2018 Out-of-hospital cardiac arrest outcomes with pitcrew and LUCAS.pdf

Accepted Manuscript
Out-of-hospital cardiac arrest outcomes with “pit crew”
resuscitation and scripted initiation of mechanical CPR
Louis Gonzales, Brandon K. Oyler, Jeff L. Hayes, Mark E. Escott,
Jose G. Cabanas, Paul R. Hinchey, Lawrence H. Brown
PII: S0735-6757(18)30666-1
DOI: doi:10.1016/j.ajem.2018.08.031
Reference: YAJEM 57740
To appear in: American Journal of Emergency Medicine
Received date: 14 May 2018
Revised date: 25 June 2018
Accepted date: 8 August 2018
Please cite this article as: Louis Gonzales, Brandon K. Oyler, Jeff L. Hayes, Mark E.
Escott, Jose G. Cabanas, Paul R. Hinchey, Lawrence H. Brown , Out-of-hospital cardiac
arrest outcomes with “pit crew” resuscitation and scripted initiation of mechanical CPR.
Yajem (2018), doi:10.1016/j.ajem.2018.08.031
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
Manual vs. Mechanical CPR
Out-of-Hospital Cardiac Arrest Outcomes with "Pit Crew" Resuscitation and Scripted Initiation
of Mechanical CPR
Louis Gonzalesa
, Brandon K. Oylerb
, Jeff L. Hayesa
, Mark E. Escotta
, Jose G. Cabanasc
, Paul R.
Hincheyc
, Lawrence H. Brownb,d
a
Office of the Medical Director, Austin-Travis County Emergency Medical Services System,
Austin, TX, USA
b
Emergency Medicine Residency Program, Department of Surgery and Perioperative Care, Dell
Medical School at the University of Texas, Austin, TX, USA
c
Wake County Emergency Medical Services, Raleigh, NC, USA
d
James Cook University, Mount Isa Centre for Rural and Remote Health, Townsville, QLD,
Australia
Presented at the National Association of EMS Physicians Annual Meeting, San Diego, CA,
January 2018.
Grant Support: None
Conflicts of Interest: None
Word Count: 4210
LG, JLH, JGC, PRH and LHB conceived and designed the study. LG, JLH, JGC and PRH supervised
the conduct of the trial and data collection. LG and LHB managed the data, including quality
control. LHB provided statistical advice on study design and analyzed the data; LG, BKO, MEE,
JLH, JGC, PRH and LB reviewed and interpreted the data. LG, BKO and LHB drafted the
manuscript, and all authors contributed substantially to its revision. LG and LHB take
responsibility for the paper as a whole.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
ABSTRACT:
Objective: To compare OHCA outcomes in patients managed with mechanical versus manual
CPR in an EMS system with a "pit crew" approach to resuscitation and a scripted sequence for
the initiation of mechanical CPR.
Methods: Through a year-long quality improvement effort we standardized the initial
resuscitative efforts for OHCA, prioritizing a "pit crew" approach to high quality manual CPR,
early defibrillation and basic airway management ahead of a scripted sequence for initiating
mechanical CPR. We then analyzed outcomes for adult, non-traumatic OHCA attended in the
following year (2016). We used a propensity score matched analysis to compare ROSC, survival
to discharge, and neurologic status among patients managed with manual versus mechanical
CPR while controlling for patient demographics and arrest characteristics.
Results: Of 444 eligible OHCAs, 217 received manual and 227 received mechanical CPR. Crude
ROSC (39.2% vs. 29.1%) and survival to discharge (13.8% vs. 5.7%) were higher with manual
CPR. In the propensity matched analysis (n = 176 manual CPR; 176 mechanical CPR), both ROSC
(38.6% vs. 28.4%; difference: 10.2%; CI: 0.4% to 20.0%) and survival to discharge (13.6% vs.
6.8%; difference: 6.8%; CI: 0.5% to 13.3%) remained significantly higher for patients receiving
manual CPR.
Conclusions: In this EMS system with a standardized, "pit crew" approach to OHCA that
prioritized initial high-quality initial resuscitative efforts and scripted the sequence for initiating
mechanical CPR, use of mechanical CPR was associated with decreased ROSC and decreased
survival to discharge.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
1. INTRODUCTION:
Out-of-hospital cardiac arrest (OHCA) resuscitation guidelines emphasize high-quality
cardiopulmonary resuscitation (CPR) with a chest compression rate of 100-120/minute and
minimal interruptions.1
The American Heart Association (AHA) does not recommend
mechanical chest compression devices except in special circumstances;2,3
neither randomized
controlled trials (RCTs)4-7
nor meta-analyses7-12
have demonstrated improved survival with
mechanical CPR compared with manual CPR. Still, given the evidence of similar outcomes7-12
and perceived logistic advantages related to manpower and safety during transport, mechanical
CPR remains popular among some prehospital providers and EMS systems.
Since 2011, the Austin-Travis County EMS (ATCEMS) system has utilized a highly
orchestrated "pit crew" (or "team-focused"/"high-performance") approach to resuscitation.
This is "a choreographed approach" to resuscitation that emphasizes early defibrillation,
optimal CPR performance metrics, limited interruptions, basic airway management, controlled
ventilation, and rescuer fatigue management.13-16
Over this same time period, use of
mechanical CPR devices has steadily increased among the various ATCEMS system first
response agencies. In 2013, a flattening in the historically upward trend in OHCA survival led
the ATCEMS system Office of the Medical Director (OMD) to undertake a comprehensive
quality improvement (QI) effort (detailed below) which revealed (1) lower OHCA survival rates
when mechanical CPR devices were available on scene, and (2) that application of mechanical
CPR devices was disrupting the choreographed approach to CPR causing interruptions in chest
compressions. As a result, the ATCEMS system and the OMD developed new protocols and
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
implemented an intensive education program re-emphasizing the standardized "pit crew"
approach to resuscitation and a scripted sequence for initiating mechanical CPR.
Here, we describe the QI effort and report the post-initiative outcomes of OHCA
patients managed with and without mechanical CPR. Our null hypothesis was that post-
initiative outcomes for OHCA patients—including return of spontaneous circulation (ROSC),
survival to discharge, and neurologic status—would not differ for patients managed with
mechanical versus manual CPR.
2. METHODS:
2.1 Design:
This was a retrospective observational cohort study. The IRB reviewed and approved
this study as exempt per CFR 46.101(b)(4).
2.2 Setting:
The ATCEMS system serves a metropolitan area of 1,023 square miles with a population
of approximately 1.2 million. Ambulances are staffed by two EMS providers, including at least
one paramedic. Fourteen career and volunteer fire departments deliver first response,
providing both CPR and defibrillation with automated external defibrillators (AEDs) for OHCA.
An accredited dispatch center staffed by certified emergency medical dispatchers prioritizes all
requests for service. Dispatchers provide telephone CPR instructions and use a computer-
assisted dispatch system to send a fire-based first response vehicle, the nearest ambulance and
a field supervisor to all presumed OHCAs. In 2016, the ATCEMS system responded to 131,825
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
incidents, including 801 attempted resuscitations of OHCAs. Approximately 75% of OHCAs
received first response by the Austin Fire Department (AFD), with the 13 (of 42) AFD first
response vehicles with a higher frequency of dispatches for OHCAs being equipped with
LUCAS® 2 Chest Compression System (Physio Control, Redmond, WA) mechanical CPR devices.
2.3 Quality Improvement Evaluation and Intervention:
The ATCEMS system maintains an Utstein-style cardiac arrest database and participates
in the Cardiac Arrest Registry to Enhance Survival (CARES). In late 2013 the OMD became aware
of a flattening in the historically upward trend in survival to discharge, accompanied by lower
survival rates when a mechanical CPR device was available on scene. This led the OMD to
undertake an evaluation of resuscitation performance to identify potential system-level
improvements. A convenience sample of 26 fire-based first responder crews was recruited to
demonstrate their management of a simulated cardiac arrest including (for crews with devices
available) the initiation of mechanical CPR. The simulations were observed in real time via video
