Share this article on:

A Comparison of Epidural Levobupivacaine 0.5% With or Without Epinephrine for Lumbar Spine Surgery

Kopacz, Dan J. MD*,; Helman, James D. MD*,; Nussbaum, Charles E. MD†,; Hsiang, John N. K. MD, PhD†,; Nora, Peter C. MD†, and; Allen, Hugh W. MD*

doi: 10.1097/00000539-200109000-00041
TECHNOLOGY, COMPUTING, AND SIMULATION: (Society for Technology in Anesthesia): Research Report

Levobupivacaine, the S(−) isomer of bupivacaine, is less cardiotoxic than racemic bupivacaine. In this prospective, randomized, double-blinded study of epidural anesthesia, we compared the onset, extent, and duration of sensory and motor blockade produced by plain 0.5% levobupivacaine (15 mL, 75 mg) with that of 0.5% levobupivacaine with the addition of 1:400,000 or 1:200,000 epinephrine in 117 patients undergoing elective spine surgery. The time to onset of adequate sensory block (T10 dermatome) was similar in all groups (12.4 ± 6.6 min for plain levobupivacaine, 13.9 ± 7.9 min for levobupivacaine with 1:400,000 epinephrine, and 12.7 ± 4.9 min for levobupivacaine with 1:200,000 epinephrine), with an average peak block height of T5. Time to complete regression of sensory blockade was also similar between groups (357 ± 119 min for plain levobupivacaine, 378 ± 98 min for levobupivacaine with 1:400,000 epinephrine, and 348 ± 80 min for levobupivacaine with 1:200,000 epinephrine). Peak serum levobupivacaine levels were reduced in each of the epinephrine-containing groups. We conclude that 0.5% levobupivacaine with or without 1:200,000 or 1:400,000 epinephrine produced effective epidural anesthesia in patients having lumbar spine surgery. Epinephrine 1:400,000 is as effective as 1:200,000 in reducing the resultant serum levobupivacaine levels after epidural anesthesia.

Departments of *Anesthesiology and †Neurosurgery, Virginia Mason Medical Center, Seattle, Washington

Supported, in part, by a grant from Celltech Chiroscience Ltd, Granta Park, Great Abington, Cambridge, UK.

May 1, 2001.

Address correspondence to Dan J. Kopacz, MD, Department of Anesthesiology, Virginia Mason Medical Center, 1100 Ninth Ave., Mailstop B2-AN, Seattle, WA 98111. Address e-mail to

IMPLICATIONS: The addition of epinephrine 1:400,000 or 1:200,000 to epidural 0.5% levobupivacaine reduces the resultant serum levobupivacaine levels without altering the duration of sensory blockade. All solutions were well tolerated and effective in producing epidural anesthesia for patients undergoing lumbar spine surgery.

Levobupivacaine, the isolated S(−) isomer of bupivacaine, was less cardiotoxic than bupivacaine in preclinical studies (1,2) and is currently available for clinical use in anesthesia and postoperative pain management. Epinephrine, most often in a 1:200,000 concentration (5 μg/mL), is often added to local anesthetic solutions to reduce absorption from the site of injection and to increase the duration, intensity, or both of epidural blockade. Levobupivacaine intrinsically produces a mild degree of vasoconstriction (3). This inherent vasoconstrictive effect may allow for a reduction in the amount of added epinephrine necessary to produce a reduction in local anesthetic absorption from the epidural space. The purpose of this study was to determine whether a reduction in the amount of epinephrine from 1:200,000 (5.0 μg/mL) to 1:400,000 (2.5 μg/mL) added to epidural levobupivacaine produces a similar decrease in local anesthetic absorption from the epidural space while retaining the same clinical efficacy (onset, duration, and intraoperative conditions) and tolerability in patients undergoing elective lumbar spine surgery.

Back to Top | Article Outline


After IRB approval and written informed consent were obtained, 117 patients with ASA physical status I–III, aged 18–85 yr, scheduled to undergo elective lumbar spine surgery with epidural anesthesia were enrolled in this prospective, single-center, parallel group, double-blinded study. Patients with known hypersensitivity to amide local anesthetics or a history of severe renal, hepatic, respiratory, or cardiac disease or a neurologic, neuromuscular, or psychiatric condition were excluded.

The sample size was determined on the basis of estimations of the primary efficacy end point—the mean duration of bilateral sensory anesthesia above the T10 dermatome (minimum block necessary to initiate and complete surgery). By using previously published data for racemic bupivacaine with epinephrine (sd of 60 min for all groups) (4) and assuming a mean difference between the Plain and Epinephrine-Containing groups of 45 min as clinically significant, 36 patients per group were considered necessary to detect statistical significance (α = 0.05, two sided) with power (1 − β) = 80%. Allowing for withdrawals, 117 patients were entered into the study.

