S

Wd

JP

a

ARRA

KDSM

1

dGsireodctt

e

i

m

0h

Resuscitation 83 (2012) 1467– 1472

Contents lists available at SciVerse ScienceDirect

Resuscitation

jo u rn al hom epage : www.elsev ier .com/ locate / resusc i ta t ion

imulation and education

ill medical examination gloves protect rescuers from defibrillation voltagesuring hands-on defibrillation?�

oseph L. Sullivan ∗, Fred W. Chapmanhysio-Control, Inc., Redmond, WA, United States

r t i c l e i n f o

rticle history:eceived 2 May 2012eceived in revised form 27 June 2012ccepted 21 July 2012

eywords:efibrillationafetyedical examination gloves

a b s t r a c t

Background: Continuing compressions during a defibrillation shock has been proposed as a method ofreducing pauses in cardiopulmonary resuscitation (CPR) but the safety of this procedure is unproven.The medical examination gloves worn by rescuers play an important role in protecting the rescuer yetthe electrical characteristics of these gloves are unknown. This study examined the response of medicalexamination gloves to defibrillation voltages.Methods: Part 1 of this study measured voltage–current curves for a small sample (8) of gloves. Part2 tested more gloves (460) to determine the voltage required to produce a specific amount of currentflow. Gloves were tested at two current levels: 0.1 mA and 10 mA. Testing included four glove materials(chloroprene, latex, nitrile, and vinyl) in a single layer and double-gloved.Results: All gloves tested in part 1 allowed little current to flow (<1 mA) as the voltage was increased untilbreakdown occurred, at which point current flow increased precipitously. In part 2, 118 of 260 (45%)

single gloves and 93 of 120 (77%) double gloves allowed at least 0.1 mA of current flow at voltages withinthe external defibrillation voltage range. Also, 6 of 80 (7.5%) single gloves and 5 of 80 (6.2%) double glovesallowed over 10 mA.Conclusions: Few of the gloves tested limited the current to levels proven to be safe. A lack of sensationduring hands-on defibrillation does not guarantee that a safety margin exists. As such, we encouragerescuers to minimize rather than eliminate the pause in compressions for defibrillation.

. Introduction

Pauses in chest compressions during CPR are known to beetrimental to survival.1 The 2005 American Heart Associationuidelines eliminated some reasons for pauses,2 but interruptionstill occur to allow intubation,3 ventilations,4 AED analyses, charg-ng, and defibrillation shocks.5 While shock delivery accounts for aelatively minor fraction of the total hands-off time in most cases,limination of the shock pause is consistent with the overall goalf minimizing hands-off time. In addition, pausing CPR for shockelivery complicates the resuscitation protocol, requiring closeoordination between multiple members of the resuscitation teamo avoid inadvertent rescuer shocks while minimizing hands-off

ime.

Recently, Lloyd et al. suggested that it may be possible for rescu-rs to continue chest compressions during a defibrillation shock.6

� A Spanish translated version of the summary of this article appears as Appendixn the final online version at http://dx.doi.org/10.1016/j.resuscitation.2012.07.031.∗ Corresponding author at: Physio-Control, Inc., 11811 Willows Road NE, Red-ond, WA 98052, United States. Tel.: +1 425 867 4405.

E-mail address: joseph.l.sullivan@physio-control.com (J.L. Sullivan).

300-9572/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.resuscitation.2012.07.031

© 2012 Elsevier Ireland Ltd. All rights reserved.

They studied the safety of “hands-on defibrillation” by simulatingchest compressions while delivering external cardioversion shocksto patients with atrial fibrillation. No harm came to the “rescuers,”and although the authors cautioned against extrapolation beyondthe limits of their model, they concluded that, “Uninterrupted man-ual chest compressions during shock delivery are feasible.”

Although defibrillator shock intensity is commonly expressedin terms of joules, it is the voltage that determines the risk of anunintended operator shock. For common biphasic external defib-rillators, the voltage applied to the patient for the first shock rangesfrom about 1300 to 1800 V.13 Maximum energy biphasic shocks canbe as high as 2700 V.7 Due to the large magnitude of these voltagescaution is wise. The best approach for providing optimal patientcare while minimizing the rescuer hazards is unclear.

