Continuous femoral nerve block during major knee surgery provides better analgesia both at rest and during mobilization, it has fewer opioid-related side effects and it improves functional recovery by comparison with intravenous (i.v.) opioid administration.1–4 Continuous femoral nerve block is usually performed with neurostimulation. In previous studies, ultrasound guidance enabled lower levels of local anaesthetic solution to be used, increased nerve block duration and reduced onset time after the first local anaesthetic bolus injection.5–8 The femoral nerve is usually visualized by obtaining a short-axis view at the level of the inguinal crease with the use of a high-frequency linear array probe and using an out-of-plane needle approach. Most of the studies5–9 concerning ultrasound-guided femoral nerve block for knee surgery have evaluated single-shot injection. Ultrasound-guided catheter placement for femoral nerve block has not been prospectively compared with conventional neurostimulation. The purpose of this prospective randomized clinical study was to compare the effectiveness of perineural femoral catheter placement using either conventional neurostimulation or neurostimulation combined with ultrasound guidance in patients undergoing total knee arthroplasty (TKA).
Patients and study design
After approval from the local ethics committee (Comité de Protection des Personnes soumises à une Recherche Biologique du Centre Hospitalo-Universitaire de Rennes) and with their written informed consent, we enrolled 93 patients with an American Society of Anesthesiologists physical status of I–III who were scheduled for elective primary unilateral TKA. This prospective study was conducted over an 11-month period (from November 2007 to September 2008). All interventions were performed by the same two trained surgeons who used a tri-compartment cemented TKA via a 12–15 cm midline incision. Exclusion criteria were contraindications to the peripheral nerve block or to nonsteroidal anti-inflammatory drugs, a preexisting neurological, cardiac or hepatic disorder, renal failure or opioid treatment. The day prior to surgery patients were randomly assigned to one of two groups using a computer-generated list of random numbers. An opaque, sealed envelope was prepared for each patient. On the same day, patients were also instructed on the use of a 100 mm visual analogue scale (VAS) graded from zero (no pain) to 100 (most severe pain). Patients were premedicated 1 h before surgery with 1 mg orally administered alprazolam. Before the nerve block, i.v. access was established, with continuous ECG, noninvasive blood pressure and oxygen saturation being continuously monitored during the procedure. All patients were in the supine position. A wide area of skin was disinfected with a 0.5% chlorhexidine digluconate–alcohol or 5% alcoholic povidone–iodine solution (according to the surgeon's protocol); the inguinal crease was protected with sterile draping. Twenty millilitres of 5 mg ml−1 levobupivacaine (Chirocaine; Abbott, Rungis Cedex, France) was administered through a 20-gauge catheter (Plexolong; Pajunk, Geisingen, Germany) inserted with a 100 mm 18-gauge short-bevelled insulated stimulating needle. Needle placement was performed for one group using neurostimulation and ultrasound guidance for the other.
All nerve blocks in the neurostimulation group were performed by two experienced anaesthetists (H.L.H. and P.V.). The puncture site was located 1 cm laterally from the femoral artery below the inguinal ligament. After skin infiltration with 2 ml of 20 mg ml−1 lidocaine, the needle was connected to a nerve stimulator (Multistim; Pajunk, Geisingen, Germany) with initial current settings of 2 mA, pulse width 0.1 ms and frequency 1 Hz. A 100 mm 18-gauge short-bevelled insulated stimulating needle was inserted 2 cm below the inguinal ligament, at 45° to the skin plane and advanced in a cephalic direction until eliciting a quadriceps femoris contraction indicated by upward movement of the patella. The stimulating current was gradually decreased to less than 0.5 mA while patella movement response was maintained. Five millilitres of 5% dextrose solution was injected through the needle and a 20-gauge catheter was inserted to a depth of 3 cm. The catheter was secured in place using Steri-Strips (3M, France) and a transparent Opsite (Smith-Nephew, France) dressing and then connected to the insertion port with an antimicrobial filter. After a negative aspiration test, 20 ml of 5 mg ml−1 levobupivacaine was slowly injected through the catheter with intermittent aspiration.
