S

Fb

DLRa

b

c

d

e

a

ARRA

KCCCCCPCCMPSMM

1

i

i

C

vSf

0h

Resuscitation 84 (2013) 1131–1136

Contents lists available at ScienceDirect

Resuscitation

journa l homepage: www.e lsev ier .com/ locate / resusc i ta t ion

imulation and education

orensic analysis of crib mattress properties on pediatric CPR quality—Can wealance pressure reduction with CPR effectiveness?�,��

ana E. Nilesa,∗, Matthew R. Malteseb, Akira Nishisakia,b, Thomas Seacriste, Jessica Leffelmana,arissa Hutchinsc, Nancy Schneckd, Robert M. Suttonb, Kristy B. Arbogaste,obert A. Bergb, Vinay M. Nadkarnia,b

Center for Simulation, Advanced Education and Innovation, The Children’s Hospital of Philadelphia, Philadelphia, PA, USADepartment of Anesthesiology and Critical Care Medicine, The Children’s Hospital of Philadelphia, Philadelphia, PA, USADepartment of Nursing, The Children’s Hospital of Philadelphia, Philadelphia, PA, USADepartment of Environmental Services, The Children’s Hospital of Philadelphia, Philadelphia, PA, USACenter for Injury Research and Prevention, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA

r t i c l e i n f o

rticle history:eceived 27 November 2012eceived in revised form 20 January 2013ccepted 28 January 2013

eywords:ardiopulmonary resuscitationhest compressionPRardiac arresthildrenediatricPR trainingPR manikinsattress

ressure reductionkin injury

a b s t r a c t

Introduction: Single mode, pressure reduction (PR) crib mattresses are increasingly employed in hospitalsto prevent skin injury and infection. However, single mode PR mattresses risk large mattress deflectionduring CPR chest compressions, potentially leading to inadequate chest compressions.Hypothesis: New, dual mode PR crib mattress technology provides less mattress deflection during chestcompressions (CCs) with similar PR characteristics for prevention of skin injury.Methods: Epochs of 50 high-quality CCs (target sternum–spine compression depth ≥38 mm) guidedby real-time force/deflection sensor (FDS) feedback were delivered to CPR manikin with realistic CCcharacteristics on two PR crib mattresses for four conditions: (1) single mode + backboard; (2) dualmode + backboard; (3) single mode − no backboard; and (4) dual mode − no backboard. Mattress dis-placement was measured using surface reference accelerometers. Mattress displacement ≥5 mm wasprospectively defined as minimal clinically important difference. PR qualities of both mattresses wereassessed by tissue interface pressure mapping.Results: During simulated high quality CC, single mode had significantly more mattress displacementcompared to dual mode (mean difference 16.5 ± 1.4 mm, p < 0.0001) with backboard. This difference wasgreater when no backboard was used (mean difference 31.7 ± 1.5 mm, p < 0.0001). Both single mode and

attress complianceattress deflection

dual mode met PR industry guidelines (mean surface pressure <50 mmHg).Conclusions: Chest compressions delivered on dual mode pressure reduction crib mattresses resultedin substantially smaller mattress deflection compared to single mode pressure reduction mattresses.Skin pressure reduction qualities of dual mode pressure reduction crib mattress were maintained. Werecommend that backboards continue to be used in order to mitigate mattress deflection during CPR on

soft mattresses.

. Introduction

Cardiopulmonary resuscitation (CPR) is a rare, although highmpact, event in critically ill children with an incidence of 1.8% of

� A Spanish translated version of the abstract of this article appears as Appendixn the final online version at http://dx.doi.org/10.1016/j.resuscitation.2013.01.033.�� Funding: Laerdal Foundation for Acute Care Medicine and Endowed Chair ofritical Care Medicine at the Children’s Hospital of Philadelphia.∗ Corresponding author at: Center for Simulation, Advanced Education and Inno-ation, Room 8NW100, Main Building, The Children’s Hospital of Philadelphia, 34thtreet and Civic Center Blvd, Philadelphia, PA 19104, USA. Tel.: +1 215 590 4039;ax: +1 215 590 4327.

E-mail address: niles@email.chop.edu (D.E. Niles).