from a remote room. The observers used a checklist to document each step in the
resuscitation, the order in which steps were performed, time intervals, and number and
duration of compression interruptions. Group debriefings were held after each simulation. As a
result of this evaluation, it became apparent that the process of applying the mechanical CPR
device was disrupting the choreographed approach to CPR and causing interruptions in chest
compressions.
Beginning in mid-2014 and into 2015, an intensive, multifaceted QI effort was
implemented to improve adherence to the "pit crew" process for resuscitation of OHCA. The
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
initiative included a new "CPR - Pit Crew" protocol (Supplementary Appendix 1) that prioritized
the initiation of high-quality, uninterrupted manual chest compressions, prompt utilization of
real-time CPR feedback devices, early defibrillation, and effective basic airway management
throughout the first 200 compressions delivered by first response personnel. Continued
uninterrupted high-quality CPR, indicated defibrillations and insertion of a supraglottic airway
were emphasized during the second set of 200 compressions, with the scripted sequence for
setting up and initiating mechanical CPR (when available) to be accomplished during the third
and fourth sets of 200 compressions. Additional interventions (IV or IO placement, medication
administration) are performed by responding ambulance personnel in parallel with CPR and
defibrillation once these responders arrive at the patient’s side, but only after high quality CPR
has been initiated/confirmed and defibrillation has been performed (if indicated). A
resuscitation checklist that scripted expected resuscitation actions, including the timing and
sequence for application of mechanical CPR when available, was developed for use as a
memory aid by first responders and EMS providers. Finally, all AFD operational field personnel
below the level of battalion chief received video-based re-training on the principles of "pit
crew" resuscitation, including high quality CPR with minimized interruptions of chest
compressions, the new "CPR - Pit Crew" protocol, the checklist, and the importance of
documenting mechanical CPR use. All AFD personnel assigned to a first response vehicle
equipped with a mechanical CPR device also received additional hands-on simulation-based
training and evaluation, practicing and demonstrating each role on the "pit crew" team, use of
the checklist, and the initiation of mechanical CPR—including the importance of securing the
device with the provided stabilization strap to keep the suction cup/pressure pad in-place.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
2.4 Population:
We subsequently evaluated outcomes among adult, non-traumatic OHCAs attended by
the ATCEMS system with first response by AFD during calendar year 2016. We used the CARES
case definition for OHCA: "a non-traumatic out-of-hospital cardiac arrest where resuscitation is
attempted by a 911 responder (CPR and/or defibrillation)."17
We excluded OHCA patients
attended by a first response agency other than AFD, as they often did not have mechanical CPR
available and were not required to participate in the hands-on simulation-based training
component of the QI initiative. We excluded patients who achieved ROSC after only bystander
CPR, or who achieved ROSC after only chest compressions and/or defibrillation without
advanced airway intervention or medication administration. Mechanical CPR would not be
indicated in these "immediate responders," which would create a bias favoring manual CPR. We
also excluded cases of EMS-witnessed arrest. EMS witnessed arrests have a higher probability
of survival, but the first responder agency's mechanical CPR device would be unlikely to be
available if the arrest occurred during ambulance transport. Thus, including EMS witnessed
arrests would likely create a bias favoring manual CPR. Finally, we excluded cases where
resuscitation was terminated on scene after only compressions/ventilations without
defibrillation, advanced airway intervention, or medication administration. Such cases typically
represent "minimal resuscitative efforts" performed while confirming DNR status or clinical
signs of obvious death (e.g., lividity), and including them would likely create a bias against
manual CPR.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
2.5 Data Sources:
All ATCEMS system data were exported from the CARES database into an Excel
Spreadsheet (Microsoft Corporation, Redmond, WA), and then imported into Stata (Stata MP,
College Station, TX) for analysis. Originally designed as a surveillance database, CARES uses
minimal data elements with standardized definitions and outcome measures primarily derived
from National EMS Information System (NEMSIS) standards and Utstein guidelines. All CARES
data fields included in this study are required fields in the ATCEMS System first responder and
transport unit electronic patient care reports, are stored within cardiac monitoring devices,
and/or are obtained by electronic data exchange with receiving hospitals, resulting in 100%
data capture. Laminated copies of the resuscitation checklist placed in all AED carrying cases
were available to guide and verify resuscitation actions and sequencing, but checklist items
were not physically 'marked off' and they did not serve as a data source for this analysis..
2.6 Intervention:
The intervention of interest was application and use of a mechanical CPR device. EMS
providers document all prehospital interventions, including manual and mechanical CPR, using
a standardized menu in the electronic patient care report. Cases where use of mechanical CPR
is not selected from the menu but is documented in the narrative are coded in the CARES
database by reviewing QI personnel as having received mechanical CPR.
2.7 Outcome Measures:
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
The primary outcome of interest was survival to hospital discharge. Secondary
outcomes included neurologic status at hospital discharge and intermediary outcomes such as
sustained ROSC at any time, termination of resuscitation on scene, pulse on arrival at the
emergency department, and survival to hospital admission. Survival to discharge was defined as
survival to the point of discharge from the hospital, whether to home or to a rehabilitation or
long-term care facility.17
Consistent with other evaluations of OHCA outcomes,5,6,18
neurologic
outcome among survivors was classified using cerebral performance category (CPC), recorded
as: (1) good cerebral performance; (2) moderate cerebral disability; (3) severe cerebral
disability; (4) coma or vegetative state.17,19
ROSC was defined as sustained ROSC for greater
than 20 consecutive minutes at any time during the resuscitation, regardless of whether re-
arrest subsequently occurred.17
2.8 Analysis:
We describe patient demographics and case characteristics using frequencies and
medians with inter-quartile ranges (none of the continuous data were normally distributed).
Chi-square and Wilcoxon rank sum test, as appropriate, were used to compare characteristics
across the mechanical and manual CPR cohorts. All continuous data were analyzed in their raw
form, without recoding into categories (e.g., age groups) or other transformation.
During the study period, mechanical CPR devices were not available on every first
response unit; use was based primarily on (1) the availability of a mechanical CPR device on the
fire-based first response vehicle, and (2) whether patients achieved ROSC after only initial
resuscitation attempts. Because allocation to mechanical versus manual CPR was not
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
randomized, we used 1:1 nearest neighbor propensity score matching to control for potential
allocation bias. Propensity score matching selects intervention and control cases with similar
baseline demographic and clinical characteristics and is common in observational studies of
OHCA.20-22
Multivariable logistic regression incorporating variables known to be associated with
cardiac arrest outcomes (age; sex; race; witnessed arrest; bystander CPR; etiology; presenting
rhythm; medication administration; location) was used to estimate the probability of receiving
mechanical CPR—the propensity score. For each included case managed with mechanical CPR,