In the ultrasound group, nerve blocks were performed by anaesthetists (C.A., A.L. and F.C.) with more than 2 years of experience in ultrasound-guided regional anaesthesia. The femoral nerve was scanned using a linear array transducer scanning at between 6 and 13 MHz, connected to a portable ultrasound unit (SonoSite, Inc., Bothell, Washington, USA). The probe in a sterile cover (Intercover; Microtek Medical, Columbus, Mississippi, USA) was inserted, a sterile gel (Asept; Asept Inmed, France) being applied on the skin to create a sterile interface with the probe head. The ultrasound probe was initially positioned at the inguinal crease (see Photo 1, Supplemental Digital Content 1, http://links.lww.com/EJA/A11, where the in-plane needle insertion in the ultrasound group is described) and moved to view short-axis femoral vessels (transverse cross-section). The transverse sonogram showed the femoral nerve that appeared as a hyperechoic structure ventrally to the iliopsoas hypoechoic muscle structure and laterally to the hypoechoic femoral artery pulsating signal. After skin infiltration of 20 mg ml−1 lidocaine, the 100 mm 18-gauge short-bevelled insulated stimulating needle was advanced using an in-plane approach with real-time ultrasound imaging guidance until the tip was close to the targeted nerve. Because of the medial direction of the needle, the needle tip was frequently close to the anterolateral femoral nerve but needle-to-nerve contact was not sought and intraneural injection avoided. The probe was tilted, rotated or both to improve the femoral nerve signal. The neurostimulator (pulse duration 0.1 ms and frequency 1 Hz) was activated and current intensity progressively increased until either a motor response was elicited (patella movement) or a maximum of 2.0 mA was attained. The minimum stimulating current level was noted, and 5 ml of 5% dextrose solution was injected to improve nerve structure. Solution dispersal was observed in real time in a hypoechoic area between the femoral nerve and the fascia iliaca. The catheter was introduced 2–3 cm beyond the needle tip and fixed securely after withdrawal of the needle. The ultrasound probe was repositioned on the inguinal crease to seek the tip of the catheter, which was carefully withdrawn. After an aspiration test, 20 ml of 5 mg ml−1 levobupivacaine was injected. Dispersal of the local anaesthetic was observed in real time by ultrasound. The catheter was viewed using ultrasound in the local anaesthetic solution area and mobilized to obtain the correct placement of the distal extremity under the fascia iliaca with local anaesthetic surrounding the femoral nerve (see Photo 2, Supplemental Digital Content 2, http://links.lww.com/EJA/A12, in which the transverse sonogram of the inguinal crease shows the lumen of the catheter posterior to the fascia iliaca and above the femoral nerve). In neither of the two groups was local anaesthetic injected by needle.
After the procedure, patients in both groups were placed in the Sims position with the operated leg uppermost; an ultrasound-guided subgluteal sciatic nerve block was performed using an in-plane approach. Briefly, according to the patient's height–weight ratio and the predicted depth of the nerve, a high-frequency 6–13 MHz linear HFL38 or a low-frequency 2–5 MHz curve transducer array was positioned in the subgluteal area and the sciatic nerve viewed in the transverse plane ventrally to the gluteus maximus and dorsally to the quadratus femoris. A 22-gauge 100 mm short-bevelled, insulated stimulating needle connected to a nerve stimulator (1 Hz and 0.1 ms) was inserted in the longer axis of the ultrasound probe. Fifteen millilitres of 3.75 mg ml−1 levobupivacaine was slowly injected with the tip of the needle positioned close to the nerve, and confirmed by either plantar or dorsal foot flexion with a stimulation current of less than 0.5 mA.
On completion of the regional nerve blocks and after checking for sensitivity and motor deficiency in the vicinity of the femoral nerve, general anaesthesia was effected using propofol (1–2.5 mg kg−1), sufentanil (0.3 μg kg−1) and a single bolus of cisatracurium (0.1 mg kg−1). Anaesthesia was maintained with 0.8–1% sevoflurane with 50% nitrogen in oxygen. Before skin incision, the patients received i.v. paracetamol (1 g), ketoprofen (100 mg), nefopam (20 mg) and droperidol (0.625 mg). All catheters were withdrawn on postoperative day (POD) 2. Enoxaparin 4000 IU was administered once a day for 2 weeks for thromboprophylaxis prevention.
