An X-ray of a proximal humerus fracture can be difficult to accept at first glance: the fragment is displaced, the bone is angulated, and it is natural to wonder how it could return to alignment without an operation.
In children, however, this is one of the sites where the initial X-ray needs to be interpreted alongside age. The proximal humeral growth plate contributes approximately 80% of the bone’s longitudinal growth, and its remodelling potential is very high. A deformity that may be acceptable at age 6 is not necessarily acceptable at age 15. [1]
A new study published in Emergency Radiology analysed 418 proximal humerus fractures in patients younger than 18 years. Only 42, or 10%, underwent surgery. The authors investigated which characteristics were associated with the decision to operate. [1]
The strongest factor was displacement: translation greater than 42% of the shaft width was associated with surgery, with an odds ratio of 13.3. Age over 12 years and an injury mechanism other than a simple fall were also independently associated. Angulation, in contrast, was no longer significant in the multivariable model. [1]
This last finding is interesting because the fracture angle is often what most catches the eye on an X-ray. The authors point out, however, that the measured angulation can change with arm position and the radiographic projection. [1]
The 42% figure is not a new threshold for surgery
This is the part of the study most likely to be misinterpreted.
The 42% figure was identified by ROC analysis as the value that best distinguished surgically treated from nonoperatively treated fractures in this series. The study did not compare a group of children above that threshold who had surgery with a similar group treated nonoperatively. Above all, it did not assess long-term outcomes. [1]
In other words, the study tells us which characteristics were associated with the surgeons’ decisions, not which treatment was better.
Turning the finding into a rule such as “displacement above 42% requires surgery” would therefore be wrong.
The series itself includes a detail that illustrates the problem: four of the 42 surgically treated fractures were Neer–Horowitz grade I, meaning they were minimally displaced. All involved older patients, aged 13.4–17.4 years, and all occurred during sporting or recreational activities. [1]
This does not mean that sport is an indication for surgery. It means that one radiographic measurement does not tell the whole story.
Why age and remaining growth matter so much
Earlier literature reaches broadly the same conclusion. A 2023 meta-analysis that included eight studies found a significant association between operative treatment, older age and severely displaced fractures. Overall, 33% of the included patients underwent surgery, rising to 60% among Neer–Horowitz grade III–IV fractures. [2]
Again, however, this describes the strategies used in the available studies rather than a universal treatment threshold.
In young children, nonoperative treatment therefore remains the most common choice, even when the initial X-ray does not look particularly reassuring. As skeletal maturity approaches, the margin narrows: less growth remains to correct residual deformity, and a severely displaced fracture is more likely to prompt discussion of surgery. [1,2]
When surgery is indicated, the most commonly used techniques are percutaneous pinning and elastic intramedullary nailing. The 2023 meta-analysis found no statistically significant preference between these two fixation methods. [2]
What I would take into clinical practice
The most useful finding of the new study is not a percentage to turn into an indication for surgery. It is a prompt to describe the fracture more precisely.
Measuring displacement relative to the shaft width, assessing age and growth plate status, considering the injury mechanism, and looking beyond the visual impression of angulation alone can make the reasoning clearer. [1]
There are nevertheless important limitations. This was a retrospective, single-centre study at a paediatric referral hospital. The X-rays were not standardised; the authors did not formally assess interobserver reproducibility of the proposed threshold, and they did not analyse long-term clinical outcomes. [1]
To determine whether a given degree of displacement truly identifies children who do better with surgery, we need studies designed around outcomes, rather than treatment decisions that have already been made.
For now, the practical message remains the one that probably matters most to a family: most proximal humerus fractures in children do not require surgery, but the same displacement does not mean the same thing in a young child and an adolescent approaching the end of growth.
References
[1] Forero-Millan J, Lerebo WT, Patel VS, Fanney L, Arkader A, Nguyen JC. Clinical and radiographic findings that impact treatment: a study of 418 proximal humerus fractures. Emerg Radiol. Published online September 1, 2026. doi:10.1007/s10140-026-02534-7.
[2] Song HR, Song MH. Operative Versus Nonoperative Management of Pediatric Proximal Humerus Fractures: A Meta-Analysis and Systematic Review. Clin Orthop Surg. 2023;15(6):1022-1028. PMID: 38045578. PMCID: PMC10689228. doi:10.4055/cios23077.
Disclaimer: this content provides general information. It does not replace an individual medical assessment.