the manual CPR case with the most similar propensity score was then selected as a control. We
assessed the success of the propensity matching by calculating post-matching standardized
differences23
and by evaluating the post-matching variances ratios (V(mechanical)/V(manual)
for each variable used in the matching process.
We report the absolute difference in the frequency of outcomes among the manual and
mechanical CPR groups in the matched cases for survival, ROSC, and other binary outcomes.
We evaluated differences in ranked CPC of survivors using Wilcoxon rank sum test. We also
report the frequency of outcomes in important subgroups of OHCA patients.
3. RESULTS:
Of the 801 attempted resuscitations of OHCAs during calendar year 2016, 700 were
confirmed adult, non-traumatic OHCAs. Forty-eight patients achieved ROSC after only initial,
basic resuscitative efforts; a non-AFD agency provided first response for 167 of the cases; and
86 cases were excluded for other reasons (See Figure 1). Table 1 shows the demographic and
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
case characteristics for the 444 included cases of adult, non-traumatic arrest. Included cases
managed with manual CPR were more likely to be witnessed OHCA than those managed with
mechanical CPR (45.6% vs. 36.1%, p=0.045) and were less likely to have received sodium
bicarbonate during the resuscitation (77.0% vs. 87.7%, p=0.003), otherwise there were no
statistically significant differences between the two groups. Table 1 also shows the
demographic and case characteristics for the 352 cases included in the propensity score
matched analysis. The propensity matching was successful: after matching all of the
standardized differences for the variables used in the propensity matching were less than 0.1
and all of the V(mechanical/V(manual) ratios were between 0.75 and 1.33, with the exception
of sodium bicarbonate administration. There were no statistically significant differences in the
characteristics of the cases managed with manual versus mechanical CPR.
3.1 Unmatched Analysis:
Table 2 shows the unadjusted outcome measures for the manual and mechanical CPR
groups. Overall, 43 (9.7%) of the included patients survived to hospital discharge. Patients
managed with manual CPR were more likely to survive than patients managed with mechanical
CPR (OR=2.6, CI: 1.3-5.3). Patients managed with manual CPR were also more likely to achieve
ROSC (OR=1.6, CI: 1.1-2.3). When limiting the analysis to only transported patients, those
managed with manual CPR were still more likely to arrive at the emergency department with a
pulse and/or survive to hospital admission and hospital discharge. There was no difference in
CPC among survivors in the manual versus mechanical CPR cohorts (Table 2; Figure 2).
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
3.2 Propensity Score Matched Analysis:
Table 2 also shows the outcomes for the propensity score matched cases. Twenty-four
(13.6%) patients managed with manual CPR survived, compared with 12 (6.8%) patients
managed with mechanical CPR (difference=6.8%, CI: 0.5%-13.3%). Sixty-eight (38.6%) patients
managed with manual CPR achieved ROSC, compared with 50 (28.4%) patients managed with
mechanical CPR (difference=10.2%, CI: 0.4%-20.0%). There was no difference in CPC among
survivors in the manual versus mechanical CPR cohorts (Table 2; Figure 2).
3.3 Sub-Strata Analyses:
Table 3 compares survival to discharge within important sub-strata of OHCA patients for
both the unmatched and propensity score matched data. There was no sub-stratum of OHCA
patients for whom mechanical CPR was beneficial.
4. DISCUSSION:
Previous studies have demonstrated that on-scene application of mechanical CPR
devices is cumbersome and logistically challenging,4,7,13,14,25-27
often leading to 20 to 30 second
(or longer) interruptions in chest compressions14,26
which are not fully appreciated by on-scene
providers.26
Despite intensive efforts to standardize the resuscitation process and script the
initiation of mechanical CPR, we were unable to demonstrate improved outcomes over manual
CPR. In fact, utilization of mechanical CPR was associated with lower rates of survival to hospital
discharge, ROSC and other intermediary outcome measures. The findings were consistent
across the unadjusted and propensity score matched analyses.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
Our results might reflect a failure of our QI initiative rather than true differences
between manual and mechanical CPR. Our results, however, are consistent with those of other
QI initiatives addressing CPR performance and/or mechanical CPR. After a state-wide effort in
North Carolina, Pearson et al.13
reported better survival to discharge (35% vs.22%) and more
survivors with good neurological outcomes (28% vs.16%) among OHCA patients treated by
systems that adopted a team-focused approach to CPR. However, patients managed with
mechanical CPR were less likely to survive to discharge. Levy et al.14
described a QI initiative in
Anchorage, Alaska that focused on high-quality manual CPR and a standardized sequence for
initiating mechanical CPR that reduced interruptions for application from 21 seconds to 7
seconds, and increased chest compression fraction from 90% to 95%, but they did not report
outcomes. Sporer et al.15
described a longitudinal series of QI interventions in Alameda County,
CA—efforts to increase bystander CPR; high-quality CPR; advanced airway intervention; use of
impedance threshold devices; post resuscitation cooling; and expanded use of mechanical CPR.
Collectively these interventions did not improve survival to hospital discharge (12% vs.10%), but
they did increase both ROSC (34% vs.29%) and the number of survivors with CPC scores of 1 or
2 (8% vs.4.5%). Hopkins et al.16
reported higher rates of neurologically intact survival (16%
vs.8%) in Salt Lake City, Utah after an initiative incorporating team-focused CPR and a modified
medication algorithm. Although the specific details varied, our QI initiative incorporated many
of the same principles as these previous efforts.
Also consistent with prior studies,5-7,13,22,25
we found no benefit of mechanical CPR in
any of the important sub-strata of OHCA patients. We particularly considered the subgroups of
witnessed and unwitnessed arrests, as well as patients with and without administration of
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
sodium bicarbonate—which is likely a marker of prolonged resuscitation. Survival among OHCA
patients with unwitnessed arrest was uniformly poor and there were more witnessed arrests
among the manual CPR cases (46% vs.36%). However, the frequency of witnessed arrest was
not significantly different in the propensity score matched analysis, and when considering only
witnessed arrest patients the survival rate remained higher for patients managed with manual
CPR (26.3% vs. 14.6% unmatched; 27.4% vs. 16.9% propensity matched) (Table 3). Similarly, the
survival rate for patients managed with manual CPR was higher whether patients did (6.0% vs.
3.5% unmatched; 6.1% vs. 4.4% matched) or did not (40.0% vs. 21.4% unmatched; 51.7% vs.
27.8% matched) receive sodium bicarbonate (Table 3). One concerning finding of the stratified
results was the substantial difference in survival among subjects presenting with a shockable
rhythm (42.4% vs. 26.7%)—a finding that was also reported in the PARAMEDIC trial (24% vs.
18%, OR=0.71, CI: 0.52-0.98).7
The significant difference in survival to discharge in our study stands in contrast to
RCTs4-7
and meta-analyses7-12
which have generally found equivalent outcomes with mechanical
and manual CPR. However, three other observational studies evaluating CARES data13,22,25
have
also reported worse outcomes among patients managed with mechanical CPR, with risk
differences ranging from 4% to 7%. One potential explanation for the different findings of
randomized versus observational studies is potential bias in the allocation of patients to
mechanical versus manual CPR in non-randomized studies. Zeiner et al.18
have suggested that
such an allocation bias would probably favor the mechanical CPR arm of a study, as patients are
more likely to receive mechanical CPR when responders perceive they have a better chance of
survival. However, in some systems, providers might resort to mechanical CPR when
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