Postoperative pain management
In the post anaesthesia care unit (PACU), catheters were connected to an ambulatory, programmable, electronic infusion pump (ambIT; Sorenson Medical, Inc., Gamida, France) that delivered a continuous infusion of 1.25 mg ml−1 levobupivacaine (infusion rate 5 ml h−1). Patients were able to self-administer additional 5 ml boluses with a lockout time of 60 min. Catheters were withdrawn 48 h after surgery and the accumulated local anaesthetic dose was noted (excluding the first bolus injected). Patients also received two tablets of paracetamol 400 mg with dextropropoxyphene 30 mg every 6 h and ketoprofen 150 mg twice a day. Up to four tablets a day of 20 mg of immediate release morphine sulphate (Actiskenan; Mayers-Squibb, France) were given to control breakthrough pain. Attending surgeons, who were unaware of group allocation, decided on the patient's discharge based on routine surgical considerations, including knee flexion above 70°, extension at 0° and ability to climb up 5–10 stairs.
Outcome and measures
All parameters were recorded by an anaesthetist and a nurse unaware of patient allocation. PACU nurses were also unaware of the analgesic catheter technique used. Distribution of the sensory block was assessed by testing loss of cold sensation every 5 min over a 30 min period following levobupivacaine administration. Motor block was assessed by ability to extend the leg with the hip passively flexed at 45°. Sensory testing was assessed by loss of ice bag sensitivity to the medial and anterior thigh above the surgical site and complete loss of cold sensation was defined as a sensitivity block. Sciatic motor nerve block was examined by dorsal and plantar flexions and sensory block with loss of cold sensitivity in the dorsal and plantar foot areas. Persistence of cold sensation in the femoral area was defined as a failed sensory block. The persistence of motor function shown by femoral testing (ability to extend the leg with the hip passively flexed at 45°) was noted as a failed motor block. In the case of sensory block failure, motor block failure or both, the femoral nerve block was considered as failed, and a new catheter was introduced using the technique attributed to the patient's group. The motor block was not systematically assessed after surgery and patients were allowed to stand under supervision by a physiotherapist from POD 1 if they were able to raise the operated limb and maintain the leg.
VAS scores at rest were assessed in PACU, and at 12, 24 and 48 h after surgery. VAS score on movement, standardized by 40° flexion of the knee during the first 2 PODs, was also noted and the maximum VAS score was recorded (VASmax). Surgeons evaluated the amplitude of knee flexion without pain on POD 2 and POD 5. The time required to perform the block was noted, defined as elapsed time from the ultrasound probe preparation until catheter placement in the ultrasound group, and from stimulating needle insertion and catheter placement in the neurostimulation group. Also noted were minimum current intensity nerve stimulation, time until first local anaesthetic bolus, time until first morphine request, total morphine dose and any episode of nausea or vomiting. The maximum degree of knee flexion without pain was measured on POD 2 and POD 5 using a goniometer with the patient supine.
The primary study endpoint was total dose of local anaesthetic administered (excluding the preoperative bolus dose) during the first 48 h. On the basis of a previous study,10 assuming a significant difference from 25 ml (SD 35 ml) in total local anaesthetic dose on POD 2 (the equivalent of 31.25 mg levobupivacaine) with a two-tailed type 1 error of 0.05 and a power of 0.9, a sample size of 43 patients per group was required. We decided to enrol a minimum of 90 patients to take into account any possible exclusions occurring during the study. Data analysis was carried out using the Statistical Package for the Social Sciences (SPSS for Windows, version 10.0; SPSS, Inc., Chicago, Illinois, USA). A Kolmogorov–Smirnov test was used, and stratified distribution plots were examined to check the distribution normality of continuous variables. Continuous variables normally distributed were compared with the Student t-test and the Mann–Whitney test of nonparametric data. Repeat measurements (VAS and amplitude of knee flexion) were evaluated using two-way variance analysis and post-hoc comparisons at various times, using Bonferroni's type I error correction for multiple comparisons. Discrete categorical data were compared with the χ2 test or Fisher's exact test (wherever appropriate). Data are expressed as means ± SD, median (25–75th percentiles), numbers and percentages. The median elapsed time before the first request for local anaesthetic with patient-controlled analgesia was estimated with the Kaplan–Meier method and compared using the log-rank test. The level of significance was set at a P value of less than 0.05. An intention-to-treat analysis was performed and all patients with an effective femoral nerve block before surgery were included.
