Resuscitation

83 (2012) 63–

69

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

Clinical

Paper

Impact

of

resuscitation

system

errors

on

survival

from

in-hospital

cardiac

arrest

夽

Joseph

P.

Ornato

a

,

∗

, Mary

Ann

Peberdy

a

,

b

,

Renee

D.

Reid

a

,

V.

Ramana

Feeser

a

,

Harinder

S.

Dhindsa

a

,

for

the

NRCPR

Investigators

c

a

Department

of

Emergency

Medicine,

Virginia

Commonwealth

University,

Richmond,

VA,

United

States

b

Department

of

Internal

Medicine

&

Emergency

Medicine,

Virginia

Commonwealth

University,

Richmond,

VA,

United

States

a

r

t

i

c

l

e

i

n

f

o

Article

history:

Received

14

May

2011

Received

in

revised

form

9

September

2011

Accepted

11

September

2011

Keywords:
Inhospital
Resuscitation
Quality
Outcomes
Errors

a

b

s

t

r

a

c

t

Background:

An

estimated

350,000–750,000

adult,

in-hospital

cardiac

arrest

(IHCA)

events

occur

annually

in

the

United

States.

The

impact

of

resuscitation

system

errors

on

survival

during

IHCA

resuscitation

has

not

been

evaluated.

The

purpose

of

this

paper

was

to

evaluate

the

impact

of

resuscitation

system

errors

on

survival

to

hospital

discharge

after

IHCA.

Methods

and

results:

We

evaluated

subjective

and

objective

errors

in

118,387

consecutive,

adult,

index

IHCA

cases

entered

into

the

Get

with

the

Guidelines

National

Registry

of

Cardiopulmonary

Resuscitation

database

from

January

1,

2000

through

August

26,

2008.

Cox

regression

analysis

was

used

to

determine

the

relationship

between

reported

resuscitation

system

errors

and

other

important

clinical

variables

and

the

hazard

ratio

for

death

prior

to

hospital

discharge.

Of

the

108,636

patients

whose

initial

IHCA

rhythm

was

recorded,

resuscitation

system

errors

were

committed

in

9,894/24,467

(40.4%)

of

those

with

an

ini-

tial

rhythm

of

ventricular

fibrillation

or

pulseless

ventricular

tachycardia

(VF/pVT)

and

in

22,599/84,169

(26.8%)

of

those

with

non-VF/pVT.

The

most

frequent

system

errors

related

to

delay

in

medication

admin-

istration

(>5

min

time

from

event

recognition

to

first

dose

of

a

vasoconstrictor),

defibrillation,

airway

management,

and

chest

compression

performance

errors.

The

presence

of

documented

resuscitation

system

errors

on

an

IHCA

event

was

associated

with

decreased

rates

of

return

of

spontaneous

circu-

lation,

survival

to

24

h,

and

survival

to

hospital

discharge.

The

relative

risk

of

death

prior

to

hospital

discharge

based

on

hazard

ratio

analysis

was

9.9%

(95%

CI

7.8,

12.0)

more

likely

for

patients

whose

ini-

tial

documented

rhythm

was

non-VF/pVT

when

resuscitation

system

errors

were

reported

compared

to

when

no

errors

were

reported.

It

was

34.2%

(95%

CI

29.5,

39.1)

more

likely

for

those

with

VF/pVT.

Conclusions:

The

presence

of

resuscitation

system

errors

that

are

evident

from

review

of

the

resuscitation

record

is

associated

with

decreased

survival

from

IHCA

in

adults.

Hospitals

should

target

the

training

of

first

responders

and

code

team

personnel

to

emphasize

the

importance

of

early

defibrillation,

early

use

of

vasoconstrictor

medication,

and

compliance

with

ACLS

protocols.

© 2011 Elsevier Ireland Ltd. All rights reserved.

1.

Introduction

The

Institutes

of

Medicine

(IOM)

landmark

publication

(“To

Err

is

Human”)

estimated

that

at

least

44,000

and

perhaps

as

many

as

98,000

Americans

die

in

hospitals

each

year

as

a

result

of

med-

ical

errors.

1

Although

the

magnitude

of

the

problem

has

been

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.2011.09.009

.

∗ Corresponding

author

at:

Department

of

Emergency

Medicine,

Virginia

Com-

monwealth

University,

1250

East

Marshall

Street

–

Main

Hospital

2nd

Floor,

Suite

500,

Richmond,

VA

23298-0401,

United

States.

Tel.:

+1

804

828

5250;

fax:

+1

804

828

8590.

E-mail

address:

[e-mail]

(J.P.

Ornato).

c

For

the

American

Heart

Association’s

Get

With

the

Guidelines

–

Resuscitation

(National

Registry

of

Cardiopulmonary

Resuscitation)

Investigators,

see

Appendix

A

.

questioned,

2

the

Canadian

Adverse

Events

(AE)

Study

confirms

an

alarming

frequency

of

in-hospital

AEs

(7.5

per

100

hospital

admissions;

95%

confidence

interval

[CI],

5.7–9.3),

36.9%

(95%

CI,

32.0–41.8%)

of

which

are

potentially

preventable.

Death

occurred

in

20.8%

(95%

CI,

7.8–33.8%)

of

cases.

The

American

Heart

Association

(AHA)

Get

with

the

Guidelines

National

Registry

of

Cardiopulmonary

Resuscitation

(NRCPR)

col-

lects

data

on

adult

and

pediatric

in-hospital

cardiac

arrest

(IHCA)

events

from

approximately

10%

of

hospitals

in

the

United

States.

