Resuscitation

 

83 (2012) 107–

 

112

Contents

 

lists

 

available

 

at

 

SciVerse

 

ScienceDirect

Resuscitation

j o

 

u

 

r n

 

a l

 

h o m

 

e p a g e

 

:

 

w w w . e l s e v i e r . c o m / l o c a t e / r e s u s c i t a t i o n

Experimental

 

paper

Pharmacokinetics

 

of

 

intraosseous

 

and

 

central

 

venous

 

drug

 

delivery

 

during

cardiopulmonary

 

resuscitation

夽,夽夽

Stephen

 

L.

 

Hoskins

a

,

 

Paulo

 

do

 

Nascimento

 

Jr.

a

,

b

,

 

Rodrigo

 

M.

 

Lima

a

,

b

,

 

Jonathan

 

M.

 

Espana-Tenorio

a

,

George

 

C.

 

Kramer

a

,

∗

a

Resuscitation

 

Research

 

Laboratory,

 

Department

 

of

 

Anaesthesiology,

 

University

 

of

 

Texas

 

Medical

 

Branch,

 

301

 

University

 

Blvd,

 

Galveston,

 

TX

 

77555-0801,

 

United

 

States

b

Sao

 

Paulo

 

Medical

 

school,

 

Department

 

of

 

Anesthesiology,

 

Unesp,

 

Botucatu,

 

SP,

 

Brazil

a

 

r

 

t

 

i

 

c

 

l

 

e

 

i

 

n

 

f

 

o

Article

 

history:

Received

 

27

 

January

 

2011

Received

 

in

 

revised

 

form

 

20

 

July

 

2011

Accepted

 

26

 

July

 

2011

Keywords:
Intraosseous
Cardiopulmonary

 

resuscitation

CPR
Pharmacokinetics
Tracers
Drug

 

delivery

a

 

b

 

s

 

t

 

r

 

a

 

c

 

t

We

 

compared

 

the

 

pharmacokinetics

 

of

 

intraosseous

 

(IO)

 

drug

 

delivery

 

via

 

tibia

 

or

 

sternum,

 

with

 

central

venous

 

(CV)

 

drug

 

delivery

 

during

 

cardiopulmonary

 

resuscitation

 

(CPR).

Methods:

 

CPR

 

of

 

anesthetized

 

KCl

 

arrest

 

swine

 

was

 

initiated

 

8

 

min

 

post

 

arrest.

 

Evans

 

blue

 

and

 

indocyanine

green,

 

each

 

were

 

simultaneously

 

injected

 

as

 

a

 

bolus

 

with

 

adrenaline

 

through

 

IO

 

sternal

 

and

 

tibial

 

needles,

respectively,

 

n

 

=

 

7.

 

In

 

second

 

group

 

(n

 

=

 

6)

 

simultaneous

 

IO

 

sternal

 

and

 

IV

 

central

 

venous

 

(CV)

 

injections

were

 

made.

Results:

 

Peak

 

arterial

 

blood

 

concentrations

 

were

 

achieved

 

faster

 

for

 

sternal

 

IO

 

vs.

 

tibial

 

IO

 

administration

(53

 

±

 

11

 

s

 

vs.

 

107

 

±

 

27

 

s,

 

p

 

=

 

0.03).

 

Tibial

 

IO

 

dose

 

delivered

 

was

 

65%

 

of

 

sternal

 

administration

 

(p

 

=

 

0.003).

Time

 

to

 

peak

 

blood

 

concentration

 

was

 

similar

 

for

 

sternal

 

IO

 

and

 

CV

 

administration

 

(97

 

±

 

17

 

s

 

vs.

 

70

 

±

 

12

 

s,

respectively;

 

p

 

=

 

0.17)

 

with

 

total

 

dose

 

delivered

 

of

 

sternal

 

being

 

86%

 

of

 

the

 

dose

 

delivered

 

via

 

CV

 

(p

 

=

 

0.22).

Conclusions:

 

IO

 

drug

 

administrations

 

via

 

either

 

the

 

sternum

 

or

 

tibia

 

were

 

effective

 

during

 

CPR

 

in

 

anes-

thetized

 

swine.

 

However,

 

IO

 

drug

 

administration

 

via

 

the

 

sternum

 

was

 

significantly

 

faster

 

and

 

delivered

a

 

larger

 

dose.

© 2011 Elsevier Ireland Ltd. All rights reserved.

1.

 

Introduction

Survival

 

from

 

out-of-hospital

 

cardiac

 

arrest

 

depends

 

on

 

a

sequence

 

of

 

therapeutic

 

interventions

 

termed

 

the

 

“chain

 

of

 

sur-

vival”

 

by

 

the

 

American

 

Heart

 

Association

 

(AHA).

 

This

 

sequence

includes

 

rapid

 

access

 

to

 

emergency

 

medical

 

care,

 

cardiopul-

monary

 

resuscitation

 

(CPR),

 

defibrillation,

 

advanced

 

care,

 

and

 

post

resuscitation

 

techniques

 

such

 

as

 

hypothermia,

 

percutaneous

 

coro-

nary

 

interventions,

 

and

 

implantable

 

cardioverter-defibrilators.

1,2

Unfortunately,

 

survival

 

rates

 

after

 

cardiac

 

arrests

 

are

 

dismal

(2.5–10.5%).

3–5

More

 

rapid

 

vascular

 

accesses

 

for

 

drug

 

delivery

 

dur-

ing

 

CPR

 

may

 

be

 

one

 

way

 

of

 

improving

 

survival.

Intravenous

 

access

 

during

 

CPR

 

can

 

be

 

difficult

 

even

 

for

 

an

experienced

 

caregiver.

 

In

 

one

 

study,

 

the

 

median

 

time

 

required

to

 

establish

 

an

 

intravenous

 

(IV)

 

line

 

by

 

well-trained

 

paramedics

夽 A

 

Spanish

 

translated

 

version

 

of

 

the

 

abstract

 

of

 

this

 

article

 

appears

 

as

 

Appendix

in

 

the

 

final

 

online

 

version

 

at

 

doi:10.1016/j.resuscitation.2011.07.041

.

夽夽 Financial

 

support:

 

American

 

Heart

 

Association

 

Texas

 

Affiliate

 

Grant-in-Aid

#0455157Y.

∗ Corresponding

 

author.

 

Tel.:

 

+1

 

409

 

772

 

3969;

 

fax:

 

+1

 

409

 

772

 

8895.

E-mail

 

addresses:

 

gkramer@utmb.edu

,

 

mtownsen@utmb.edu

 

(G.C.

 

Kramer).

in

 

the

 

field

 

was

 

2

 

min

 

for

 

first

 

attempts

 

and

 

5

 

min

 

when

 

further

attempts

 

were

 

required.

6

The

 

overall

 

success

 

rate

 

to

 

establish

 

an

IV

 

line

 

in

 

the

 

field

 

for

 

medical

 

emergencies

 

is

 

less

 

than

 

75%.

6–8

There

 

remains

 

a

 

need

 

for

 

more

 

rapid

 

vascular

 

accesses

 

for

 

drug

delivery

 

during

 

CPR

 

may

 

be

 

one

 

way

 

of

 

improving

 

survival.

