By Mark W. Kroll

2154-2287/12/$31.00©2012IEEE

Date of publication: 24 September 2012

In the summer of 2006, manufacturers of a simple respira-

tor mask costing US$1 told the U.S. Congress that Americans

would find a shortage of these masks if

there was another flu pandemic. The reason

for this was that suing the makers of these

dust masks became a major business for trial

lawyers. By 2006, there had been more than

326,000 lawsuits filed. It has been reported in

many cases that the law firms worked with cooperative physi-

cians and contracted with X-ray labs to screen individuals for

lung problems. If an individual had an abnormal lung X-ray

and claimed that they had used a certain brand of respirator—

and the manufacturer was still in business

and had economic resources—a lawsuit was

filed. Some manufacturers went out of busi-

ness while others simply stopped making

the masks.

The 3M Company had 17,700 mask law-

suits filed against them, but it chose to fight rather than settle

for nuisance amounts that were far less than the costs of fighting

the lawsuits. The company had spent about US$16 million an-

nually fighting these cases, even though its record includes nine

victories out of nine cases that have gone to trial. Because of its

Editorial noteThe opinions presented in

this article are those of the

author and do not neces-

sarily represent those of

IEEE and its affiliates,

including IEEE Pulse.

To serve our mission as a

forum for a range of ideas,

the editors of IEEE Pulse

welcome your opinions.

Please contact us with

comments or for informa-

tion on how to submit an

opinion article.

Digital Object Identifier 10.1109/MPUL.2012.2205778

QUESTION MARK COURTESY OF STOCK.XCHNG/CHRIS BAKER.

SCALE COURTESY OF WIKIMEDIA COMMONS/TOBY HUDSON.

SEPTEMBER/OCTOBER 2012 ▼ IEEE PULSE 27

From Breast Implants to TASER Weapons

28 IEEE PULSE ▼ SEPTEMBER/OCTOBER 2012

willingness to fight these cases, the number of mask lawsuits is

now down to 2,200 after spending US$123 million defending

its product.

In a typical mask case, a plaintiff’s lawyer would retain a

physician expert witness who opined that the lung problem (or

abnormal X-ray) was caused by breathing some sort of dust (and

not due to cigarette smoking). Engineer expert witnesses would

also be retained who opined that they would have designed the

mask differently and conclude that the existing mask design was

suboptimal. In some cases, experts would cite

marginally related medical studies or engineer-

ing papers to attempt to buttress their opinions.

The defense lawyers would often argue—usu-

ally correctly—that the cited papers represented

“junk science” and strive to keep the citations and

the experts’ opinions out of court on a ruling of

the expert’s credentials, methodology, or opin-

ions being unreliable.

In these situations, manufacturers face a dif-

ficult decision between settling a frivolous nui-

sance value lawsuit and spending potentially

millions of dollars to go to trial. It is important

to note that the battle is asymmetrical as plain-

tiffs’ lawyers usually lose nothing but their time

(and expert-witness fees and minimal expenses) if they lose a

suit while a manufacturer has a risk of losing millions and, oc-

casionally, billions of dollars because of a sympathetic and/or

emotional jury or judge.

To help the biomedical engineer understand how these bat-

tles over money are fought, some time was spent in analyzing

the highly publicized lawsuits over breast implants and also the

TASER electronic control device (ECD). In both the cases, the

common threads of junk science, unreliable peer-reviewed piec-

es, media sensationalism, experienced plaintiffs’ lawyers, and

plaintiffs’ expert witnesses were seen.

Silicone Breast ImplantsMaria Stern sued Dow Corning, claiming that her breast im-

plants gave her an autoimmune disease. A jury agreed and,

in 1984, awarded her US$211,000. Her lawyer told the jury

that they needed to send a message to Dow Corning to punish

them for making a defective product, and the jury awarded

her another US$1.5 million in punitive damages. In 1990, the

U.S. Congress held hearings on the safety of breast implants.

