bs_bs_banner
Emergency Medicine Australasia (2014) 26, 585–590
doi: 10.1111/1742-6723.12307
ORIGINAL RESEARCH
Chest compression with kneeling posture in hospital cardiopulmonary resuscitation: A randomised crossover simulation study Jaehoon OH,1 Youngjoon CHEE,2 Taeho LIM,1 Youngsuk CHO3 and In Young KIM4 1 Department of Emergency Medicine, College of Medicine, Hanyang University, Seoul, Korea, 2School of Electrical Engineering, University of Ulsan, Ulsan, Korea, 3Department of Emergency Medicine, College of Medicine, Hallym University, Seoul, Korea, and 4Department of Biomedical Engineering, College of Medicine, Hanyang University, Seoul, Korea
Abstract Purpose: We suggest an alternative chest compression (CC) in kneeling posture using a ‘kneeling stool’ on which the performer kneels beside the patient on a bed in-hospital. In kneeling posture, we can maintain high quality cardiopulmonary resuscitation (CPR) without the bed height adjustment, which is necessary and inconvenient in standing posture. Methods: This study is a randomised crossover trial with 38 participants working in one ED. The first group knelt on the kneeling stool beside a manikin placed on a bed, whereas the second group stood on a step stool with the manikin at knee level using bed height adjustment. All the participants performed continuous chest compression for 5 min without audiovisual feedback. After that, the posture was changed in each group. The parameters of CC quality (CC depth, rate, accuracy, and incomplete chest recoil), visual analogue scale (VAS) for fatigue and pain, and preference of participants were compared between the two groups. Results: The data of 33 participants in both postures were analysed
following exclusion of five participants. In the comparisons overall and per minute between the two postures, the parameters and VAS do not differ significantly (all P > 0.05) except for the median 1st CC rate that was faster in kneeling posture than in standing posture, P = 0.01). Twenty-three performers preferred the kneeling posture. Conclusions: A kneeling posture with a kneeling stool were preferred by participants, which have shown similar results in CC parameters and VAS with a standing posture on a stepstool with bed height adjustment during inhospital CPR. Key words: cardiac arrest, cardiopulmonary resuscitation, compression, posture, simulation.
Introduction The suggested posture for chest compression (CC), which is the key factor for high quality cardiopulmonary resuscitation (CPR), known as the ‘kneeling posture’, requires the rescuer to kneel as close to the patient as possible, place his or her body directly above the patient’s chest, straighten both arms, and place the heel of the
Correspondence: Professor Youngjoon Chee, Biomedical Engineering Office, University of Ulsan, 102 Daehak-Ro, Nam-Gu, Ulsan 680-749, Korea. Email: yjchee @ulsan.ac.kr Jaehoon Oh, MD, PhD, Fellow, Emergency Physician; Youngjoon Chee, PhD, Associate Professor; Taeho Lim, MD, PhD, Professor, Emergency Physician; Youngsuk Cho, MD, Clinical Associate Professor, Emergency Physician; In Young Kim, MD, PhD, Professor. Trial registration: Clinicaltrials.gov (NCT02043028) Accepted 24 August 2014
Key findings • Participants could perform the chest compression using the kneeling stool equivalent to the standing posture with bed height adjustment and stepstool those are not needed for kneeling stool during in-hospital CPR. • There was no difference in fatigue and pain in both postures. • Participants preferred the kneeling posture on the kneeling stool compared to standing posture.
palm on the lower half of the sternum with the fingers interlocked.1 The CC performers in a hospital setting typically use the ‘standing posture’, which requires the rescuer to stand beside the patient instead of kneeling.2,3 Several studies have shown that the bed height should be adjusted to the knee to midthigh level of the performer for high quality CPR in the standing posture.4–6 A step stool and bed adjustment mechanism facilitate adjustment to the proper level of the performer.7–9 Appropriate posture might be important for reducing potential injuries and minimising fatigue of CC performers. The quality of chest compressions during CPR for cardiac arrest decreases dramatically after a short time primarily because of rescuer fatigue.10,11 Foo et al. found that the kneeling posture is better than the standing posture in terms of rescuer fatigue.12 In these studies, CC performers knelt beside a patient on the floor or on a bed with a kneeling posture or stood beside a patient on
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
586
J OH ET AL.
a bed with a standing posture using a step stool or bed adjustment mechanism. Kneeling on the bed beside the patient might not be suitable for the CC performer and patient, and the kneeling posture on the floor is not typically performed in hospitals. In the standing posture, it is not easy for a hospital CC performer to maintain the proper bed height level especially during a change for CC performer role every 2 min. A step stool with the proper height for the wide height range of rescuers is not always available. The adjustment of bed height requires much effort and time and is not practical for rotational CPR in hospitals. These factors might lead to low quality CC and aggravation of fatigue and pain in rescuers. To perform CC in the kneeling posture in hospital, we designed a stage with stairs, named the ‘kneeling stool’, on which a CC performer kneels beside a patient on a bed. This work is a validation study to demonstrate that the kneeling stool could be used for high quality hospital CPR with the kneeling posture. We hypothesised that the quality of chest compression with a kneeling posture using the kneeling stool is equal to or superior to CC with a standing posture using the height adjustment mechanism of the bed.
Materials and methods Study design and setting This randomised controlled crossover trial was approved by the Institutional Review Board at Hanyang University Hospital (Seoul, Korea) and was conducted 13–28 August 2013. Written informed consent was obtained before enrollment for all participants.
Heinrich University, Düsseldorf, German) with a 0.05 α error and a 0.8 power was used to detect a difference of 5 mm 13 chest compression depth (CCD) between the two groups through our pilot study with eight participants who have not been included in data of this study before starting of this study.
