Continuous interscalene brachial plexus block consists of inserting a catheter directly within the brachial plexus in the interscalene groove, between the middle and anterior scalene muscles, which is then infused with local anesthetic. This technique provides potent and site-specific analgesia for several days after major shoulder surgery, along with increased shoulder range of motion. A major side-effect of this procedure is the paralysis of the ipsilateral diaphragm, which is the main muscle of the respiration. This paralysis occurs at a rate of 90 to 100%, and is due to the spread of the local anesthetic towards the phrenic nerve, lying anteriorly to the plexus, between the sternocleidomastoid and anterior scalene muscles. Therefore, many patients suffering from respiratory syndromes, such as chronic obstructive pulmonary disease, will not have a continuous interscalene brachial plexus block, thus forfeiting some, or all, of the aforementioned benefits associated regional anesthesia.

It is conventionally taught to place the needle tip between the nerve roots of the brachial plexus to obtain an effective block (intra-plexic injection). Recently, we have published a new technique of single-shot injection for interscalene brachial plexus block where the needle tip was positioned at a distance of 4 mm of the lateral part of the brachial plexus (extra-plexic injection) and resulted in a reduction rate of 70% of hemidiaphragmatic paresis and a preservation of spirometric values, while providing similar analgesia, when compared to a conventional injection. Contrary to single-shot injections where high volume and high concentration of local anesthetics are injected (eg, ropivacaine 0.5%, 20 mLs), continuous blocks required low volume and low concentration (eg, ropivacaine 0.1-2%, 2-6 mLs/h) and therefore the aforementioned concept of extra-plexic technique might not be suitable.

Recently, new catheter-over-the needle (Contiplex® C; B. Braun Melsungen AG, Germany) have been released on the market and allows the physician to place the catheter tip exactly where he desires, contrary to the previous generations where the catheters were inserted blindly, despite the use of ultrasound.

The first objective of that randomised controlled trial is to demonstrate that an extra-plexus catheter placement produces less respiratory complications than an intra-plexus catheter placement. The second objective is to confirm that both techniques provide similar analgesia.

Patients will be brought to the block room theatre. Before nerve blockade, standard monitoring including non-invasive blood pressure, electrocardiogram, and pulse oximetry will be applied and IV access established. Sedation and anxiolysis will be achieved with IV midazolam (incremental doses of 1 mg) and IV fentanyl (incremental doses of 25 µg).

Continuous interscalene brachial plexus block will be performed with the patient lying lateral on the non-operative side. The ultrasound probe will be placed in the interscalene region to visualize the carotid artery and brachial plexus in the transverse sectional view as is routine in our institution. The C5-C6-C7 roots will be identified following the description of Martinolo et al. After sterilization and injection of local anesthetic (1-2% lidocaine) into the skin, a catheter-over-the-needle (Contiplex® C; B. Braun Melsungen AG, Germany) will be placed on the lateral side of the probe and advanced along the long axis of the probe in the same plane as the ultrasound beam. For patients in the intra-plexus group, the catheter tip will be positioned beyond the sheath, between C5 and C6 roots. In the extra-plexus group, the catheter tip will be positioned 2 mm away from the lateral sheath of the brachial plexus. After the injection of dose-test to exclude an intravascular placement, twenty milliliters of local anesthetic containing ropivacaine 0.5% will be injected over 3-5 min under ultrasound visualization.

Intraoperative and postoperative procedure After application of routine monitors in the operating theatre, patients will receive a standard general anesthetic. Anesthesia will be induced using fentanyl 1-2 µg/kg IV and propofol 2-4 mg/kg IV with endotracheal intubation facilitated by rocuronium 0.6 mg/kg IV. Maintenance of anesthesia will be ensured via inhaled sevoflurane 1.6-2.5% in a 40:60 mixture of oxygen and air. Positive pressure ventilation will be initiated with tidal volume and rate adjusted to maintain an end-tidal PCO2 of 30-40 mmHg. Fentanyl will be administered as needed to treat increases in blood pressure and/or heart rate of more than 15% above pre-induction baseline values. Prophylaxis of postoperative of nausea and vomiting will be performed with dexamethasone 0.15 mg/kg after the induction, droperidol 1 mg and ondansetron 4 mg at the end of the surgery, as per routine practice in our institution. Muscle relaxation will be antagonized with neostigmine 50 µg/kg and glycopyrrolate 5-10 µg/kg routinely.