300-9572/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.resuscitation.2013.01.033

© 2013 Elsevier Ireland Ltd. All rights reserved.

pediatric intensive care unit admissions.1 High quality CPR, specif-ically adequate chest compression (CC) rate and depth, improvesresuscitation outcome.2–6 Several studies have indicated that themovement of the mattress, or mattress deflection, under the patientduring CPR may decrease the effectiveness of chest compressions(CCs).7–10 Further, some real-time CPR feedback devices usingaccelerometers on the chest may overestimate depth of CPR com-pression on a soft surface.6,11

Recently, the pressure reduction mattress has been widely usedin adult and pediatric hospitals to minimize skin breakdown and

pressure ulcers. Those have been identified as a major healthcarechallenge with an estimated cost of US$2.2–3.6 billion per year.12

While CCs are more effective when the patient is on a firm surface,clearly the risk of pressure ulcers increases especially in patients

1 itation

wc

toPtcsowqOtmw

2

T

2

2

pp(mifioifaiocl

2

dassd5wLSbc(wttmCbtbbCf

132 D.E. Niles et al. / Resusc

ith reduced mobility. Balancing those two factors is an arduoushallenge for front-line critical care providers.

In our institution, clinical observations in the pediatric ICU iden-ified excessive movement and bouncing of the patient during CPRn single-mode (SM) pressure reduction (PR) ICU crib mattresses.roper CC depth was difficult to achieve due to excessive mat-ress deflection. Our team, which includes pediatric ICU leaders,linical staff, resuscitation experts, and equipment supply teams,ubsequently notified the mattress manufacturer. Based on our rec-mmendations, the mattress manufacturers agreed to collaborateith us to develop a new crib mattress designed to enhance CPR

uality with similar PR characteristics to the SM PR crib mattress.ur objective in this investigation was to quantify mattress deflec-

ion during simulated CCs on two hospital crib mattresses: SM cribattresses compared to a new dual-mode (DM) crib mattress, bothith and without a backboard using forensic engineering analysis.

. Methods

This study was approved by the Institutional Review Board athe Children’s Hospital of Philadelphia.

.1. Data collection

.1.1. Crib mattressesThe crib mattress that was in clinical use was a single mode

ressure reduction crib mattress (SM). This mattress operates in theressure reduction mode only. The new dual mode CPR mattressDM) is designed with two modes of operation: pressure reduction

ode and CPR mode. This dual mode PR/CPR mattress transitionsmmediately from pressure reduction mode to CPR mode with therst chest compression (high force and rapid movement). This typef mattress (viscoelastic flexible polyurethane foam) is typified byts slow recovery after compression. When a weight is positioned onoam, the foam progressively conforms to the shape of the object,nd after the weight is removed, the foam slowly reassumes itsnitial shape. However, with the rapid, high force and short durationf a CC, the foam does not quickly yield to the weight/force. Theentral characteristic include the foam’s ability to absorb shock andack of surface “springiness”.13

.1.2. Chest compression and mattress deflectionData were collected using the HeartStart4000 monitor/

efibrillator (Laerdal Medical, Stavanger, Norway) equipped withCC force and deflection sensor (FDS) and additional mattress

urface reference accelerometer. The device provides audiovi-ual feedback for CC quality (rate, depth, and leaning force) byouble-integration of an FDS internal accelerometer. Epochs of0 high quality CCs were delivered to a simulation manikinith realistic chest deflection/force characteristics (Resusci Anne,

aerdal Medical AS, Stavanger, Norway) for each condition: (1)M crib mattress with backboard; (2) DM crib mattress withackboard; (3) SM crib mattress without backboard; and (4) DMrib mattress without backboard. A clinically used CPR backboard59 cm × 50 cm × 1 cm) was utilized and the backboard/mattressas weighted to ensure that each condition had a simulated patient

orso mass of 12.6 kg. CCs were targeted using real-time feedbacko achieve ≥38 mm internal (sternum to spine) chest depth in the

anikin with a rate of 100 CC min−1 and complete release betweenCs. Our measurement of mattress displacement was performedy accelerometer-based measures. To measure mattress deflec-ion, a surface reference accelerometer was firmly attached to the

ackboard or the internal back plate of the manikin if no back-oard was used.10 Accurate manikin internal (sternum-to-spine)C depth was measured by subtracting the displacement the sur-

ace accelerometer from the displacement (movement) of FDS on

84 (2013) 1131–1136

the chest. This method was previously published to measure themattress displacement during simulated CCs.7 The proportion oftotal FDS motion that was attributed to internal CC was calculatedfor each compression.

