Purpose. Assess and report the functional and post-operative outcomes of complex acute radial head fractures with elbow instability treated by arthroplasty using an uncemented modular anatomic prosthesis. Methods. Over a 3-year period (2007–2010), 21 patients (mean age 51.9 years) were treated primarily with modular radial head arthroplasty (mean follow up of 27.1 months). Data was collected retrospectively using clinical notes, operation documentation and prospectively using validated scoring systems namely the Oxford Elbow Index, Quick DASH and the Mayo Elbow Performance Score. Associated elbow fractures, ligamentous injury and short to mid term post-operative outcomes including radiographic assessment were recorded. Results. The mean Oxford Elbow Score was 34.80 (range 20–48). The mean Quick Dash score was 26.01 (range 0–68.2). The Mayo Performance score showed 6 scored excellent, 5 scored good, 3 scored fair and 2 scored poor. Regarding post-operative outcomes, 1 patient had a radial head dislocation, 1 patient had prosthesis removal for ongoing pain and 1 patient had a total elbow replacement due to associated proximal ulna fracture non-union. 11 patients had an associated ligamentous injury of which 6 had an associated coronoid fracture. Of note, 7 patient's radiographs showed early signs of implant loosening; this was mainly asymptomatic. Conclusions. With regard to complex radial head fractures with elbow instability, patient outcome measures showed good functionality and overall patient satisfaction despite radiographic evidence of loosening. Post-operative complication rates were low. These findings support the use of this radial head prosthesis in arthoplasty surgery for the treatment of complex acute radial head fractures with elbow instability

Postero-lateral rotator instability (PLRI) is the most common pattern of recurrent elbow instability. Unfortunately, current imaging to aid PLRI diagnosis is limited. We have developed an ultrasound (US) technique to measure ulnohumeral joint gap with and without stress of the lateral ulnocollateral ligament. We sought to define lateral ulnohumeral joint gap measurements in the resting and stressed state to provide insight into how US may aid diagnosis of PLRI. Sixteen elbows were evaluated in eight healthy volunteers. Lateral ulnohumeral gap was measured on US in the resting position and with posterolateral drawer stress test maneuver applied. Joint laxity was calculated as the difference between stress and rest conditions. Measurements were performed by two independent readers with comparison performed between stress and rest positions. A highly significant difference in ulnohumeral gap was seen between stress and rest conditions (Reader 1: p < 0 .0001 and Reader 2: p=0.0002) with median values of 2.93 mm and 2.50 mm at rest and 3.92 mm and 3.40 mm at stress for Reader 1 and 2 respectively. Median joint laxity was 1.02 mm and 0.74 mm respectively for each reader. Correlation and agreement between readers was good. This study provides key new insight into use of US for diagnosis as PLRI as it defines normal ulnohumeral distances and demonstrates widening when applying the posterolateral drawer stress maneuver. Further evaluation of this technique is required in patients with PLRI

In the setting of traumatic elbow injuries involving coronoid fractures, the relative size of the coronoid fragment has been shown to relate to the stability of the joint. Currently, the challenge lies in accurately classifying the amount of bone loss in coronoid fractures. In comminuted fractures, bone loss is difficult to measure with plain radiographs or computed tomography. The purpose of this study is to describe a novel radiographic measure, the Coronoid Opening Angle (COA), on lateral elbow radiographs. We demonstrate the relationship of the COA to coronoid height and describe how this measure can be used to estimate bone loss and potentially predict elbow instability following coronoid fracture. Radiographs were drawn from a regional database in a consecutive fashion. Candidate radiographs were excluded on the basis of radiographic evidence of degenerative changes, previous surgery or injury, bony deformity, and inadequate lateral view of the elbow. The COA was measured as the angle between the long axis of the ulna at the level of the trochlear notch, and the tip of coronoid, from a common origin at the posterior cortex of the olecranon. Images were reviewed by a fellowship trained upper extremity surgeon, an upper extremity fellow, and a junior resident. Normal COA, coronoid height, and calculated COA at varying amounts of bone loss were calculated by three reviewers. A sensitivity analysis was performed to determine how the COA can most effectively predict bone loss at varying coronoid heights. Intraclass correlation coefficient (ICC) was calculated for 39 subjects. Seventy-two subjects were included for analysis (M=40, F=32). The normal coronoid opening angle is 33.19 degrees [32.2 – 34.2]. Coronoid height is 18.8 mm [18.1 – 19.6]. Extrapolating this baseline data, the COA at 20%, 33%, and 50% of coronoid bone loss was calculated to be 27.5, 23.5, and 18 degrees, respectively. ICC was found to be 0.90 or higher. Cutoff values were determined to maximize the sensitivity of the COA. A cutoff value of 21 degrees has a 92% sensitivity in detecting a minimum of 50% bone loss. The COA with similar sensitivity in predicting 20% and 33% bone loss are 32 and 27 degrees. The coronoid opening angle is a novel technique that can be used on a lateral elbow radiograph to predict the minimum coronoid bone loss. This can be used to guide clinical decision making and potentially predict instability. Future research will aim to validate this tool in the clinical setting in predicting instability

