Ropivacaine in ultrasound-guided femoral nerve block: what is the minimal effective anaesthetic concentration (EC90)?


  • This article is accompanied by an editorial titled “Peripheral nerve block and local anaesthetic dose, how much is enough?” by O' Donnell and Szücs. Anaesthesia 2014; 69: 665–8.
  • You can respond to this article at


The objective of this study was to estimate the minimal effective anaesthetic concentrations of ropivacaine required to block the femoral nerve in 90% of patients. Forty-five patients who had knee surgery received ultrasound-guided femoral nerve block using 15 ml ropivacaine. The ropivacaine concentration given to a patient relied on the efficacy of the block in the previous patient, using the biased-coin design up–down sequential method. In the event of successful block, the next patient was randomly assigned to receive either the same ropivacaine concentration or a concentration 0.02% w/v less. In the event of a failed block, the next patient received a concentration 0.02% w/v higher. Successful block was defined as complete sensory and motor block before surgery together with pain-free surgery. The minimal effective ropivacaine concentration was estimated to be 0.167% w/v (95% CI 0.14–0.184%). Perineural injection of 15 ml ropivacaine 0.167% w/v under ultrasound guidance can provide successful femoral nerve block in 90% of patients.


Peripheral nerve block can provide adequate anaesthesia and postoperative analgesia for limb surgery with few side-effects [1, 2]. Femoral nerve block is commonly performed, relatively simple, effective and with numerous indications [2-4]. Ropivacaine is a long-acting local anaesthetic that is widely used including for femoral nerve block [4-8]. The success rate of a nerve block relies on multiple interacting variables [9-14]; increasing the local anaesthetic volume or concentration generally improves block success rate, but at the expense of an increase in the local anaesthetic dose [9-14]. Furthermore, multiple nerve blocks may be required for lower limb surgery [12] leading to the frequent use of maximum recommended total doses. Clinical investigations have therefore been performed to address whether it is possible to reduce local anaesthetic doses but still reliably achieve successful nerve block [9-14]. The objective of this study was to estimate the minimal effective anaesthetic concentration of ropivacaine required to produce ultrasound-guided femoral nerve block in 90% of patients (EC90).


The Research and Ethical Committee of Abu Dhabi Knee and Sport Medicine Centre approved this study, which was conducted from January to April 2013. Patients of ASA physical status 1–2, aged > 18 years, scheduled for total knee arthroplasty or knee ligament reconstruction were included. Written informed consent was obtained from the patients.

In the block room, routine monitoring consisting of ECG, pulse oximeter and non-invasive arterial pressure was applied. Supplemental oxygen was administered via a mask at 4 l.min−1. Intravenous access was established and 3 mg midazolam administered. Ultrasound-guided parasacral sciatic, proximal interfacial obturator and lateral femoral cutaneous nerve (lateral cutaneous nerve of the thigh) blocks were performed in all patients using, respectively, 20, 10 and 2 ml ropivacaine 0.5% containing 5 μ−1 adrenaline [15-17].

All femoral nerve blocks were performed by an experienced anaesthetist (AMT) using an S-Nerve machine with linear ultrasound probe (HFL 38, 13–6 MHz; SonoSite Inc, Bothell, WA, USA) and a 21-G 5-cm-long needle (Locoplex, Vygon, Ecouen, France) [18]. Under aseptic conditions, the ultrasound probe was placed on the inguinal crease, with a slight cephalic tilt, to identify the femoral artery and nerve. The needle was advanced using an in-plane approach (from lateral to medial), and 15 ml study solution (see below) was slowly injected; 10 ml was placed deep and 5 ml superficial to the nerve. Adequate local anaesthetic spread around the nerve was confirmed and intraneural injection was avoided.

Sensory and motor block were assessed at 15, 25 and 35 min after the injection. Motor block was classified as one of three grades according to the patient's ability to extend their knee: 0 (normal motor power); 1 (motor weakness); and 2 (complete motor paralysis). Femoral sensory block was assessed by testing pinprick sensation using a 23-G hypodermic needle at four points on the medial aspect of the leg, approximately 1, 8, 13 and 18 cm cephalic to the medial malleolus. Sensory block was classified as one of three grades: 0 (normal sensation); 1 (decreased sensation); and 2 (absent sensation).

Before taking the patient to the operating room, successful block of the sciatic, obturator and lateral femoral cutaneous nerves was confirmed [19] and incomplete blocks were supplemented. No intra-operative sedatives were used except before hammering the femoral implant in patients with total knee arthroplasty, or if the patient complained of backache during multiligament reconstruction surgery. Neurological assessment was performed in all patients before hospital discharge, and also during physiotherapy visits for three weeks postoperatively.

The patients were considered to have a successful block only if they had grade-2 sensory and motor block within 35 min and completely pain-free surgery with no systemic opioid administration. Otherwise, they were considered to have a failed block and received a supplementary femoral nerve block or light general anaesthesia. Ropivacaine solutions were prepared by the operator anaesthetist. The success and failure of the block and any complications were assessed and recorded by an assistant who was unaware of the study solution used.

Based on the biased-coin design up–down sequential method, ropivacaine concentration given to a patient depended on the response of the previous patient [13]. When a patient had a failed block, the ropivacaine concentration was increased by 0.02% w/v in the next patient. When a patient had a successful block, the next patient was randomly assigned to receive the same ropivacaine concentration (with a probability of 0.89) or to receive a concentration 0.02% w/v less (with a probability of 0.11). To our knowledge, there has only been one clinical trial designed to estimate EC50 of ropivacaine required for femoral nerve block, but its results have not yet been published [9]. The initial ropivacaine concentration (0.12% w/v) was therefore based on our past experience. To avoid local anaesthetic toxicity, we set a maximum concentration of 0.5% ropivacaine.