2.1.3. Tissue interface pressure image mappingPR qualities of mattresses were assessed by the mattress

industry standard, tissue interface pressure (TIP) image mapping(mmHg).14 Pressure image recordings were collected at Hard Man-ufacturing Co. (Buffalo, NY) using the XSensor® X3 6912 PressureMapping System (SDL, Seven Hills, NSW). Data were collected onthe highest average tissue interface pressure (HAP) of a 3 by 3-sensor group (9 sensors) in a 1.5-in. square area for each of thebody zones (heels, sacrum, and scapula). Using a random numbergenerator, 100 random image frames were selected of each cribmattress recording session and calculated for HAP. Mean pressuresand percentage of recording frames that met pressure reductioncriteria were calculated for each of the body zones and entirebody. Mean HAPs for all body zones were evaluated for PR proper-ties against the target (mean pressure ≤ 50 mmHg) as per researchand industry guidelines.13,15,16 We also evaluated an incidence ofHAP > threshold (50 mmHg) at each condition.

2.2. Statistical analysis

Our primary outcome measure was the difference in mattressdisplacement during CCs between the SM and DM crib mattresses.The mattress displacement and CC data were extracted to an Excelspreadsheet (Microsoft Corp, Redmond, WA). Summary data werereported as mean ± standard deviation (SD). An unpaired t-testwas used to compare the distribution of CCs between SM andDM. The difference between mattresses with and without a back-board was reported as the reduction in mattress displacement. Weprospectively defined a 5 mm reduction in mattress displacementas a minimally clinically significant difference from the existingliterature.3,6,17 p-Values less than 0.05 were considered statisticallysignificant.

3. Results

Four conditions of 50 compressions for a total 200 compressionswere analyzed: (1) SM with backboard; (2) DM with backboard;(3) SM without backboard; and (4) DM without backboard. Meanmanikin CC depth was 45.8 ± 4.3 mm and mean force 42.6 ± 1.7 kg.Fig. 1 illustrates sample waveforms demonstrating total FDS com-pression, mattress deflection and force. Table 1 displays the totalFDS compression depth, measured CC mattress deflection, manikinCC depth, the percentage of manikin CC depth of total FDS depth, aswell as force under various conditions. The difference in means ofmattress deflection measurements during CC (mm) are displayedin Table 2.

3.1. Backboard placement

With a backboard placed between the crib mattress and themanikin, the SM crib mattress had more mattress displacement ofclinical significance (i.e. ≥5 mm) compared to the DM crib mattress(16.5 ± 1.4 mm, p < 0.0001). The percent of total FDS compressionthat was delivered to the manikin chest was 63 ± 1% on the SM cribmattress and 82 ± 1% on the DM crib mattress (difference 20 ± 1%,p < 0.0001).

3.2. Without backboard placement

Without the use of a backboard, the SM crib mattress had evengreater mattress displacement of clinical significance compared to

D.E. Niles et al. / Resuscitation 84 (2013) 1131–1136 1133

Fig. 1. Waveform demonstrating total FDS compression, mattress deflection and force.

Table 1Compression deflection measurements of total FDS, mattress, manikin chest, manikin chest as percentage of total FDS, and force (mean ± SD).

Total FDS (mm) Mattress (mm) Manikin chest (mm) Manikin chest of total FDS (%) Force (kg)

Backboard Single mode 70.6 ± 2.3 26.4 ± 1.0 44.2 ± 2.1 63 ± 1 42.2 ± 1.6Dual mode 55.4 ± 3.6 9.8 ± 0.7 45.6 ± 3.0 82 ± 1 44.4 ± 3.0

No backboard Single mode 87.1 ± 4.0 42.1 ± 1.5 45.0 ± 2.7 52 ± 2 40.4 ± 6.0Dual mode 52.0 ± 2.2 10.3 ± 1.0

Mean

Table 2Difference in means of mattress deflection measurements during CC (mm)(mean ± SD).