Equipment and environmental characteristics We designed and implemented the kneeling stool to perform CPR with the kneeling posture for a patient on a bed. The frame is constructed of an aluminum alloy. The size is 570 (width) × 598 (depth) × 600 mm (height). The weight is 9.3 kg, which is easily moved as needed. For easy rotation between rescuers, a step plate was inserted. To prevent pain to the knees of the kneeling CC performer, a 40 mm thick sponge cushion was placed on the top plate and fixed to the frame. A standard hospital bed frame (Transport stretcher®, 760 × 2110 mm, 228 kg, Stryker Co., Kalamazoo, MI, USA), a foam mattress (660 × 920 × 80 mm, soft foam with polyurethane coverage, Stryker Co.), a backboard (450 × 600 × 10 mm, 3 kg Lifeline Plastic, Sung Shim Medical Co., Bucheon, Korea), and a step stool (395 × 450 × 410 mm, Gunica Co., Gyeongsangnam-do, Korea) were used in the experiment. A Resusci Annie Modular System Skill Reporter ® manikin (9.9 kg, Laerdal Medical, Orpington, UK) was used to perform CPR in the evaluations. We collected the data to a laptop computer using VAM software version 1.30.19 Beta (Laerdal Medical). We added weights to the manikin to equal 34 kg for simulating the upper body weight of an adult human as in a previous study.14
Study participants Forty doctors and nurses with American Heart Association Basic Life Support (AHA BLS) provider certification working at one ED participated in this study. Participants with heart, wrist, low back disease, who were pregnant were excluded. The minimum sample size was 30 subjects, and an analysis with G-power 3.1.2® (Heine
Data collection Participants were divided into first and second group. By random drawing lots, subjects of odd number were allocated to the first group randomly with subjects of even number in the second group allocation. Each performer in the first group knelt on the kneeling stool beside the manikin on a bed and
compressed the chest of the manikin (CCs with a kneeling posture, PK). Each performer in the second group stood on a step stool beside the manikin on a bed and compressed the chest of the manikin (CCs with a standing posture, PS). When the performer stood beside the manikin on a bed, the height of the manikin’s back was adjusted in height to the knee level of the provider using the step stool and the bed height adjustment mechanism. When performing CPR, it is recommended to change compression performers every 2 min to avoid fatigue.1 In this study, all the participants performed continuous CCs for 5 min without audio-visual feedback to estimate the CC parameters and the fatigue of performers in every minute. After 2 weeks, the posture used for the CCs by each group was changed to minimise any carryover effects. Data indicating the quality of CCs during 5 min in each participant were collected by analysing the graph produced by the Resusci Annie Modular System Skill Reporter® and separated into five consecutive 1 min sectors. Each participant recorded his or her fatigue and pain levels on a visual analogue scale (VAS, score 0 indicates ‘no pain and fatigue’ and score 10 indicates ‘unbearable pain and fatigue’) for each minute during 5 min of CCs. After conclusion of the experiment, the participants selected a preferred CC posture in terms of the safety and the quality of the CCs.
Primary outcomes We measured: (i) CCD; (ii) the proportions of accurate compressions depth (ACD, % of the numbers of the accurate compressions depth ≥ 5 cm/ the total number of compressions); (iii) the rate of compressions per minute (CCR); and (iv) the numbers of incomplete chest decompression (ICD) ≥ 1 cm of residual leaning/ the total number of compressions, with analysing the graph produced by the Laerdal PC Skill Reporting System. Parameters of CC quality were analysed overall and per minute. We collected the VAS score of the overall fatigue and pain at each posture per minute and for the total CC time as well as the preference of the
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
587
CHEST COMPRESSION WITH KNEELING POSTURE IN-HOSPITAL
performers for PK or PS as secondary outcomes.
assessed for eligibility (n = 40)
excluded (n = 2) • exclusion criteria (n = 2)
Statistical analysis The data were compiled using a standard spreadsheet application (Excel, Microsoft, Redmond, WA, USA) and analysed using the Statistical Package for the Social Sciences (SPSS) 18.0 KO for Windows (SPSS Inc., Chicago, IL, USA). We generated descriptive statistics and presented them as frequency and percentages for categorical data and medians with interquartile ranges (IQR) (non-normal distribution) or mean with standard deviation (SD) (normal distribution) for continuous data. The characteristics and data for the participants were compared between the groups with the Wilcoxon signed rank test (non-normal distribution) or paired t-test (normal distribution) for continuous measures and with McNemar tests for categorical measures. P < 0.05 was considered statistically significant.
enrolled and randomised (n = 38)
allocated to stare with the first group (n = 19) CC with kneeling posture (PK, n = 19)
1st day
crossover, lost to follow up (n = 1)
allocated to stare with the second group (n = 19) CC with standing posture (PS, n = 19) crossover, lost to follow up (n = 1)
allocated to continue with the first group (n = 18) CC with standing posture (PS, n = 18)
15th day
allocated to continue with the second group (n = 18) CC with kneeling posture (PK, n = 18)
analysis with exclusion due to recorder error (n = 4)
analysed (n = 66) PK, n = 33, PS, n = 33
Results Forty participants were eligible, and 38 were enrolled in this study after the exclusion of two participants for having a history of herniated lumbar disc history. One participant in each group was lost to the crossover test after 2 weeks. The data of four participants were excluded because of technical recording problems in the manikin in both groups. The data of 33 participants in both PK and PS were analysed following the exclusion of five participants (Fig. 1). The baseline characteristics of the analysed participants are shown in Table 1.
Overall and every minute comparison of CC quality parameters between PK and PS The median of the CCDs (IQR) was 51.0 mm (39.5–57.0) in P K and 50.0 mm (36.0–57.0) in PS (P = 0.30). The proportion of ACD was 52.7% in PK and 50.3% in PS (P = 1.00). The median of the CCRs was 106.0 times/ min (100.0–113.0) in PK and 105.0 times/min (97.5–109.5) in PS (P = 0.06) (Fig. 2). The comparison of the quality of CCs per minute between PK and PS
Figure 1. Experimental flow chart of the participants. CC, chest compression; PK, CC with kneeling posture on the kneeling stool; PS, CC with standing posture and bed height adjustment at the knee level.
TABLE 1.