 

Intra-

venous

 

access

 

during

 

cardiopulmonary

 

resuscitation

 

(CPR)

 

can

 

be

difficult

 

even

 

for

 

an

 

experienced

 

caregiver.

 

Intraosseous

 

vascular

(IO)

 

access

 

is

 

an

 

established

 

rapid,

 

safe,

 

and

 

effective

 

alternative

for

 

peripheral

 

intravenous

 

drug

 

delivery.

8,9

The

 

American

 

Heart

Association

 

and

 

the

 

European

 

Resuscitation

 

Council

 

Guidelines

 

for

Pediatric

 

Life

 

Support

 

recommend

 

IO

 

access

 

via

 

the

 

tibia

 

for

 

pedi-

atric

 

patients.

12,13

In

 

the

 

last

 

10

 

years,

 

several

 

large

 

bore

 

IO

 

needles

for

 

adult

 

patients

 

have

 

become

 

available

 

that

 

use

 

IO

 

access

 

via

 

the

sternum,

 

tibia

 

and

 

humerus.

 

These

 

devices

 

have

 

been

 

evaluated

 

in

both

 

patients

 

and

 

animals.

8,10,11

Use

 

of

 

these

 

devices

 

provides

 

rapid

access

 

to

 

the

 

systemic

 

circulation

 

during

 

normovolemia.

7,8,10,14

However,

 

the

 

effectiveness

 

of

 

IO

 

drug

 

delivery

 

via

 

different

 

anatom-

ical

 

sites

 

during

 

CPR

 

has

 

been

 

under

 

evaluation.

We

 

used

 

a

 

swine

 

model

 

of

 

cardiac

 

arrest

 

to

 

determine

 

the

 

phar-

macokinetics

 

of

 

IO

 

delivery

 

of

 

a

 

double

 

dye

 

tracer

 

method

 

during

CPR

 

using

 

simultaneous

 

IO

 

injections

 

in

 

the

 

sternum

 

and

 

tibia.

 

We

also

 

compared

 

the

 

pharmacokinetics

 

of

 

tracer

 

administration

 

via

the

 

sternum

 

vs.

 

central

 

venous

 

IV

 

administration.

0300-9572/$

 

–

 

see

 

front

 

matter ©

 

 2011 Elsevier Ireland Ltd. All rights reserved.

doi:

10.1016/j.resuscitation.2011.07.041

108

S.L.

 

Hoskins

 

et

 

al.

 

/

 

Resuscitation

 

83 (2012) 107–

 

112

2.

 

Methods

2.1.

 

Animal

 

preparation

The

 

study

 

protocol

 

was

 

approved

 

by

 

the

 

University

 

of

 

Texas

Medical

 

Branch’s

 

Institutional

 

Animal

 

Care

 

and

 

Use

 

Committee

(IACUC).

 

UTMB

 

animal

 

facilities

 

are

 

accredited

 

by

 

the

 

American

Association

 

for

 

the

 

Accreditation

 

of

 

Laboratory

 

Animal

 

Care.

The

 

experimental

 

model

 

was

 

Yorkshire

 

swine

 

(25–35

 

kg).

 

The

night

 

before

 

the

 

experiment

 

food

 

was

 

withheld

 

from

 

the

 

animals,

though

 

they

 

had

 

free

 

access

 

to

 

water.

 

Pre

 

sedation

 

was

 

induced

the

 

day

 

of

 

the

 

experiment

 

by

 

an

 

intramuscular

 

injection

 

of

 

telazol,

ketamine,

 

and

 

xylazine.

 

A

 

22

 

gauge

 

peripheral

 

intravenous

 

catheter

was

 

placed

 

in

 

the

 

ear

 

vein

 

to

 

deliver

 

fluids

 

and

 

alpha

 

chloralose.

 

The

animals

 

were

 

anesthetized

 

for

 

the

 

surgical

 

prep

 

with

 

2–4%

 

isoflu-

rane

 

by

 

facial

 

mask

 

and

 

then

 

intubated

 

orotracheally

 

using

 

direct

laryngoscopy.

 

Animals

 

were

 

placed

 

supine

 

on

 

a

 

heating

 

blanket

 

to

maintain

 

body

 

temperature

 

between

 

38

 

and

 

39

◦

C.

 

Surgical

 

areas

were

 

scrubbed

 

and

 

covered

 

with

 

sterile

 

surgical

 

drapes.

 

Mechani-

cal

 

ventilation

 

was

 

established

 

at

 

a

 

tidal

 

volume

 

of

 

15–20

 

ml/kg

 

and

a

 

ventilatory

 

rate

 

of

 

12–16

 

breaths/min

 

to

 

maintain

 

end

 

tidal

 

car-

bon

 

dioxide

 

between

 

30

 

and

 

40

 

mmHg.

 

Thereafter,

 

isoflurane

 

was

discontinued

 

and

 

anesthesia

 

was

 

maintained

 

with

 

an

 

IV

 

infusion

 

of

1%

 

alpha

 

chloralose

 

via

 

the

 

catheter

 

in

 

the

 

ear,

 

administered

 

as

 

an

initial

 

bolus

 

of

 

50

 

mg/kg

 

and

 

sustained

 

with

 

a

 

continuous

 

infusion

at

 

10

 

mg/kg/h.

The

 

carotid

 

artery

 

was

 

cannulated

 

for

 

arterial

 

blood

 

sampling

via

 

an

 

incision

 

of

 

the

 

right

 

side

 

of

 

the

 

neck.

 

A

 

central

 

venous

catheter

 

was

 

placed

 

via

 

the

 

external

 

jugular

 

vein

 

to

 

provide

 

dye

tracer

 

administration

 

into

 

the

 

central

 

venous

 

circulation.

 

Catheters

were

 

placed

 

into

 

the

 

aorta,

 

via

 

right

 

femoral

 

artery,

 

and

 

femoral

vein

 

for

 

acute

 

monitoring

 

and

 

recording

 

of

 

mean

 

arterial

 

pres-

sures

 

and

 

drug

 

delivery

 

by

 

sampling

 

arterial

 

blood,

 

respectively.

IO

 

needles

 

Jamshidi

 

(Baxter,

 

Deerfield,

 

IL)

 

or

 

EZ-IO

®

(VidaCare,

 

San

Antonio,

 

TX)

 

were

 

placed

 

in

 

the

 

manubrium

 

5

 

cm

 

caudal

 

of

 

the

 

ster-

nal

 

notch,

 

and

 

at

 

3

 

cm

 

distal

 

of

 

the

 

tibial

 

tuberosity,

 

respectively.

Correct

 

placement

 

was

 

confirmed

 

by

 

cross

 

section

 

at

 

necropsy.

 

Lac-

tated

 

Ringer’s

 

solution

 

was

 

administered

 

at

 

a

 

rate

 

of

 

15

 

ml/kg/h

during

 

surgery.

 

Standard

 

hemodynamics

 

were

 

monitored

 

(Hewlett

Packard,

 

Andover,

 

MA)

 

throughout

 

the

 

experiments.