This was apparently at the behest of plaintiffs’ lawyers as the

only three testifying scientific experts had previously served

as expert witnesses for plaintiffs’ lawyers. Just before the

hearings, the television show Face-to-Face with Connie Chung

had an episode on the alleged dangers of silicone implants.

This show produced a flurry of lawsuits, and, in 1991, Brenda

Toole prevailed with a US$5.4 million settlement. Interesting-

ly, she did not claim any injuries but merely alleged that she

had an increased risk of cancer and autoimmune disease. That

December, Mariann Hopkins was awarded US$7.3 million in

an implant case.

At this point, there were no published peer-reviewed lit-

erature suggesting any dangers from silicone breast implants.

There were scattered anecdotes that suggested a potential risk,

such as a 1981 case report of a Japanese woman who died of

kidney failure sometime after breast augmentation. There

was also a 1988 case report of a Florida patient who devel-

oped breast cancer 13 years after she had a silicone breast im-

plant. The authors stated that they could not conclude that the

implant had caused her cancer. It should be noted that these

anecdotes, or case reports, only suggest that scientific stud-

ies need to be done. For example, 1,000 women with breast

implants would be followed for ten years, and

their disease development would be compared

with 1,000 similar women who did not receive

breast implants.

The most famous breast implant lawsuit was

that of Pamela Johnson. She was a cigarette

smoker and had no autoimmune disease. She

testified that she had muscle and joint pain,

headaches, dizziness, and fatigue, which even

her lawyers admitted were basic flu symptoms.

She may have gained some jury sympathy

from the fact that her physician did not per-

form the implant procedure properly and one

of her implants ruptured. Instead of merely su-

ing the physician, her lawyer, John O’Conner,

sued the more attractive financial target—the manufacturer

of the implants. He also hired a public relations firm and ap-

peared on the Phil Donahue and 60 Minutes shows. This may

have influenced the pool of potential jurors. These tactics

were successful as the jury gave Johnson as well as her at-

torneys US$5 million in compensation and an additional

US$20 million in punitive damages for a total of US$25 mil-

lion. The trial was broadcast on the television show Court TV,

and the O’Conner Law firm filed hundreds of new cases in

1992. The same year, manufacturers Dow Corning, Bristol

Myers Squibb, and Bioplasty dropped out of the breast im-

plant business.

Coincidently, also in 1992, major studies appeared in promi-

nent medical journals showing that women with breast im-

plants did not have an increased risk of cancer. Lawyers then

changed their claims to focus on the more nebulous immune-

system diseases.

By the end of 1993, more than 12,000 lawsuits had been

filed against Dow Corning alone. In the fall of that year,

the defendant companies got together and agreed to pay

US$4.75 billion to settle all breast implant lawsuits. This

large pool of money attracted so many additional plaintiffs

that the settlement agreement collapsed. In March 1994, the

Dow Corning Company was forced to file for bankruptcy.

However, they did eventually win a Supreme Court deci-

sion in Daubert v. Dow, establishing reliability requirements

for expert testimony admissibility in the courtroom. The

remaining implant manufacturers settled their lawsuits for

US$3.4 billion, setting a new record for class action lawsuits

(also in March of 1994).

Just three months later, the New England Journal of Medi-

cine published a study showing that there was, in fact, zero

connection between these implants and connective tissue

In these situations, manufacturers face a difficult decision between settling a frivolous nuisance value lawsuit and

spending potentially millions of dollars to

go to trial.

SEPTEMBER/OCTOBER 2012 ▼ IEEE PULSE 29

disease. In 1995, the journal published an even larger study

with the same results. In the same year, the American Col-

lege of Rheumatology stated that the evidence was compel-

ling that there was no relationship between silicone implants

and either connective tissue or autoimmune disease. They

also asked that the U.S. Food and Drug Ad-

ministration and the courts stop using anec-

dotes to imply such a connection. Neverthe-

less, the lawsuits continued. Finally, in 1996,

a federal judge took input from an impartial

scientific panel that found that there was no

connection between these implants and dis-

ease. Later, another federal judge appointed

his own expert panel. They spent two years

and US$800,000 and also concluded that there

was no connection between breast implants

and disease.