After an IV infusion of 500 mL of lactated Ringer’s solution, all patients were premedicated with IV midazolam (1–5 mg). Lidocaine 1% (3 mL) was used to infiltrate the skin and subcutaneous tissues at the L1-2 or L2-3 interspace. The epidural space was identified while patients were in the lateral decubitus position by use of an 18-gauge Tuohy needle and a loss of resistance to saline technique. After negative aspiration, patients were randomized to receive one of three blinded study solutions: 0.5% levobupivacaine without epinephrine, 0.5% levobupivacaine with 1:400,000 epinephrine, or 0.5% levobupivacaine with 1:200,000 epinephrine incrementally over a 3-min period (5-mL study drug “test dose,” wait 2 min, 5-mL injection, wait 1 min, 5-mL injection). The initial total volume of study drug administered was 15 mL, providing a total dose of 75 mg.

The end of injection of study drug was termed “Time 0” for the purposes of subsequent patient assessment. A 20-gauge catheter was advanced 3–4 cm into the epidural space, and the needle was removed. If the height of sensory blockade failed to reach the T10 dermatome after 20 min, patients could be administered one re-dose with 5 mL of study drug if necessary. Intraoperative sedation was provided with additional IV midazolam and propofol (25–100 μg · kg−1 · min−1) as needed at the discretion of the anesthesiologist. To assess the degree of anesthetic quality produced by the epidural drugs, neither IV muscle relaxants nor volatile anesthetics were used, and the patients were allowed to breathe spontaneously throughout the surgical procedure. Opiates and nonsteroidal antiinflammatory drugs were withheld until the patient’s first request for analgesic therapy in the postoperative period.

Epidural anesthetic blockade adequate to initiate surgery was defined as a sensory block bilaterally to dermatome T10. The primary efficacy measure was the duration of effective anesthesia, defined as the period of time from achieving bilateral T10 blockade to the time of bilateral regression to T10. Secondary efficacy measures included peak block height, time to reach peak block, time to two-segment regression, time to regression to T10, and time to complete regression (total duration of sensory block). Sensory block was measured with the blunt end of a 27-gauge dental needle at 0, 2, 5, 10, 15, 20, 25, 30, and 60 min after injection and every 30 min thereafter until complete regression of sensory block was observed.

The onset, degree, and duration of motor blockade were measured in both legs by using a modified Bromage scale and scored as “0” (no paralysis, full flexion of hips, knees, and ankles), “1” (inability to raise extended leg, able to move knees), “2” (inability to flex knees, able to flex ankles), or “3” (inability to move any portion of the lower limb). Motor block was mea-sured at 0, 10, and 20 min after dosing (before surgery) and every 30 min after surgery until the patient returned to a score of “0” in both legs.

The overall quality of intraoperative anesthetic quality (poor, fair, good, or excellent) was graded separately by the surgeon or anesthesiologist at the end of surgery.

Hemodynamic variables were recorded at baseline (preinjection), at the end of injection, and at 30-min intervals until complete resolution of the sensory block. Hypotension was defined as a decrease in systolic blood pressure of at least 30% from baseline and was treated with IV fluids or vasopressor drugs at the discretion of the attending anesthesiologist. Tachycardia was defined as a 30% increase in heart rate compared with baseline. All other adverse events were recorded throughout the study.

Venous blood samples were obtained from 30 patients who had received only 15 mL of solution (10 per group) at the end of injection, at 5, 10, 20, 30, 45, and 60 min after the end of injection, and at 2, 4, 6, 8, 10, and 12 h after the end of injection for pharmacokinetic analysis (Cmax, Tmax, t1/2, area under the plasma drug concentration-time curve, clearance, Vd [volume of distribution]) of plasma levobupivacaine levels. Levobupivacaine levels were determined with liquid chromatography-mass spectrometry, with limits of determination of 10 ng/mL and coefficients of variation at these limits of approximately 0.8%.

All statistical comparisons were two tailed and performed on the intent-to-treat patient population, with significance set at P ≤ 0.05. Differences between groups were analyzed by using analysis of variance with pairwise comparisons, with 95% confidence intervals. Logistic regression analysis or χ2 statistics were used to analyze categoric and incidence data. Kaplan-Meier survival analysis was used to compare the time to first request for postoperative analgesics.