 

Data

 

were

recorded

 

via

 

a

 

multi

 

channel

 

analog-digital

 

data

 

acquisition

 

pro-

gram

 

using

 

PowerLab

 

(AD

 

Instruments,

 

UK).

2.2.

 

Protocol

Two

 

protocols

 

were

 

employed

 

with

 

simultaneous

 

injections;

both

 

of

 

them

 

were

 

terminal

 

studies.

 

Protocol

 

I

 

(sternum

 

vs.

tibia)

 

compared

 

the

 

pharmacokinetics

 

of

 

two

 

different

 

dye

 

tracers

administered

 

intraosseously

 

and

 

simultaneously

 

via

 

the

 

sternum

and

 

the

 

tibia,

 

respectively.

 

Protocol

 

II

 

(sternum

 

vs.

 

central

 

venous

IV)

 

compared

 

the

 

pharmacokinetics

 

of

 

IO

 

administration

 

of

 

dye

tracers

 

via

 

the

 

sternum

 

with

 

a

 

simultaneous

 

administration

 

via

central

 

venous

 

IV.

 

A

 

60-min

 

baseline

 

time

 

period

 

was

 

established

after

 

completion

 

of

 

instrumentation.

 

Lactate

 

and

 

blood

 

gas

 

vari-

ables

 

were

 

monitored

 

to

 

ensure

 

that

 

the

 

animals

 

had

 

sufficiently

recovered

 

from

 

the

 

surgical

 

procedure

 

and

 

reached

 

a

 

physio-

logic

 

baseline

 

before

 

experimental

 

data

 

was

 

collected.

 

Heparin,

10,000

 

units

 

was

 

administered

 

IV

 

prior

 

to

 

the

 

induction

 

of

 

cardiac

arrest.

 

During

 

low

 

flow

 

states

 

such

 

as

 

cardiac

 

arrest,

 

blood

 

sampling

can

 

be

 

difficult

 

if

 

the

 

lines

 

become

 

clotted.

 

Prior

 

to

 

the

 

induction

of

 

cardiac

 

arrest,

 

the

 

animals

 

were

 

administered

 

a

 

ketamine

 

bolus

(30

 

mg/kg)

 

to

 

achieve

 

a

 

deeper

 

anesthesia

 

plane

 

and

 

avoid

 

any

 

dis-

tress

 

during

 

the

 

cardiac

 

arrest

 

and

 

resuscitation.

Cardiac

 

arrest

 

was

 

induced

 

by

 

rapid

 

IV

 

administration

 

of

 

10

 

ml

of

 

saturated

 

potassium

 

chloride

 

(KCl)

 

(Hospira

 

Inc.,

 

Lake

 

Forest,

 

IL)

solution

 

via

 

central

 

venous

 

catheter

 

followed

 

by

 

a

 

10

 

ml

 

saline

 

flush.

Immediately

 

following

 

the

 

injection

 

of

 

KCl

 

the

 

electrocardiogram

(EKG)

 

displayed

 

a

 

typical

 

ventricular

 

fibrillation

 

(VF)

 

waveform.

Ventilator

 

support

 

was

 

terminated

 

at

 

this

 

time.

 

Cardiac

 

arrest

 

was

followed

 

by

 

an

 

8-min

 

period

 

of

 

untreated

 

ventricular

 

fibrillation.

CPR

 

was

 

then

 

initiated

 

and

 

delivered

 

by

 

a

 

mechanical

 

chest

 

com-

pression

 

device

 

(Thumper

®

Michigan

 

Instruments,

 

Grand

 

Rapids,

MI)

 

at

 

100

 

compressions

 

per

 

min

 

(without

 

supplemental

 

O

2

)

 

and

at

 

duty

 

cycle

 

rate

 

of

 

50%.

 

A

 

compression

 

depth

 

was

 

set

 

at

 

2-in.

 

and

chest

 

compressions

 

were

 

delivered

 

in

 

an

 

anterior/posterior

 

posi-

tion

 

centered

 

on

 

the

 

sternal

 

body.

 

After

 

1-min

 

of

 

CPR

 

pre-tracer

arterial

 

blood

 

samples

 

were

 

taken.

 

The

 

volume

 

of

 

solution

 

utilized

was

 

1.5

 

ml

 

followed

 

by

 

a

 

1.0

 

ml

 

of

 

saline

 

flush.

2.3.

 

Tracers

Evans

 

blue

 

(EB)

 

(Sigma–Aldrich,

 

St.

 

Louis,

 

MO)

 

5.0

 

mg/ml,

 

and

indocyanine

 

green

 

(ICG)

 

(Alkorn,

 

Buffalo

 

Grove,

 

IL)

 

2.5

 

mg/ml

 

were

used

 

randomly

 

in

 

each

 

site

 

for

 

the

 

consecutive

 

experiments

 

as

 

trac-

ers

 

to

 

determine

 

the

 

relative

 

arterial

 

appearance

 

times

 

and

 

dose

delivered

 

from

 

the

 

IO

 

and

 

central

 

venous

 

routes.

 

Both

 

ICG

 

and

 

EB

dyes

 

are

 

inert

 

and

 

have

 

no

 

known

 

biological

 

activity.

 

Each

 

bolus

 

of

tracer

 

contained

 

0.014

 

mg/kg

 

of

 

adrenaline

 

(epinephrine).

 

At

 

2-min

post

 

CPR

 

(0

 

time

 

point)

 

the

 

tracers

 

EB

 

and

 

ICG

 

were

 

co-administered

simultaneously

 

to

 

the

 

designated

 

two

 

paired

 

sites

 

in

 

Protocol

 

I

(sternal

 

IO

 

and

 

tibial

 

IO)

 

and

 

in

 

Protocol

 

II

 

(central

 

venous

 

IV

 

and

sternal

 

IO).

 

Rapid

 

injection

 

of

 

the

 

2–3

 

ml

 

of

 

tracer

 

solution

 

was

immediately

 

followed

 

by

 

a

 

1

 

ml

 

flush

 

to

 

clear

 

the

 

needle.

 

Arterial

blood

 

samples

 

were

 

taken

 

every

 

10-s

 

for

 

5

 

1/2

 

min

 

and

 

then

 

at

 

every

30-s

 

for

 

the

 

remainder

 

of

 

the

 

8-min

 

time

 

period.

 

After

 

completion

of

 

the

 

study

 

CPR

 

was

 

stopped

 

and

 

the

 

animal

 

was

 

euthanized

 

with

a

 

high

 

dose

 

of

 

ketamine

 

and

 

KCl.

Plasma

 

tracer

 

concentrations

 

in

 

arterial

 

blood

 

were

 

determined

spectrophotometrically

 

(Beckman

 

Coulter

 

DU

 

800

 

spectropho-

tometer,

 

Brea,

 

CA)

 

using

 

absorbance

 

wavelengths

 

of

 

805

 

nm

 

for

 

ICG

and

 

620

 

nm

 

for

 

EB.

 

Calibration

 

standards

 

of

 

EB

 

and

 

ICG

 

were

 

pre-

pared

 

in

 

plasma

 

and

 

used

 

to

 

calculate

 

the

 

concentrations

 

of

 

EB

 

and

ICG

 

from

 

arterial

 

blood

 

samples.