Baseline features of participants
Number of participants (n = 33) Qualification, n (%) Physicians
Attending Resident Intern
4 13 4 12
Nurse Gender, n (%) M:F Age, mean (SD) (years) Height, mean (SD) (cm) BMI, mean (SD) (kg/m2) Low bodyweight (BMI < 18.5), n (%)
(12%) (40%) (12%) (36%)
19 (58%):14 (42%) 30.5 (5.2) 169.3 (7.2) 27.5 (2.7) 1 (3%)
BMI, body mass index; SD, standard deviation.
are shown in Table 2. The overall and every minute median of the ICD were 0.00 (0.00–0.00) in both postures (all P > 0.05). The CCD and ACD decreased with time in all cases whereas the CCR and ICD did not changed.
When we compared the results per minute between postures, the parameters were not significantly different for the 1st median CC rate, which was faster in kneeling posture than in standing posture with P < 0.01 (all P > 0.05).
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
588
30
60
20
40
10
20
0
0 (98.5–113.0) (97.5–108.5)
(34.0–57.0) (34.5–57.0)
(100.0–111.5) (97.0–108.0)
50.0 47.0 0.20 51.5 45.5 0.63 106.0 105.0 0.09 (102.5–114.5) (97.0–109.0)
2nd minute
(37.5–57.5) (37.5–58.0) (48.5–58.5) (45.0–59.0)
1st minute
51.0 49.0 0.13 54.5 45.5 0.25 105.0 105.0 0.05 CCR median (IQR) (times/min)
The quality parameters of CC with kneeling posture using the kneeling stool (PK) are not different from those in standing posture using a step stool and an adjustment of the bed height (PS) during CCs in a manikin simulation of hospital CPR (all P > 0.05) except for the median 1st CCR (P = 0.01). The median 1st CCR rate of 106.0 times/min (102.5–114.5) in PK was more than 100 times/min. This
ACD (%)
Discussion
54.0 56.0 0.31 72.7 69.7 1.00 106.0 104.0 0.01*
The median of the fatigue and pain score was 5.0 (4.0–7.0) in PK and 7.0 (4.0–8.0) in PS (P = 0.13). The comparison per minute of the fatigue and pain score between the two postures is shown in Table 3. There was no significant difference between the two postures (all P > 0.05). Twenty-three performers (70%) preferred PK, seven performers (21%) preferred PS and three performers (9%) indicated having no preference.
finding is in compliance with current guidelines, rather than in PS.1,3 The CC posture beside the bed using the kneeling stool did not show the inferiority in all CC parameters over a standing posture. This posture might be an alternative technique for highquality CC in terms of the CC quality parameters. According to current CPR guidelines, rescuers should switch CC roles every 2 min to decrease fatigue and pain as well as to perform high quality CPR.3,15 Our results that the CCD and ACD decreased with time in all cases, especially beyond 2 min could support this switch role. Several studies have reported various results on the correlation of performer fatigue or pain with CC postures.16,17 These studies used the heart rate, oxygen consumption (VO2), lactate level and subjective scoring (such as the VAS and the Borg scale) to estimate the fatigue and pain of CPR performers. Foo et al. found that the kneeling posture is better than the standing posture in terms of rescuer fatigue.12 In our study, the fatigue and pain score using the VAS is not different between the two postures during and after 5 min CPR (P > 0.05). A kneeling stool did not show increased
PK PS P PK PS P PK PS P
Overall and every minute comparison of fatigue and pain score with the CC performer preference for CC postures PK and PS
CCD median (IQR) (mm)
Figure 2. Overall comparison of the quality of CCs between the kneeling stool (PK) posture and the standing posture (PS). CCs, chest compressions; CCD, chest compression depth; CCR, chest compression rate; PK, CC with kneeling posture on the kneeling stool; PS, CC with standing posture and bed height adjustment at the knee level. P < 0.05 is significant.
Parameter of CCs
type of posture
3rd minute
PS A comparison of the chest compression parameters at each minute for the kneeling and standing postures
PK
PS
TABLE 2.
PK
*P < 0.05 is significant. ACD, the numbers of the accurate compressions depth ≥ 5 cm/the total number of compressions; CCD, chest compression depth; CCR, chest compression rate; CCs, chest compressions; IQR, interquartile range; PK, CC with kneeling posture on the kneeling stool; PS, CC with standing posture and bed height adjustment at the knee level.
80
(99.5–115.5) (99.0–110.5)
40
(35.5–57.0) (32.0–54.5)
100
5th minute
50
44.0 41.0 0.51 42.4 45.5 1.00 110 106.0 0.14
120
4th minute
60
(35.5–57.0) (33.0–56.0)
140
rate [times/min]
depth [mm]
CCR P=0.06
47.0 45.0 0.52 42.4 45.5 1.00 107.0 105.0 0.06
CCD P=0.30
70
(99.0–115.0) (99.0–112.5)
J OH ET AL.
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
589
CHEST COMPRESSION WITH KNEELING POSTURE IN-HOSPITAL
TABLE 3. Self-reported fatigue and pain scores using visual analogue scale (VAS) during the 5 min experimental session between the kneeling (PK) and standing postures (PS)
VAS of fatigue and pain median (IQR)
PK PS P
1st minute
2nd minute
3rd minute
4th minute
5th minute
Overall
3.0 (2.0–4.0) 3.0 (2.0–4.0) 0.20
4.0 (3.0–6.0) 5.0 (4.0–6.0) 0.16
6.0 (4.0–7.0) 6.0 (4.0–7.5) 0.06
7.0 (5.5–8.0) 8.0 (6.0–8.0) 0.05
8.0 (7.0–9.0) 9.0 (7.5–9.0) 0.23
5.0 (4.0–7.0) 7.0 (4.0–8.0) 0.13
P < 0.05 is significant. IQR, interquartile range; PK, CC with kneeling posture on the kneeling stool; PS, CC with standing posture and bed height adjustment at the knee level.