During postoperative recovery in hospital, pain (Numeric rating scale [NRS] ≥ 4 or patient request for analgesia) will be treated with morphine 2 mg every 10 min as needed, and the ropivacaine 0.2% infusion will be set at a rate of 2 ml/h with bolus of 4 ml available every 30 minutes, as per our routine institutional practice. Once oral intake will be initiated, patients will receive acetaminophen 1000 mg PO every 6 h and breakthrough oxycodone 5 mg per os (PO) as needed, max 8 times. On the ward, infusion rate of ropivacaine will be increased to 4 ml/h and then 6 ml/h in case of pain scores ≥ 4. The catheter will be removed on the morning of postoperative day 3.

On postoperative day 1, 2, 3, and 4 a blinded research assistant will visit patients and record data. Patients will also be contacted on postoperative day 30 to capture any block-related complications such as persistent paresthesia, weakness, bruising, or non-surgical pain in the operative extremity.

All these managements represent the current standard of care at Centre Hospitalier Universitaire Vaudois.

Measurement of respiratory outcomes Diaphragmatic movement will be assessed by real-time M-mode ultrasonography on each side using a curvilinear 2 - 5 MHz US probe, following previous published descriptions. Patients will be examined in the lying position and scanned from a low intercostal or subcostal approach using the liver or spleen as an acoustic window. The range of diaphragmatic movement from the resting expiratory position (functional residual capacity) to deep and quiet inspiration (sigh test) will be recorded as will the range of diaphragmatic movement from resting expiratory position when quickly inspirating through the nose (sniff test). The movement of the diaphragm will be measured in centimeters.

• diaphragmatic movement reduction of more than 75%, no movement, or paradoxical movement will be considered to be " complete paresis " ;
• diaphragmatic movement reduction of both sigh and sniff test between 25% and 75% will be considered to be " partial paresis " ;
• diaphragmatic movement of less than 25% will be considered to be " no paresis ".

Normal caudad movement will be designated as positive, whereas paradoxical cephalad movement will be designated as negative. Each test will performed 3 times, and values will be averaged. All measurements will be performed before the surgery (baseline), after the surgery in phase 1 recovery (postanesthetic care unit) and on postoperative day 1.

A bedside spirometer (EasyOneTM Spirometer; ndd Medical Technologies, Andover, UK) will be used to assess ventilatory function (pulmonary function tests). After instructions, the full vital capacity (VC) in lying and sitting upright positions will be measured. Other measurements performed in patients in upright sitting position will be forced expiratory volume at 1 sec (FEV1), forced vital capacity (FVC), and peak expiratory flow (PEF); the tests will be repeated 3 times. The best value will be recorded.

All measurements will be performed before the surgery (baseline), after the surgery in phase 1 recovery (postanesthetic care unit) and on postoperative day 1, 2, 3 and 4.

The percent fall of vital capacity from sitting to supine position will be considered as an index of diaphragmatic dysfunction.

Inclusion Criteria:

• ASA physical status I-III
• Major shoulder surgery: rotator cuff repair, shoulder arthroplasty

Exclusion Criteria:

• contraindications to brachial plexus block (e.g., allergy to local anesthetics, coagulopathy, malignancy or infection in the area);
• existing neurological deficit in the area to be blocked;
• history of neck surgery or radiotherapy;
• severe respiratory disease;
• chest deformity,
• pregnancy;
• inability to understand the informed consent and demands of the study;
• patient refusal.