| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anaesthesia, HCI International Medical Centre, Clydebank, Scotland; Department of Anaesthesia, Crosshouse Hospital, Kilmarnock, Scotland; and Department of Anaesthesia, Royal National Orthopedic Hospital, Stanmore, England
Address correspondence and reprint requests to John A. C. Murdoch, Specialist Registrar in Anesthesia, South Glasgow University Hospitals, Victoria Infirmary, Langside Road, Glasgow, G42 9TV, Scotland, UK. Address e-mail to john-murdoch{at}btinternet.com
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
IMPLICATIONS: Postoperative epidural infusion of levobupivacaine can provide safe and effective analgesia for patients having hip or knee joint replacement. Of the three concentrations we infused at a constant rate, 0.25% provided significantly better pain relief.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Bupivacaine contains an asymmetric carbon atom that gives it a chiral center. Thus it is a racemic mixture of two enantiomers: levo- or S(-)-bupivacaine and dextro- or R(+)-bupivacaine. These have identical physical and chemical properties, but there is evidence of stereospecificity of action, in particular with relation to cardiotoxicity. Preclinical animal studies comparing levobupivacaine with either the dextro enantiomer alone (3) or together with the racemate (4) suggest that it is less cardiotoxic than either of these. In a study by Huang et al. (5), the mean convulsive threshold was significantly higher and ventricular arrhythmias fewer and easier to treat for levobupivacaine relative to bupivacaine in a sheep model. Human studies have also found fewer cardiac effects from levobupivacaine, with 40%–60% less reduction in myocardial contractility relative to the racemate (6). It is important to note that levobupivacaine and racemic bupivacaine seem to have similar clinical effects. No significant differences in onset time, dermatomal spread, or duration of sensory or motor blockade have been seen in comparative studies of brachial plexus blockade (7) or epidural blockade for lower limb surgery (8). This pattern of reduced toxicity, but equal clinical efficacy to bupivacaine, should make levobupivacaine a useful drug for epidural infusion for postoperative analgesia. Therefore, in this study, we aimed to assess the efficacy and safety of this new regional anesthetic administered as continuous epidural infusions for pain relief in patients undergoing elective orthopedic surgery.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects were premedicated with temazepam 20 mg and ranitidine 150 mg orally the night before surgery and again 1 h preoperatively. Nonsteroidal antiinflammatory drugs were discontinued the night before surgery.
Intraoperative monitoring included pulse oximetry, ECG, and noninvasive arterial blood pressure monitoring. A 16-gauge IV cannula was inserted and an IV crystalloid infusion started in the anesthetic room before regional blockade. An epidural was then placed, under local lidocaine infiltration, between T12 and L3 by use of a Tuohy needle and a 3 side-hole extradural catheter advanced 3–5 cm into the epidural space. After a negative aspiration test and a negative test dose of 3 mL of 0.75% levobupivacaine, a further 7 mL of the same solution was injected. After 15 min, the sensory block to pinprick was assessed with a 27-gauge dental needle. If inadequate for surgery, an additional 5 mL of the 0.75% levobupivacaine was given and the block reassessed after a further 15 min. If the block was still inadequate, the patient was withdrawn from the study. If it was adequate (taken to be above dermatomal level T10 for both hips and knees), sedation with IV propofol was commenced before surgical incision. During all bolus injections, patients were monitored for any symptoms or signs of local anesthetic toxicity.
Full monitoring was continued intraoperatively, and a five-lead ECG was recorded during surgery. Hypotension was defined as a decrease of more than 30% from the baseline systolic blood pressure or was confirmed by measurement if the patient became symptomatic (symptoms such as faintness or nausea). It was treated with vasopressors (ephedrine boluses) or IV fluids at the discretion of the investigator. Bradycardia was defined as a heart rate of <50 bpm.
Thirty minutes after completion of the last epidural injection, patients were randomized, according to a computer generated schedule, to one of three groups to receive an epidural infusion of 0.0625%, 0.125%, or 0.25% levobupivacaine. The randomization was stratified by joint to be replaced by using a randomized block procedure, and coded bags were prepared by the hospital pharmacy. This infusion ran continuously for 24 h at 6 mL/h, and all investigators and study patients were blinded as to the concentration of levobupivacaine in the infusion.
Sensory block (to pinprick), motor block, and standard 100-mm visual analog scale (VAS) pain scores were assessed hourly for the first 4 h after the start of the epidural arterial infusion (time 0). Subsequently, they were performed every 2 h for the next 8 h and then every 6 h thereafter up to 24 h, provided that the patient was awake. Motor block was assessed on the nonoperated limb by using the modified Bromage scale (0 = no motor block, 1 = inability to raise extended leg, 2 = inability to flex knee, 3 = inability to move lower limb). Heart rate and oxygen saturation were monitored continuously and noninvasive arterial blood pressure recorded hourly for 24 h post-operatively.
