Accuracy and Precision of Planned Glenoid Guide-Pin Placement in Reverse Total Shoulder Arthroplasty: a Cadaveric Comparison of Fluoroscopy-Guided Freehand Technique, Patient-Specific Instrumentation, and Navigation

Background This study aimed to compare the accuracy and precision of CT-based planned central glenoid guide-pin placement among fluoroscopy-guided freehand insertion, CT-based patient-specific instrumentation (PSI), and CT-based navigation in a cadaveric reverse total shoulder arthroplasty (rTSA) model. Methods Six cadaveric shoulders were used. Three surgeons performed all three techniques in each shoulder (18 insertions per technique) under a shared preoperative plan. The sequence was fixed: fluoroscopy-guided freehand insertion, PSI, then navigation. After each insertion, guide-pin angle and entry-point position were measured by an independent assessor using the same navigation platform; recorded measurement values were not disclosed to the operating surgeon. Before the next insertion, the previous pin track was filled and visually/tactilely masked. Accuracy was assessed as deviation from the plan, and precision as the dispersion of deviations and the frequency of threshold-defined outliers. Overall successful placement was defined as angular deviations within ±5° in both inclination and retroversion and an entry-point offset magnitude ≤2.0 mm. Overall outlier status was defined as >10° deviation in either angular axis and/or an entry-point offset magnitude >4.0 mm. Results Resultant angular deviation was 6.8° (SD 4.2) for fluoroscopy, 5.7° (SD 4.6) for PSI, and 2.3° (SD 1.5) for navigation (p = 0.002). Entry-point offset magnitude was 3.2 mm (SD 1.3), 2.2 mm (SD 1.2), and 1.3 mm (SD 0.5), respectively (p < 0.001). Navigation showed smaller angular and positional deviations than both fluoroscopy and PSI. PSI improved entry-point offset magnitude versus fluoroscopy (p = 0.021) but not resultant angular deviation (p = 0.384). Overall successful placement occurred in 4/18 fluoroscopy insertions, 7/18 PSI insertions, and 17/18 navigation insertions (p < 0.001). Overall outliers occurred in 7/18, 4/18, and 0/18 insertions, respectively (p = 0.030). Conclusion In this fixed-sequence cadaveric rTSA model, navigation showed the smallest angular and entry-point deviations relative to the preoperative plan, and no threshold-defined outliers were observed. PSI improved entry-point accuracy compared with fluoroscopy-assisted freehand insertion but did not significantly improve angular accuracy. These findings characterize plan-to-pin execution at the guide-pin stage and require confirmation in studies evaluating final implant position and clinical outcomes. Level of Evidence Basic Science Study; Validation of Surgical Technique; Cadaver Model