Safety, feasibility, and hemodynamic and blood flow effects of

active compression– decompression of thorax and abdomen in

patients with cardiac arrest*
Christof Havel, MD; Andrea Berzlanovich, MD; Fritz Sterz, MD; Hans Domanovits, MD; Harald Herkner, MD;
Andrea Zeiner, MD; Wilhelm Behringer, MD; Anton N. Laggner, MD

Objective: During closed chest compression for cardiac arrest,
any increase in coronary perfusion pressure accounts for a pro-
portional increase in myocardial blood flow and therefore the
resuscitability of the patient. The objectives of this study were to
evaluate the safety, feasibility, and hemodynamic effects of
phased chest and abdominal compression– decompression and to
compare it with mechanical chest compression during cardiopul-
monary resuscitation.
Design: In this prospective, single-center, phase II study, we
compared patients treated with the Datascope Lifestick Resusci-
tator with patients who had been treated with mechanical pre-
cordial compression.
Setting: Emergency department of a tertiary care university
hospital.
Patients: We included 31 patients with cardiac arrest who had
received cardiopulmonary resuscitation in the emergency depart-
ment.
Interventions: The Lifestick device was used in 20 patients. In
11 patients, mechanical chest compression with the Thumper
device was used as a control intervention.
Measurements and Main Results: We evaluated the safety,
feasibility, and hemodynamic effects of both interventions and
observed, with the help of echocardiography, the mechanisms
through which blood flow was generated. We found no significant
difference between the use of the Lifestick device and standard
chest compression with the Thumper device in resuscitations.
Most operators regarded the Lifestick as a feasible alternative to
the Thumper. We could observe a mean increase in coronary
perfusion pressure of 9.33 mm Hg (interquartile range, 1.96 –
14.36; p ⴝ .08) and an increase of end-tidal CO2 of 10 mm Hg
(interquartile range, 5–16; p ⴝ .003) (1333Pa [interquartile range,
665–2133]) during resuscitation with the Lifestick compared with
using the Thumper.
Conclusion: In this preliminary study, resuscitation with
the Lifestick was found to be safe and feasible. The design of
the study and small number of patients included in it limit the
conclusions about the hemodynamic effects of the Lifestick.
(Crit Care Med 2008; 36:1832–1837)
KEY WORDS: cardiopulmonary resuscitation; death; sudden;
echocardiography

C losed chest compression has
remained the standard for hu-
man resuscitation (1–3) since
the 1960 landmark study by
Kouwenhoven et al (4, 5). Crile and Dol-
ley (6) recognized the importance of ad-
*See also p. 1974.
From the Departments of Emergency Medicine
(CH, FS, HD, HH, AZ, WB, ANL) and Forensic Medicine
(AB), Medical University of Vienna, Vienna General
Hospital, Vienna, Austria.
The authors have not disclosed any potential con-
flicts of interest.
Supported, in part, by grants from the Austrian
Science Foundation (P11405-MED) and by an unre-
stricted grant from Datascope Cardiac Assist Division,
Fairfield, NJ. Robert B. Schock (Datascope Corpora-
tion, Cardiac Assist Division, Fairfield, NJ) provided
technical support and the Lifestick.
For information regarding this article, E-mail:
fritz.sterz@meduniwien.ac.at
Copyright © 2008 by the Society of Critical Care
Medicine and Lippincott Williams & Wilkins
DOI: 10.1097/CCM.0b013e3181760be0
equate myocardial perfusion for cardiac
resuscitation. Myocardial blood flow is
highly correlated with the coronary per-
fusion pressure (7–10) (i.e., the pressure
gradient between the aorta and the right
atrium during compression diastole). An
increase in coronary perfusion pressure
leads to a corresponding increase in myo-
cardial blood flow and therefore enhances
the resuscitability of a patient (9, 11, 12).
The report of a successful cardiopul-
monary resuscitation using a toilet
plunger was the trigger for the develop-
ment of a suction cup, which produces
not only active chest compressions but
also decompressions (13). In subsequent
animal experiments (14, 15), and later in
human cardiac arrest studies (14, 16, 17),
the beneficial hemodynamic effects of ac-
tive compression and decompression
were clearly demonstrated (18).
Active out-of-phase compression and
decompression both of the thorax and
abdomen was described to yield maximal
coronary and carotid perfusion (19 –21).
This concept was refined by a modeling
study of Lin et al (22). Based on this
study, the Lifestick resuscitator was de-
veloped (23).
