a b s t r a c t

Background: The classic technique of high quality chest compression (HQCC) during cardiopulmonary resuscita- tion (CPR) is based on the International Liaison Committee on Resuscitation guidelines which specify that the rescuer’s hands should maintain constant contact with the chest surface but should not lean upon it, in order to provide full chest recoil. Since end-tidal CO2 values have been shown to be a reliable indica- tor of CPR quality, we examined a method where classic HQCC was modified by a high impulse and palm lifting (HIPL) technique which merged rapid forceful compression with disconnection of the rescuer’s palm from the patient’s sternum during the recoil phase. The object of the study was to detect any differences in HIPL EtCO2 values in comparison with those from classic HQCC.

Methods: We report a prospective pilot study in which we compared EtCO2 readings achieved during 2 min of classic HQCC technique with readings after implementing 2 min of the HIPL technique during out-of-hospital CPR, provided by medical emergency response teams for cases of cardiac arrest.

Results: EtCO2 values obtained from16 cases who received HQCC followed by HIPL compressions showed a sig- nificant difference (p = 0.037) between the two techniques. Mean +- SD EtCO2 values after 2 min of each tech- nique were: HQCC: 18 +- 9 mmHg; HIPL: 27 +- 11 mmHg; followed by a further 2 min of HQCC: 19 +- 11 mmHg. Linear regression showed that the differences in EtCO2 were associated with non - significant changes in venti- lation rate (p = 0.493) and chest compression rate (p = 0.889).

Conclusions: The results obtained suggest that modifying HQCC with the HIPL technique led to a significant in- crease in EtCO2 values in comparison with classic HQCC, indicating an improvement in circulation during CPR. We think that these encouraging early results warrant a larger multi - centre study of HIPL.

(C) 2021

1. Introduction

The International Liaison Committee on Resuscitation (ILCOR) 2015 guidelines [1] for high quality chest compressions (HQCC) during Cardio-pulmonary resuscitation indicate the need for the assur- ance of the proper rate [2] and depth [3] of compressions. Also, the guidelines underline the need for full chest recoil [4] achieved by not leaning on the sternum at the end of the decompression phase [1]

Abbreviations: ALS, advanced life support; BLS, basic life support; CA, cardiac arrest; CCR, chest compression rate; CPR, cardio-pulmonary resuscitation; EtCO2, end-tidal car- bon dioxide; HIPL, high impulse and palm lifting; HQCC, high quality chest compressions; ILCOR, International Liaison Committee on Resuscitation; OHCA, out-of-hospital cardiac arrest; PEA, pulseless electrical activity; ROSC, return of spontaneous circulation; VR, ven- tilation rate.

* Corresponding author.

E-mail addresses: [email protected] (P. Woloszyn), [email protected] (I. Baumberg).

while maintaining constant contact of the rescuer’s hand with the chest wall.

It is believed that complete chest wall recoil improves hemodynam- ics during CPR by generating an appropriate negative intrathoracic pres- sure and thus drawing venous blood back to the chest and heart, providing adequate cardiac pre-load prior to the next chest compres- sion phase [7].

The ILCOR guidelines say little about the technique of the rescuer’s movements while providing chest compression/decompression, speci- fying only the intermediate parameter, a Duty cycle (the ratio of com- pression to decompression time) of 50:50. No indications are given about the specific kinetic details related to both phases.

Recommendations, other than those of ILCOR, of different chest

compression techniques have been published with the objective of achieving the best blood flow from chest compressions during the com- pression phase with a high impulse [8,9] and during decompression with a hands-off technique [10]. The high impulse concept was aban- doned in time, even though it was potentially a practical solution for

https://doi.org/10.1016/j.ajem.2021.08.030

0735-6757/(C) 2021

changing the duty cycle. The advantages of the hands-off technique were proven partially in a manikin study [10].

The ILCOR guidelines also emphasize that measurement of end-tidal carbon dioxide provides a non-invasive estimation of cardiac output and organ perfusion during cardiac arrest and may therefore be used to monitor the Quality of CPR [5,6].