Mattress (mm)

Single mode vs.dual mode

Backboard 16.5 ± 1.4*

No backboard 31.7 ± 1.5*

Backboard vs. nobackboard

Single mode 15.6 ± 1.9*

Dual mode 0.7 ± 0.7

≥

tFat

i(rb

calculated to determine the pressure reduction properties of each

TT(

5 mm = clinically significant.* p < 0.0001 for ≥5 mm difference.

he DM crib mattress (31.7 ± 1.5 mm, p < 0.0001). Percent of totalDS compression that was delivered to the manikin chest withoutbackboard was 52 ± 2% on the SM crib mattress and 80 ± 2% on

he DM crib mattress (difference 28 ± 2%, p < 0.0001).In only the SM condition, use of backboard demonstrated a clin-

cally significant decrease (i.e. ≥5 mm) in mattress displacement

15.6 ± 1.9 mm, p < 0.0001). There was not a clinically significanteduction in deflection in the DM crib mattress with the use of aackboard (0.7 ± 0.7 mm, p = 1.0).

able 3issue interface pressure (TIP) mapping pressures (mean ± SD) and percentage of fram≤50 mmHg).

Heels

Mean pressure ± SD(mmHg)

Single mode 25 ± 16Dual mode 40 ± 10

Frames ≤ 50 mmHg(%)

Single mode 98Dual mode 92

41.6 ± 2.0 80 ± 2 43.4 ± 6.7

45.8 ± 4.3 42.6 ± 1.7

3.3. Tissue interface pressure image mapping

Tissue interface pressure image mapping recordings were col-lected on a single male, 4.9 V, 32 kg, 115 cm and lightly clothed.Due to requirement for the subject to remain motionless duringthe recordings, use of a younger subject was not possible. A total of600 random frames (100 frames per body zone per condition) wereanalyzed from the recordings (Table 3).

3.3.1. Mean pressuresThe average of the 100 frames provided the mean pressure for

each body zone and the entire body per condition. Both SM and DMaverage mean pressures were within industry guideline pressuresof ≤50 mmHg.

3.3.2. Pressure reduction property frequencyFor each condition, the frequency of readings ≤50 mmHg was

mattress. Both SM and DM had a high frequency of pressure reduc-tion properties with 99% and 96% of entire body zone readings≤50 mmHg.

es with recording results within “pressure reduction” guidelines for body zones

Sacrum Scapula Entire body

36 ± 2 20 ± 1 27 ± 840 ± 4 18 ± 2 32 ± 13

100 100 9996 100 96

1 itation

4

pIilcmtsoDm

tittbitfopatmiHtsdnaap

bhiaEwfiC“amtpwmplrmtm

iCtofa

134 D.E. Niles et al. / Resusc

. Discussion

This study reports the deflection characteristics of two types ofressure reduction hospital crib mattresses during simulated CC.

n an attempt to satisfy both pressure reduction and CPR qualityssues, the clinical and equipment supply teams at our hospital col-aborated with the mattress manufacturers to design an improvedrib mattress. Our study demonstrated that by utilizing a DM cribattress we were able to maintain sufficient pressure reduction,

hus satisfying the most frequent requirement of long stay admis-ions, and decrease mattress deflection, thereby improving depthf delivered CC. Specifically, we determined that CC delivered onM crib mattresses resulted in less deflection compared to SM cribattresses both with and without a backboard.CPR guidelines recommend that CC be performed with the vic-

im on a hard surface.18,19 New SM crib mattresses prevent skinnjury, but may also absorb the depth and force of CC during CPR,hus increasing the amount of work that a CC provider would haveo provide. In other words, more of the CC provider’s energy woulde absorbed by the SM mattress than the DM mattress thereby

ncreasing the CC provider’s rate of fatigue. Many studies reporthat practitioners are not aware of the deterioration in their per-ormance of CCs and only reported subjective fatigue after 3–4 minf CCs.2,4,5,20 The suggestion from these early studies was that CCrovider fatigue due to the amount of work required to performdequate CCs could be a significant and unrecognized factor con-ributing to poor CC quality. As such, the recommendation has been

ade for CC providers to switch every 2 min in the hopes of min-mizing fatigue and improving the continuity of high quality CCs.owever, this may inadvertently increase detrimental hands-off

ime, which has been shown to adversely affect coronary perfu-ion and contribute to worsening post-resuscitation myocardialysfunction.21 For that reason, increased efforts must be made toot only serve the long-term needs of these critical patients, but tolso ensure their environment (i.e. mattress, location of bed/crib,nd access to head of bed) can facilitate the delivery of the bestossible care in an emergent situation.