Over a 10-month period, 93 patients were enrolled in the study. One patient withdrew from the study subsequently. There were two failed blocks (failure of motor block on femoral testing, both in the neurostimulation group), but, in accordance with study protocol, they were included in the analysis. These two patients received a second femoral catheter under neurostimulation alone (according to randomization and the protocol), both of which were effective. Two other patients (one in each group) had a catheter prematurely removed before the end of the study at 36 and 43 h through disconnection of the antimicrobial filter on the injection port. They have also been included in the analysis. Demographics and surgical features reported in Table 1 showed no differences between groups. Mean intensity pain scores before patient randomization were similar between groups. The mean VAS score at rest measured the day before surgery was 34 ± 11 mm in the ultrasound group and 37 ± 9 mm in the neurostimulation group [between-group difference = −3.0 mm, 95% confidence interval (CI) −7.2 to +1.4 mm, P = 0.7]. The mean duration of block performance was 6.3 ± 2.8 min in the neurostimulation group and 9.3 ± 4.1 min in the ultrasound group (between-group difference = −3.0 min, 95% CI −4.4 to −1.6 min, P = 0.0007). Femoral catheter placement was less painful in the ultrasound group than in the neurostimulation group [median VAS scores 1 (0–3) vs. 3 (1–4) cm, P < 0.003]. Catheters were easily viewed via ultrasound in 36 out of 46 patients. Minimum intensity of stimulating current was also lower in this group (Table 2). In the ultrasound group, femoral nerve block was achieved on the first attempt in 37 out of 46 patients and after one needle redirection in the remaining nine patients. In the neurostimulation group, stimulation of the femoral nerve was achieved on the first needle insertion in 23 patients, on the second needle insertion in 21 patients and on the third needle insertion in two patients. The median time to achieve complete sensitivity and motor femoral blocks was shorter in the ultrasound group [11 (6–17) vs. 16 (11–23) min in the ultrasound and neurostimulation groups, respectively, P = 0.009].
Concerning the primary endpoint, the mean cumulative local anaesthetic consumption over 48 h was lower in the ultrasound group (299 ± 45 vs. 333 ± 48 ml in the ultrasound and neurostimulation groups, respectively; between-group difference = 34.0 ml; 95% CI 15.5–52.5 ml, P = 0.0003). The first requirement for local anaesthetic bolus by perineural catheter was lengthened (P < 0.02 using the log-rank test, and P = 0.0327 with the Breslow–Gehan–Wilcoxon test, Fig. 1). The median time before the first local anaesthetic bolus administration was 11.0 (7–13) h in the ultrasound group and 7 (4–12) h in the neurostimulation group (P = 0.034).
Postoperative analgesia and knee flexion
Postoperative VAS scores are shown in Fig. 2. Patients in the ultrasound group expressed significantly less pain at rest, 12, 24 and 48 h (P < 0.05). VAS scores during mobilization were also reduced in the ultrasound group at 48 h [14.5 (11.0–23.1) vs. 28.5 (21.0–43.5) mm, P < 0.0001]. The median total immediate release morphine sulphate dose was significantly lower in the ultrasound group [20 ([0–40) vs. 40 (20–60) mg in ultrasound and neurostimulation groups, respectively, P = 0.0065] but the elapsed time before the first request for morphine was equivalent in both groups with 17.8 ± 9.6 min in the ultrasound group and 17.1 ± 7.1 min in the neurostimulation group (between-group difference = −0.7 h, 95% CI −4.3 to +2.8 h, P = 0.85). The median angle of knee flexion was higher on POD 2 in the ultrasound group than in neurostimulation group [50° (43–58°) vs. 45° (39–49°), P = 0.0064, Table 2] but there was little difference by POD 5 [79° (68–87°) vs. 73° (65–79°), P = 0.08, Table 2]. Time before hospital discharge was comparable in both groups.
The incidence of postoperative nausea and vomiting was low and not group dependent (Table 2). Three patients (two in the ultrasound group and one in the neurostimulation group) required transient bladder catheterization. One patient in the ultrasound group suffered from paraesthesia in the saphenous nerve, persisting until POD 5. No patient developed septic complications at the operation site and no local symptomatic infection on the femoral catheter orifice was noted.