3

From

this

registry,

NRCPR

investigators

have

documented

lower

survival

from

adult

in-hospital

cardiac

arrest

(1)

on

nights

and

weekends

likely

due,

at

least

in

part,

to

system

factors,

3

(2)

when

defibrillation

is

delayed

greater

than

2

min

in

patients

whose

ini-

tial

IHCA

rhythm

is

ventricular

fibrillation

or

pulseless

ventricular

tachycardia

(VF/pVT),

4

and

(3)

in

certain

hospital

locations.

5

The

purpose

of

this

paper

was

to

determine

whether

the

presence

of

resuscitation

system

errors

reported

to

NRCPR

are

0300-9572/$

–

see

front

matter ©

2011 Elsevier Ireland Ltd. All rights reserved.

doi:

10.1016/j.resuscitation.2011.09.009

64

J.P.

Ornato

et

al.

/

Resuscitation

83 (2012) 63–

69

associated

with

lower

likelihood

of

survival

in

adult

patients

who

experience

an

IHCA.

2.

Methods

2.1.

Data

collection

and

integrity

NRCPR

is

a

prospective,

observational,

multi-center

perfor-

mance

improvement

registry

of

IHCA

events.

Hospitals

join

voluntarily

and

pay

an

annual

fee

for

data

support

and

report

gen-

eration.

Hospital

medical

records

on

sequential

IHCA

events

are

abstracted

by

trained,

NRCPR-certified,

performance

improvement

personnel

at

each

participating

institution.

All

data

elements

have

standardized

definitions

allowing

aggregate

data

analysis

from

multiple

sites,

and

all

data

transfer

is

in

compliance

with

the

Health

Insurance

Portability

and

Accountability

Act.

Oversight

for

opera-

tions

is

provided

by

the

American

Heart

Association,

a

scientific

advisory

board,

and

an

executive

database

steering

committee.

Multiple

efforts

are

taken

to

assure

data

integrity,

including

data

abstractor

certification

prior

to

allowing

data

entry,

over

300

soft-

ware

checks

and

smart

skips

to

assist

with

accurate

data

entry,

and

ongoing

abstractor

training

with

monthly

user’s

group

calls

and

an

annual

user’s

group

conference.

Even

though

the

most

challeng-

ing

data

points

to

collect

during

resuscitation

are

event

times

and

intervals,

these

are

documented

in

a

high

percentage

of

cases

in

NRCPR

hospitals.

For

example,

in

this

analysis

the

time

intervals

from

IHCA

onset

to

start

of

CPR

and

first

vasopressor

administra-

tion

were

captured

in

89%

and

76%,

respectively.

Further

details

of

the

NRCPR

database

and

data

integrity

can

be

found

in

previous

publications.

6,7

2.2.

Study

outcomes

The

primary

study

outcome

was

survival

to

hospital

discharge.

Secondary

outcomes

were

return

of

spontaneous

circulation

(ROSC)

and

survival

for

24

h

after

IHCA.

2.3.

Inclusion/exclusion

criteria

The

current

analysis

includes

all

consecutive,

adult

(age

≥18

years),

initial,

pulseless

IHCA

events

entered

from

549

dif-

ferent

hospitals

from

January

1,

2000

through

August

26,

2008.

All

adults

(

≥18

years

of

age)

who

experienced

an

in-hospital

resuscitation

event

and

who

had

documentation

of

initial

heart

rhythm

were

eligible

for

inclusion.

An

event

is

defined

as:

(1)

cardiopulmonary

arrest

requiring

chest

compressions

and/or

defib-

rillation,

or

(2)

acute

respiratory

compromise

requiring

emergency

assisted

ventilation

leading

to

cardiopulmonary

arrest

requiring

chest

compressions

and/or

defibrillation.

All

events

must

also

elicit

a

resuscitation

response

by

facility

personnel

and

have

a

resusci-

tation

record

completed.

Events

are

excluded

if

the

arrest

begins

outside

of

the

hospital,

is

limited

to

a

shock

delivered

by

an

implanted

cardioverter-defibrillator

(ICD),

or

occurs

on

a

patient

with

a

pre-existing

do

not

attempt

resuscitation

order.

For

patients

having

multiple

IHCA

events

during

the

same

hospitalization,

only

the

first

event

was

analyzed.

2.4.

Categorization

and

counting

of

resuscitation

system

error

types

The

NRCPR

database

contains

a

section

for

the

data

abstractor

at

each

hospital

site

to

capture

“self-reported”

system

errors

that

were

noted

during

or

following

the

resuscitation

effort

on

the

code

record

and/or

hospital

chart

or

to

document

errors

reported

by

resuscitation

members.

Types

of

error

categories

included

alert-

ing

hospital-wide

resuscitation

response,

airway

management,

vascular

access,

chest

compression,

defibrillation,

medications,

leadership,

protocol

deviation,

and

equipment

function

issues,

with

specific

subcategories

listed

in

Table

1

.

Each

type

of

individual

error

had

a

checkbox

that

was

to

be

filled

in

by

the

data

abstractor

if

supported

by

the

code

record

and/or

hospital

chart

documenta-

tion,

or

conveyed

by

personnel

on

scene.

For

analysis

of

this

data,

one

point

was

assigned

for

each

individual

error

box

checked

on

a

given

IHCA

event.

We

totaled

the

number

of

cases

with

individual

system

errors

and

created

subtotals

for

the

various

types

of

errors

to

calculate

the

number

of

cases

in

which

any

error

occurred

and

to

catalogue

and

quantify

the

specific

types

of

errors

that

occurred.

We

counted

and

added

one

point

on

each

case

for

each

NRCPR

“process

of

care

exception”

error

defined

as:

(1)

delay

of

>5

min

from

IHCA

event

recognition

to

the

first

dose

of

a

vasoconstric-

tor

(epinephrine

or

vasopressin)

on

events

that

had

a

duration

of

>5

min;

or

(2)

delay

of

>2

min

from

IHCA

event

recognition

to

the

first

defibrillation

shock

in

patients

whose

initial

documented

IHCA

rhythm

was

VF/pVT.