transporting OHCA patients for whom resuscitation is viewed as futile, which could create a
bias against mechanical CPR in observational studies. Our system supports termination on
scene when resuscitative efforts fail, and transport of cases perceived as futile is rare. We
found no differences in the scene providers' commitment to the resuscitation process when
measured as scene time or termination of resuscitation on scene. Most importantly, propensity
score matching to control for allocation bias did not meaningfully alter the results of our
analysis—nor those of Youngquist et al.22
Other explanations for the differences between the observational and randomized trials
can be found in the details of the RCTs. First, the intention to treat approach used by the RCTs is
considered "anti-conservative" for non-inferiority studies.28-30
That is, patients who respond to
only initial resuscitative attempts are counted as survivors in both the manual and mechanical
CPR arms, although mechanical CPR might never be initiated and is arguably not indicated for
such patients. Our approach evaluates manual and mechanical CPR only in those patients who
do not respond to initial resuscitative efforts. Both the LINC trial5
and the CIRC trail6
delayed
rhythm analysis and defibrillation in lieu of compressions for between 90 seconds and 3
minutes. Our study emphasized prompt rhythm analysis and (when indicated) defibrillation. In
the LINC trial, subjects were excluded—rather than included in the intention to treat analysis—
if the mechanical CPR device would not fit.5
In the PARAMEDIC trial,6
nearly 40% (638/1652) of
subjects allocated to the mechanical CPR arm received only manual CPR, while just 4%
(11/2819) of subjects allocated to the manual CPR arm crossed-over. Finally, in all four RCTs the
point estimate for survival was greater in the manual CPR arm, although the differences were
not statistically significant.4-7
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
As a result of this analysis, our system now uses mechanical CPR only for moving
patients, managing rescuer fatigue in prolonged resuscitations, and/or for ongoing
resuscitations during transport to hospital. Still, mechanical CPR remains popular among many
EMS systems and providers, including first responders in our system. Some argue there might
be a subgroup of patients who do benefit from mechanical CPR, although no studies to date
have identified such a subgroup,5-12,22,25,31,32
and we did not find any sub-strata of patients for
whom mechanical CPR was clearly beneficial. There is some evidence that mechanical CPR
generates better CPR performance metrics during ambulance transport,33-36
and one small
before-and-after study from Japan reported better outcomes after adopting mechanical CPR for
helicopter transport of OHCA patients.37
Finally, there is a manpower advantage to mechanical
CPR. Any of these could be reasonable arguments for using mechanical CPR if outcomes were in
fact equivalent between the two strategies when used in clinically relevant scenarios. The
observational studies to date raise questions about that equivalence. One study has suggested
that EMS systems that use mechanical CPR in a higher proportion of their OHCAs are more
likely to achieve survival rates equivalent to those of manual CPR, but the actual difference in
mechanical CPR survival rates for the highest use (>75% of cases) and lowest use (<25% of
cases) systems is small: 8.7% vs. 8.5%.25
4.1 Limitations:
This was a single system, retrospective QI study. The observational, non-randomized
nature of our analysis presents some risk of bias in the allocation of patients to manual or
mechanical CPR, particularly depending on the perceived utility or futility of prolonged CPR. In
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
our study, crew commitment to the resuscitative effort did not appear to differ for the manual
and mechanical CPR patients: there was no significant difference in the frequency of
termination of resuscitation on scene or in median on scene times for transported cases.
Observational studies of OHCA are also at risk for "resuscitation time bias"—that is,
early responders are less likely to be exposed to an intervention, which creates a bias favoring
non-exposure.24
To minimize this risk, we excluded patients who achieved ROSC after only CPR
and/or defibrillation during the initial compression cycles. While this limits our analysis to
OHCAs where initial resuscitative efforts fail to produce an immediate response, that is equally
true for the manual and mechanical CPR cohorts in our study—and is the clinically relevant
scenario. If early ROSC can be achieved with only basic resuscitative efforts, then initiation of
mechanical CPR is unwarranted. We also excluded EMS witnessed arrests and patients who
received only minimal resuscitation attempts before termination on scene. As expected,
initiation of mechanical CPR was less common in excluded cases (17.6%), and there was no
group of excluded patients for whom mechanical CPR was grossly beneficial (Table 4). There
might be other, unmeasured factors associated with the decision to utilize mechanical CPR that
could explain differences in outcomes, but the propensity matching in our study has accounted
for the known, common confounders in OHCA research.
Our analysis is based on reported CARES data elements, not direct observation of
resuscitation performance and/or protocol compliance. Although all OHCA cases are reviewed
by QI personnel—and we did not identify any systematic trends in protocol non-compliance
during the study period—it is possible that reported care differs from actual care. Further,
some data points that are not required CARES elements might be viewed as potential
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
determinants of OHCA outcomes, for example: CPR performance metrics, the number of
defibrillations delivered and the number of medication doses administered, and detailed time
intervals. Our providers use real-time CPR performance feedback to guide their resuscitative
efforts; although we did not collect and analyze performance data (e.g., compression fraction,
pauses, capnography), our first response and ambulance personnel were trained to
immediately respond to real time CPR performance feedback. The CARES data do not include
how many defibrillations were administered in each individual patient. We do know which
medications they received but not the actual dosing or number of administrations. We only
analyzed scene time for subjects transported to hospital: on-scene times for OHCAs that are
terminated on scene are often protracted for logistical, non-clinical reasons. Finally, we did not
track the identities of individual crew members. It is possible that some fire-based first
response personnel were consistently assigned to first response vehicles with or without
mechanical CPR devices. Ambulance staffing in our system is more dynamic, and individual
paramedics would have equal probability of being dispatched to OHCAs with and without
mechanical CPR available on-scene.
We did not control how hospitals dealt with patients who arrived with mechanical CPR
in progress, but there was only one survivor who did not achieve ROSC in the field. There were
only minor, non-significant differences in the post-resuscitation care of subjects who survived
to hospital admission (Table 5), which are likely explained by individual patient characteristics,
clinical indications and contraindications.
Finally, our system exclusively uses the battery-driven LUCAS® 2 mechanical CPR device.
Perhaps our outcomes and the effectiveness of our QI initiative would differ if we had used
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
some other mechanical CPR device, but the clinical trials and meta-analyses to date have not
suggested a differential effect of varying mechanical CPR devices.5-13,22,25
5. CONCLUSION:
Despite our comprehensive efforts to standardize a "pit crew" approach to resuscitation
and to script the sequence for initiating mechanical CPR, survival to discharge remained more
likely for OHCA patients managed with manual CPR. Manual CPR was also associated with
higher rates of ROSC and other intermediary outcome measures, although the neurologic status
of survivors did not differ for patients managed with manual versus mechanical CPR.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