A minimum of 45 successful blocks (the smallest multiple of 9 that is > 40) were needed to calculate the EC90 [13]. Isotonic regression with bias corrected 95% confidence interval (CI) derived by bootstrapping was used to estimate the EC90 [20]. The target dose estimator μ3, used in this study, was shown to be more advantageous relative to other estimators [20-22]. This estimator was calculated by linear interpolation from the two consecutive concentrations, the success rates of which enclosed the value of probability of effect ‘Г’ (which is 0.9 in this study). We preferred to state the mean value of the estimate (μ3) with CI rather than the value obtained from the original sample. The mean value was obtained from the 4000 bootstrap samples. Statistical analysis was performed using Minitab® 16.1 Statistical Software 2010 (Minitab, Inc., State College, PA, USA) and Microsoft® Excel 2007 (Microsoft, Seattle, WA, USA).


Fifty-two patients completed the study (Table 1); 45 had a successful block and seven had a failed block. The mean EC90 was estimated to be 0.167% w/v (95% CI 0.14–0.184%; Fig. 1). Ropivacaine 0.167% w/v was estimated to produce a successful block in 89.3% (95% CI 76.9–97.5%) of patients. Grade-2 motor block was not achieved in the seven patients with a failed block. In the 45 patients with successful block, grade-2 motor block was always preceded by grade-2 sensory block and was always associated with painless surgery. Grade-2 motor block required more than 25 min in 38 (84%) patients. Twenty-four (53%) patients required inflation of a tourniquet. This was always painless if a successful grade-2 motor block was present. No complications were recorded during the study.

Table 1. Characteristics of 52 patients. Values are mean (SD) or number (proportion).
  1. BMI, body mass index; ACLR, anterior cruciate ligament reconstruction.

Age; years33.3 (12.6)
BMI; kg.m−227.7 (5.3)
ASA physical status 146 (88%)
Male44 (85%)
ACLR using hamstring graft34 (65%)
ACLR using patellar tendon9 (17%)
Total knee arthroplasty6 (12%)
Multi-ligament reconstruction3 (6%)
Sciatic nerve block supplementation4 (8%)
Figure 1.

Plot of successful (image) and failed (image) blocks with different ropivacaine concentrations. The horizontal line is the calculated minimal concentration of ropivacaine providing successful femoral block in 90% of patients (EC90); error bars represent 95% CI.


In this study, 0.167% w/v ropivacaine was estimated to induce complete femoral nerve block in 90% of patients.

Nerve blocks can provide adequate anaesthesia and postoperative analgesia for knee and hip surgery with minimal disturbance of haemodynamic status unless local anaesthetic toxicity occurs. Local anaesthetic toxicity is a potentially fatal complication resulting from direct intravenous injection of small doses of local anaesthetic or injection of an overdose into the correct perineural space [23]. A near-maximum dose of local anaesthetic is usually required to achieve painless knee surgery as multiple nerve blocks (femoral, sciatic and obturator) are usually needed [12].

To minimise the required local anaesthetic dose, many studies have been performed to estimate the minimal effective concentration or minimal effective volume [9-14]. The up–down sequential method is commonly used to estimate the EC50 and effective volume for 50% success (EV50) [10], but both are of little clinical importance as by definition they have a 50% failure rate. Higher quantiles cannot be accurately estimated by using up–down sequential analysis [13]. Instead, we used the biased-coin design up–down sequential method that can accurately and directly estimate the effective concentration or effective volume at any quantile [13]. Many factors can affect effective concentration including the volume injected, use of adjuvant, the nerve to be blocked, block level, the localising technique and the injection device (needle or catheter) [9-14].

The minimal effective volume of ropivacaine but not the minimal effective concentration to block the femoral nerve successfully has been published [24]. High ropivacaine concentrations of 0.5–1% are usually used to provide anaesthesia [6]. Ropivacaine 0.25% has been reported to induce dense lumbar plexus block [7]. The advance of ultrasound technology with higher image resolution in the last few years allows further reduction in the required local anaesthetic dose [11, 12]. We found that ropivacaine 0.167% provided complete femoral nerve block adequate for surgery in 90% of patients. This low concentration has not been used before to provide anaesthesia. Infusion of 0.2% ropivacaine through a femoral catheter is commonly used for postoperative analgesia; however, it is not associated with the dense quadriceps block achieved in this study [8]. This contradiction may be related to the use of different injection devices (needle vs catheter). In this study, the ability of dynamic adjustment of the needle tip under ultrasound guidance allows precise placement of local anaesthetic at multiple sites all around the nerve [11]. This advantage cannot be achieved by injection through the catheter [14]. Onset of femoral motor block in this study, although not precisely measured, was markedly delayed. However, once achieved, it was always associated with painless surgery.

The following factors may limit the general application of our results. The definition of successful block, success rate and the estimated EC90 relied on a fixed 15-ml volume of ropivacaine and a fixed 35-min time limit; using a smaller volume or shorter time limit might not be associated with similar results. All blocks were performed by one anaesthetist and most of the patients studied were middle-aged healthy males. Patient factors such as age, obesity, sex and coexisting diabetic neuropathy affect block duration and success rate, and may or may not affect the minimal effective concentration [25-28].

In conclusion, we found that perineural injection of 15 ml 0.167% w/v ropivacaine under ultrasound guidance provided successful femoral nerve block in 90% of patients.


Study equipment support was provided by departmental sources. We sincerely thank Dr Rania Hamza and Dr Raham Hasan (Lecturers in Anaesthesia, Ain Shams University, Cairo, Egypt), Amy Gilbert and the staff of Abu Dhabi Knee and Sport Medicine Centre for their tremendous help.

Competing interests

No external funding and no competing interests declared.