“Resuscitation

system

errors”

were

defined

as

the

total

number

of

reported

system

errors

plus

the

number

of

“process

of

care

exceptions”

in

each

case.

When

the

process

of

care

exception

matched

a

self

reported

error

(e.g.,

delay

in

time

to

first

shock,

time

to

first

vasoconstrictor

administration),

only

one

point

was

assigned

for

the

error

to

avoid

double

counting.

2.5.

Statistical

analysis

All

data

analyses

were

performed

using

PASW

Statistics

ver-

sion

17.0.2

(SPSS,

Inc.,

Chicago,

IL).

Chi

square

and

95%

confidence

intervals

were

used

for

comparison

of

descriptive

variables.

ANOVA

using

Scheffe’s

test

for

multiple

comparisons

was

used

to

analyze

the

relationship

between

0,

1,

2,

3

or

more

errors

and

outcome

variables.

Cox

regression

analysis

was

used

to

determine

the

rela-

tionship

between

reported

resuscitation

system

errors

as

well

as

other

important

clinical

variables

and

the

hazard

ratio

for

death

prior

to

hospital

discharge.

The

status

of

the

primary

outcome

(i.e.,

life

or

death)

was

ascertained

at

three

time

points

follow-

ing

resuscitation:

(1)

whether

ROSC

occurred;

(2)

24

h

following

the

IHCA

event;

and

(3)

at

hospital

discharge.

Cox

regression

anal-

ysis

was

also

used

to

determine

the

hazard

ratio

for

the

time

of

day

(day/evening

or

7a–11p

vs.

night

or

11p–7a)

and

day

of

week

(weekday

or

M–F,

weekend

or

S–Su)

on

the

reporting

of

resuscita-

tion

system

errors

during

a

IHCA

event.

3.

Results

A

total

of

118,387

in-hospital,

adult,

index

IHCA

cases

were

entered

into

the

NRCPR

database

from

January

1,

2000

through

August

26,

2008.

Of

these,

84,440

(71.3%)

had

no

system

errors

recorded

and

33,947

(28.7%)

had

one

or

more

system

errors

recorded.

Of

the

cases

with

system

errors,

26,919

(22.7%)

had

1

error,

5614

(4.7%)

had

2

system

errors,

and

1414

(1.2%)

had

3

or

more

system

errors.

Of

all

cases

in

which

the

initial

rhythm

was

recorded,

84,169/108,636

(77.5%)

had

non

VF/pVT

and

24,467/108,636

(22.5%)

had

VF/pVT.

Errors

were

committed

in

22,599/84,169

(26.8%)

of

non

VF/pVT,

and

in

9894/24,467

(40.4%)

of

those

with

VF/pVT

as

the

first

documented

rhythm.

The

distribution

of

system

errors

for

all

IHCA

patients

by

cate-

gory

is

noted

in

Table

1

.

The

most

frequent

system

errors

related

to

delay

in

medication

administration

(>5

min

time

from

event

recognition

to

first

dose

of

a

vasoconstrictor),

defibrillation,

air-

way

management,

and

chest

compression

performance

errors.

The

5.4%

of

cases

having

a

delay

in

defibrillation

of

>2

min

repre-

sents

the

percentage

of

patients

with

defibrillation

delays

using

all

J.P.

Ornato

et

al.

/

Resuscitation

83 (2012) 63–

69

65

Table

1

Distribution

of

resuscitation

system

errors.

Error

category

N

%

of

cases

%

of

reported

resuscitation

system

errors

Alerting

Delay

in

alerting

code

team

232

0.2

1.1

Pager

issue

66

0.1

Other

266

0.2

Airway

Aspiration

related

to

provision

of

airway

225

0.2

17.0

Airway

insertion

delay

2672

2.3

Delayed

recognition

of

airway

misplacement

328

0.3

Intubation

attempted,

not

achieved

319

0.3

Multiple

intubation

attempts

3314

2.8

Other

1939

1.6

Vascular

access

Delay

in

obtaining

access

846

0.7

3.4

Inadvertent

arterial

cannulation

37

<0.1

Infiltration

or

inadvertent

disconnection

of

IV

line

415

0.4

Other

480

0.4

Chest

compression

Compression

rate

of

∼100/min

not

maintained

2530

2.1

7.2

Interruption

of

compressions

>10

s

at

any

time

177

0.1

Delay

in

starting

chest

compressions

155

0.1

No

board

used

beneath

patient

during

compressions

758

0.6

Other

120

0.1

Defibrillation

Defibrillation

delay

>2

min

from

event

recognition

in

patients

with

an

initial

CA

rhythm

of

VF/pVT

due

to

insufficient

trained

personnel

or

defibrillator

not

immediately

available

6364

5.4

19.0

Energy

level

too

high/low

based

on

AHA

guidelines

485

0.4

Incorrect

defibrillator

paddle/pad

placement

32

<0.1

Defibrillator

malfunction

131

0.1

Shock

given,

not

indicated

957

0.8

Shock

indicated,

not

given

607

0.5

Other

1068

0.9

Medications

First

vasopressor

(epinephrine

or

vasopressin)

delay

>5

min

from

event

recognition

in

cases

with

event

duration

>5

min

20,035

16.9

42.5

Wrong

route

of

administration

202

0.2

Wrong

dosage

575

0.5

Wrong

medication

selection

1190

1.0

Code

team

leadership

Delay

in

identifying

team

leader

276

0.2

2.8

Knowledge

of

equipment

107

0.1

Knowledge

of

medications/protocols

539

0.5

Knowledge

of

team

member

roles

223

0.2

Code

team

oversight

196

0.2

Too

many

individuals

present

in

room

105

0.1

Protocol

deviation

from

AHA

Guidelines

Deviation

from

AHA

ACLS

guideline

recommendations

1521

1.3

3.1

Other

84

0.1

Equipment

issues

Availability

1100

0.9

3.9

Malfunction

540

0.5

Other

387

0.3

patients

as

the

denominator

including

those

without

initial

shock-

able

rhythms.

