esophageal gastric tube airway vs endotracheal tube in prehospital cardiopulmonary arrest chest 1986

8/10/2019 Esophageal Gastric Tube Airway vs Endotracheal Tube in Prehospital Cardiopulmonary Arrest Chest 1986

1/7

Esophageal Gastric Tube Airway vs

Endotracheal Tube in Prehospital

Cardiopulmonary Arrest

lruin E Coldenberg M .D .; Brian C . Campion M .D. ;

Constance M Siebold R.N.;]ohn

W

McBride M.D.;and

Linda Long M .D .

We evaluated the efficacy of the esophageal airway (EA) by

prospectively randomizing 175 prehospital cardiopulmo-

nary arrest patients to receive either an esophageal gastric

tube

airway (EGTA) or an endotracheal tube (ET). If

attempts with the initid airway failed, the alternateairway

was attempted. The cost of training paramedics in EA use

was considerably less than the ET ( 80vs 1,000).Survival to

the emergency room, to hospitalization and to discharge in

ET and EGTA

p u p s were 64.4percent,

25.6

percent, l l . 1

percent, and 54.1 percent, 27.1 percent,

B.9

percent,

he optimal airway for respiratory management of a

T pa t i en t with cardiopulmonary arrest is a well-

placed endotracheal tube (ET). However, the place-

ment of an ET requires a certain level of skill with

training and recertification requiring substantial time

and resources. Therefore, alternative methods for

ventilation have been sought. The esophageal airway,

ie

both the esophageal obturator airway (EOA)and the

esophageal gastric tube airway (EGTA), were devel-

oped s such alternatives to the ET. Despite the

esophageal airway s acceptance s a useful airway

adjunct by the National Conference on Cardiopulmo-

nary Resuscitation and Emergency Cardiac Care in

1973 and 1979,0 the efficacy of this airway remains

c o n t r o ~ e r s i a l . ~o evaluate the efficacy of the

esophageal airway (EA), we performed a study compar-

ing morbidity and mortality in patients prospectively

randomized to receive either the EGTA or the ET in

prehospital cardiopulmonary arrest. We also compared

these patients to a group of unrandomized patients

who were resuscitated by paramedics trained only in

esophageal airway intubation.

METHODS

St

Popuhtion

D u ri ng a m y e a r p er io d, 175 patients with out-of-hospital

*From the Section of Cardiolo and Office of Emerg ency Medical

Services, S t. Pau l-Ramse M g c a l ~ e n t e r , t.

Paul

and University

of Minnesota Medical ~ X o o l ,Minneapolis.

This work was supported in part by a grant from the Medical

Education and Research Foundation of St. Paul-Ramsey Medical

Cent er and Ramse Clinic, St. Paul, MN.

Manuscript receivedr ~ u g u s t6; revision accepted anuary

28.

Reprint requests: Dr

ong St.

Paul-Ramey Me

Center 640

Jackson Street, St. Poul5 901

d d

respectively-differences not statistically significant. The

incidence of neurologic residual (ET50 percent, EGTA 36.4

percent)and congestive heart failure (ET

40

percent, EGTA

45.5 percent) in surviving ET and EGTA patients did not

differ (NS).

An

additional onsecutive patients with only

the opportunity to receive an EA were also evaluated and

did not differ inmortality, neurologic residual, or congestive

heart failure from ET patients. We conclude that the EA is a

satisfactory alternative to the ET for short-term prehospital

use in cardiopulmonary arrest patients.

cardiopulmonary arrest were prospectively randomized

at

the time

of parame dic arrival at the scene to receive the ET (group 1)or the

EGTA (group

2)

for initial airway management. On paramedic

arrival, a

card

was drawn. The writing on

this

card determined

whe ther th e patient's initial airway attem pt was to be an EGTA or an

ET. Th e random order of thes e cards

was

dete rniin ed from a table of

random numbers. If tw attempts with the initial airway failed to

adequately ventilate the patient, the alternative airway

was

at

tempted. Inclusion criteria required

that

the patient's

initf l

airway

management with an esophageal airway or an endotracheal tube be

done by paramedics participating in the study and that any patient

entered into the study be at least five feet tall. The 35 paramedics

participating in this part of the study w ere trained in both E T and

EGTA insertion. All patients were treated by paramedics in the St.

