The pre-operative ECG in day surgery: a habit?

FORUM

Effect of epidural bupivacaine on the relationship

between the bispectral index and end-expiratory

concentrations of des¯urane

P. Hans, J. P. Lecoq, J. F. Brichant, P. Y. Dewandre and M. Lamy

University Department of Anaesthesia and Intensive Care Medicine, CHR de la Citadelle, University Hospital,

4000 Liege, Belgium

Summary

We compared the relationship between the bispectral index and end-tidal des¯urane concentrations

in 20 patients undergoing elective surgery. Patients received epidurally either 10 ml saline (group S)

or 10 ml bupivacaine 0.125% with epinephrine 1/800 000 (group B) before induction of

anaesthesia with sufentanil (0.15 mg.kgÿ1) and propofol (2 mg.kgÿ1); muscle relaxation was

obtained with cisatracurium (0.2 mg.kgÿ1). Patients lungs were ventilated to maintain end-tidal

des¯urane at 3% in O2/N2O (50/50) until 5 min after skin incision, followed by two consecutive

10 min periods at end-tidal des¯urane 6% and 9%. bispectral index values were recorded before

induction, at 3% des¯urane before and 5 min after skin incision, and at 6% and 9% end-tidal

des¯urane. Bispectral index decreased with increasing end-tidal des¯urane concentration (ANOVA:

p< 0.05). The decrease in bispectral index was signi®cant between pre-induction, 3% and 6%

des¯urane. No signi®cant difference was observed at 3% des¯urane before and after skin incision,

or between 6 and 9% des¯urane. The relationship between bispectral index and end-tidal

des¯urane concentration was ®tted by a linear regression in each group. No signi®cant difference

in bispectral index was observed between the groups at any time. We conclude that bispectral

index decreases with increasing des¯urane concentration and that this relationship is not affected

by epidural 0.125% bupivacaine.

Keywords Anaesthesia; depth. Monitoring; bispectral index. Anaesthetic, volatile; des¯urane.

Anaesthetic, local; bupivacaine.

......................................................................................

Correspondence to: Professor P. Hans

Accepted: 1 January 1999

Des¯urane is a volatile anaesthetic agent known to pro-

duce electroencephalographic (EEG) changes comparable

with those observed with equipotent levels of iso¯urane

and sevo¯urane [1±3]. Whereas the EEG is not commonly

recorded in routine anaesthetic practice, the use of the

bispectral index derived from the bispectral analysis of the

EEG is gaining increasing popularity. Bispectral index

values are correlated with the hypnotic component of

anaesthesia and bispectral index has been proposed as a

guide to the administration of intravenous and volatile

anaesthetic agents [4±6]. However, the effect of des¯urane

on bispectral index has never been investigated. The present

study was designed ®rst, to examine the relationship

between bispectral index and end-tidal des¯urane concen-

tration, and second, to determine if that relationship was

affected by epidural bupivacaine.

Methods

After receiving their informed consent to this protocol

approved by the hospital ethics committee, 20 ASA I

patients (12 males, 8 females), aged 17±62 years, were

enrolled. All were scheduled to undergo elective lumbar

disc surgery under general anaesthesia combined with

epidural analgesia. They were allocated to one of two

groups of 10 patients according to the solution given

Anaesthesia, 1999, 54, pages 899±908................................................................................................................................................................................................................................................

899Q 1999 Blackwell Science Ltd

epidurally: saline (group S) or bupivacaine (group B).

Random allocations to these groups were made according

to a computer-generated number table.

Premedication consisted of alprazolam 1 mg and atro-

pine 0.5 mg given orally 1 h before surgery. Upon arrival

in the operating room, patients were monitored with ECG,

pulse oximetry and noninvasive blood pressure (Datex

AS-3). EEG electrodes (ZipPrep: Aspect Medical Systems)

were placed on the scalp using the bipolar frontomastoid

montage con®guration. The impedance of each electrode

was > 2 Q. Bispectral index was recorded continuously

using an Aspect A-1000 System.

After intravenous cannulation, an epidural catheter was

inserted at the L1±L2 intervertebral space, with 4 cm left in

the epidural space. After intravenous administration of

500 ml of lactate±Ringer, the study solution was injected

epidurally: 10 ml solution of bupivacaine 0.125%�epinephrine 1/800 000 in group B (12.5 mg bupivacaine),

or 10 ml solution of saline in group S. Patients

were anaesthetised with sufentanil (0.15 mg.kgÿ1) and pro-

pofol (2 mg.kgÿ1), and paralysed with cisatracurium

(0.2 mg.kgÿ1) for intubation. The lungs were ventilated

with 50% nitrous oxide and 3% end-tidal des¯urane in

oxygen. Patients were then placed in the prone position

and surgery started within 15 min of induction of anaes-

thesia. Five minutes after skin incision, the inspired

concentration of des¯urane was increased to achieve a

6% end-tidal concentration for 10 min equilibration and

then 9% for 10 min. Thereafter, the end-tidal des¯urane

concentration was reduced and adjusted to keep an appro-

priate level of anaesthesia. Bispectral index was recorded

before the induction of general anaesthesia (control), at 3%

end-tidal des¯urane concentration before and 5 min after

skin incision, and at the end of each equilibration period of

6% and 9% des¯urane. The level of sensory block of the

epidural analgesia was assessed at awakening in each patient.