A number of factors contribute to the risk of hands-on defibril-lation. Two key factors are, (1) the fraction of the shock voltagepresented across the rescuer and (2) the voltage standoff capabilityof the gloves. Neither of these factors is known.

The fraction of the shock voltage presented across the rescuer is

dependent on the electrical circuit formed by the patient and therescuer. This is illustrated in Fig. 1.

The possible current paths through a rescuer have not been wellstudied but one model has been proposed that includes current flow

1468 J.L. Sullivan, F.W. Chapman / Resuscitation 83 (2012) 1467– 1472

F ck. Thg ry sk

tai

rdudaoarcpr

ew

2

2

ctots

aaawaurr

ig. 1. At least two points of contact with the patient are required for a rescuer shoround or inadvertent patient contact. High-voltage current is known to puncture d

hrough the rescuer’s chest.17 In this model the gloves can serve as barrier to current flow through the rescuer and may provide anmportant safety mechanism.

In general, medical examination gloves are designed as a bar-ier to bodily fluids, not electricity, and manufacturers typicallyo not provide electrical specifications for their gloves. One studysing the AC (alternating current) voltage from an electrocauteryevice found latex and neoprene glove breakdown voltages as lows 2000 V,8 but it is unclear how these results relate to hands-n defibrillation. Little other information is available that wouldllow the safety of hands-on defibrillation to be assessed. A recenteview paper by Petley et al. suggested, “Further work needs to bearried out to determine the necessary electrical and mechanicalroperties required of an insulator suitable to be used to protectescuers.”9

The purpose of this study was to compare the DC (direct current)lectrical characteristics of common medical examination glovesith the output voltages of biphasic defibrillators.

. Methods

.1. Part 1

Part one of this study examined the shape of the voltage–currenturve for several types of gloves by applying a range of DC voltageso each glove and recording the current flow at each level. The goalf this part of the study was to determine if the current flow inhese gloves typically exhibits a “breakdown” characteristic and, ifo, to determine the voltage level at which it occurs.

A dielectric analyzer (Vitrek 944i, San Diego, CA) was used topply the voltage to the gloves and monitor current flow. The volt-ge across the glove was manually increased and allowed to dwellt each level long enough for the current to stabilize. Current valuesere recorded at 100 V, 200 V, 500 V, and 500 V increments there-

fter until the current exceeded the device maximum (20 mA) orntil a breakdown occurred. At breakdown the dielectric analyzerapidly removes the voltage to prevent excessive current flow and,ather than reporting the actual current flow, simply reports that

e gloves form one point of contact. The second point of contact may be through thein, penetrate clothing, and jump across a small air gap.23,24

a breakdown was detected. All measurements in this part of thestudy were made by inserting a roughly palm-sized metal electrodeinto the glove, placing the unstretched glove on a metal sheet, andapplying the voltage between the electrode and the sheet. Minimalcompressive force (<0.5 pounds) was applied for each measure-ment.

Gloves made of four commonly used polymers were tested:

• latex (Safe-Touch powder-free latex examination gloves,Dynarex Corporation, Orangeburg, NY; and Ambidex Powderfree cleanroom latex gloves, Ambidex Corporation, Thailand),

• nitrile (Esteem Stretchy Nitrile power-free nitrile exam gloves,CardinalHealth, McGaw Park, IL; and Safe-Touch power-freenitrile examination gloves, Dynarex Corporation, Orangeburg,NY),

• chloroprene (NeoPro powder-free chloroprene examinationgloves, Microflex Corporation, Reno, NV), and

• vinyl (Safe-Touch powder-free vinyl examination gloves,Dynarex Corporation, Orangeburg, NY; and Disposable vinylgloves powdered, Magid Corporation, China).

Part 1 of the study tested two gloves of each polymer.

2.2. Part 2

Part 2 measured the voltage necessary to produce a specificamount of current flow in each glove. This was done by program-ming the dielectric analyzer to increase the voltage until a pre-setcurrent limit was reached. The voltage required to produce thatcurrent level was then recorded. If the desired current flow wasnot achieved before reaching the maximum output voltage of thedielectric analyzer (5000 V), then 5000 V was recorded.