If during this time the patient requested rescue analgesia, the time of request relative to time 0 was recorded. They then received 2-mg boluses of morphine every 3–4 min until they were comfortable and morphine patient-controlled analgesia (PCA) was instigated (1-mg boluses with a 5-min lockout time). Droperidol 2.5 mg was added to each 60 mg of morphine PCA solution, as an antiemetic. A VAS pain score was completed before the initial administration of morphine. The total amount of morphine the patient received and the number of requests for analgesia from the pump during the study period were recorded.
At the end of the 24-h study period, the infusions of levobupivacaine were discontinued, alternative analgesia was provided, and the biochemical and hematologic investigations were repeated. Any adverse events (such as hypotension, bradycardia, nausea and vomiting, and urinary retention) occurring during the period of the study were documented. In addition, patients were followed up 12 h postinfusion and then 3–5 days after completion of the study to detect any subsequent adverse events or changes to previously noted events.
On the basis of previous experience, it was estimated that the between-patient SD for first request for analgesia would be 2 h. It was calculated that a minimum of 21 evaluable patients would be required to be recruited per group to detect a 2-h difference with a power of 80% and corrected P < 0.05. Therefore, we aimed to recruit a minimum of 30 evaluable patients per treatment group. Time to first request for analgesia and normalized doses of morphine were analyzed initially with analysis of variance and then with Student’s t-test and Wilcoxon’s two-sample test, respectively. Because the first observation contained censored results (i.e., patients who did not request any analgesia in the 24-h period), secondary analysis using survival techniques was performed. Kaplan-Meier survival curves were constructed and a Cox’s proportional regression model used to make pairwise comparisons between groups. VAS pain scores were analyzed by calculating the normalized area under the curve for VAS/time, and the data were analyzed with Wilcoxon’s two-sample tests. The maximum grade of motor blockade was analyzed with a logit model, and pairwise comparisons were performed with the Wald test statistic. Time to achieve a motor block of zero was analyzed with survival analysis and Wilcoxon’s test. Correction for multiple comparisons was made where appropriate by using the Bonferroni-Holm method to preserve the overall significance level as P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Early mobilization of patients after major joint replacement is desirable, and avoidance of excessive motor blockade from epidural analgesia is clearly important for facilitating this. Therefore, for any postoperative epidural infusion regimen, the problem is to balance adequate analgesia against the increase in motor blockade that can occur with increasing the concentration of local anesthetic. In our study, there was a trend toward a denser maximum and longer-acting motor blockade in the group receiving the largest concentration, and both the 0.125% and 0.25% groups had significantly denser maximum motor blockade than the 0.0625% group, indicating a degree of dose-dependent response. However, there was no significant difference in degree of maximum motor blockade or in time to reach minimum motor blockade between the 0.125% group and those receiving 0.25% levobupivacaine. Nevertheless, there was a significant improvement in analgesia in the latter group. This may be evidence of a desirable sensory/motor blockade ratio of effect for levobupivacaine. A previous study by Cox et al. (8) on 96 patients undergoing lower limb surgery under epidural anesthesia found no significant differences in the nature and quality of blockade produced by 0.5% levobupivacaine compared with 0.5% racemic bupivacaine, although there did seem to be a trend toward less motor blockade with the former drug. However, because our study did not make a comparison with the racemate, we cannot comment any further on any difference of activity with the racemic bupivacaine.
Although approximately 47% of the group receiving 0.25% levobupivacaine infused at a constant 6 mL/h did not require rescue analgesia, further pain relief was required in the majority of cases. It is possible that by increasing the rate of infusion, a larger percentage of our patients would have remained pain free but at the cost of increased motor blockade. Raj et al. (9) compared epidural infusion of 0.25% racemic bupivacaine with conventional opioid postoperative analgesia in 30 patients undergoing knee replacement surgery. Only 20% of their epidural group required supplementary analgesia. However, their mean bupivacaine dosage over the first 24 hours was 21 mg/h, compared with our constant 15 mg/h of levobupivacaine. Notably, the percentage of patients with complete motor blockade in their epidural group was much larger at 60%–70% on the first postoperative day, compared with our 22% in the 0.25% group recording complete paralysis as their maximum block. To achieve adequate analgesia without excessive motor blockade, a combination of levobupivacaine with another analgesic, such as an opioid, or another adjunct, such as clonidine, may be advisable. Additive effects allowing a reduction in dose of each drug have previously been described for racemic bupivacaine and fentanyl in patients undergoing joint replacement surgery (10) and for the racemate in combination with clonidine (11), and it is likely that similar effects will allow reduced local anesthetic dosage for levobupivacaine.
