Introduction. Osteoarthritis of the glenohumeral joint leads to global degeneration of the shoulder and often results in humeral or glenoid osteophytes. It is established that the axillary neurovascular bundle is in close proximity to the glenohumeral capsule. Similar to other compressive neuropathies, osteophytic impingement of the axillary nerve could result in axillary nerve symptoms. The purpose of this study was to compare the proximity of the axillary neurovascular bundle to the inferior humerus in shoulders to determine distance of the neurovascular bundle as the osteophyte (goat's beard) of glenohumeral osteoarthritis develops. Methods. In this IRB approved study, preoperative MRI's of 98 shoulders (89 patients) with primary osteoarthritis (OA group) were compared to 91 shoulders (86 patients) with anterior instability (Control group). For MRI measurements (mm) two coronal-oblique T1 or proton density weighted images were selected for each patient located at 5 and 6 o'clock position of the glenoid in the parasagittal plane. Humeral head diameter to standardize the glenohumeral measurements, size of the spurs, and 6 measurements between osseus structures and axillary neurovascular bundle were obtained on each image using a calibrated measurement system (Stryker Office PACS Power Viewer). Level of significance was set at p>.05. Results. Since results were both significant at 5 and 6 o'clock, for clarity we will only give the results at 6 o'clock. Humeral head osteophytes were present in 52% (51/86) of arthritic patients with an average size of 9.90 mm (range 0–24.31). Distance between humeral head or inferior osteophyte and neurovascular bundle was significantly decreased (p<0.05) in the OA group, 19.74 mm (range 2.80–35.12) compared to the control group 23.8 mm (14.25–31.89). If we compare the same distance between the Control group, OA group with a spur and OA group without a spur, the difference is only significant between the Control group and OA with spur. (p<0.05) In non-arthritic patients, the neurovascular bundle was closest to the inferior glenoid rim in all patients (91/91). By contrast, the neurovascular bundle was closest to the humeral head in 26.5% (26/98) of arthritic patients. Among these 26 patients, a large humeral head osteophyte was present in 96% (25/26). The neurovascular bundle distance and humeral head osteophyte size were inversely correlated (r=−0.45 at 5 o'clock, r=−0.546 at 6 o'clock) in the arthritic group (p<0.05). Discussion. The axillary neurovascular bundle was significantly closer to the osseous structures (humerus) in arthritic patients compared to non-arthritic patients (p<0.05). The neurovascular bundle was significantly closer to the bone when there was a humeral osteophyte, and the distance was inversely proportional to humeral osteophyte size (p<0.05). This study indicates humeral osteophytes are capable of encroaching on the axillary nerve. Axillary nerve entrapment may be a contributing and treatable factor of pain in patients with glenohumeral osteoarthritis

Aims

The purpose of this study was to develop a convolutional neural network (CNN) for fracture detection, classification, and identification of greater tuberosity displacement ≥ 1 cm, neck-shaft angle (NSA) ≤ 100°, shaft translation, and articular fracture involvement, on plain radiographs.