fatigue and pain than a standing posture at his or her knee level throughout CPR. However, the kneeling posture using a kneeling stool might be better than the standing posture without strict control of the bed height. The over the head technique and straddle technique might be helpful for CPR in confined spaces.2 The over the head technique could be useful for a single CPR rescuer who should compress the chest and ventilate the breath of the patient on the floor.18,19 This technique for CPR is impossible when the patient is placed on a bed. A rescuer with straddle technique who could not be separated rapidly during defibrillation will be exposed to high energy shock and delay rapid defibrillation for patient. These techniques could cause an interruption to other procedures in hospital CPR such as endotracheal intubation and injection. With these techniques, a kneeling posture on the bed, not on a kneeling stool, could be used for a patient on a bed such as in hospital CPR. This posture is impossible for especially small bed or an obese patient. This posture might cause injury to the patient and performer and require more time for a switch, resulting in interruption of the CCs. Use of the kneeling stool could solve these problems. There are several limitations to the study. We used a manikin that differs from actual human patients in chest stiffness in spite of adjustment for the manikin weight. We did not find the quality of CPR and fatigue of a performer with consideration of a deformity and patient chest stiffness timely. We also did not know the usefulness of this kneeling stool according to various bodyweights or heights
of CC performers. We examined this study with the fixed height of a kneeling stool to be set at the lowest level of a bed. Some time is required for a bed to be adjusted to the lowest level. The kneeling posture on a kneeling stool, which differs from the kneeling posture on the floor, causes the performer’s body to rise from the floor and might induce anxiety in some performers. The volume of a kneeling stool was approximately three times larger than a stepstool. This bulky device also could not be used in limited resuscitation space or in a bed transportation. For clinical application, the device should provide stability and safety for the patient and rescuer. The hands-off time that occurs for during a shift between CC performers each 2 min during CPR should be minimised for high quality CPR. Longer hands off time could be required in a kneeling posture with a kneeling stool than that in a standing posture. After this experiment was finished, we measured the hands-off time for a switch of performers from the last CC of one performer to the first CC of the next performer with a stopwatch ten times in kneeling stool. The median (IQR) of hands-off time in kneeling stool was 3.3 s (3.0–3.5). We hypothesise that training in use of the kneeling stool might decrease this time. In conclusion, we designed a kneeling stool to enable a CC performer to kneel beside a patient on the bed and compress the patient’s chest in hospital CPR. A kneeling posture with a kneeling stool was preferred by participants, which have shown similar results in CC parameters and VAS fatigue and pain with a standing posture on a stepstool with bed height adjustment of the patient during inhospital CPR.
Acknowledgements This work was supported by the research fund of the University of Ulsan (2013-0141).
Author contributions JO and YC designed the study and drafted this article. TL and IYK made the kneeling stool and collected the data. YC analysed and interpreted the data. All authors revised it critically for important intellectual content and final approval of the version to be submitted.
Competing interests None declared.
References 1. Koster RW, Baubinb MA, Bossaert LL et al. European Resuscitation Council Guidelines for Resuscitation 2010 Section 2. Adult basic life support and use of automated external defibrillators. Resuscitation 2010; 81: 1277–92. 2. Handley AJ, Handley JA. Performing chest compressions in a confined space. Resuscitation 2004; 61: 55– 61. 3. Berg RA, Hemphill R, Abella BS et al. Part 5: adult basic life support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010; 122: S685– 705. 4. Cho J, Oh JH, Park YS et al. Effects of bed height on the performance of chest compressions. Emerg. Med. J. 2009; 26: 807–10. 5. Lewinsohn A, Sherren PB, Wijayatilake DS. The effects of bed height and time
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
590
6.
7.
8.
9.
J OH ET AL.
on the quality of chest compressions delivered during cardiopulmonary resuscitation: a randomised crossover simulation study. Emerg. Med. J. 2012; 29: 660–3. Perkins GD, Smith CM, Augre C et al. Effects of a backboard, bed height, and operator position on compression depth during simulated resuscitation. Intensive Care Med. 2006; 32: 1632–5. Edelson DP, Call SL, Yuen TC et al. The impact of a step stool on cardiopulmonary resuscitation: a crossover mannequin study. Resuscitation 2012; 83: 874–8. Oh JH, Kim CW, Kim SE et al. Comparison of chest compressions in the standing position beside a bed at knee level and the kneeling position: a nonrandomised, single-blind, cross-over trial. Emerg. Med. J. 2014; 31: 533– 5. Lee DH, Kim CW, Kim SE et al. Use of step stool during resuscitation improved the quality of chest compression in simulated resuscitation. Emerg. Med. Australas. 2012; 24: 369– 73.
10. Ashton A, McCluskey A, Gwinnutt CL et al. Effect of rescuer fatigue on performance of continuous external chest compressions over 3 min. Resuscitation 2002; 55: 151–5. 11. Min MK, Yeom SR, Ryu JH et al. A 10-s rest improves chest compression quality during hands-only cardiopulmonary resuscitation: a prospective, randomized crossover study using a manikin model. Resuscitation 2013; 84: 1279–84. 12. Foo NP, Chang JH, Lin HJ et al. Rescuer fatigue and cardiopulmonary resuscitation positions: a randomized controlled crossover trial. Resuscitation 2010; 81: 579–84. 13. Edelson DP, Abella BS, KramerJohansen J et al. Effects of compression depth and pre-shock pauses predict defibrillation failure during cardiac arrest. Resuscitation 2006; 71: 137–45. 14. Oh J, Chee Y, Song Y et al. A novel method to decrease mattress compression during CPR using a mattress compression cover and a vacuum pump. Resuscitation 2013; 84: 987– 91.
15. Manders S, Geijsel FE. Alternating providers during continuous chest compressions for cardiac arrest: every minute or every two minutes? Resuscitation 2009; 80: 1015–8. 16. Chi C-H, Tsou J-Y, Su F-C. Effects of rescuer position on the kinematics of cardiopulmonary resuscitation (CPR) and the force of delivered compressions. Resuscitation 2008; 76: 69– 75. 17. Jones AY, Lee RY. Rescuer’s position and energy consumption, spinal kinetics, and effectiveness of simulated cardiac compression. Am. J. Crit. Care 2008; 17: 417–25. 18. Perkins GD, Stephenson BT, Smith CM et al. A comparison between overthe-head and standard cardiopulmonary resuscitation. Resuscitation 2004; 61: 155–61. 19. Hüpfl M, Duma A, Uray T et al. Overthe-head cardiopulmonary resuscitation improves efficacy in basic life support performed by professional medical personnel with a single rescuer: a simulation study. Anesth. Analg. 2005; 101: 200–5.