In the study reported here, the objective was to examine how EtCO2 values recorded during out-of-hospital cardiac arrest (OHCA) differed between two techniques of chest compressions, namely classic HQCC, in accordance with the ILCOR 2015 recommendations, where the res- cuer maintains constant contact of his hand with chest wall and tries to spend an equal time on the compression and decompression phases and extended HQCC, where a high impulse and palm lifting (HIPL) tech- nique was used, modifying the classic HQCC approach. The assumption was made that both the chest compression rate (CCR) and ventilation rate (VR) as parameters affecting EtCO2 values during CPR, remained within the same range whilst performing the first and second compres- sion technique in each patient.

Our hypothesis was that EtCO2 values will depend only on the tech- nique of chest compression applied.

1. Methods

This prospective, experimental pilot study was done in Poland in the Gdansk, Elblag and Lodz areas of EMS operations between June 2018 and March 2020.

The research project was approved by the Bioethics Committee of the Medical University of Lodz, Poland (RNN/332/19/KE).

• 1. Patients

Because of the nature of this study, the study size was not previously established. The only patient inclusion criterion was to be in cardiac ar- rest from any cause, irrespective of No-flow time. Thus, all patients who underwent OHCA and received CPR provided by paramedical teams and two of three authors during their routine EMS shifts were enrolled into the research protocol.

Two exclusion criteria were applied. These were lack of the possibil- ity of making reliable EtCO2 measurements and return of spontaneous circulation (ROSC) or death before or during first stage (first 2 min) of the study protocol. After exclusions, a total of 16 patients were studied and the results reported below.

• 1. Study protocol and patient management

In every patient, before the protocol was applied, ILCOR 2015 ad- vanced life support (ALS) procedures with manual or mechanical HQCC were performed for some time by local, experienced emergency crews (three paramedics or two paramedics with a physician). All res- cue crews were elements of the national EMS system responding to car- diac arrest (CA) patients. Team leaders had a minimum experience of 5000 h practice in the pre - hospital setting and were regularly trained to follow current ILCOR guidelines.

All rescuers, except the two authors who participated in the research project were not aware of the purpose of the study and thus acted nor- mally according to standard CPR recommendations.

Fig. 1. Study protocol.

Description of the study protocol with each stage presented on the CPR time axis. Additionally, the bar trends of EtCO2 values (case no 15.) and data from the monitor screen printouts show how chest compressions/decompressions waveforms changed according to each of the techniques applied (case no 7.). For better visualization of the differences between the end of the decompression phase in both techniques the palm configuration is shown in each case. During a 6 min study sequence, ventilation was provided by means of bag-valve device by the same person maintaining a constant the ventilation rate in each patient as well as constant flow of O2 to the bag at 15 l/min. Timing of drug administration, rhythm checking and defibrillations was not affected by the research protocol and remained in accordance with ILCOR recommendations.

Fig. 2. Flow chart of the study protocol.

In each case, one of the authors was a member of the second medical team dispatched as a supportive crew on - scene and started the study protocol after arrival.

In cases where artificial ventilation had been started with a supraglottic airway device in place, an endotracheal tube was insert- ed by a member of the second medical crew. Following this, the EtCO2 monitoring device was connected and the research protocol started.

One of the paramedics from the first crew was always designated to perform bag-valve ventilation through all 3 stages of the protocol with as constant rate and tidal volume as possible. Ventilation using a porta- ble ventilator was not a technique of choice but if ALS procedures had been started with use of this device, ventilation was maintained in this manner until the end of the research protocol. During bag-valve ventilation it was not possible to measure the delivered tidal volume and observed chest rise and fall were used as indicators of consistent ventilation. The study protocol is shown in Figs. 1 and 2.

Animation 1 shows details of the differences between the classic HQCC and HIPL techniques.

• 1. Data collection

The following parameters were collected for each patient: age, gen- der, cause of cardiac arrest, initial rhythm, implementation of BLS before ALS or not, whether or not Mechanical chest compressions were imple- mented within the 1st stage of research protocol, indications for defi- brillation after 2 min of HIPL in cases with initial Pulseless electrical activity or asystole, and return of spontaneous circulation (ROSC) at or after the 2nd stage of the study protocol.