Clinical studies have demonstrated a positive associationetween the CC depth and CPR outcome.2–6 In adult out-of-ospital CPR, Kramer-Johansen et al. reported that each 1 mm

ncrement of CC depth was associated with improved hospitaldmission rate.6 In adult in-hospital and out-of-hospital CPR,delson et al. reported that each 5 mm increment of CC depthas associated with improved shock success for ventricularbrillation.3 Hence, even a small incremental improvement inC depth would be clinically significant. Thus, in order to avoidstatistically significant but not clinically important” results, we

priori defined the minimal clinically important difference inattress displacement as ≥5 mm, based on previous clinical inves-

igations (i.e. how much depth loss in the mattress would affectatient outcome). We found that even with a backboard, thereas a clinically significant difference between the SM and DM cribattresses, which was even greater when there was no backboard

laced. Interestingly, without a backboard, the DM crib mattressost only about 2% manikin CC depth of the total compression (aeduction from 82 ± 1% to 80 ± 2%, p < 0.0001) whereas the SM cribattress lost 11% (63 ± 1% to 52 ± 2%, p < 0.0001). We surmise that

he compressions on SM mattress resulted in excessive extraneousattress deflection regardless of backboard use.There have been several studies evaluating the effects of the var-

ous support and mattress surfaces on the effectiveness of CC duringPR. Particularly, an early study by Perkins found that compared to

he most stable surface, the floor, CC depth was significantly lessn all mattresses and were not guideline depth-compliant.22 Theyound that the 10–20% reduction in CC depth between the floornd mattresses reported would seem likely to cause a reduction

84 (2013) 1131–1136

in the cardiac output associated with CPR, which could potentiallyhave an adverse effect on outcome. Most recently, Perkins and histeam found that when utilizing automated CPR feedback/promptdevices, under-compression occurred due to the failure for thedevice to compensate for the underlying mattress deflection, whichrepresented 35–40% of total compression depth.8 Furthermore, ina previous report from our group of in-hospital real pediatric CPRevents, we found that with realistic forensic engineering recon-struction, deflection of the mattress contributed approximately28% of measured CC depth on ICU beds and 10% of measured CCdepth on stretchers with backboards in place, resulting in over-estimation of CC depth based on the sternal placement of anFDS during real CC.7 For this study, we used a previously pub-lished technique to determine the amount of mattress deflection oftwo crib mattresses during simulated CPR on manikins. We foundthat during high quality CC with a backboard, the SM crib mat-tress had nearly threefold the mattress displacement comparedto the newly designed DM crib mattress. In addition, when notusing a backboard under the manikin, mattress displacement wasover 4 times greater in the SM compared to DM. These differ-ences may lead CC providers to underestimate delivered CC depth,due to the overall movement of the patient on the SM mattress,which many result in ineffective CCs. Importantly, the increasedmovement of the patient on the SM mattress during CCs may com-promise quality, security and safety of other critical resuscitationinterventions such as endotracheal tubes and intravenous/arterialcatheters.

The development of DM mattresses and introduction to the hos-pital industry has been a recent one. We found that with thesemattresses, the placement of a backboard during CPR did notimprove compression depth of the chest significantly. Well knownto resuscitation teams, finding and placing a backboard duringCPR distracts from the critical task of providing adequate CC, mayincrease detrimental hands-off time, and introduces the danger ofdislodging vital tubes and indwelling lines. The use of a DM-typemattress exclusively in units such as the ICU could obviate the needfor the backboard and the delays that come with its use during CPR.However, at this time, not all hospitals have critical care units witha DM-type mattress as standard equipment. Thus, backboard place-ment should continue to be standard of care in order to mitigatethe risk of mattress deflection during CPR, when soft mattresses aredeployed in a hospital setting.

