In 2020, a trial demonstrated that a 2-injection technique constitutes the optimal method for the costoclavicular block.

This study aims to determine the minimum amount of medication required to achieve a successful double-injection costoclavicular nerve.

Block dose assignment will be done using an up-and-down sequential method, called the Biased Coin Design (BCD).

The double-injection technique for US-guided costoclavicular block consists in depositing two thirds of the LA volume in the deep compartment (to anesthetize the posterior and medial cords), and one third of the injectate in the superficial compartment (to anesthetize the lateral cord).

The ED90 will be calculated using isotonic regression with a 95% confidence interval derived by bootstrapping.

With the approval of the ethics committee of McGill University Health Centre, a total of 60 patients undergoing upper extremity surgery (elbow or below elbow) will be recruited.

The costoclavicular block will be performed under direct US vision until the tip is between the medial cord and the axillary artery. Two thirds of the volume of LA will be injected in this first location. Subsequently, the needle will be advanced until its tip is positioned next to the lateral cord. The remaining third of the LA volume will be injected in this second location.

Dose assignation will be done using an up-and-down sequential method, where the dose of each subsequent patient depends on the response of the previous patient, called the Biased Coin Design. The first subject recruited will receive a total of 30 mL of lidocaine 1.5 % with epinephrine 1: 200 000. Injection will be carried out by slow increments (5 mL) with negative aspiration between each increment.

The assignment of each subsequent dose will be based on the response of the previous patient as follows: if the previous patient did not have a successful block, the patient will receive the next higher dose, which is the previous dose incremented by 3.0 ml (i.e., an increment of 2.0 mL between the medial cord and the axillary artery, and an increment of 1.0 mL next to the lateral cord). If the previous patient had a successful block, the patient will be randomized to either receive the next lower dose, which is the previous dose decremented by 3.0 ml (i.e., a decrement of 2.0 mL between the medial cord and the axillary artery, and a decrement of 1.0 mL next to the lateral cord), with a probability b = 0.11, or to receive the same previous dose, with a probability 1 - b = 0.89. These are the probabilities required for assigning doses under the BCD for estimating ED90. A maximal dose of 40 mL will not be exceeded to avoid LA toxicity. If the previous patient did not respond and had been given the maximal dose the patient will also get the maximal dose. As soon as we recruit 45 patients with random dose assignment, we will terminate the enrollment procedure.

A research assistant will prepare and administer the injectate using syringes connected to the block needle. The operator and the patient will be blinded to the volume injected.

After LA injection at the second site, a catheter will be advanced 3 cm past the needle tip and secured to the skin. The purpose of the catheter is to allow LA supplementation for unsuccessful blocks, LA reinjection for lengthy surgical procedures or postoperative LA infusion for pain control.

The primary outcome is success rate which is defined as a minimal composite score of at least 14 points (out of a maximum of 16) is achieved at 30 minutes. The ED90 will be calculated using isotonic regression with a 95% confidence interval derived by bootstrapping .

Inclusion Criteria:

• Age between 18 and 70 years
• American Society of Anesthesiologists classification 1-3
• Body mass index between 20 and 30

Exclusion Criteria:

• Adults who are unable to give their own consent
• Pre-existing neuropathy (assessed by history and physical examination)
• Coagulopathy (assessed by history and physical examination and, if deemed clinically necessary, by blood work up i.e. platelets≤ 100, International Normalized Ratio≥ 1.4 or partial prothrombin time ≥ 50)
• Renal failure (assessed by history and physical examination and, if deemed clinically necessary, by blood work up i.e. creatinine≥ 100)
• Hepatic failure (assessed by history and physical examination and, if deemed clinically necessary, by blood work up i.e. transaminases≥ 100)
• Allergy to LA
• Pregnancy
• Prior surgery in the infraclavicular region
• Chronic pain syndromes requiring opioid intake at home

A research assistant (a licensed anesthesiologist) will prepare the local anesthetic (lidocaine) solutions according to the sequential allocation. The operator, patient, and outcome assessor will remain blinded to the assignment.