Shoulder arthroscopy is one of the most common orthopedic procedures, and the quality of the surgeon's view inside the joint is critical to its safety and success. Unlike limb surgery, no tourniquet can be applied to the shoulder, so surgical visibility depends almost entirely on how much bleeding occurs within the joint - which is itself influenced by the anesthetic drugs used to keep the patient asleep.

Two standard techniques exist for maintaining general anesthesia: inhaled anesthesia (sevoflurane) and total intravenous anesthesia (TIVA, using propofol). Evidence from other types of keyhole surgery (nasal, ear) suggests that propofol may produce better surgical visibility, possibly because sevoflurane causes slightly greater dilation of the smallest blood vessels in tissue, leading to more bleeding into the joint. However, in shoulder arthroscopy specifically, the data are scarce and contradictory.

CLEAR-SHOULDER is a randomized, single-blind trial designed to determine whether propofol-based intravenous anesthesia provides superior intraoperative visibility compared to sevoflurane-based inhaled anesthesia during arthroscopic rotator cuff repair. Surgical visibility will be assessed from blinded video recordings of each procedure using the validated modified Fromme-Boezaart score. An exploratory artificial intelligence analysis of the arthroscopic videos will also be conducted.

Secondary outcomes include operating time, irrigation fluid consumption, hemodynamic stability, vasopressor requirements, surgeon satisfaction, postoperative recovery, and environmental impact of each anesthetic strategy.

Inclusion Criteria:

1. Patient aged 18 or over
2. Patient scheduled for elective rotator cuff repair arthroscopy (RCRA), performed in the "beach-chair" position under general anesthesia combined with an interscalene block (with or without standard associated procedures such as acromioplasty and/or tenotomy/tenodesis of the long biceps). Procedures for instability (Bankart/Latarjet), fractures, capsulitis, simple lavage, or any arthroscopy not focused on rotator cuff repair (RCR) are not eligible.
3. Patient with an ASA score of I to IV
4. Patient affiliated to a social security scheme,
5. Patient having given written consent following written and oral information.

Exclusion Criteria:

1. Patients with a contraindication to any of the study agents, including known hypersensitivity or allergy to any of these agents or to any of their excipients.
2. Patients with known coagulopathy (abnormal PT/INR; platelets < 100,000/L) or
3. Patients with active preoperative hemarthrosis or local infection.
4. Non-elective (emergency) surgery or a procedure whose initial objective is not rotator cuff repair arthroscopy (RCRA). Note: Intraoperative conversion to open surgery is not an exclusion criterion; it will be considered an intraoperative event and managed in the analysis according to the intention-to-treat (ITT) principle, with a sensitivity analysis procedure described in the statistical plan.
5. Uncooperative patient, inability to consent
6. Pregnant women
7. Patient protected by law under guardianship or curatorship, or unable to participate in a clinical study under article L. 1121-16 of the French Public Health Code.