There

were

6364

patients

with

defibrillation

delays

of

>2

min

out

of

20,125

patients

with

an

initial

shock-able

rhythm

documented,

yielding

a

delay

in

defibrillation

in

31.6%

of

patients

in

whom

defibrillation

was

indicated.

Table

2

displays

the

demographic

characteristics

of

IHCA

events

with

and

without

reported

resuscitation

system

errors.

Cases

in

which

system

errors

were

documented

were

more

likely

to

be

male,

not

witnessed/monitored

at

the

time

of

arrest,

during

the

night

(11p–7a),

on

weekends

(S–Su),

initially

in

VF/pVT,

medical

(vs.

sur-

gical

or

cardiac)

type

admissions,

or

non-ICU

patients

(vs.

ICU,

ED,

or

OR/PACU

patients)

than

cases

in

which

no

system

errors

were

documented.

Resuscitation

system

errors

occurred

in

the

highest

percentage

of

IHCA

events

in

non-ICU

inpatient

areas

(9838/24,378,

40.4%);

they

were

least

frequently

noted

on

IHCA

events

occurring

in

ICU/stepdown

units

(17,692/69,643,

25.8%),

EDs

(3491/12,830,

27.2%),

or

Operating

Room/Post-Anesthesia

Care

Units

(685/2913,

23.5%)

(p

=

.0001).

The

presence

of

documented

resuscitation

system

errors

on

an

IHCA

event

was

associated

with

decreased

rates

of

ROSC,

survival

to

24

h,

and

survival

to

hospital

discharge

(

Fig.

1

)

in

all

patients

as

a

group

and

in

those

whose

initial

documented

IHCA

rhythm

was

separated

into

VF/pVT

or

non-VF/pVT

categories.

Characteristics

that

increased

the

likelihood

of

death

prior

to

hospital

discharge

in

patients

whose

initial

rhythm

was

non-VF/pVT

included:

male

gender,

when

the

event

occurred

at

night

or

on

a

weekend,

or

when

there

were

documented

resuscitation

system

errors

(

Fig.

2

).

There

was

a

lower

likelihood

of

death

prior

to

hospital

discharge

if

the

event

was

witnessed

or

monitored,

if

the

type

of

patient

was

car-

diac

or

surgical

(as

opposed

to

general

medical),

and

if

the

patient

location

was

OR/PACU.

66

J.P.

Ornato

et

al.

/

Resuscitation

83 (2012) 63–

69

Table

2

Demographic

characteristics

of

in-hospital

CA

events

with

and

without

resuscitation

system

errors.

No

resuscitation

system

errors

Resuscitation

system

errors

p

N

(total

=

118,387)

84,440

(71.3%)

33,947

(28.7%)

–

Age

[mean,

95%

CI]

66.07

[65.96,

66.18]

66.36

[66.20,

66.53]

.008

Male

48,428/84,440

(57.4%)

19,951/33,947

(58.8%)

.0001

Witnessed

or

monitored

arrest

69,361/70,152

(98.9%)

25,729/26,636

(96.6%)

.0001

Initial

documented

CA

rhythm

VF/pVT

14,573/76,143

(19.1%)

9894/32,493

(30.4%)

.0001

Patient

type

Cardiac

29,329/81,750

(35.9%)

11,487/33,757

(34.0%)

.0001

Medical

34,777/81,750

(42.5%)

15,483/33,757

(45.9%)

Surgical

17,644/81,750

(21.6%)

6787/33,757

(20.1%)

CA

event

location

ICU

or

stepdown/telemetry

area

50,951/77,057

(66.1%)

17,692/31,707

(55.8%)

.0001

Non-ICU

inpatient

area

14,539/77,057

(18.9%)

9838/31,707

(31.0%)

Emergency

department

9339/77,057

(12.1%)

3491/31,707

(11.0%)

Operating

room

or

PACU

2228/77,057

(2.9%)

685/31,707

(2.2%)

Time

of

day

Day-evening

(7a–11p)

54,888/80,480

(68.2%)

22,387/33,777

(66.3%)

.0001

Night

(11p–7a)

25,592/80,480

(31.8%)

11,390/33,777

(33.7%)

Day

of

week

Weekday

(M–F)

56,783/81,970

(69.3%)

23,236/10.711

(68.4%)

.003

Weekend

(S–Su)

25,187/81,970

(30.7%)

10,711/33,947

(31.6%)

Fig.

1.

Effect

of

any

resuscitation

system

errors

on

an

IHCA

event

and

the

rate

of

ROSC,

survival

for

24

h,

and

survival

to

hospital

discharge

for

all

patients

and

those

with

an

initial

documented

IHCA

rhythm

of

non-VF/pVT

and

VF/pVT.

Fig.

2.

Association

between

the

number

of

resuscitation

system

errors

during

an

IHCA

event

and

a

significantly

increased

hazard

ratio

for

death

prior

to

hospital

discharge

stratified

by

patients

with

an

initial

IHCA

rhythm

of

non-VF/pVT

or

VF/pVT.

J.P.

Ornato

et

al.