The objectives of this study were to
evaluate the safety, feasibility, and he-
modynamic effects (with additional
echocardiographic analyses) of sequen-
tial active compression– decompression
of the thorax and abdomen with the
Lifestick resuscitator (Datascope Cor-
poration, Fairfield, NJ) during human
cardiopulmonary resuscitation in a
clinical setting.
MATERIALS AND METHODS
In this prospective (not randomized, not
blinded), single-center, phase II study in
which we compared patients receiving 5 mins
of mechanical chest compression (Thumper

1832 Crit Care Med 2008 Vol. 36, No. 6

1000, Michigan Instruments, Grand Rapids,
MI) (8) with patients who received 10 mins of
“treatment” with the Lifestick. To assess the
safety of the Lifestick, we also included in the
study patients who were only treated with me-
chanical chest compression (Thumper) under
identical circumstances. Ethics committee ap-
proval was obtained before commencement
of the study; informed consent was waived
pursuant to the Declaration of Helsinki,
Section II.5.
Subjects. Patients for this study were se-
lected from the population served by our de-
partment of emergency medicine at a tertiary
care university hospital (24). Patients ⬎18 yrs
of age presenting to the emergency depart-
ment in nontraumatic, normothermic, wit-
nessed cardiopulmonary arrest were eligible
for inclusion into this study. The specific in-
clusion criteria were: continued cardiac arrest
and absence of pulse and blood pressure de-
spite ⱖ10 mins of standard advanced cardiac
life support— defined as coronary perfusion
pressure of ⬍15 mm Hg or end-tidal CO2
pressure of ⬍15 mm Hg (1200 Pa), or both.
Patients with known terminal conditions,
chronic mental or physical disabilities, hemo-
dialysis treatment, or midline sternotomy
scars were not included.
All patients were managed in accordance
with the guidelines of the American Heart
Association and conventional medical man-
agement of the department of emergency
medicine, regardless of which type of external
cardiac compression was applied (2, 24).
Volume-controlled ventilation (Servo Ven-
tilator 300, Siemens-Elema AB, Life Support
Systems Division, Solina, Sweden) was set at
an FIO2 of 1.0, tidal volume of 10 mL/kg body
weight, respiratory rate of 15 breaths/min, and
ventilation pressure limit of 50 mm Hg.
Cardiopulmonary Resuscitation. While
standard advanced cardiac life support was be-
ing performed, a special Lifestick code team
prepared for the trial of standard cardiopul-
monary resuscitation and Lifestick resuscita-
tion. Thoracoabdominal compression– decom-
pression Lifestick resuscitation was performed
by means of a hand-held and manually oper-
ated device (Fig. 1) by a specially trained Life-
stick code team. The device consists of a single
rigid frame attached to two disposable adhe-
Figure 1. Thoracoabdominal compression– decom-
pression Lifestick resuscitator.
Crit Care Med 2008 Vol. 36, No. 6
sive-backed pads (one pad each for adhesion to
the sternum and abdomen). The device incor-
porates force monitoring and a metronome to
facilitate the delivery of optimal forces and
timing of load application. Mechanical chest
compression in the control group was per-
formed by an automatic chest compression
device (Thumper) that delivered 4 –5 cm of
deflection (40 – 60 kg of force) at a rate of 80
compressions/min and a compression/ventila-
tion ratio of 5:1. The compression duty cycle
was 50% and not interrupted for ventilation.
Safety Criteria. Safety was evaluated by
assessment and analysis of adverse events in
patients treated with the Lifestick by means of
autopsies following predefined criteria.
Feasibility. After the completion of resus-
citation with the Lifestick, the operator of the
Lifestick device completed a questionnaire,
which included the ability of the device to
deliver appropriate forces and pad adhesion to
the skin.
Hemodynamic Measurements. Two pigtail
catheters (Cordis Corporation, Miami, FL)
were inserted percutaneously, one via the fem-
oral artery and one via the femoral vein, to
obtain aortic and right atrial pressures. A cor-
onary perfusion pressure analysis was per-
formed by using the area under the curve
(aortic minus venous pressure) during dias-
tole. Diastole was defined as the time during
which no chest compression was carried out.
Using interval frames of 4 secs for on-line
calculations with various filter techniques and
signal processing features (windowing of sig-
nal, differential, and fast Fourier transform
algorithm) on the coronary perfusion pressure
result, it was possible to extract exactly defined
parts of the signal and evaluate the blood
pressure within the coronary arteries during
external chest compression.