Age, weight and height were compared between groups

using an unpaired Student's t-test. The distribution by sex

was analysed using the Chi-squared test. Bispectral index

values were analysed in both groups using ANOVA for

repeated measurements and Turkey±Kramer post test.

The relationship between bispectral index and end-tidal

des¯urane during surgery was tested in each group using

least squares regression. The effect of epidural bupivacaine

was analysed by comparing bispectral index values

between groups at each end-tidal concentration of des-

¯urane with the Student's t test. A level of p< 0.05 was

considered statistically signi®cant.

Results

Demographic data were similar in both groups (Table 1).

At awakening, patients from group B had an upper level

of analgesia ranging between T6 and T10. Bispectral index

values recorded in both groups at each end-tidal des¯urane

concentration are shown in Table 2. Patients from both

groups showed a decrease in bispectral index with increas-

ing concentrations of des¯urane (Fig. 1). Bispectral index

decreased signi®cantly between pre-induction, 3% and 6%

end-tidal des¯urane concentration. No signi®cant differ-

ence was observed in bispectral index values before and after

skin incision at 3% des¯urane, or between 6% and 9%

des¯urane concentration. Least squares regression analysis

demonstrated a signi®cant linear relationship between

bispectral index and end-tidal concentration of des¯urane

in group S (r�ÿ0.39, p� 0.03) and group B (r�ÿ0.43,

p� 0.01). Comparison of bispectral index values between

the two groups at each end-tidal des¯urane concentration

did not reveal any signi®cant difference.

Discussion

In the present study we found that increasing the end-tidal

des¯urane concentration from 3 to 9% produced a decrease

Forum Anaesthesia, 1999, 54, pages 899±908................................................................................................................................................................................................................................................

900 Q 1999 Blackwell Science Ltd

Surgery

End-tidal Control Pre-incisionDes¯urane 0% 3% 3% 5% 9%

Group B 97.3 (1.9) 41.8 (5.2)* 41 (4.5)* 31 (5.3)*²³ 30.5 (8.8)*²³Group S 95.4 (2.9) 40.2 (10.4)* 41.3 (7.1)* 28.1 (3.8)*²³ 32.2 (8.9)*³

Data are means (SD). * Signi®cantly different from control (0% des¯urane); ² signi®cantlydifferent from 3% des¯urane pre-incision; ³ signi®cantly different from 3% des¯urane afterincision.

Table 2 Bispectral index values in bothgroups

Table 1 Demographic summary of subjects enrolled in the studyprotocol

Group B Group S

n 10 10Male/female 8/4 2/6Age; years 42.2 (12.4) 41.5 (15.0)Weight; kg 76.7 (14.6) 76.6 (14.2)Height; cm 174 (9.3) 170.5 (7.0)

Values are mean (SD). No signi®cant difference between groups.

in bispectral index ®tted by a linear regression and that this

relationship was not affected by epidural bupivacaine.

To the best of our knowledge, the present study is the

®rst to report the effect of des¯urane on bispectral index.

The decrease in bispectral index as a function of des¯urane

concentration is in agreement with the electroencephalo-

graphic effects of des¯urane in humans. It has been

demonstrated that des¯urane at increasing concentrations

produced a pattern of decreasing EEG activity and induced

prominent burst suppression at 1.24 MAC (9% end-tidal

concentration) and higher [1]. Des¯urane has also been

reported to produce dose-related increments in EEG burst

suppression similar to those produced by iso¯urane or

sevo¯urane at equipotent concentrations [2, 3]. In our

study, no signi®cant difference was observed in bispectral

index values between 6 and 9% end-tidal des¯urane

concentration. This ®nding could be explained either by

the release of catecholamines, which may be associated

with the rapid increase of des¯urane concentration, or by

the concomitant use of nitrous oxide. Indeed, rapid

increases in des¯urane concentration can increase sympa-

thetic activity and plasma epinephrine concentrations,

probably as a result of activation of tracheopulmonary

and systemic receptors [7, 8]. This catecholamine release

may affect the electrical activity of the brain [9]. On the

other hand, nitrous oxide is also known to reduce the

cortical depressive effect of hypnotic agents. It has been

shown that substitution of nitrous oxide for 3% des¯urane

was associated with increased EEG activity and reduced

the degree of EEG suppression relative to the equipotent

des¯urane concentrations and oxygen [1]. The ®ndings of

this study are supported by other reports in the literature

addressing the relationship between bispectral index and

other volatile anaesthetics such as iso¯urane and sevo¯ur-

ane. Glass et al. [5] reported a linear decrease in bispectral

index with increasing end-tidal iso¯urane concentration in

oxygen until 1.2%. Katoh et al. [10] described a linear

decrease in bispectral index with increasing end-tidal

sevo¯urane concentration from 0.5 to 1.5% in oxygen,

and did not observe any further decrease in bispectral

index beyond 1.8% end-tidal sevo¯urane.