Back to Top | Article Outline


One-hundred-seventeen patients were enrolled into the trial and randomized to treatment. Three patients did not demonstrate any evidence of epidural anesthesia after the initial injection and were excluded. An additional six patients did not reach the minimal T10 block height and were excluded from further analysis. The remaining 108 patients who were included in efficacy analyses were evenly distributed among the three groups. Eight patients required the additional allowed 5-mL reinjection (Plain Levobupivacaine, 2 of 35, 6%; Levobupivacaine + 1:200,000 Epinephrine, 1 of 37, 3%; Levobupivacaine + 1:400,000 Epinephrine, 3 of 36, 8%). There were no differences in age, sex, or body mass among the groups (Table 1).

Table 1

Table 1

The mean time to onset of sensory block adequate for surgery (T10) and duration of effective anesthesia (T10-10) were equivalent in the three groups (Table 2). The peak block height was between T5 and T6 in all treatment groups. Block variables often could not be obtained from 30 to 90 min, as patients were undergoing lumbar spine surgery (Fig. 1). There were also no differences among groups in time to two-segment or complete sensory regression (Table 2).

Table 2

Table 2

Figure 1

Figure 1

Levobupivacaine 0.5% produces relatively little motor blockade. In fact, in 53% (57 of 108) of all patients studied, no motor block of the lower extremities could be demonstrated (Table 3). Although the addition of either 1:200,000 or 1:400,00 epinephrine tended to increase the degree of motor blockade, this did not reach statistical significance. Furthermore, in those patients who did develop some degree of motor blockade, its duration was not different among the three groups (Table 3).

Table 3

Table 3

The amount of perioperative midazolam and propofol used for sedation did not differ among groups (Table 1). The overall quality of intraoperative epidural block was assessed by the investigator (Fig. 2) as “excellent” or “good” in 85% of patients (Plain Levobupivacaine, 77%; Epinephrine 1:200,000, 88% [P = 0.03 versus Plain Levobupivacaine], and Epinephrine 1:400,000, 94% [P = 0.01 versus Plain Levobupivacaine]). There was very strong agreement between block quality ratings by the anesthesiologist and the surgeon (P = 0.0001). Although patients in the Plain Levobupivacaine group had the shortest time before requesting analgesics, this also did not reach statistical significance (Table 1, P = 0.07 and P = 0.06 vs 1:200,000 and 1:400,000 Epinephrine-Containing groups, respectively). There were no differences between groups in time to first spontaneous voiding after surgery (P = 0.58).

Figure 2

Figure 2

No serious adverse events were related to study medication. Adverse events considered to be study drug related were reported by 64% of patients administered levobupivacaine, 64% of patients administered levobupivacaine with 1:200,000 epinephrine, and 72% of patients administered levobupivacaine with 1:400,000 epinephrine. Nausea (26%) and hypotension (18%) were the most common side effects attributed to study drug. Of the patients receiving levobupivacaine with 1:200,000 epinephrine, 15% became tachycardic and 23% hypotensive within the first 30 min, compared with 3% tachycardia and 10% hypotension in the Plain Levobupivacaine patients and 10% tachycardia and 10% hypotension in the Levobupivacaine with 1:400,000 Epinephrine patients (P = 0.15 and 0.08, respectively). Ephedrine was administered to 1 of 37 (3%) patients receiving plain levobupivacaine, 7 of 38 (18%) of patients administered levobupivacaine with 1:200,000 epinephrine, and 4 of 36 (11%) patients receiving levobupivacaine with 1:400,000 epinephrine (P = 0.09). There were no clinically significant electrocardiogram changes related to study drug, and no patients were withdrawn because of side effects.

There were no differences in plasma concentrations of local anesthetic in patients who received plain levobupivacaine compared with patients in the Epinephrine-Containing groups (Fig. 3), with peak concentrations (Levobupivacaine, 0.36 ± 0.15 μg/mL; Levobupivacaine with 1:200,000 Epinephrine, 0.33 ± 0.19 μg/mL; Levobupivacaine with 1:400,000 Epinephrine, 0.25 ± 0.10 μg/mL;P = 0.21) occurring 0.84 ± 1.19 h, 1.02 ± 1.12 h, and 0.43 ± 0.29 h after the end of injection, respectively (Table 4, P = 0.39). The t1/2 and the area under the plasma drug concentration-time curve calculated did not significantly differ among groups (Table 4).