Many hospitals have implemented the use of SM PR mattressesto reduce the complications of skin injury and infection during thecourse of a hospital admission, particularly for long stay, critically illpatients.23–27 Since critically ill patients often experience low car-diac output, the body compensates by shunting blood away fromnonvital organs; specifically, the skin. This vasoconstriction mayjeopardize a patient’s response to local compression ischemia.28

However, the patient’s hemodynamic instability is often cited asthe primary reason why intensive care nurses do not always fol-low unit-based protocols to reposition patients every 2 h.29,30 Inan effort to reduce incidence of pressure ulcers response, mattressmanufactures have designed improved pressure relieving supportsurfaces, which have been widely implemented by hospitals,25,31

particularly in the acute care setting where many patients are lessmobile. One study has shown that implementing such changesresulted in a decrease in the quarterly hospital-acquired pressureulcer prevalence rates from a high of 29% to near 0%.32

Literature from the mattress industry states that to be labeledas providing pressure relief, a mattress has to have the pressureof 32 mmHg or lower, whereas pressure reduction performance is

to occur between 32 mmHg and 50 mmHg.13 In this study, weselected ≤50 mmHg as the cutoff for pressure reduction based onindustry guidelines and previous studies.15,16 In our study, tissueinterface pressure image mapping analysis demonstrated that both

itation

mrcy

ctwiatmrmPMdoaoC

5

c24l<wt(tpfaCacawfwaos

6

prpattCthcTisb

1

1

1

11

1

1

1

1

D.E. Niles et al. / Resusc

attresses were consistent with industry guidelines for a pressure-educing surface. Importantly, since implementing these new DMribs mattresses in our institution’s pediatric intensive care unit 2ears ago, there has been NO increase in pressure ulcer cases.

This study was designed to reach higher CC target depths,onsistent with American Heart Association (AHA) and Interna-ional Liaison Committee on Resuscitation (ILCOR) 2005 Guidelines,hich necessitated the use of a manikin that could reach an

nternal CC depth of ≥38 mm. There are no manikins that were aware of that characterize the deflection/force relationship ofhose under the age of 8 years. Consequently the use of an adult

anikin with realistic CC properties which was able to reliablyeach such depth was required (Resusci Anne, PRO-RP01-2575:ax CC depth 58–62 mm, force 51–58 kg versus Resusci Junior,

RO-RP01-2936: max CC depth 42–43 mm, force 30–33 kg; Laerdaledical, Stavanger, Norway). We achieved a mean manikin CC

epth for all conditions (200 CCs) of 45.8 ± 4.3 mm with mean forcef 42.6 ± 1.7 kg. For the purpose of comparison, our recent studynalyzing actual CC depth in real pre-pubertal patients (8–14 yearsld) demonstrated that during real cardiac arrests, the mean actualC depth was 36.2 ± 9.6 mm with a mean CC force of 30.7 ± 7.6 kg.33

. Limitations

There are several limitations to this study. Experiments wereompleted at a target depth of ≥38 mm, based on AHA and ILCOR005 guidelines. Our mean manikin CC depth during this study was5.8 ± 4.3 mm which lies between the current 2010 CPR Guide-

ine depths of ≥50 mm for those ≥1 year and ≥40 mm for those1 year. Yet, mattress deflection at each of those target depthsas not assessed. In order to reach depths ≥38 mm with realis-

ic CC force/deflection curves, we utilized the adult Resusci AnneLaerdal Medical AS, Stavanger, Norway) simulation manikin sincehere are no manikins with force/deflection characteristics of thisediatric age group available. Therefore, we are not certain that theorce/deflection relationship results from this study are generaliz-ble to the younger pediatric population <8 years. In addition, theC provider was coached to the depth displayed by the FDS and notmanikin internal potentiometer that may have led to shallower

ompressions in SM mattress group. Furthermore, we only evalu-ted two types of crib mattresses and variation in weight conditionas not assessed. Finally, qualitative evaluation of the tissue inter-

ace pressure image mapping was conducted with a healthy subjectho may not adequately represent the physical characteristics ofcritically ill child. In addition, this approach provides informationn surface pressure only, revealing nothing about tissue compres-ion, perfusion or distortion.

