Podcast: Play in new window

Subscribe: RSS

Reference:  Perry DC, et al. Surgical fixation versus non-surgical care for children with a displaced medial epicondyle fracture of the elbow (the SCIENCE study): a multicentre, randomised controlled, superiority trial and economic evaluation. The Lancet 2026

Date: July 14, 2026

Guest Skeptics: Dr. Megan Terle is an orthopedic surgeon and Assistant Professor for Pediatrics and Trauma services at UC Davis Health. She completed her orthopedics residency at the University of Washington and completed two fellowships, one in pediatrics at the University of Washington and one in trauma at UC Davis.

Dr. Megan Terle

Dr. Holly Leshikar is also an orthopedic surgeon and Associate Professor for Pediatrics and Trauma services at UC Davis Health. She completed orthopedics residency at UC Davis and a pediatric orthopedic surgery fellowship at Vanderbilt Children’s Hospital in Nashville.

Case: An 11-year-old gymnast comes to the emergency department (ED) after a fall from the bars. She felt a “pop” in the elbow and now refuses to move the arm. The elbow is swollen and tender on your examination. You perform a careful neurovascular exam and note intact distal perfusion and no neurological deficits. You temporarily immobilize the arm and order radiographs.

X-ray shows a displaced medial epicondyle fracture. Her parent asks you: “Does she need surgery to put the bone back where it belongs?”

Dr. Holly Leshikar

Background: Medial epicondyle fractures are pediatric elbow injuries involving the bony prominence on the inside of the distal humerus. In children, this area is a secondary growth center. The typical mechanism is an avulsion injury from valgus stress at the elbow, often after a fall onto an outstretched hand or a sports-related injury. These fractures are extra-articular in most cases, but they can be associated with elbow dislocation and, occasionally, the fragment can become trapped inside the joint.

Key early priorities are pain control, neurovascular assessment, recognition and reduction of any elbow dislocation, and post-reduction imaging to ensure the medial epicondyle fragment is not incarcerated in the joint.

Management has traditionally varied widely. Minimally displaced fractures are usually treated non-operatively with immobilization, often in an above-elbow backslab or cast with the elbow flexed around 900. Surgery, usually open reduction and internal fixation (ORIF) with a screw or wires, has generally been reserved for clearer indications such as an incarcerated fragment, open fracture, ulnar nerve entrapment, gross elbow instability, or selected high-demand upper extremity athletes.

The real controversy has been the isolated displaced fracture. Supporters of ORIF argue that surgery restores anatomy, improves the reliability of healing, improves stability, and speeds return to activity. Observational studies have supported both approaches, driving uncertainty. Pediatric orthopedics has historically had few clinical trials, leaving decisions shaped by tradition, expert opinion, and retrospective data. Much of the older literature on the medial epicondyle has methodological limitations (e.g., follow-up issues, selection bias, limited patient-reported outcomes).

One smaller RCT from Finland (n=72) suggested non-operative care was non-inferior at 1 year, but a single small trial still left uncertainty.

Clinical Question:  In children (age 7 to 15 years) with a displaced medial epicondyle fracture, does surgical fixation improve patient-important outcomes compared with non-surgical immobilization, and is it cost-effective?

Reference:  Perry DC, et al. Surgical fixation versus non-surgical care for children with a displaced medial epicondyle fracture of the elbow (the SCIENCE study): a multicentre, randomised controlled, superiority trial and economic evaluation. The Lancet 2026

 This was the same lead author as the FORCE trial covered in SGEM #372 that taught us it is reasonable to treat a buckle fracture of the forearm with a soft bandage and the CRAFFT trial covered in SGEM #510 that taught us a cast-first strategy for young children with displaced distal radius fractures can be a reasonable approach.