6. REFERENCES:
1. American Heart Association. Adult basic life support and CPR quality: HCP BLS. In: Highlights
of the 2015 American Heart Association Guidelines Update for CPR and ECC. Dallas, TX:
American Heart Association, 2015, pp: 8-11.
2. American Heart Association. Alternative techniques and ancillary devices for CPR. In:
Highlights of the 2015 American Heart Association Guidelines Update for CPR and ECC. Dallas,
TX: American Heart Association, 2015, pp: 11-3.
3. Brooks SC, Anderson ML, Bruder E, et al. Part 6: alternative techniques and ancillary devices
for cardiopulmonary resuscitation: 2015 American Heart Association Guidelines Update for
Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation.
2015;132(suppl. 2):S436–43.
4. Smekel D, Johansson J, Huzevka T, Rubertsson S. A pilot study of mechanical chest
compressions with LUCASTM
device in cardiopulmonary resuscitation. Resuscitation.
2011;82:702-6.
5. Rubertsson S, Lindgren E, Smekal D, et al. Mechanical chest compressions and simultaneous
defibrillation vs.conventional cardiopulmonary resuscitation in out-of-hospital cardiac arrest:
The LINC randomized trial. JAMA. 2014; 311:53-61.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
6. Wik L, Olsen J-A, Persse D, et al. Manual vs. integrated automatic load-distributing band CPR
with equal survival after out of hospital cardiac arrest. The randomized CIRC trial. Resuscitation.
2014;85:741-8.
7. Perkins GD, Lall R, Quinn T, et al. Mechanical versus manual chest compression for out-of-
hospital cardiac arrest (PARAMEDIC): a pragmatic cluster randomized controlled trail. Lancet.
2015;385:947-55.
8. Bonnes JL, Brouwer MA, Navarese EP, et al. Manual cardiopulmonary resuscitation versus
CPR including a mechanical chest compression device in out-of-hospital cardiac arrest: A
comprehensive meta-analysis from randomized and observational studies. Ann Emerg Med.
2016;67:349-60.
9. Li H, Wang D, Yu Y, Zhao X, Jing X. Mechanical versus manual chest compressions for cardiac
arrest: a systematic review and meta-analysis. Scand J Trauma, Resus & Emerg Med.
2016;24:10. DOI: 10.1186/s13049-016-0202-y.
10. Gates S, Quinn T, Deakin CD, Blair L, Couper K, Perkins GD. Mechanical chest compression
for out-of-hospital cardiac arrest: Systematic review and meta-analysis. Resuscitation.
2015;94:91-7.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
11. Brooks SC, Hassan N, Bigham BL, Morrison LJ. Mechanical versus manual chest
compressions for cardiac arrest. Cochrane Database of Systematic Reviews. 2014; CD007260.
DOI: 10.1002/14651858.CD007260.pub3.
12. Ong ME, Mackey KE, Zhang ZC, et al. Mechanical CPR devices compared to manual CPR
during out-of-hospital cardiac arrest and ambulance transport: a systematic review. Scand J
Trauma, Resus & Emerg Med. 2012;20:39. http://www.sjtrem.com/content/20/1/39.
13. Pearson DA, Nelson RD, Monk L, et al. Comparison of team-focused CPR vs. standard CPR in
resuscitation from out-of-hospital cardiac arrest: Results from a statewide quality improvement
initiative. Resuscitation. 2016;105:165-72.
14. Levy M, Yost D, Walker RG, Scheunemann E, Mendive SR. A quality improvement initiative
to optimize use of a mechanical chest compression device within a high-performance CPR
approach to out-of-hospital cardiac arrest resuscitation. Resuscitation. 2015;92:32-7.
15. Sporer K, Jacobs M, Derevin L, Cuval S, Pointer J. Continuous quality improvement efforts
increase survival with favorable neurologic outcome after out-of-hospital cardiac arrest.
Prehosp Emerg Care. 2017;21:1-6.
16. Hopkins CL, Burk C, Moser S, Meersman J, Baldwin C, Youngquist ST. Implementation of pit
crew approach and cardiopulmonary resuscitation metrics for out-of-hospital cardiac arrest
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
improves patients survival and neurological outcome. J Am Heart Assoc. 2016;5:e002892. DOI:
10.1161/JAHA.115.002892.
17. Cardiac Arrest Registry to Enhance Survival. Complete Data Set for EMS, Hospital, and CAD
Participants and Instructions for Abstracting and Coding Data Elements. Atlanta, GA: CARES
Program, 2013.
18. Zeiner S, Sulzgruber P, Datler P, et al. Mechanical chest compression does not seem to
improve outcome after out-of-hospital cardiac arrest. A single center observational trial.
Resuscitation. 2015;96:220-5.
19. Safar P. Resuscitation after Brain Ischemia. In: Grenvik A, Safar P (eds). Brain Failure and
Resuscitation. New York, NY: Churchill Livingstone, 1981, pp: 155-184.
20. Lamhaut L, Hutin A, Puymirat E, et al. A pre-hospital extracorporeal cardio pulmonary
resuscitation (ECPR) strategy for treatment of refractory out hospital cardiac arrest: An
observational study and propensity analysis. Resuscitation. 2017;117;109-17.
21. Shin SD, Ahn KO, Song KY, Park CB, Lee EJ. Out-of-hospital airway management and cardiac
arrest outcomes: A propensity score matched analysis. Resuscitation. 2012;83:313-9.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
22. Youngquist ST, Ockerse P, Hartsell S, Stratford C, Taillac P. Mechanical chest compression
devices are associated with poor neurological survival in a statewide registry: A propensity
score analysis. Resuscitation. 2016;106:102-7.
23. Haukoos JS, Lewis RJ. The propensity score. JAMA. 2015;314:1637-8.
24. Andersen LW, Grossestreuer AV, Donnino MW. "Resuscitation time bias" – A unique
challenge for observational cardiac arrest research. Resuscitation. 2018;125:79-82.
25. Buckler DG, Burke RV, Naim MY, et al. Association of mechanical cardiopulmonary
resuscitation device use with cardiac arrest outcomes. Circulation. 2016;134:2131-3.
26. Yost D, Phillips RH, Gonzales L, et al. Assessment of CPR interruptions from transthoracic
impedance during use of the LUCASTM
mechanical chest compression system. Resuscitation.
2012;83:961-5.
27. Esibov A, Banville I, Chapman FW, Boomars R, Box M, Rubertsson S. Mechanical chest
compression improved aspects of CPR in the LINC trial. Resuscitation. 2015;91:116-21.
28. Gillespie D, Farewell D, Barrett-Lee P, et al. The use of randomisation-based efficacy
estimators in non-inferiority trials. Trials. 2017;18:117.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
29. Jones B, Jarvis P, Lewis J, Ebbutt A. Trials to assess equivalence: the importance of rigorous
methods. BMJ. 1996;313(7048):36.
30. ICH Steering Committee. Statistical principles for clinical trials (E9). Geneva, Switzerland:
International Conference on Harmonisation of Technical Requirements for Registration of
Pharmaceuticals for Human Use. 1998.
31. Hardig BM, Lindgren E, Ostlund O, Herlitz J, Karlsten R, Rubertsson S. Outcome of VF/VT
patients in the LINC (LUCAS IN cardiac arrest) trial—A randomized, controlled trial.
Resuscitation. 2017;115:155-62.
32. Rubertsson S, Lindgren E, Smekal D, et al. Per-protocol and pre-defined population analysis
of the LINC study. Resuscitation. 2015;96:92-9.
33. Lyon RM, Crawford A, Crookston C, Short S, Clegg GR. The combined use of mechanical CPR
and a carry sheet to maintain quality resuscitation in out-of-hospital cardiac arrests patients
during extrication and transport. Resuscitation. 2015;93:102-6.
34. Kim TH, Shin SD, Song KJ, et al. Chest compression fraction between mechanical
compressions on a reducible stretcher and manual compressions on a standard stretcher during
transport in out-of-hospital cardiac arrests: The ambulance stretcher innovation of Asian
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
cardiopulmonary resuscitation (ASIA-CPR) pilot trial. Prehosp Emerg Care. 2017;(e-pub ahead of
print). DOI: 10.1080/10903127.2017.1317892.
35. Kim TH, Hong KJ, Shin SD, et al. Quality between mechanical compression on reducible
stretcher versus manual compression on a standard stretcher in small elevator. Am J Emerg
Med. 2016;34:1604-9.
36. Sunde K, Wik L, Steen PA. Quality of mechanical, manual standard and active compression-
decompression CPR on the arrest site and during transport in a manikin model. Resuscitation.
1997; 34:235-42.
37. Omori K, Sato S, Sumi Y, et al. The analysis of efficiency for AutoPulseTM
system in flying
helicopter. Resuscitation. 2013;84:1045-50.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
FIGURE LEGENDS:
Figure 1: Included cases.
*Some cases met more than one exclusion criteria.
Figure 2: Neurologic outcomes among survivors in the unmatched and propensity score
matched analyses.
ACCEPTED MANUSCRIPT