Figure 3

Figure 3

Table 4

Table 4

Back to Top | Article Outline


This study demonstrates that 0.5% levobupivacaine, with or without epinephrine, is a suitable anesthetic for use in lumbar spine surgery. The addition of epinephrine tends to increase the duration of blockade, reduce the resultant local anesthetic concentration, and improve intraoperative anesthetic quality, although statistical significance was not reached for any of these measures. Mean onset of sensory blockade to T10 was achieved within 15 minutes of administering the epidural injection in all patient groups. Certain institutions routinely and successfully use both epidural and spinal anesthesia for lumbar spine surgery. Proposed advantages of neuraxial blockade include retaining the patient’s ability to turn to the prone position; guarding against compressive injuries; verbal contact with the patient, allowing precise localization of nerve root involvement; a decrease in intraoperative blood loss; and long-lasting postoperative analgesia (5–7).

Proposed disadvantages of regional anesthesia for lumbar spine surgery include the inability to immediately assess lower extremity motor function and a possible delay in bladder function. Yet, in this study, 53% of patients did not attain any degree of motor blockade.

This is the first study to evaluate the effect of adding epinephrine to epidural levobupivacaine. Our results are in agreement with previous studies looking at the effect of adding epinephrine to local anesthetic solutions. Both concentrations of epinephrine in this study decreased the peak plasma levobupivacaine concentrations, although this did not reach statistical significance. The 1:400,000 concentration also shortened the time required to reach this peak. Epinephrine consistently decreases peak plasma local anesthetic concentrations during epidural anesthesia, although it does not consistently reduce or prolong the time to reach these peak concentrations (4,8–10). Furthermore, vasoconstrictors have their greatest clinical effect on short-acting local anesthetics and do not always prolong the duration of blockade when added to long-acting drugs such as etidocaine or racemic bupivacaine (11).

Although a dose-response relationship of epinephrine added to local anesthetics has been established for subcutaneous (12) and intrathecal injections (13), this has not been established for the epidural injection of local anesthetics. In this study, two different concentrations of epinephrine were used. The addition of 2.5 μg/mL of epinephrine proved to be as effective as 5.0 μg/mL in reducing plasma levobupivacaine concentrations. Whether this is because levobupivacaine intrinsically produces a mild degree of vasoconstriction is not known. Both epinephrine concentrations also tended to increase the duration of sensory blockade above T10, although statistical significance was not reached for either (Levobupivacaine, 186 ± 67 minutes; Levobupivacaine with 1:200,000 Epinephrine, 202 ± 62 minutes; Levobupivacaine with 1:400,000 Epinephrine, 200 ± 67 minutes;P = 0.29).

Our results also show that the addition of either concentration of epinephrine to levobupivacaine increases the quality of intraoperative block as assessed by the anesthesiologist and surgeon. This increase in block density and reliability was also demonstrated when epinephrine (5 μg/mL) was added to racemic bupivacaine (14). According to multivariate analysis, the probability of a patient experiencing pain during surgery with epidural anesthesia is reduced by the addition of epinephrine (15).

The onset, peak block height, and regression of block in the patients who received the plain solution in this study are in agreement with previous studies in which plain 0.5% levobupivacaine was administered (16,17). All of these trials also confirm that effective clinical anesthesia is achieved with levobupivacaine and that the sensory blockade is substantially longer than any degree of motor blockade that may develop.

This study also confirms that levobupivacaine has a favorable profile with regard to adverse events. As expected, a decrease in systolic blood pressure, attributable to sympathetic blockade accompanying the epidural anesthesia, was the most common. Hemodynamic effects of epidural anesthesia are complex, because the volume, type, and concentration of local anesthetic, the level of blockade, the inclusion of vasoconstrictors, and physical status of the patient are all contributory. The addition of epinephrine generally produces a mild increase in heart rate and a greater decrease in blood pressure than plain epidural local anesthetics (4,8). In this study, reducing the concentration of epinephrine from 5.0 to 2.5 μg/mL decreased the incidence of hypotension to the same incidence as patients in the Plain Levobupivacaine group (from 23% to 10%), although this did not reach statistical significance (P = 0.08).

No cases of cardiac depression or central nervous system (CNS) toxicity caused by vascular absorption or direct intravascular injection of local anesthetic occurred in this study. One case of accidental intravascular injection of levobupivacaine (142.5 mg) has been reported, with the development of minimal CNS symptoms (transient agitation and disorientation) and no cardiovascular toxicity (18). In a study of 14 volunteers, the intentional intravascular administration (10 mg/min, until mild CNS symptoms developed) of levobupivacaine produced significantly fewer effects on myocardial function than racemic bupivacaine. Notably, changes in stroke index, acceleration index, and ejection fraction were less marked (19).

