 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Intra-Articular Injection of Bupivacaine in Knee-Replacement Operations. Results of Use for Analgesia and for Preemptive Blockade * , LONDON, ONTARIO, CANADAInvestigation performed at the London Health Sciences Centre, University Campus, London.
The effectiveness of an intra-articular injection of bupivacaine, administered before the incision or after closure of the wound, was studied in an effort to decrease the need for postoperative narcotics and to improve analgesia for patients who have elective knee replacement. Eighty-two patients received two intra-articular injections in a random, double-blind fashion. Twenty-eight of them received thirty milliliters of 0.5 per cent bupivacaine and 1:200,000 epinephrine in saline solution before the incision and an injection of thirty milliliters of plain saline solution after closure of the wound (Group 1). Twenty-seven patients received an injection of thirty milliliters of plain saline solution before the incision and thirty milliliters of 0.5 per cent bupivacaine and 1:200,000 epinephrine in saline solution after closure of the wound (Group 2). Twenty-seven patients were given thirty milliliters of plain saline solution (a placebo) for both injections (Group 3). The patients who had received bupivacaine after closure of the wound (Group 2) used less morphine from the patient-controlled analgesia pumps than the patients who had received bupivacaine before the incision (Group 1) and the patients who had received the placebo (Group 3). In the first twenty-four hours after the operation, the administration of morphine (mean and standard deviation) was 59 ± 27 milligrams for Group 2 compared with 68 ± 30 milligrams for Group 1 (p = 0.26) and 81 ± 30 milligrams for Group 3 (p = 0.006). At the time of discharge from the hospital, the patients in Group 2 also had a significantly greater mean range of motion (85.2 ± 8.0 degrees) compared with that of the patients in Groups 1 (80.6 ± 6.8 degrees, p = 0.02) and 3 (80.1 ± 6.2 degrees, p = 0.009). However, there was no difference among the groups with respect to the effectiveness of the analgesia, as measured with use of either the visual-analog or the verbal pain-rating scale, or in the prevalence of side effects, including somnolence, urinary retention, nausea and vomiting, or pruritus. Serum concentrations of bupivacaine were well below toxic levels. It was our conclusion that that an intra-articular injection of thirty milliliters of 0.5 per cent bupivacaine and 1:200,000 epinephrine in saline solution after closure of the wound decreases the need for narcotics and increases the range of motion after an elective knee replacement. The clinical importance of the amount of increased motion is questionable and needs longer-term monitoring.
Knee replacement is a major operation associated with considerable postoperative pain and need for analgesics3. Narcotics are routinely administered for the relief of pain, but they may cause respiratory depression. As a result, optimum analgesia often is not achieved because of a fear that this side effect will be precipitated. Measures to decrease both the pain and the use of narcotics are thus beneficial. One such method is the local injection of an anesthetic into the operative site. This is done routinely, with an intra-articular injection, after arthroscopic operations on the knee. Such injections have improved analgesia, decreased the need for postoperative narcotics, and led to earlier discharge from the hospital2,9,12,14,15,18. Research conducted with animal models has also suggested that trauma to tissue, such as that occurring during an operation, can lead to long-lasting changes in the perception of pain by the central nervous system. Inhibition of neural input by local injection of an anesthetic before the trauma of an operation may prevent these changes and minimize subsequent pain postoperatively21. This technique has been termed preemptive blockade. Recent clinical investigations involving patients who were having inguinal herniorrhaphy have supported this theory8,20. They showed that local injection of an anesthetic subcutaneously before the incision is superior, in terms of providing postoperative analgesia, to a similar injection after closure of the wound. To our knowledge, however, no attempts have been made to study the effect of a local intra-articular injection of an anesthetic either before or after the incision for knee replacement. Therefore, the present study was designed to investigate the analgesic effects of intra-articular injection of bupivacaine in patients having elective knee replacement as well as to compare the effectiveness of such an injection before the incision with that after closure of the wound in a prospective, random, double-blind fashion.
