The purpose of this cadaveric study was to compare standard and modified coracoid transfer procedures, bicortical and tricortical iliac crest autografts, and tibial plafond and glenoid allografts with respect to glenoid surface curvature restoration. Computed tomography scans of 8 cadaveric shoulders were acquired in 9 conditions: (1) intact, (2) 25% width defect, (3) classic Latarjet, (4) modified congruent-arc Latarjet, (5) tricortical iliac crest inner table, (6) outer table, (7) bicortical iliac crest, (8) distal tibia, and (9) glenoid allograft. Outcome measures included articular surface area, width, depth, axial and coronal radius of curvature, and subchondral articular step-off, analyzed in bone and soft-tissue window. Reconstruction of the articular surface area was optimal with the glenoid allograft (99.4%), classic Latarjet (97.4%), and iliac crest bicortical graft (93.2%). Depth was best restored by the congruent-arc Latarjet (101.0%), tibial (98.9%), and glenoid (95.3%) allografts. Axial curvature was closely matched by the glenoid allograft (97.5%), classic Latarjet (108.7%), and iliac bicortical graft (91.2%). Coronal curvature was most accurately restored by the glenoid allograft (102.6%), the tibial allograft (115.0%), and the classic Latarjet (55.9%). The articular step-off was smallest using the glenoid allograft. Overall, glenoid allografts most accurately restored articular geometry. Alternative grafts provided restoration of some parameters but not others. Classic Latarjet performed well in axial and coronal curvature on average but exhibited large variability. Tibial allograft produced the poorest results in axial curvature, despite excellent coronal curvature reconstruction. The congruent-arc Latarjet did not restore the axial curvature accurately and overcorrected coronal curvature. Graft geometry must be weighed against availability, morbidity, and the role of additional stabilizers. Accurate graft morphology may help prevent postoperative osteoarthritis. Grafts differ significantly regarding geometric parameters. The findings of this study will help surgeons select the most appropriate graft for glenoid reconstruction.
Read full abstract