Resuscitation 64 (2005) 103–108
CPREzyTM: an evaluation during simulated cardiac
arrest on a hospital bed?
Gavin D Perkins∗, Colette Augr´ e, Helen Rogers, Michael Allan, David R Thickett
Division of Medical Sciences, University of Birmingham, Birmingham B152TT, UK
Received 23 July 2004; received in revised form 23 August 2004; accepted 23 August 2004
CPREzyTMis a new adjunct designed to improve the application of manual external chest compressions (ECC) during cardiopulmonary
resuscitation (CPR). The aim of this study was to determine the effect of using the CPREzyTMdevice compared to standard CPR during the
simulated resuscitation of a patient on a hospital bed. Twenty medical student volunteers were randomised using a cross over trial design to
perform 3min of continuous ECC using CPREzyTMand standard CPR. There was a significant improvement in ECC depth with CPREzyTM
compared to standard CPR 42.9 (4.4)mm versus 34.2 (7.6):mm, P = 0.001; 95% CI d.f. 4.4–12.9mm. This translated to a reduction in the
percentage of shallow compressions (<38mm) with CPREzyTM16 (23)% compared to standard CPR 59 (44)%, P = 0.003. There was a small
increase in the percentage of compression regarded excessive (>51mm): CPREzyTM6.5 (19)% versus standard CPR 0 (0.1)%. P = 0.012).
There was no difference in compression rate or duty cycle between techniques. Equal numbers of participants (40% in each group) performed
one of more incorrectly placed chest compression. However the total number of incorrect compressions was higher for the CPREzyTMgroup
animal and clinical studies are required to validate this finding in vivo and to see if it translates to an improvement in outcome in human
victims of cardiac arrest.
© 2004 Elsevier Ireland Ltd. All rights reserved.
Keywords: Basic life support; Cardiac massage; Chest compression; Cardiopulmonary resuscitation
The performance and quality of basic life support are im-
Most in-hospital arrests occur while the victim is on a hos-
pital bed . We have demonstrated previously that external
chest compressions (ECC) performed on a hospital bed are
inferior to those undertaken with the victim on the floor .
flation of air filled mattresses ; by using a back-board; by
kneeling on the bed next to the victim or by altering bed
?A Spanish and Portuguese translated version of the Abstract and
Keywords of this article appears at 10.1016/j.resuscitation.2004.08.011.
∗Corresponding author. Fax: +44 121 443 2494.
E-mail address: firstname.lastname@example.org (G.D. Perkins).
CPREzyTMis a portable device designed to improve the
efficacy of manual chest compressions during cardiopul-
monary resuscitation (CPR). Early evaluations of this device
reported improvements in compression rate, the number of
correct compressions and a reduction in the deterioration of
CPR performance over time . The aim of this study was
to evaluate the effectiveness of the CPREzyTMdevice to im-
prove chest compression efficacy in a cardiac arrest model
simulating a patient in cardiac arrest on a hospital bed.
external chest compressions (Fig. 1). It has a series of lights
0300-9572/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved.
G.D. Perkins et al. / Resuscitation 64 (2005) 103–108
Fig. 1. (a) CPREzyTMdevice: (A) compression pad, (B) light indicator; (b)
the CPREzyTMdevice is placed on the sternum during CPR.
on its upper surface that illuminate depending on the amount
of pressure generated by each chest compression. Activation
pressures for the lights are quoted as: 1 light (child) 23kg;
2 lights (small adult) 32kg, 3 lights (average adult) 41kg, 4
lights (large adult) 50kg, 5 lights (caution) 54kg . When
compression pressure is released the lights switch off. In ad-
dition to the description of patient size next to the lights (i.e.
child to large adult), the approximate weight of the patient
is also identified. This is not the same as the activation pres-
sure for the lights. The device contains an audible tone that
bleeps at a rate of 100 times per minute. It is designed to be
placed on the sternum during chest compression in order to
improve the accuracy of CPR. A diagram on the front of the
device shows where it should be placed on the patient when
2.2. Pilot study
surface. However, a greater force is required when perform-
the additional effect of mattress compression. In order to de-
termine the optimal compression force required to achieve a
compression depth of 40–50mm, six volunteers performed
1min of chest compressions aiming to illuminate 1, 2, 3, 4
or all 5 lights.