After our study had been approved by the institutional ethics committee, patients who were scheduled for elective knee replacement and who had signed an informed-consent form were considered eligible for enrollment. Patients were excluded if they were more than seventy-five years old, weighed more than 110 kilograms, had a physical status rated class IV or V according to the system of the American Society of Anesthesiologists16, had a psychiatric illness, had a dependence on alcohol or narcotics, or had a known allergy to morphine or to any local anesthetic. Eighty-two patients who were scheduled for elective knee replacement were enrolled in the study and randomly assigned to one of three groups. Other than the distribution of the preoperative deformities of the knee (p = 0.046), none of the demographic data (age, height, weight, gender distribution, diagnosis, and score on The Knee Society clinical-rating scale10) differed significantly among the three groups (Table I). There were also no significant differences among the three groups with regard to the operative data, including the duration of the operation and of the anesthesia, the duration for which a tourniquet had been used, the type of implant (anatomic modular knee [AMK; DePuy, Warsaw, Indiana], self-aligning knee [SAL; Protek, Bern, Switzerland], Genesis [Smith and Nephew Orthopaedics, Memphis, Tennessee], or Miller-Galante I hemiarthroplasty [Zimmer, Warsaw, Indiana]), the method of fixation, the prevalence of lateral release, or the intraoperative use of narcotics (Table I).
The patients were randomly and prospectively assigned to the three groups by means of a computer-generated randomization code. The patients in Group 1 received an intra-articular injection of thirty milliliters of 0.5 per cent bupivacaine and 1:200,000 epinephrine in saline solution before the skin incision was made and thirty milliliters of plain saline solution after the wound was closed. The patients in Group 2 received thirty milliliters of plain saline solution before the incision and thirty milliliters of 0.5 per cent bupivacaine and 1:200,000 epinephrine in saline solution after closure. The patients in Group 3 received thirty milliliters of plain saline solution (a placebo) for both injections. The syringes were prepared by the pharmacy at University Hospital in a double-blind fashion and were labeled only as the preincisional injection or the postincisional injection. Immediately after the induction of general anesthesia, the patient received the initial intra-articular injection. The lower limb was then prepared and draped in the usual fashion, it was exsanguinated, and a tourniquet was applied. The knee replacement was performed through a midline approach with a medial capsular incision. A drain was not inserted at closure of the wound. Postoperatively, on transfer to the postanesthesia-care unit, the patient received morphine intravenously (0.02 milligram per kilogram of body weight at five-minute intervals) until he or she was comfortable. The patient then received morphine through a patient-controlled analgesia pump with a dose of 0.02 milligram per kilogram of body weight, a lock-out period (a duration of time when no doses were delivered) of eight minutes, and no basal infusion. The patients assessed the intensity of the pain at one, two, four, and twenty-four hours after they had arrived in the postanesthesia-care unit. The assessments were made with use of two rating scales: a 4-point verbal scale with which the severity was identified as none (0 points), mild (1 point), moderate (2 points), severe (3 points), or very severe (4 points), and a visual-analog scale on which zero was equivalent to "no pain" and 100 millimeters, to "the worst pain possible." Data on the total amount of morphine administered were stored in the patient-controlled analgesia pump and were recorded by automatic printout after twenty-four hours. An assessment of somnolence, which was performed at the same times as the assessment of pain, was carried out with use of a 4-point scale. One point was given if the patient was oriented to person, place, and time and initiated conversation; 2 points were given if the patient was sleeping but awoke and responded to verbal commands and was oriented to person, place, and time; 3 points, if the patient was drowsy and responded to verbal commands and painful stimulation but was disoriented to person, place, or time; and 4 points, if the patient responded to painful stimulation only. The presence of nausea and vomiting, pruritus, or urinary retention was also determined at the same times as the measurements of pain and somnolence. The range of motion of the involved knee was measured, with use of a goniometer accurate to within 1 degree, preoperatively and at the time of discharge from the hospital. This measurement was performed by one of two staff physiotherapists who were not connected with the study. A sample of blood for serum bupivacaine assay was drawn when the patient arrived in the postanesthesia-care unit and again two hours later. The samples were centrifuged and were stored at -20 degrees Celsius for later analysis. The final analysis of all of the samples was performed with use of high-performance liquid chromatography and standard techniques1. In our laboratory, the assay was sensitive to one nanogram per milliliter with an interassay variability of 4 per cent. Statistical evaluation consisted of two-tailed analysis of variance with the Fisher protected least-significant-difference post hoc test for the demographic data (age, height, weight, knee score, duration of the operation, and intraoperative use of narcotics), the use of morphine through the patient-controlled analgesia pump, the scores on the visual-analog scale, the range of motion at the time of discharge from the hospital, and the concentrations of bupivacaine in the serum. Chi-square analysis was used to compare the distributions of demographic data (gender, diagnosis, preoperative deformity, type of implant, method of fixation, and lateral release) and the prevalence of side effects. The Kruskal-Wallis test was used for analysis of the verbal-rating-scale and somnolence scores. A finding was considered significant if p was less than 0.05.
