Intramedullary Allograft Augmentation of Plated Unstable Proximal Humerus Fractures: Biomechanical and Clinical Findings

This study evaluated the biomechanical competence of locked plating (LP) with intramedullary graft (IMG) augmentation and its clinical application for treating unstable proximal humerus fractures (PHFs). An unstable four-part PHF was simulated in 30 artificial humeri assigned into three groups (n=10): Group 1 had medial column insufficiency, Group 2 had severe cancellous bone loss, and Group 3 combined both features. Quasi-static axial loading (150–400 N, 50-N increments) was performed with X-ray imaging, followed by retesting after inserting a 3D-printed intramedullary graft. Forty-seven patients with unstable PHFs treated with LP and IMG were retrospectively evaluated with a minimum 12-month follow-up. Radiological evaluation included changes in neck-shaft angle (NSA) and humeral head height (HHH). Local bone quality was assessed using the deltoid tuberosity index (DTI). The three most reliable Hertel's criteria for predicting humeral head AVN were analyzed. Fracture reduction was categorized as anatomical, acceptable, or malreduced. Patients were classified as non-geriatric (<65 years) or geriatric (≥65 years). Functional outcomes were measured using DASH and relative Constant-Murley Score (CS rel ). The biomechanical experiment showed that grafting significantly increased axial stiffness in Groups 1 and 3 (P ≤ 0.01), but not in Group 2, and reduced varus deformation in all groups (P ≤ 0.04). Nongrafted specimens in Group 2 had higher stiffness and lower varus deformation than Groups 1 and 3. Clinical results showed mean follow-up of 28.5 months, with a mean age of 63.2 years. 28 patients (59.6%) had four-part fractures, 36 had calcar comminution, 27 had initial varus displacement, and 29 had DTI<1.4. Twenty-two patients had the three Hetel's criteria combined. Functional outcomes showed a CS rel of 73.4 and DASH of 16.6. NSA decreased by 7.98° and HHH by 0.96 mm. The most common complication was humeral head AVN (13 patients). Younger patients and those with anatomical/acceptable reductions had better outcomes, while bone quality had no impact on radiological or functional results. From a biomechanical standpoint, LP combined with an IMG enhances stability against varus collapse in unstable PHFs, compared to standard LP fixation. From a clinical perspective LP supplemented with IMG is a reliable treatment method for unstable PHFs, only when good fracture reduction is achieved. The most common complication was AVN and it was also dependent on fracture reduction. Our biomechanical and clinical results demonstrated that osteoporosis did not significantly affect construct stability or clinical outcomes.