Transesophageal Echocardiography. In
patients receiving Lifestick resuscitation,
transesophageal echocardiography was per-
formed with a HP Sonos 2500 (Hewlett-
Packard, Palo Alto, CA) using a biplane trans-
esophageal echocardiography probe according
to the guidelines of the American Society of
Echocardiography (25). All examinations were
performed by two investigators trained and
experienced in this ultrasound technique. A
force curve of the Lifestick resuscitator was
displayed on the ultrasound screen in addition
to echocardiographic tracings to allow for
time phasing of the echo with the resuscita-
tion cycles. The variables selected for this
study were: 1) ventricular deformation during
the compression phase of cardiopulmonary re-
suscitation, 2) mitral valve position during the
compression phase of cardiopulmonary resus-
citation, 3) transmitral flow during the com-
pression phase of cardiopulmonary resuscita-
tion, and 4) aortic valve position.
Statistics. The data are presented as median
and the interquartile range or number and per-
centage, where appropriate. Fisher’s exact test
and the Mann-Whitney U test were used to test
for differences between the Lifestick and
Thumper resuscitation groups for the safety
analysis. For the analysis on hemodynamic ef-
fects, a one-sample Wilcoxon’s test was used to
allow for the paired nature of data. SPSS (SPSS,
Chicago, IL) for Windows was used for data anal-
ysis. A two-sided p value of ⬍.05 was considered
as statistically significant.
RESULTS
The study was 10 months in duration.
During this period, 214 cardiac arrest pa-
tients were admitted to the emergency
department, out of which 31 patients
could be included in the study. Treat-
ment with thoracoabdominal compres-
sion– decompression Lifestick resuscita-
tion was applied to 20 patients. To assess
the safety of the Lifestick, 11 patients
were prospectively selected for mechani-
cal chest compression with the Thumper.
The demographic data for both groups
are presented in Table 1. There were no
significant differences among the pa-
tients with regard to most variables; the
only exception was the cause of their car-
diac arrest.
In patients receiving Lifestick resusci-
tation, the median time that had elapsed
from their collapse until the application
of the Lifestick resuscitator was 52 mins
(34 –78 mins). None of these patients sur-
vived. The compression force applied dur-
ing mechanical chest compression with
the Thumper was 55 kg (50 – 60 kg), fol-
lowed by 54 kg (49 – 61 kg) in the Life-
stick resuscitation phase.
Safety Criteria. Adverse events as de-
termined by autopsy are presented in Ta-
ble 2. None of the patterns reached sta-
tistical significance between the groups.
Feasibility. The operators’ ratings of
the ability to deliver appropriate forces
during thoracoabdominal compression–
decompression Lifestick resuscitation
and of the adhesion of the pads to the
skin are presented in Figures 2 and 3.
Overall, the feasibility of using the Life-
stick was rated as easy and very good by
the majority of operators.
Hemodynamic Measurements. In five
patients, the arterial and venous cathe-
ters could be correctly placed in the aor-
tic arch and right atrium, respectively.
Mean increase in coronary perfusion
pressure was 9.33 mm Hg (1.96 –14.36,
p ⫽ .08). In all but one patient, the cor-
onary perfusion pressure increased dur-
ing Lifestick resuscitation (Fig. 4).
The end-tidal CO2 pressure was deter-
mined in 15 patients. The mean values
are presented in Figure 5. We could ob-
serve an increase in end-tidal CO2 pres-
1833
Table 1. Demographic data
Age (years), median (IQR)
Female gender, n (%)
Height (m), median (IQR)
Weight (kg), median (IQR)
Pre-arrest status (OPCa), median (IQR)
Medical history
Smoker, n (%)
Diabetes, n (%)
Heart disease, n (%)
Skin dampness, n (%)
Moderate or severe hirsute, n (%)
Initial cardiac rhythm
Asystole, n (%)
Pulsless electric activity, n (%)
Ventricular fibrillation, n (%)
Total defibrillation energy J, median
(IQR)
Prehospital
In-hospital prior Lifestick/prior
inclusion
In-hospital during Lifestick/after
inclusion
Total amount of epinephrine mg,
median (IQR)
Prehospital
In-hospital prior Lifestick/prior
inclusion
In-hospital during Lifestick/after
inclusion
Cause of cardiac arrestb
Myocardial infarction, n (%)
Pulmonary embolism, n (%)
Miscellaneous, n (%)
a
Overall performance category; bthe cause of cardiac arrest was in each case confirmed by autopsy.
IQR, interquartile range.
Table 2. Adverse events as determined by autopsy
Bilateral rib fracture – n/n (%)
Number of fractured ribs – median (IQR)
Sternum fracture – n/n (%)
Liver rupture – n/n (%)
Gastric dilation – n/n (%)
Skin abrasion – n/n (%)
Skin contusion – n/n (%)
IQR, interquartile range.
sure of 10 mm Hg (5–16, p ⫽ .003) (1333
Pa [665–2133]) during Lifestick resusci-
tation.