The tests in Part 2 were performed using one of two currentlimits. The first current limit was 0.1 mA. This is the maximum

direct current (DC) allowed to flow incidentally through an oper-ator from a medical device with no faults as established by thewidely recognized standards agency, the International Electrotech-nical Commission (IEC).10 A current flow less than this amount is

J.L. Sullivan, F.W. Chapman / Resuscitation 83 (2012) 1467– 1472 1469

40003500300025002000150010005000

Cur

rent

Flo

w (m

A)

Applied Voltage (Volts)

Nitrile

Nitrile

Chloroprene

Chloroprene

Vinyl

Vinyl

Latex

Latex

Breakdown Voltage

1mA

F n of aT of illu

gtv

tactwm

tgecs

1dlco

3

3

vcbaaf

3

4etl3

ig. 2. Measured current flow (mA) through the gloves tested in Part 1 as a functiohe current flow at breakdown is represented here at about 10 mA for the purpose

enerally considered to be safe. For this study the voltage requiredo produce this amount of current flow is referred to as the “leakageoltage.”

The second current limit was set at 10 mA. This is more currenthan is allowed by any regulatory standards of which we are awarend is also above the threshold of perception (about 1 mA).11 Thisurrent limit was also chosen because Part 1 of our study showedhat 10 mA is above the knee of the curve for all glove types. In otherords, it represents a complete electrical breakdown of the gloveaterial.While Part 1 measurements were all made on single gloves,

he measurements in Part 2 were made on both single and doubleloves. The double-glove measurements were made by placing thelectrode on top of the glove (rather than inside the glove) so thaturrent flowed through two layers of material to reach the metalheet.

The testing in Part 2 involved all four polymers listed in Part. Each polymer was tested at both current limits in single andouble-layer configurations, giving a total of 16 polymer–current-

ayer combinations. At least 20 measurements were made of eachombination, resulting in a total of 460 measurements in Part 2 ofur study.

. Results

.1. Part 1

All gloves exhibited a highly nonlinear “hockey-stick” shapedoltage–current relationship that is characteristic of an electri-al breakdown (Fig. 2). The current flow for every glove stayedelow about 1 mA until breakdown, at which point current flowbruptly jumped up and the dielectric analyzer removed the volt-ge. The breakdown voltage varied greatly among gloves, rangingrom 2500 V to 4000 V.

.2. Part 2

Some single gloves (32/80 = 40% chloroprene, 2/60 = 3% latex,2/60 = 70% nitrile, and 42/60 = 70% vinyl) of each polymer showed

ither a leakage voltage or a breakdown voltage (or both) withinhe output voltage range of biphasic defibrillators (Fig. 3). Theeakage voltage of most double gloves (18/20 = 90% chloroprene,5/40 = 87% nitrile, 40/40 = 100% vinyl gloves, and 0/20 = 0% latex)

pplied voltage. The 1 mA line represents the approximate threshold of perception.stration.

also fell within this range. The breakdown voltage for all (20each) double latex, double nitrile, and double vinyl gloves testedincreased to a level higher than the highest biphasic defibrillatoroutput voltage.

4. Discussion

The main findings of this study are that current flow throughmedical examination gloves is highly non-linear with respect tovoltage, and that the current flow is very inconsistent among glovetypes and between samples of a single glove type. This makes therisks of hands-on defibrillation difficult to predict (Fig. 4).

This study was the first to measure voltage–current curves formedical examination gloves. These data are relevant because theyillustrate the difficulty of assessing the safety of hands-on defibril-lation. During this procedure the rescuer may be relying partially orcompletely on the gloves to avoid an electrical shock. In our studythe current flow for the gloves tested in Part 1 stayed low – belowthe threshold of perception – until an arc developed and the cur-rent precipitously increased. Importantly, this demonstrates thatrescuers should not construe the absence of sensation as an indica-tion of a safety margin. It is likely that hands-on defibrillation withmedical examination gloves will produce no sensation at all unlessthe gloves completely break down, at which time the current willbe limited by factors other than the gloves. In our study the currentwas limited by the dielectric analyzer but current flow through arescuer could be far higher.

