Anaesthesia, 1991, Volume 46, pages 275-277
Comparison of lumbar plexus block versus conventional opioid analgesia after total knee replacement
M. G. SERPELL, F A. MILLAR
AND
M. F. THOMSON
Summary A randomised controlled study was undertaken to assess the analgesic eficacy of continuous lumbar plexus block for the$rst 48 hours after total knee replacement surgery. Boluses of 0.5% bupivacaine with adrenaline 1 in 200000 (0.3 mlikg) were administered through a cannula inserted into the neurovascular sheath of the femoral nerve. Thirteen patients who received this block required significantly less morphine than a control group of 16 patients. Pain scores were similar and there were no complications related to this technique.
Key words Anaesthetic techniques, regional; lumbar plexus block. Anaesthetics, local; bupivacaine. Analgesia, postoperative.
Postoperative pain is difficult to treat effectively, as suggested by the many techniques available,’ but regional analgesia avoids many of the complications associated with conventional methods of treatment. Femoral nerve block provides good analgesia for open knee surgery,2 and our practice is to repeat it as required for pain relief, but this relies on the availability of an experienced person to perform the block. The ‘3 in 1 block13 is a simple technique to block the femoral, lateral cutaneous nerve of thigh and obturator nerves, and the placement of an indwelling catheter allows continuous block of the lumbar plexus; this has been used successfully for pain relief after knee The purpose of this study was to evaluate the analgesic effect of continuous lumbar plexus block after total knee replacement.
Methods We studied patients of ASA grade 1 or 2, aged between 18 and 85 years, undergoing elective total knee replacement under spinal anaesthesia, Patients were not studied if they had a history of neurological disease, hypersensitivity to bupivacaine or were unsuitable for spinal anaesthesia. The study was approved by the hospital Ethics Committee. A full explanation of the On-Demand Analgesic Computer (ODAC, Janssen Scientific Instruments, Beerse, Belgium) was given and informed consent obtained.
The anaesthetic technique was standardised and comprised temazepam premedication 0.3 mg/kg to the nearest 10 mg orally 1 hour before surgery. All patients underwent spinal anaesthesia with 0.75% bupivacaine 2.75-3.25 ml. Peri-operative oxygen therapy was administered and sedation provided with either intravenous midazolam in 1-2-mg increments or a propofol infusion as required. No supplementary analgesia was administered. Patients were allocated randomly into two groups. Group A had an 18-gauge cannula threaded over a 22gauge metal spinal needle which was placed accurately into the femoral nerve sheath using a nerve stimulator. The minimum current required to elicit a contraction in the quadriceps muscle was used. The block was performed immediately after surgery while the effects of the spinal were still present. The cannula was connected, after withdrawal of spinal needle, to a short extension tube and bacterial filter. The insertion site was then covered with a sterile occlusive dressing. A dose of 0.3 ml/kg 0.5% bupivacaine with adrenaline 1 : 200 000 was administered in 5-ml increments following aspiration, while observing cardiovascular parameters to further exclude intravenous injection. The cannula was left in situ to enable top-ups of the same dose of bupivacaine at 6-8 hour intervals during the next 48 hours. Group B did not receive this block nor did they have a placebo injection at the femoral nerve site. Both groups
M.G. Serpell, MB, ChB, FCAnaes, F.A. Millar, MB, ChB, Registrars, M.F. Thomson, MB, ChB, FFARCSI, Consultant, Department of Anaesthesia, Ninewells Hospital and Medical School, Dundee D D 1 9SY Accepted 2 July 1990. 0003-2409/91/040275 + 03 $03.00/0
@ 1991 The Association of Anaesthetists of Gt Britain and Ireland
275
276
M.G. Serpell, F.A. Millar and M.F. Thomson
were attached to the ODAC soon after returning to the ward and analgesic requirements were recorded by means of a printer. The ODAC was set to deliver morphine in 2-mg intravenous boluses to a maximum of 12 mg/hour with a lockout interval of 9 minutes. If pain relief was inadequate, intramuscular morphine 10 mg or paracetamol 1 g orally was administered. The patient’s assessment of postoperative pain was evaluated at 24 and 48 hours after surgery using a linear analogue on a scale from 0 to 100 (0, no pain, 100, worst pain imaginable). Any adverse effects were also noted at these times. Results were analysed statistically by Student’s t-test, Chi-squared test with Yates’ correction or two-tailed Wilcoxon test as appropriate.
