-in technology measuring patient AP diameter. Today, they perform CC as absolute depth in mm but could be modified to compress a percentage of the AP diameter. Both AP diameter and the amount of force needed to reach a specific compression depth vary between individuals, Body Mass Index, and sex.19 The CorPuls CPR, Weil MCC, Life-Stat, Thumper and AutoPulse can adjust the compression depth, but this has not been investigated in a controlled RCT for outcome. The AutoPulse administers whole chest compressions through a Load Distributing Band at a consistent rate of 80±5 compressions per minute. Compression depth causes a chest displacement equal to a 20% reduction in AP chest diameter. This is calculated for each patient according to their chest size.
The devices in the RCT followed ERC/AHA guidelines recommendation of 100-120 CC min-1 based on improved survival and interdependency between rate and depth except for those studies using the AutoPulse (80 CC min-1).7 For manual CPR, increased CC rate reduces compression depth in a dose dependent manner, hastens rescuer fatigue, and increases incidence of incomplete chest release. Kilgannon et al. found that the highest rate of restoration of spontaneous circulation (ROSC) for hospital cardiac arrest was 120-130 CC min-1.20
By retrospectively evaluating 5 minutes of CPR of 20% of their sample, Idris et al documented that CC rate was associated positively with ROSC (p=0.01, p=0.012), with the highest ROSC (but not survival) occurring at 125 CC min-1.21 These results are inconclusive. The CC devices tested in RCT or used in clinical practice have fixed CC rates of 80 and 100 min-1 which seems reasonable until more conclusive data are provided.1-3,5
All RCT followed the ERC/AHA guidelines recommendation for ventilation but no description is available on how this was quality controlled. Whether ventilations are provided with bag mask (BMV), supraglottic device (SGDV), or endotracheal tube, CC and ventilations impact each, potentially decreasing filling of the heart, reducing stroke volume, causing gas leaks, increasing intrathoracic pressure and causing lung barotrauma (bleeding). Martin et al. instrumented twenty OHCA patients with thoracic aortic (Ao) and right atrial (RA) catheters on arrival in the emergency department.22 Five patients had one-minute trials of simultaneous compression and ventilation CPR (SCV-CPR). Ao-RA end diastolic gradients decreased in four of the five during SCV-CPR, leading the authors to concluded that, although SCV-CPR had been shown to improve carotid blood flow in human beings, it appeared to adversely affect myocardial perfusion.21
A much larger study, in which ambulance crews were randomly assigned to use simultaneous compression-ventilation SCV-CPR or conventional CPR to treat 924 prehospital patients, found that survival to hospital admission and to discharge were significantly higher in the conventional CPR group than the SCV-CPR group (p<.01). 23 There was statistically significant difference in the Glasgow coma scores between survivors in the two groups, both at 24 h post-hospital admission and at discharge. The lower survival in the SCV-CPR group likely reflects a deleterious effect of this resuscitation technique. Also noted was that 14% of the control patients and 6% of the experimental patients survived with manual CPR alone.23
The providers of manual CC and ventilation may adjust to each other to avoid impact. During continuous mechanical CC with manual ventilation this adjustment is challenging. When providing mechanical CC with SGDV, CC must be paused to avoid negatively impacting the positive pressure ventilation (PPV). Otherwise, oxygen gas is pressed into the esophagus, or leaks out of the mouth, and the patient is not ventilated adequately. When ventilating through an endotracheal tube, it is recommended to deliver one PPV every 10 CC.
However, this is exceedingly difficult to accomplish manually and has the same negative consequences as described with SGDV occur. Therefore, I suggest using a 30:2 ratio (mechanical CC: ventilations) for patients with BMV, SGDV or endotracheal tubes, at least until mechanical CC devices can operate with a 10:1 or 20:2 ratio (that is, with a longer pause after every 10th or 20th CC to facilitate ventilation).