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Copyright of Emergency Medicine Australasia is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Emergency Medicine Australasia (2014) 26, 585–590
doi: 10.1111/1742-6723.12307
ORIGINAL RESEARCH
Chest compression with kneeling posture in hospital cardiopulmonary resuscitation: A randomised crossover simulation study Jaehoon OH,1 Youngjoon CHEE,2 Taeho LIM,1 Youngsuk CHO3 and In Young KIM4 1 Department of Emergency Medicine, College of Medicine, Hanyang University, Seoul, Korea, 2School of Electrical Engineering, University of Ulsan, Ulsan, Korea, 3Department of Emergency Medicine, College of Medicine, Hallym University, Seoul, Korea, and 4Department of Biomedical Engineering, College of Medicine, Hanyang University, Seoul, Korea
Abstract Purpose: We suggest an alternative chest compression (CC) in kneeling posture using a ‘kneeling stool’ on which the performer kneels beside the patient on a bed in-hospital. In kneeling posture, we can maintain high quality cardiopulmonary resuscitation (CPR) without the bed height adjustment, which is necessary and inconvenient in standing posture. Methods: This study is a randomised crossover trial with 38 participants working in one ED. The first group knelt on the kneeling stool beside a manikin placed on a bed, whereas the second group stood on a step stool with the manikin at knee level using bed height adjustment. All the participants performed continuous chest compression for 5 min without audiovisual feedback. After that, the posture was changed in each group. The parameters of CC quality (CC depth, rate, accuracy, and incomplete chest recoil), visual analogue scale (VAS) for fatigue and pain, and preference of participants were compared between the two groups. Results: The data of 33 participants in both postures were analysed
following exclusion of five participants. In the comparisons overall and per minute between the two postures, the parameters and VAS do not differ significantly (all P > 0.05) except for the median 1st CC rate that was faster in kneeling posture than in standing posture, P = 0.01). Twenty-three performers preferred the kneeling posture. Conclusions: A kneeling posture with a kneeling stool were preferred by participants, which have shown similar results in CC parameters and VAS with a standing posture on a stepstool with bed height adjustment during inhospital CPR. Key words: cardiac arrest, cardiopulmonary resuscitation, compression, posture, simulation.
Introduction The suggested posture for chest compression (CC), which is the key factor for high quality cardiopulmonary resuscitation (CPR), known as the ‘kneeling posture’, requires the rescuer to kneel as close to the patient as possible, place his or her body directly above the patient’s chest, straighten both arms, and place the heel of the
Correspondence: Professor Youngjoon Chee, Biomedical Engineering Office, University of Ulsan, 102 Daehak-Ro, Nam-Gu, Ulsan 680-749, Korea. Email: yjchee @ulsan.ac.kr Jaehoon Oh, MD, PhD, Fellow, Emergency Physician; Youngjoon Chee, PhD, Associate Professor; Taeho Lim, MD, PhD, Professor, Emergency Physician; Youngsuk Cho, MD, Clinical Associate Professor, Emergency Physician; In Young Kim, MD, PhD, Professor. Trial registration: Clinicaltrials.gov (NCT02043028) Accepted 24 August 2014
Key findings • Participants could perform the chest compression using the kneeling stool equivalent to the standing posture with bed height adjustment and stepstool those are not needed for kneeling stool during in-hospital CPR. • There was no difference in fatigue and pain in both postures. • Participants preferred the kneeling posture on the kneeling stool compared to standing posture.
palm on the lower half of the sternum with the fingers interlocked.1 The CC performers in a hospital setting typically use the ‘standing posture’, which requires the rescuer to stand beside the patient instead of kneeling.2,3 Several studies have shown that the bed height should be adjusted to the knee to midthigh level of the performer for high quality CPR in the standing posture.4–6 A step stool and bed adjustment mechanism facilitate adjustment to the proper level of the performer.7–9 Appropriate posture might be important for reducing potential injuries and minimising fatigue of CC performers. The quality of chest compressions during CPR for cardiac arrest decreases dramatically after a short time primarily because of rescuer fatigue.10,11 Foo et al. found that the kneeling posture is better than the standing posture in terms of rescuer fatigue.12 In these studies, CC performers knelt beside a patient on the floor or on a bed with a kneeling posture or stood beside a patient on
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
586
J OH ET AL.
a bed with a standing posture using a step stool or bed adjustment mechanism. Kneeling on the bed beside the patient might not be suitable for the CC performer and patient, and the kneeling posture on the floor is not typically performed in hospitals. In the standing posture, it is not easy for a hospital CC performer to maintain the proper bed height level especially during a change for CC performer role every 2 min. A step stool with the proper height for the wide height range of rescuers is not always available. The adjustment of bed height requires much effort and time and is not practical for rotational CPR in hospitals. These factors might lead to low quality CC and aggravation of fatigue and pain in rescuers. To perform CC in the kneeling posture in hospital, we designed a stage with stairs, named the ‘kneeling stool’, on which a CC performer kneels beside a patient on a bed. This work is a validation study to demonstrate that the kneeling stool could be used for high quality hospital CPR with the kneeling posture. We hypothesised that the quality of chest compression with a kneeling posture using the kneeling stool is equal to or superior to CC with a standing posture using the height adjustment mechanism of the bed.
Materials and methods Study design and setting This randomised controlled crossover trial was approved by the Institutional Review Board at Hanyang University Hospital (Seoul, Korea) and was conducted 13–28 August 2013. Written informed consent was obtained before enrollment for all participants.
Heinrich University, Düsseldorf, German) with a 0.05 α error and a 0.8 power was used to detect a difference of 5 mm 13 chest compression depth (CCD) between the two groups through our pilot study with eight participants who have not been included in data of this study before starting of this study.
