Br. J. Anaesth.-1994-LEITH-552-8 (1)

8/16/2019 Br. J. Anaesth.-1994-LEITH-552-8 (1)

1/7

British Journal of

Anaesthesia

1994; 73: 552-558

COMMENTARIES

Extradural infusion analgesia for postoperative pain relief

S.

LEITH,

R. G.

WHEATLEY,

I. J. B.

JACKSON,

T. H.

MADEJ AND

D.

HUNTER

Summary

W e describe 4-yr experience providing extradural

infusion analgesia in a distr ict hospital for treatment

of postoperative

pain.

A total of 770 patients

recovering from major surgery were treated on

general surgical wards between April 1989 and

March 1993. The results of a retrospective audit

showed

that pain

control,

assessed with both

a

visual analogue scale (VAS score (0-10 cm)) and a

verbal rating scale (VRS), was good. At rest, more

than 80 of patients scored pain as absent or mild

(VAS

score 0-3) during the first 24 h, with only 4

experiencing

severe pain (VAS score 7-10). On

movement, 50 of patients had good pain control

(VAS

score 0-3) while

20

of

patients experienced

severe

pain (VAS score 7-10). Minor

com plications

such as

emetic

sequelae

and

pruritis

were c ommon ;

these

conditions were mild and rarely required

treatment.

Hypotension (


2/7

Extradural analgesia

for

postoperative pain

553

between patients receiving PCA and extradural

analgesia.

To

minimize this risk

and the

time

involved in refilling the syringes, a new system was

introduced. The updated system used an ambulatory

PCA device (Bard Ltd, Crawley, W. Sussex, UK)

which included a 250-ml infusion bag. T his system

was again updated

in 1992 and now

extradural

infusion analgesia

is

maintained using ambulatory

Abbott Pancretec Provider pumps (Abbott Labora-

tories Ltd ) which use a 500-ml infusion bag prepared

by

our

pharm acy. T hi s system differs physically

from the PCA system, avoiding confusion, and

provides analgesia for up to 5 days w ithout refilling.

SOLUTIONS

Initially

the

extradural infusate consisted

of a

solution of 0.15% bupivacaine with diamorphine

5

mg in

50

ml

(0.01 % ).

The

dose

of

diamorphine

was reduced to 2.5 mg in 50 ml (0.005 % ) in patients

more than

65

yr of age.

In

1990, after the introdu ction

of the larger infusion bags, the dose of diamorphine

added to 0.15% bupivacaine was reduced and all

patients, irrespective

of

age, received d iamo rphine

at

the lower concentration of 0.005

(25 mg in

500 ml). T his change

was

designed

to

reduce

the

daily dose of diamorphine administered. The first

year's experience with

the

service demonstrated that

the lower concentration was effective in the majority

of patien ts. Infusion rates varied from 4 to 10 ml h

1

.

NURSING OBSERV TIONS

While patients were

in the

recovery room, nursing

observations (ventilatory frequencies, pain score

(0-3)

[2],

derma tomal level, arterial pressu re

and

sedation score (0-3)) were assessed at 15-min

intervals. In the ward, similar observations were

made by the nursing staff, hourly for the first 4 h and

then 4-hourly for the duration of the infusion. Each

set

of

observations

was

signed

for by the

recording

nurse. The extradural forms were assessed daily on

die A PT ward round

for

occurrence of com plications

and completeness of recording. The compliance of

die nursin g staff

in

recording

the

data was extremely

high (approaching 100 % ). In the first year,

1989-1990, sedation

was not

assessed

but a

simple

sedation score was included subsequently in the

postoperative observations. Each extradural chart

contained guidelines

for the

management

of

complications. Guidelines were available also on the

wards

and the

nursing staff attended in-house

training on the management of extradurals (See

appendix). A mem ber

of

th e

APT

was available

on a

24-h basis (including weekends) to give advice and

treatment if required. Bolus doses and changes in the

infusion ra te could be given only by a member of th e

pain team.