2.3. Principle study design
The study was approved by the Executive Dean at the
Medical school, University of Birmingham. Students gave
was a randomised controlled cross-over trial. Medical stu-
dent volunteers, trained as European Resuscitation Council
BLS/AED Instructors as part of our peer led resuscitation
training initiative were recruited [7,8]. Students performed,
dard CPR. Randomisation was performed using odd-even al-
(SPSSinc, IL, USA). Each phase of testing was separated by
a period of 7 days.
Participants received 2min instruction on the use of the
CPREzyTMdevice at the start of the study and were al-
lowed a short 1min period of familiarisation/revision with
CPREzyTM/standard CPR. Participants were instructed to il-
luminate four lights when using the CPREzyTMdevice. The
study required participants to perform 3min of continuous
ECC (with and without CPREzyTM) on a Laerdal Resusci
Anne manikin placed on a hospital bed and SoftformTMmat-
in the intubated patient. The bed was adjusted to its lowest
position, which was 45cm above the ground.
Data on compression performance were collected using
the VAM system software (version 1.30.19 Beta) and down-
loaded to a laptop computer (Dell Latitude D600, Dell, UK)
as previously described . The VAM system was used for
Data for each participant were exported to a Microsoft Excel
2000 spreadsheet. The VAM system measures the following
compression variables (depth, duty cycle, rate, hand posi-
tion). Hand positioning is recorded as correct if pressure is
of incorrect compressions that are low (below the simulated
xiphoid process) are also recorded.
At the conclusion of each test, participants were asked to
of efficacy and fatigue . (Statements—(i) the CPR I per-
with using the CPREzyTMdevice.
2.4. Statistical analysis
From a previous study  we calculated that we would
compression depth at a significance level of 0.05 and 90%
inc, IL, USA). Data were tested for normality using Sharpiro
Wilks test. For normally distributed data repeated measure-
ments over time were analysed by two-factor repeated mea-
sure ANOVA, the two factors being compression variable
G.D. Perkins et al. / Resuscitation 64 (2005) 103–108
and time. Huynh–Feldt epsilon was used when spherecity
conditions were not met. Paired t-tests were used to com-
pare overall data for the two groups. These results are pre-
sented as mean (standard deviation) and 95% confidence in-
tervals of the difference between groups (95% CI d.f.). Non-
and Wilcoxon signed rank test and were presented as me-
dian (interquartile range). Pearsons correlation coefficient
was used to assess linear associations. Nominal data were
compared using McNemar’s test. A P-value of <0.05 was
considered statistically significant.
3.1. Pilot study
There was a significant linear relationship between com-
pression depth and force (indicated by the number of lights
illuminated on the CPREzyTMdevice), r = 0.998, P = 0.002.
Optimal chest compression depth (40–50mm) was achieved
when four lights were illuminated (Fig. 2).
3.2. Principle study
Due to logistic problems, three participants from the first
phase of the study had to be replaced by volunteers matched
for age, sex, height and weight during the second phase.
The participants age, sex, height and weight are displayed
in Table 1.
Fig. 2. Pilot study results—chest compression depth corresponding to the
number of lights illuminated on the CPREzyTMdevice. Optimal chest com-
illuminated. Data presented are the mean (standard error) from six subjects.
Demographics of study participants
Data are mean (standard deviation).
There was a significant improvement in chest compres-
sion depth with CPREzyTMcompared to standard CPR
42.9mm (4.4) versus 34.2mm (7.6), P = 0.001; 95% CI d.f.
4.4–12.9mm. This effect was maintained throughout the 3-
min test (Fig. 3A). This translated to a reduction in the per-
centage of shallow compressions (<38mm) with CPREzyTM
16 (23)% compared to standard CPR 59 (44)%, P = 0.003.
There was a small increase in the percentage of compression
regarded to be excessive (>51mm): CPREzyTM6.5 (19)%
versus standard CPR 0 (0.1)%, P = 0.012.