The current prospective, randomized study demonstrates that ultrasound guidance improves the quality of the procedure, extending the elapsed time before the first patient-controlled local anaesthetic bolus is requested, reducing the total dose of local anaesthetic solution and opioid consumption. Continuous femoral nerve block reduces postoperative analgesia requirements after TKA by comparison with i.v. opioid analgesic and also reduces side effects.1,3,11,12 Until recently, neurostimulation for continuous femoral nerve block remained the gold standard for achieving effective analgesia after major knee surgery with a high success rate compared with i.v. analgesic or intraarticular infiltration.1,13 Previous studies7,8 have demonstrated that, by comparison with neurostimulation, ultrasound guidance reduces single femoral block onset time and improves the quality of analgesia in adults after hip surgery. Ultrasound-guided femoral nerve block can also reduce the amount of local anaesthetic administered (0.15 vs. 0.3 ml kg−1) in children.5 Most studies are based on an out-of-plane approach for needle insertion and only single-shot injections have been evaluated.5,7,8 The advantage of the in-plane needle approach is the possibility of viewing the entire needle shaft and tip in a longitudinal scan but frequent transducer manipulations are required to maintain both needle and target nerve in the same view. In the current study, the in-plane technique with a transverse view of the femoral nerve enabled correct placement of the needle tip on the first attempt in the majority of cases (36/46 patients). The site of needle insertion was lateral to the nerve and some distance from the femoral vessels. Dextrose solution injection allowed fascia iliaca and femoral nerve movements to be viewed and the needle tip positioning to be optimized. The catheters were specifically put in place under real-time control to maintain their position close to the nerve structure. Catheters inserted blindly are indeed prone to be placed at some distance from nerve structures.3,14,15 On the contrary, as levobupivacaine consumption was lower in the ultrasound group of patients, it is likely that catheters placed under ultrasound guidance were closer to nerve structures. We deliberately injected local anaesthetic via the catheter rather than by needle to reinforce the concept of proximity between nerve structure and local anaesthetic. In addition, ultrasound-guided catheter placement resulted in a more extended sensory block as shown in previous studies.5,7–9,16 A recent qualitative systematic review17 of those studies evaluating ultrasonography and neurostimulation as peripheral nerve blocks documents more rapid block onset when using ultrasound guidance. This was previously demonstrated by Marhofer et al.7,8 in patients administered femoral nerve block for hip surgery.
VAS scores in the neurostimulation group were comparable to those previously reported after knee surgery.1,3,12,16 Lower VAS scores observed in our study in the ultrasound group (Table 2) were associated with improved knee flexion on POD 2 as previously reported.1 However, the improvement in knee flexion in the ultrasound group was transient and limited to the perineural infusion period, suggesting an advantage in the ultrasound-guided catheter insertion with a better nerve-to-catheter proximity. The knee flexions at POD 5 and at discharge were not affected by randomization or the duration of hospital stay. The improvement in VAS scores at rest from 12 h to POD 2 in patients who received an ultrasound-guided catheter was associated with an oral morphine reduction (Table 2). This was previously observed with a conventional femoral perineural catheter by comparison with systemic opioid.1,3,12 The dispersal of the local anaesthetic injected by catheter under the fascia iliaca gave more effective diffusion of local anaesthetic around the femoral nerve and may have provided a more extensive and earlier neural block contributing to reduced peripheral and central sensitizations, which are known to influence the postoperative pain score and rehabilitation.
The use of neurostimulation in combination with ultrasound guidance remains controversial. In the current study, the combined technique provided better results than neurostimulation alone in terms of analgesia and ease of performing the block, as indicated by the number of needle adjustments. However, achieving a complete block was slightly more rapid in the ultrasound group. In another recent comparative study of patients undergoing a partial or complete femoral block, Sites et al.18 reported no difference between ultrasound alone vs. ultrasound combined with nerve stimulation. The difference in findings between the two studies might be explained by the fact that it is easier to block the femoral nerve using a single injection through a needle than it is through a catheter that requires correct positioning. Other authors have demonstrated that the combined technique was more effective in cases of more complex anatomy such as infraclavicular location.19 Nevertheless, the results from combined use of the ultrasound probe and neurostimulation also show a reduction in patient satisfaction and comfort when performing the block. Continuous femoral block is associated with better knee flexion when compared with i.v. analgesia but the advantages of single administration nerve block are still controversial and frequently limited to the first postoperative day.1,12,20
The lack of double blinding is a limitation in this study and may have introduced a bias in data collection. However, the majority of recent studies16,18,19 evaluating ultrasound guidance and neurostimulation have pointed out the impossibility of double blinding in the case of ultrasound. We have not evaluated the number of bolus failures in patient-controlled perineural analgesia. However, the prolonged and more effective analgesia observed with the first bolus injection (administered before incision and via the catheter) probably reduced the risk of failed administration.
Ultrasound guidance combined with neurostimulation for perineural femoral catheter placement improves postoperative analgesia and reduces femoral block installation time, as well as local anaesthetic consumption and total morphine doses by comparison with the conventional neurostimulation technique.