Nearly a decade after the Connie Chung

show, the media began to change course and

The New York xTimes reported that manufactures

had won 80% of the breast implant cases that

went to trial. Eventually, the tactics of these trial lawyers were

exposed. They had used a small number of “silicone doctors”

to diagnose diseases that were supposedly caused by the im-

plants. Some lawyers had flown these physicians around the

country to examine their clients using their law office rooms

for the examinations. Further questionable practices included

the lawyers paying the medical bills, which is actually illegal

in several states. Often the physicians agreed to

be paid by the patients only after the patient

received lawsuit proceeds, a practice that cre-

ated a clear conflict in the physician’s objec-

tivity. One physician treated almost 5,000

women with implants and admitted that

more than 90% of his patients had been

referred by lawyers. He opined that

93% of the women examined by

him had been harmed by the

breast implants.

Beyond the obvious ques-

tions of the physician’s di-

agnostic objectivity, there is

a high likelihood that thousands of

women were unnecessarily given painful and

expensive testing. It is one thing to decide that a lawyer’s

client has a given disease but now the physician is obligat-

ed to prescribe a treatment for the diagnosed disease. The

silicone doctors prescribed treatments, including the cancer

drug Cytoxan and intravenous gamma globulin, which car-

ry significant side-effect risks.

How Publication Pressures Can Create Junk (Unreliable) ScienceThe prestige of a scientific journal is measured by how often its

papers are cited in other journals. The simplest calculation is that

of the impact factor, which is the average number of times a jour-

nal article is cited in later articles. For example, if a given journal

published 200 articles in a given two-year period, and these ar-

ticles were cited a total of 600 times in the following year, then

the impact factor was 3.0.

The competition for a high impact factor incentivizes

journals to publish high-quality papers on important top-

ics. Unfortunately, it also can incentivize some

journals to publish sensationalistic anecdotes.

The more sensational the anecdote—as long as

it seems plausible—the higher the potential im-

pact factor for the journal. The most prestigious

journals are not immune to this incentive as

they have high impact factors to protect [1].

TASER ECDsThe TASER ECD has revolutionized law enforce-

ment. The majority of law-enforcement agencies

in the United States and Canada—and every

agency in the United Kingdom—now use this de-

vice. Although the TASER ECD is not a medical

device, it is designed and tested according to the

principles of bioelectricity and biomedical engi-

neering and, thus, provides a meaningful example. This article’s

author serves on the TASER (International, Inc.) corporate as

well as the Scientific and Medical Advisory Boards, by serving

as a consultant and expert, and can provide an insider’s look at

this example.

Law-en forcement officers often need to capture, control, and

restrain a violent and paranoid criminal who is often under the

influence of illegal drugs or alcohol or serious psycho-

logical distress (SPD). The main

methods used by officers in

such situations in the past re-

lied on inflicting pain or trau-

matic injury. The old techniques

included punching, painful

joint-twists, grounding, wres-

tling, chemical irritants, impact

tools, and firearms.

Unfortunately, many illegal

drugs (and alcohol) are painkillers,

and, as a result, standard subdual tech-

niques are often ineffective. Even worse,

many of the dangerously drug-addled per-

petrators and persons in SPD exhibit superhu-

man stamina and strength. There are numerous

accounts of such people manhandling several law-enforce-

ment officers at once. Many officers have been injured along

with those they are trying to restrain.

When the trigger of a TASER ECD is pulled, a blast of

compressed nitrogen launches its two barbed darts at 55 m/s.