CUTE P IN TE M W RD ROUND

After referral

to the APT,

each patient receiving

extradural infusion analgesia was seen twice daily on

the morning and afternoon ward rounds for the

duration of treatment. The acute pain service (AP S),

which

has

been

in

operation since

1989

[2],

is

based

on the existing obstetric anaesthetic service and is

provided

by

three consultant a naesthetists,

an

acute

pain sister, and die on-call obstetric trainee an-

aesthetist.

An

audit

of the

entire workload

of the

A PS

for the

year 1989-1990 showed that approxi-

mately

30

of each con sultant session was spent on

the ward rounds and work related to the provision of

the service (both ex tradural analgesia and PCA). The

trainee anaesthetists spent

26% of

their working

time on acute pain work. A substantial part of the

time spent

by the

trainee anaesthetist

in the

year

1989—1990 was refilling and attending to problems

with

the

PCA p um ps which were also being used

for

extradural analgesia during this year. This pro-

portion of work has decreased over the past 3 yr widi

the increasing involvement of die nursing staff.

However, the increase in die number of patients

referred

to the

service

(660 in

1989-1990,

1362 in

1992-1993)

has

kept

the

workload

of the

trainee

anaesthetist constant.

The

assessment

at 24 h in-

cluded: (1) VAS score (0-10) for pain at rest; (2)

VAS score (0-10)

for

pain

on

movement;

(3) VRS

for satisfaction (poor, fair, adequate, good, excel-

lent); and (4) VRS for pain (absent, mild, dis-

comforting, distressing, excruciating). Both (2) and

(4) were introduced in 1990 after the first year's

experience with

die

service.

The

inclusion

of a

movem ent component improved the measurement of

efficacy without significantly increasing die workload

of the APT. In addition, the incidences of emetic

sequelae

and

pruritus were recorded.

Results

P TIENT D T

In die 4-yr period, April 1989 to March 1993, 770

patients received extradural infusion analgesia

on

general wards in York District Hospital (table 1).

T he num ber of patients receiving extradural infusion

analgesia increased from

150 in

1989-1990

to 226

in 1992-1993. The mean duration of treatment

increased from 55 to 80 h over die same period.

The majority of patients (65-82% ) were referred

to

the

service after general abdominal su rgery .

Non-

surgical referrals included pancreatitis

and

chest

trauma.

ASA status ranged from ASA I to IV with the

majority

of

patients

(73%) of ASA

I I -IV status.

There was an annual decrease in the percentage of

ASA

I

patients receiving extradural infusion

an-

algesia (40% in 1989-1990, 16% in 1992-1993).

Patients satisfaction with the method of analgesia

was high, widi more dian 88

of

patients regarding

it as adequate, good or excellent.

P IN SSESSMENT

VAS

SCORE

0 - 10 )

In the first year of the service, 75% of patients

receiving extradural analgesia scored pain as mild,

with only

7%

experiencing severe pain (table

2).

After the introduction in 1990 of separate VAS

b y g u e s t onM a y1 8 ,2 0 1 6

h t t p : / / b j a . oxf or d j o ur n a l s . or g /

D o wnl o a d e d f r om

http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/


3/7

554

British Journal of Anaesthesia

Table 1

Patient data (mean (SD or range))

Male

Female

N o .

of patients

Age(yr)

Weight (kg)

Duration of treatment (h)

Dose of opioid mg/24 h (1989)

Dose of opioid mg/24 h (1990-93)

433

65(14-92)

70(13)

70 (1-240)

9.7

7.23

33 7

61(17-99)

63(13)

64(1-576)

9.7

6.8

Table 2

Twenty-four-hour pain assessment (VAS 0-10) at rest

(% patients). M989-1990 single VAS at 24 h

VAS score

1989-90 1990-91 1991-92 1992-93

0-3 (mild)

4—6 (moderate)

7-10 (severe)