The second result of our study is the absence of an effect

of epidural bupivacaine 0.125% on the relationship

between bispectral index and des¯urane. Bispectral index

values in patients receiving epidural bupivacaine were not

signi®cantly different from those observed in patients

receiving epidural saline at equivalent des¯urane concen-

trations. This result suggests that the bispectral index is an

index of the hypnotic component of anaesthesia and that

the level of analgesia was adequate in both groups. Epi-

dural bupivacaine 0.125% combined with volatile anaes-

thesia is able to control weak painful stimulations during

lumbar disc surgery. However, the intravenous administra-

tion of sufentanil at induction, followed by des¯urane and

nitrous oxide for maintenance, has also been shown to

provide appropriate analgesia during lumbar disc surgery,

even in patients who do not receive epidural bupivacaine.

The interest of an epidural catheter during such a surgical

procedure is limited and its major advantage is essentially

to provide high quality postoperative analgesia. Therefore,

we cannot infer what could happen in a more painful

procedure requiring higher bupivacaine concentrations.

Indeed, bispectral index could be affected by an inadequate

level of analgesia and show a different pattern in the

presence or absence of ef®cient epidural analgesia. In

contrast, it has also been shown that the administration

of opioids confounds the bispectral index as a measure of

Anaesthesia, 1999, 54, pages 899±908 Forum................................................................................................................................................................................................................................................


Figure 1 Linear regression between bispectral index and end-

tidal-des¯urane in groups B and S.

anaesthetic adequacy, at least when movement response to

skin incision is taken as the primary endpoint [4]. In this

respect, the absence of modi®cation in bispectral index

values before and after incision in patients receiving

epidural saline also suggests that bispectral index essentially

re¯ects the hypnotic effect of anaesthetic agents, or that

the level of analgesia was appropriate in those patients.

In the conditions in the present study, des¯urane produced

a dose-related decrease in bispectral index ®tted by linear

regression and this relationship was not affected by epidural

bupivacaine. Therefore, bispectral index may be used as a

reliable guide to the depth of des¯urane anaesthesia in

patients who bene®t from epidural analgesia.

References

1 Rampil IJ, Lockhart SH, Eger EI, Yasuda N, Weiskopf RB,

Cahalan MK. The electroencephalographic effects of

des¯urane in humans. Anesthesiology 1991; 74: 434±9.

2 Hoffman WE, Edelman G. Comparison of iso¯urane and

des¯urane anesthetic depth using burst suppression of

electroencephalogram in neurosurgical patients. Anesthesia

and Analgesia 1995; 81: 811±16.

3 Schwender D, Daunderer M, Klasing S, Finsterer U,

Peter K. Power spectral analysis of the

electroencephalogram during increasing end-expiratory

concentration of iso¯urane, des¯urane and sevo¯urane.

Anaesthesia 1998; 58: 335±42.

4 Sebel PS, Lang E, Rampil IJ, et al. A multicenter study of

bispectral electroencephalogram analysis for monitoring

anesthetic effect. Anesthesia and Analgesia 1997; 84: 891±9.

5 Glass PSA, Bloom M, Kearse L, Rosow C, Sebel P,

Manberg P. Bispectral analysis measures sedation and

memory effects of propofol, midazolam, iso¯urane, and

alfentanil in healthy volunteers. Anesthesiology 1997; 86:

836±47.

6 Struys M, Versichelen L, Mortier E, et al. Comparison of

spontaneous frontal EMG, EEG power spectrum and

bispectral index to monitor propofol drug effect

emergence. Acta Anaesthesiologica Scandinavica 1998; 43:

94±102.

7 Weiskopf RB, Eger II EI, Daniel M, Noorani M.

Cardiovascular stimulation induced by rapid increase in

des¯urane concentration in humans results from activation

of tracheopulmonary and systemic receptors. Anesthesiology

1995; 83: 1173±8.

8 Moore MA, Weiskopf RB, Eger II EI, Noorani M,

McKay L, Damask M. Rapid 1% increases of end-tidal

des¯urane concentration to greater than 5% transiently

increase heart rate and blood pressure in humans.

Anesthesiology 1994; 81: 94±8.

9 Kanaya N, Nakayama M, Fujita S, Namiki A.

Haemodynamic and EEG changes during rapid-sequence

induction of anaesthesia. Canadian Journal of Anaesthesia

1994; 41: 699.

10 Katoh T, Suzuki A, Ikeda K. Electroencephalographic

derivatives as a tool for predicting the depth of sedation

and anesthesia induced by sevo¯urane. Anesthesiology 1998;

88: 642±50.

11 Gan TJ, Glass PS, Windsor A, et al. The BIS Utility Study

Group. Bispectral index monitoring allows faster

emergence and improved recovery from propofol,

alfentanil, and nitrous oxide anesthesia. Anesthesiology 1997;

87: 808±15.

FORUM

The effects of ropivacaine hydrochloride on coagulation

and ®brinolysis An assessment using thromboelastography

J. M. Porter,1 J. McGinley,2 B. O'Hare2 and G. D. Shorten1

1 Department of Anaesthesia & Intensive Care Medicine, Cork University Hospital, University College Cork,

Wilton, Cork, Republic of Ireland

2 Department of Anaesthesia & Intensive Care Medicine, Our Lady's Hospital for Sick Children, Crumlin,

Dublin, Republic of Ireland

Summary

Amide local anaesthetics impair coagulation by inhibition of platelet function and enhanced

®brinolysis. The potential therefore exists that the presence of amide local anaesthetics in the

epidural space could contribute to the therapeutic failure of an epidural autologous blood patch.