There was no difference in chest compression rate (me-
dian[IQR]): CPREzyTM102[101–104]min−1versus stan-
dard CPR 98[90.5–107], P = 0.407. This did not change over
time P = 0.861 (Fig. 3B). There was no difference in aver-
age duty cycle CPREzyTM47.7(3.4)% versus standard CPR
48.1(5.5)%, P = 0.845. This did not change over time P =
0.789 (Fig. 3C).
Equal numbers of participants (40% in each group) per-
formed one or more incorrectly placed chest compression.
total number of incorrectly placed compressions was higher
for the CPREzyTMgroup (26% versus 3.9% standard CPR,
P < 0.001.). This was due to a higher number of low com-
pressions (26% of total compressions for CPREzyTMversus
1% for standard CPR, P < 0.001).
3.3. Participant feedback
There was no difference in participants perceptions of
achieving adequate depth of compression between tech-
niques: CPREzyTM49 (S.D. 30) versus standard CPR 60
(S.D. 22) P = 0.149. Participants found using the CPREzyTM
device caused greater fatigue than standard CPR 75 (S.D.
22) versus 60 (S.D. 25), 95% CI d.f. 4.5–26.4, P = 0.008.
Ninety-five percent of participants reported discomfort in
the heels of their hands and wrists in association with the
CPREzyTMdevice. One participant sustained a soft tis-
sue injury when the skin covering the fifth metacarpal be-
came trapped between moving compression pad (Fig. 1,
part A) and fixed lights section (Fig. 1, part B) of the de-
vice (Fig. 4 shows injury). The bruising settled within 2
weeks and there were no long lasting sequelae from this
device was associated with a significant (average 8.7mm)
improvement in chest compression depth during simulated
resuscitation of a patient on a hospital bed. This was
associated with a substantial reduction in the number of
“shallow” (<38mm) compressions and only a slight increase
in the number of “excessive” compressions (>51mm). This
is an important finding since it is to our knowledge the
first intervention shown to improve the efficacy of external
G.D. Perkins et al. / Resuscitation 64 (2005) 103–108
Fig. 3. Chest compression depth over time. Black, dashed line represents
CPREzyTM, grey solid line represents standard CPR. Data are presented
as mean (triangles/circles) and standard error (bars). Data were analysed by
difference between CPREzyTMand standard CPR (F = 16.2, P = 0.001).
There was a significant decline in chest compression depth over time (F =
10.97, P <0.0001). (B) Rate—there was no significant difference between
CPREzyTMand standard CPR (F = 0.72, P = 0.410). There was a significant
decline in chest rate over time (F = 9.7, P = 0.0001). (C) Duty cycle: there
was no significant difference between CPREzyTMand standard CPR (F =
0.904, P = 0.904). There was a significant decline in chest rate over time (F
= 5.3, P = 0.0001).
Fig. 4. Photograph showing soft tissue injury over outer aspect of fifth
metacarpal after the hand became trapped between the mobile CPREzyTM
compression plate and fixed light indicator box.
chest compressions during the resuscitation of a simulated
in-hospital cardiac arrest.
External chest compressions yield at best 30% of normal
ity of CPR have been shown to be important determinants of
a linear relationship between compression depth and cardiac
output, mean arterial blood pressure and coronary artery per-
fusion [11,12]. A 10mm improvement in compression depth
was associated with a 50% relative increase in cardiac out-
put and 30% relative increase in mean arterial blood pressure
Compared to CPR performed on a manikin placed on the
floor, we and others have demonstrated previously that com-
pression depth deteriorates markedly when the manikin is
placed on a hospital bed [4,3,15]. One potential explanation
for this finding is that when CPR is performed on a hospital
bed, the compression force not only causes compression of
the sternum, but also of the underlying mattress . It is
for this reason that current international guidelines recom-
mend that a back-board is placed underneath the victim .
However in our previous studies, neither emergency defla-
tion of an air filled mattress (such that it then becomes a firm,
beneath the manikin  led to any improvement in compres-
rescuer body position. In a bench model we found that there
was an inverse relationship between maximal compression
victim above the floor had an adverse effect on compression
performance . However when we subsequently went on to
study the effect of kneeling on the bed next to the victim or
lowering the height of the bed neither intervention had any
significant impact on compression efficacy .
One clear theme emerging from this study and our previ-
ous studies was the failure of the CPR provider to recognise
G.D. Perkins et al. / Resuscitation 64 (2005) 103–108
study, we found no difference in participant’s perceptions of
the efficacy of their chest compressions between the stan-
dard and CPREzyTMgroup despite clearly different levels of
efficacy. The failure of the CPR provider to recognise sub-
lem by providing continuous visual and audible feedback on
tors that have studied the effect of feedback during CPR and
CPR is the Laerdal Voice Advisory Mannikin system. This
system analyses CPR performance using a micro-computer
and gives verbal instructions to the CPR provider on how to
optimise their technique during the resuscitation attempt e.g.
“Press a little deeper” etc. Early evaluations of this device
have yielded promising results with reported improvements
improvements in ventilation volumes and chest compression
depth during simulated CPR performed by nurses .
The present study supports and extends the findings by
Boyle et al. . Their study evaluated the use of CPREzyTM
during simulated CPR on a resuscitation manikin placed on
the floor. The principle findings were a significant improve-
ment in compression rate, the proportion of “effective com-
pressions” (a composite score of correct position, force and
release between compressions) and compression position-
ing. In contrast to their study however, we found that the
CPREzyTMdevice was associated with an increased total
than compressions) that performed low compressions. When
considered in these terms, 40% of our subjects performed
one or more low compressions, which is comparable to the
finding in their control group. What we failed to demonstrate
therefore was any improvement with the CPREzyTMdevice.
sensitive to small errors in compression placement and as we
evance, or potential for harm, from this observation is likely
to be small. We suggest that to overcome this potential prob-
lem, participants are briefed carefully to continually evaluate
the position of the device during CPR to avoid misplaced
It was a concern that 95% of our participants reported dis-
comfort in their hands and wrist whilst using the CPREzyTM
device. This may reflect the additional effort that is required
to correctly perform CPR. It may also be due to the hard-
ness or narrowness of the compression plate. Of particular
concern was the soft tissue injury sustained by one of our
participants whilst using the device. We suggest that users
device and take care to avoid placing their hands to close
to the moveable compression plate and fixed housing. The
manufacturers should also re-visit the design of the devise
and take steps to minimise the risk of this recurring.
Although the results of this early evaluation of the
CPREzyTMdevice are promising, there are several limita-
eration. Most importantly, this was a laboratory-based study
using a resuscitation manikin rather than a clinical study of
require verification in animal cardiac arrest models before
undertaking clinical studies in humans in cardiac arrest. This
is particularly topical when one considers the early promise
spite evidence of improved CPR efficacy on manikins and
improvements in haemodynamics during CPR, large clinical
trials failed to demonstrate any consistent improvement in
ical students recently trained in BLS as the CPR providers
rather than qualified healthcare providers. Although at the
time of undertaking the study few had first hand experience
in performing CPR for real, the quality of CPR appears com-
vious study .
In conclusion, the CPREzyTMdevice was associated with
simulated resuscitation of a victim in a hospital bed. To our
knowledge, this is the first intervention that has been shown
to improve the performance of chest compressions when un-
dertaken with the victim on a hospital bed. Further animal
and clinical studies are required to validate this finding in
vivo and to see if it translates to an improvement in outcome
in human victims of cardiac arrest.
6. Conflict of interest statement
GDP, CA, MA and HR received a travel grant from the
UK distributors of CPREzyTM, Health Affairs Ltd.
This study was supported by a small project grant from
design, data collection and interpretation or the decision to
submit for publication. The sponsor kindly provided Fig. 1a
and b. We gratefully acknowledge the support of Dr Ken
Moralle (Laerdal Medical) for supplying the VAM system
108 Download full-text
G.D. Perkins et al. / Resuscitation 64 (2005) 103–108
 Wik L, Steen PA, Bircher NG. Quality of bystander cardiopulmonary
resuscitation influences outcome after prehospital cardiac arrest. Re-
 Jones AY. Can cardiopulmonary resuscitation injure the back? Re-
 Tweed M, Tweed C, Perkins GD. The effect of differing support
surfaces on the efficacy of chest compressions using a resuscitation
manikin model. Resuscitation 2001;51:179–83.
 Perkins GD, Benny R, Giles S, Gao F, Tweed MJ. Do different mat-
tresses affect the quality of cardiopulmonary resuscitation? Intensive
Care Med 2003;29:2330–5.
 Smith CM, Stephenson BTF, Gao F, Perkins GD. Using a backboard
during CPR on a bed does not improve CPR performance (Abstract).
 Boyle AJ, Wilson AM, Connelly K, McGuigan L, Wilson J, Whit-
bourn R. Improvement in timing and effectiveness of external cardiac
compressions with a new non-invasive device: the CPR-Ezy. Resus-
 Perkins GD, Hulme J, Shore HR, Bion JF. Basic life support training
for health care students. Resuscitation 1999;41:19–23.
 Perkins GD, Hulme J, Bion JF. Peer-led resuscitation training for
healthcare students: a randomised controlled study. Intensive Care
 Perkins GD, Stephenson BT, Smith CM, Gao F. A comparison be-
tween over-the-head and standard cardiopulmonary resuscitation. Re-
 Robertson C, Holmberg S. Compression techniques and blood flow
during cardiopulmonary resuscitation. A statement for the Advanced
Life Support Working Party of the European Resuscitation Council.
 Wik L, Naess PA, Ilebekk A, Nicolaysen G, Steen PA. Effects of
various degrees of compression and active decompression on haemo-
dynamics, end-tidal CO2, and ventilation during cardiopulmonary
resuscitation of pigs. Resuscitation 1996;31:45–57.
 Babbs CF, Voorhees WD, Fitzgerald KR, Holmes HR, Geddes LA.
Relationship of blood pressure and flow during CPR to chest com-
pression amplitude: evidence for an effective compression threshold.
Ann Emerg Med 1983;12:527–32.
 Bellamy RF, DeGuzman LR, Pedersen DC. Coronary blood
flow during cardiopulmonary resuscitation in swine. Circulation
 Ornato JP, Levine RL, Young DS, Racht EM, Garnett AR, Gonza-
lez ER. The effect of applied chest compression force on systemic
arterial pressure and end-tidal carbon dioxide concentration during
CPR in human beings. Ann Emerg Med 1989;18:732–7.
 Larsen PD, Perrin K, Galletly DC. Patterns of external chest com-
pression. Resuscitation 2002;53:281–7.
 Boe JM, Babbs CF. Mechanics of cardiopulmonary resuscitation per-
formed with the patient on a soft bed vs a hard surface. Acad Emerg
 European Resuscitation Council. Part 3: adult basic life sup-
port. European Resuscitation Council. Resuscitation 2000;46:29–
 Allen M, Augre C, Rogers H, Perkins GD. Does bed height ef-
fect the efficacy of chest compressions? (Abstract) Resuscitation
 Hightower D, Thomas SH, Stone CK, Dunn K, March JA. Decay
in quality of closed-chest compressions over time. Ann Emerg Med
 Elding C, Baskett P, Hughes A. The study of the effective-
ness of chest compressions using the CPR-plus. Resuscitation
 Wik L, Thowsen J, Steen PA. An automated voice advisory manikin
system for training in basic life support without an instructor.
A novel approach to CPR training. Resuscitation 2001;50:167–
 Wik L, Myklebust H, Auestad BH, Steen PA. Retention of basic life
support skills 6 months after training with an automated voice ad-
visory manikin system without instructor involvement. Resuscitation
 Handley AJ, Handley SA. Improving CPR performance using an
audible feedback system suitable for incorporation into an automated
external defibrillator. Resuscitation 2003;57:57–62.
 Ornato JP. Efficacy vs. effectiveness: the case of active compression-
decompression (ACD) CPR. Resuscitation 1997;34:3–5.
 Lafuente-Lafuente C,Melero-Bascones
Cochrane Database Syst Rev 2001; CD002751.