Evaluation of Blood Flow During CPR
Direct measurement of blood flow generated during CPR is not yet feasible. Therefore, indirect measurements such as ETCO2,1711,24 intra-arterial blood pressure,25 and cerebral oximetry26 are used as surrogate measures of the hemodynamic effect of CC. Wide use of intra-arterial blood pressure is hindered by the need for advanced skills to insert catheters during CC, and use of cerebral oximetry may be hindered by its cost. ETCO2 is easy to apply and inexpensive and should be used in all cases to guide the provider to deliver best care. ETCO2 may also be used to identify the best point on the chest to deliver piston-based CC; no RCT has yet focused on this use of ETCO2. However, one study used ETCO2 as an endpoint comparing mechanical CC with manual and found a significant increase in favor of mechanical CC (p=0.04, no values provided).11
Three of the RCT delivered defibrillation without stopping the mechanical CC device. Guidelines advise a short period of CPR followed by rhythm analysis and shock delivery.7 Be careful to not let the arrival or deployment of a mechanical CC device delay delivery of the first shock. Steinberg et al. documented that manually delivering a shock without stopping CC is not beneficial.27 There was a significantly lower termination of fibrillation when the shock occurred during the compression phase of the compression decompression cycle.26 Until automatic technology becomes available coordinating shock delivery with CC phases, CC should be paused a maximum of 2 seconds and the shock delivered during this pause.
There is a large and growing body of evidence related to mechanical chest compression device use, but many important topics remain to be studied. I suggest that a high-performance user of these devices should:
- Focus on excellent BLS with manual compressions first, for at least the first two CPR cycles, before applying the device.
- Apply the device with only a brief pause in compressions; train, practice and review data to reduce this pause to less than 10 seconds.
- Find the individuals’ optimal compression point by palpating the CC pulse, measuring ETCO2 or using ultrasound (if available).
- Promptly reposition the device if it is visually out of correct position, ETCO2 is low, or palpated CC pulse is weak.
- Use 30:2 mode to facilitate ventilation, whether ventilating with bag/valve/mask, supraglottic device, or endotracheal tube.
- Pause briefly to defibrillate.
Conflict of Interest
The author is a member of the medical advisory board of Stryker. PI for the CIRC and LUCAS2 AD trials. Holds patents licenced to Zoll and Stryker from Oslo University Hospital Inven2.
1. Rubertsson S, Lindgren E, Smekal D, et al. Mechanical Chest Compressions and Simultaneous Defibrillation vs Conventional Cardiopulmonary Resuscitation in Out-of-Hospital Cardiac Arrest. The LINC Randomized Trial. JAMA. 2014;311(1):53-61.214.
2. Perkins GD, Lall R, Quinn T, et al. Mechanical versus manual chest compression for out-of-hospital cardiac arrest (PARAMEDIC): a pragmatic, cluster randomised controlled trial. Lancet 2015;385:947-955.
3. Wik L, Olsen JA, Persse D, et al. Manual vs. integrated automatic load-distributing band CPR with equal survival after out of hospital cardiac arrest. The randomized CIRC trial. Resuscitation 2014;85:741–748.
4. Gao C, Chen Y, Peng H, et al. Clinical evaluation of the AutoPulse automated chest compression device for out‐of‐hospital cardiac arrest in the northern district of Shanghai, China. Archives of Medical Science 2016;12(3):563‐70.
5. Hallstrom A, Rea TD, Sayre MR, et al. Manual chest compression vs use of an automated chest compression device during resuscitation following out‐of‐hospital cardiac arrest: a randomized trial. JAMA 2006;H295(22):2620‐8.
6. Halperin HR, Tsitlik JE, Gelfand M, et al. A preliminary study of cardiopulmonary resuscitation by circumferential compression of the chest with use of a pneumatic vest. New England Journal of Medicine 1993;329(11):762‐8.
7. Kleinman ME, Brennan EE, Goldberger ZD, et al. Part 3: adult basic life support and automated external defibrillation: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Resuscitation 2015;95:e43-69.
8. Buckler DG, Burke RV, Naim MY, et al. Association of Mechanical Cardiopulmonary Resuscitation Device Use With Cardiac Arrest Outcomes. A Population-Based Study Using the CARES Registry (Cardiac Arrest Registry to Enhance Survival). Circulation. 2016;134:2131–2133.
