The impact of 3-dimensional humeral planning and standard transfer instrumentation on reconstruction of native humeral anatomy for anatomic total shoulder arthroplasty.

ABSTRACT

BACKGROUND: Preoperative 3-dimensional (3D) computed tomography (CT)-based planning for anatomic total shoulder arthroplasty (TSA) has grown in popularity in the past decade with the primary focus on the glenoid. Little research has evaluated if humeral planning has any effect on the surgical execution of the humeral cut or the positioning of the prosthesis. METHODS: Three surgeons performed a prospective study using 3D-printed humeri printed from CTs of existing patients, which were chosen to be -3, -1, 0, 1, and 3 standard deviations of all patients in a large database. A novel 3D printing process was used to 3D print not only the humerus but also all 4 rotator cuff tendons. For each surgical procedure, the printed humerus was mounted inside a silicone shoulder, with printed musculature and skin, and with tensions similar to human tissue requiring standard retraction and instruments to expose the humerus. Three phases of the study were designed. In phase 1, humeral neck cuts were performed on all specimens without any preoperative humeral planning; in phase 2, 3D planning was performed, and the cuts and implant selection were repeated; in phase 3, a neck-shaft angle (NSA) guide and digital calipers were used to measure humeral osteotomy thickness to aid in the desired humeral cut. All humeri were digitized. The difference between the prosthetic center of rotation (COR) and ideal COR was calculated. The percentage of patients with a varus NSA was calculated for each phase. The difference in planned and actual cut thickness was also compared. RESULTS: For both 3D change in COR and medial to lateral change in COR, use of preoperative planning alone and with standard transfer instrumentation resulted in a significantly more anatomic restoration of ideal COR. The deviations from planned cut thickness decreased with each phase phase 1 2.6 ± 1.9 mm, phase 2 2.0 ± 1.3 mm, phase 3 1.4 ± 0.9 mm (P = .041 for phase 3 vs. phase 1). For NSA, in phase 1, 7 of 15 (47%) cases were in varus; in phase 2, 5 of 15 (33%) were in varus; and in phase 3, 1 of 15 (7%) cases was in varus (P = .013 for phase 3 vs. phase 1). CONCLUSIONS: Use of preoperative 3D humeral planning for stemless anatomic TSA improved prosthetic humeral COR, whether performed with or without standard transfer instrumentation. The use of an NSA cut guide and calipers to measure cut thickness significantly reduced the percentage of varus humeral cuts and deviation from planned cut thickness.