Part 2 of this study measured leakage and breakdown voltages.One strategy for ensuring the safety of hands-on defibrillation isto use only gloves that are guaranteed to have a leakage voltagegreater than the defibrillator shock voltage. With biphasic defibril-lators this appears to be possible with double latex gloves. Noneof the double latex gloves exceeded the leakage voltage thresholdat less than 5000 V, giving these sample gloves at least a two-to-one voltage margin over most biphasic defibrillation voltages.All other polymers showed a distribution of leakage voltages thateither straddles or is lower than some of biphasic shock voltages,even with gloves doubled up.

However, the risk to the operator from leakage current is prob-ably small, even in the worst case, one in which a current pathincludes the heart. The 0.1 mA leakage current limit set by the IECwas selected to ensure that 1–3 s of current flow would induce

1470 J.L. Sullivan, F.W. Chapman / Resuscitation 83 (2012) 1467– 1472

F r singld er, ran

Vmltpifprabgdi

trfpc

ig. 3. Histograms of leakage (hatched bars) and breakdown (clear bars) voltages foefibrillation shock voltages are shown. Monophasic shock voltages are much high

F in less than 5% of the population.12 Defibrillation shocks areuch shorter, typically less than 20 ms.13 In order for a defibril-

ation pulse to induce fibrillation the shock must be timed to hithe vulnerable period of the cardiac cycle. VF may be triggered ineople with heart disease if they receive a well-timed low-energy

nternal defibrillation shock (≤0.2 J)14 or pacing pulse (0.5 V)15; butor healthy adults it is estimated that, even if the defibrillationulse is appropriately timed, as much as 500 mA of current flow isequired to induce VF.16 Petley et al. propose a limit of 1 mA through

rescuer9, and Hoke et al. assert that, “currents up to 200 mA shoulde regarded as safe.”17 If this is true, then any voltage less than thelove breakdown voltage would not present a hazard. By this stan-ard, the double nitrile, double vinyl, and double latex gloves tested

n our study appear safe.Although the risk from leakage current is debatable, the poten-

ial harm from a glove breakdown is another matter. Hoke et al.

eport that healthy non-rescuers have suffered serious injuriesrom accidental or erroneous defibrillator shocks.17 All of theseeople apparently placed the defibrillator paddles directly on theirhest, head, or other body part, delivered a shock, and suffered

e and double gloves. Median values are given for each group. For reference, biphasicging up to 5000 V.

burns, arrhythmias, or death. This situation is somewhat differentfrom a rescuer shock from hands-on defibrillation because duringresuscitation the main current path is believed to be through thepatient, with a smaller, secondary path through the rescuer. Thereare no data available that allow the amount of current on any of thepossible secondary paths to be quantified.

Hoke et al. report that, although minor injuries are not rare, themedical literature contains no examples of a “life-threatening con-dition or long-term disability” suffered by a rescuer delivering anexternal defibrillation shock to a patient.17 This is a remarkablerecord considering that there are approximately 60,000 VF car-diac arrests in the United States annually.18 It should be pointedout, though, that these rescuers were trained to stand clear dur-ing a shock. It is not known how many of those rescuers may havebeen injured had they been trained to continue compressions dur-ing defibrillation. Even a relatively small glove failure rate could

translate into a large number of rescuer shocks.

It is also possible that a rescuer could be harmed in other waysthan immediate induction of VF. There are reports of arrhyth-mias, conduction disturbances, elevated cardiac enzymes, coronary

J.L. Sullivan, F.W. Chapman / Resusc

Glove Breakdown Vol tage

Glove Leakage Vol tage

Applied Voltage > Glove Breakdown

Voltage

Glove Breakdown Voltage > Applied Voltage > Glove

Leakage Vol tage

Glove Leakage Vol tage > Applied Voltage

KnownSafe

SafetyUncertain

Dangerous

Low Voltage

HighVoltage

Fg

oslctt

drfceclnote

samhe

catti2a

fc

ig. 4. The safety of hands-on defibrillation depends on the voltage applied to theloves, the glove leakage voltage, and the glove breakdown voltage.

cclusive thrombosis, and myocardial infarction due to coronarypasm as a result of an electrical shock from either an AC powerine or a lightning strike.19 Unfortunately, the level and duration ofurrent flow is not known in any of those cases so it is impossibleo say whether an accidental defibrillation shock can cause any ofhose effects.