Ropivacaine is an aminoamide local anaesthetic increasingly used for epidural analgesia and

anaesthesia, particularly in obstetric practice. This study was undertaken to investigate whether



concentrations of ropivacaine in blood, which could occur clinically in the epidural space, alter

coagulation or ®brinolysis. Thromboelastography was used to assess clotting and ®brinolysis of

blood to which ropivacaine had been added. Although modest alterations in maximum amplitude,

coagulation time and alpha angle were observed, the effect of ropivacaine on clotting and

®brinolysis was not clinically signi®cant. We conclude that it is unlikely that the presence of

ropivacaine in the epidural space would reduce the ef®cacy of an early or prophylactic epidural

blood patch.

Keywords Anaesthetics, local; ropivacaine. Anaesthetic techniques, regional; epidural. Blood; coagulation.

......................................................................................

Correspondence to: Dr J. M. Porter

Accepted: 22 January 1999

Epidural autologous blood patch is commonly used in the

treatment of postdural puncture headache. It is thought

that a blood patch works by sealing the dural perforation

and preventing further leakage of cerebrospinal ¯uid.

Although relief of postdural puncture headache by epi-

dural blood patch actually occurs through several mechan-

isms [1, 2], the long-lasting therapeutic effect is dependent

on the ability of the autologous blood to coagulate in the

epidural space [3]. The amide local anaesthetics, bupiva-

caine and lignocaine, both impair clotting by inhibition of

platelet function [4±6] and by enhanced ®brinolysis in the

case of lignocaine [7]. These effects occur at local anaes-

thetic concentrations that could occur in the epidural

space a few hours after an epidural block is established.

Ropivacaine is an aminoamide local anaesthetic, prepared

as a pure S-enantiomer, recently introduced into clinical

practice and increasingly used for epidural analgesia and

anaesthesia in obstetric practice. To date, its effects on

platelet function and whole blood coagulation have not

been studied. Conventional coagulation tests end with the

formation of ®brin strands, whereas thromboelastographic

assessment begins at this point [8]. Thromboelastography

(TEG) is a more sensitive test of ®brinolytic activity than

conventional measurement of ®brin degradation products

[9]. The effects of clinically achievable concentrations of

ropivacaine in blood on clotting and ®brinolysis were

assessed using TEG.

Methods

With institutional ethical approval and having obtained

informed consent from each, 10 ASA grade I volunteers

were selected for study. Individuals with a history of

coagulation abnormalities were excluded as were those

taking medication known to in¯uence platelet function or

coagulation.

An 18 G intravenous cannula was inserted in the non-

dominant forearm. Whole blood was withdrawn using a

two-syringe technique, with the ®rst 2 ml aliquot being

discarded to reduce tissue thromboplastin contamination

from the venepuncture. The whole blood was analysed

within 4 min of collection. Ropivacaine solutions were

prepared from ropivacaine hydrochloride 10 mg.mlÿ1

(NaropinÒ, Astra Pharmaceuticals Ltd, Hertfordshire,

UK) in 0.9% saline to create ®ve serial end concentrations

in blood, ranging from 3.75 mg.mlÿ1 to 0.23 mg.mlÿ1. The

highest ropivacaine concentration was chosen because

it corresponds to that in a 50% admixture of 0.75%



Figure 1 Typical thromboelastographic

tracing, showing calculation of reaction time

(r), coagulation time (k), maximum amplitude

(MA) and alpha angle (a).

ropivacaine in blood, a mixture that might theoretically

occur in the epidural space if an epidural blood patch were

performed immediately after injection of ropivacaine

0.75%. To each of ®ve TEG cuvettes, prepared saline

solutions of ropivacaine (180 ml) and whole blood (180 ml)

were added, forming end concentrations of ropivacaine of

3.75 mg.mlÿ1, 1.88 mg.mlÿ1, 0.94 mg.mlÿ1, 0.47 mg.mlÿ1

and 0.23 mg. mlÿ1 (ropivacaine/blood preparations).

Normal saline (180 ml) was added to 180 ml of whole

blood in a sixth cuvette and this was used as a volumetric

control (saline control). A seventh cuvette contained only

whole blood (360 ml).

Blood coagulation and ®brinolysis was measured by

thromboelastography using a computerised dual channel

TEG (Haemoscope Corporation 1990). The TEG para-

meters measured were as follows (Fig. 1).

Reaction time

Reaction time (r: normal range 10±14 min) is the distance

measured from the beginning of the trace to the point of

2 mm divergence. It corresponds to an early stage in clot

formation and is in¯uenced by coagulation factor activity.

Coagulation time

Coagulation time (k: normal range 3±6 min) is the distance

from the point of 2 mm divergence to a point of 20 mm

divergence and represents the time the forming clot takes

to achieve a ®xed degree of viscoelasticity. The parameter

k is in¯uenced by coagulation factor activity, number and

function of platelets and ®brinogen.

Maximum amplitude

Maximum amplitude (MA: normal range 59±68 mm) is

the width of the curve at its maximal divergence and

represents the strength of the ®brin clot. The MA re¯ects

the structural components of the clot, i.e. platelet number

and function, and ®brinogen. Maximum amplitude is

in¯uenced by platelet count and activity and correlates

with tests of platelet aggregation responses to adenosine

diphosphate (ADP) and collagen [10].

