The limited value of routine radiographs in the follow-up treatment of proximal humerus fractures.

PURPOSE: Proximal humerus fractures (PHFs) are common injuries in adults and occur particularly often in older patients. In our hospital, patients are routinely scheduled for follow-up radiographs at approximately two and six to eight weeks after injury, although the management yield of this practice is uncertain. METHODS: We performed a retrospective cohort study of consecutive adults (≥ 18 years) presenting with PHFs to the OLVG emergency department (ED) between January and April 2022. Patients with multiple fractures, pathological fractures, fractures initially treated or followed up at another hospital, and fractures other than proximal humerus fractures were excluded. The primary outcome was a documented treatment-plan change during the observed outpatient treatment period, including conversion from conservative to operative treatment, cancellation of planned surgery, or additional surgery after initial operative treatment. The primary outcome proportion was reported with a two-sided exact 95% confidence interval. Descriptive variables included fracture classification, comminution, imaging use, follow-up timing, secondary displacement, and treatment duration. RESULTS: One hundred patients were included (mean age 60.4 ± 18.0 years; 60% female). Most fractures were one-part (63%) or two-part (33%), 45% showed comminution, and 87% of patients were treated conservatively. The observed treatment-plan change rate was 0/100 (0%; exact 95% confidence interval, 0.0-3.6%). Among the 87 patients initially treated conservatively, none converted to operative treatment. Secondary displacement occurred in nine patients, including one radiographic non-union and one malunion, but no treatment-plan change was documented in these patients. No revision surgery was performed. Patients underwent a mean of 3.0 radiographs, including the initial ED radiograph. CONCLUSION: In this cohort, routine follow-up radiographs had a low observed management yield: none was associated with a documented treatment-plan change. Because the upper 95% confidence limit was 3.6% and management reflected practice within a single hospital organisation, these data do not establish the safety of omitting radiographs. They support larger multicentre evaluation of clinically triggered rather than universally scheduled follow-up imaging.