/

Resuscitation

83 (2012) 63–

69

67

Characteristics

that

increased

the

likelihood

of

death

prior

to

hospital

discharge

in

patients

whose

initial

rhythm

was

VF/pVT

included:

when

the

event

occurred

at

night

or

on

a

weekend,

or

when

there

were

documented

resuscitation

system

errors

(

Fig.

2

).

There

was

a

lower

likelihood

of

death

prior

to

hospital

discharge

if

the

type

of

patient

was

cardiac

or

surgical

(as

opposed

to

gen-

eral

medical)

or

if

the

patient

location

was

ED.

The

relative

risk

of

death

prior

to

hospital

discharge

based

on

hazard

ratio

analysis

was

9.9%

(95%

CI

7.8,

12.0)

more

likely

for

patients

whose

initial

docu-

mented

rhythm

was

non-VF/pVT

when

resuscitation

system

errors

were

reported

compared

to

when

no

errors

were

reported.

It

was

34.2%

(95%

CI

29.5,

39.1)

more

likely

for

those

with

VF/pVT.

Only

certain

types

of

resuscitation

system

errors

were

asso-

ciated

with

an

increased

hazard

ratio

for

death

prior

to

hospital

discharge.

For

patients

whose

initial

rhythm

was

non-VF/pVT,

delays

in

obtaining

vascular

access

(i.e.,

an

intravenous

line)

and

medication

errors

increased

the

hazard

ratio

for

death

prior

to

hospital

discharge

(

Fig.

3

).

For

patients

whose

initial

rhythm

was

VF/pVT,

defibrillation

problems

and

medication

errors

increased

the

hazard

ratio

for

death

prior

to

hospital

discharge.

4.

Discussion

The

principal

finding

in

this

study

is

that

the

presence

of

resus-

citation

system

errors

is

associated

with

decreased

survival

from

IHCA

in

adults.

More

errors

were

noted

in

patients

whose

initial

documented

IHCA

rhythm

was

VF/pVT

as

opposed

to

those

with

non-shock-able

rhythms.

This

finding

is

particularly

relevant

clin-

ically,

given

that

the

majority

of

survivors

of

IHCA

are

those

with

initial

VF/pVT.

3

Our

findings,

although

much

broader,

support

those

reported

by

Chan

et

al.

4

who

evaluated

6789

sequential

patients

with

VF

as

the

first

documented

rhythm

in

the

NRCPR

database

and

found

that

30.1%

of

this

cohort

underwent

defibrillation

more

than

2

min

after

initial

recognition

of

their

IHCA.

Patients

with

delayed

defib-

rillation

had

a

significantly

lower

likelihood

of

ROSC

(adjusted

odds

ratio,

0.55;

95%

CI,

0.49–0.62;

p

<

0.001)

and

survival

at

24

h

after

the

cardiac

arrest

(adjusted

odds

ratio,

0.52;

95%

CI,

0.46–0.58;

p

<

0.001).

The

Chan

study

was

the

first

large

scale

analysis

of

data

in

IHCA

patients

identifying

a

specific

defibrillation

delay

time

cut-

off

(2

min

or

less

after

event

recognition)

that

negatively

impacts

survival.

Our

analysis,

in

a

larger

sample

size

from

the

same

reg-

istry,

confirms

that

defibrillation

system

errors,

including

a

>2

min

delay

from

IHCA

recognition

to

initial

defibrillation

accounts

for

higher

mortality

in

the

initial

VF/pVT

group,

but

it

also

found

an

association

between

medication

errors

and

a

lower

likelihood

of

survival.

Peberdy

et

al.

3

found

that

survival

to

discharge

following

in-

hospital

cardiac

arrest

is

lower

during

nights

(14.7%

[95%

CI,

14.3–15.1%]

vs.

19.8%

[95%

CI,

19.5–20.1%])

or

weekends

(20.6%

[95%

CI,

20.3–21%]

vs.

17.4%

[95%

CI,

16.8–18%])

compared

with

day/evening

or

weekdays,

even

after

accounting

for

many

poten-

tially

confounding

IHCA

event

and

hospital

factors.

Our

current

study

confirms

that

nights

and

weekends

are

associated

with

an

increased

hazard

ratio

for

death

prior

to

hospital

discharge

in

patients

regardless

of

the

initial

documented

IHCA

rhythm

and

goes

a

step

beyond

the

previously

reported

data

in

demonstrating

an

increased

hazard

ratio

for

death

during

those

times

in

which

there

is

an

increase

in

resuscitation

system

errors,

thus

suggesting

a

link

between

increased

error

on

nights

and

weekends

and

decreased

survival

during

those

times.

Kayser

et

al.,

5

demonstrated

that

ED

location

was

an

inde-

pendent

predictor

of

improved

survival,

speculating

that

this

was

due

to

both

the

requirement

for

ED

staff

to

receive

basic

and

the

advanced

cardiac

life

support

training,

and

their

frequent

experience

in

performing

resuscitation

compared

to

clinicians

working

on

general

hospital

floors.

Our

paper

supports

this

hypoth-

esis

by

confirming

that

IHCA

events

occurring

in

the

ED

have

a

relatively

low

percentage

of

cases

with

resuscitation

system

errors.

Not

all

types

of

errors

were

associated

with

an

increased

hazard

ratio

for

death

prior

to

hospital

discharge

in

our

study.

Olasveen-

gen

et

al.,

17

randomized

out-of-hospital

cardiac

arrest

patients

to

receive

ACLS

treatment

with

and

without

intravenous

drug

admin-

istration

and

determined

that

survival

to

discharge

was

the

same

in

both

groups.

However,

they

were

not

able

to

identify

the

timeframe

in

which

the

drug

therapy

was

given

in

the

group

that

received

an

intravenous

line

and

medication.