Transesophageal Echocardiographic
Observations. Transesophageal echocar-
diography was performed on five patients
during mechanical chest compression
with the Thumper and on 11 patients
during resuscitation with the Lifestick.
The proposed variables could be analyzed
only in a limited number of recordings
due to difficulties during the ongoing re-
suscitation. During standard chest com-
pression with the Thumper and during
1834
resuscitation, the mitral valve was open
during the compression phase in three
Lifestick, n ⫽ 20 Thumper, n ⫽ 11
patients and closed in eight patients. In one
61 (49–71) 65 (50–71)
patient, the valve was open during chest
6 (30) 6 (55) compression with the Thumper and closed
1.70 (1.63–1.79) 1.72 (1.62–1.77) during Lifestick resuscitation. Quantitative
80 (74–90) 68 (62–90) Doppler flow over the mitral valve (Table 3)
1.0 (1.0–1.3) 1.0 (1.0–1.0) could be measured only in three patients
7 (35) 2 (18) during chest compression with the
3 (15) 1 (9) Thumper and in 11 patients during Life-
9 (45) 5 (45) stick resuscitation. During abdominal com-
11 (55) 8 (73) pression, atrioventricular reflux in the be-
4 (20) 2 (18)
ginning of the compression phase, then
7 (35) 2 (18) antegrade flow from the left atrium to the
5 (25) 4 (36) left ventricle in the mid-compression
8 (40) 5 (45) phase, and finally, another bout of ante-
grade flow from the left atrium to the left
2660 (1,000–4,505) 2360 (1,700–3,680)
ventricle in the early decompression phase
1360 (960–1,960) 2020 (1,440–4,650) was observed. One patient had exclusively
antegrade flow from the left atrium to the
960 (720–4,560) 2520 (1,460–5,400) left ventricle during the thoracic compres-
sion phase of chest compression with the
Thumper. The velocity–time integral of an-
8 (4–12) 9 (7–11)
tegrade atrioventricular flow was .174
2 (1–2) 2 (2–2) m/sec (.147–.181 m/sec). The aortic valve
opened, with forward aortic flow during
2 (1–2) 1 (1–2) chest compression, in all patients studied.
14 (70) 2 (20) DISCUSSION
2 (10) 1 (10)
4 (20) 7 (70) Four-phase thoracoabdominal compres-
sion– decompression resuscitation with the
Lifestick, a technique combining the effects
of interposed abdominal compression (26)
and active compression– decompression
(14) cardiopulmonary resuscitation is fea-
Lifestick Controls to Assess Safety sible and easily accepted by users after a
n ⫽ 20 n ⫽ 11 p
brief period of training. Injuries due to car-
16/20 (80) 9/11 (75) .36
diopulmonary resuscitation, such as rib
8 (5–12) 6 (5–10) .56 fractures or internal bruising, as deter-
16/18 (89) 7/10 (64) .19 mined by autopsy, were not significantly
1/20 (5) 0/0 1.00 different for Lifestick resuscitation and me-
5/20 (25) 3/10 (30) .32 chanical chest compression with the
13/18 (72) 10/11 (91) .20
3/18 (18) 1/11 (9) .39 Thumper only. Although all patients had
prolonged cardiopulmonary resuscitation,
Lifestick resuscitation seemed to improve
hemodynamic effects compared with me-
chanical chest compression with the
thoracoabdominal compression– decom- Thumper only. The few transesophageal
pression Lifestick resuscitation, the right echocardiographic observations point to
atrium and ventricle were almost com- the conclusion that in most patients, the
pletely compressed and thus evacuated effectiveness of blood flow mechanisms and
during the compression phase in all pa- the thoracic and cardiac pump mechanism
tients. Also, the left ventricle appeared (27, 28) are mutually dependent on each
empty, although to a lesser extent than other.
the right ventricle. We made no attempts Standard manual chest compression
to measure the magnitude of the reduc- would be the appropriate technique with
tion of the left ventricle. During chest which the Lifestick should be compared. To
compression with the Thumper, the mi- achieve accurate, consistent, and compara-
tral valve was open during the compres- ble pressures during chest compression, we
sion phase in two patients and closed in instead used mechanical chest compres-
three patients (Table 3). During Lifestick sion with the Thumper device as control
Crit Care Med 2008 Vol. 36, No. 6

Figure 2. Feasibility of the thoracoabdominal compression– decompression Lifestick resuscitation:
operators’ judgment of the ability to deliver appropriate forces during the use of the Lifestick
resuscitator assessed by a questionnaire immediately after the procedure.