75*

18 *

7*

91

9

0

93

6

1

83

13

4

Table 3

Twenty-four-hour pain assessment (VAS 0—10) on

movement (% patients)

VAS score

1989-90 1990-91 1991-92 1992-93

0-3 (mild) — 57 61 47

4-6 (moderate) — 36 23 32

7-10 (severe) — 14 16 21

Table

5 Patient assessment of pain severity (VRS)

(% patients)

1989-90 1990-91 1991-92 1992-93

Absent —

Mild —

Discomforting —

Distressing —

Excruciating —

21

38

29

10

2

25

41

19

11

4

26

25

32

14

3

Table 6

Incidence of vomiting in the first 24 h (% patients)

1989-90 1990-91 1991-92 1992-93

Females

Males

26

13

23

2

19

11

24

16

scores (0-10)

for

assessing pain

at

rest

and on

movement,

the

results showed that more than

80

%

of patients scored their pain as mild

at

rest (table

2),

but this declined to 50% on movement, with

16-21 % experienc ing severe pain (table 3). Mean

pain scores varied from 0.85 to 1.3 at rest and from

3.14

to

3.84

on

mov emen t (table 4). On assessing

the

severity

of

pain experienced

by

patients using

the

V R S , 50%

experienced

no

pain

or

mild pain

and

less than

4%

described dieir pain

as

excruciating

(table 5).

MINOR COMPLICATIONS

Hypotension

secondary

to

relative hypovolaemia was

a frequent finding;

34% of

patients

had a

systolic

arterial pressure

of

less tha n 100 mm

Hg at

some

time in the first 24 h (see discussion).

Mean ventilatory frequency

varied betw een 14

and

15 b.p.m ., with 2.6

of

patien ts w ith frequencies

of

8 b.p.m.

or

less.

Emetic sequelae. The

incidence

of

vomiting

in

females

was

(19-26 % ) compared with

males

(2-16 % )

(table

6).

Pruritis.

T he incidence of pruritus varied between

12 and 2 2% , but treatment was rarely required.

Prolonged block Two patients in the first year of

the study developed prolonged unilateral motor

block lasting 4—10 days which delayed mobilization

[2].

Both patients

had

lumbar extradural catheters

sited

for

pain relief after abdominal hysterectomy.

While formal assessment

of

motor block

was not

made, it was noted frequently on the APT ward

round that after hysterectomy, patients with extra-

dural infusion analgesia had difficulty mobilizing

because

of

motor block.

It has

since become

departmental policy

to use PGA for

pain control

after this type

of

surgery, avoiding the use

of

lumbar

extradural catheters.

Urinary retention

was

not

studied as most patients

receiving extradural infusion analgesia were

catheterized before operation.

MAJOR COMPLICATIONS (TABLE 7)

In the 4-yr period 1989-1993, four patients (0.52 % )

developed

severe respiratory depression

associated

with excessive sedation, airway obstruction and

hypoxaemia. One patient subsequently died (patient

N o. 3).

High block

One p atient with

a

thoracic extradural

catheter

in situ for

analgesia after laparotomy

de-

veloped hypotension associated with slurred speech

because of block to the third thoracic dermatome

(patient No. 5).

Two patients (Nos 6, 7) with major postoperative

intra-abdominal haemorrhage

were managed

in-

adequately as hypotension was attribu ted incorrectly

Table 4

Twenty-four-hour VAS 0-10 (mean (range)).