Paul, Minnesota Paramedic Program.

During the sam e period, an additional

25

consecutive patients

with out-of-hospital cardiopulmonary arrest who were not ran-

domized but who had an EOA attem pted by paramedics trained only

in esophageal airway placement were evaluated (group 3 . These

patients we re ventilated by an esophageal airway, or

if

this failed, by

an oral airway. These additional 25 atients were compared to group

1patients t o see if they differed in morbidity o r mortality.

Esophageal

i w y

The esophageal airway and its insertion have been described in

detail elsewhere.LPg

Proofder

o

Care

The St. Paul Paramedic Program has 25 paramedics. Thirty-five

of

these paramedics were trained in both endotracheal tube and

esophageal airway intubation. The remainder were trained only in

esophageal airway intubation. Endotracheal intubation training

consisting of

U

hours of didactic work, mannequin practice, and

12

to 16 intubations in th e operating room cost approximately 1,000per

paramedic. Completion of training

was based

on an evaluation for

proficiency by the Department of Anesthesiology prior to using the

ET in the field. Each paramedic

was

required to return to the

operating room each year to demonstrate endotracheal intubation

wnloaded From: http://journal.publications.chestnet.org/ by a University of Minnesota Libraries User on 01/02/2014


2/7

under supervision. The EGTA and EOA training, which consisted of

three hours of didactic work and mannequin training, cost approx-

imately

80

per paramedic. During management of all cardiac

arrests, paramedics received orders through radio and telemetry

contact with S t Paul-Ramsey Medical Center physicians. The

Medical Center serves as the medical control base station for

Ramsey, Washington, and Dakota Counties. Cardiac arrests were

treated according to American Heart Association Advanced Cardiac

Life Support guidelines.'

Data were obtained on all patients from special protocol study

reports, routine paramedic run reports, emergency mom (ER)

records, and hospital charts. The data collected included the

following: population characteristics (age, sex, initiator of CPR,

initial rhythm, time from arrest to initiation of CPR, ie downtime,

time from arrest to paramedic arrival, time

required

for intubation,

cause of arrest, preintubation vomiting, past medical history),

hospital course (neurologic sequelae, incidence of congestive heart

failure, incidence of aspiration pneumonia), survivaldata, ditficulties

with airway establishment, airway complications, and in some

patients, arterial blood gas levels (ABGs).

The downtime was estimated by the bystander at the scene. It

equalled the time between the onset of the cardiac arrest and the

beginning of basic

life

support measures (CPR). The time to

paramedic arrival equalled the time between the onset of the cardiac

arrest and the time the paramedics arrived at the scene.

Statistical analysiswas performed using chi square tests, Fischer's

exact probability tests, and Student's t-tests. Statistical analysiswas

performed (1) between the patients randomized to ET (group 1) and

patients randomized to EGTA (group 2), and (2)between the patients

randomized to ET (group

1

and the

115

consecutive patients who

were not randomized but had the EOA attempted (group 3).

The data

in

this paper are presented by intent to treat groups

(groups1 o 3). but incorrect randomization occurred in 17.1percent

of the patients. Therefore, statistical analysis

was

also performed

between the following groups:

(3) Correctly randomized ET patients vs correctly randomized

EGTA patients.

4) Randomized patients receiving an ET as final airway vs ran-

domized patients receiving an EGTA as final airway.

(5) Correctly randomized patients vs incorrectly randomized pa-

tients.

(6) Correctly randomized ET patients vs group 3 patients.

(7)

Randomized patients receiving ET as final airway vs group 3

patients. The

data

presented in this paper reflect intent to treat

unless otherwise stated. However, whenever significant results were

obtained by any of the above analyses, they are reported.

One hundred seventy-five patients were ran-

domized (Table1 .Ninety were randomized to ET and

85

to EGTA. However, only 145 of these 175 patients

were correctly randomized (81 to ET, 64 to EGTA).