The projectiles, which weigh 1.6 g each, have a 9- to 13-mm-

long tip to penetrate clothing and the insulating outer layer

of skin. The hair-thin (127-nm diameter) wires trail behind

for usually up to 9 m, forming an electrical connection to

the weapon. (There is also a less-used drive-stun mode in

which the ECD-fixed electrodes are pressed directly against

the skin.)

the physicians agreed to

only after the patient

ds, a practice that cre-

the physician’s objec-

reated almost 5,000

and admitted that

atients had been

opined that

mined by

by the

ques-

’s di-

here is

t thousands of

arily given painful and

one thing to decide that a lawyer’s

influence of illegal drugs or al

logical

meth

such

lied on

matic i

includ

joint-tw

tling, ch

tools, and

Unfort

drugs (and

and, as a resul

niques are often

many of the dange

petrators and persons

man stamina and streng

Law-enforcement officers often need to capture, control,

and restrain a violent and paranoid criminal

who is often under the influence of illegal

drugs or alcohol or serious psychological

distress.

©IN

GR

AM P

UBL

ISH

ING

Because the barbs get stuck in clothing and fail to reach

the skin about 30% of the time, the ECD is designed to gen-

erate a brief arcing pulse, which ionizes the intervening

air to establish a conductive path. The arcing phase has an

open-circuit peak voltage of 50 kV; that is, the voltage is

50  kV only until the arc appears or until the barbs make

contact with conductive flesh, which offers approximately

600 Ω of resistance.

The target’s body is never exposed to 50-kV voltage. The X26

ECD—the TASER model most commonly used by law enforce-

ment agencies—delivers a pulse voltage of 600 V to the body.

Once the barbs establish a circuit, the ECD

generates a series of 100-ns pulses at a rate of

19 pulses/s). Each pulse carries 100 nC (micro-

coulombs) of charge, so the aggregate (and actual

average) current is 1.9 mA. The newer X2 and X3

ECD models deliver a metered charge of 63 nC.

The ECD takes advantage of two primary

natural protections against electrocution that

arise from the difference between skeletal and

cardiac muscle. The first—anatomy—is so obvi-

ous that it is typically overlooked. The skeletal

muscles are on the outer shell of the body, and

the heart is nestled farther inside beneath the

thoracic cage. In the upper body, the skeletal

muscles are arranged in bands surrounding the

rib cage. Because of the skeletal muscle’s natu-

ral inclination to conduct current along its fibers (known as

anisotropy), current injected into such a muscle tends to follow

the grain around the chest rather than penetrating toward the

heart. In addition, the ribs and lungs surrounding the heart are

essentially insulators.

The second protection results from the different timing

requirements of the nerves that trigger muscle contractions

(motor neurons) versus cardiac cells. The chronaxie (the pulse

duration that stimulates with the minimum energy) is around

100 ns for the motor neurons, and it is far higher at around

2 ms for cardiac cells. Both of these values decrease for elec-

trodes closer to the target neurons.

These devices have now been applied to human beings more

than 3 million times. About one half of those applications were

to law-enforcement officers for training purposes and the other

half to subjects in the field. More than 12 deaths occurred due

to brain injuries from uncontrolled falls out of >1.6 million

subject applications. As the ECD is electrical in nature, the pri-

mary remaining concern lies with the potential for electrocu-

tion, which is the electrical induction of ventricular fibrillation

(VF). The 100-nC charge per pulse (which is what determines

the potential to affect the heart) can be compared with electric

fence energizers that can provide an output up to 1.1 mC, or

11 times more charge [2]. Granted, the ECD delivers pulses at

a more rapid rate than does the electric fence, but, even when

factoring that in, the most popular ECD (TASER X26) delivers

only 44% of the current allowed for electric fences by Under-

writer’s Laboratory standards.

Scientifically, there should be no scientific debate on the

ability of these devices to cause VF since modern electric fences

do not cause VF. There should be no debate except for the usual

junk science.