Alpha angle

The alpha angle (a: normal range 54±678) is formed by

the slope of the tracing from the r- to the k-value. Similar

to k, a is an indicator of the rate of solid clot formation and

is dependent on platelet count and ®brinogen.

Clot lysis

Clot lysis can be described in terms of Ly30, de®ned as

clot lysis at 30 min after the maximum amplitude (normal

range < 7.5%) [11].

For each patient, seven preparations were analysed:

360 ml of each of the ®ve ropivacaine/blood preparations

and 360 ml of the two controls (whole blood and saline)

were maintained at 37 8C in prewarmed plastic TEG

cuvettes. The pins of the TEG were raised and lowered

®ve times to ensure adequate mixing of the blood and

saline solutions. The morphology and data for ®ve TEG

parameters (r, k, a, MA and Ly30) were collected and

stored on TEG software. The saline control and the

ropivacaine/blood preparations were subjected to a stan-

dardised dilution (180 ml of blood and 180 ml of saline/

ropivacaine) to minimise volume-related effects.



Table 1 Prothrombin times and activated partial thromboplastintimes in subjects. Values are given as mean (range)

Prothrombin time; sSubjects 13.72 (12.3±15.0)Normal range 12.0±15.5

Activated partial thromboplastin time; sSubjects 29.68 (21.9±37.0)Normal range 29.0±46.0

Reaction Coagulation Maximum Alpha Lysis 30,time, r time, k amplitude, MA angle, a Ly30(min) (min) (mm) (8) (%)

Normal range 10.0±14.0 3.0±6.0 59.0±68.0 54.0±67.0 < 7.5Whole blood 10.8 (4.6) 3.3 (0.5) 73.0 (4.5) 68.2 (3.2) 0.8 (0.5)Saline control 10.6 (2.6) 3.6 (0.8) 63.0 (5.6)² 69.7 (4.1) 0.3 (0.4)Ropivacaine

3.75 mg.mlÿ1 9.6 (4.8) 5.3 (1.7)* 59.7 (9.0)² 62.8 (6.4) 0.7 (1.3)1.88 mg.mlÿ1 10.9 (6.0) 5.4 (1.2)* 64.9 (7.7)² 59.1 (6.2)* 0.5 (1.4)0.94 mg.mlÿ1 12.3 (5.4) 4.2 (1.0) 66.2 (4.0)² 64.6 (5.7) 1.0 (2.8)0.47 mg.mlÿ1 14.4 (7.5) 4.6 (1.1) 61.4 (6.7)² 62.8 (6.4) 0.6 (1.9)0.23 mg.mlÿ1 17.5 (8.0)* 5.3 (2.8)* 62.0 (7.8)² 60.7 (12.3)* 2.4 (4.0)

* p< 0.05 when compared with blood and saline controls; ²p< 0.05 when compared withblood control only.

Table 2 Thromboelastography parametersin ropivacaine/blood preparations, wholeblood and saline-treated controls. Data aregiven as mean (SD)

Results of the data were subjected to one-way analysis

of variance for repeated measures. The Bonferroni post hoc

correction was applied. A p-value of < 0.05 was con-

sidered to be statistically signi®cant.

Results

Ten ASA grade I volunteers aged 25±45 years, ®ve males

and ®ve females, were included in the study. Baseline

values for prothrombin time and activated partial throm-

boplastin time are summarised in Table 1.

None of the subjects had measured baseline abnormali-

ties in coagulation variables (TEG parameters, prothrombin

time or activated partial thromboplastin time). When com-

pared with whole blood, ropivacaine solutions in blood

produced effects in the following TEG parameters: k, a and

MA. No signi®cant effects were demonstrated on reaction

time, r (with the exception of the most dilute concentration

of ropivacaine) or Ly30. Of those effects where ropivacaine

produced a difference (i.e. k, a and MA), saline dilution

produced changes similar in nature and degree to ropiva-

caine in MA, indicating an effect attributable to dilution

alone. In contrast, changes in a (a decrease) and in k

(prolongation) induced by ropivacaine in blood were not

reproduced by saline dilution. Group means, standard

deviations and analysis of variance for repeated measures

with post hoc correction are displayed in Table 2.

Discussion

Although statistically signi®cant changes in TEG parameters

(k, a, MA) were observed in the ropivacaine/blood pre-

parations, it is unlikely that these are of clinical importance

in terms of the ef®cacy of an epidural blood patch. With

one exception (ropivacaine 0.23 mg.mlÿ1: r� 17.5 min),

mean TEG parameters for all of the ropivacaine/blood

preparations were within normal limits. This single excep-

tion is likely to have arisen by chance as it occurred at the

lowest concentration of ropivacaine in blood and the value

of r obtained (17.5 min) lay within 2 SD (i.e. 2 ´ 4.6 mm) of

the mean seen in our whole blood control (10.8 mm).