That is, 30 patients either &led to meet inclusion

criteria or the randomized airwaywas not attempted in

able 1 Croups nalyted

Total ET EGTA

Initially randomized

175

90

85 NS

Correctly randomized

145 81 6 p


3/7

Table 4 C a r ~ n u n u a y Arresi Charectcrbtics Table Morbidity in

H a a p i d b d

Patients

Group

1

Group2 Group3

Number 90 85

125

Initiator o f CPR

Paramedic

57 52 71

Public

12 12

20

Emergency medical technician

18

13

27

Nurse

2 6 5

Physician

0 1 0

Police

1 1 1

Unknown

0 0 1

Initial rhythm

Ventricular

tachIFib* 50 45 66

~rad~cardialblbcks 23 20

19

Asystole

16 14 29

Electromechanical dissociation

1

4

10

Unknown

0 2 1

Cause of arrest

Cardiac 73

67 97,

Rimary respiratory arrest

6 10 10

Other

9

7 7

Unknown

2 1 11

Reintubation vomiting

6 5 4

*Ventricular

tachycardia/fibrillation

includes only one patient

with

ventricular tachycardia in group

2.

table were similar between patients randomized to ET

or EGTA and between the patients randomized to ET

and group

3

patients.

Eble 4 lists the initiator of CPR, the initial rhythm,

the cause of arrest, and the incidence of preintubation

vomiting in these groups. There were no statistically

significant differences in any of these characteristics

between the patients randomized to ET and the

patients randomized to EGTA. When the group

1

patients were compared to the group

3

patients, the

frequency distribution of the initial rhythm

w s

differ-

ent between the groups. The d rence w s almost

entirely due to the high incidence of electromechani-

cal dissociation in the group 3 patients. The incidence

of ventricular fibrillation/ventricular tachycardia w s

similar between these two groups. Group 1and group3

patients were otherwise similar.

Arterial Blood

as

Levels

The purpose of this study

w s

to determine if there

w s a difference in morbidity or mortality in patients

randomized to or receiving the ET or EGTA. The

drawing of ABGs immediately after arrival to the ER

Table A d

Blood

as

V

ET EGTA

PH

7.21 20.22

7.2120 .25

pot

121 127

149k 132

Pc o ,

51234

46

29

Hc o ,

1 9 2 5 1 8 k 8

X m e to intubation* min)

5.97k3.86 6.00 2 1.26

Group

1

Group

2

Group3

Number

90

85

125

Hosp italized patients

23 23 34

Neurolog ic residual

Initial even t

16 10

20

Prior to death or discharge

17 14

20

Cong estive heart failure 13

7 19

Aspiration pneumonia

5

8

N

A

*NA,

data

not obtained.

was not an integral part of the study. However,67of the

randomized patients and 31of the group

3

patients had

ABGs drawn soon after arrival in the ER. There

was

no

statistically significant difference in the ABGs or in

time elapsed from intubation to arterial sampling in

patients receiving the endotracheal tube or esophageal

airway (Table 5).

Hospital Course

Eble

6

lists the incidence of neurologic residual,

congestive heart failure

CHF),

and aspiration pnsu-

monia in hospitalized patients. The incidence of neu-

rologic residual secondary to the initial cardiopulmo-

nary arrest and the incidence of neurologic

residhal

secondary to any event after and including the initial

event were similar between group 1 and group

patients and between group

1

and group 3 patients.

Analysis, however, between patients correctly ran-

domized to ET and patients comctly randomized to

EGTA showed a significmtly higher incidence of

neurologic residual that could be attributed to the

initial event in the ET patients

72.7

percent vs

31.6

percent)

p


4/7

able

8 Sumival by A i n w y Gr oup

able

1 0 - D i H with A i n w y

Establishment

Group

1

Group2 Group

3

N N

Number 90

85 125

Survival to ER

58

64.4

46

54.1 78 62.4

Survivaltohospitalization 3 25.6

3

27.1 34 27.2

Survival to discharge

10 11.1 11 12.9 15 12.0

Unknown

1 1.2

groups analyzed. Table 9 shows no difference in

survival to discharge in randomized patients by intent

to treat groups, final airway received, and correctly

randomized patients.

Complications and Di f icult ies wi th i m y

Establishment

When airway difficulties were evaluated, we found

that the duration of the intubation procedure was

similar for the patients receiving the ET, EGTA, and

EOA,

and that there

w s

approximately

90

percent

successful airway placement with all tubes (Table 10).

