Low glenoid vault bone density is a risk factor for failure of proximal humerus fracture fixation

We aimed to determine whether computed tomography Hounsfield units (CTHU) are an independent predictor of proximal humerus fracture fixation failure following open reduction and internal fixation (ORIF). We hypothesize that CTHU of the glenoid vault <160 Hounsfield units (HU) would be associated with a higher rate of failure of ORIF and need for subsequent revision to arthroplasty.

A retrospective cohort study was performed at a single academic center, including patients who sustained a displaced proximal humerus fracture treated with ORIF using locking plates over a 20-year period. A minimum follow-up of 6 months was required. Exclusion criteria included concomitant glenoid fractures, humeral shaft fractures, primary bone malignancies, metastatic disease, or lack of preoperative CT scan. CTHU were measured at 4 distinct regions with the glenoid vault to assess bone density. Patients were categorized into those requiring subsequent revision to arthroplasty and those who did not. Statistical analyses were performed using standard tests for categorical and continuous variables, with significance set at P <.05.

A total of 135 patients were identified, with 95 patients who successfully underwent ORIF and 40 who required revision to arthroplasty. Patients who required revision to arthroplasty had lower mean CTHU than those without revision (110.9 ± 42.6 vs. 170.5 ± 45.9, P < .001). Patients with 2-part (166.9 ± 44.9 vs 113.9 ± 43.3, P < .0001) and 3-part (175.7 + 47.3 vs 118.4 + 54.0) proximal humerus fractures who underwent revision arthroplasty had significantly decreased mean CTHU than patients who did not undergo revision. CTHU of the glenoid vault <160 HU significantly increased the risk of revision to arthroplasty after ORIF (P < .001, relative risk 3.00, 95% confidence interval 1.28-7.02). Within the failed cohort who required revision to arthroplasty, there was no difference in CTHU between those who failed within 6 months or after 6 months from ORIF (92.9 ± 37.7 vs. 134.3 ± 50.9, P = .057); there was a significant difference in revision rates of 2-part proximal humerus fractures with CTHU <160 than those with CTHU >160 (P = .004). CTHU of the glenoid vault were found to be positively correlated to the deltoid tuberosity index (R = 0.72, P < .001) and combined cortical thickness of the humerus (R = 0.21, P = .02).

CTHU of the glenoid vault <160 HU is an independent risk factor for ORIF failure in patients with displaced proximal humerus fractures, leading to a higher likelihood of requiring revision to arthroplasty. This opportunistic measurement CTHU can assist surgeons in informed decision making between ORIF and primary arthroplasty in the management of these fractures.