Figure 3. Feasibility of the Lifestick resuscitation: judgment of the pad adhesion to the skin during the
use of the Lifestick resuscitator.
25
mmHg
20
15
10
5 20
0 10
mechanical chest lifestick
compression
Figure 4. Mean coronary perfusion pressure
available from five patients, comparing mechan-
ical chest compression with Thumper (cardiopul-
monary resuscitation) followed by thoracoab-
dominal compression– decompression phase with
the Lifestick. The bold line represents the mean
of these five patients.
intervention. This was considered better
then manual chest compression, even if the
piston of the Thumper might somewhat
hinder the relaxation phase (29).
Safety Criteria. Because of the force
necessary to circulate blood, it is com-
Crit Care Med 2008 Vol. 36, No. 6
sion Lifestick resuscitation and standard
chest compression Thumper resuscita-
tion alone (Table 2). Depending on the
definition of injuries and the quality of
the autopsy, the rate of reported injuries
varied between 21% and 65% (30). In a
series of 705 autopsies after conventional
resuscitation, thoracic complications of
various degrees were observed in 43% of
the cases. Rib fractures were found in
32% and sternal fractures in 21% of the
autopsies. In 18% of the cases, mediasti-
nal hemorrhages were observed. More-
over, abdominal injuries were described
in 31% and pulmonary complications in
13%. In nearly 0.5% of the cases, the
injuries were categorized as life threaten-
ing (31). In this study, we found a high
rate of rib and sternal fractures, which is
in line with the findings reported else-
where (32, 33). However, there was no
indication of an increase in abdominal
lesions with the new technique.
Feasibility. The thoracoabdominal
compression– decompression Lifestick
resuscitation technique was easy to learn
and everyone on our team was able to
operate the device after a short period of
training. The device was well accepted
and the operators delivered appropriate
forces during thoracoabdominal com-
pression– decompression Lifestick resus-
citation. The adhesion of the pads to the
skin was satisfactory. Lifestick users did
not complain about exhaustion or stren-
50 uous handling during the 10 mins of op-
mmHg eration. There were also no reports of
45
discomfort in the back or other com-
40
plaints by the operators. There were a few
35 reports of instability of the device because
30 of unsatisfactory fixation of the pads to
the skin (Fig. 3). No problems with re-
25
spect to defibrillation were reported. The
metronome giving the thoracic–abdomi-
15 nal phase shift was found to be useful. No
handling errors were observed.
Hemodynamic Effects and Mecha-
5
nism of Blood Flow. There are two theo-
0
ries about the mechanisms that generate
mechanical chest Lifestick
blood flow. The cardiac pump theory pos-
compression
tulates direct compression of the heart be-
Figure 5. Mean end-tidal CO2 pressure in 15 pa-
tients, comparing mechanical chest compression tween the sternum and spine. The thoracic
with Thumper (cardiopulmonary resuscitation) fol- pump theory postulates that blood flow is
lowed by thoracoabdominal compression– decom- caused by intrathoracic pressures exceed-
pression phase with the Lifestick. The bold line ing extrathoracic vascular pressure (27, 28,
represents the mean of these 15 patients. 34). Compression and decompression of ei-
ther the chest or the abdomen can help to
move blood during cardiac arrest. Negative
mon for victims to receive injuries. Au- thoracic and positive abdominal pressures
topsies of the patients have shown that prime the chest pump. Positive thoracic
these have been comparable between tho- and negative abdominal pressures prime
racoabdominal compression– decompres- the abdominal pump (20).
1835
Table 3. Transesophageal echocardiographic observations in patients receiving first mechanical chest compression with the Thumper followed by thoraco
abdominal compression-decompression Lifestick resuscitation

Thumper Lifestick

Patient # Age Rhythm Lifestick MVa Position AVb Regurgitation Antero-grade VTIc MVa Position AVb Regurgitation Antero-grade VTIc

1 79 Asystd — — — Closed Yes —

4 67 PEAe — — — Open No —
6 80 Asystd Open — — Closed Yes .188
7 64 VFf — — — Closed Yes .174
8 48 VFf — — — Open Yes .224
9 41 Asystd — — — Closed Yes —
10 77 Asystd — — — Closed Yes .128
11 70 VFf Closed Yes .158 Closed Yes —
14 57 Asystd Closed — — Closed Yes .174
16 57 PEAe Open No — Open Yes .148
17 70 PEAe Closed Yes — Closed Yes .146
a
mitral valve; batrioventricular; cmeasurements were performed in the beginning of the compression phase by means of color Doppler flow and
anterograde mitral velocity time integral in m/s (VTI); dasystole; epulsless electrical activity; fventricular fibrillation.