 

The

 

area

 

under

 

the

 

curve

 

(AUC)

 

of

arterial

 

tracer

 

concentration

 

divided

 

by

 

the

 

tracer

 

dose

 

was

 

used

 

as

a

 

measure

 

of

 

the

 

drug

 

delivered

 

to

 

the

 

systemic

 

circulation

 

during

the

 

first

 

8

 

min

 

after

 

drug

 

injection

 

(0–480

 

s).

 

The

 

ratio

 

of

 

the

 

AUC

for

 

both

 

tracers

 

was

 

used

 

as

 

a

 

measure

 

of

 

the

 

relative

 

drug

 

delivery.

2.4.

 

Statistics

Summary

 

data

 

are

 

expressed

 

as

 

means

 

±

 

standard

 

error

 

of

 

the

mean

 

(SEM).

 

To

 

test

 

for

 

differences

 

of

 

appearance

 

times

 

a

 

paired

Student’s

 

t-test

 

was

 

conducted.

 

Correlation

 

coefficients

 

for

 

the

 

rela-

tionship

 

of

 

mean

 

arterial

 

pressure

 

(MAP)

 

to

 

appearance

 

time

 

were

calculated

 

utilizing

 

Sigma

 

plot

 

software

 

(Systat

 

Software

 

Inc.,

 

Ver-

sion

 

11,

 

San

 

Jose,

 

CA).

 

A

 

two-sided

 

alpha

 

level

 

of

 

significance

 

of

<0.05

 

was

 

used

 

for

 

assessing

 

statistical

 

significance.

3.

 

Results

Data

 

on

 

appearance

 

time

 

and

 

dose

 

delivered

 

for

 

all

 

individual

animals

 

and

 

groups

 

are

 

presented

 

in

 

figures

 

and

 

tables.

3.1.

 

Appearance

 

times

Fig.

 

1

(A

 

and

 

C)

 

and

 

Table

 

1

 

display

 

data

 

for

 

each

 

experiment

of

 

appearance

 

times

 

calculated

 

in

 

seconds,

 

between

 

injection

 

and

time

 

to

 

peak

 

tracer

 

concentration,

 

in

 

Protocol

 

I—sternal

 

IO

 

and

 

tibial

IO

 

injections

 

(n

 

=

 

7).

 

Mean

 

time

 

to

 

maximum

 

concentration

 

was

53

 

±

 

11

 

s

 

for

 

the

 

sternal

 

injection

 

compared

 

to

 

107

 

±

 

27

 

s

 

the

 

tibial

injection.

 

The

 

range

 

was

 

from

 

20

 

to

 

90

 

s

 

and

 

40

 

to

 

240

 

s

 

for

 

the

sternal

 

and

 

tibial

 

routes,

 

respectively

 

(p

 

=

 

0.03).

 

Time

 

to

 

half

 

(50%)

S.L.

 

Hoskins

 

et

 

al.

 

/

 

Resuscitation

 

83 (2012) 107–

 

112

109

Fig.

 

1.

 

The

 

two

 

upper

 

graphs

 

show

 

appearance

 

times

 

of

 

tracers

 

vs.

 

time:

 

Protocol-I

 

(tibial

 

IO

 

vs.

 

the

 

sternal

 

IO):

 

appearance

 

times

 

of

 

tracers

 

tibia

 

(Graph-A)

 

vs.

 

sternum

(Graph-C).

 

Concentrations

 

were

 

normalized

 

in

 

this

 

figure

 

to

 

the

 

maximal

 

concentration

 

in

 

order

 

to

 

better

 

visualize

 

time

 

differences

 

to

 

peak

 

concentration.

 

The

 

two

 

lower

graphs

 

show

 

dose

 

delivered

 

to

 

the

 

arterial

 

blood

 

calculated

 

as

 

dose

 

injected

 

(mg)

 

by

 

aortic

 

blood

 

concentration

 

(

␮g/ml)

 

for

 

the

 

same

 

protocol

 

tibia

 

(Graph-B)

 

and

 

sternum

(Graph-D).

maximum

 

concentration

 

was

 

22

 

±

 

3

 

s

 

using

 

the

 

sternal

 

route

 

and

50

 

±

 

8

 

s

 

for

 

the

 

tibial

 

route

 

(p

 

=

 

0.006).

Fig.

 

2

(A

 

and

 

C)

 

and

 

Table

 

2

 

show

 

the

 

appearance

 

times

 

of

 

trac-

ers

 

for

 

Protocol

 

II,

 

sternal

 

IO

 

and

 

central

 

venous

 

IV

 

injections

 

(n

 

=

 

6).

Mean

 

peak

 

time

 

to

 

the

 

maximum

 

tracer

 

concentrations

 

after

 

simul-

taneous

 

injections,

 

via

 

IO

 

and

 

central

 

vein

 

were

 

not

 

significantly

different

 

97

 

±

 

17

 

s

 

and

 

70

 

±

 

12

 

s,

 

respectively

 

(p

 

=

 

0.17).

 

Times

 

for

tracers

 

to

 

reach

 

their

 

50%

 

maximal

 

concentrations

 

were

 

36

 

±

 

4

 

s

for

 

sternal

 

IO

 

and

 

30

 

±

 

4

 

s

 

for

 

the

 

central

 

vein

 

routes

 

(p

 

=

 

0.06).

3.2.

 

Dose

 

delivered

Dose

 

delivered

 

was

 

determined

 

by

 

using

 

an

 

area

 

under

 

the

 

curve

analysis

 

(AUC)

 

for

 

aortic

 

concentration

 

divided

 

by

 

injected

 

dose.

Fig.

 

1

(B

 

and

 

D)

 

and

 

Table

 

3

 

show

 

the

 

doses

 

of

 

tracer

 

delivery

 

to

the

 

aortic

 

blood,

 

for

 

each

 

animal

 

of

 

Protocol

 

I,

 

calculated

 

as

 

AUC.

The

 

ratio

 

of

 

the

 

AUC

 

between

 

Protocol

 

I

 

(tibial

 

IO

 

vs.

 

sternal

 

IO)

 

is

a

 

measure

 

of

 

the

 

relative

 

effectiveness

 

of

 

dose

 

delivery

 

via

 

the

 

two

routes.

 

The

 

tibial

 

IO

 

route

 

delivered

 

less

 

dose

 

to

 

the

 

arterial

 

blood

 

or

Fig.

 

2.

 

The

 

two

 

upper

 

graphs

 

show

 

appearance

 

times

 

of

 

tracers

 

vs.

 

time:

 

Protocol-II

 

(sternal

 

IO

 

vs.

 

central

 

venous

 

IV):

 

appearance

 

times

 

of

 

tracers

 

central

 

venous

 

(Graph-A)

vs.

 

sternum

 

(Graph-C).

 

Concentrations

 

were

 

normalized

 

in

 

this

 

figure

 

to

 

the

 

maximal

 

concentration

 

in

 

order

 

to

 

better

 

visualize

 

time

 

differences

 

to

 

peak

 

concentration.