Table 2. Postoperative analgesic requirements, mean (SEM). * p i 0.05 (Wilcoxin two-tailed test). ~
~
Total 48 hour consumption of morphine; mg Number of refused ODAC requests in 48 hours Total number of intramuscular supplements; 10 mg Range of frequency of administration of morphine supplements; mean Total number of paracetamol supplements; 1 g Range of frequency of administration of paracetamol supplements; mean
60 (8.7)*
~~~
Group B (n = 16)
Group A (n = 13)
~~~
91 (9.8)
6.2 (1.4)
12.9 (2.6)
10
17
0-3 (0.8)
0-5 (1.1)
13
10
0 4 (I)
0-2 (0.6)
Results
Thirty patients entered the trial, but one in Group A was withdrawn because of postoperative urinary retention which required bouginage under sedation and local anaesthetic to dilate a urethral stricture. Patient data, duration of surgery and peri-operative sedation requirements are summarised in Table 1 . Two in each group received a propofol infusion for sedation at 2-3 mg/kg/hour. Patients with the lumbar plexus block required a third less morphine than the control group in the 48 hour postoperative period (p < 0.05). The ODAC system provided most of the analgesia because the doses of intramuscular morphine and oral paracetamol were low in both groups (Table 2). Postoperative pain scores at 24 and 48 hours were similar, even though Group B had twice the number of refused ODAC requests compared to Group A. The incidence of nausea and vomiting were comparable, and although there was less antiemetic administered to the study group, this did not reach significance (Table 3). Discussion We have demonstrated a significant reduction in postoperative analgesic requirements by using a continuous femoral nerve block (p < 0.05). Use of the ODAC system provides a good indication of the magnitude of pain since the demand rate reflects a balance between attaining analgesia and avoiding side effects. The similar mean pain scores and incidence of nausea and vomiting in both groups supports this. We found, despite training the patients to use the ODAC as soon as pain was appreciated, that when the spinal analgesia wore off, it did so very rapidly and the ODAC could not deliver a large enough
loading dose of morphine. Patients would often request an intramuscular supplement, and were then able to control pain by a maintenance demand rate. The knee joint is innervated by components of the femoral, obturator and sciatic nerves. Only the first two are involved in the lumbar plexus block so complete analgesia is unlikely. Quadriceps muscle spasm aggravates pain in the knee joint (femoral nerve) and abolition of this muscle spasm by femoral nerve blockade may well be the main benefit of this analgesic technique. Other effective regional techniques used for postoperative analgesia include epidural opioids and (or) local anaesthetics. The complications of urinary retention, pruritus and respiratory depression from epidural opioids requires close observation of the patient for at least 24 hours. Epidural local anaesthetics provide the most effective analgesia but the bilateral motor blockade makes the patient more nurse dependent and the sympathetic blockade necessitates cardiovascular monitoring. Several studies have shown the benefits of the ‘3 in I’ block in patella, knee ligament surgery and arthroscopy,’6 but Dahl stated there was no advantage when used in knee arthr~plasty.~ There were only six patients in each of his groups and the block was performed after operation. There is evidence*that abolishing sensory input before the occurrence of painful stimuli can reduce postoperative analgesic requirements and we performed the block before spinal anaesthesia wore off. Our study was not double-blind, because we thought it was unethical to submit a patient to a placebo which involved an invasive procedure. Secondly, any trial that Table 3. Pain scores (0 = no pain, 100 = worst pain imaginable), incidence of side effects and antiemetic administration, mean (SEM). Chi-squared and Wilcoxon test.
Table 1. Demographic data, mean (SEM). No significant differences (p < 0.05) were found (Student’s r-test).
