ANESTH ANALG 1984;631053-8

1053

Effects of Intravenous Bupivacaine on Cardiovascular Function and Plasma Catecholamine Levels in Humans
Lars J. Hasselstrom, MD, Torben Mogensen, Niels J. Christensen, MD
HASSELSTR0M LJ, MOGENSEN T, KEHLET H, CHRISTENSEN NJ. Effects of intravenous bupivacaine on cardiovascular function and plasma catecholamine levels in humans. Anesth Analg 1984;63:1053-8.

MD,

Henrik Kehlet,

MD,

and

The effects of the intravenous infusion of bupivacaine (2 mglmin for 3 hr) on cardiovascular function and various endocrine metabolic parameters were studied in a randomized single-blind crossover study in eight normal subjects. Bupivacaine infusion resulted in plasma concentrations about 1-2 pglml. Heart rate increased significantly from approximately 60 to 70 beatslmin. Mean arterial blood pressure increased from 87 to about 100 mm Hg, and cardiac output decreased about 20%, both of these statistically significant. Oxygen consumption did not change significantly. Plasma

epinephrine concentrations increased signijicantly from 0.03 to 0.08 ng/ml, but plasma norepinephrine levels did not change sjgnificantly, nor did blood concentrations of cose, lactate and plasma cortisol, and free fatty acids. These results demonstrate that systematically administered bupivacaine results in plasma concentrations of bupivacaine comparable to those that may be observed during regional anesthesia, including a positive chronotropic and an arterial vasoconstrictiveeffect. Theseeffects areprobnbly caused mainly by a direct effect of bupivacaine on the cardiovascular system, because only quantitatively minor changes in plasma catecholamines and other endocrine metabolic parameters were observed.
Key Words: LOCAL ANESTHETICS-bupivacaine.

Studies of the effects of systematically administered bupivacaine on cardiovascular function in animals have shown a decrease in contractility in a rabbit heart preparation (l),a constrictive effect on the portal vein (2), and vasodilatation in the dog hind limb after intraarterial injection (3). In humans, the infusion of bupivacaine has been associated with either no significant changes or increases in heart rate, blood pressure, and cardiac output (4,5). Systemic vascular resistance has been reported to increase at low concentrations (6)and decrease at high concentrations of plasma bupivacaine (6). Infusion studies resulting in plasma bupivacaine concentrations comparable to those observed during regional anesthesia have shown either a slight increase (4) or decrease in peripheral vascular resistance (5). In addition, oxygen consumption and splanchnic uptake of glycerol and lactate increased (7), effects that were hypothesized as being

related to (unmeasured) increases in plasma levels of epinephrine. Because of the uncertainty surrounding the physiologic and biochemical consequences of the presence of bupivacaine in peripheral blood, we studied the effects of intravenous infusion of bupivacaine on heart rate, blood pressure, cardiac output, oxygen consumption, and plasma concentrations of epinephrine, norepinephrine, glucose, lactate, free fatty acids, and cortisol in normal volunteers in a randomized singleblind crossover study.

Material and Methods

Subjects
Eight healthy volunteers, four of each sex, took part in the study after informed consent was obtained. Age was 28 yr (median), range 23-44 yr; weight was 71 kg (median), range 56-79 kg. The investigation was approved by the local ethics committee.

This study was supported by a grant from AB Astra, Sweden. Received from the Departments of Anesthesia and Internal Medicine and Endocrinology, Herlev Hospital, University of Copenhagen, and Hvidare Hospital, Klampenborg, Denmark. Accepted for publication August 21, 1984. Address correspondence to Dr. Hasselstrem, Department of Anesthesia, Herlev Hospital, DK 2730 Herlev, Denmark.
0

Experimental Protocol
The protocol consisted of a randomized single-blind crossover design with at least one week between each

1984 by the International Anesthesia Research Society

1054

ANESTH ANALG 1964;63: 1053-8

HASSELSTR0M ET AL.