 

The

two

 

lower

 

graphs

 

show

 

dose

 

delivered

 

to

 

the

 

arterial

 

blood

 

calculated

 

as

 

dose

 

injected

 

(mg)

 

by

 

aortic

 

blood

 

concentration

 

(

␮g/ml)

 

for

 

the

 

same

 

protocol

 

central

 

venous

(Graph-B)

 

and

 

sternum

 

(Graph-D).

110

S.L.

 

Hoskins

 

et

 

al.

 

/

 

Resuscitation

 

83 (2012) 107–

 

112

Table

 

1

Appearance

 

times

 

in

 

seconds

 

from

 

injection

 

to

 

maximum

 

tracer

 

concentrations

 

and

half

 

(50%)

 

maximal

 

concentration.

Tibial

 

IO

 

vs.

 

sternal

 

IO

 

injection

Animal

 

(n

 

=

 

7)

 

Peak

 

concentration

*

50%Peak

 

concentration

§

Sternum

 

Tibia

 

Sternum

 

Tibia

86

 

80

 

110

 

36

 

57

21

90

150

22

68

18

80

240

 

25

 

85

34

20

 

40

 

15

 

25

35

 

30

 

100

 

18

 

50

39

 

20

 

50

 

22

 

33

36

 

50

 

60

 

13

 

35

Mean

53

107

22

50

SEM

11

27

3

8

CI

30–75

 

55–158

 

16–27

 

34–65

CI,

 

confidence

 

interval

 

(confidence

 

level

 

=

 

95%);

 

SEM,

 

standard

 

error

 

of

 

the

 

mean.

*

p

 

=

 

0.03

 

–

 

peak

 

concentration

 

–

 

tibia

 

vs.

 

sternum.

§

p

 

=

 

0.006

 

–

 

50%

 

peak

 

concentration

 

–

 

tibia

 

vs.

 

sternum.

65

 

±

 

5%

 

as

 

compared

 

with

 

the

 

sternal

 

route,

 

mean

 

AUC’s

 

difference

was

 

statically

 

significant

 

(p

 

=

 

0.003).

Fig.

 

2

(B

 

and

 

D)

 

and

 

Table

 

4

 

show

 

the

 

actual

 

values

 

and

 

ratio

 

of

 

the

AUC

 

between

 

Protocol

 

II

 

(sternal

 

IO

 

vs.

 

central

 

venous

 

IV).

 

The

 

ster-

nal

 

IO

 

route

 

was

 

86

 

±

 

10%

 

as

 

effective

 

as

 

the

 

central

 

venous

 

route

in

 

tracer

 

delivery,

 

although

 

the

 

mean

 

AUCs

 

were

 

not

 

significantly

different

 

(p

 

=

 

0.22).

4.

 

Discussion

To

 

the

 

best

 

of

 

our

 

knowledge

 

the

 

present

 

study

 

is

 

the

 

first

 

to

 

use

a

 

double

 

tracer

 

technique

 

to

 

assess

 

effectiveness

 

of

 

simultaneous

drug

 

delivery,

 

during

 

CPR

 

into

 

two

 

IO

 

sites.

Overall

 

the

 

study

 

demonstrated

 

that

 

the

 

intraosseous

 

(IO)

 

route

is

 

an

 

effective

 

means

 

of

 

delivering

 

drugs

 

during

 

CPR

 

for

 

tibia

 

and

sternum

 

IO

 

sites.

Peripheral

 

IV

 

lines

 

are

 

the

 

most

 

commonly

 

used

 

routes

 

for

drug

 

delivery

 

by

 

EMS

 

personnel.

 

An

 

absence

 

of

 

venous

 

blood

 

flow

and

 

low

 

pressure

 

during

 

cardiac

 

arrest

 

can

 

lengthen

 

the

 

time

 

to

obtain

 

peripheral

 

IV

 

access

 

and

 

delay

 

critically

 

needed

 

drug

 

ther-

apy.

 

Experienced

 

medics

 

can

 

achieve

 

IV

 

access

 

rapidly

 

under

 

ideal

conditions.

 

However,

 

prehospital

 

conditions

 

in

 

the

 

field

 

transport

to

 

hospital,

 

and

 

the

 

skill

 

levels

 

of

 

medics

 

can

 

vary

 

widely.

 

Clini-

cal

 

studies

 

have

 

shown

 

that

 

peripheral

 

IV

 

access

 

times

 

can

 

range

from

 

2

 

to

 

49

 

min.

6–8,15

The

 

success

 

rate

 

for

 

establishing

 

periph-

eral

 

IV

 

access

 

after

 

cardiac

 

arrest

 

and

 

difficult

 

IV

 

is

 

variable

 

and

ranges

 

broadly

 

between

 

30

 

and

 

75%

 

in

 

adult

6–8

patients,

 

with

 

lower

Table

 

2

Appearance

 

times

 

in

 

seconds

 

from

 

injection

 

to

 

maximum

 

tracer

 

concentrations

 

and

half

 

(50%)

 

maximal

 

concentration.

Sternal

 

IO

 

vs.

 

central

 

venous

 

IV

 

injection

Animal

 

(n

 

=

 

6)

 

Peak

 

concentration

 

50%Peak

 

concentration

Sternum

 

IV

 

Sternum

 

IV

87

 

100

 

50

 

36.4

 

24

89

 

70

 

50

 

34

 

23

105

 

60

 

50

 

29

 

28

95

110

110

52

 

48

110

 

70

 

90

 

28

 

27

92

 

170

 

110

 

38

 

36

Mean

 

97

 

70

 

36

 

30

SEM

 

17

 

12

 

4

 

4

CI

 

64–129

 

45–94

 

28–42

 

22–37

p

 

=

 

0.17

 

–

 

peak

 

concentration

 

–

 

sternum

 

vs.

 

central

 

venous

 

infusion.

p

 

=

 

0.06

 

– 50%

 

peak

 

concentration

 

–

 

sternum

 

vs.

 

central

 

venous

 

infusion.

CI,

 

confidence

 

interval

 

(confidence

 

level

 

=

 

95%);

 

SEM,

 

standard

 

error

 

of

 

the

 

mean.

Table

 

3

Dose

 

delivered

 

for

 

tibial

 

vs.

 

sternal

 

IO

 

injections

 

calculated

 

as

 

area

 

under

 

the

 

curve

for

 

aortic

 

concentration

 

␮g/ml

 

divided

 

by

 

dose

 

injected

 

(mg)

 

over

 

480

 

s

 

after

 

injec-

tion.

 

The

 

relative

 

effectiveness

 

of

 

the

 

two

 

routes

 

is

 

shown

 

as

 

a

 

ratio

 

of

 

the

 

area

 

under

the

 

curve

 

(AUC),

 

tibial

 

IO

 

divided

 

by

 

sternal

 

IO.

Relative

 

dose

 

delivered

 

of

 

tracers

 

(Tibial

 

IO

 

vs.

 

sternal

 

IO

 

injection—AUC

0–480

 

s

)

Animal

 

AUC

*

(

␮g

 

s/ml)

 

Ratio

Sternum

Tibia

Tibia/sternum

21

912

 

450

 

0.49

18

 

776

 

382

 

0.49

34

 

601

 

400

 

0.67

35

 

645

 

368

 

0.57

39

 

509

 

423

 

0.83

36

511

418

0.82

86

783

545

0.70

Mean

 

677

 

427

 

0.65

SEM

57

 

22

 

0.05

CI

 

564–789

 

383–470

 

0.6–0.7

CI,

 

confidence

 

interval

 

(confidence

 

level

 

=

 

95%);

 

SEM,

 

standard

 

error

 

of

 

the

 

mean.