Recorded numerical variables were EtCO2, VR and CCR. They all were collected using a LIFEPAK 15 defibrillator (Physio-Control, Inc., Redmond, WA, USA) fitted with a side stream capnometer with microstream measurement technology allowing recording of EtCO2 and VR. The LIFEPAK15 performs self - testing and calibration after being switched on. A 3-lead array attached to the patient’s chest by means of self adhesive pads designed for ECG monitoring enabled re- cording of CCR and chest compression/decompression waveforms. These waveforms were created by frontal chest wall movements during compression and decompression.

Table 1

Measured results.

Patient number	EtCO2 mmHg (VR/CCR) in	EtCO2 mmHg (VR/CCR) in	EtCO2 mmHg (VR/CCR) in
	1-st stage	2-nd stage	3-rd stage
1	23 (14/129)	28 (16/113)	19 (16/-)
2	10 (13/-)	17 (13/123)	7 (13/147)
3	14 (20/131)	39 (17/138)	17 (17/-)
4	6 (6/116)	12 (14/116)	7 (14/117)
5	16 (9/101)	19 (9/101)	11 (10/120)
6	16 (13/-)	22 (16/-)	15 (16/-)
7	15 (14/139)	19 (12/128)	12 (6/133)
8	38 (11/114)	47 (11/124)	36 (11/121)	PEA - VF - Death
9	26 (8/101)	33 (8/108)	22 (7/102)
10	7 (11/107)	16 (14/102)	12 (6/111)
11	10 (13/127)	18 (15/127)	11 (12/123)
12	36 (16/102)	32 (17/102)	39 (17/102)
13	22 (12/-)	41 (13/95)	33 (13/92)	ASY-PEA-VF-ROSC
14	9 (9/106)	16 (14/108)	9 (13/100)	ASY - VF
15	18 (14/-)	31 (14/100)	23 (14/102)	ventilator, not bag-valve
16	21 (20/-)	45 (15/110)	33 (16/115)	ASY - VF - Death
mean +- SD EtCO2 CI	18 +- 9 (13 +- 4/116 +- 14) 14-22	27 +- 11 (14 +- 3/113 +- 13) 22-34	19 +- 11 (13 +- 4/114 +- 15) 14-24

Measured parameters obtained for each of the enrolled patients in each stage of the study. EtCO2 One-Way Kruskal-Wallis ANOVA p = 0.037, Effect Size - 0.1, Priori Power - 0.280. VR One-Way Kruskal-Wallis ANOVA p = 0.493, Effect Size - 0.013, Priori Power - 0.281. CCR One-Way Kruskal-Wallis ANOVA p = 0.889, Effect Size - 0.049, Priori Power - 0.217.

For analysis, numerical and graphical trend data of the observed pa- rameters were used. Values of CCR and VR were taken from the trend re- port printed by the monitoring device after all ALS procedures were finished. Also, the monitor printout strip was used to show how the chest compression waveforms differed between the classic HQCC and HIPL techniques.

• 1. Outcome

The main outcome was the comparison of EtCO2 values obtained from both chest compression techniques in each patient.

• 1. Statistical analysis

Each patient had three measurements of all numerical variables at 3 time points, during the 3 research stages. The values from 1st stage served as a reference. Values from 3rd stage served as a confirmation that those obtained during 2nd stage were not a matter of coincidence.

Descriptive statistics were determined for numerical variables col- lected from all included patients and expressed as mean +- SD and with 95% confidence interval after Shapiro-Wilk Normality Test analy- sis. Because of the small sample size and non-normal distribution within values obtained in the 3rd research stage, comparisons between them were conducted by using the Kruskal-Wallis ANOVA test. In addition, linear regression expressed as R² was performed to establish if changes in EtCO2 values were related to CCR and VR. A p value of <0.05 was con- sidered to indicate a significant difference between parameters ob- tained in both techniques. Effect Size and Priori Power were also performed. All calculations were made using free, online statistical soft- ware applications [11] and calculation charts in the desktop applications Numbers and Excel for Mac (Apple Inc. Cupertino, CA, USA).

1. Results

Results from the 16 patients included in the study are shown in Table 1.

Fig. 3. Scatter plots with linear regression lines for CCR vs EtCO2. In addition, R² values for each regression are presented.

Fig. 4. Scatter plots with linear regression lines for VR vs EtCO2. In addition, R² values for each regression are presented.