. Conclusions

Chest compressions delivered on dual mode (provides bothressure reduction and CPR support when needed) crib mattressesesulted in less mattress deflection compared to single mode (onlyrovide pressure reduction) crib mattresses. Forensic engineeringnalysis of crib mattress properties suggests characteristics of mat-resses impact pediatric CPR depth. Importantly, our data showhat it is possible to balance skin pressure reduction qualities withPR effectiveness. Mattress deflection testing should be performedo assure safety and effectiveness of CPR on such mattresses asospitals move to pressure reduction mattress design. Backboardsontinue to be important for CPR when performed on soft surfaces.

hus, we recommend that a backboard be used in order to mit-gate mattress deflection during CPR. The impact of crib featuresuch as mattress height, CPR mode, and backboard design shoulde evaluated for their effect on CPR performance.

1

84 (2013) 1131–1136 1135

Conflict of interest statement

The authors acknowledge the following potential conflicts ofinterest: Dana Niles, Akira, Nishisaki, Matthew R. Maltese and VinayNadkarni receive unrestricted research grant support from theLaerdal Foundation for Acute Care Medicine. Finally, Hard Manu-facturing Co. nor Otis Bed Manufacturing Co. had control or accessto the data, and did not have censor authority on the data analysisor reporting.

Acknowledgments

Funding provided by Laerdal Foundation for Acute CareMedicine and the Endowed Chair of Critical Care Medicine at theChildren’s Hospital of Philadelphia. Special thanks to Stephanie Tut-tle MS, MBA, Lori Boyle RN, Eileen Nelson RN, Jamie Sklar RN, TerriWentzell, Lori Greco, John Roma of Otis Bed Manufacturing, HardManufacturing, and the Pediatric ICU staff at The Children’s Hospi-tal of Philadelphia for their diligence, support, and contributions tothis study.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.resuscitation.2013.01.033.

References

1. Slonim AD, Patel KM, Ruttimann UE, Pollack MM. Cardiopulmonary resuscitationin pediatric intensive care units. Crit Care Med 1997;25:1951–5.

2. Wik L, Kramer-Johansen J, Myklebust H, et al. Quality of cardiopulmonary resus-citation during out-of-hospital cardiac arrest. JAMA 2005;293:299–304.

3. Edelson DP, Abella BS, Kramer-Johansen J, et al. Effects of compression depthand pre-shock pauses predict defibrillation failure during cardiac arrest. Resus-citation 2006;71:137–45.

4. Abella BS, Sandbo N, Vassilatos P, et al. Chest compression rates duringcardiopulmonary resuscitation are suboptimal: a prospective study during in-hospital cardiac arrest. Circulation 2005;111:428–34.

5. Abella BS, Alvarado JP, Myklebust H, et al. Quality of cardiopulmonary resusci-tation during in-hospital cardiac arrest. JAMA 2005;293:305–10.

6. Kramer-Johansen J, Myklebust H, Wik L, et al. Quality of out-of-hospital car-diopulmonary resuscitation with real time automated feedback: a prospectiveinterventional study. Resuscitation 2006;71:283–92.

7. Nishisaki A, Nysaether J, Sutton R, et al. Effect of mattress deflection on CPR qual-ity assessment for older children and adolescents. Resuscitation 2009;80:540–5.

8. Perkins GD, Kocierz L, Smith SC, McCulloch RA, Davies RP. Compression feed-back devices over estimate chest compression depth when performed on a bed.Resuscitation 2009;80:79–82.

9. Oh J, Song Y, Kang B, et al. The use of dual accelerometers improvesmeasurement of chest compression depth. Resuscitation 2012;83:500–4,http://dx.doi.org/10.1016/j.resuscitation.2011.09.028.

0. Nishisaki A, Maltese MR, Niles DE, et al. Backboards are important when chestcompressions are provided on a soft mattress. Resuscitation 2012;83:1013–20.

1. Aase SO, Myklebust H. Compression depth estimation for CPR quality assessmentusing DSP on accelerometer signals. IEEE Trans Biomed Eng 2002;49:263–8.

2. Vandenkerkhof EG, Friedberg E, Harrison MB. Prevalence and risk of pressureulcers in acute care following implementation of practice guidelines: annualpressure ulcer prevalence census 1994–2008. J Health Care Qual 2011;33:58–67,http://dx.doi.org/10.1111/j.1945-1474.2011.00127.x.