ACCEPTED
M
ANUSCRIPT
Table 1: Demographic and case characteristics for the 444 included cases and 352 propensity matched cases.
444 Included Cases 352 Propensity Matched Cases
Manual
CPR
Mechanical
CPR
Manual
CPR
Mechanical
CPR
Standardized
Difference
V(mechanical)/
V(manual)
N 217 227 176 176
Demographics
Age (years), median (IQR) 64 (52-76) 65 (55-77) 63 (51-75) 62 (55-76) -0.06 0.84
Male, N (%) 130 (59.9) 142 (62.6) 106 (60.2) 110 (62.5) -0.05 0.98
White non-Hispanic, N (%)* 106 (57.0) 120 (61.5) 100 (56.8) 104 (59.1) -0.05 0.99
Location
At Home, N (%) 128 (59.0) 138 (60.8) 104 (59.1) 108 (61.4) -0.05 0.98
At Medical Facility, N (%) 15 (6.9) 11 (4.9) 11 (6.3) 9 (5.1) 0.05 0.83
At Nursing Home, N (%) 31 (14.3) 35 (15.4) 27 (15.3) 25 (14.2) 0.03 0.94
Other/Public Places, N (%) 43 (19.8) 43 (18.9) 34 (19.3) 34 (19.3) -- --
Arrest Characteristics
ACCEPTED MANUSCRIPT