study. After an overnight fast, at 8 AM the seminude subjects were placed in a room with constant temperature (23-25C). A 17-gauge intravenous catheter was inserted percutaneously into a peripheral vein to be used for a 3-hr infusion of either bupivacaine (Macain; 2 mg/min) or isotonic saline (25 ml/hr) by a Perfusor pump (B. Braun Melsungen AG, West Germany). A supplementary infusion of isotonic saline (100 ml/hr) was given through the same catheter to both groups. A 16-gauge central venous catheter inserted percutaneously via the cephalic vein was used for continuous monitoring of central venous pressure (CVP) with a Siemens Elema pressure tranducer 746 (Siemens Elema, Sweden) and an S & W Pressure Display PDU P 410 and pressure amplifier BAP 001 (Simonsen & Weel, Copenhagen, Denmark). Blood pressure was measured automatically with a Dinamap (Critikon, USA). ECG was continuously monitored on a Memory Trendscope MTS 216 with a warning unit WUP 306 (Simonsen & Weel, Denmark) and with phonocardiogram (Siemens Elema, EMT 25 C, Sweden). Cardiac output was measured by electrical transthoracic impedance (8) using the Minnesota Impedance Cardiograph (model 304 A) with a fourelectrode impedance system. Two aluminium band electrodes (3 M tape) separated as widely as possible, were placed around the neck of each subject. A third electrode was placed around the thorax 5 cm below the ziphoid process, and a fourth electrode 5 cm below that. All tracings were recorded (Siemens Elema Mingograph 34) at a paper speed of 50 mdsec. Measurements of transthoracic impedance were made during breath-holding after a normal expiration. Stroke volume was calculated from the following equation
(8):

ik 3

1
h

2-

1-

60

120

minutes

180

Figure 1. Concentrations of plasma bupivacaine (pg/ml) during intravenous infusion of bupivacaine (2 mg/min).

centrations in blood were determined by routine laboratory enzymatic methods (9,lO). Plasma-free fatty acids were measured calorimetrically after extraction with methoxyethanol and butylether (11). Plasma concentrations of cortisol were measured by a competitive protein binding method (12) and plasma catecholamine levels were measured by a single isotope-derivative method (13). Plasma bupivacaine concentrations were determined by selected ion monitoring (mass fragmentography) (Astra, Sodertalje, Sweden).

Statistical Methods
The trial design implied multiple assessments in the same person under varying circumstances, so that the relevant statistical method is analysis of variance for a two-factor design with repeated measurements on both factors (treatment and time). Because of problems with regard to assumptions, we decided to employ a corresponding nonparametric method (Bradley) (14) in which treatments were compared with a Pratt (15) test disregarding time, while both the factor time and the interaction between time and treatment were assessed by Friedmann tests (16). Before performing this analysis we, however, used the method by Koch (17) to exclude the presence of period and carry-over effects. Single missing datum points during bupivacaine infusion in patient no. 2 have been calculated according to Sokal and Rohlf (18). The correlations between plasma concentrations of bupivacaine and the cardiovascular changes were made using a Spearman rank correlation (19).

AV = P x - x T x
zo2

L2

where AV represents stroke volume (ml), P the resisL tivity of blood (a x cm), the mean distance (cm) between the inner pair of electrodes, Zo the basal thoracic impedance between the two inner electrodes (a),(dzldt) the rate of change of impedance (Wsec), and T the ventricular ejection time (sec). Stroke volume was calculated as mean of eight consecutive measurements using a P value of 135. Oxygen uptake (mymin) was measured over a 5-min period using an expirograph (Statham NV, type 16000).

Blood Analyses
Central venous blood was sampled 30 and 15 min before infusion, at infusion, and 30, 60, 90, 120, 150, and 180 min after initiation of the infusion of bupivacaine (or isotonic saline). Glucose and lactate con-

Results
Plasma Bupivacaine
Concentrations of bupivacaine in plasma increased progressively in all seven subjects in which it was

BUPIVACAINE AND CARDIOVASCULAR FUNCTION

ANESTH ANALG 1984;63:1053-8

1055

"1

n=8
median

-- + sallns
I

bupivacaine

n=B medlan

-do

60

la

1 180 minutes

L
-30
0

60

180

lM

minutes

"1

n=8 median

L
measured (Fig. l), reaching a final median value of 1.84 pg/ml (range 1.62-2.44) 180 min after beginning infusion.

