Total knee arthroplasty (TKA) results in significant postoperative pain. Even with patient-controlled IV opioids (1,2) and nonsteroidal antiinflammatory drugs (NSAIDs) (3), moderate-to-severe pain can persist, especially during mobilization. Continuous epidural analgesia provides excellent pain control and improves rehabilitation in this setting (4). However, the adoption of low-molecular weight heparin thromboprophylaxis protocols has raised concerns regarding spinal hematoma formation with the use of continuous lumbar epidural analgesia, and led to examination of peripheral nerve block techniques as an alternative (5). Among individual blocks for this indication, femoral nerve block has been widely studied and found to significantly improve pain control in the first 24 h after surgery compared with systemic analgesia alone (6–9). Some have reported that the analgesic benefits may extend to 48 h (8,9), and that functional outcomes such as knee flexion and length of hospital stay may be improved (9). Previous studies have also evaluated adding single-injection obturator block to femoral (10) or femoral-sciatic (11) blockade. We sought to compare the analgesic efficacy and functional outcome effects of obturator and femoral nerve blocks, using equal local anesthetic doses, to each other and placebo after TKA in a randomized, prospective, double-blind manner.
With IRB approval and informed written consent, 60 patients undergoing elective, unilateral TKA under spinal anesthesia were enlisted in this randomized, placebo-controlled, double-blind, parallel-group study. Patients were enlisted on the day of surgery and were randomly assigned to receive one of three study interventions: obturator nerve block, femoral nerve block, or placebo (sham block).
Exclusion criteria were age <18 yr, inability to communicate lucidly in English or French, revision knee replacement, morbid obesity (body mass index >35), contraindication to spinal anesthesia (back fusion, coagulopathy, local infection), renal failure, or hypersensitivity to bupivacaine, fentanyl, or NSAIDs.
Randomization and Blinding Methods
Eligible patients were randomly assigned to treatment groups in blocks of different sizes (4, 6, and 8), according to a list preprepared by the study epidemiologist (AV). Study treatment was given upon completion of the surgery under spinal anesthesia. The anesthesiologist prepared and administered the nerve blocks or sham injection. Both patients and nurse observers collecting data were blinded to the intervention group.
Intraoperative sedation with an IV propofol infusion was administered at the discretion of the anesthesiologist, and the total dose given throughout the case recorded. Spinal anesthesia was induced in the sitting position at the L2–5 level using a 25-gauge Whitacre spinal needle (BD Medical, Franklin Lakes, USA) with a dose of 12 mg plain 0.5% bupivacaine (Hospira, Montreal, Canada), and patients were immediately positioned supine. No opioids were given intraoperatively. At the end of surgery, after the surgical dressing and the patient's view was still blocked by the surgical field drape barrier, the study nerve blocks were performed using a 5-cm, 22-gauge Stimuplex insulated needle (B. Braun Medical, Bethlehem, USA) connected to a nerve stimulator set at an output of 2 mA, 2 Hz with a 100 μs square-wave pulse. Blocks were only performed if spinal anesthesia to needle prick persisted at the site. No extra sedation was given at the time of the blocks, so that all patients were responsive. Patients requiring supplemental analgesia for incomplete or insufficiently long spinal anesthesia were excluded from the study.
Femoral nerve blocks were performed using the classic paravascular technique described by Winnie et al. (12), using contraction of the quadriceps muscle as an end-point. For obturator blocks, the needle was inserted 2 cm caudal and lateral to the pubic tubercle, “walking” inferiorly off the superior pubic ramus if contacted, until the obturator canal was entered and contraction of the thigh adductor muscles was obtained (10).
The end-point for nerve localization was contraction of target musculature with 0.5 mA stimulating current. Upon localization, the current was reduced to the lowest level that still produced visible contractions of the target muscle and this value was recorded separately, unseen by postoperative data collectors. Nerve blocks were performed with 20 mL bupivacaine 0.5% with 1/200,000 epinephrine. Patients in the placebo group did not receive any injection, but the patient's inguinal area was prepared, and a sham block performed behind a surgical drape blocking the patient's view. For all groups, a bandage was placed over the inguinal area, to maintain double-blinding in the recovery period.