Equipment and environmental characteristics We designed and implemented the kneeling stool to perform CPR with the kneeling posture for a patient on a bed. The frame is constructed of an aluminum alloy. The size is 570 (width) × 598 (depth) × 600 mm (height). The weight is 9.3 kg, which is easily moved as needed. For easy rotation between rescuers, a step plate was inserted. To prevent pain to the knees of the kneeling CC performer, a 40 mm thick sponge cushion was placed on the top plate and fixed to the frame. A standard hospital bed frame (Transport stretcher®, 760 × 2110 mm, 228 kg, Stryker Co., Kalamazoo, MI, USA), a foam mattress (660 × 920 × 80 mm, soft foam with polyurethane coverage, Stryker Co.), a backboard (450 × 600 × 10 mm, 3 kg Lifeline Plastic, Sung Shim Medical Co., Bucheon, Korea), and a step stool (395 × 450 × 410 mm, Gunica Co., Gyeongsangnam-do, Korea) were used in the experiment. A Resusci Annie Modular System Skill Reporter ® manikin (9.9 kg, Laerdal Medical, Orpington, UK) was used to perform CPR in the evaluations. We collected the data to a laptop computer using VAM software version 1.30.19 Beta (Laerdal Medical). We added weights to the manikin to equal 34 kg for simulating the upper body weight of an adult human as in a previous study.14
Study participants Forty doctors and nurses with American Heart Association Basic Life Support (AHA BLS) provider certification working at one ED participated in this study. Participants with heart, wrist, low back disease, who were pregnant were excluded. The minimum sample size was 30 subjects, and an analysis with G-power 3.1.2® (Heine
Data collection Participants were divided into first and second group. By random drawing lots, subjects of odd number were allocated to the first group randomly with subjects of even number in the second group allocation. Each performer in the first group knelt on the kneeling stool beside the manikin on a bed and
compressed the chest of the manikin (CCs with a kneeling posture, PK). Each performer in the second group stood on a step stool beside the manikin on a bed and compressed the chest of the manikin (CCs with a standing posture, PS). When the performer stood beside the manikin on a bed, the height of the manikin’s back was adjusted in height to the knee level of the provider using the step stool and the bed height adjustment mechanism. When performing CPR, it is recommended to change compression performers every 2 min to avoid fatigue.1 In this study, all the participants performed continuous CCs for 5 min without audio-visual feedback to estimate the CC parameters and the fatigue of performers in every minute. After 2 weeks, the posture used for the CCs by each group was changed to minimise any carryover effects. Data indicating the quality of CCs during 5 min in each participant were collected by analysing the graph produced by the Resusci Annie Modular System Skill Reporter® and separated into five consecutive 1 min sectors. Each participant recorded his or her fatigue and pain levels on a visual analogue scale (VAS, score 0 indicates ‘no pain and fatigue’ and score 10 indicates ‘unbearable pain and fatigue’) for each minute during 5 min of CCs. After conclusion of the experiment, the participants selected a preferred CC posture in terms of the safety and the quality of the CCs.
Primary outcomes We measured: (i) CCD; (ii) the proportions of accurate compressions depth (ACD, % of the numbers of the accurate compressions depth ≥ 5 cm/ the total number of compressions); (iii) the rate of compressions per minute (CCR); and (iv) the numbers of incomplete chest decompression (ICD) ≥ 1 cm of residual leaning/ the total number of compressions, with analysing the graph produced by the Laerdal PC Skill Reporting System. Parameters of CC quality were analysed overall and per minute. We collected the VAS score of the overall fatigue and pain at each posture per minute and for the total CC time as well as the preference of the
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
587
CHEST COMPRESSION WITH KNEELING POSTURE IN-HOSPITAL
performers for PK or PS as secondary outcomes.
assessed for eligibility (n = 40)
excluded (n = 2) • exclusion criteria (n = 2)
Statistical analysis The data were compiled using a standard spreadsheet application (Excel, Microsoft, Redmond, WA, USA) and analysed using the Statistical Package for the Social Sciences (SPSS) 18.0 KO for Windows (SPSS Inc., Chicago, IL, USA). We generated descriptive statistics and presented them as frequency and percentages for categorical data and medians with interquartile ranges (IQR) (non-normal distribution) or mean with standard deviation (SD) (normal distribution) for continuous data. The characteristics and data for the participants were compared between the groups with the Wilcoxon signed rank test (non-normal distribution) or paired t-test (normal distribution) for continuous measures and with McNemar tests for categorical measures. P < 0.05 was considered statistically significant.
enrolled and randomised (n = 38)
allocated to stare with the first group (n = 19) CC with kneeling posture (PK, n = 19)
1st day
crossover, lost to follow up (n = 1)
allocated to stare with the second group (n = 19) CC with standing posture (PS, n = 19) crossover, lost to follow up (n = 1)
allocated to continue with the first group (n = 18) CC with standing posture (PS, n = 18)
15th day
allocated to continue with the second group (n = 18) CC with kneeling posture (PK, n = 18)
analysis with exclusion due to recorder error (n = 4)
analysed (n = 66) PK, n = 33, PS, n = 33
Results Forty participants were eligible, and 38 were enrolled in this study after the exclusion of two participants for having a history of herniated lumbar disc history. One participant in each group was lost to the crossover test after 2 weeks. The data of four participants were excluded because of technical recording problems in the manikin in both groups. The data of 33 participants in both PK and PS were analysed following the exclusion of five participants (Fig. 1). The baseline characteristics of the analysed participants are shown in Table 1.
Overall and every minute comparison of CC quality parameters between PK and PS The median of the CCDs (IQR) was 51.0 mm (39.5–57.0) in P K and 50.0 mm (36.0–57.0) in PS (P = 0.30). The proportion of ACD was 52.7% in PK and 50.3% in PS (P = 1.00). The median of the CCRs was 106.0 times/ min (100.0–113.0) in PK and 105.0 times/min (97.5–109.5) in PS (P = 0.06) (Fig. 2). The comparison of the quality of CCs per minute between PK and PS
Figure 1. Experimental flow chart of the participants. CC, chest compression; PK, CC with kneeling posture on the kneeling stool; PS, CC with standing posture and bed height adjustment at the knee level.
TABLE 1.
Baseline features of participants
Number of participants (n = 33) Qualification, n (%) Physicians
Attending Resident Intern
4 13 4 12
Nurse Gender, n (%) M:F Age, mean (SD) (years) Height, mean (SD) (cm) BMI, mean (SD) (kg/m2) Low bodyweight (BMI < 18.5), n (%)
(12%) (40%) (12%) (36%)
19 (58%):14 (42%) 30.5 (5.2) 169.3 (7.2) 27.5 (2.7) 1 (3%)
BMI, body mass index; SD, standard deviation.
are shown in Table 2. The overall and every minute median of the ICD were 0.00 (0.00–0.00) in both postures (all P > 0.05). The CCD and ACD decreased with time in all cases whereas the CCR and ICD did not changed.
