An Estimation of the Minimum Effective Anesthetic Volume of 2% Lidocaine in Ultrasound-guided Axillary Brachial Plexus Block

The Cork Teaching Hospitals Research Ethics Committee granted approval for this study. Patients undergoing upper limb surgery to the forearm and hand were invited to participate in the study. Inclusion criteria included age 18-80 yr, American Society of Anesthesiologists Grade I-III, weight 40-110 kg, unilateral surgery, and expected duration of surgery of less than 90 min. Exclusion criteria included language barrier, contraindication to regional anesthesia, bleeding diathesis, inability to visualize one or more nerves in the axilla on ultrasound, psychiatric history, chronic pain history, and pregnancy.

After securing written informed consent, patients were included in the study. The starting dose was chosen arbitrarily as 4 ml per nerve of 2% LidoEpi on the basis of clinical experience. Using a step-up/step-down model, the dose used for following patients was determined by the outcome of the preceding block. In the case of block failure, the dose was increased by 0.5 ml per nerve. Conversely, block success resulted in a reduction in dose by 0.5 ml per nerve. Each of the four nerves was treated as a separate entity. Dose adjustments were made to each nerve individually.

Intravenous access was established in the nonoperative upper limb and standard monitoring was attached (noninvasive blood pressure, electrocardiography, and pulseoximetry). The operative arm was abducted and externally rotated, and the elbow flexed to 90 degrees. The skin of the axilla and the medial upper arm was prepared in an aseptic fashion. A sterile transparent cover was placed on the ultrasound transducer, and sterile water-based ultrasound gel was used as an acoustic couplant. Ultrasound examination of the axilla was performed using a SonoSite Titan unit (SonoSite®, Bothwell, WA) with a 38-mm high frequency (7-10 MHz) linear array transducer (L38).

The axillary block was performed under ultrasound guidance alone, and nerve stimulation was not used. The ultrasound transducer was placed in a vertical orientation at the level of the anterior axillary fold. The axillary artery was identified and placed in the center of the image. Minor adjustment of scanning planes facilitated the identification of the median, ulnar, and radial nerve complexes surrounding the axillary artery. The musculocutaneous nerve was then identified in a connective tissue plane between the biceps and the coracobrachialis muscles. The identity of each nerve was confirmed by tracing the nerve from the axilla to a fixed reference point and then back to the axilla. The ulnar nerve was traced to and from the medial epicondyle. The median nerve was traced to and from the antecubital fossa, where it lies medial to the brachial artery. The radial nerve may be seen to emerge in a fascial plane within the triceps muscle at the midhumeral level and may be traced back to the axilla. The musculocutaneous nerve can be identified in a fascial plane between biceps and coracobrachialis muscles and traced backwards to the axilla.

A 50-mm 24-gauge Stimuplex® (BBraun, Melsungen, Germany) insulated blunt regional anesthesia needle was introduced percutaneously at the center of the transducer, directly parallel to the scanning beam. A needle-out-of-plane approach was used, and the needle was advanced to positions adjacent to the median, ulnar, and radial nerves in this order. The study volume of 2% LidoEpi was injected adjacent to each nerve. The injectate was administered slowly in 0.5-ml aliquots, and evidence of inadvertent intraneural injection was sought. Intraneural injection was deemed to occur if the target nerve increased in size after injection of 0.2-0.5 ml of local anesthetic agent.9The needle was repositioned during injection and circumferential perineural injectate spread was ensured. After blockade of the ulnar, median, and radial nerves, the block needle was withdrawn to the subcutaneous tissues and redirected toward the musculocutaneous nerve by using a needle-in-plane approach. The study volume of 2% LidoEpi was injected adjacent to the musculocutaneous nerve in 0.5-ml aliquots ensuring perineural local anesthetic placement (fig. 1). A single skin puncture was used for the entire block procedure. No subcutaneous local anesthetic infiltration was used before block needle insertion.

An assessor blinded to injectate volume evaluated the presence of motor and sensory blockade in each nerve territory. Sensory and motor function was assessed in the innervation of each of the nerves. Simultaneous comparison of sensory and motor function in the contralateral limb was used as a point of reference. Block assessment was performed at 5-min intervals up to 30 min after completion of the last perineural injection. Block assessment was halted either after surgical anesthesia was achieved or after 30 min had elapsed. Motor function was graded on a modified Bromage Scale, and sensory function was assessed using pinprick (25-gauge hypodermic needle), soft touch (gauze), and cold (ethyl chloride spray) (table 1). Sensory function was scored as being present or absent. Surgical anesthesia was defined as a motor score of 2 or lower, with absent appreciation of cold and pinprick sensation. The time interval at which surgical anesthesia was achieved was noted.

Block failure was defined as absence of surgical anesthesia at 30 min in one or more of the four nerve territories. Block failure was managed by either the performance of a rescue block at a point distal to the axilla or conversion to general anesthesia at the patient’s discretion.

In the event of a successful block, the patient was asked to remember either the time of day that purposeful movement returned to their upper limb, or the time they first noticed sensation or pain returning. This time was used to estimate the duration of axillary bock. It was not possible to perform postoperative sensory testing on the operative limb due to the presence of surgical dressings.

Sedation was provided on patient request using a combination of midazolam 2 mg with a low-dose propofol infusion, if required, titrated to a level of responsiveness to voice command.