Sensory and motor testing of the operative leg was performed after the spinal block had regressed to the point where the patient was able to raise and extend the knee, with normal sensation to the thigh, on the nonoperative side. Sensation to light touch at mid-thigh on the anterior and medial aspects was assessed. Motor block was assessed by the ability of the patient to contract the quadriceps muscle and adduct the operative leg from a 30° abducted position to the midline. These examinations were performed on all patients by the anesthesiologist, and the results recorded separately, to maintain blinding.
Postoperative Pain Management
Postoperatively, patients were provided with an IV patient-controlled analgesia (PCA) system (Graseby) with fentanyl 50 μg/mL set to deliver 25 μg every 5 min as needed. Time to first dose of fentanyl from arrival in the recovery room was noted. Cumulative doses of PCA fentanyl were recorded every 4 h for 48 h after surgery. Celecoxib 100 mg and acetaminophen 650 mg PO were given on arrival in the recovery room, then round-the-clock every 12 and 6 h, respectively. If pain at rest was rated >6/10 after receiving all of the above, breakthrough medication with ketorolac 10 mg IM was given every 4 h PRN. Ketorolac was chosen to spare escalating opioid use and because its parenteral administration was an advantage for patients experiencing opioid-related nausea. Time to first dose of ketorolac for breakthrough pain and total doses required over 48 h were recorded.
Assessment and Data Collection
Knee pain was assessed at rest and with movement (30° knee flexion) using an 11-point numeric rating scale (NRS: 0 = no pain, 10 = worst pain imaginable). Baseline preoperative values were recorded by the anesthesiologist for each case. Pain scores with movement were recorded at recovery room discharge and 24 and 48 h postoperatively by the study nurse (SP), who was blind to group assignment.
Side effects recorded by the study nurse daily during the 48 h study period were: level of sedation on a five-point NRS (1 = fully alert, 2 = drowsy when undisturbed, 3 = consistently drowsy, 4 = rousable only with stimulation, 5 = unarousable); nausea or pruritus requiring treatment; urinary retention requiring catheterization (bladder ultrasound showing more than 500 mL volume); or any numbness or weakness in the operative leg.
Starting the day after surgery, intermittent manual physiotherapy was performed daily. Surgical drains were not used. Maximum knee flexion was measured with a goniometer by the physiotherapist and recorded on the second postoperative day (when the compressive dressing was first removed), and at discharge from hospital. Hospital discharge time was at the discretion of the surgeon, who was blinded to study group assignment. Total duration of hospitalization in days was noted.
Sample size was calculated to detect a difference between mean group rest pain scores of 2.5 on a 0–10 scale with a sd of 2.6. These assumptions were based on the results of a previous study of single-shot femoral nerve block (9) of similar design. For an α of 0.05 and 80% power, these assumptions resulted in a projected sample size of 20 per group.
Baseline data were compared among groups using one-way ANOVA for continuous measures and Pearson's χ2 or Fisher's exact test for categorical measures. All outcome variables were compared among groups using Proc GLM general linear model (ANOVA). Primary outcome was defined as effect on first rest and dynamic pain scores after recovery from spinal anesthesia. Analysis of primary outcome was on an intention-to-treat basis to compare among groups, adjusting for preoperative baseline pain. Bonferonni correction was made for repeated secondary outcome observations such as rest pain scores and fentanyl consumption. Multivariate analysis was used to evaluate effects of baseline variable differences among groups. All analyses were performed using SAS software version 9.1.
Eighty-seven patients were approached to participate. Of the 60 who were eligible, all consented to participate. One patient randomized to the femoral group received a successful block but was excluded from further study because of postoperative confusion in the recovery room. Among the 59 patients completing the study, there were no significant differences among groups in demographic variables, surgical time, or propofol dosage (Table 1).
All femoral blocks resulted in quadriceps motor block, but only 75% had sensory block to the anterior thigh. Eighty-four percent of all femoral blocks, and 93% of those with sensory block resulted in sensory block to the medial thigh. Only one patient (5%) in the femoral block group developed adductor motor block. All obturator blocks produced adductor blockade, but only 20% had any medial-thigh sensory block. One (5%) obturator block produced quadriceps motor block, but without anterior thigh anesthesia.