Maleslfemales Age; years Height; cm Weight; kg Operation time; minutes Midazolam; mg (n = 11 and 14 respectively)
Group A (n = 13)
Group B (n = 16)
211 1 68 (2.7) 160 (1.5) 70 (2.3) 105 (5.1) 4.0 (0.9)
4/12 70 (2.5) 165 (3.0) 72 (3.8) 105 (4.8) 2.9 (0.5)
Pain score; 24 hours Pain score; 48 hours Nausea; number of patients Vomiting; number of patients Total number of doses of prochlorperazine; 12.5 mg Range of frequency of prochlorperazine administrations; mean
Group A (n = 13)
Group B (n = 16)
32 (8.8) 34 (7.6) 9 7 11
40 (5.5) 37 (6.5) 12 8 24
0-2 (0.8)
0-3 ( I .5)
Lumbar plexus block versus opioid analgesia involved local anaesthetic drugs will produce paraesthesia and weakness which is obvious to both doctor and patient. It was difficult to assess for paraesthesia over the distribution of the affected nerves because of extensive dressings around the leg. Previous studies have demonstrated that there is usually evidence of femoral nerve anaesthesia but this is variable for lateral cutaneous nerve of thigh and rare for the obturator n e r ~ e . ~ . ~ The 6-8 hour top-up regimen was devised to be as workable as possible so that the resident doctors would not be called to administer a dose during 'unsociable' hours (2400-0800). This problem could be solved by setting up an infusion or allowing suitably trained nurses to give the top-ups. We did not measure plasma concentration of bupivacaine, but group A received an average of 5.8 top-ups (range 3-8) or 0.18 mg/kg/hour (SEM 0.01) over the 48hour period. This is lower than the amount given in another study5 which demonstrated safe subtoxic concentrations, so there is scope for giving more frequent doses if required. There were no complications of prolonged anaesthesia or infection at the cannula site, although damage to the femoral nerve has been reported in isolated cases.'0.'' We used a nerve stimulator for accurate placement of the cannula, and it is suggested that the avoidance of eliciting paraesthesia during nerve blocks reduces the incidence of damage.12 In conclusion, we have demonstrated that continuous lumbar plexus blockade is a safe and simple technique to perform and provides similar analgesia to parenteral morphine, but with significantly reduced opioid requirements.
277
References I . DODSONME. A review of methods for relief of postoperative pain. Annals ofthe Royal College of Surgeons of England 1982; 64: 324-7. 2. TIERNEY E, LEWISG, HURTIGJB. Femoral nerve block with bupivacaine 0.25 per cent for postoperative analgesia after open knee surgery. Canadian Journal of Anaesthesia 1987; 3 4 455-8. 3. WINNIEAP, RAMAMURTHY S, DURANNIZ. The inguinal paravascular technic of lumbar plexus anesthesia: The '3-in-I ' block'. Anesthesia and Analgesia 1973; 52: 989-96. 4. ROSENBLATT RM. Continuous femoral anesthesia for lower extremity surgery. Anesthesia and Analgesia 1980; 5 9 63 1-2. 5. DAHLJB, CHRISTIANSEN CL, DAUGAARD JJ, SCHULTZP, CARLSONP. Continuous blockade of the lumbar plexus after knee surgery-postoperative analgesia and bupivacaine plasma concentrations. A controlled clinical trial. Anaestkesia 1988; 4 3 1015-8. MH, PASKINS, GROSSMAN R. A 6. PATALNJ, FLASHBURG regional anesthetic technique compared to general anesthesia for outpatient knee arthroscopy. Anesthesia and Analgesia 1986; 65: 185-7. 7. DAHLJB, ANKER-MOLLER E, SPANSBERG N, SCHULTZP, CHRISTENSEN EF. Continuous lumbar plexus block after arthroplasty. Anaesthesia 1988; 43: 989. 8. RINGROSE NH, CROSSMJ. Femoral nerve block in knee joint surgery. American Journal of Sports Medicine 1984; 1 2 398-402. 9. MADEJTH, ELLISFR, HALSALLPJ. Evaluation of '3 in I' lumbar plexus block in patients having muscle biopsy. British Journal of Anaesthesia 1989; 6 2 515-7. 10. JOHRM. Spaete komplikation der kontinuierlichen blockade des n. femoralis. Regional Anaesthesia 1987; 10: 37-8. 11. NIELSENBF. Iatrogenic femoral nerve neuropathy after lumbar plexus block. Ugeskrift for Laeger 1988; 150: 748-9. 12. SELANDER D, EDSHAGES, WOLFF T. Paresthesiae or no paresthesiae. Nerve lesions after axillary blocks. Acta Anaesthesiologica Scandinavica 1919; 23: 27-33.