*

p

 

=

 

0.003

 

– comparison

 

between

 

AUC

0–480

– tibia

 

vs.

 

sternum.

success

 

rates

 

for

 

the

 

pediatric

 

patient

 

population

 

18–65%.

16,17

A

prospective

 

study

 

of

 

successful

 

prehospital

 

IV

 

placement

 

in

 

583

patients

 

showed

 

that

 

the

 

success

 

rate

 

at

 

first

 

attempt

 

was

 

74%

 

(368

patients).

6

Physicians

 

have

 

long

 

sought

 

alternate

 

routes

 

for

 

the

 

rapid

administration

 

of

 

drugs

 

during

 

cardiac

 

emergencies,

 

circulatory

shock,

 

and

 

low

 

flow

 

states.

 

The

 

endotracheal

 

route

 

is

 

often

 

used

as

 

a

 

convenient

 

and

 

rapid

 

alternative

 

for

 

IV

 

delivery

 

of

 

selected

drugs.

 

However,

 

efficacy

 

of

 

endotracheal

 

delivery

 

of

 

drugs

 

is

controversial.

18,19

The

 

IO

 

route

 

provides

 

access

 

to

 

systemic

 

circu-

lation

 

via

 

the

 

bone

 

marrow

 

cavity

 

which

 

provides

 

a

 

noncollapsible

delivery

 

point

 

into

 

the

 

central

 

circulation

 

for

 

emergency

 

infu-

sions

 

and

 

for

 

drug

 

delivery

 

in

 

the

 

operation

 

room

 

setting.

20

Current

 

American

 

Heart

 

Association

 

guidelines

 

and

 

the

 

Interna-

tional

 

Resuscitation

 

Council

 

Guidelines

 

recommend

 

the

 

IO

 

route

as

 

first

 

vascular

 

access

 

in

 

pediatric

 

emergencies

 

such

 

a

 

cardiac

arrest.

13–21

For

 

adult

 

cardiac

 

arrest

 

IO

 

is

 

the

 

first

 

alternative

 

when

intravenous

 

access

 

is

 

delayed

 

or

 

impossible.

13,22

The

 

success

 

rate

when

 

IO

 

access

 

is

 

used

 

is

 

81–100%

8,10,11

and

 

the

 

time

 

to

 

establish

a

 

IO

 

line

 

varies

 

between

 

20

 

s

 

and

 

1.5

 

min.

8,10,23

The

 

most

 

common

adverse

 

effect

 

associated

 

with

 

IO

 

infusion

 

is

 

extravasation

 

and

 

this

complication

 

has

 

been

 

reported

 

in

 

12%

 

of

 

patients.

24

Compartment

syndrome,

 

osteomyelitis,

 

and

 

tibial

 

fracture

 

are

 

rare,

 

but

 

have

 

been

reported.

9,24,25

Table

 

4

Dose

 

delivered

 

for

 

sternal

 

IO

 

versus

 

central

 

venous

 

IV

 

injections

 

calculated

 

as

 

area

under

 

the

 

curve

 

for

 

aortic

 

concentration

 

␮g/ml

 

divided

 

by

 

dose

 

injected

 

(mg)

 

over

480

 

seconds

 

after

 

injection.

 

The

 

relative

 

effectiveness

 

of

 

the

 

two

 

routes

 

is

 

shown

 

as

a

 

ratio

 

of

 

the

 

area

 

under

 

the

 

curve

 

(AUC),

 

sternal

 

IO

 

divided

 

by

 

central

 

venous

 

IV.

Relative

 

dose

 

delivered

 

of

 

tracers

 

(sternal

 

IO

 

vs.

 

central

 

venous

 

IV

injection—AUC

0–480

 

s

)

Animal

 

AUC

 

␮g

 

s/ml

 

Ratio

IV

 

Sternum

 

Sternum/IV

89

 

694

 

589

 

0.85

105

 

855

 

939

 

1.10

95

 

879

 

805

 

0.92

110

 

854

 

783

 

0.92

92

 

956

 

923

 

0.97

87

934

 

385

 

0.41

Mean

 

862

 

737

 

0.86

SEM

 

38

 

87

 

0.10

CI

788–935

 

566–907

 

0.7–1.0

p

 

=

 

0.22

 

– comparison

 

between

 

AUC

0–480

–

 

sternum

 

vs.

 

central

 

venous

 

infusion.

CI,

 

confidence

 

interval

 

(confidence

 

level

 

=

 

95%);

 

SEM,

 

standard

 

error

 

of

 

the

 

mean.

S.L.

 

Hoskins

 

et

 

al.

 

/

 

Resuscitation

 

83 (2012) 107–

 

112

111

Voelckel

 

et

 

al.

 

showed

 

that

 

bone

 

marrow

 

blood

 

flow

 

was

reduced

 

by

 

70–80%

 

after

 

hemorrhage.

26

During

 

CPR

 

the

 

bone

 

mar-

row

 

flow

 

is

 

expected

 

to

 

be

 

lower

 

than

 

in

 

hemorrhagic

 

shock.

 

Sato

et

 

al.

 

and

 

Del

 

Guercio

 

et

 

al.

 

showed

 

in

 

dogs

 

and

 

humans,

 

respec-

tively

 

that

 

during

 

CPR

 

the

 

cardiac

 

output

 

is

 

only

 

approximately

20–30%

 

of

 

normal.

27,28

In

 

our

 

study

 

mean

 

aortic

 

appearance

 

times

to

 

the

 

peak

 

concentration

 

of

 

the

 

tracer

 

was

 

97

 

±

 

17

 

s

 

for

 

the

 

sternal

IO

 

route

 

which

 

was

 

not

 

statistically

 

significant

 

(p

 

=

 

0.17)

 

compared

to

 

70

 

±

 

12

 

s

 

for

 

central

 

venous

 

route.

 

Barsan

 

et

 

al.

 

showed

 

similar

result

 

in

 

dogs

 

with

 

mean

 

time

 

to

 

peak

 

times

 

for

 

central

 

venous

infusion

 

of

 

84

 

s

 

with

 

range

 

between

 

53

 

and

 

100

 

s.

29

Kuhn

 

et

 

al.

showed

 

that

 

the

 

peak

 

concentration

 

of

 

dye

 

obtained

 

with

 

central

venous

 

injection

 

of

 

indocyanine

 

green

 

during

 

CPR

 

in

 

humans

 

was

at

 

30

 

s.

 

However,

 

only

 

three

 

patients

 

were

 

included

 

on

 

the

 

study.

30

Emerman

 

et

 

al.

 

demonstrated

 

in

 

dogs

 

that

 

the

 

interval

 

of

 

central

venous

 

injection

 

to

 

first

 

appearance

 

of

 

the

 

indocyanine

 

green

 

dur-

ing

 

CPR

 

was

 

37

 

±

 

17

 

s.

31

Zuercher

 

et

 

al.