9. Wang PL, Brooks SC. Mechanical versus manual chest compressions for cardiac arrest. Cochrane Database of Systematic Reviews 2018, Issue 8. Art. No.: CD007260. DOI: 10.1002/14651858.CD007260.pub4.
10. Koster RW, Beenen LF, van der Boom EB, et al. Safety of mechanical chest compression devices AutoPulse and LUCAS in cardiac arrest: a randomized clinical trial for non‐inferiority. European Heart Journal 2017;38(40):3006‐13. [DOI: 10.1093/eurheartj/ehx318]
11. Dickinson ET, Verdile VP, Schneider RM, Salluzzo RF. Effectiveness of mechanical versus manual chest compressions in out‐of‐hospital cardiac arrest resuscitation: a pilot study. American Journal of Emergency Medicine 1998;16(3):289‐92.
12. Olsen JA, Lerner EB, Persse D, et al. Chest compression duration influences outcome between integrated load‐distributing band and manual CPR during cardiac arrest. Acta Anaesthesiologica Scandinavica 2015;60:222-229.
13. Levy M, Kern KB, Yost D, et al. Metrics of mechanical chest compression device use in out-of-hospital cardiac arrest. JACEP Open 2020; 1:1214-1221. https://doi.org/10.1002/emp2.12184
14. Levy M, Yost D, Walker RG, et al. A quality improvement initiative to optimize use of mechanical chest compression device within a high-performance CPR approach to out-of-hospital cardiac arrest resuscitation. Resuscitation 2015;92:32-37.
15. Nestaas S, Stensæth KH, Rosseland V, et al. Radiological assessment of chest compression point and achievable compression depth in cardiac patients. SJTREM 2016;24:54. DOI 10.1186/s13049-016-0245-0
16. Cha KC, Kim HJ, Shin HJ, et al. Hemodynanic effect of external chest compressions at the lower end of the sternum in cardiac arrest patients. J Emerg Med 2013;44:691-697
17. Hwang SO, Zhao PG, Choi HJ, et al. Compression of the left ventricular outflow tract during cardiopulmonary resuscitation. Acad Emerg Med 2009;16:928-933
18. Stiell IG, Brown SP, Nichol G, et al. What is the optimal chest compression depth during out of hospital cardiac arrest resuscitation of adult patients? Circulation 2014;130:1962-1970.
19. Beesems SG, Hardig BM, Nilsson A, Koster RW. Force and depth of mechanical chest compressions and their relation to chest height and gender in an out-of-hospital setting. Resuscitation 2015;91:67-72.
20. Kilgannon JH, Kirchhoff M, Pierce L, et al. Association between chest compression rates and clinical outcomes following in-hospital cardiac arrest at an academic tertiary hospital. Resuscitation 2017;110:154-161
21. Idris AH, Guffey D, Aufderheide TP, et al. The relationship between chest compression rates and outcomes from cardiac arrest. Circulation 2012;125:3004-3012.
22. Martin GB, Carden DL, Nowak RM, et al. Aortic and right atrial pressures during standard and simultaneous compression and ventilation CPR in human beings. Annals of Emerg Med 1986;15:125-130.
23. Krischer JP, Fine EG, Weisfeldt ML, et al. Comparison of prehospital conventional and simultaneous compression-ventilation cardiopulmonary resuscitation. Critical Care Medicine 1989, 17(12):1263-1269.
24. Falk JL, Rackow EC, Weil MH. End-tidal carbon dioxide concentration during cardiopulmonary resuscitation. NEJM 1988;318;607-611.
25. Paradis NA, Martin GB, Rivers EP, et al. Coronary Perfusion Pressure and the Return of Spontaneous Circulation in Human Cardiopulmonary Resuscitation. JAMA. 1990;263(8):1106-1113.
26. Wik L. Near-infrared spectroscopy during cardiopulmonary resuscitation and after restoration of spontaneous circulation: a valid technology? Curr Opin Crit Care 2016, 22:191–198.
27. Steinberg MT, Olsen JA, Brunborg C, et al. Defibrillation success during different phases of the mechanical chest compression cycle. Resuscitation 2016;103:99-105. doi: 10.1016/j.resuscitation.2016.01.031.