1 Capdevila X, Barthelet Y, Biboulet P, et al
. Effects of perioperative analgesic technique on the surgical outcome and duration of rehabilitation after major knee surgery. Anesthesiology 1999; 91:8–15.
2 Liu SS, Wu CL. The effect of analgesic technique on postoperative patient-reported outcomes including analgesia: a systematic review. Anesth Analg 2007; 105:789–808.
3 Singelyn F, Gouverneur JM. Extended ‘three-in-one’ block after total knee arthroplasty: continuous versus patient-controlled techniques. Anesth Analg 2000; 91:176–180.
4 Heid F, Muller N, Piepho T, et al
. Postoperative analgesic efficacy of peripheral levobupivacaine and ropivacaine: a prospective randomized double-blind trial in patients after knee arthroplasty. Anesth Analg 2008; 106:1559–1561.
5 Oberndorfer U, Marhofer P, Bösenberg A, et al
. Ultrasonographic guidance for sciatic and femoral nerve blocks in children. Br J Anaesth 2007; 86:797–801.
6 Casati A, Baciarello S, Di Cianni S, et al
. Effects of ultrasound guidance on the minimum effective anaesthetic volume required to block the femoral nerve. Br J Anaesth 2007; 98:823–827.
7 Marhofer P, Schrögendorfer K, Wallner T, et al
. Ultrasonographic guidance reduces the amount of local anesthetic for 3-in-1 blocks. Reg Anesth Pain Med 1998; 23:584–588.
8 Marhofer P, Schrögendorfer K, Koinig H, et al
. Ultrasonographic guidance improves sensory block and onset time of three-in-one blocks. Anesth Analg 1997; 85:854–857.
9 Marhofer P, Greher M, Kapral S. Ultrasound guidance in regional anaesthesia
. Br J Anaesth 2005; 94:7–17.
10 Casati A, Vinciguerra F, Cappelleri G, et al
. Adding clonidine to the induction bolus and postoperative infusion during continuous femoral nerve block delays recovery of motor function after total knee arthroplasty. Anesth Analg 2005; 100:866–872.
11 Chelly J, Greger J, Gebhard R, et al
. Continuous femoral blocks improves recovery and outcome of patients undergoing total knee arthroplasty. J Arthroplasty 2001; 16:436–445.
12 Syngelyn F, Deyaert Y, Joris D, et al
. Effects of intravenous patient-controlled analgesia with morphine, continuous epidural analgesia, and continuous ‘3-in-1’ block on postoperative pain and knee rehabilitation after total knee arthroplasty. Anesth Analg 1998; 87:88–92.
13 Dauri M, Fabbi E, Mariani P, et al
. Continuous femoral nerve block provides superior analgesia compared with continuous intra-articular and wound infusion after anterior cruciate ligament reconstruction. Reg Anesth Pain Med 2009; 34:95–99.
14 Capdevila X, Biboulet P, Moreau D, et al
. Continuous three-in-one block for postoperative pain after lower limb orthopedic surgery: where do the catheters go? Anesth Analg 2002; 94:1001–1006.
15 Dauri M, Sidiropoulou T, Fabbi E, et al
. Efficacy of continuous femoral nerve block with stimulating catheters versus nonstimulating catheters for anterior cruciate ligament reconstruction. Reg Anesth Pain Med 2007; 32:282–287.
16 Dolan J, Williams A, Murney E, et al
. Ultrasound guided fascia iliaca block: a comparison with the loss of resistance technique. Reg Anesth Pain Med 2008; 33:526–531.
17 Liu SS, Ngeow JE, YaDeau JT. Ultrasound-guided regional anesthesia and analgesia. A qualitative systematic review. Reg Anesth Pain Med 2009; 34:47–59.
18 Sites BD, Beach ML, Chinn CD, et al
. A comparison of sensory and motor loss after a femoral nerve block conducted with ultrasound with ultrasound versus ultrasound and nerve stimulation. Reg Anesth Pain Med 2009; 34:508–513.
19 Dingemans E, Williams S, Archand G, et al
in ultrasound guided infraclavicular block: a prospective randomized trial. Anesth Analg 2007; 104:1275–1280.
20 Salinas FV, Liu SS, Mulroy MF. The effects of single-injection femoral nerve block versus continuous femoral nerve block after total knee arthroplasty on hospital length of stay long-term functional recovery within an established clinical pathway. Anesth Analg 2006; 102:1234–1239.