When we compared the results per minute between postures, the parameters were not significantly different for the 1st median CC rate, which was faster in kneeling posture than in standing posture with P < 0.01 (all P > 0.05).
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
588
30
60
20
40
10
20
0
0 (98.5–113.0) (97.5–108.5)
(34.0–57.0) (34.5–57.0)
(100.0–111.5) (97.0–108.0)
50.0 47.0 0.20 51.5 45.5 0.63 106.0 105.0 0.09 (102.5–114.5) (97.0–109.0)
2nd minute
(37.5–57.5) (37.5–58.0) (48.5–58.5) (45.0–59.0)
1st minute
51.0 49.0 0.13 54.5 45.5 0.25 105.0 105.0 0.05 CCR median (IQR) (times/min)
The quality parameters of CC with kneeling posture using the kneeling stool (PK) are not different from those in standing posture using a step stool and an adjustment of the bed height (PS) during CCs in a manikin simulation of hospital CPR (all P > 0.05) except for the median 1st CCR (P = 0.01). The median 1st CCR rate of 106.0 times/min (102.5–114.5) in PK was more than 100 times/min. This
ACD (%)
Discussion
54.0 56.0 0.31 72.7 69.7 1.00 106.0 104.0 0.01*
The median of the fatigue and pain score was 5.0 (4.0–7.0) in PK and 7.0 (4.0–8.0) in PS (P = 0.13). The comparison per minute of the fatigue and pain score between the two postures is shown in Table 3. There was no significant difference between the two postures (all P > 0.05). Twenty-three performers (70%) preferred PK, seven performers (21%) preferred PS and three performers (9%) indicated having no preference.
finding is in compliance with current guidelines, rather than in PS.1,3 The CC posture beside the bed using the kneeling stool did not show the inferiority in all CC parameters over a standing posture. This posture might be an alternative technique for highquality CC in terms of the CC quality parameters. According to current CPR guidelines, rescuers should switch CC roles every 2 min to decrease fatigue and pain as well as to perform high quality CPR.3,15 Our results that the CCD and ACD decreased with time in all cases, especially beyond 2 min could support this switch role. Several studies have reported various results on the correlation of performer fatigue or pain with CC postures.16,17 These studies used the heart rate, oxygen consumption (VO2), lactate level and subjective scoring (such as the VAS and the Borg scale) to estimate the fatigue and pain of CPR performers. Foo et al. found that the kneeling posture is better than the standing posture in terms of rescuer fatigue.12 In our study, the fatigue and pain score using the VAS is not different between the two postures during and after 5 min CPR (P > 0.05). A kneeling stool did not show increased
PK PS P PK PS P PK PS P
Overall and every minute comparison of fatigue and pain score with the CC performer preference for CC postures PK and PS
CCD median (IQR) (mm)
Figure 2. Overall comparison of the quality of CCs between the kneeling stool (PK) posture and the standing posture (PS). CCs, chest compressions; CCD, chest compression depth; CCR, chest compression rate; PK, CC with kneeling posture on the kneeling stool; PS, CC with standing posture and bed height adjustment at the knee level. P < 0.05 is significant.
Parameter of CCs
type of posture
3rd minute
PS A comparison of the chest compression parameters at each minute for the kneeling and standing postures
PK
PS
TABLE 2.
PK
*P < 0.05 is significant. ACD, the numbers of the accurate compressions depth ≥ 5 cm/the total number of compressions; CCD, chest compression depth; CCR, chest compression rate; CCs, chest compressions; IQR, interquartile range; PK, CC with kneeling posture on the kneeling stool; PS, CC with standing posture and bed height adjustment at the knee level.
80
(99.5–115.5) (99.0–110.5)
40
(35.5–57.0) (32.0–54.5)
100
5th minute
50
44.0 41.0 0.51 42.4 45.5 1.00 110 106.0 0.14
120
4th minute
60
(35.5–57.0) (33.0–56.0)
140
rate [times/min]
depth [mm]
CCR P=0.06
47.0 45.0 0.52 42.4 45.5 1.00 107.0 105.0 0.06
CCD P=0.30
70
(99.0–115.0) (99.0–112.5)
J OH ET AL.
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
589
CHEST COMPRESSION WITH KNEELING POSTURE IN-HOSPITAL
TABLE 3. Self-reported fatigue and pain scores using visual analogue scale (VAS) during the 5 min experimental session between the kneeling (PK) and standing postures (PS)
VAS of fatigue and pain median (IQR)
PK PS P
1st minute
2nd minute
3rd minute
4th minute
5th minute
Overall
3.0 (2.0–4.0) 3.0 (2.0–4.0) 0.20
4.0 (3.0–6.0) 5.0 (4.0–6.0) 0.16
6.0 (4.0–7.0) 6.0 (4.0–7.5) 0.06
7.0 (5.5–8.0) 8.0 (6.0–8.0) 0.05
8.0 (7.0–9.0) 9.0 (7.5–9.0) 0.23
5.0 (4.0–7.0) 7.0 (4.0–8.0) 0.13
P < 0.05 is significant. IQR, interquartile range; PK, CC with kneeling posture on the kneeling stool; PS, CC with standing posture and bed height adjustment at the knee level.