The addition of the local anaesthetic agent to whole

blood altered coagulation as measured by TEG as follows:

changes were noted in k, a and MA but not in lysis or

reaction time. A signi®cant reduction in MA from the

control value was observed in all ®ve ropivacaine/blood

preparations (p< 0.05). However, the effects on MA were

similar in magnitude and direction to those produced by

the saline control, indicating an effect attributable to

dilution. Maximum amplitude re¯ects the structural integ-

rity of the clot (platelets, ®brinogen) and therefore one or

both of these were in¯uenced by dilution with saline with

or without ropivacaine.

Ropivacaine tended to increase coagulation time (k)

and reduce the alpha angle (a). This did not occur with

the saline control. We therefore conclude that the changes

observed were due to the effects of ropivacaine. Both

coagulation time (k) and alpha angle (a) are measures of

clot formation, the former dependent upon the activity of

clotting factors, ®brinogen and platelets, the latter on

®brinogen and platelets. Coagulation time (k) is a measure

of the time taken for a ®xed degree of viscoelasticity to be

achieved by the forming clot as a result of ®brin build up

and cross-linking. The alpha angle (a) denotes the speed at

which the clot forms. Coagulation time (k) was prolonged

in all ®ve ropivacaine/blood preparations, three of which

(3.75 mg.mlÿ1, 1.88 mg.mlÿ1 and 0.23 mg.mlÿ1) reached

statistical signi®cance. The ®ve local anaesthetic concentra-

tions studied signi®cantly prolonged coagulation time (k).

Alpha angle (a) was reduced in all ®ve ropivacaine/blood

preparations, two of which (1.88 mg.mlÿ1, 0.23 mg.mlÿ1)

reached statistical signi®cance. The magnitude of these

effects was small when compared with the normal values

for whole blood or with those produced by lignocaine in a

previous report [7], in which the concentrations studied

were equivalent to those employed in the current study.

All samples in our study (with the exception of the

whole blood control) were subjected to a standard dilution

of 50%. This degree of dilution (50% blood and ropiva-

caine) might occur within the epidural space following

the performance of an epidural blood patch. The effects of

dilution on coagulation have been studied: 50% dilution of

blood with saline produces an increase in coagulability in

vitro (r and k reduced, a increased) [12]. Blood loss of 15%

with crystalloid replacement [13] and 30% haemodilution

with crystalloid have been associated with a similar TEG

pro®le (r and k reduced, a increased) [11, 14]. Sixty

per cent dilution with normal saline produced a reduction

in coagulability (r, k, MA reduced, a increased), possibly

due to a dilution of coagulation factor activity and plate-

lets, but all TEG indices remained within the normal range

[11, 14]. In our study, the 50% dilution of blood with

saline produced a signi®cant reduction only in MA, but

mean MA values remained within normal limits.

Bupivacaine and lignocaine have previously demonstrated

potent inhibition of platelet function in vitro [5]. In vivo

inhibition of platelet aggregation has also been demonstrated

with epidural bupivacaine [6]. An ex vivo study of concen-

trations of bupivacaine similar to serum levels of local

anaesthetic associated with epidurally administered bupiva-

caine failed to alter TEG parameters [15]. The lowest

concentration of ropivacaine in blood chosen in our study

(0.23 mg.mlÿ1) exceeds by 100-fold that which might be

achieved in serum during ropivacaine epidural anaesthesia

(2.5 mg.mlÿ1) [15] but such concentrations could be realised

in the epidural space when a blood patch is performed.



We did not correct for the effect of dilution on calcium

concentration or pH induced by 50% dilution in our

preparations. Other workers have examined the effects

of 30% and 60% dilution of whole blood with saline on

ionised calcium levels and pH and have noted minimal

changes [14].

The concentrations of ropivacaine were chosen to simu-

late the admixtures that might be produced in the epidural

space by starting with an equal volume of ropivacaine

0.75% and autologous blood (3.75 mg.mlÿ1). Prophylactic

epidural blood patch is a recognised technique in which

autologous blood is injected into the epidural space shortly

after a dural tap is recognised. This may result in such an

admixture of blood and local anaesthetic [16]. The mass of

ropivacaine initially placed in the epidural space is known.

However, estimation of its concentration as a function of

time is not easy. If ropivacaine uptake from the epidural

space proceeds more slowly than uptake of its diluent, the

concentration of ropivacaine may increase rather than

decrease with the passage of time. Alternatively, cerebro-

spinal ¯uid from the inadvertent dural puncture may dilute

the epidural ropivacaine. The more dilute solutions of

ropivacaine assessed simulate the situation if an epidural

blood patch is performed following the recent discon-

tinuation of an epidural infusion of, for example, 0.15%

ropivacaine (0.23 mg.mlÿ1).

The results of this study indicate that ropivacaine, in the

concentrations studied, does not produce clinically sig-

ni®cant alterations in clotting or ®brinolysis. The clinical

implication of this is that the presence of ropivacaine in the

epidural space will not reduce the duration of ef®cacy of

an epidural blood patch.

Acknowledgments

We are indebted to Paul Smith, Our Lady's Hospital for

Sick Children, Crumlin, Dublin, who assisted us in running

the TEG analysis and to Kathleen O'Sullivan, University

College Cork, who provided us with statistical advice.

References

1 Di Giovanni AJ, Galbert MW, Wahle WM. Epidural

injection of autologous blood for post lumbar puncture

headache. II. Additional clinical experience and

laboratory investigation. Anesthesia and Analgesia 1972; 51:

226±32.