In

our

IHCA

population,

we

were

able

to

demonstrate

decreased

survival

when

the

first

vaso-

constrictor

was

administered

>5

min

after

IHCA

onset

in

patients

whose

arrest

lasted

for

at

least

5

min.

The

difference

in

our

find-

ings

is

likely

due

to

the

fact

that

the

time

from

event

onset

to

first

drug

administration

is

usually

much

shorter

in-

vs.

out-of-hospital.

In

our

study,

the

mean

time

from

event

onset

to

first

epinephrine

was

2.4

(95%

CI,

2.3,

2.4)

min.

In

contrast,

the

time

interval

from

collapse

to

first

epinephrine

in

standard

vs.

high

dose

epinephrine

pre-hospital

trials

is

approximately

20

min.

18

Much

of

the

focus

on

patient

safety

and

error

prevention

in

hos-

pitals

focuses

on

interventions

that

relate

to

patient

interactions

that

are

of

higher

volume

and

lower

acuity

compared

to

resus-

citation.

In

our

study,

the

impact

of

resuscitation

team

errors

on

survival

varied

widely

by

both

error

type

and

the

initial

docu-

mented

heart

rhythm,

with

some

errors

causing

little

or

no

impact

and

others

being

associated

with

a

significantly

lower

survival,

par-

ticularly

in

the

group

of

patients

with

initial

VF/pVT.

Given

that

resuscitation

practices

need

to

occur

consistently

well

throughout

all

areas

of

the

hospital

and

that

responders

will

always

have

vary-

ing

degrees

of

expertise

and

experience,

our

findings

suggest

that

resuscitation

training

should

be

targeted

to

emphasize

avoiding

the

types

of

errors

having

the

greatest

impact

on

survival

(e.g.,

delays

in

initial

defibrillation

and

medication

administration

and

adherence

to

ACLS

protocols).

An

increasing

body

of

evidence

indicates

that

effective

lead-

ership

and

team

work

rather

than

just

individual

knowledge,

skills,

and

attitudes

are

required

to

optimize

outcomes

and

min-

imize

errors

in

a

variety

of

medical

emergencies.

8–15,16

Specific

behaviors

have

been

identified

that

contribute

to

effective

leader-

ship

including

providing

orienting

remarks,

inviting

team

member

contributions,

promoting

exchange

of

information

and

clear

com-

munication,

and

avoiding

performing

physical

tasks

that

can

be

assigned

to

others

during

the

emergency.

8

In

addition,

there

are

four

teamwork

behavior

principles

that

can

help

to

avoid

medi-

cal

errors:

(1)

the

leader

should

voice

specific

findings

rather

than

diagnosing

the

problem

prematurely;

(2)

all

members

of

the

team

should

“think

out

loud”

and

“talk

to

the

room”

as

the

case

unfolds;

(3)

the

leader

should

direct

period

reviews

of

quantitative

infor-

mation

(e.g.,

drug

dose,

time,

response);

and

(4)

all

members

of

the

team

should

double-check

crucial

data.

8

Unfortunately,

the

majority

of

these

leadership

and

team

behav-

iors

could

not

be

measured

in

this

study.

This

is

the

likely

explanation

for

why

we

were

unable

to

demonstrate

an

association

with

reported

“poor

team

leadership”

and

survival

in

our

study.

The

definition

of

“good

team

leadership”

is

subjective

and

may

not

be

reported

consistently

among

institutions

or

that

that

magnitude

of

the

effect

on

survival

of

other

resuscitation

errors

drowns

out

the

impact

of

poor

team

leadership.

Our

observations

confirm

the

association

between

the

presence

of

resuscitation

system

errors

that

are

evident

from

review

of

the

resuscitation

record

and

decreased

survival

from

IHCA

in

adults.

However,

they

do

not

point

to

a

specific

solution

to

the

problem.

A

number

of

recent

simulator-based

studies

have

identified

qualitatively

and

quantitatively

similar

problems

to

those

noted

68

J.P.

Ornato

et

al.

/

Resuscitation

83 (2012) 63–

69

Fig.

3.

Association

between

various

types

of

resuscitation

system

errors

and

the

hazard

ratio

for

death

prior

to

hospital

discharge

in

patients

whose

initial

IHCA

rhythm

was

non-VF/pVT

and

VF/pVT.

during

clinical

resuscitation.

4,19–24

Team

training

can

improve

per-

formance

during

simulation

of

medical

emergencies,

25–28

but

not

all

training

is

equal

or

effective.

12–14

Siassakos

et

al.,

12–14

have

iden-

tified

specific

elements

required

to

improve

outcome,

including

multi-professional

training

of

all

healthcare

providers

who

man-

age

an

emergency

in

a

realistic

simulation

setting.

These

elements

need

to

be

incorporated

into

team

and

leadership

training,

which

are

now

recommended

in

the

2010

American

Heart

Association

Guidelines

for

Adult

and

Pediatric

Advanced

Cardiovascular

Life

Support.

29,30

4.1.

Limitations

General

limitations

of

NRCPR

include:

(1)

registry

hospitals

may

not

be

representative

of

all

hospitals;

(2)

there

is

no

on-site

valida-

tion

of

data

collection;

(3)

and

there

is

no

follow-up

after

hospital

discharge.

In

addition,

although

medication

use

is

tracked,

NRCPR

does

not

attempt

to

assess

clinical

eligibility

for

each

medication.

These

limitations

are

similar

to

those

of

other

contemporary

in-

hospital

registries.

The

specific

limitation

pertinent

to

this

analysis

is

that

NRCPR

data

is

self-reported

by

having

a

trained

abstractor

review

hospi-

tal

charts

and

code

records.

The

Get

with

the

Guidelines

NRCPR

is

a

voluntary

data

collection/analysis

quality

improvement

project.