fatigue and pain than a standing posture at his or her knee level throughout CPR. However, the kneeling posture using a kneeling stool might be better than the standing posture without strict control of the bed height. The over the head technique and straddle technique might be helpful for CPR in confined spaces.2 The over the head technique could be useful for a single CPR rescuer who should compress the chest and ventilate the breath of the patient on the floor.18,19 This technique for CPR is impossible when the patient is placed on a bed. A rescuer with straddle technique who could not be separated rapidly during defibrillation will be exposed to high energy shock and delay rapid defibrillation for patient. These techniques could cause an interruption to other procedures in hospital CPR such as endotracheal intubation and injection. With these techniques, a kneeling posture on the bed, not on a kneeling stool, could be used for a patient on a bed such as in hospital CPR. This posture is impossible for especially small bed or an obese patient. This posture might cause injury to the patient and performer and require more time for a switch, resulting in interruption of the CCs. Use of the kneeling stool could solve these problems. There are several limitations to the study. We used a manikin that differs from actual human patients in chest stiffness in spite of adjustment for the manikin weight. We did not find the quality of CPR and fatigue of a performer with consideration of a deformity and patient chest stiffness timely. We also did not know the usefulness of this kneeling stool according to various bodyweights or heights
of CC performers. We examined this study with the fixed height of a kneeling stool to be set at the lowest level of a bed. Some time is required for a bed to be adjusted to the lowest level. The kneeling posture on a kneeling stool, which differs from the kneeling posture on the floor, causes the performer’s body to rise from the floor and might induce anxiety in some performers. The volume of a kneeling stool was approximately three times larger than a stepstool. This bulky device also could not be used in limited resuscitation space or in a bed transportation. For clinical application, the device should provide stability and safety for the patient and rescuer. The hands-off time that occurs for during a shift between CC performers each 2 min during CPR should be minimised for high quality CPR. Longer hands off time could be required in a kneeling posture with a kneeling stool than that in a standing posture. After this experiment was finished, we measured the hands-off time for a switch of performers from the last CC of one performer to the first CC of the next performer with a stopwatch ten times in kneeling stool. The median (IQR) of hands-off time in kneeling stool was 3.3 s (3.0–3.5). We hypothesise that training in use of the kneeling stool might decrease this time. In conclusion, we designed a kneeling stool to enable a CC performer to kneel beside a patient on the bed and compress the patient’s chest in hospital CPR. A kneeling posture with a kneeling stool was preferred by participants, which have shown similar results in CC parameters and VAS fatigue and pain with a standing posture on a stepstool with bed height adjustment of the patient during inhospital CPR.
Acknowledgements This work was supported by the research fund of the University of Ulsan (2013-0141).
Author contributions JO and YC designed the study and drafted this article. TL and IYK made the kneeling stool and collected the data. YC analysed and interpreted the data. All authors revised it critically for important intellectual content and final approval of the version to be submitted.
Competing interests None declared.
References 1. Koster RW, Baubinb MA, Bossaert LL et al. European Resuscitation Council Guidelines for Resuscitation 2010 Section 2. Adult basic life support and use of automated external defibrillators. Resuscitation 2010; 81: 1277–92. 2. Handley AJ, Handley JA. Performing chest compressions in a confined space. Resuscitation 2004; 61: 55– 61. 3. Berg RA, Hemphill R, Abella BS et al. Part 5: adult basic life support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010; 122: S685– 705. 4. Cho J, Oh JH, Park YS et al. Effects of bed height on the performance of chest compressions. Emerg. Med. J. 2009; 26: 807–10. 5. Lewinsohn A, Sherren PB, Wijayatilake DS. The effects of bed height and time
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
590
6.
7.
8.
9.
J OH ET AL.
on the quality of chest compressions delivered during cardiopulmonary resuscitation: a randomised crossover simulation study. Emerg. Med. J. 2012; 29: 660–3. Perkins GD, Smith CM, Augre C et al. Effects of a backboard, bed height, and operator position on compression depth during simulated resuscitation. Intensive Care Med. 2006; 32: 1632–5. Edelson DP, Call SL, Yuen TC et al. The impact of a step stool on cardiopulmonary resuscitation: a crossover mannequin study. Resuscitation 2012; 83: 874–8. Oh JH, Kim CW, Kim SE et al. Comparison of chest compressions in the standing position beside a bed at knee level and the kneeling position: a nonrandomised, single-blind, cross-over trial. Emerg. Med. J. 2014; 31: 533– 5. Lee DH, Kim CW, Kim SE et al. Use of step stool during resuscitation improved the quality of chest compression in simulated resuscitation. Emerg. Med. Australas. 2012; 24: 369– 73.
10. Ashton A, McCluskey A, Gwinnutt CL et al. Effect of rescuer fatigue on performance of continuous external chest compressions over 3 min. Resuscitation 2002; 55: 151–5. 11. Min MK, Yeom SR, Ryu JH et al. A 10-s rest improves chest compression quality during hands-only cardiopulmonary resuscitation: a prospective, randomized crossover study using a manikin model. Resuscitation 2013; 84: 1279–84. 12. Foo NP, Chang JH, Lin HJ et al. Rescuer fatigue and cardiopulmonary resuscitation positions: a randomized controlled crossover trial. Resuscitation 2010; 81: 579–84. 13. Edelson DP, Abella BS, KramerJohansen J et al. Effects of compression depth and pre-shock pauses predict defibrillation failure during cardiac arrest. Resuscitation 2006; 71: 137–45. 14. Oh J, Chee Y, Song Y et al. A novel method to decrease mattress compression during CPR using a mattress compression cover and a vacuum pump. Resuscitation 2013; 84: 987– 91.
15. Manders S, Geijsel FE. Alternating providers during continuous chest compressions for cardiac arrest: every minute or every two minutes? Resuscitation 2009; 80: 1015–8. 16. Chi C-H, Tsou J-Y, Su F-C. Effects of rescuer position on the kinematics of cardiopulmonary resuscitation (CPR) and the force of delivered compressions. Resuscitation 2008; 76: 69– 75. 17. Jones AY, Lee RY. Rescuer’s position and energy consumption, spinal kinetics, and effectiveness of simulated cardiac compression. Am. J. Crit. Care 2008; 17: 417–25. 18. Perkins GD, Stephenson BT, Smith CM et al. A comparison between overthe-head and standard cardiopulmonary resuscitation. Resuscitation 2004; 61: 155–61. 19. Hüpfl M, Duma A, Uray T et al. Overthe-head cardiopulmonary resuscitation improves efficacy in basic life support performed by professional medical personnel with a single rescuer: a simulation study. Anesth. Analg. 2005; 101: 200–5.
© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Copyright of Emergency Medicine Australasia is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