ACCEPTED
M
ANUSCRIPT
Witnessed, N (%) 99 (45.6) 82 (36.1)†
75 (42.6) 68 (38.6) 0.08 0.97
Bystander CPR, N (%) 84 (38.7) 88 (38.8) 68 (38.6) 70 (39.8) -0.03 1.01
AED Applied, N (%) 69 (31.8) 88 (38.8) 56 (31.8) 71 (40.3) -- --
Any AED Shock, N (%) 22 (10.1) 22 (9.7) 16 (9.1) 19 (10.8) -- --
Cardiac Etiology, N (%) 183 (84.3) 196 (86.3) 146 (83.0) 151 (85.8) -0.08 0.86
Asystole, N (%) 122 (56.2) 116 (51.1) 33 (18.8) 30 (17.1) 0.04 1.00
Shockable Rhythm, N (%) 41 (18.9) 37 (16.3) 101 (57.4) 100 (56.8) 0.02 0.93
EMS Management
Times (min), median (IQR)
Response Time 8.8 (7.1-10.6) 8.6 (6.7-10.5) 8.7 (7.0-10.3) 8.6 (6.6-10.3) 0.03 1.02
Scene Time** 28 (21-38) 27 (22-34) 27 (22-35) 29 (25-39) -- --
Advanced Airway, N (%) 172 (79.3) 189 (83.3) 149 (80.1) 153 (82.3) -0.06 0.97
ETT, N (%) 16 (7.4) 19 (8.4) 14 (7.5) 17 (9.1) -0.06 1.19
I-gel, N (%) 156 (71.9) 170 (74.9) 135 (72.6) 136 (73.1) -0.01 1.03
ITD used, N (%) 191 (88.0) 207 (91.6) 165 (88.7) 168 (90.3) -- --
Received Medications, (%) 216 (99.5) 227 (100.0) 176 (100.0) 176 (100.0) 0.00 n/a
ACCEPTED MANUSCRIPT

