Twelve patients who had undergone costal osteochondral graft reconstruction of the proximal pole of scaphoid were evaluated with clinical examination, patient-reported outcome scores and radiographs with an average follow-up of 10 years (range 3.5–18). The range of wrist motion was not significantly changed compared with the preoperative range of motion and functional outcomes scores were acceptable. The patients reported low pain scores despite the universal presence of radiographic changes of reduced carpal height and arthritis of the midcarpal and radiocarpal joints. Costal osteochondral graft reconstruction of the proximal pole of scaphoid offers good long-term pain relief and function. Level of evidence: IV

Upper brachial plexus injuries to the C5/6 roots or axillary nerve can result in severe deficits in upper limb function. Current techniques to reinnervate the deltoid muscle utilise the well-described transfer of radial nerve branches to triceps to the axillary nerve. However, in around 25% of patients, there is a failure of sufficient deltoid reinnervation. It is unclear in the literature if deltoid reanimation should be attempted with a nerve transfer from a weak but functioning triceps nerve. The authors present the largest series of triceps to axillary nerve transfers for deltoid reanimation in order to answer this clinical question. Seventy-seven consecutive patients of a single surgeon were stratified and analysed in four groups: (1) normal triceps at presentation, (2) abnormal triceps at presentation recovering to clinically normal function preoperatively, (3) abnormal triceps at presentation remaining abnormal preoperatively, and lastly (4) where pre-operative triceps function was deemed insufficient for use, requiring alternative reconstruction for deltoid reanimation. The authors considered deltoid re-animation of ≥ M4 as successful for the purpose of this study. Median Medical Research Council (MRC) values demonstrate group 1 achieves this successfully (M5), while median values for groups 2–4 result in M4 power (albeit with decreasing interquartile ranges). Median post-operative shoulder abduction active range of motion (AROM) values were represented by 170° (85–180) in group 1, 117.5° (97.5–140) in group 2, 90° (35–150) in group 3, and 60° (40–155) in group 4. For both post-operative assessments, subgroup analyses demonstrated statistically significant differences when comparing group 1 with groups 3 and 4 (p < 0.05), while all the other group to group pairwise comparisons did not reach significance. The authors postulated that triceps deficiency can act as a surrogate marker of a more extensive plexus injury and may predict poorer outcomes if the weakness persists representing the trending differences between groups 2 and 3. However, given no statistical differences were demonstrated between groups 3 and 4, the authors conclude that utilising an abnormal triceps nerve that demonstrates sufficient strength and redundancy intraoperatively is preferable to alternative transfers for deltoid reanimation. Lastly, in group 4 patients where triceps nerves are damaged and unusable for nerve transfer, alternative operations can also achieve sufficient outcomes and should be considered for restoration of shoulder abduction.