-&

60

lhl

minules

Figure 2. Effects of intravenous infusion of bupivacaine (2 mg/min) on heart rate, mean arterial blood pressure, cardiac output, and oxygen consumption.

Cardiac Output Heart Rate

There was a significant (P < 0.0005) interaction between treatment and time (Fig. 2). Bupivacaine resulted in a 10-15% increase in heart rate, evident after 60 min of infusion. When time was disregarded in the statistical analysis, the effect of treatment was not significant (P > 0.10), while the effect of time was significant ( P < 0.025) when treatment was disregarded. There was a significant ( P < 0.005) interaction between treatment and time (Fig. 2). Bupivacaine resulted in a decrease of as much as 20% in cardiac output, evident after 60 min of infusion. When time was disregarded in the statistical analysis, the effect of treatment was significant ( P < 0.005) as was the effect of time ( P < 0.005) when treatment was disregarded.

Central Venous Pressure (CVP) Mean Arterial Blood Pressure ( M A P ) There was a significant ( P < 0.005) interaction between treatment and time (Fig. 2). Bupivacaine resulted in a steady and progressive increase in blood pressure, 10-15 mm Hg, evident after 60 min of infusion. When time was disregarded in the statistical analysis, the effect of treatment was significant (P < 0.02) as was the effect of time ( P < 0.005) when treatment was disregarded. Central venous pressures were, for reasons such as malpositioning of catheters, only available in four patients. There was no significant interaction (0.05 < P < 0.1) between treatment and time. Bupivacaine did not result in changes in CVP.

Oxygen Consumption
Oxygen consumption remained constant during both saline and bupivacaine infusions, without any significant difference between groups (Fig. 2).

1056

ANESTH ANALG
1984;63:105M3

HASSELSTRQM ET AL.

0.30 1

0.20.
0.15-

0.104

o.K5{

,
-30
0

Figure 3. Effects of intravenous infusion of bupivacaine (2 mghnin) on concentrationsof plasma catecholaminesand cortisol and blood glucose and lactate.

Plasma Catecholamines
The plasma concentration of norepinephrine increased from 0.19 ng/ml to 0.24 ng/ml during bupivacaine infusion, but there was no significant interaction (0.05 < P < 0.1) between treatment and time (Fig. 3). Plasma epinephrine concentration increased from 0.03 ng/ml to 0.08 ng/ml during bupivacaine infusion, and there was a significant (P < 0.001) interaction between treatment and time (Fig. 3). When time was disregarded in the statistical analysis, the effect of treatment was significant (P < 0.02), as was the effect of time (P < 0.0005) when treatment was disregarded.

Plasma Cortisol
There was no significant interaction between treatment and time (Fig. 3). There was a decrease in concentration of plasma cortisol in both groups (normal diurnal variation).

----------+-~

0.25-

norepinephrine

n-8
medlan

-krpvacaJne
P-*saline

n=8
median

60

120

180

rninules

0.5

0.4

0.3

n-8 median
4

-a

80

120

180 lllkum

Blood Glucose
Blood glucose remained constant and without significant differences between the two infusion regimens (Fig. 3).

Blood Lactate
Blood lactate levels decreased significantly in the saline group, but remained constant during bupivacaine infusion (Fig. 3). There was no significant difference between groups.

Plasma-Free Fatty Acids (FFA)

FFA remained constant and without significant differences between the two groups.

Discussion
Our results demonstrate that a 3-hr intravenous infusion of bupivacaine at a rate of 2 mg/min leads to significant and progressive increases in heart rate and mean arterial blood pressure and a decrease in cardiac output. Dueng the same period, plasma bupivacaine concentrations increased steadily. to 1.8 pg/ml, Con-

