A prospective clinical study of proximal humerus fractures treated with a locking proximal humerus plate.

To the Editor: We read with interest the article ‘‘A prospective clinical study of proximal humerus fractures treated with a locking proximal humerus plate’’ by Huiguang Yang et al. In this article, the authors have done a commendable job in conducting a quality prospective study that provides early information on the effectiveness, complications, and the predictive factors for the failure of these implants. However, we have a few concerns regarding this study and seek further clarification. Our first concern regards the intraoperative technique adopted by the authors. In all their cases, the authors found it necessary to elevate the humeral head and reduce it using a lateral cortical window, followed by filling the defect with cancellous chips or calcium phosphate cement. This technique seems a bit aggressive, as we believe it is uncommon for the articular surface of the humeral head to undergo such a significant impaction as to demand this procedure. Also, it is unclear from the text whether the calcium phosphate cement was used to support a deficient medial cortex. Contrary to the authors’ opinion, Gardner et al have suggested futility of medial cement augmentation in maintaining the humeral head height. Furthermore, there is no indication of whether the authors employed any additional fixation methods in cases of associated greater tuberosity fractures (tension band wiring and/or transosseous sutures). A malreduction or inadequate fixation of these osseous appendages can potentially impair the function of the rotator cuff tendons and thereby have a negative impact on the final functional outcome. Second, the authors report poorer functional outcome and lower Constant scores in the inadequate medial support group. Gardner at al suggested 3 situations that lead to lack of medial support after proximal humerus stabilization. We would have appreciated more details on the situations that led to deficiency in medial reinforcement in the 30 patients in whom this was reported. Such information would assist readers in evaluating the effectiveness of techniques like medial shift of the humeral shaft or additional inferomedial screws in cases where there is medial insufficiency. Third, there is lack of details on some of the factors that are believed to influence shoulder outcomes. The incidence of osteoporosis in patients with failed procedures is not provided. Also, we would have liked to see inclusion of an assessment of the influence of significant associated traumatic rotator cuff injuries on the functional outcome achieved and specific complications encountereddimpingement, rotator cuff rupture, frozen shoulder, and the like. In one of our previously published studies, we found that a more cranial initial placement of the locking plate on the humeral head resulted in a significant increase in the occurrence of postoperative subacromial impingement. Outcomes were also found to be negatively influenced (significantly) by the initial varus malalignment (denoted by the head-shaft angle < 120 degrees). Another significant factor influencing outcomes was surgeon experience. We would have appreciated more information on these factors as well. Finally, the authors have chosen to apply locking plates in 24 of their Neer’s type 4 fractures and replace the humeral head in 4 patients. The indications for either of these management modalities were not mentioned by the authors but would have been helpful. Nevertheless, the appeal of the article remains unchallenged. We believe that this study will pioneer further prospective trials on proximal humeral locking plates, paving the way for improved biomechanical understanding and technical advancements.