Similar to the breast implant litigation, the plaintiffs’ law-

yers have their favorite TASER doctors. One of the most popular

such experts has charged more than US$100,000—in a single

case—to opine that an ECD was causal in a death. As there are

no peer-reviewed papers suggesting a risk of VF in humans, this

expert packaged up his paid expert-witness opinions and—after

some two rejections—found a medical journal that would accept

the advocate litigation case series as a paper [3]. This retired car-

diologist requested that the most knowledgeable authorities be

excluded from reviewing his case series. Hence,

there is some debate as to whether this case se-

ries can be truly considered peer reviewed.

ECD AnecdotesThe best-known ECD anecdote is found in a two-

paragraph letter to the editor in the New England

Journal of Medicine published in September 2005,

with regard to a February 2005 incident [1]. It

read as follows:

The question of the safety of the use of

“stun guns” by law-enforcement agencies

has been raised in the news. Deaths after dis-

charges from such devices (Tasers) have been

reported, although no definite causative link

between death and the use of a stun gun has

been made. An adolescent was subdued with a Taser

stun gun and subsequently collapsed. Paramedics found

the adolescent to be in ventricular fibrillation and began

performing cardiopulmonary resuscitation within two

minutes after the collapse. After four shocks and the ad-

ministration of epinephrine, atropine, and lidocaine, a

perfusing rhythm was restored. The adolescent made a

nearly complete recovery and was discharged from the

hospital several days later. This case of ventricular fibril-

lation after a discharge from a stun gun suggests that the

availability of automated external defibrillators to law-

enforcement personnel carrying stun guns should be

considered.

The misperception left by this letter is that the ECD elec-

trically induced VF and that officers should carry automated

external defibrillators, as the sentence regarding the finding

of VF is placed right after the sentence relating the subdual

with the ECD. The use of the word “collapse” also gave a false

impression, as this word is typically used to signify a cardio-

vascular collapse. However, the primary function of an ECD is

to cause a postural collapse, i.e., cause someone to collapse to

the ground to stop resisting as well as threatening behavior.

The documented facts are significantly different from the

misleading impressions given by this letter. The violent psychi-

atric subject, a ward of the state, had a psychotic episode and

punched through a glass door and gave himself significant lac-

erations requiring emergency medical care. Paramedics and po-

lice were called to the scene. The subject refused medical care

and jumped at a police officer, who then used an ECD to control

him. The subject was handcuffed to the paramedics’ gurney. The

The 100 µC charge per pulse (which is

what determines the potential to affect the heart) can be

compared to electric fence energizers that can provide an output

up to 1.1 mC, or 11 times more charge.

30 IEEE PULSE ▼ SEPTEMBER/OCTOBER 2012

SEPTEMBER/OCTOBER 2012 ▼ IEEE PULSE 31

subject stopped responding to communication. It was not clear

whether he was faking or had fainted. Paramedics checked his

pulse and respiration with a 15-s vitals check and found it to

be normal. This is documented in written records and sworn

testimony. Obviously, had the ECD induced VF,

there would have been no pulse or respiration

afterward.

The subject was taken down to the ground floor

and placed in the ambulance where a second vi-

tals check was taken and found to be normal, fol-

lowed by his developing VF and being eventually

resuscitated. It has also been demonstrated that

the cardiac rhythm strip shown in the letter (sup-

posedly demonstrating a return to sinus rhythm

by a defibrillation shock) was cropped after what

were actually three premature ventricular con-

tractions followed by asystole. It has not been

established whether this was intentional, acci-

dental, or the fault of the author or the journal.

Finally, the rhythm strip cited as representing the rhythm sur-

rounding the fourth shock has also been shown to be the strip

of the second shock.

This anecdote has been rebuked in the literature and is no

longer taken seriously by most scientists. However, despite the

manifested errors in the letter, the anecdote continues to be ag-

gressively cited in litigation.