Participating

hospitals

pay

a

fee

to

the

American

Heart

Associa-

tion

to

have

the

data

analyzed,

benchmarked,

and

reported

back

to

them

quarterly.

With

hundreds

of

hospitals

involved

through-

out

the

entire

United

States

and

very

minimal

hospital

user

fees,

it

has

not

been

feasible

to

provide

independent

data

validation

at

the

hospital

level.

The

abstractors

work

with

uniform

definitions

and

uniform

methods

of

data

acquisition.

The

abstractors

review

the

physi-

cian

and

nurse

narrative

notes

in

the

hospital

chart

and

the

“code

record”

on

each

event.

Thus,

all

of

the

errors

identified

came

from

a

review

of

documentation

in

the

patient

charts.

The

abstractors

do

not

capture

whether

the

errors

were

“recognized”

in

the

narrative

notes

by

the

doctors

and

nurses

running

the

code

or

are

evident

from

the

documentation

of

events

themselves

(which

is

the

case

for

the

majority

of

errors

that

we

looked

at).

There

is

no

way

for

the

abstractor

to

verify

whether

errors

may

have

occurred

but

were

not

documented.

This

may

have

resulted

in

an

under-reporting

of

resuscitation

team

errors.

If

this

occurred,

if

anything,

it

would

increase

the

potential

importance

of

our

findings.

Finally,

we

cannot

exclude

the

possibility

that

individu-

als

completing

the

code

sheets

could

have

checked

off

more

“error

boxes”

in

patients

who

did

not

achieve

ROSC

than

the

boxes

they

would

have

checked

had

the

patient

been

resuscitated.

4.2.

Public

health

importance

The

public

health

importance

of

these

findings

is

con-

siderable.

Eisenberg

and

Mengert,

31

estimated

that

there

are

350,000–750,000

adult,

IHCA

events

per

year

in

the

United

States.

Eliminating

resuscitation

system

errors

has

the

potential

to

save

21,000–44,000

additional

lives

per

year

in

the

United

States

from

IHCA.

This

figure

is

ten

times

larger

than

the

estimated

2000–4000

additional

lives

saved

per

year

from

out-of-hospital

cardiac

arrest

that

drove

widespread

deployment

of

public

access

defibrillation

programs

throughout

the

country.

13

5.

Conclusions

We

conclude

that

the

presence

of

resuscitation

system

errors

that

are

evident

from

review

of

the

resuscitation

record

is

associ-

ated

with

decreased

survival

from

IHCA

in

adults.

Hospitals

should

target

their

training

of

first

responders

and

code

team

personnel

to

emphasize

the

importance

of

early

defibrillation

when

indi-

cated,

early

use

of

vasoconstrictor

medication,

and

compliance

with

established

AHA

ACLS

resuscitation

protocols.

Conflict

of

interest

statement

None

of

the

authors

have

any

relevant

conflicts.

Funding

sources

None.

Acknowledgement

None.

Appendix

A.

Get

with

the

Guidelines

– National

Registry

of

Cardiopulmonary

Resuscitation

(NRCPR)

investigators:

J.P.

Ornato

et

al.

/

Resuscitation

83 (2012) 63–

69

69

Mary

E.

Mancini,

Robert

A.

Berg,

Emilie

Allen,

Elizabeth

A.

Hunt,

Vinay

M.

Nadkarni,

Scott

Braithwaite,

Graham

Nichol,

Kathy

Dun-

can,

Tanya

Lane

Truitt,

Melinda

Smyth,

Brian

Eigel,

Paul

S.

Chan,

Tim

Mader,

Karl

B.

Kern,

Sam

Warren,

Thomas

Noel,

Romergryko

Geocadin,

Dana

Edelson,

Vince

Mosesso

and

Comilla

Sasson.

References

1. Institute

of

Medicine.

To

err

is

human:

building

a

safer

health

system.

Washing-

ton,

DC:

National

Academy

Press;

2000.

2. University

study

identifies

problems

with

IOM

report.

Hosp

Case

Manag

2000;8(Suppl.

3–4):146.

3. Peberdy

MA,

Ornato

JP,

Larkin

GL,

et

al.

Survival

from

in-hospital

cardiac

arrest

during

nights

and

weekends.

JAMA

2008;299:785–92.

4. Chan

PS,

Krumholz

HM,

Nichol

G,

Nallamothu

BK.

Delayed

time

to

defibrillation

after

in-hospital

cardiac

arrest.

N

Engl

J

Med

2008;358:9–17.

5.

Kayser

RG,

Ornato

JP,

Peberdy

MA.

Cardiac

arrest

in

the

emergency

depart-

ment:

a

report

from

the

National

Registry

of

Cardiopulmonary

Resuscitation.

Resuscitation

2008;78:151–60.

6. Nadkarni

VM,

Larkin

GL,

Peberdy

MA,

et

al.

First

documented

rhythm

and

clin-

ical

outcome

from

in-hospital

cardiac

arrest

among

children

and

adults.

JAMA

2006;295:50–7.

7.

Peberdy

MA,

Kaye

W,

Ornato

JP,

et

al.

Cardiopulmonary

resuscitation

of

adults

in

the

hospital:

a

report

of

14720

cardiac

arrests

from

the

National

Registry

of

Cardiopulmonary

Resuscitation.

Resuscitation

2003;58:297–308.

8.

Hunziker

S,

Johansson

AC,

Tschan

F,

et

al.

Teamwork

and

leadership

in

cardiopul-

monary

resuscitation.

J

Am

Coll

Cardiol

2011;57:2381–8.

9.

Schaefer

HG,

Helmreich

RL.

The

importance

of

human

factors

in

the

operating

room.