ACCEPTED
M
ANUSCRIPT
Epinephrine 207 (95.4) 223 (98.2) 176 (100.0) 176 (100.0) 0.00 n/a
Amiodarone 41 (18.9) 46 (20.3) 37 (21.0) 40 (22.7) -- --
Lidocaine 5 (2.3) 10 (4.4) 4 (2.3) 8 (4.6) -- --
Atropine 1 (0.5) 1 (0.4) 1 (0.6) 0 (0.0) -- --
Dextrose 33 (15.2) 37 (16.3) 29 (16.5) 35 (19.9) -- --
Sodium Bicarbonate 167 (77.0) 199 (87.7)†
147 (83.5) 158 (89.8) -0.19 0.67
Other 6 (2.8) 4 (1.8) 0 (0.0) 0 (0.0) -- --
IQR = inter-quartile range; Resp. Time = response time; min = minutes; ETT = endotracheal intubation; ITD = impedance threshold device; CPC =
cerebral performance category; Resus = resuscitation; ED = emergency department; †
p<0.05, chi-square; *race/ethnicity data missing for 95
(13.6%) cases; **scene time and resuscitation time available for only 177 transported cases; -- = not used in the propensity matching.
ACCEPTED MANUSCRIPT

ACCEPTED
M
ANUSCRIPT
Table 2: Outcomes for the 444 included cases and the 352 propensity matched cases.
All 444 Included Cases 352 Propensity Matched Cases
Manual CPR
(N=217)
Mech. CPR
(N=227)
Unadjusted
Difference
Manual CPR
(N=176)
Mech. CPR
(N=176)
Adjusted
Difference (CI)
Outcomes
Survival to Discharge, N (%) 30 (13.8) 13 (5.7) 8.1% 24 (13.6) 12 (6.8) 6.8% (0.5%, 13.3%)
CPC = 1 or 2, of Survivors, n/N (%) 25/30 (83.3) 11/13 (84.6) -1.3% 21/24 (87.5) 10/12 (83.3) 4.2% (-20.4, 28.8%)
CPC, of Survivors, Median [IQR] 1 (1-1) 1 (1-1) n/a‡
1 (1-2) 1 (1-1) n/a‡
Sustained ROSC, N (%) 85 (39.2) 66 (29.1) 10.1% 68 (38.6) 50 (28.4) 10.2% (0.4%, 20.0%)
Resus. Terminated on Scene, N (%) 111 (51.2) 123 (54.2) -3.0% 94 (53.4) 101 (57.4) -4.0% (-6.4%, 14.4%)
Transported Cases Only, N 106 104 82 75
Pulse upon Arrival at ED, N (%) 73 (68.9) 50 (48.1) 20.8% 59 (72.0) 36 (48.0) 24.0% (9.0%, 38.9%)
Hospital Admission, N (%) 69 (65.1) 51 (49.0) 16.1% 55 (67.1) 37 (49.3) 17.7% (2.5%, 33.0%)
Survival to Discharge, N (%) 30 (28.3) 13 (12.5) 15.8% 24 (29.3) 12 (16.0) 13.3% (0.4%; 23.8%)
Mech CPR = Mechanical CPR; IQR = inter-quartile range; Resus = resuscitation; ED = emergency department; ‡
= p>0.05, Wilcoxon Rank Sum test
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
Table 3: Survival to discharge (%) in important sub-strata of OHCA.
Unmatched Cases
N=444
Propensity Score Matched Cases
N=352
Manual CPR Mechanical CPR Manual CPR Mechanical CPR
Male 13.9 5.6 15.1 7.3
Female 13.8 5.9 11.4 6.1
White, non-Hispanic 13.2 7.5 12.0 8.7
Non-white and/or Hispanic 15.0 5.3 15.8 4.2
Witnessed 26.3 14.6 28.0 16.2
Unwitnessed 3.4 0.7 3.0 0.9
Bystander CPR 20.2 6.8 19.1 8.6
No Bystander CPR 9.8 5.0 10.2 5.7
AED Applied* 20.3 8.0 19.6 9.9
At Home 10.9 2.2 9.6 2.8
At Medical Facility 13.3 9.1 18.2 11.1
At Nursing Home 3.2 2.9 3.7 4.0
Cardiac Etiology 13.7 5.1 13.0 6.6
Non-cardiac Etiology 14.7 9.7 16.7 8.0
Shockable Rhythm 41.5 21.6 42.4 26.7
Non-shockable Rhythm 7.4 2.6 7.0 2.7
Asystole 2.5 1.7 2.0 1.0
Sodium Bicarbonate Given 6.0 3.5 6.1 4.4
No Sodium Bicarbonate 40.0 21.4 51.7 27.8
* Bystander or First Responder
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
Table 4: Survival among excluded subjects.
Exclusion Criteria* Manual CPR Survival Mechanical CPR Survival
ROSC after CPR/Defibrillation Only 30/43 (69.8%) 2/5 (40.0%)
Other First Response Agency 16/148 (10.8%) 2/19 (10.5%)
EMS Witnessed 9/39 (23.1%) 1/25 (4.0%)
Terminated after Minimal Efforts 0/22 (0.0%) 0/1 (0.0%)
Any Exclusion 44/211 (20.9%) 5/45 (11.1%)
* Some cases met more than one exclusion criteria.
ACCEPTED MANUSCRIPT

A
C
C
E
P
T
E
D
M
A
N
U
S
C
R
I
P
T
Table 5: In-hospital interventions for subjects who survived to hospital admission.
Manual CPR
(N=55)
Mechanical CPR
(N=37) Sig.*
Induced Hypothermia 41 (74.5) 26 (70.3) 0.448
Angiography 17 (30.9) 12 (32.4) 0.972
Stent Placement 6 (10.9) 8 (21.6) 0.198
* Chi Square
ACCEPTED MANUSCRIPT

2018 Out-of-hospital cardiac arrest outcomes with pitcrew and LUCAS.pdf

Recommended

Recommended

More Related Content

Similar to 2018 Out-of-hospital cardiac arrest outcomes with pitcrew and LUCAS.pdf

Similar to 2018 Out-of-hospital cardiac arrest outcomes with pitcrew and LUCAS.pdf (20)

More from Robert Cole

More from Robert Cole (20)

Recently uploaded

Recently uploaded (20)

2018 Out-of-hospital cardiac arrest outcomes with pitcrew and LUCAS.pdf