A more recent anecdote, published in the British Emergency

Medicine Journal, also claimed that a teenager had VF induced by

an ECD [4]. The lead author inappropriately relied on the teen-

ager’s mother for information rather than on the emergency

medical services (EMSs) records. This VF allegation was com-

pletely contradicted by the EMSs records and sworn testimony

of the paramedics on scene. Interestingly, this anecdote has been

repudiated by other emergency physicians in the same hospital

in their case report published in American Journal of Emergency

Medicine [5]. They published that the subject, presented with

asystole—not VF—which is consistent with his extreme blood

concentration of alcohol at 80% of the median lethal level. Note

that asystole cannot be induced by electrical stimulation.

How Medical Examiner Caution Can Encourage LawsuitsThe medical examiner (ME) is highly trained to recognize me-

chanical trauma and find toxic substances in the body. There

is a presumption that the ME’s autopsy report

is dispositive as to the cause of death. Hence, if

the autopsy report mentions the ECD as causal

or contributory (generally out of error) or could

not be ruled out (out of misplaced caution), then

an expensive lawsuit is typical.

As electrical current does not linger or accu-

mulate in the body, some MEs have, in the past,

decided to err on the side of including the ECD

even though they had no explanation—and no

scientific support—for how it could cause or con-

tribute to someone’s death. However, as more

data and published studies have come out, MEs

quickly began to make more accurate judgments

about the causes of death.

A total of 301 arrest-related death (ARD) autopsy reports

covering deaths that occurred between January 2001 and De-

cember 2006 were studied. The findings of almost all ARDs

were analyzed, in which an ECD was used at some point during

the arrest attempt.

There were 39 cases (9.4%) where the autopsy report listed

the ECD as possibly contributory or as an unknown factor. The

listing rate declined from 33% in 2001 to 3.3% in 2006 (r2 =

0.73, P = 0.031 for linear fit) as seen in Figure 1. The decedents

were primarily male (38 male/one female) with mean age 35.0 !

10.9 years (median = 32), which is consistent with other reported

ARD data. The errors were counted based on the present state

of knowledge and the well-known presentation of electrocution

(electrical induction of VF). These counted errors were as follows:

1) delayed collapse ignored (VF causes a cardiovascular col-

lapse within 5–10 s)

2) non-VF cardiac rhythm ignored (asystole and pulseless elec-

trical activity cannot be generated by electrical stimulation)

3) failure of defibrillation ignored (electrically induced VF is

easily reversed electrically)

The ECD takes advantage of two primary natural

protections against electrocution that

arise from the difference between skeletal and cardiac

muscle.

FIGURE 2 The scored error rate has fallen significantly from 3.5 per autopsy in 2002–2004 to a rate of one to two in 2006. A quadratic fit trend line is shown.

6

5

4

3

2

1

0

Jan.

2001

Jan.

2002

Jan.

2003

Jan.

2004

Jan.

2005

Jan.

2006

Jan.

2007

Date of Death

Err

ors

Per

Aut

opsy

FIGURE 1 The rate at which MEs have mentioned an ECD as possibly contributing to an ARD declined significantly over the study period.

35

30

25

20

15

10

5

02001 2002 2003

Year of Death

2004 2005 2006

EC

D M

entio

ned

Rat

e (%

)

4) discharge duration or number of discharges stressed (in all

relevant circumstances, VF is either induced or not in 5 s)

5) drive-stun mode ignored (as it only causes pain and does not

deliver current to any organs)

6) assumed drug–ECD dysynergy (cocaine and methamphet-

amine actually make it harder to be electrocuted)

7) cardiac damage blamed on an ECD (permanent cellular

effects require extreme currents from lightning strikes or

power lines)

8) impaired respiration assumed (an ECD does not interfere

with human breathing)

9) nebulous equivocal comments such as “could not be ruled

out” or “straw that broke the camel’s back.”

A mean of 2.9 ± 1.3 scored errors per autopsy report was found

with a range of one to six. This error rate declined steadily over

time, as seen in Figure 2 (P = 0.035).