Analysis of these indicates that the HILP technique applied during 2nd stage of the protocol provided significantly higher EtCO2 mean values than classic HQCC technique applied during 1st and 3rd stage (18 +- 9 mmHg vs 27 +- 11 mmHg vs 19 +- 11 mmHg; p = 0.037). Linear regression showed a weak direct relationship between values of EtCO2 and VR and EtCO2 and CCR. This indicates that the rise of EtCO2 was re- lated to the technique of chest compression and decompression and was not influenced by changes in VR (p = 0.493) and CCR (p = 0.889). Data showing how EtCO2 values were affected by these two pa- rameters are shown in Figs 3 and 4.

• 1. General observations

In all enrolled cases, intravenous access was gained and artificial ventilation was started by the first responding ambulance crew at the beginning of ALS, 15 to 35 min before the study protocol was com- menced. The authors had no influence on the duration of that period as they were dispatched as part of the second, medical supportive team. None of the first rescue crews responded twice during the study period.

Recordings of the CCR parameter were sometimes affected by exter- nal artefacts such as additional movements or a poor connection be- tween ECG adhesive pads and the skin. For these reasons we were not able to establish complete CCR data in 6 cases. In one case (patient no. 6) we did not collected CCR values at all because of a monitoring cable failure.

All necessary efforts were made to sustain CCR at a comparable level in all stages of the study protocol, even if CCR were over the recom- mended range. This was done by using the authors’ prompt commands to slow down or speed up CCR if necessary.

This was not necessary in the case of VR which was performed by the same rescuer over all the protocol time.

In 4 cases (patients no. 8, 13, 14, 16), primary PEA or asystole in the 1st stage were converted into VF at the end of the 2nd stage and then defibrillated. It is possible that this conversion to a shockable rhythm in patients presenting with a non-defibrillation rhythm for prolonged time before and during the first stage of the protocol could be a casual result or the effect of HIPL technique applied in the 2nd stage of the protocol.

In two other cases, temporary ROSC was observed shortly after the 2nd stage of the protocol (EtCO2 change: 19-28 mmHg in the first

case and 29-52 mmHg in the second). In these cases the 3rd stage of re- search protocol was omitted. These ROSCs lasted 3 min and 1 min re- spectively and then reverted to CA again. For this reason, values of observed parameters obtained in these cases were not included in the statistical analysis.

In case no. 13, after the 3rd stage resulted in a lowering of EtCO2 values, HIPL was applied again when the EtCO2 readings rose again and ROSC was observed (after 20 min of CPR).

In another case (patient no. 12) classic HQCC gave higher values of EtCO2 compared with HIPL. There is no obvious explanation for this finding.

Data from cases 5. and 7. show how the classic HQCC and HIPL com-

pression/decompressions waveforms changed according to each of the compression techniques used (Fig. 1). When classic HQCC was applied, the waveforms appeared sinusoidal and regular. In the HIPL technique the waveform was narrowed and with higher amplitude as an example of shorter and forceful, high impulse compression.

1. Discussion

This pilot study indicates that the HIPL technique provides statisti- cally significant higher EtCO2 values than classic HQCC. These EtCO2 dif- ferences were related to the technique of chest compression and decompression and were not influenced by non-significant changes in ventilation and chest compression rate.

The current ILCOR recommendation suggesting a 50:50 duty cycle of chest compression is based essentially on animal and mathemati- cal studies [1], although one paper with human CPR data suggested that a different duty cycle of up to 40% for the compression phase was more effective in restoring spontaneous circulation [13]. Since we did not find any specific recommendations about how exactly Manual compressions of a 40% duty cycle should be made we believe that, in practice these could be probably achieved by means of short- ening the compression phase. Compression movements should then be performed briefly, more rapidly and forcefully and therefore imitating the contractions of the heart during systole as shown in Animation 1 in the supplementary materials. This technique of compression was described in the past as a high impulse technique. We believe it switches the duty cycle to a different ratio but our method of obtaining data did not allow adequate analysis of this hy- pothesis which should be the object of a future study. We only

gained graphical data from two cases showing changes in the mor- phology of the chest compression waveform caused by the high im- pulse (a narrow complex with high amplitude).

