Patient-specific instrumentation vs. a free-hand technique for glenoid baseplate and peripheral screw placement in reverse total shoulder arthroplasty using the Exactech implant system: a multicenter randomized controlled trial

Accurate positioning of the glenoid baseplate and peripheral screws is critical in reverse total shoulder arthroplasty (rTSA). Patient-specific instrumentation (PSI) has been introduced to improve surgical accuracy, but its benefits over conventional free-hand techniques remain uncertain. This study aimed to evaluate the accuracy of glenoid baseplate and peripheral screw placement between PSI-assisted and conventional free-hand rTSA. We conducted a prospective, multicenter, parallel-group, single-blinded, superiority randomized controlled trial across 7 tertiary centers between March 2022 and December 2024. Patients aged &#x2265;65 years undergoing primary rTSA were randomized 1:1 to PSI or free-hand group through a centralized web-based electronic case report form. No cross-over occurred and all analyses were performed on an intention-to-treat basis. A standardized implant system (Exactech, Gainesville, FL, USA) was used in all cases. The primary outcome was the proportion of patients with postoperative computed tomography-based deviation error from preoperative planning, including version (&#x2265;5&#xb0;), inclination (&#x2265;5&#xb0;), or position offset (&#x2265;2.5 mm) errors. The secondary outcomes included continuous deviation values and accuracy of peripheral screw placement (anteroposterior [AP] gap, superoinferior (SI) gap, and length gap deviations). Effect sizes with 95% confidence intervals (CIs) were reported. Of 106 patients, 53 were included in the PSI and free-hand groups each. Baseline demographics and clinical characteristics were comparable between the groups. Baseplate positioning was similar between the groups, with version error 5 &#xb1; 5&#xb0; vs. 5 &#xb1; 5&#xb0;, inclination error 6 &#xb1; 5&#xb0; vs. 5 &#xb1; 4&#xb0; (mean difference +0.9&#xb0;, 95% CI -0.8 to +2.6; P = .309), and positional offset 3 &#xb1; 2 mm vs. 4 &#xb1; 2 mm (mean difference -0.3 mm, 95% CI -1.1 to +0.5; P = .496). Conversely, PSI significantly improved peripheral screw placement accuracy, including AP gap (-1 &#xb1; 3&#xb0; vs. 1 &#xb1; 6&#xb0;; mean difference -1.9&#xb0;, 95% CI -3.7 to -0.1; P = .041), SI gap (-2 &#xb1; 4&#xb0; vs. 1 &#xb1; 6&#xb0;; mean difference -2.7&#xb0;, 95% CI -4.7 to -0.8; P = .008), superior length gap (1 &#xb1; 6 mm vs. -2 &#xb1; 8 mm; mean difference +3.0 mm, 95% CI 0.2 to 5.8; P = .039), and inferior AP gap (1 &#xb1; 6&#xb0; vs. -3 &#xb1; 6&#xb0;; mean difference +4.2&#xb0;, 95% CI 2.0 to 6.4; P < .001). Additionally, clinically relevant errors (>5&#xb0;) were significantly less frequent in the PSI group for anterior (11.3% vs. 43.4%; P < .001) and superior (20.8% vs. 41.5%; P = .035) SI gap. In primary rTSA, PSI did not improve glenoid baseplate positioning compared to the free-hand technique but significantly enhanced the accuracy of peripheral screw placement.