All patients received 1 g of intravenous acetaminophen and 75 mg of diclofenac sodium during surgery. Postoperative analgesia consisted of 1 g of oral acetaminophen every 6 hours and 75 mg of diclofenac sodium twice daily for 72 h after surgery. Oxycodone 5 mg was prescribed for rescue analgesia after block regression.

Two study-stopping rules were chosen. Based on previous nonprobability sequential dosing, up-and-down dose finding studies with similar binary outcomes,6,10,11we estimated that a minimum of five independent negative-positive up-and-down deflections was required to calculate MEAV⁵⁰. It was arbitrarily agreed that reducing the dose beyond 1 ml per nerve was of little clinical significance. Therefore five consecutive patients with successful axillary block at a dose of 1 ml per nerve was chosen as the second stopping point.

Summary data were calculated using EpiInfo 2002 (Centers for Disease Control and Prevention, Atlanta, GA) and presented as median (range) or mean (SD) as appropriate. It was intended to calculate the MEAV⁵⁰of 2% LidoEpi for each of the four nerves in the axilla. The MEAV⁵⁰of 2% LidoEpi was defined as the midpoint of pairs of volumes from consecutive patients in which a negative response (inadequate block within 30 min) is followed by a positive one (adequate block within 30 min).12 

Eleven patients completed the study protocol. No patients who met the inclusion criteria were excluded from the study. No patients were excluded on the basis of inability to visualize nerve structures in the axilla. The study group baseline characteristics are summarized in table 2. Recruitment to the study was terminated when five consecutive patients had successful ultrasound-guided axillary block using 1 ml of 2% LidoEpi per nerve (fig. 2). It was not possible to calculate the MEAV⁵⁰of 2% LidoEpi because the secondary stopping rule was reached.

For the entire group (n = 11), the median (range) block onset time was 5 (5-20) min, the median (range) block duration was 190 (120-310) min (fig. 3), and the mean (SD) block performance time was 445 (100) s. The dose-related block onset times for each nerve are summarized in table 3. Intraoperative sedation was provided on patient request for 3 of 11 cases. The block success rate was 100%, and no patients required either a distal rescue block or conversion to general anesthesia.

Four of the five patients in the 1 ml per nerve subgroup had surgical anesthesia at 5 min, and one patient at 10 min. The median (range) block duration for the subgroup of patients at 1 ml per nerve was 180 (120-190) minutes. One patient in the 1 ml per nerve subgroup received intraoperative sedation for anxiolysis.

All surgical procedures were performed under regional anesthesia. One patient (using 2 ml per nerve) required subcutaneous infiltration of 2 ml of 1% lidocaine under the skin of the anterior axillary fold due to discomfort associated with tourniquet inflation. No patients required opiate analgesics after block regression. There were no adverse incidents during the study period. No patients discharged on the day of surgery required hospital readmission.

Ultrasound technology permits nerve, needle, and injectate visualization, which facilitates a dynamic block-needle endpoint and accurate perineural local anesthetic placement. Ultrasound guidance has been associated with high block success rates, reduced bock onset time, and reduced local anesthetic dose.1,3,6This study demonstrates the utility of ultrasound guidance in achieving successful axillary brachial plexus block with low doses of local anesthetic agent. A total of 4 ml of 2% lidocaine with 1:200,000 epinephrine produced upper limb anesthesia in adults undergoing elective and emergency upper limb surgery.

The duration of upper limb block, defined as the time taken from block performance to the return of either motor or sensory function, is sufficient for most upper limb procedures (180 min; range 120-190 min). This relatively short block duration may have benefits in the ambulatory setting. Despite evidence demonstrating safe hospital discharge of patients with an insensate limb,13,14discharge without protective reflexes remains controversial because concern exists that accidental injury may occur.15Therefore, axillary brachial plexus of short block duration, with effective balanced multimodal analgesia, may be beneficial for ambulatory upper limb surgery. The onset and offset profiles of ultrasound-guided axillary block with 1 ml per nerve is the subject of further study by our group.

Regional anesthesia has unwanted side effects, such as systemic toxicity, which may be dose-related.16Recent reports of local anesthetic systemic toxicity have demonstrated the efficacy of lipid rescue in the management of this serious complication.17,18,19,20Rather than focus on the treatment of systemic toxicity, modern regional anesthesia practice might develop strategies for the prevention of this potentially fatal complication. Reducing the dose of local anesthetic is one such logical step.

We acknowledge that the study has a number of limitations. The small number of patients included (n = 11) may limit the external validity of the data. Nonetheless, the planned endpoint of achieving successful block with 1 ml per nerve in five consecutive patients was reached. Therefore stopping the study was appropriate. A single anesthesiologist performed all of the blocks. Although this eliminates interoperator variability, it may limit the generalizability of the results. We acknowledge that this may reflect the proficiency of the individual rather than the efficacy of the low-dose technique. The study failed to calculate the MEAV⁵⁰of 2% LidoEpi; however, we believe that reducing the dose lower than 1 ml per nerve is of questionable clinical significance.

In summary, our study demonstrates that ultrasound-guided axillary brachial plexus block can be accomplished with as little as 1 ml of 2% lidocaine with 1:200,000 epinephrine per nerve (4 ml total). The widespread clinical applicability of this technique is unknown because substantial expertise in ultrasound imaging and clear visualization of the individual nerves are prerequisite. However, with ongoing educational efforts and continuous improvement in ultrasound technology, a substantial reduction in local anesthetic dose conferring increased patient safety may be possible in the future.