The average scores on the visual-analog scale ranged from thirty to sixty millimeters during the study period. The patients who had received bupivacaine after closure of the wound (Group 2) showed a trend toward lower mean scores during the twenty-four-hour period of the study. However, with the numbers available, we could not detect a significant difference among the scores for the groups (p = 0.14). The need for morphine during the first twenty-four hours was significantly lower for Group 2 (59 ± 27 milligrams [mean and standard deviation]) than for Group 3 (81 ± 30 milligrams, p = 0.006) but, with the numbers available, we could not detect a significant difference between Group 1 (68 ± 30 milligrams) and Group 3 (p = 0.09). The range of motion at the time of discharge from the hospital was significantly greater in Group 2 (85.2 ± 8.0 degrees) than in Group 1 (80.6 ± 6.8 degrees, p = 0.02) or Group 3 (80.1 ± 6.2 degrees, p = 0.009). There was no difference in the scores on the verbal rating scale, the prevalence of side effects, or the level of somnolence among the three groups at any of the measurement times. The mean concentrations of bupivacaine in serum obtained after the patients had arrived in the postanesthesia-care unit and two hours later ranged from 100 to 250 nanograms per milliliter. The maximum level was 849 nanograms per milliliter, which was found at two hours in a patient in Group 2.
Knee replacement is a major operative procedure that is known to be associated with considerable postoperative pain and need for analgesics3. In an effort to diminish both of these problems, we investigated the use of a local intra-articular injection of an anesthetic, bupivacaine, in patients having a knee replacement. We also attempted to determine if the timing of the injection, either before the incision or after closure of the wound, altered the effect of this technique. We found that the intra-articular injection of bupivacaine after closure of the wound reduced the need for narcotics and increased the range of motion at the time of discharge from the hospital. However, with the numbers available, we were unable to demonstrate significantly better analgesia with this injection, as measured by scores on the visual-analog or verbal rating scales for intensity of pain. We also did not find that an intra-articular injection before incision decreased the use of morphine or increased the range of motion. The concept of preemptive blockade, whereby a local injection of an anesthetic before trauma to the tissue (such as occurs during an operation) produces better analgesia than a similar injection after trauma, is in fact highly controversial6. Clinically, injection before incision has been shown to be more effective than such an injection after closure of the wound in patients having repair of an inguinal hernia8. However, another study, which was identical in design and involved patients having a similar procedure, demonstrated no benefit to injection before incision7. To our knowledge, the effect of preemptive blockade with the use of an intra-articular injection has not been studied in patients having either arthroscopy or arthroplasty. We used thirty milliliters of 0.5 per cent bupivacaine as this dose has been shown to be effective for patients having an arthroscopic operation on the knee12,18. The dose was also chosen because studies in which a smaller volume (twenty milliliters) or a lower concentration of bupivacaine (0.25 per cent) had been used revealed inconsistent findings11,13,15. A general anesthetic was used for all of the patients in our study, in order to eliminate the known differences in postoperative analgesia that occur when spinal anesthesia is used instead of general anesthesia9,20. Also, in order to prevent the local anesthetic from leaking out of the knee before it became effective in the patients who had received the injection after closure of the wound, a suction drain was not used. Closing the wound without placement of a drain has been shown to be a safe practice in knee replacement4. The mean serum concentrations of bupivacaine in both treatment groups (Groups 1 and 2) at the measured times were well below the toxic level