Besides shocks from AC power lines, lightning, and externalefibrillators, there is one report in the literature of a nitrile-glovedescuer suffering permanent nerve damage as a result of a shockrom an implanted defibrillator.20 This report is of particular con-ern because internal defibrillator voltages are much lower thanxternal defibrillator voltages. The defibrillator, a Medtronic Con-erto C174AWK, outputs a maximum peak voltage of 800 V21 –ower than even the first shock voltage of most common exter-al defibrillators.13 The risk of rescuer injury would be even moref a concern when treating patients with subcutaneous defibrilla-ors which likely possess a higher output voltage and may utilizelectrodes closer to the surface.

There is also a chance that gloves may contain a pinhole ormall tear. Although gloves may be punctured by the wearer, someuthors have suggested that as many as 15% of gloves may beanufactured with defects.8 We did not notice any defects in the

undreds of gloves that we tested, but our experiment was set up toxamine the properties of glove materials, not to look for pinholes.

While it is clear that hazards exist for rescuers performingompressions during defibrillation, the potential patient benefitppears to be small. It should be possible for well trained rescuerso charge the defibrillator while performing chest compressions,ake their hands off, shock, and resume compressions with a totalnterruption of perhaps four seconds.22 If a shock is delivered every

min CPR period, then these four second interruptions would only

dd 3.3% to the total hands-off time.

Our results indicate a need for further study of the other keyactor not addressed in this study: the voltage to which the res-uer is exposed. Had we found that glove leakage and breakdown

itation 83 (2012) 1467– 1472 1471

voltages were consistently higher than the shock voltages, then wecould have concluded that hands-on defibrillation is safe. Becausewe instead found that many gloves allow current flow and somebreak down completely when faced with defibrillation voltages, thesafety of this situation is much less clear. Without additional infor-mation about the voltages presented to rescuers in resuscitationscenarios, continuing compressions during defibrillation cannot beadvised.

5. Limitations

This study was limited primarily because it characterized onlyone factor relating to the safety of hands-on defibrillation: theelectrical characteristics of medical examination gloves. The otherfactor – the fraction of the shock voltage presented to the rescuer –is equally important. Currently, there are no data available on thatfactor.

In order for current to flow through a rescuer there must bea current path. This study is not predicated on a particular currentpath but merely postulates that one or more paths may exist. In theinterests of safety, we believe this assumption is prudent unless itis proven otherwise.

For the sake of simplicity the gloves in our study were not com-pressed as they would be during chest compressions nor were theystretched as they would be when worn by a person. These factorswould likely reduce the leakage and breakdown voltages for thegloves, increasing the risk to the wearer.

We applied voltages to these gloves relatively slowly comparedto the rise time of a defibrillation shock. It is not known whetherthe rate of voltage application influences the leakage or breakdownvoltages.

6. Conclusions

Few of the gloves tested limited the current to levels proven tobe safe. A lack of sensation during hands-on defibrillation does notguarantee that a safety margin exists. As such, we encourage rescu-ers to minimize rather than eliminate the pause in compressionsfor defibrillation.

Conflict of interest statement

The authors are both employees of Physio-Control, Inc., a man-ufacturer of external defibrillators.

References

1. Christenson J, Andrusiek D, Everson-Stewart S, et al. Resuscitation Out-comes Consortium Investigators. Chest compression fraction determinessurvival in patients with out-of-hospital ventricular fibrillation. Circulation2009;120:1241–7.

2. American Heart Association. 2005 American Heart Association guidelines forcardiopulmonary resuscitation and emergency cardiovascular care. Circulation2005;112:IV12–8.

3. Wang HE, Simeone SJ, Weaver MD, Callaway CW. Interruptions in cardiopul-monary resuscitation from paramedic endotracheal intubation. Ann Emerg Med2009;54:645–52.

