WALANT technique versus general anesthesia in frozen shoulder manipulation: a prospective randomized controlled trial

Abstract

Manipulation under anesthesia can rapidly improve pain and motion in refractory frozen shoulder but usually requires operating-room and anesthesiology resources. This trial compared an ultrasound-guided wide-awake local anesthesia no tourniquet (WALANT)-based outpatient protocol with manipulation under general anesthesia. This single-center, prospective randomized controlled trial enrolled 72 adults with primary frozen shoulder persisting despite at least 6 months of conservative treatment. Participants were randomized 1:1 to ultrasound-guided WALANT-based manipulation in an outpatient setting or manipulation under general anesthesia in the operating room. The primary outcome was shoulder forward flexion at 4 weeks. Secondary outcomes included other range-of-motion measures, the visual analog scale (VAS), University of California Los Angeles (UCLA) Shoulder, Disabilities of the Arm, Shoulder and Hand (DASH), Constant–Murley, and six-item State-Trait Anxiety Inventory (STAI-6) scores, and early complications. Six participants in the general-anesthesia group did not complete follow-up; complete-case analyses included 36 WALANT and 30 general-anesthesia participants. At the prespecified 4 week primary endpoint, forward flexion was 170.56° ± 15.01° in the WALANT group and 159.60° ± 31.37° in the general-anesthesia group (between-group mean difference, 10.96°; 95% CI, − 0.83–22.74°; p = 0.068). Exploratory analyses favored WALANT for forward flexion at week 3 ( p = 0.040), internal rotation at week 1 ( p = 0.002), pain at weeks 2 and 3 ( p = 0.005 and p = 0.035), and Constant–Murley scores at weeks 2 and 3 ( p = 0.007 and p = 0.039). These differences were no longer evident at week 4. UCLA Shoulder and DASH scores did not differ between groups. Preprocedural anxiety was lower in the WALANT group ( p = 0.007). No procedure-related complications, infections, early recurrences, or reinterventions were observed. No statistically significant between-group difference was observed in forward flexion at the prespecified 4-week primary endpoint. Nevertheless, the WALANT-based outpatient protocol was feasible and achieved short-term outcomes comparable to those of manipulation under general anesthesia, with transient early advantages in pain, function, anxiety, and selected range-of-motion measures. Larger multicenter trials with longer follow-up are needed to confirm these findings. ClinicalTrials.gov, NCT07731646, registered on July 23, 2026. Retrospectively registered.