The role of anconeus in elbow stability has been a long-standing debate. Anatomical and electromyographic studies have suggested a potential role as a stabilizer. However, to our knowledge, no clinical or biomechanical studies have investigated its role in improving the stability of a lateral collateral ligament (LCL) deficient elbow. Seven cadaveric upper extremities were mounted in an elbow motion simulator in the varus position. An LCL injured model was created by sectioning of the common extensor origin, and the LCL. The anconeus tendon and its aponeurosis were sutured in a Krackow fashion and tensioned to 10N and 20N through a transosseous tunnel at its origin. Varus-valgus angles and ulnohumeral rotations were recorded using an electromagnetic tracking system during simulated active elbow flexion with the forearm pronated and supinated. During active motion, the injured model resulted in a significant increase in varus angulation (5.3°±2.9°, P=.0001 pronation, 3.5°±3.4°, P=.001 supination) and external rotation (ER) (8.6°±5.8°, P=.001 pronation, 7.1°±6.1°, P=.003 supination) of the ulnohumeral articulation compared to the control state (varus angle −2.8°±3.4° pronation, −3.3°±3.2° supination, ER angle 2.1°±5.6° pronation, 1.6°±5.8° supination). Tensioning of the anconeus significantly decreased the varus angulation (−1.2°±4.5°, P=.006 for 10N in pronation, −3.9°±4°, P=.0001 for 20N in pronation, −4.3°±4°, P=.0001 for 10N in supination, −5.3°±4.2°, P=.0001 for 20N in supination) and ER angle (2.6°±4.5°, P=.008 for 10N in pronation, 0.3°±5°, P=.0001 for 20N in pronation, 0.1°±5.3°, P=.0001 for 10N in supination, −0.8°±5.3°, P=.0001 for 20N in supination) of the injured elbow. Comparing anconeus tensioning to the control state, there was no significant difference in varus-valgus angulation except with anconeus tensioning to 20N with the forearm in supination which resulted in less varus angulation (P=1 for 10N in pronation, P=.267 for 20N in pronation, P=.604 for 10N in supination, P=.030 for 20N in supination). Although there were statistically significant differences in ulnohumeral rotation between anconeus tensioning and the control state (except with anconeus tensioning to 10N with the forearm in pronation which was not significantly different), anconeus tensioning resulted in decreased external rotation angle compared to the control state (P=1 for 10N in pronation, P=.020 for 20N in pronation, P=.033 for 10N in supination, P=.001 for 20N in supination). In the highly unstable varus elbow orientation, anconeus tensioning restores the in vitro stability of an LCL deficient elbow during simulated active motion with the forearm in both pronation and supination. Interestingly, there was a significant difference in varus-valgus angulation between 20N anconeus tensioning with the forearm supinated and the control state, with less varus angulation for the anconeus tensioning which suggests that loads less than 20N is sufficient to restore varus stability during active motion with the forearm supinated. Similarly, the significant difference observed in ulnohumeral rotation between anconeus tensioning and the control state suggests that lesser degrees of anconeus tensioning would be sufficient to restore the posterolateral instability of an LCL deficient elbow. These results may have several clinical implications such as a potential role for anconeus strengthening in managing symptomatic lateral elbow instability