There was no significant difference between the lowest current producing nerve stimulation among groups (0.32 ± 0.02 and 0.34 ± 0.02 mA, femoral and obturator, respectively).
Pain at rest was not significantly different among groups at any time period (Fig. 1). When pain scores were analyzed by intention-to-treat according to the difference from preoperative baseline, there was a significant decrease in the femoral group compared with placebo at recovery room discharge (NRS difference from baseline 2.0 ± 0.4 vs 3.4 ± 0.4, respectively; between group difference = −2.0, 95%CI: −3.7 to −0.4, P = 0.02). The difference between obturator and placebo groups remained insignificant.
Pain with movement was not significantly different among groups at baseline (Fig. 2). At discharge from the recovery room, movement pain scores were significantly lower after femoral nerve block compared with obturator (P = 0.03). When these results were analyzed as difference from baseline, a borderline significant difference between femoral and placebo groups remained (NRS difference from baseline 2.6 ± 0.6, 4.2 ± 0.6; between group difference = −1.6, 95% CI: −3.3 to 0.02, P = 0.05), and there was no difference between obturator and placebo groups. There was no significant difference in dynamic pain, or change from baseline, among groups at 24 or 48 h.
Cumulative fentanyl PCA use was not significantly different among groups at any time period (Fig. 3). When PCA use was converted to 4-h incremental dosage and the analysis was repeated, there was again no difference. Time to first dose of PCA (127 ± 19, 99 ± 11, 112 ± 14 min; femoral, obturator, placebo), time to first breakthrough analgesic (1055 ± 243, 677 ± 243, 1073 ± 199 min), and total doses of breakthrough-pain analgesic (1.2 ± 0.2, 0.55 ± 0.2, 0.95 ± 0.2) showed no significant difference among groups.
There were no differences in functional outcome variables: time in recovery room, maximum knee flexion on day 2 and at discharge, and total days in hospital (Table 2).
One patient who received a femoral nerve block reported persistent numbness over the anterior thigh persisting at 48 h. This resolved spontaneously by the fifth postoperative day. No other sequelae related to nerve blocks were noted. There was no significant difference among groups in the incidence of side effects at any time (Table 3).
The five patients with femoral block who did not develop anterior thigh sensory anesthesia were included in all analyses on an intention-to-treat basis.
This double-blind, placebo-controlled trial found no clinical advantage to performing individual obturator block after TKA under spinal anesthesia. Two previous studies had reported improved outcomes when obturator blockade was added to femoral (10) or femoral-sciatic (11) block after TKA. Both used general anesthesia for surgery. Macalou et al. (10), in a single-blinded study, reported improvements in rest pain scores and PCA morphine consumption for 6 h after obturator block with 7 mL of a mixture of bupivacaine 0.5% and lidocaine 2% with epinephrine was performed in addition to femoral nerve block. Their results with femoral nerve block alone were no different from placebo, suggesting obturator nerve block alone may be more effective than femoral blockade in treating TKA pain. McNamee et al. (11) studied the effects of adding an obturator block with 5 mL 0.75% ropivacaine to femoral-sciatic blocks. There was no placebo control group. They reported no change in pain scores with movement, but there was a decrease in 48-h PCA morphine consumption and delay in its first use. However, as in our study, there was no difference in opioid use over the first 20 h. Also, as in our experience, there was no significant effect on opioid-related side effects.
In both these positive studies, obturator block was combined with other blocks. It is possible that there is some synergistic effect between obturator and femoral block. Our study did not include a treatment group that received both.
Femoral nerve block decreased pain in the recovery room, but did not provide any improvement in outcome at 24 h and beyond. This is consistent with the findings of others (6,7). In two placebo-controlled trials that found an effect of single-shot femoral block on postoperative pain beyond 24 h after TKA (8,9), surgery was under general anesthesia, and no postoperative NSAIDs were used. It is possible that our results represent a false negative, or Type II error, but our group size of 20 was more than that of either of the positive studies. An explanation may be our use of bupivacaine spinal anesthesia and commencing regular multimodal analgesia before the onset of pain, which may have provided sufficient analgesia to allow little incremental benefit from nerve blocks.