 

showed

 

mean

 

time

 

from

adrenaline

 

injection

 

to

 

peak

 

coronary

 

perfusion

 

of

 

60

 

±

 

6

 

s

 

when

the

 

drug

 

was

 

delivered

 

via

 

IO

 

vs.

 

43

 

±

 

4

 

after

 

IV

 

injection

 

during

CPR.

32

These

 

results

 

are

 

similar

 

to

 

our

 

finding

 

of

 

time

 

to

 

the

 

50%

peak

 

concentration,

 

i.e.

 

central

 

venous

 

(30

 

s),

 

sternal

 

(22

 

s—Protocol

I;

 

36

 

s—Protocol

 

II),

 

and

 

tibia

 

(50

 

s).

Some

 

factors

 

can

 

affect

 

the

 

appearance

 

times

 

and

 

the

 

dose

delivery

 

in

 

this

 

study.

 

One

 

is

 

that

 

sternum

 

is

 

located

 

closer

 

to

the

 

central

 

circulation

 

when

 

compared

 

with

 

the

 

tibia

 

location,

which

 

may

 

facilitate

 

the

 

faster

 

appearance

 

of

 

the

 

drug

 

on

 

the

systemic

 

circulation

 

when

 

the

 

drug

 

is

 

delivered

 

into

 

the

 

ster-

num.

 

Second,

 

there

 

is

 

a

 

difference

 

of

 

blood

 

perfusion

 

between

the

 

two

 

bones.

 

It

 

is

 

likely

 

that

 

the

 

sternum

 

perfusion

 

is

 

better

than

 

the

 

tibia

 

perfusion

 

and

 

this

 

may

 

facilitate

 

the

 

absorption

 

of

the

 

drug

 

to

 

the

 

systemic

 

circulation.

 

Gross

 

et

 

al.

 

showed

 

a

 

wide

heterogeneity

 

of

 

bone

 

blood

 

flow

 

comparing

 

hematopoietic

 

can-

cellous

 

bones

 

(red

 

marrow)

 

such

 

as

 

sternum,

 

rib,

 

ilium,

 

and

 

femur

epiphysis

 

(24

 

ml

 

min

−1

100

 

g

−1

)

 

vs.

 

nonhematopoietic

 

bones

 

(yel-

low

 

marrow)

 

such

 

as

 

tibia

 

and

 

mandible

 

(2

 

ml

 

min

−1

100

 

g

−1

).

 

The

authors

 

also

 

described

 

a

 

significant

 

decrease

 

in

 

blood

 

flow

 

and

an

 

increase

 

in

 

vascular

 

resistance

 

in

 

bone

 

during

 

hemorrhagic

hypotension.

33

A

 

key

 

point

 

during

 

the

 

CPR

 

maneuvers

 

is

 

the

 

quality

 

of

 

the

chest

 

compressions.

 

To

 

give

 

effective

 

chest

 

compression

 

is

 

impor-

tant

 

that

 

the

 

rescuers

 

or

 

the

 

devices

 

used

 

to

 

perform

 

the

 

CPR

 

push

hard

 

(

≥5

 

cm)

 

and

 

fast

 

(

≥100/min).

22

The

 

chest

 

should

 

be

 

allowed

 

to

recoil

 

freely

 

after

 

each

 

compression.

 

Besides,

 

approximately

 

equal

compressions

 

and

 

relaxation

 

times

 

should

 

be

 

used

 

and

 

interrup-

tions

 

in

 

chest

 

compressions

 

should

 

be

 

minimized.

 

If

 

these

 

chest

compressions

 

are

 

not

 

effective

 

all

 

the

 

circulatory

 

blood

 

flow

 

can

 

be

affected

 

including

 

the

 

bone

 

marrow

 

flow.

22,34

Any

 

anatomic

 

dif-

ference

 

between

 

the

 

animals

 

or

 

any

 

other

 

factor

 

that

 

impair

 

the

dynamic

 

of

 

the

 

chest

 

compressions

 

might

 

result

 

in

 

differences

 

in

cardiac

 

output

 

during

 

this

 

period,

 

which

 

might

 

consequently

 

delay

the

 

appearance

 

time

 

of

 

tracers

 

on

 

the

 

systemic

 

circulation.

The

 

dose

 

delivered

 

of

 

tracer

 

via

 

the

 

IO

 

route

 

was

 

similar

 

to

 

that

delivered

 

by

 

central

 

venous

 

route.

 

The

 

sternal

 

IO

 

route

 

delivered

86%

 

of

 

the

 

tracer

 

to

 

the

 

aorta

 

compared

 

with

 

central

 

vein

 

drug

 

deliv-

ery.

 

However,

 

in

 

one

 

animal,

 

the

 

ratio

 

between

 

sternum/central

venous

 

infusions

 

was

 

0.41

 

(

Table

 

4

).

 

When

 

we

 

exclude

 

this

 

outlier

data

 

point

 

from

 

the

 

analysis,

 

the

 

resultant

 

sternum

 

dose

 

delivered

via

 

the

 

route

 

was

 

95%

 

that

 

of

 

the

 

central

 

venous.

 

The

 

effectiveness

 

of

the

 

IO

 

sternal

 

route

 

for

 

drug

 

delivery

 

during

 

CPR

 

may

 

be

 

due

 

to

 

one

or

 

more

 

factors.

 

The

 

red

 

bone

 

marrow

 

of

 

the

 

sternum

 

could

 

pro-

vide

 

sufficient

 

blood

 

flow

 

for

 

rapid

 

delivery

 

of

 

drugs

 

to

 

the

 

great

veins.

 

Further,

 

chest

 

compressions

 

may

 

facilitate

 

the

 

drug

 

egress

out

 

of

 

the

 

marrow

 

and

 

into

 

the

 

vasculature.

35

Alternatively,

 

the

IO

 

delivery

 

of

 

tracer

 

may

 

be

 

independent

 

of

 

marrow

 

blood

 

flow.

It

 

may

 

be

 

that

 

a

 

1.5

 

ml

 

bolus

 

of

 

tracer

 

followed

 

by

 

the

 

1

 

ml

 

flush

used

 

in

 

our

 

study

 

is

 

sufficient

 

volume

 

to

 

advance

 

most

 

of

 

the

 

tracer

through

 

the

 

marrow,

 

out

 

of

 

the

 

injection

 

site

 

and

 

into

 

the

 

venous

circulation.

The

 

mean

 

dose

 

delivered

 

via

 

the

 

tibial

 

route

 

was

 

65%

 

and

 

53%

of

 

the

 

drug

 

delivery

 

via

 

the

 

sternum

 

and

 

central

 

venous

 

route,

respectively.

 

However,

 

even

 

for

 

the

 

tibial

 

route

 

the

 

half

 

maxi-

mal

 

concentrations

 

were

 

achieved

 

in

 

less

 

than

 

1

 

min.

 

Andropoulos

et

 

al.

 

used

 

HPLC

 

analysis

 

for

 

the

 

determination

 

of

 

tibial

 

adrenaline

delivery

 

during

 

CPR

 

in

 

lambs.