3. In-Touch.Polyurethane foam association I, vol. 1, June 2003 ed. 2003.4. Bain D, Ferguson-Pell M, McLeod A. Evaluation of mattresses using interface

pressure mapping. J Wound Care 2003;12:231–5.5. Braden B, Bergstrom N. A conceptual schema for the study of the etiology of

pressure sores. Rehabil Nurs 1987;12:8–12.6. Bergstrom N, Braden BJ, Laguzza A, Holman V. The braden scale for predicting

pressure sore risk. Nurs Res 1987;36:205–10.7. Sutton RMNA, Niles DE, Leffelman JL, et al. Deep chest compressions ≥50 mm

improve hemodynamic outcomes during actual pediatric and adolescent arrest.Crit Care Med 2010 2010;38:453.

8. Berg MD, Schexnayder SM, Chameides L, et al. Part 13: pediatric basic life sup-

port: 2010 American Heart Association Guidelines for cardiopulmonary resus-citation and emergency cardiovascular care. Circulation 2010;122:S862–75.

9. Berg RA, Hemphill R, Abella BS, et al. Part 5: adult basic life support: 2010American Heart Association Guidelines for cardiopulmonary resuscitation andemergency cardiovascular care. Circulation 2010;122:S685–705.

1 itation

2

2

2

2

2

2

2

2

2

2

3

3

3

136 D.E. Niles et al. / Resusc

0. Hamrick JT, Fisher B, Quinto KB, Foley J. Quality of external closed-chestcompressions in a tertiary pediatric setting: missing the mark. Resuscitation2010;81:718–23.

1. Berg RA, Sanders AB, Kern KB, et al. Adverse hemodynamic effects of interruptingchest compressions for rescue breathing during cardiopulmonary resuscitationfor ventricular fibrillation cardiac arrest. Circulation 2001;104:2465–70.

2. Perkins GD, Benny R, Giles S, Gao F, Tweed MJ. Do different mattresses affectthe quality of cardiopulmonary resuscitation? Intensive Care Med 2003;29:2330–5.

3. Curley MA, Razmus IS, Roberts KE, Wypij D. Predicting pressure ulcer risk inpediatric patients: the Braden Q Scale. Nurs Res 2003;52:22–33.

4. Schindler CA, Mikhailov TA, Kuhn EM, et al. Protecting fragile skin: nursing inter-ventions to decrease development of pressure ulcers in pediatric intensive care.Am J Crit Care 2011;20:26–34, quiz 5.

5. de Laat EH, Pickkers P, Schoonhoven L, Verbeek AL, Feuth T, van Achterberg T.Guideline implementation results in a decrease of pressure ulcer incidence incritically ill patients. Crit Care Med 2007;35:815–20.

6. Pressure ulcer guidelines on performing prevalence and incidence studies pre-sented. Dermatol Nurs 2009;21:219.

3

84 (2013) 1131–1136

7. Shahin ES, Dassen T, Halfens RJ. Pressure ulcer prevention in intensive carepatients: guidelines and practice. J Eval Clin Pract 2009;15:370–4.

8. Maklebust J. Pressure ulcers: etiology and prevention. Nurs Clin North Am1987;22:359–77.

9. Batson S, Adam S, Hall G, Quirke S. The development of a pressure area scor-ing system for critically ill patients: a pilot study. Intensive Crit Care Nurs1993;9:146–51.

0. Neidig JR, Kleiber C, Oppliger RA. Risk factors associated with pressure ulcersin the pediatric patient following open-heart surgery. Prog Cardiovasc Nurs1989;4:99–106.

1. Russell LJ, Reynolds TM, Park C, et al. Randomized clinical trial comparing 2support surfaces: results of the prevention of pressure ulcers study. Adv SkinWound Care 2003;16:317–27.

2. Hiser B, Rochette J, Philbin S, Lowerhouse N, Terburgh C, Pietsch C. Implementing

a pressure ulcer prevention program and enhancing the role of the CWOCN:impact on outcomes. Ostomy Wound Manage 2006;52:48–59.

3. Niles DE, Nishisaki A, Sutton RM, et al. Comparison of relative and actual chestcompression depths during cardiac arrest in children, adolescents, and youngadults. Resuscitation 2012;83:320–6.