1989-90 1990-91

Rest 2.2*

Movement —

0.85 (0-6)

3.23 (0-10)

M989-1990 single 24 h VAS 0-10

1991-92

1.02(0-9)

3.14(0-10)

1992-93

1.3 (0-10)

3.84 (0-10)

1990-93

1.05(0-10)

3.40 (0-10)

b y g u e s t onM a y1 8 ,2 0 1 6





4/7

Table 7 Serious complications with extradural infusion analgesia 1989-1993 . *Patients 3, 6 and 7 died while receiving extradural analgesia

Patient

No. Age Sex Past history Operatio n

Problem

Hours Ventilatory Infusion rate

postop.

freq. (b.p.m.) (ml h

1

) Treatment Outcome Remarks

B

I

3

79 F Obstructive

jaundice, atrial

fibrillation

82 F Hypertension

Laparotomy for

carcinoma of pancreas

Rt. hemicolectomy Ca.

of colon, transverse

incision

73 F Reasonably healthy Laparotom y abandoned

because of inoperable

gastric carcinoma

75 F Parkinson's disease, Laparotomy. Long

obesity midlinc incision,

apendicetomy

76 M Reasonably healthy Laparotom y abandone d

because of inoperable

gastric carcinoma

Unrousable, sedation 51 10

score = 3, previous

scores = 1

8 h postop. sedation score 21 6

= 2, infusion rate I

from 6 to 2 ml h

1

.

Patient remained

drowsy, sedation score

= 3 at 21 h

Drowsy postop. 7 20

Remained drowsy

despite decreases in

infusion r ate. 7 h

postop. sedation score

-

2. AP = 90/60 mm Hg

30 h postop. sedation 30 20

score = 3, AP =

80/50 mm Hg. Previous

scores = 1

16 h posto p. AP = 16 14

75/45 mm Hg, slurred

speech, level = T3

81 M Inferior MI 18 yr Cholecystectomy

previously, atrial

fibrillation

7

Reasonably healthy Abdominal-perineal

resection

10 h post op. hypotensiv e 10

AP = 90/60 mm Hg

12 h postop. hypotensi ve 12

AP = 85/45 mm Hg

I.v. naloxone Good, no adverse

sequelae

I.v. naloxone Good, no adverse

sequelae

I.v. fluids, Good recovery

Resp. depression can

occur at any time

Some elderly patients

are very sensitive to

opioids

infusion

discontinued

initially. At 13 h

postop. cardiac arrest

PM = Basal

congestion/aspiration

I.v. naloxone, Good, no adverse

i.v. fluids sequelae

Overscdated pt who

became hypoxaemic

due to fluid overload.

Aspiration occurring

as part of terminal

event

Resp. depression can

occur at any time

Infusion

discontinued,

i.v. fluids,

ephedrine

Infusion

discontinued,

i.v. fluids

Infusion

discontinued,

i.v. fluids

Good response to

treatment but patient

developed CCF and

pul. oedema 2 h later.

Treated successfully

Inadequately

resuscitated. Cardiac

arrest 2 h later.

PM—Large intra-

abdominal

haemorrhage

Inadequately

resuscitated. Cardiac

arrest 4 h later.

PM—Large intra-

abdominal

haemorrhage

CCF due to fluid

overload and rebound

vasoconstriction on

stopping infusion

Other causes of

hypotension should

be suspected when

there is an inadequate

response to simple

measures

As above

b y g u e s t onM a y1 8 ,2 0 1 6



http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/http://bja.oxfordjournals.org/


5/7

556


to the effects of extradural block. A surgical cause

was not considered.

Discuss ion

Efficacy and safety are major considerations in the

provision of pain

relief.

This is particularly relevant

with the use of invasive techniq ues such as extrad ural

infusion analgesia where hypotension and respir-

atory depression are potential complications. It is

therefore important to demonstrate a high benefit to

risk ratio.

EFFIC CY

T he VAS score (0-10) at 24 h has been used w idely

to quantify postoperative pain and has been shown to

be an acceptable indication of a patient's pain after

surgery [6]. The visual analogue and verbal de-

scriptive score techniques have been shown to

successfully assess the sensory intensity of exper-

imental pain [7]. These two methods of pain

assessment were used for the 3-yr period 1990-1993.