It would seem from the results of those patients not requiring rescue analgesia that the majority of these in the 0.25% group were undergoing hip replacement. In many cases this was despite having detectable sensory block over the operative dermatomes. From this and our clinical impression it would seem that the postoperative pain from a knee replacement is more intense and requires a denser sensory block or adjuvant analgesia in the majority of cases. Because the use of continuous passive movement during the study period was one of the exclusion criteria, this could not account for the difference. It is a potential weakness of our study that we included both hip and knee joint replacement patients and that despite block randomization, ultimately, as a result of patient withdrawal after randomization, there were not exactly equal numbers of each surgical procedure among groups. However, in all three groups there was a clear majority of hip replacement patients and approximately equal numbers of knee joint replacements. We, therefore, believe that the large and highly statistically significant reduction in supplementary analgesia required in the 0.25% group represents a real and clinically important finding.
The most frequently reported adverse event in our study was hypotension, which occurred in up to 60% of patients in the 0.25% group. This appears relatively frequently. A retrospective audit of four years’ experience of epidural infusions of 0.15% bupivacaine for postoperative analgesia in general surgical patients reported an overall incidence of 34% for documented hypotension in the first 24 hours (12). However, in a dose-finding study for ropivacaine for patients undergoing major lower limb orthopedic surgery, the control group, which received an infusion of normal saline, recorded an incidence of hypotension of 45%, with an incidence of 54% in the group receiving 0.3% ropivacaine at 10 mL/h (13). Despite our lack of a placebo control group, it is likely that the relatively frequent incidence in all three of the study groups reflects, at least in part, a frequent background occurrence of perioperative hypotension in this population of predominantly elderly patients undergoing major orthopedic surgery. The reduced mean arterial pressures both intra- and postoperatively in patients undergoing total hip replacement under epidural anesthesia may actually contribute to smaller blood losses relative to patients who receive general anesthesia (14). However, this also reinforces the need for close monitoring of cardiovascular variables and treatment of hypotensive episodes after major orthopedic surgery.
In general, the block height achieved with our protocol was above that required for lower limb surgery, with median preoperative block heights between T6 and T7. It is likely that either using a smaller volume of 0.75% levobupivacaine or a smaller concentration of the drug would be more appropriate to establish initial blockade. Also it is possible that the 15-minute interval allowed between bolus injection and assessment of block height was not adequate to allow full extension of the sensory blockade. A longer time interval may have reduced the need for the second injection in some cases. Indeed, the high sensory block, combined with preoperative ß-blockade, was implicated in the bradycardia recorded as a severe adverse event, although it was considered that such an event could have occurred with any local anesthetic and was not specific to levobupivacaine. Overall, levobupivacaine proved to be well tolerated when administered as a continuous epidural infusion.
In conclusion, this study found that 0.25% levobupivacaine, as a continuous epidural infusion, provided significantly longer analgesia than 0.125% or 0.0625% levobupivacaine, without any significant increase in maximum degree or duration of motor blockade relative to the 0.125% dose. However, a smaller dose of levobupivacaine (0.125%) in combination with an opioid or nonopioid analgesic may achieve adequate analgesia without excessive motor blockade. Safety data were similar among the three groups.
|
Acknowledgments |
|---|
We would like to thank Inveresk Clinical Research for their assistance in coordinating the data collection and statistical analysis.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Gupta, A. Perniola, K. Axelsson, S. E. Thorn, K. Crafoord, and N. Rawal Postoperative Pain After Abdominal Hysterectomy: A Double-Blind Comparison Between Placebo and Local Anesthetic Infused Intraperitoneally Anesth. Analg., October 1, 2004; 99(4): 1173 - 1179. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Casati, F. Vinciguerra, G. Cappelleri, G. Aldegheri, C. Grispigni, M. Putzu, and P. Rivoltini Levobupivacaine 0.2% or 0.125% for Continuous Sciatic Nerve Block: A Prospective, Randomized, Double-Blind Comparison with 0.2% Ropivacaine Anesth. Analg., September 1, 2004; 99(3): 919 - 923. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Senard, A. Kaba, M. J. Jacquemin, L. M. Maquoi, M.-P. N. Geortay, P. D. Honore, M. L. Lamy, and J. L. Joris Epidural Levobupivacaine 0.1% or Ropivacaine 0.1% Combined with Morphine Provides Comparable Analgesia After Abdominal Surgery Anesth. Analg., February 1, 2004; 98(2): 389 - 394. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Dernedde, M. Stadler, F. Bardiau, and J. G. Boogaerts Continuous Epidural Infusion of Large Concentration/Small Volume Versus Small Concentration/Large Volume of Levobupivacaine for Postoperative Analgesia Anesth. Analg., March 1, 2003; 96(3): 796 - 801. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