Anesthesiology

1994;80:479.

10.

Schaefer

HG,

Helmreich

RL,

Scheidegger

D.

Human

factors

and

safety

in

emer-

gency

medicine.

Resuscitation

1994;28:221–5.

11.

Schaefer

HG,

Helmreich

RL,

Scheidegger

D.

Safety

in

the

operating

theatre

–

part

1:

interpersonal

relationships

and

team

performance.

Curr

Anaesth

Crit

Care

1995;6:48–53.

12.

Siassakos

D,

Bristowe

K,

Draycott

TJ,

et

al.

Clinical

efficiency

in

a

simulated

emer-

gency

and

relationship

to

team

behaviours:

a

multisite

cross-sectional

study.

BJOG

2011;118:596–607.

13.

Siassakos

D,

Draycott

TJ,

Crofts

JF,

Hunt

LP,

Winter

C,

Fox

R.

More

to

teamwork

than

knowledge,

skill

and

attitude.

BJOG

2010;117:1262–9.

14.

Siassakos

D,

Fox

R,

Crofts

JF,

Hunt

LP,

Winter

C,

Draycott

TJ.

The

management

of

a

simulated

emergency:

better

teamwork,

better

performance.

Resuscitation

2011;82:203–6.

15.

Tourgeman-Bashkin

O,

Shinar

D,

Zmora

E.

Causes

of

near

misses

in

critical

care

of

neonates

and

children.

Acta

Paediatr

2008;97:299–303.

16.

Manser

T.

Teamwork

and

patient

safety

in

dynamic

domains

of

health-

care:

a

review

of

the

literature.

Acta

Anaesthesiol

Scand

2009;53:

143–51.

17.

Olasveengen

TM,

Sunde

K,

Brunborg

C,

Thowsen

J,

Steen

PA,

Wik

L.

Intravenous

drug

administration

during

out-of-hospital

cardiac

arrest:

a

randomized

trial.

JAMA

2009;302:2222–9.

18.

Brown

CG,

Martin

DR,

Pepe

PE,

et

al.

A

comparison

of

standard-dose

and

high-

dose

epinephrine

in

cardiac

arrest

outside

the

hospital.

The

Multicenter

High-

Dose

Epinephrine

Study

Group.

N

Engl

J

Med

1992;327:1051–5.

19. Abella

BS,

Alvarado

JP,

Myklebust

H,

et

al.

Quality

of

cardiopulmonary

resusci-

tation

during

in-hospital

cardiac

arrest.

JAMA

2005;293:305–10.

20. Abella

BS,

Sandbo

N,

Vassilatos

P,

et

al.

Chest

compression

rates

during

cardiopulmonary

resuscitation

are

suboptimal:

a

prospective

study

during

in-

hospital

cardiac

arrest.

Circulation

2005;111:428–34.

21.

Kramer-Johansen

J,

Myklebust

H,

Wik

L,

et

al.

Quality

of

out-of-hospital

car-

diopulmonary

resuscitation

with

real

time

automated

feedback:

a

prospective

interventional

study.

Resuscitation

2006;71:283–92.

22. Chan

PS,

Nichol

G,

Krumholz

HM,

Spertus

JA,

Nallamothu

BK.

Hospital

varia-

tion

in

time

to

defibrillation

after

in-hospital

cardiac

arrest.

Arch

Intern

Med

2009;169:1265–73.

23. Hunziker

S,

Tschan

F,

Semmer

NK,

Howell

MD,

Marsch

S.

Human

factors

in

resuscitation:

lessons

learned

from

simulator

studies.

J

Emerg

Trauma

Shock

2010;3:389–94.

24.

Hunziker

S,

Tschan

F,

Semmer

NK,

et

al.

Hands-on

time

during

cardiopulmonary

resuscitation

is

affected

by

the

process

of

teambuilding:

a

prospective

ran-

domised

simulator-based

trial.

BMC

Emerg

Med

2009;9:3.

25.

Thomas

EJ,

Taggart

B,

Crandell

S,

et

al.

Teaching

teamwork

during

the

Neonatal

Resuscitation

Program:

a

randomized

trial.

J

Perinatol

2007;27:

409–14.

26.

Thomas

EJ,

Williams

AL,

Reichman

EF,

Lasky

RE,

Crandell

S,

Taggart

WR.

Team

training

in

the

neonatal

resuscitation

program

for

interns:

teamwork

and

quality

of

resuscitations.

Pediatrics

2010;125:539–46.

27.

Cooper

S.

Developing

leaders

for

advanced

life

support:

evaluation

of

a

training

programme.

Resuscitation

2001;49:33–8.

28.

Hunziker

S,

Buhlmann

C,

Tschan

F,

et

al.

Brief

leadership

instructions

improve

cardiopulmonary

resuscitation

in

a

high-fidelity

simulation:

a

randomized

con-

trolled

trial.

Crit

Care

Med

2010;38:1086–91.

29.

Bhanji

F,

Mancini

ME,

Sinz

E,

et

al.

Part

16:

education,

implementation,

and

teams:

2010

American

Heart

Association

Guidelines

for

Cardiopul-

monary

Resuscitation

and

Emergency

Cardiovascular

Care.

Circulation

2010;122:S920–33.

30.

Mancini

ME,

Soar

J,

Bhanji

F,

et

al.

Part

12:

education,

implementation,

and

teams:

2010

International

Consensus

on

Cardiopulmonary

Resuscitation

and

Emergency

Cardiovascular

Care

Science

with

Treatment

Recommendations.

Cir-

culation

2010;122:S539–81.

31.

Eisenberg

MS,

Mengert

TJ.

Cardiac

resuscitation.

N

Engl

J

Med

2001;344:1304–13.