2 de Abouleish E, la Vega S, Blaudinger I, Tio TO. Long

term follow up of epidural blood patch. Anesthesia and

Analgesia 1975; 54: 459±63.

3 Rosenberg PH, Heavner JE. In vitro study of epidural

blood patch on leakage through a dural puncture.

Anesthesia and Analgesia 1985; 64: 501±4.

4 Feinstein MB, Fiekers J, Fraser C. An analysis of the

mechanisms of local anaesthetic inhibition of platelet

aggregation and secretion. The Journal of Pharmacology

and Experimental Therapeutics 1976; 197: 215±28.

5 Borg T, Modig J. Potential anti-thrombotic effects of

local anaesthetics due to their inhibition of platelet

aggregation. Acta Anaesthesiologica Scandinavia 1985; 29:

739±42.

6 Henny CP, Odoom JA, Ten Cate H, Oosterhof RJF,

Dabhoiwala NF, Sih IL. Effects of extradural bupivacaine

on the haemostatic system. British Journal of Anaesthesia

1986; 58: 301±54.

7 Tobias MD, Pilla MA, Rogers C, Jobes DR. Lidocaine

inhibits blood coagulation: implications for epidural blood

patch. Anesthesia and Analgesia 1996; 82: 766±9.

8 Mallett V, Cox JA. Thromboelastography. British Journal of

Anaesthesia 1992; 69: 307±13.

9 Whitten CW, Latson TW, Gulden RH, Burkhardt D,

Hyndman V. Thromboelastographic ®brinolysis does not

correlate with levels of D-dimers after cardiopulmonary

bypass. Anesthesiology 1991; 75: A432.

10 Tuman KJ, McCarthy RJ, Patel RV, Ivankovich AD.

Comparison of thromboelastography and platelet

aggregometry. Anesthesiology 1991; 75: A433.

11 Egli GA, Zollinger A, Popovich D, Pasch T, Spahn R.

Effect of progressive haemodilution with hydoxyethyl

starch, gelatin and albumin on blood coagulation. British

Journal of Anaesthesia 1997; 78: 684±9.

12 Tobias MD, Pilla MA, Rogers C. Effects of sequential

hemodilution upon TEG assessed coagulation. Anesthesia

and Analgesia 1995; 80: S503.

13 Tuman KJ, Speiss BD, McCarthy RJ, Ivankovich AD.

Effects of progressive blood loss on coagulation as measured

by thrombelastography. Anesthesia and Analgesia 1987; 66:

856±63.

14 Jamnicki M, Zollinger A, Seifert B, Popovic D, Pasch T,

Spahn DR. The effect of potato starch derived and corn

starch derived hydroxyyethyl starch on in vitro blood

coagulation. Anaesthesia 1998; 53: 638±44.

15 Gibbs NM, Sear JW. Effect of ketorolac, bupivacaine and

low-dose heparin on thromboelastographic variables in

vitro. British Journal of Anaesthesia 1995; 75: 27±30.

16 Berger CW, Crosby ET, Grodecki W. North American

survey of the management of dural puncture occurring

during labour epidural analgesia. Canadian Journal of

Anaesthesia 1998; 45: 110±14.



FORUM

The pre-operative ECG in day surgery: a habit?

C. J. Murdoch,1,2 D. R. Murdoch,3 P. McIntyre,4 H. Hosie5 and C. Clark5

1 Specialist Registrar in Anaesthesia, 5 Consultant Anaesthetist, Department of Anaesthesia, and

4 Consultant Physician and Cardiologist, Department of Cardiology, Southern General Hospital, Glasgow, UK

2 Department of Anaesthetics, Glasgow Royal In®rmary NHS Trust, 82±84 Castle Street, Glasgow G4 0SF, UK

3 Specialist Registrar in Cardiology, Department of Cardiology, Western In®rmary, Glasgow, UK

Summary

As the population presenting for day-case surgery and anaesthesia increases, so does the challenge

of adequate pre-operative assessment. Although an electrocardiogram is frequently performed, its

value in day-case surgery remains unproven. One thousand, one hundred and eighty-®ve patients

presenting for day-case surgery were assessed. One hundred and ®fty-four (13%) were referred for

electrocardiogram according to well-recognised criteria for the prediction of coronary artery

disease. They were read independently by the anaesthetist responsible for the case and by an

experienced cardiologist. A signi®cant abnormality was noted in 26% of electrocardiograms, most

frequently in patients referred with hypertension. There was a good correlation between the

reports of the anaesthetist and cardiologist. Only 20% of those patients with an abnormal

electrocardiogram had their surgery postponed. No adverse events occurred in patients proceeding

to surgery despite the abnormalities. We conclude that a resting electrocardiogram is of limited

value in risk strati®cation of patients undergoing day-case surgery.

Keywords Monitoring; electrocardiography. Anaesthesia, outpatient; evaluation, pre-operative.

......................................................................................

Correspondence to: Dr C. J. Murdoch

*Present address: Auckland Hospital, Auckland, New Zealand

Accepted: 16 March 1999

The advantages of day-case, as opposed to inpatient,

surgery for both hospital and patient are clear. The

increasing popularity of this technique has, however,

inevitably led to an expansion of the eligible population

to include patients at higher anaesthetic risk. Unfortu-

nately the opportunity for accurate, time-consuming, pre-

operative risk assessment of this group is limited. We

wished to determine the value of the resting 12-lead

electrocardiogram (ECG) because, although it is known

to be relatively insensitive and frequently normal even in

the presence of severe coronary artery disease [1], it is

quick, easy to perform and frequently requested. Its value

in day-case surgery, however, remains unproven.

Methods

We prospectively studied the in¯uence of the pre-operative

ECG on anaesthetic management and outcome of patients

presenting to the day surgery unit between June 1994 and

October 1997. Patients were selected for an ECG on the

basis of age (> 60 years), a history of hypertension or heavy

cigarette consumption, or a history or symptoms of

ischaemic heart disease, respiratory disease, peripheral or

cerebrovascular disease. The anaesthetist responsible for

the case reported the ECG and any alteration of manage-

ment related to the ECG ®ndings. The tracing was later

read independently by an experienced cardiologist blinded

to patient identity. Follow-up was conducted by telephone

after 48 h.

Results

Of 1185 eligible patients, ECGs were requested and

available for analysis in 154 (13%; 91 male; mean age

54.8 years). Anaesthetist and cardiologist concurred on

110 normal and 35 abnormal ECGs. Five tracings thought



by the anaesthetist to show nonspeci®c ST changes and

voltage criteria for left ventricular hypertrophy (LVH) were

considered normal by the cardiologist, whereas four

reported normal by the anaesthetist were interpreted by

the cardiologist as unifocal ventricular ectopics, nonspeci®c

ST changes, LVH and myocardial ischaemia, respectively.

The investigators felt that the most signi®cant alteration

in management in a day-case setting was the postponement

of surgery because the patient was un®t to undergo day-case

anaesthesia. The anaesthetist responsible for the case was

also asked to provide further information about changes in

anaesthetic practice based on the ECG appearance.

Eight (20%) of the 40 patients considered to have an

abnormal ECG by the anaesthetist had their surgery

postponed. This ranged from postponing surgery pending

further investigation (six patients), referral for in-patient

surgery (one patient) and admission to the coronary care

unit for treatment of a ventricular arrhythmia (one patient).

The participating anaesthetists did not document other

changes in management.

The ECG abnormalities that were identi®ed but did not

lead to postponement of surgery were LVH (10 patients),

nonspeci®c ST changes (three patients), myocardial

ischaemia (®ve patients), left bundle branch block (one

patient), previous inferior myocardial infarction (three

patients) and previous anteroseptal myocardial infarction

(one patient). None of the patients with ECG evidence

of myocardial infarction gave a positive history. Despite

this, there were no cases of peri-operative cardiovascular

complications or admissions to hospital.

The majority of ECGs were performed on the grounds

of hypertension and age (Table 1).

Twenty-two of 57 ECGs in hypertensive patients were

abnormal (38.6%), and in this group four previously

undiagnosed myocardial infarctions were identi®ed.

Abnormal ECGs were noted in 10 of 39 carried out in

patients aged over 60 years (25.6%). The other referral

groups were either small or yielded few abnormalities.

Discussion

This is the ®rst prospective study to consider the in¯uence

of the ECG on the anaesthetic management of day surgery

patients, while also allowing independent comparison of

the ECG by the anaesthetist and cardiologist. The inci-

dence of ECG abnormalities in these patients (26%) was

similar to that found in other studies [2, 3]. Nevertheless,

in 80% of our patients the abnormalities did not prevent

the patient proceeding directly to anaesthesia and surgery.

Furthermore, they did not predict intra-operative or post-

operative complications or hospital admission following

the procedure. In contrast to previous studies [2, 3], we

found that a history of hypertension and not age was more

likely to predict an abnormal ECG.

We conclude the ECG is of limited value in the risk

strati®cation of patients selected for day-case surgery.

Acknowledgments

We would like to thank the nursing staff, Day surgery

unit, Southern General Hospital for their work in data

collection and the ECG department for providing copies

of the ECGs.

References

1 Norell M, Lythall D, Coghlan G, et al. Limited value of the

resting electrocardiogram in assessing patients with recent

onset chest pain ± lessons from a chest pain clinic. British

Heart Journal 1992; 67: 53±6.

2 Golub R, Cantu R, Sorrento JJ, Stein HD. Ef®cacy of

preadmission testing in ambulatory surgical patients.

American Journal of Surgery 1992; 163: 565±70.

3 Gold BS, Young ML, Kinman JL, Kitz DS, Berlin J,

Schwartz J. The utility of preoperative electrocardiograms in

the ambulatory surgical patients. Archives of Internal Medicine

1992; 152: 301±5.



Table 1 ECG request categories, results and management

Reason Alteredfor request Normal Abnormal management Total

Hypertension 35 22 4 57Age 29 10 2 39CHD 1 0 0 1CVA 1 0 0 1Palpitations 15 3 1 18PVD 3 1 1 4Smoker 13 2 0 15Effort dyspnoea 4 0 0 4VHD 5 1 0 6Angina/AMI 8 1 0 9

CHD, coronary heart disease; CVA, cerebrovascular disease; PVD,peripheral vascular disease; VHD, valvular heart disease; AMI, acutemyocardial infarction.

Documents

The pre-operative ECG in day surgery: a habit?