BUPIVACAINE AND CARDIOVASCULAR FUNCTION

ANESTH ANALG 1984;63: 1053-8

1057

centrations similar to those observed during regional anesthesia with bupivacaine. For example, after a single-dose epidural or intercostal block, a short lasting (1-2 hr) plasma bupivacaine concentration between 1 and 2 pg/ml has been reported (20,21). Furthermore, during continuous epidural analgesia, bupivacaine levels range between 1and 3 pg/ml(22,23). The finding of an increase in heart rate is in accordance with a similar study using same dose rate for 2.5 hr (5). In contrast, short-term (10-20 min) infusions of bupivacaine at a rate of 4-7.5 mg/min did not alter heart rate (4,24). Similarly, long-term infusion of bupivacaine at a rate of 2 mg/min (5) was associated with an increase in mean arterial blood pressure that was also observed during short-term infusions (4,24). Although cardiac output has been reported as remaining unchanged during the short-term infusion of bupivacaine (4,24), we found a progressive decrease in cardiac output amounting to 20% during a 3-hr infusion, a finding that contrasts the results reported by Wiklund (5) using the same rate and time of infusion. We have no explanation for this difference except that our study was more rigidly controlled. The combined effects of bupivacaine infusion on heart rate, mean arterial blood pressure, and cardiac output are difficult to explain by a single mechanism. The increase in heart rate after systemic infusion of local anesthetics has been considered a central nervous system effect evoked through the sympathetic nervous system (3). We observed a slight but significant increase in plasma epinephrine concentrations after bupivacaine administration, which may in part be responsible for the observed increase in heart rate (25).However, the increase in plasma norepinephrine levels was relatively small, and activation of the sympathoadrenal system is unlikely to be solely responsible for the increase in arterial blood pressure after bupivacaine. Lastly, our finding of unchanged plasma cortisol concentrations indicates that bupivacaine does not result in a general central nervous system "arousal" reaction. The increase in mean arterial blood pressure is probably explained by a direct vasoconstrictoreffect on the arterial side of the circulation as demonstrated during various experimental settings after bupivacaine (2,6) and other local anesthetics (3). This vasoconstrictor effect of local anesthetics is predominantly seen at low concentrations, while higher concentrations lead to vasodilatation. An explanation for the decrease in cardiac output after bupivacaine may be a direct myocardial depressive effect as shown in experimental studies (1). Furthermore, bupivacaine may inhibit cardiac sympathetic nerve activity as demonstrated during lidocaine infusion (26). Finally, our findings of unchanged concentrations of blood

glucose, lactate, FFA, and oxygen consumption h r ther argue against a major activation of the sympathetic nervous system by bupivacaine. There was a significant relationship between the plasma concentration of bupivacaine and decrease in cardiac output (Rs = -0.33, P < 0.05), and increase in heart rate (Rs = 0.35, P < 0.05), while the relationship to increase in blood pressure was insignificant (R, = 0.25, P = 0.1). The cardiovascular effects of bupivacaine were only poorly correlated to total plasma bupivacaine concentrations, suggesting that factors such as differences in protein binding of bupivacaine in plasma, intracellular transport, and sensitivity may be of importance. In conclusion, our results have shown a definite effect of systemic infusion of bupivacaine on heart rate, mean arterial blood pressure, and cardiac output. The underlying mechanisms for these changes are complex and remain unclarified, but a central nervous system activation evoked through the sympathetic nervous system is unlikely to be the predominant cause.
The authors are indebted to 8 . Saurbrey for her technical assistance and to Dr. B. Andersen for his extensive statistical advice.

