SAFE (Subarachnoid-Alternative Anaesthesia for Endoprosthesis): A Motor-Sparing and Opioid-Sparing Anesthetic...

Background: Anesthetizing frail patients for hip surgery is challenging; spinal (SA) and general anesthesia (GA) often cause hemodynamic instability. Traditional nerve blocks provide analgesia but rarely complete surgical anesthesia without motor block. We evaluate the clinical feasibility of the SAFE (Subarachnoid-alternative Anaesthesia For Endoprosthesis) protocol-combining Anterior Pericapsular Nerve Group (A-PENG), POsterior pericapsular Nerve Group (PONG), and Local Infiltration Analgesia (LIA) under intravenous sedation-as a primary anesthetic preserving motor function and avoiding SA/GA. Methods: This single-center retrospective series analyzed patients undergoing elective or trauma-related hip surgery using the SAFE protocol between September 2022 and April 2026. The primary outcome was success rate (completion without SA/GA conversion). Secondary outcomes included procedural timings, recovery room (RR) transit, and motor preservation. Variables are reported as medians [IQR]. Results: We included 48 patients (median age 83.5 years [IQR: 68.7-87.2]; 66.7% female) undergoing hip hemiarthroplasty (n = 28) or total hip arthroplasty (n = 20). The success rate was 100%, without SA/GA conversion or advanced airway management. Median anesthetic preparation and surgical durations were 55 [IQR: 50-76.2] and 85 min [IQR: 74-110], respectively. RR transit times (recorded for 35 patients) were brief (40 min [IQR: 34.0-67.5]). Crucially, lower-limb motor capacity was preserved in 100% of cases. The technique also proved opioid-sparing, substantially reducing postoperative opioid consumption. Conclusions: The SAFE protocol is a clinically feasible primary anesthetic strategy for hip surgery. By preserving motor function and enabling rapid fast-tracking, it aligns with ERAS pathways, offering a promising alternative to conventional anesthesia for elective and frail trauma patients. Randomized controlled trials are warranted to validate these outcomes.