During 2005, conference presentations began to address the

speculated safety concerns regarding these ECDs. As seen in Fig-

ure 3, it was not until 2006 that the first MedLine-indexed pa-

pers on ECD effects appeared. These factors may explain why the

autopsy report mentions—of an ECD contribution to an ARD—

fell to 3.3% in 2006.

Advice for the Biomedical EngineerObviously, a manufacturing company will do its best to design

a high-performance and high-quality device. Sadly, that may

not protect these companies from being dragged into a product

liability lawsuit. An important issue that is often overlooked is

e-mail informality.

Imagine that you and a colleague derive different estimates

for the number of samples of a certain capacitor that should be

tested for a medical device you are developing. Your statistical

calculations say that 173 samples should be tested and your

colleague comes up with 151. At the end of a good-natured

e-mail banter, you finally agree with your col-

league’s calculations and settle on 151 samples.

To show that you have no hard feelings, you

send a friendly—but flippant—e-mail saying, “I

sure hope no one dies because we only tested

151 capacitors.”

The product goes to market and someone

dies with your product inside of them. The per-

formance of your product may or may not have

had been causal with the death. However, what

is clear is that your capacitor never failed, and so

the question of testing 151 versus 173 samples is

scientifically, but definitely not litigatiously, ir-

relevant. A lawsuit is filed, and under discovery

rules, the plaintiff’s lawyers get every e-mail re-

motely related to this product. It is not unusual

for a million e-mails to be provided to the other

side. They will search through these e-mails for

relevant evidence, including flippant, salacious,

and in-artfully drafted or ambiguous comments,

and then try to turn such comments into a scan-

dal for the jury.

Even though your capacitor had nothing to

do with the death, your e-mail may be presented to a jury as

evidence that your company thought that safety testing was a

big joke. One cannot assume that a jury will be able to factor in

the irrelevance of your e-mail. Remember that every e-mail you

send could someday be waved in front of a jury by a skilled law-

yer who knows how to generate an emotional reaction.

ConclusionsMedical device litigation is a large industry in the United

States. In many cases, there was no true product defect behind

the litigation. Beyond designing high-quality devices, the bio-

medical engineer must understand the realities of this litiga-

tion environment and be cautious with the use of humor or

irony in e-mails.

Mark W. Kroll (mark@kroll.name) is with the Department of Bio-

medical Engineering at the University of Minnesota, Minneapolis.

References[1] P. J. Kim and W. H. Franklin, “Ventricular fibrillation after stun-gun

discharge,” N. Engl. J. Med., vol. 353, no. 9, pp. 958–959, Sept. 2005.

[2] A. J. Nimunkar and J. G. Webster, “Safety of pulsed electric de-

vices,” Physiol. Meas., vol. 30, no. 1, pp. 101–114, Jan. 2009.

[3] D. P. Zipes, “Sudden cardiac arrest and death following application

of shocks from a TASER electronic control device,” Circulation, vol.

125, no. 20, pp. 2417–2422, May 2012.

[4] R. S. Naunheim, M. Treaster, and C. Aubin, “Ventricular fibrilla-

tion in a man shot with a Taser,” Emerg. Med. J., vol. 27, no. 8, pp.

645–646, Aug. 2010.

[5] E. S. Schwarz, M. Barra, and M. M. Liao, “Successful resuscitation of

a patient in asystole after a TASER injury using a hypothermia proto-

col,” Am. J. Emerg. Med., vol. 27, no. 4, pp. 515.e1-2, May 2009.

FIGURE 3 The first MedLine-indexed studies of ECDs appeared in 2006, and this literature body has grown rapidly since then.

20

18

16

14

12

10

8

6

4

2

0

2004 2005 2006 2007 2008 2009 2010

Publication Year

Human

Animal

Modeling

Num

ber

of N

ew P

aper

s

32 IEEE PULSE ▼ SEPTEMBER/OCTOBER 2012