The results showed that more than 80% of our

patients were comfortable at rest (mean VAS 1.05),

with 50 % (47-54% ) being comfortable on move-

ment (mean VAS 3.4). Using the VRS to assess the

quality of pain experienced by the patient, more than

5 0 % (51-6 7% ) described their pain as absent to

mild, with less than 2 0% (12-17 % ) describing their

pain as distressing or excruciating. Direct com-

parison of our results with similar studies is difficult

because of the different methods used for pain

assessment [8,

9].

The use of a local anaesthetic with

an opioid in the extradural space results in better

pain relief than one of the agents alone [10], while

minimizing the side effects associated with the

administration of each drug alone. Administration

by infusion also produces effects in a more gradual

manner.

Extradural infusion analgesia provides improved

postoperative pain control compared with parenteral

methods. In a study using i.m. analgesia, Kuhn and

colleagues [3] found that only 13 % of patients had

little or no pain while 40 % described pain as very

painful. Similarly, Owen, McMillan and Rogowski

[4],

using a pain rating scale, found that only 26 of

patients described pain as mild or absent while 37 %

rated pain as unbearable or severe. Jayr and

co-workers, [11] in a recent study comparing the

effects of extradural bupivacaine and opioids with

parenteral opioids (s.c. morphine 2.5 mg h~')j found

that pain control was significantly better in the

extradural group than in the s.c. group, especially

during recovery and on the first and second post-

operative days. The results of our first year's

experience with the APT [2] showed that patients

receiving extradural analgesia were more likely to

have no pain or mild pain than those receiving PCA

(chi-square, P


6/7

Extradural analgesia for postoperative pain

55 7

results confirm the observation of Ready and co-

workers th at excessive sedation is a valuable clinical

sign of impending respiratory depression [16]. In

this series, extradural infusion analgesia tende d to be

used in older, high-risk patients with associated

cardiorespiratory disease, undergoing extensive sur-

gery, who were more prone to the depressive effects

of opioids. In patients receiving morphine orally or

parenterally (i.v., i.m., s.c.) the incidence of life-

threatening respiratory depression has been reported

to be 0.9% [17], which is higher than in our study.

Opioid-induced respiratory depression is not the

only hazard to the surgical patient. Myocardial

ischaemia secondary to tachycardia and hypertension

with poor analgesia may pose a greater risk to the

elderly surgical patient than the rare case of res-

piratory depression w ith extradural opioids. In high-

risk patients undergoing non-cardiac surgery, early

postoperative myocardial ischaemia is an important

correlate of adverse cardiac outcome [18].

Hypotension secondary to hypovolaemia from

inadequate fluid replacement and sympathetic vaso-

dilatation was a frequent finding: 34 of patients

devejoped hy potension of less than 100 mm Hg at

some time during the first 24 h. W hile Hob bs and

R oberts [8] recorded an incidence of only 2.6 % for

hypotension less than 95 mm H g, Jayr and co-

workers [11] found that 2 1 % of his patients with

extradura l infusion analgesia had hy potensio n of less

than 80 mm Hg which responded rapidly to treat-

ment. This is significantly greater (P


7/7

558


4. Arterial pressure

I F B P I S L E S S T H A N T H E N

I N C R E A S E I .V . I N F U S I O N R A T E

L I E P A T I E N T F L A T

A D M I N I S T E R O X Y G E N

BLEEP M TERNITY N ESTHETIST

5. Level = dermatome at which cold sensation

returns

Dermatomes: T5— Xiphistemum

T10—Umbilicus

T12—Pubis

If level is greater than then inform maternity anaesthetist

PROBLEMS

If there are any problems with the extradural please contact the

junior or consultant maternity anaesthetist.

N.B. Extradural analgesia is an effective method of pain relief

which relies on the adm inistration of local anaesthetic and opioid

drugs via a catheter in the extradural space. The settings on the

pum p are chosen by the anaesthetist with reference to the patient's

age, general condition and type of surgery. The settings should

only be altered by members of the acute pain team.

References

1. Report of a W orking Party of the C ommission on the

Provision of Surgical Services. Pain after Surgery. London:

Royal College of Surgeons of England and College of

Anaesthetists, 1990.