Table

ll

ists the complications in patients receioing the

EGTA and ET. The incidence of complications w s

similar in the two groups.

Paramedic ssessm ent of irwa y dequ acy

More patients were judged to be adequately venti-

lated with the ET than the EGTA when ventilation w s

assessed by paramedics in the field (Table 10). Of the

patients who had the ET attempted, 89 percent were

judged to be adequately ventilated, while only 70

percent of the patients who had the EGTA attempted

were judged to be adequately ventilated. If in the

paramedics' opinion the patient

w s

being inade-

quately ventilated with the EGTA, the protocol

allowed placement of the ET. Therefore, most of the

patients assessed by the paramedics to be inadequately

ventilated with the EGTA subsequently received the

ET. However, when paramedics who were trained only

in esophageal airway intubation judged adequacy of

ventilation in the field, they felt that

90

percent of the

patients were adequately ventilated.

Adequate ventilation is essential for a successful

outcome in patients with cardiopulmonary arrest. At

able 9 Sumival to Discharge Randomized

Patients

ET E

GTA

N

Intent to treat

10 11.1 11 12.9 NS

Final airway received

13 10.9 15.4 NS

Correctly randomized patients

10 12.3 10 15.6 NS

ET EGTA EOA

Duration of intubation

procedure*

19211 16212 21+19

Number of patients with

tube attempted 132 125

Total number of patients with

successll intubations

119(90.2 ) 71(92.2 ) llB(Q4.476)

Number of patients with tube

attempted adequately

ventilated

t 118(89.4 ) 54(70.1 ) 113(90.4 )

*Mean seconds)

standard

de v i ko n .

tAs assessed by paramedics in the field.

present the optimal airway for respiratory manage-

ment in an apneic patient is an ET. However, the place-

ment of an ET requires a certain level of skill, the use of

a laryngoscope, and frequent use or retraining to

maintain proficiency. Because of the difficulties with

learning and maintaining ET intubation skills, alter-

native methods for ventilation were developed. In

1968, Don Michael et alu described a device that

occluded the esophagus and simultaneously allowed

ventilation of the lungs. This device with madifications

is now known as the esophageal airway EOA and

EGTA). Since its introduction, the esophageal airway

has been used in approximately

two

million cardiopul-

monary arrests.

The advantages of the esophageal airway over the

ET have been stated to be the relative ease and speed

of insertion, shorter time required b r raining and skill

maintenance, lack of need for a laryngoscope, and its

safer insertion in patients with cervical trauma. -C I . ~n

addition, in our program, training is less costly for the

esophageal airway. In communities where resources

for training are limited, the esophageal airway would

be an attractive alternative for short-term prehospital

airway management i its efficacy could be docu-

mented.

To determine whether the EA is a suitable alter-

native to the ET in the prehospital cardiopulmonary

arrest patient, we evaluated these different airway

managements in patients experiencing out-of-hospital

cardiopulmonary arrest. Our results (Tables6 through

9) showed that patients randomized to or receiving an

EGTA or ET had similar survival rates, incidence of

congestive heart failure, neurologic residual, and aspi-

ration pneumonia. However, analysis between patients

correctly randomized to the T and patients correctly

randomized to the EGTA showed a significantly higher

incidence of neurologic residual that occurred during

the initial cardiopulmonary arrest in the ET patients.

This finding would tend to favor the EA. However,

because of our multiple statistical analyses and because

this difference

w s

not present between patients ran-

domized to or receiving the EGTA or ET, we do not

HEST 9 1

JULY.

11986

93



5/7

wish to over emphasize it.

Because patients in our randomized study could

receive the alternative airway if difficulties were en-

countered with the initial airway, we evaluated an

additional

125

patients who had an esophageal airway

attempted (group 3) and who had no chance to receive

the ET. This was done to see if such patients were at a

significant disadvantage. These (group 3) patients had

similar survival rates, incidence of congestive heart

failure, and neurologic residual as patients randomized

to or receiving the ET (Thbles 6 through 8).