 

The

 

authors

 

determined

 

that

 

the

maximum

 

arterial

 

plasma

 

adrenaline

 

concentrations

 

were

 

similar

between

 

central

 

venous

 

and

 

tibial

 

IO

 

delivery.

 

However,

 

they

 

noted

reduced

 

appearance

 

time,

 

after

 

central

 

venous

 

administration

 

com-

pared

 

to

 

tibial

 

IO

 

injection

 

after

 

adrenaline

 

injection.

36

Our

 

measurements

 

of

 

appearance

 

times

 

and

 

doses

 

delivered,

coupled

 

with

 

an

 

additional

 

one

 

or

 

more

 

minutes

 

for

 

establishing

a

 

peripheral

 

IV,

 

suggest

 

that

 

even

 

when

 

using

 

the

 

slower

 

tibial

IO

 

route,

 

one

 

would

 

effectively

 

deliver

 

drugs

 

into

 

the

 

arterial

 

cir-

culation

 

during

 

CPR

 

in

 

a

 

shorter

 

time

 

than

 

the

 

time

 

needed

 

to

successfully

 

start

 

a

 

peripheral

 

IV.

 

As

 

such,

 

the

 

tibial

 

IO

 

route

 

is

both

 

an

 

efficacious

 

and

 

rapid

 

means

 

of

 

delivering

 

drug

 

therapy

 

dur-

ing

 

CPR.

 

The

 

size

 

of

 

the

 

saline

 

bolus

 

after

 

the

 

drug

 

infusion

 

may

also

 

have

 

an

 

important

 

role

 

on

 

the

 

time

 

for

 

maximum

 

concentra-

tion

 

of

 

the

 

dye.

 

If

 

we

 

had

 

used

 

a

 

larger

 

flush

 

the

 

effectiveness

 

of

the

 

IO

 

tibial

 

delivery

 

may

 

have

 

increased.

 

Wenzel

 

et

 

al.

 

demon-

strated

 

comparable

 

vasopressin

 

plasma

 

level

 

and

 

hemodynamic

variables

 

when

 

the

 

drug

 

was

 

delivered

 

both

 

by

 

the

 

intravenous

and

 

the

 

tibial

 

IO

 

routes

 

during

 

CPR.

 

However,

 

the

 

authors

 

infused

20

 

ml

 

of

 

saline

 

bolus

 

compared

 

with

 

1.0

 

ml

 

used

 

in

 

the

 

present

study.

37

Based

 

on

 

the

 

present

 

data,

 

we

 

recommend

 

that

 

sternal

 

IO

 

route

be

 

considered

 

as

 

the

 

first

 

choice

 

of

 

drug

 

delivery

 

during

 

CPR

 

when

IV

 

access

 

has

 

not

 

been

 

established,

 

and

 

that

 

the

 

tibial

 

IO

 

route

is

 

also

 

justified

 

as

 

second

 

choice.

 

The

 

practical

 

choices

 

of

 

which

route

 

to

 

use

 

in

 

adults

 

also

 

depend

 

on

 

which

 

IO

 

devices

 

are

 

avail-

able.

 

There

 

are

 

currently

 

6

 

adult

 

IO

 

devices

 

allowed

 

for

 

marketing

by

 

the

 

Food

 

and

 

Drug

 

Administration

 

(FDA).

 

This

 

includes

 

two

 

IO

devices

 

for

 

adult

 

sternal

 

access

 

(FAST1

 

(Pyng

 

Medical

 

Corp.,

 

Rich-

mond,

 

BC,

 

Canada)

 

and

 

Sternal

 

EZ-IO

 

(Vidacare

 

Corp.,

 

San

 

Antonio,

TX))

 

and

 

four

 

IO

 

devices

 

for

 

tibial

 

access

 

(SurFast

 

(Cook

 

Criti-

cal

 

Care,

 

Bloomington,

 

IN),

 

Jamishidi

 

(Baxter

 

Allegiance,

 

McGraw

Park),

 

Bone

 

Injection

 

Gun

 

(B.I.G.,

 

Waismed,

 

Houston,

 

TX),

 

EZ-IO

(Vidacare

 

Corp.,

 

San

 

Antonio,

 

TX)).

9,38,39

In

 

pediatric

 

patients,

 

stan-

dard

 

butterfly

 

needle,

 

spinal

 

needle,

 

and

 

pediatric

 

versions

 

of

 

adult

IO

 

needles

 

can

 

be

 

used.

 

Most

 

recently

 

the

 

humerus

 

has

 

been

 

sug-

gested

 

as

 

a

 

route

 

for

 

IO

 

delivery.

 

Further

 

work

 

will

 

be

 

required

 

to

assess

 

the

 

relative

 

success

 

of

 

this

 

route

 

vs.

 

the

 

sternal

 

and

 

the

 

tibial

route.

There

 

are

 

limitations

 

to

 

our

 

study.

 

First,

 

swine

 

are

 

not

 

humans

and

 

conclusive

 

extrapolation

 

to

 

human

 

patient

 

responses

 

cannot

be

 

made.

 

The

 

shape

 

of

 

the

 

pig

 

thorax

 

is

 

different

 

from

 

the

 

human

thorax.

 

In

 

pigs,

 

the

 

ventricles

 

are

 

positioned

 

in

 

the

 

center

 

of

 

the

 

tho-

racic

 

cavity,

 

surrounded

 

by

 

lung

 

tissues

 

on

 

all

 

sides.

 

In

 

humans,

 

the

right

 

ventricle

 

is

 

positioned

 

just

 

under

 

the

 

sternum.

 

This

 

anatomic

difference

 

makes

 

it

 

more

 

difficult

 

to

 

get

 

a

 

compression

 

effect

 

on

the

 

heart

 

of

 

pigs.

 

Chest

 

compressions

 

in

 

pigs

 

increase

 

intratho-

racic

 

pressure

 

(thoracic

 

pump

 

mechanism),

 

which

 

in

 

turns

 

affects

the

 

heart.

 

In

 

humans

 

we

 

have

 

not

 

only

 

the

 

thoracic

 

pump

 

effect

but

 

also

 

the

 

direct

 

heart

 

pump

 

mechanism

 

affecting

 

the

 

heart

 

by

chest

 

compression.

34

Moreover,

 

we

 

did

 

not

 

measure

 

the

 

plasma

concentrations

 

of

 

adrenaline.

 

We

 

used

 

dye

 

tracers

 

as

 

a

 

surrogate

of

 

drug

 

delivery

 

in

 

place

 

of

 

the

 

biologically

 

active

 

drug.

 

However,

measurement

 

of

 

adrenaline

 

would

 

preclude

 

comparison

 

of

 

simulta-

neous

 

injections.

 

The

 

significant

 

variability

 

of

 

cardiac

 

output

 

during

CPR

 

results

 

in

 

an

 

animal

 

to

 

animal

 

variability

 

of

 

time

 

to

 

peak

 

con-

centration

 

and

 

dose

 

delivered;

 

while

 

simultaneous

 

2

 

tracer

 

paired

studies

 

provides

 

for

 

great

 

precision

 

for

 

comparing

 

differences.

 

Fur-

ther,

 

high

 

background

 

levels

 

of

 

endogenous

 

adrenaline

 

during

 

CPR