4. Odegaard S, Pillgram M, Berg NE, Olasveengen T, Kramer-Johansen J. Time usefor ventilation in two-rescuer CPR with a bag-valve-mask device during out-of-hospital cardiac arrest. Resuscitation 2008;77:57–62.

5. Sullivan JL, Walker RG, Chapman FW. How much can hands-off time be reducedby performing rhythm analysis during CPR? Circulation 2009;120:S1479.

6. Lloyd MS, Heeke B, Walter PF, Langberg JJ. Hands-on defibrillation: an analysisof electrical current flow through rescuers in direct contact with patients duringbiphasic external defibrillation. Circulation 2008;117:2510–4.

7. Zoll R series ALS operator’s guide, rev C. Chelmsford, MA: Zoll Medical Corpora-

tion; January, 2010. p. A-13.

8. Tucker RD, Ferguson S. Do surgical gloves protect staff during electrosurgicalprocedures? Surgery 1991;110:892–5.

9. Petley GW, Cotton AM, Deakin CD. Hands-on defibrillation: theoretical and prac-tical aspects of patient and rescuer safety. Resuscitation 2012;83:551–6.

1 esusc

1

1

1

1

1

1

1

1

1

1

2

22

23. Gibbs W, Eisenberg M, Damon SK. Dangers of defibrillation: injuries toemergency personnel during patient resuscitation. Am J Emerg Med 1990;8:

472 J.L. Sullivan, F.W. Chapman / R

0. International Electrotechnical Commission. Medical electrical equipment—Part1: General requirements for basic safety and essential performance, clause 8.7Leakage current and patient auxiliary currents. IEC 60601-1 ED 3.0 B; 2005. p.49–64.

1. Olson WH. Medical instrumentation application and design. Chapter 13: Elec-trical safety. Boston, MA: Houghton Mifflin Company; 2000. p. 668.

2. International Electrotechnical Commission. Medical electrical equipment—Part1: General requirements for basic safety and essential performance, clause 8.7Leakage current and patient auxiliary currents. IEC 60601-1 ED 3.0 B; 2005. p.182–186.

3. Achleitner U, Rheinberger K, Furtner B, Amann A, Baubin M. Waveform analysisof biphasic external defibrillators. Resuscitation 2001;50:61–70.

4. Hou CJ, Chang-Sing P, Flynn E, et al. Determination of ventricular vulner-able period and ventricular fibrillation threshold by use of T-wave shocksin patients undergoing implantation of cardioverter/defibrillators. Circulation1995;92:2558–64.

5. Day GA, Padanilam BJ, Fogel RI, et al. Pacing threshold testing induced ventri-

cular fibrillation following acute rate control of atrial fibrillation. J CardiovascElectrophysiol 2009;20:1405–7.

6. International Electrotechnical Commission. Basic safety publication. Effects ofcurrent on human beings and livestock—Part 1: General aspects. IEC479-1 ED3.0; 1994. p. 49–53.

2

itation 83 (2012) 1467– 1472

7. Hoke RS, Heinroth K, Trappe HJ, Werdan K. Is external defibrillation an electricthreat for bystanders? Resuscitation 2009;80:395–401.

8. Rea TD, Eisenberg MS, Sinibaldi G, White RD. Incidence of EMS-treatedout-of-hospital cardiac arrest in the United States. Resuscitation 2004;63:17–24.

9. Fineschi V, Di Donato S, Mondillo S, Turillazzi E. Electric shock: cardiac effectsrelative to non fatal injuries and post-mortem findings in fatal cases. Int J Cardiol2006;111:6–11.

0. Stockwell B, Bellis G, Morton G, et al. Electrical injury during hands-ondefibrillation—A potential risk of internal cardioverter defibrillators? Resusci-tation 2009;80:832–4.

1. Personal conversation with Bill Havel, Medtronic scientist, 25.10.10.2. Edelson DP, Roberston-Dick BJ, Yuen T, et al. Safety and efficacy of defibrillator

charging during ongoing chest compressions: a multi-center study. Resuscita-tion 2010;81:1521–6.

101–4.4. Fish RM, Geddes LA. Conduction of electrical current to and through the human

body: a review. Eplasty 2009;9:e44.