The patients randomized to either the ET or EGTA

were comparable except for age (Table 3). The EGTA

patients (group 2) were older (69 vs 65years). We think

that because of the multiple analyses done on these

patients, the significance of this finding also should not

be over emphasized. Age was not different between

the patients correctly randomized to or receiving the

ET or the EGTA. The older age in the patients

randomized to the EGTA might, if anything bias the

results against the EGTA.

Table 3 shows the downtime and time from arrest to

paramedic arrival in groups 1, 2 and 3. While there

was

no significant difference among the groups, the

mean time is shorter in group 3 patients. Since groups

1and 2 patients were treated by the same paramedics

in the same locale, one would expect these response

times to be almost identical. The shorter mean times in

the group 3 patients, while not statistically rent

from group 2 patients, may be due to these paramedics

responding to a slightly different locale.

Our study also showed that the duration of the

intubation procedure and the percent of successful

airway placements were similar in

all groups analyzed

(Table 10). However, more patients receiving an EGTA

were judged to be inadequately ventilated compared

to patients receiving an ET. One explanation is that,

despite proper placement of the EGTA, providing

adequate ventilation is difficult. The majority of the

paramedics stated that adequate ventilation

was

easier

to provide with the ET because a tight Eace mask seal is

not needed, and ventilation could be done with one

hand. While this may be correct, we believe that our

data do suggest that adequate ventilation can be

obtained with an EA. In support, we found that

oxygenation and ventilation assessed by ABG mea-

surements were similar in patients receiving the

esophageal airway or endotracheal tube. In addition,

paramedics trained only in EOA intubation felt they

were able to adequately ventilate 90 percent of their

patients. This is similar to the 89 percent of patients

who had an ET attempted that were felt to be ade-

quately ventilated. Therefbre, we believe that some of

the patients in the randomized study who were fklt to

be inadequately ventilated with the EGTA may have

received the ET due to paramedic bias.

Table U Complicationr

in dimta

eceiving th EGT

nd ET

ET tu e

n

esophagus

1 ET

balloon above cords

1

ET

in right

mainstem

1

Esophageal

tear

To determine paramedic preference, we asked each

paramedic participating in the randomized study to

select the tube he would choose

if llowed

to use only

one, the EGTA or ET. While most paramedics felt it

was

easier to ventilate the patient once the ET was in

place, only 18 of the 35participating paramedics chose

the ET.

The incidence of complications was similar in pa-

tients receiving the ET or the EGTA (Thble ll . Our

study suggests that both tubes can be used with only

minimal complications in the prehospital setting. This

finding is supported by

This article is the first to compare patients prospec-

tively randomized to the ET or to the EGTA in regard

to morbidity and mortality. Despite problems with

randomization and potential paramedic bias, we be-

lieve our results re reliable. To assure that incorrectly

randomized patients did not distort our data, we did

extensive intergroup analysis as outlined in our

method section. Despite this extensive analysis, no

major differences were detected between the groups.

Where differences existed, they were noted. While

our data showed essentially no difference in morbidity

and mortality in the groups analyzed, we cannot

exclude the possibility of a beta error;

ie that a

significant difference w s not found because the sam-

ple size

was

too small. In further support of our claim

that there is no difference in morbidity and mortality,

we have recently submitted data from a retrospective

study of 317 patients who had an out-of-hospital

cardiopulmonary arrest and fbund no difference in

morbidity and mortality between patients receiving an

EOA or ET Also, Shea et a18 recently retrospectively

evaluated 296 nonrandomized patients with out-of-

hospital cardiopulmonary arrests and found similar

survival rates and neurologic residual in patients

receiving the EGTA or ET.

While as stated we believe our data provide evi-

dence that the EA may be a suitable alternative to the

ET in cardiopulmonary arrest patients, our study has

many limitations. There were a significant number

(17.1 percent) of crossovers in this study, making

interpretation of our data difficult and requiring us to

perform multiple group comparisons. We cannot ex-

clude the possibility that some very ill patients who

could have received the EGTA actually received the

ET because of paramedic bias (Table 2). This, of course,

Pmhop adopulmonaty rrest

Oddenberg

t d



6/7


7/7

esophageal gastric tube airway vs endotracheal tube in prehospital cardiopulmonary arrest chest 1986

Documents