2. Wheatley R G, Madej TH , jackson IJB, Hunter D . Th e first

year's experience of an acute pain service.

British

Journal of

naesthesia 1991; 67: 353-359.

3.

Kuhn S, Cooke K, Collins M , Jones JM , Mucklow JC.

Perceptions of pain relief after surgery.

British Medical

Journal 1990; 300: 1687-1690.

4. Owen H, McMillan V, Rogowski D. Postoperative pain

therapy: a survey of patients' expectations and their

experiences. Pain 1990; 41 : 303-307.

5.

M organ M. Epidural and intrathecal opioids. naesthesia and

Intensive Care 1987; 15: 60-67.

6. Liu W H D , Aitkenhead AR . Comparison of contemporaneous

and retrospective assessment of postoperative pain using the

visual analogue scale. British Journal of naesthesia 1991; 67 :

768-777.

7. Duncan GH , Bushnell M C, Lavinge GJ. C omparison of

verbal and visual analogue scales for measuring the intensity

and unpleasantness of experimental pain. Pain 1989; 37:

295-303.

8. H obbs GJ , Roberts FL . Epidural infusion of bupivacaine and

diamorphine for postoperative analgesia—Use on general

surgical wards.

naesthesia

1992; 47: 58-62.

9. Stuart-Taylor M E, Billingham I S, BarTert RF , Church J J.

Epidural diamorphine for postoperative analgesia: Audit of

a

nurse administered service to patients in a District General

Hospital. British Journal

of

naesthesia 1992; 68: 429-432.

10. George KA , Chisakuta AM , Gamble JAS , Browne GA.

Thoracic epidural infusion for postoperative pain relief

following abdom inal aortic surgery: bupivacaine, fentanyl or

a mixture of both? naesthesia 1992; 47: 388-394.

11. Jayr C, Thomas H , Rey A, Farhat F, Lasser P, Bourgain JL.

Postoperative pulmonary complications: epidural analgesia

using bupivacaine and opioids versus parenteral opioids.

nesthesiology 1993; 78: 666-676.

12. Laveaux MM D , Hasenbos MA WM , Harbers JBM , Liem T .

Thoracic epidural bupivacaine plus sufentanil: High

concentration/low volume versus low concentration/high

volume. Regional naesthesia 1993; 18: 39-43.

13.

W hite MJ, Berghausen EJ, Dumont SW , Tseuda K ,

Schroeda JA , Vogel RL , Heine M F, H aung K C. Side effects

during continuous epidural infusion of morphine and

fentanyl. Canadian Journal

of

naesthesia 1992; 39: 576-582.

14. Semple AJ, Macrae DJ, M unishankarappa S, Burrow L M ,

M ilne MK , Gra nt IS . Effect of addition of adrenaline to

extradural diamorphine analgesia after Caesarean section.

British Journal of

naesthesia 1988; 60: 632-638.

15.

Lee A, Simpson D, Whitfield A, Scott DB. Postoperative

analgesia by continuous extradural infusion of bupivacaine

and diamorphine. British Journal of naesthesia 1988; 60:

845-850.

16. Ready BL, Oden R, Chadwick HS, Benedetti C, Rooke G,

Caplan R, Wild L. Development of an anesthesiology based

postoperative pain management service. nesthesiology 1988;

68:

100-106.

17. M iller R R, G reenblart D J. Drug effects in Hospitalised

Patients. New York: Joh n W iley and Son s, 1976; 151-152.

18. Mergano D T , Browner W S, Hollenberg M , London M J,

Tu bau J F , Tateo IM , SPI Research Grou p. Association of

perioperative myocardial ischemia with cardiac morbidity

and mortality in men undergoing noncardiac surgery.

New

England Journal of Medicine 1990; 323: 1781-1788.

b y g u e s t onM a y1 8 ,2 0 1 6



Documents

Br. J. Anaesth.-1994-LEITH-552-8 (1)