of 2000 nanograms per milliliter (two micrograms per milliliter). Although we measured the serum levels only after the patient had arrived in the postanesthesia-care unit and two hours later (at approximately fifteen minutes and at two hours and fifteen minutes after injection for Group 2 and at approximately two hours and fifteen minutes and at four hours and fifteen minutes after injection for Group 1), the mean serum levels were similar to those reported by Butterworth et al. after injection of an identical solution of thirty milliliters of 0.5 per cent bupivacaine with 1:200,000 epinephrine during arthroscopic operations on the knee. Because the serum levels in that study were found to peak thirty minutes after injection, it is unlikely that our measurements missed higher concentrations. The low levels may have resulted from the addition of the 1:200,000 epinephrine in saline solution, which has been shown to be effective for this purpose19. We recommend the use of the 1:200,000 epinephrine in saline solution until studies have been performed on the use of bupivacaine without its addition. Our protocol was limited by the fact that the patients in Group 1 had approximately fifteen minutes before incision and thirty minutes before arthrotomy for the bupivacaine to take effect. After that, the remaining, unabsorbed bupivacaine was lost through the wound. In comparison, the patients in Group 2 had an unlimited time for the bupivacaine to become absorbed and to take effect. In order to overcome this shortcoming, the patients in Group 1 would have had to receive the injection before induction of general anesthesia, which we did not think was clinically acceptable. We also did not study the effect of the combination of bupivacaine and epinephrine on either the healing of the wound or the fixation of the prosthesis with cement. We are unaware of studies of the effect of local anesthetics on the fixation of a prosthesis with cement. However, the topical use of lidocaine spray after inguinal herniorrhaphy has been shown to be safe with regard to the healing of the wound17. Although the use of our technique did not improve analgesia or decrease the prevalence of side effects, the need for narcotics was decreased by 25 per cent in the first twenty-four hours after the operation. These results are superior to those reported after the use of this technique in arthroscopy of the knee9,14,15. The use of narcotics in those studies was found to decrease significantly (p < 0.05) in the first few hours with only trends toward a 15 per cent reduction during the first twenty-four hours. Still, the technique is used routinely in arthroscopy of the knee. On the basis of these findings, we think that it should also be used routinely in knee-replacement operations. The mean improvement of 5 degrees in the range of motion at the time of discharge from the hospital may not be clinically important. Longer-term monitoring of these patients is necessary to determine if this improvement is maintained and, therefore, whether this technique offers any benefits for improving the range of motion. In summary, we have demonstrated that intra-articular injection of thirty milliliters of 0.5 per cent bupivacaine and 1:200,000 epinephrine in saline solution in patients having elective knee replacement can safely reduce the postoperative need for narcotics and increase the range of motion at the time of discharge. The clinical importance of the amount of increased motion is questionable and needs longer-term monitoring. We also found no evidence of preemptive blockade with use of this technique.
* No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding sources were The Physicians Services Incorporated Foundation and the Canadian Anaesthetists' Society Research Foundation Zeneca Pharma Incorporated Award.
  CiteULike  Connotea  Del.icio.us  Technorati What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
Department of Anaesthesia (N. H. B. and J. A. D.) and Department of Orthopedic Surgery (R. B. B.C. H. R., and S. J. MacD.), London Health Sciences Centre, University Campus, P.O. Box 5339, London, Ontario N6A 5A5, Canada. 
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |