| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Departments of *Anesthesiology and Critical Care Medicine, Orthopaedic Surgery, and Clinical Research, Children’s Hospital of Philadelphia, Pennsylvania
Address correspondence to Giovanni Cucchiaro, MD, Children’s Hospital of Philadelphia, Room, 9th Floor, 34th St. and Civic Center Blvd., Philadelphia, PA 19104. Address e-mail to Cucchiaro{at}email.chop.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Reconstruction of the ACL is often associated with significant immediate postoperative pain. Traditional methods of analgesia include the intraarticular (IA) administration of medications, either local anesthetics, opioids, or a combination of the two, and regional anesthesia. Both techniques improve immediate postoperative pain control when compared with sham intervention (4,5). However, when the effects of the femoral nerve block on postoperative analgesia have been compared with the IA infiltration IA (6–9), data have been contradictory.
In this randomized, prospective study, we compared the effects of a femoral and sciatic nerve block (FSNB) versus the IA infiltration of a combination of medications on postoperative analgesia in children undergoing ACL reconstruction. The primary aim of this study was to determine whether a FSNB using bupivacaine and clonidine could decrease the consumption of morphine in the immediate postoperative period compared with the IA administration of a combination of bupivacaine-clonidine-morphine. The secondary aims were to determine whether the pain scores and the incidence of side effects, such as vomiting, pruritus, and sedation, were affected by the two different techniques.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Patients were randomized into one of two study groups: FSNB or IA. The FSNB group received a combined FSNB using bupivacaine and clonidine. The IA group received IA injection of bupivacaine-clonidine-morphine. The FSNB were performed by anesthesiologists, and surgeons performed the IA injections. Only two anesthesiologists and two surgeons participated in the study to assure uniform anesthetic and surgical techniques.
The following data were collected by the nurses in the recovery room and on the patient care units: (a) pain score, assessed using a visual analog scale (VAS) ranging from 0 to 10 cm with 10 cm being the worst pain imaginable, (b) the amount of fentanyl used during the surgery, (c) the amount of morphine administered, (d) time to first demand dose of morphine patient-controlled analgesia (PCA), (e) incidence of pruritus, (f) number of episodes of vomiting, (g) doses of ketorolac, diphenhydramine, and ondansetron administered, (h) sedation level using a De Kock scale (10), (i) duration of sensory block testing sensation in the distribution areas of both the femoral and sciatic nerve, and (j) duration of motor block using a modified Bromage score (11). The evaluating nurses had been trained to assess and document the outcomes studied.
General anesthesia was induced with IV propofol (3–5 mg/kg). Patients randomized to receive a FSNB were endotracheally intubated without neuromuscular blocking drugs to identify femoral and sciatic nerve twitch responses during block placement; patients randomized to receive an IA were endotracheally intubated after the administration of vecuronium (0.1 mg/kg). The anesthesiologist could use muscle relaxants (vecuronium, 0.1/mg/kg) during the procedure in both groups of patients. Neuromuscular block was reversed with atropine (20 µg/kg) and neostigmine (30–70 µg/kg) before endotracheal extubation. IV fentanyl (1 µg/kg) was administered before intubation in every patient. Anesthesia was maintained with desflurane in nitrous and oxygen. Additional fentanyl (1 µg/kg) was allowed during surgery in the case of persistent tachycardia and hypertension (defined as values 20% more than baseline) that did not respond to increases of the inspired desflurane concentration. Every patient received ondansetron 50 µg/kg (maximum, 4 mg) at the end of the operation.
In patients randomized to the FSNB group, femoral and sciatic nerves were localized with a nerve stimulator (Stimuplex Dig RC, B. Braun Medical Inc., Bethlehem, PA) and insulated needle. After sterile prepping and draping, the femoral nerve was localized with a 24-gauge, 1.5-in. insulated needle (Stimuplex, B. Braun Medical), which was introduced just lateral to the femoral artery and 2 cm inferior to the inguinal ligament. The nerve block was performed when contractions were lost at a stimulator intensity of 0.5–0.7 mA using 0.5 mL/kg of 0.125% bupivacaine with 1:200,000 epinephrine (maximum, 40 mL) and clonidine 1 µg/kg (maximum, 100 µg) after negative aspiration for blood and negative heart rate responses to a test dose of 3 mL of the local anesthetic solution. The sciatic nerve was approached from the lateral thigh (12). A 21-gauge, 4-in. needle was used. Common peroneal or tibial nerve stimulation was sought with a stimulator intensity of 0.5–0.7 mA. The block was performed using 0.5 mL/kg of 0.125% bupivacaine with 1:200,000 epinephrine (maximum, 20 mL) and clonidine 1 µg/kg (maximum, 100 µg) after negative aspiration and test dose. In those patients randomized to the IA group, surgeons performed the injection of medication after tracheal intubation and before prepping the leg. A thigh tourniquet was first inflated to 300 mm Hg. Patients then received 1 mL/kg of 0.25% bupivacaine (maximum, 30 mL), morphine 5 mg, and clonidine 1 µg/kg. The tourniquet was kept inflated for 15 min after injection.
Before starting the arthroscopic portion of the surgery, a quadrupled hamstring autograft or an Achilles tendon allograft was prepared, depending on patient preference. A graft 12 cm in length and 9 or 10 mm in diameter was used, and the grafts were tubularized using suture material. The ruptured ACL was debrided with an arthroscopic curved full-radius shaver, a straight biter, and a pituitary rongeur. A tibial ACL guide was set at the tibial ACL footprint. A guide pin and a 10-mm cannulated drill bit were used to create the tibial tunnel. The femoral tunnel was then placed at the usual 11 o’clock position for the right knee and the 1 o’clock position for the left knee, leaving a small rim of bone at the posterior wall of the femoral tunnel. A femoral guide pin was overdrilled with a 10-mm cannulated drill bit and drilled to 40 mm in length. A femoral guide for the transverse femoral fixation system (Rigidfix Cross Pin System; Mitek, Westwood, MA) was then placed such that the bioresorbable cross pins were placed proximal to the femoral physis. Two transverse tunnels were drilled using the femoral guide and cannula. The arthroscopic camera was then placed within the femoral tunnel to visualize the appropriate location of the transverse fixation points, proximal to the level of the femoral physis. The camera was also placed within the tibial tunnel to evaluate the bone stock and the tibial physis to ensure fixation of the graft distal to the tibial physis. The graft was passed through the tibial and femoral tunnels and held in position while two bioabsorbable cross pins were tamped lateral to medial across the grafts in the femoral tunnel. The knee was placed in full extension as the distal length of the graft was pulled distally. Soft tissue screw fixation of the graft was then performed distal to the tibial physis.
Upon arrival to the recovery room, patients were started on a morphine PCA (demand bolus, 20 µg/kg, lockout interval 8 min, and maximum dose in 1 h 100 µg · kg–1 · h–1). If patients were in severe pain on arrival to the recovery room (VAS >3 cm), a single dose of IV ketorolac (0.5 mg/kg; maximum, 30 mg) was given. In case of persistent pain, boluses of morphine 50 µg/kg (maximum, 3 mg per dose) were given until the VAS score was <3 cm. After discharge to the care unit, morphine PCA was continued using the same settings for the next 18 h. Patients with persistent pain (VAS >3 cm) initially received IV ketorolac every 6 h (0.5 mg/kg; maximum, 30 mg/dose). When ketorolac was not effective and the VAS score was still more than 3 cm, patients received repeated doses of IV morphine (50 µg/kg; maximum, 3 mg per dose). Morphine PCA was discontinued after 18 h, and patients were transitioned to oral oxycodone (0.1 mg/kg; maximum, 10 mg) and acetaminophen (15 mg/kg; maximum, 650 mg) every 4 h and discharged home. Patients complaining of pruritus were given IV diphenhydramine (0.5 mg/kg; maximum, 50 mg) every 8 h as required. Patients complaining of more than one episode of vomiting were given ondansetron (0.1 mg/kg; maximum, 4 mg per dose) every 8 h.
Sample size was based on the primary outcome measure involving the amount of morphine used in the 24 h after ACL reconstruction. Presuming a 75% reduction of use of morphine in patients receiving a FSNB, a sample size of 15 in each group would have an 80% power to detect a difference between the groups.
The outcomes between the groups were compared with Student’s t-test or Fisher’s exact test, where appropriate, for continuous data or proportions. Kaplan-Meier survival curves were made, and the log-rank test was used to compare the time to first PCA use. Analysis of variance for repeated measures was performed to compare differences in pain scores between the two treatments with time. Univariate linear regression was performed to verify the existence of a correlation between some of the variable considered in the final analysis. Patients were not analyzed by intention-to-treat because patients whose blocks failed were excluded from the final analysis a priori. Data are presented as mean ± sd. Statistical significance was defined as P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
The intraoperative fentanyl requirement was significantly less after FSNB (50 ± 40 µg) compared with IA (80 ± 50 µg) (P = 0.04). Fewer patients required rescue doses of morphine in the recovery room after FSNB compared with patients who received IA (one versus six patients, respectively; P = 0.061). The consumption of morphine in the 18 h after the discharge from the recovery room, including both PCA usage and rescue doses, was significantly less in the FSNB group (7 ± 13 mg) compared with IA group (21 ± 21 mg) (P = 0.03). Only three patients in the FSNB group required morphine in the postoperative period compared with 15 of the IA patients (P = 0.0001) (Table 2). There was a significant difference in the duration of analgesia between the two groups (P = 0.004) (Fig. 1). Patients in the FSNB group had a significantly longer time to first analgesic request compared with the IA group (449 ± 312 versus 116 ± 222 min, respectively). After arrival to the patient care units, more IA patients (10 of 18 or 55%) required nurse-administered morphine rescues than FSNB patients (3 of 16 or 19%) (P = 0.04). Administration of ketorolac (doses and timing) was similar between the two groups in the recovery room and patient care units, although there was a trend towards less ketorolac administration in the recovery room (Table 2). The average use of morphine during the course of the study was significantly larger in patients in whom the surgeons used an Achilles tendon allograft (16 ± 15 mg) compared with patients receiving a hamstring tendon autograft (5 ± 7 mg) over the course of the study (P = 0.02).
|
|
The mean pain score in the FSNB group was less than in the IA group on arrival to the recovery room (1.8 ± 3 versus 5.4 ± 3; P = 0.0002) and in the 18 h of the study period (1.6 ± 1 versus 2.9 ± 2; P = 0.01; Fig. 2). There was no correlation between pain score and surgical technique. Analysis of variance showed significantly different VAS scores between the two groups from the time patients reached the recovery room to 12 h after the operation (P = 0.04).
|
The incidence of vomiting was significantly different in the two groups. One patient in the FSNB and two patients in the IA group vomited while in the recovery room. These three patients had further episodes of vomiting during the study period. Two patients in the FSNB (11%) and nine patients in the IA group (50%) (P = 0.03) vomited after their discharge from the recovery room.
The incidence of pruritus and sedation were similar in the two groups (Table 3), and no significant differences were found in the usage of ondansetron and diphenhydramine. One patient in the IA group suffered from urinary retention. None of the FSNB patients experienced prolonged paresthesia. The duration of the sensory block was of 17 ± 4 h in the area of the femoral nerve distribution and 20 ± 4 h in the area of the sciatic nerve distribution. Fifty percent of the patients experienced a motor block on the femoral or sciatic nerve distribution. The duration of the motor block was 4 ± 6 h in the area of the femoral nerve distribution and 6 ± 7 h in the area of the sciatic nerve distribution.
|
No correlation was found between sex or administration of neostigmine and episodes of vomiting (P = 0.68). However, a significant correlation was found between treatment (FSNB versus IA) and vomiting (P = 0.014), as well as between consumption of morphine and vomiting (P = 0.001).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Opposite conclusions have been reached by other authors who have shown that a femoral nerve block provides a better analgesia when compared with the IA infiltration (17,18). However, in both these studies, the IA infiltration group received only local anesthetic when it had been demonstrated that a combination of local anesthetic, morphine, and clonidine results in better pain control than a local anesthetic alone in patients undergoing arthroscopic knee surgery (19–21).
Our data show that children who received a FSNB had a significantly lower pain score in both the immediate and late postoperative course compared with children who received IA medications. This resulted in a significant less consumption of IV morphine and incidence of side effects. This study does not confirm recent data from Williams et al. (14) who, surprisingly, failed to show any advantage of a simple femoral or combined FSNB over IA injections after allograft ACL. This may be explained by the intrinsic limitations of studies based on retrospective data collections.
After an extensive review of the literature, we decided to combine in the two study groups, FSNB and IA infiltration, sets of treatments and medications that have been shown to be effective when administered independently from each other. (a) We added a sciatic to the femoral nerve block in the regional anesthesia group. (b) We combined local anesthetic and clonidine in the FSNB in an attempt to extend the duration of the sensory block (22–24) and improve the quality of the sensory block (25). It should be emphasized that the role of clonidine in peripheral nerve blocks is controversial, with data arguing against any additional benefit from combining clonidine to bupivacaine or ropivacaine (26,27). (c) We combined three different medications for the IA infiltration. The addition of clonidine (19) or morphine (5,28) to a local anesthetic improves the quality of analgesia, and one study showed a significant analgesic benefit from the IA administration of both morphine and clonidine (20). (d) Finally, the IA infiltration was performed before the surgical procedure (28,29), and the tourniquet was left inflated after the IA infiltration to allow binding of the medications to the local tissue (30).
We found a significant correlation among the analgesia technique, consumption of morphine, and the incidence of vomiting, clearly indicating that the FSNB provides better analgesia in children undergoing ACL reconstruction, which results in less consumption of narcotics and less frequent incidence of side effects. We can not draw any conclusion on the effects of the surgical technique (Achilles tendon allograft versus hamstrings autograft) on postoperative pain level because of the small number of patients in the hamstrings group.
The most common complaints after orthopedic procedures are pain along with nausea and vomiting, which are also the most common causes of costly unplanned admissions (31,32). Our report shows that a proper use of regional anesthesia can minimize these postoperative symptoms.
Potential criticisms of this study are the fact that it was not blinded, that there were more patients who received neostigmine in the IA compared to the FSNB group, and the potential for nerve injury when performing peripheral nerve blocks in patients under anesthesia. The study was not blinded because it is impossible to blind patients and evaluating personnel when comparing regional anesthesia versus other techniques that do not cause sensory or motor blocks because of the obvious effects of the nerve block. Moreover, the measured outcomes (consumption of morphine, episodes of vomiting, and presence of pruritus) are objective findings that could have not been influenced by the observer’s interpretation.
Neostigmine-induced increased motility of the gastrointestinal tract has been suggested to contribute to postoperative nausea and vomiting. However, the effect of reversal drugs on postoperative vomiting is unclear. We did not find any correlation between use of neostigmine and postoperative vomiting. Several studies have failed to show an increased incidence of vomiting after administration of neostigmine (33,34). Surprisingly, authors also found that smaller doses of neostigmine and atropine may prevent postoperative vomiting (35,36).
A large percentage of our patients in the FSNB group experienced a motor block in the postoperative period. This may have been related to the relative large concentration of bupivacaine we used or the addition of clonidine to the local anesthetic. The role of clonidine in inducing or extending the duration of a motor block is controversial. Studies have shown that adding clonidine to a local anesthetic may delay the recovery of the motor functions (37). Other authors have failed to show any effect of clonidine on the duration of motor blocks (38). With respect to the local anesthetic, ropivacaine has been shown to induce less motor block compared with bupivacaine (39,40), at least when administered in the epidural space and at diluted concentrations. In our experience, the presence of a prolonged motor block is a source of great anxiety for families and children, particularly in outpatient settings. However, a proper education for both the families and patients about this side effect may help overcome their concerns, and we should take into consideration the type of local anesthetic and the concentration used when performing peripheral nerve blocks in children to minimize the incidence of motor blocks.
The performance of regional anesthesia in children is not as widely accepted as in adults because these blocks must often be performed under general anesthesia. Peripheral nerve blocks are relatively safe when performed in awake patients (41). There are only few data in the literature on the safety of peripheral nerve blocks in children. A retrospective study (42), conducted in 1996 by the French-Language Society of Pediatric Anesthesiologists, did not report any significant complication from peripheral nerve blocks performed in children. Although based on small populations, more recent studies have confirmed that peripheral nerve blocks and placement catheters can be safely performed in anesthetized children (43,44). None of the patients in our study experienced complications related to the nerve block itself. Larger prospective studies are required to confirm the safety of this practice.
In summary, this prospective, randomized study demonstrated that a FSNB with bupivacaine-clonidine, as opposed to IA with bupivacaine-clonidine-morphine, decreased immediate postoperative pain and resulted in decreased opioid requirements and the incidence of side effects in children getting ACL reconstruction. Further studies are required to determine the effects of improved pain control on rehabilitation and patient satisfaction.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