We could demonstrate a marked in-
crease in coronary perfusion pressure,
which indicates that Lifestick resuscitation
does increase blood flow. Abdominal com-
pression, however, may increase arterial
and venous pressures via diaphragmatic
motion but not actually increase perfusion
in the coronaries. Thus, the calculated cor-
onary perfusion pressure may not reflect
actual coronary perfusion if the aortic valve
is open. Unfortunately, this is currently the
only way to arrive at an estimate of coro-
nary perfusion pressure in real life.
Animal and human data (34, 35) dem-
onstrated that a reduction in the size of the
left ventricle, the opening of the mitral
valve during cardiac release, and atrioven-
tricular regurgitation support the cardiac
pump theory. Porter et al. (36) performed
transesophageal echocardiography on 17
human cardiac arrest victims during chest
compressions and found that the mitral
valve closed during the compression phase
in 12 patients, who also had peak transmi-
tral flow in the release phase of the chest
compression cycles. Redberg et al. (34)
found closed mitral valves in all 20 patients
during cardiopulmonary resuscitation in-
cluded in their study. In the current study,
two out of five patients had open mitral
valves during the compression phase under
mechanical chest compressions and three
out of 11 patients under Lifestick resusci-
tation. The difference between our data and
that of Redberg et al. (34) might be due to
the brief time interval between the start of
cardiopulmonary resuscitation and the
start of echocardiography (ⱕ7 mins) in
their study, which was substantially longer
in our study (collapse to Lifestick resusci-
tation: mean 52 mins [34 –78 mins]). One
of our patients (patient 6) changed from
open mitral valve during compression un-
der standard cardiopulmonary resuscita-
tion to closed mitral valve under Lifestick
resuscitation. Because the arteriovenous
pressure difference remained unchanged,
we do not believe that this represents a
substantial change in the blood flow mech-
anism. It could have been caused by in-
creased compression force (37), changes in
intrathoracic pressure (10), or application
of the compression forces to regions of the
chest wall that were overlying the heart
(38). However, we did not observe any dif-
ferences in pressure forces in our study.
The most likely explanation is that active
thoracic decompression combined with si-
multaneous abdominal compression in-
creased the filling of the cardiac chambers
and thus produced higher stroke volumes.
However this remains a conjecture because
we lack quantitative Doppler measurement
data for the aortic outflow tract.
A recent study of blood flow during
cardiopulmonary resuscitation (28)
claims that both mechanisms might be
involved in the generation of blood flow.
The transmitral flow pattern observed in
all our patients during Lifestick resusci-
tation, independent of the position of the
mitral valve, can be interpreted to be
caused by both blood flow mechanisms
operating jointly. Initially, the cardiac
pump starts blood flow, and as soon as
the intrathoracic pressure exceeds a cer-
tain level, then the thoracic pump pro-
vides additional pressure. We can only
surmise that this was the predominant
mechanism for blood flow in our patients
with prolonged cardiac arrest.
Laboratory studies (39 – 41) evaluating
phased chest and abdominal compression–
decompression have arrived at similar find-
ings as in our study with regard to the
hemodynamic effects and mechanisms of
blood flow, and this without the optimal
combination of ventilation with the phased
compressions and decompressions (39) and
the use of Lifestick resuscitation after pro-
longed standard cardiopulmonary resusci-
tation. Therefore, future studies need to
focus on the earlier application of this tech-
nique with different ventilation compres-
sion ratios. The reports with regard to the
operation of the phased chest and abdomi-
nal compression– decompression tech-
nique confirm our findings of its feasibility.
Nevertheless, larger, randomized, prospec-
tive studies to answer the question on
whether phased thoracoabdominal com-
pression can achieve any real hemody-
namic improvement over conventional
cardiocerebral resuscitation (42) are inevi-
table.
Limitations. The conclusions of our
study are somewhat weakened by the
facts that the control groups were not
randomized, the physicians were not
blinded to the respective interventions,
and the relatively small sample size. Also,
informed consent was waived, and appro-
priately so, according to the Declaration
of Helsinki. Furthermore, we only in-
cluded patients after prolonged cardio-
pulmonary resuscitation; therefore, we
do not know whether our conclusions
would also be valid for patients after
short-term cardiopulmonary resuscita-
tion. However, there is usually a marked
deterioration in coronary perfusion pres-
sure and end-tidal CO2 pressure as the
time of unsuccessful resuscitation in-
creases. Therefore improvements with
the Thumper device are beyond what is
expected by a usual course.

1836 Crit Care Med 2008 Vol. 36, No. 6

CONCLUSION
Despite the statistical limitations due
to the relatively small number of patients
included in this study, it is reasonable to
conclude that the use of the Lifestick
device in resuscitations is both safe and
beneficial. Laboratory studies support
our findings about the positive hemody-
namic effects of the Lifestick. Clearly, fur-
ther studies covering a larger number of
patients will be needed to provide the
necessary statistical underpinning to our
conclusions.
REFERENCES
1. Proceedings of the 1985 National Conference
on Standards and Guidelines for Cardiopul-
monary Resuscitation and Emergency Car-
diac Care. July 11–13, Dallas, Texas. Circu-
lation 1986; 74:IV1–IV153
2. Guidelines 2000 for Cardiopulmonary Resus-
citation and Emergency Cardiovascular Care:
An international consensus on science. Cir-
culation 2000; 102:A3–I370
3. Guidelines 2000 for cardiopulmonary resus-
citation and emergency cardiovascular care:
An international consensus on science. Re-
suscitation 2000; 46:1– 430
4. Kouwenhoven WB, Jude JR, Knickerbocker
GG: Closed chest cardiac massage. JAMA
1960; 173:1064 –1067
5. Kouwenhoven WB, Jude JR, Knickerbocker
GG: Landmark article July 9, 1960: Closed-
chest cardiac massage. By W. B. Kouwen-
hoven, James R. Jude, and G. Guy Knicker-
bocker. JAMA 1984; 251:3133–3136
6. Crile G, Dolley DH: Experimental research
into resuscitation of dogs killed by anesthet-
ics and asphyxia. J Exp Med 1906; 8:713–715
7. Ditchey RV, Winkler JV, Rhodes CA: Relative
lack of coronary blood flow during closed-
chest resuscitation in dogs. Circulation
1982; 66:297–302
8. Bellamy RF, DeGuzman LR, Pedersen DC:
Coronary blood flow during cardiopulmo-
nary resuscitation in swine. Circulation
1984; 69:174 –180
9. Ralston SH, Voorhees WD, Babbs CF: In-
trapulmonary epinephrine during prolonged
cardiopulmonary resuscitation: Improved re-
gional blood flow and resuscitation in dogs.
Ann Emerg Med 1984; 13:79 – 86
10. Halperin HR, Tsitlik JE, Guerci AD, et al:
Determinants of blood flow to vital organs
during cardiopulmonary resuscitation in
dogs. Circulation 1986; 73:539 –550
11. Sanders AB, Ogle M, Ewy GA: Coronary per-
fusion pressure during cardiopulmonary re-
suscitation. Am J Emerg Med 1985; 3:11–14
12. Ditchey RV, Lindenfeld J: Failure of epineph-
rine to improve the balance between myocar-
dial oxygen supply and demand during
closed-chest resuscitation in dogs. Circula-
tion 1988; 78:382–389
Crit Care Med 2008 Vol. 36, No. 6
13. Lurie KG, Lindo C, Chin J: CPR: The P stands
for plumber’s helper. JAMA 1990; 264:1661
14. Cohen TJ, Tucker KJ, Lurie KG, et al: Active
compression-decompression: A new method
of cardiopulmonary resuscitation. Cardiopul-
monary Resuscitation Working Group. JAMA
1992; 267:2916 –2923
15. Sunde K, Wik L, Naess PA, et al: Effect of
different compression– decompression cycles
on haemodynamics during ACD-CPR in pigs.
Resuscitation 1998; 36:123–131
16. Callaham M: Active compression-decompres-
sion cardiopulmonary resuscitation. N Engl
J Med 1999; 341:1770 –1771
17. Skogvoll E, Wik L: Active compression-
decompression cardiopulmonary resuscita-
tion: A population-based, prospective ran-
domised clinical trial in out-of-hospital
cardiac arrest. Resuscitation 1999; 42:
163–172
18. Malzer R, Zeiner A, Binder M, et al: Hemo-
dynamic effects of active compression-
decompression after prolonged CPR. Resus-
citation 1996; 31:243–253
19. Babbs CF, Ralston SH, Geddes LA: Theoreti-
cal advantages of abdominal counterpulsa-
tion in CPR as demonstrated in a simple
electrical model of the circulation. Ann
Emerg Med 1984; 13:660 – 671
20. Babbs CF: CPR techniques that combine
chest and abdominal compression and de-
compression: Hemodynamic insights from a
spreadsheet model. Circulation 1999; 100:
2146 –2152
21. Babbs CF: Efficacy of interposed abdominal
compression-cardiopulmonary resuscitation
(CPR), active compression and decompres-
sion-CPR, and Lifestick CPR: Basic physiol-
ogy in a spreadsheet model. Crit Care Med
2000; 28:N199 –N202
22. Lin CK, Levenson H, Yamashiro SM: Optimi-
zation of coronary blood flow during cardio-
pulmonary resuscitation (CPR). IEEE Trans
Biomed Eng 1987; 34:473– 481
23. Arntz HR, Agrawal R, Richter H, et al: Phased
chest and abdominal compression-decom-
pression versus conventional cardiopulmo-
nary resuscitation in out-of-hospital cardiac
arrest. Circulation 2001; 104:768 –772
24. Bur A, Mullner M, Sterz F, et al: The emer-
gency department in a 2000-bed teaching
hospital: Saving open ward and intensive
care facilities. Eur J Emerg Med 1997;
4:19 –23
25. Pearlman AS, Gardin JM, Martin RP, et al:
Guidelines for physician training in trans-
esophageal echocardiography: Recommenda-
tions of the American Society of Echocardi-
ography Committee for Physician Training
in Echocardiography. J Am Soc Echocar-
diogr 1992; 5:187–194
26. Mateer JR, Stueven HA, Thompson BM, et al:
Interposed abdominal compression CPR ver-
sus standard CPR in prehospital cardiopul-
monary arrest: Preliminary results. Ann
Emerg Med 1984; 13:764 –766
27. Chandra N, Guerci A, Weisfeldt ML, et al:
Contrasts between intrathoracic pressures
during external chest compression and car-
diac massage. Crit Care Med 1981; 9:
789 –792
28. Chandra NC, Tsitlik JE, Halperin HR, et al:
Observations of hemodynamics during hu-
man cardiopulmonary resuscitation. Crit
Care Med 1990; 18:929 –934
29. Koetter KP, Maleck WH: Effectiveness of me-
chanical versus manual chest compressions
in out-of-hospital cardiac resuscitation. Am J
Emerg Med 1999; 17:210
30. Sommers MS: Potential for injury: Trauma
after cardiopulmonary resuscitation. Heart
Lung 1991; 20:287–293
31. Nagel EL, Fine EG, Krischer JP, et al: Com-
plications of CPR. Crit Care Med 1981; 9:424
32. Baubin M, Rabl W, Pfeiffer KP, et al: Chest
injuries after active compression-decompres-
sion cardiopulmonary resuscitation (ACD-
CPR) in cadavers. Resuscitation 1999; 43:
9 –15
33. Krischer JP, Fine EG, Davis JH, et al: Com-
plications of cardiac resuscitation. Chest
1987; 92:287–291
34. Redberg RF, Tucker KJ, Cohen TJ, et al:
Physiology of blood flow during cardiopul-
monary resuscitation: A transesophageal
echocardiographic study. Circulation 1993;
88:534 –542
35. Deshmukh HG, Weil MH, Rackow EC, et al:
Echocardiographic observations during car-
diopulmonary resuscitation: A preliminary
report. Crit Care Med 1985; 13:904 –906
36. Porter TR, Ornato JP, Guard CS, et al: Trans-
esophageal echocardiography to assess mi-
tral valve function and flow during cardio-
pulmonary resuscitation. Am J Cardiol 1992;
70:1056 –1060
37. Hackl W, Simon P, Mauritz W, et al: Echo-
cardiographic assessment of mitral valve
function during mechanical cardiopulmo-
nary resuscitation in pigs. Anesth Analg
1990; 70:350 –356
38. Feneley MP, Maier GW, Gaynor JW, et al:
Sequence of mitral valve motion and trans-
mitral blood flow during manual cardiopul-
monary resuscitation in dogs. Circulation
1987; 76:363–375
39. Kern KB, Hilwig RW, Berg RA, et al: Opti-
mizing ventilation in conjunction with
phased chest and abdominal compression-
decompression (Lifestick) resuscitation. Re-
suscitation 2002; 52:91–100
40. Wenzel V, Lindner KH, Prengel AW, et al:
Effect of phased chest and abdominal com-
pression-decompression cardiopulmonary
resuscitation on myocardial and cerebral
blood flow in pigs. Crit Care Med 2000; 28:
1107–1112
41. Tang W, Weil MH, Schock RB, et al: Phased
chest and abdominal compression-decom-
pression: A new option for cardiopulmo-
nary resuscitation. Circulation 1997; 95:
1335–1340
42. Ewy GA: Cardiocerebral resuscitation: The
new cardiopulmonary resuscitation. Circula-
tion 2005; 111:2134 –2142
1837