A recent study by Salinas et al. (13), performed in the context of multimodal analgesia, showed that maintaining analgesic effects from femoral nerve blockade beyond 24 h required a continuous infusion. As in our study, they found no improvement in functional outcome, suggesting that established clinical pathways of mobilization and hospital discharge may determine these outcomes as much as analgesia.
There are persistent references in the literature to femoral nerve blocks supposedly producing obturator and lateral femoral cutaneous nerve anesthesia by “3-in-1” paravascular spread (12). Other studies have shown that the lateral femoral cutaneous is frequently blocked, whereas the obturator nerve rarely is (14–16). Bouaziz et al. (14) reported that obturator nerve block produces hypoesthesia of the medial thigh in only 20% of patients, identical to our results, making this a poor test of obturator blockade. Adductor motor block has been shown to be the only reliable test of obturator nerve blockade (14–16). We did not quantify adductor muscle strength, but patients' ability to adduct the leg while extended in a heavy surgical dressing provided unequivocal results. Using this as an end-point, we found that blockade of the femoral nerve rarely (5%) produced obturator nerve block and vice versa. Since we did not image the spread of local anesthetic, we cannot make any conclusion about the mechanism of this occasional occurrence. Our study methodology also could not distinguish between blockade of the main trunk of the obturator nerve and its anterior and posterior branches, although both lie together in the obturator canal where our blocks were performed. The posterior branch contributes sensation to the knee joint.
All our femoral blocks resulted in quadriceps motor block, but only 75% had sensory block to the anterior thigh, served by the anterior division of the femoral nerve. This finding reflects a limitation of relying on motor stimulation alone to identify the main trunk of the femoral nerve. The posterior branch, which supplies the quadriceps, contributes to sensation of the knee joint. The lateral femoral cutaneous nerve overlaps innervation of the anterior thigh in up to 40% of patients, and it has been suggested that quadriceps motor block be used as the definitive test of femoral nerve blockade (17).
The sciatic nerve provides sensation to the posterior aspect of the knee, which can be a significant source of pain in certain TKA patients (18,19). We did not investigate this, but presumably this confounding variable was equally distributed among the study groups, which may be a factor in the minimal benefit observed with femoral block alone.
We performed our blocks under spinal anesthesia, which spares patients feeling the injection, but eliminates their ability to report a needle-induced paresthesia. Using our nerve stimulator settings, contact with the nerve is still possible (20). It is unclear whether the transient thigh numbness one patient experienced 48 h after femoral block was related to the block or the thigh tourniquet. There was no pain or persistent motor weakness attributable to any of the blocks.
In conclusion, femoral nerve block rarely blocks the obturator nerve. We were unable to demonstrate any clinically significant benefits of obturator nerve block alone after TKA. Individual, single-injection femoral nerve block with 20 mL bupivacaine 0.5% and epinephrine provides no clinical benefit beyond the day of surgery for TKA patients in the context of regular multimodal analgesia after spinal anesthesia.
The authors gratefully acknowledge the cooperation of Drs. John Antoniou, Olga Huk, and Len Rosen of the Department of Orthopedics, SMBD-Jewish General Hospital.