In conclusion, levobupivacaine is being developed as a local anesthetic with a similar pharmacodynamic profile to racemic bupivacaine, but with less risk of systemic toxicity. This study demonstrates that 0.5% levobupivacaine, with or without the addition of epinephrine, provides good epidural anesthesia for lumbar spine surgery, and all solutions are suitable for clinical use. The addition of epinephrine decreases plasma local anesthetic levels and improves intraoperative anesthetic quality. Decreasing the amount of added epinephrine from 5.0 to 2.5 μg/mL reduced the incidence of hypotension and tachycardia from 23% to 10% and 15% to 10%, respectively, although neither reduction was statistically significant.

The authors thank Jormain Cady, ARNP, Maureen Beaulieu, RN, Teresa Guthrie, RN, BSN, Pearl Washburn, RN, Rose Lopez, RN, and Jill Zeller, RN, for their assistance and dedication.

Back to Top | Article Outline


1. Huang Y, Pryor M, Mather L, Veering B. Cardiovascular and central nervous system effects of intravenous levobupivacaine and bupivacaine in sheep. Anesth Analg 1998; 86: 797–804.
2. Chang D, Ladd L, Wilson K, et al. Tolerability of large-dose intravenous levobupivacaine in sheep. Anesth Analg 2000; 91: 671–9.
3. Burke D, MacKenzie M, Newton D, et al. A comparison of vasoactivity between levobupivacaine and bupivacaine [abstract]. Int Monitor 1998; 10: 66A.
4. Baranowski A, Dean Y, Pither C. The effect of priming epidural injection of adrenaline on epidural blockade with bupivacaine. Anaesthesia 1991; 46: 1023–9.
5. Greenbarg P, Brown M, Pallares V, et al. Epidural anesthesia for lumbar spine surgery. J Spinal Disord 1988; 1: 139–43.
6. Tetzlaff J, O’Hara J, Bell G, et al. Influence of baricity on the outcome of spinal anesthesia with bupivacaine for lumbar spine surgery. Reg Anesth Pain Med 1995; 20: 533–7.
7. Gal R, Cundrle I, Stibor B, Smrcka M. Epidural anesthesia for lumbar disc surgery [abstract]. Br J Anaesth 1999; 82: A373.
8. Salevsky F, Whalley D, Kalant D. Epidural epinephrine and the systemic circulation during peripheral vascular surgery. Can J Anaesth 1990; 37: 160–5.
9. Burm A, van Kleef J, Olthof G, Spierdijk J. Epidural anesthesia with lidocaine and bupivacaine: effects of epinephrine on the plasma concentration profiles. Anesth Analg 1986; 65: 1281–4.
10. Keir L. Continuous epidural analgesia in prostatectomy: comparison of bupivacaine with and without epinephrine. Acta Anaesthesiol Scand 1974; 18: 1–4.
11. Tucker G. Pharmacokinetics of local anesthetics. Br J Anaesth 1986; 58: 717–31.
12. Liu S, Carpenter R, Chiu A, et al. Epinephrine prolongs duration of subcutaneous infiltration local anesthesia in a dose-related manner: correlation with magnitude of vasoconstriction. Reg Anesth 1995; 20: 378–84.
13. Racle J, Benkhadra A, Poy J, Gleizal B. Effect of increasing amounts of epinephrine during isobaric bupivacaine spinal anesthesia in elderly patients. Anesth Analg 1987; 66: 882–6.
14. Laishley R, Morgan B. A single dose epidural technique for caesarean section: a comparison between 0.5% bupivacaine plain and 0.5% bupivacaine with adrenaline. Anaesthesia 1988; 43: 100–3.
15. Curatolo M, Orlando A, Zbinden A, et al. A multifactorial analysis to explain inadequate surgical analgesia after extradural block. Br J Anaesth 1995; 75: 274–81.
16. Cox C, Faccenda K, Gilhooly C, et al. Extradural S(−)-bupivacaine: comparison with racemic RS-bupivacaine. Br J Anaesth 1998; 80: 289–93.
17. Bader A, Tsen L, Camann WR, et al. Clinical effects and maternal and fetal plasma concentrations of 0.5% epidural levobupivacaine versus bupivacaine for cesarean delivery. Anesthesiology 1999; 90: 1596–601.
18. Kopacz D, Allen H. Accidental intravenous levobupivacaine. Anesth Analg 1999; 89: 1027–9.
19. Bardsley H, Gristwood R, Baker H, et al. A comparison of the cardiovascular effects of levobupivacaine and rac-bupivacaine following intravenous administration to healthy volunteers. Br J Clin Pharmacol 1998; 46: 245–9.
© 2001 International Anesthesia Research Society