References
1. Block A, Covino BG. Effect of local anaesthetic agents on cardiac conduction and contractility. Reg Anaesth 1981;6:55-61. 2. Aberg G, Wahlstrom B. Mechanical and electrophysiological effects of some local anesthetic agents and their isomers on the rat portal vein. Acta Pharmacol Toxicol 1972;31:255. 3. Blair MR.Cardiovascular pharmacology of local anaesthetics. Br J Anaesth 1975;47247-52. 4. Jorfeldt L, Lofstrom B, Pernow B, Persson B, Wahren J, Widman B. The effect of local anaesthetics on the central circulation and respiration in man and dog. Acta Anaesthesiol Scand 1968;12153-69. 5. Wiklund L. Human hepatic blood flow and its relation to systemic circulation during intravenous infusion of bupivacaine or etidocaine. Acta Anaesthesiol Scand 1977;21:189-99. 6. Aps C, Reynolds F. The effect of concentration on vasoactivity of bupivacaine and lignocaine. Br J Anaesth 1976;48:1171-4. 7. Wiklund L, Jorfeldt L. Splanchnic turn-over of some energy metabolites and acid-base balance during intravenous infusion of lidocaine, bupivacaine or etidocaine. Acta Anaesthesiol Scand 1981;25:200-8. 8. Kubicek WG, Kamegis JN, Patterson RP, Witsoe DA, Mattson RH. Development and evaluation of an impedance cardiac output system. Aerosp Med 1966;371208-12. 9. Leon LP, Chu DK, Snyder LR, Horvath C. Continuous flow analysis for glucose in serum, with use of hexokinase, glucose6-phosphate dehydrogenase coimmobilied in tubular form. Clin Chem 1980;26:123-9. 10. Noll F. Determination of I-lactate with lactate-dehydrogenase, glutamatepyruvate transaminase and NAD. In: Bergmeyer HE,

1058

ANESTH ANALG 1984;63:1053-8

HASSELSTR0M ET AL.

ed. Methods of enzymatic analysis, 2nd ed. (trans from 3 German ed). New York, London: Verlag Chemie and Academic Press, 19741475-9. 11. Bergmann SR, Carlson E, Dannen E, Sobel BE. An improved assay with 4-(2-thiazolylazo)-resorcinol non-esterified fatty for acids in biological fluids. Clin Chem Acta 1980;104:53-63. 12. Kehlet H, Binder C, Engbaek C. Cortisol binding capacity in plasma during anaesthesia and surgery. Acta Endocrinol (Copenh) 1974;75:119-24. 13. Christensen NJ, Vestergaard P, Ssrensen T, Rafaelsen OJ. Cerebrospinal fluid adrenaline and noradrenaline in depressed patients. Acta Psychiatr Scand 1980;61:178-82. 14 Bradley JV. Distribution free statistical tests. Englewood Cliffs, NJ: Prentice-Hall, 1968:138-45. 15. Rahe AJ. Tables of critical values for the Pratt matched pair signed rank statistic. J Am Stat Assoc 1974;69:368-73. 16. Bradley JV. Distribution free statistical tests. Englewood Cliffs, NJ: Prentice-Hall, 1968388. 17. Koch GG. The use of non-parametric methods in the statistical analysis of the two-period change-over design. Biometrics 1972;28:577-84. 18. Sokal RR, Rohlf FJ. Biometry. San Francisco: WH Freeman, 196933740. 19. Maxwell AE. Analysing qualitative data. London: Methuen 1961:163.

20. Moore DC, Mather LE, Bridenbaugh PO, Bridenbaugh LD, Balfour RI, Lysons DF, Horton WG. Arterial and venous plasma levels of bupivacaine following epidural and intercostal nerve blocks. Anesthesiology 1976;45:3945. 21. Moore DC, Mather LE, Bridenbaugh LD, Balfour RI, Lysons DF, Horton WG. Arterial and venous plasma levels of bupivacaine following periferal nerve blocks. Anesth Analg 1976;55:763-8. 22. Ross RA, Clarke JE, Armitage EN. Postoperative pain prevention by continuous epidural infusion. Anaesthesia 1980;35663-8. 23. Seeling W, Altemeyer KH, Berg S, Dick W, Kossmann B. Bupivacain Konzentrationen im Serum von Patienten mit Kontinuerlicher Thorakaler Katheterperiduralanaesthesie. Anaesthesist 1982;31:434-8. 24. Mather LE, Tucker GI, Murphy TM, Stanton-Hicks MdA, Bonica JJ. Cardiovascular and subjective central nervous system effects of long-acting local anaesthetics in man. Anaesth intensive Care 1979;7:215-21. 25. Clutter WE, Bier DM, Shah SD, Cryer PE. Epinephrine plasma metabolic clearance rates and physiologic thresholds for metabolic and hernodynamic actions in man. J Clin Invest 1980;66:94-101. 26. Miller BD, Thames MD, Mark AL. Inhibition of cardiac sympathetic nerve activity during intravenous administration of lidocaine. J Clin Invest 1983;71:1247-53.