1. Chelly JE, Greger J, Gebhard R, Coupe K, Clyburn TA, Buckle R, Criswell A. Continuous femoral blocks improve recovery and outcome of patients undergoing total knee arthroplasty. J Arthroplasty 2001;16:436–45
2. Singelyn FJ, Deyaert M, Joris D, Pendeville E, Gouverneur JM. Effects of IV patient-controlled analgesia with morphine, continuous epidural analgesia, and continuous three-in-one block on postoperative pain and rehabilitation after unilateral total knee arthroplasty. Anesth Analg 1998;87:88–92
3. Capdevila X, Barthelet Y, Biboulet P, Ryckwaert Y, Rubenovitch J, d'Athis F. Effects of perioperative analgesic technique on the surgical outcome and duration of rehabilitation after major knee surgery. Anesthesiology 1999;91:8–15
4. Williams-Russo P, Sharrock NE, Haas SB, Insall J, Windsor RE, Laskin RS, Ranawat CS, Go G, Ganz SB. Randomized trial of epidural versus general anesthesia: outcomes after primary total knee replacement. Clin Orthop 1996;331:199–208
5. Enneking FK, Chan V, Greger J, Hadzi A, Lang SA, Horlocker TT. Lower-extremity peripheral nerve blockade: essentials of our current understanding. Reg Anesth Pain Med 2005;30:4–35
6. Hirst GC, Lang SA, Dust WN, Cassidy JD, Yip RW. Femoral nerve block: single injection versus continuous infusion for total knee arthroplasty. Reg Anesth 1996;21:292–7
7. Allen HW, Liu SS, Ware PD, Nairn CS, Owens BD. Peripheral nerve blocks improve analgesia after total knee replacement surgery. Anesth Analg 1998;87:93–7
8. Ng HP, Cheong KF, Lim A, Lim J, Puhaindran ME. Intraoperative single-shot “3-in-1” femoral nerve block with ropivacaine 0.25%, ropivacaine 0.5% or bupivacaine 0.25% provides comparable 48-hr analgesia after unilateral total knee replacement. Can J Anaesth 2001;48:1102–8
9. Wang H, Boctor B, Verner J. The effect of single-injection femoral nerve block on rehabilitation and length of hospital stay after total knee replacement. Reg Anesth Pain Med 2002; 27:139–44
10. Macalou D, Trueck S, Meuret P, Heck M, Vial F, Ouologuem S, Capdevila X, Virion JM, Bouaziz H. Postoperative analgesia after total knee replacement: the effect of an obturator nerve block added to the femoral 3-in-1 nerve block. Anesth Analg 2004;99:251–4
11. McNamee DA, Parks L, Milligan KR. Postoperative analgesia following total knee replacement: an evaluation of the addition of an obturator nerve block to combined femoral and sciatic nerve block. Acta Anaesthesiol Scand 2002;46:95–9
12. Winnie AP, Ramamurthy S, Durrani Z. The inguinal paravascular technic of lumbar plexus anesthesia: the “3-in-1 block.” Anesth Analg 1973;52:989–96
13. Salinas FV, Liu SS, Mulroy MF. The effect of single-injection femoral nerve block versus continuous femoral nerve block after total knee arthroplasty on hospital length of stay and long-term functional recovery within an established clinical pathway. Anesth Analg 2006;102:1234–9
14. Bouaziz H, Vial F, Jochum D, Macalou D, Heck M, Meuret P, Braun M, Laxenaire MC. An evaluation of the cutaneous distribution after obturator nerve block. Anesth Analg 2002; 94:445–9
15. Atanassoff PG, Weiss BM, Brull SJ, Horst A, Kulling D, Stein R, Theiller I. Electromyographic comparison of obturator nerve block to three-in-one block. Anesth Analg 1995;81:529–33
16. Lang SA, Yip RW, Chang PC, Gerard MA. The femoral 3-in-1 block revisited. J Clin Anesth 1993;5:292–6
17. Jochum D, O'Neill T, Jabbour H, Diarra PD, Cuignet-Pourel E, Bouaziz H. Evaluation of femoral nerve blockade following inguinal paravascular block of Winnie: are there still lessons to be learnt? Anaesthesia 2005;60:974–7
18. Ben-David B, Schmalenberger K, Chelly J. Analgesia after total knee arthroplasty: is continuous sciatic blockade needed in addition to continuous femoral blockade? Anesth Analg 2004;98:747–9
19. Levesque S, Delbos A. Sciatic nerve block for total-knee replacement: is it really necessary in all patients? Reg Anesth Pain Med 2005;30:410–1
20. Choyce A, Chan VWS, Middleton WJ, Knight PR, Peng P, McCartney CJL. What is the relationship between paresthesia and nerve stimulation for axillary brachial plexus block? Reg Anesth Pain Med 2001;26:97–9