The ILCOR guidelines recommend that the rescuer should maintain constant contact of his hands with the chest wall and should not lean on the chest during the decompression phase. This recommendation may lead to confusion if compared with practical observations de- scribed by other authors [14-17] which estimated that, along with the suggested technique the average residual force applied to chest wall at the end of decompression phase was significant. This is important in CPR since restricted recoil of the chest decreases the lung capacity caus- ing an increase in intrathoracic pressure, a decrease in coronary perfu- sion pressure, Myocardial blood flow and venous return and thus cardiac output [18].

Immediate and full chest recoil may be quickly facilitated by remov- ing the residual force from the chest during the decompression phase. This was previously called the ‘hands-off’ technique [10]. Based partly on a manikin study, it has been shown that this technique is 129 times more likely to allow complete chest wall recoil compared with the standard, constant contact hand position. No differences in accuracy of Hand placement and depth of compressions were observed.

The hands-off technique also increases Chest compression release velocity which is thought to be an important factor influencing a greater survival rate to hospital discharge after cardiac arrest [19,20]. Aufderheide [21] hypothesized that achieving faster chest recoil would augment intrathoracic suction and thus improve venous return to the heart.

Our study may support the hypothesis that achieving as much chest recoil as possible enhances blood flow as indicated by an increase in the observed ETCO2 levels with the HIPL technique.

Keeping CCR and VR steady in all stages of the study protocol pro- vided data about EtCO2 differences related to the chest compression techniques used.

As the CCR and the VR parameter differences and low R² value de- rived from Linear regression analysis were not significant, both of these parameters had little influence on EtCO2 values. Our results sup- port the hypothesis that differences in kinetics of HIPL vs classic HQCC applied to individual patients played a dominant role in observed differ- ences in EtCO2 readings.

Changing to a new technique of chest compression to allow full chest recoil may not be easy. Complete disconnection of the rescuer’s hands from the patient’s sternum in the hands-off technique may lead to con- fusion among rescuers, since the current ILCOR recommendation is to keep constant contact between the hands and the chest.

However, we think that natural chest wall recoil could be also achieved using a similar technique by lifting the rescuer’s metacarpus above the sternum immediately after finishing the high impulse com- pression phase and with only the distal parts of rescuer’s fingers to maintain the recommended constant contact of the hand with the chest wall. We suggest calling this technique ‘palm lifting’.

Palm lifting may be compared with the five finger fulcrum technique described previously by Aufderheide [6] but with one difference; with palm lifting, 4 fingers (not always crossed) remain in constant contact with the chest wall to fulfill the constant contact requirement.

Some limitations of our study should be noted.

It was done on patients presenting with CA during normal emer- gency response practice and thus entry selection to the study was re- stricted. In addition, we report data from only a limited number of cases. Clearly, a larger scale multi - center study is required to validate our preliminary results.

The exact no-flow duration was unknown due to lack of the accurate data in medical records of the first rescue crew.

Only the VR was measured and no exact measurements of expired tidal volume were available due to the use of a self-reforming bag. The exact minute ventilation as a one of the factors influencing EtCO2 values remained unknown.

The HIPL technique was applied by the second rescue team after a prolonged period of conventional CPR. Thus its influence on circulation during the initial stages of cardiac arrest remains unknown.

The authors of the study were the only ones who performed HIPL and they did not deliver HQCC.

The emerging HIPL technique does not represent current standard practice. However, it may be considered in the clinical scenarios when classic HQCC technique is used without success.

1. Conclusions

The results obtained in this pilot study suggest that modifying HQCC with HIPL significantly increases EtCO2 values, indicating an improved circulation. Moreover, recorded differences were unrelated to non- significant changes in ventilation and chest compression rates.

We feel that these encouraging preliminary results highlight the im- portance of the modification of chest compression technique and war- rant future larger scale, Multi-center trials.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influ- ence the work reported in this paper.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2021.08.030.

References

1. Perkins GD, Handley AJ, Koster RW, et al. European Resuscitation Council Guidelines for Resuscitation 2015: Section 2. Adult basic life support and automated external defibrillation. Resuscitation. 2015;95:81-99.
2. Idris AH, Guffey D, Pepe PE, et al. Chest compression rates and survival following out-of-hospital cardiac arrest. Crit Care Med. 2015;43:840-8.
3. Stiell IG, Brown SP, Nichol G, et al. What is the optimal chest compression depth dur- ing out-of-hospital cardiac arrest resuscitation of adult patients? Circulation. 2014; 130:1962-70.
4. Niles DE, Sutton RM, Nadkarni VM, et al. Prevalence and hemodynamic effects of leaning during CPR. Resuscitation. 2011;82:S23-6.
5. Sheak KR, Wiebe DJ, Leary M, et al. Quantitative relationship between end-tidal car- bon dioxide and CPR quality during both in-hospital and out-of-hospital cardiac ar- rest. Resuscitation. 2015;89:149-54.
6. Sandroni C, De Santis P, D’Arrigo S. Capnographyduringcardiacarrest. Resuscitation. 2018;132:73-7.
7. Aufderheide TP, Pirrallo RG, Yannopoulos D, et al. Incomplete chest wall decompres- sion: a clinical evaluation of CPR performance by EMS personnel and assessment of alternative manual chest compression-decompression techniques. Resuscitation. 2005;64:353-62.
8. Maier GW, Tyson Jr GS, Olsen CO, et al. The physiology of external cardiac massage: High-impulse cardiopulmonary resuscitation. Circulation. 1984;70:86-101.
9. Swenson RD, Weaver WD, Niskanen RA, et al. Hemodynamics in humans during conventional and experimental methods of cardiopulmonary resuscitation. Circula- tion. 1988;78:630-9.
10. Aufderheide TP, Pirrallo RG, Yannopoulos D, et al. Incomplete chest wall decompres- sion: a clinical evaluation of CPR performance by EMS personnel and assessment of alternative manual chest compression-decompression techniques. Resuscitation. 2005;64:353-62.
11. Statistics. http://www.statskingdom.com/index.html; 2020https://www. socscistatistics.comhttps://www.danielsoper.com/statcalc/default.aspx. (accessed 03.09.2020).
12. Woloszyn P, Baumberg I. Cardiac arrest following electrocution using unconven- tional CPR: A case study. Anestezjologia i Ratownictwo. 2015;9:381-6.
13. Johnson B, Coult J, Fahrenbruch C, et al. Cardiopulmonary resuscitation duty cycle in out-of-hospital cardiac arrest. Resuscitation. 2015;87:86-90.
14. Tomlinson A, Nysaether J, Kramer-Johansen J, et al. compression force-depth rela- tionship during out-of-hospital cardiopulmonary resuscitation. Resuscitation. 2007;72:364-70.
15. Fried DA, Leary M, Smith DA, et al. The prevalence of chest compression leaning dur- ing in-hospital cardiopulmonary resuscitation. Resuscitation. 2011;82:1019-24.
16. Niles D, Nysaether J, Sutton R, et al. Leaning is common during in-hospital pediatric CPR, and decreased with automated corrective feedback. Resuscitation. 2009;80: 553-7.
17. Zuercher M, Hilwig RW, Ranger-Moore J, et al. Leaning during chest compressions impairs cardiac output and left ventricular myocardial blood flow in piglet cardiac arrest. Crit Care Med. 2010;38:1141-6.
18. Yannopoulos D, McKnite S, Aufderheide TP, et al. Effects of incomplete chest wall de- compression during cardiopulmonary resuscitation on coronary and cerebral perfu- sion pressures in a porcine model of cardiac arrest. Resuscitation. 2005;64:363-72.
19. Cheskes S, Common MR, Byers AP, et al. The association between chest compression release velocity and outcomes from out-of-hospital cardiac arrest. Resuscitation. 2015;86:38-43.
20. Kovacs A, Vadeboncoeur TF, Stolz U, et al. Chest compression release velocity: Asso- ciation with survival and Favorable neurologic outcome after out-of-hospital cardiac arrest. Resuscitation. 2015;92:107-14.
21. Aufderheide TP, Frascone RJ, Wayne MA, et al. Standard cardiopulmonary resuscita- tion versus active compression-decompression cardiopulmonary resuscitation with augmentation of negative intrathoracic pressure for out-of-hospital cardiac arrest: a randomised trial. Lancet. 2011;377:301-11.