  • Population: Children 7 to 15 years of age with a displaced medial epicondyle fracture (across 59 hospitals in the UK/Australia/NZ)
    • Exclusion: Injury >2 weeks old; incarcerated fragment in joint; complex intra-articular elbow fracture; additional fractures outside the elbow.
  • Intervention: Surgical fixation under general anaesthesia and fixation with screws/wires. A cast, splint, or bandage was applied afterward, and the duration was at the surgeon’s discretion.
  • Comparison: Non-surgical care which was immobilization of the elbow elbow ~90° flexion with cast/splint/sling. Duration was at the discretion of the surgeon. They were also allowed mobilization under clinical supervision and as tolerated base don pain. Casting >4 weeks was discouraged.
  • Outcome:
    • Primary Outcome: Functional recovery assessed by the Patient Report Outcomes Measurement System (PROMIS) Upper Extremity Score for Children Computer Adaptive Test (CAT) at 12 months. A higher score means better limb function.
    • Secondary Outcomes: Sports and performing arts participation assessed with the Disability of the Arm Shoulder, and Hand (DASH), pain assessment with the Wong-Baker Faces Pain Scale, health-related quality of life assessed with EQ-5D-Y, complications, health-care utilization, parental leave, days of purchased childcare, school absence
  • Trial: multicentre, randomised controlled superiority trial

Authors’ Conclusions: “The SCIENCE trial demonstrates that surgical fixation offers no clinical benefit and is not cost-effective compared with non-surgical care, while exposing children to avoidable surgical risks. These findings suggest that non-surgical care should be adopted as the default management strategy for these injuries, regardless of initial elbow dislocation status.”

Quality Checklist for Randomized Clinical Trials:

  1. The study population included or focused on those in the emergency department. No
  2. The patients were adequately randomized. Yes
  3. The randomization process was concealed. Yes
  4. The patients were analyzed in the groups to which they were randomized. Yes
  5. The study patients were recruited consecutively (i.e. no selection bias). Unsure
  6. The patients in both groups were similar with respect to prognostic factors. Yes
  7. All participants (patients, clinicians, outcome assessors) were unaware of group allocation. No
  8. All groups were treated equally except for the intervention. Unsure
  9. Follow-up was complete (i.e. at least 80% for both groups). Yes
  10. All patient-important outcomes were considered. Yes
  11. The treatment effect was large enough and precise enough to be clinically significant. No
  12. Trial Funding. Funded by the National Institute for Health and Care Research Health Technology Assessment programme, with additional NIHR Academy, Oxford NIHR Biomedical Research Centre, and Starship Foundation support.
  13. Financial conflicts of interest. Declared NIHR grant support; DCP and MLC had NIHR HTA roles; KW reported Zimmer Biomet and Medartis fellowship support; JGW reported other grants and consultancy fees.

Results: A total of 647 patients were assessed for eligibility, with 335 being randomized, and 334 were included in the ITT analysis. There were 166 children assigned to non-surgical care and 168 assigned to surgical fixation. Mean age was 11.7 years, 51% were female, 58% of injuries were sports-related, and 24% presented with elbow dislocation. The dominant arm was injured in 51% of participants.

Primary Outcome: (PROMIS UE at 12 months)

  • The non-surgical group had a score of 53.1 (SD 7.8), while the surgical group had a score of 54.3 (SD 5.7).
  • This was a mean difference of 1.57 (95% CI –0.01 to 3.14), p=0.052
  • The pre-specified clinically important difference was 4 points, so the observed difference was not clinically meaningful.

There was a slight difference in the per-protocol analysis. Non-surgical group had a score of 53 (SD 7.9). The surgical group had a score of 54.5 (SD 5.6). There was a mean difference of 1.76 (95% CI 0.1 to 3.42), p=0.0372

There was not much difference in the PROMIS scores as they tracked the patients over time. There was improvement in the scores up until around 6 months and the scores started to plateau.

Secondary Outcomes: There were many secondary outcomes. We will highlight a few.

They used different tools like the DASH, Wong-Baker Faces Pain Scale, or the EQ-5D-Y to measure return to activity, pain, and quality of life between the two groups. There were no significant differences between the two groups.

There was less school absence in the non-surgical group (5 vs 3.1 days) with a mean difference of 2.2 days (95% CI 0.7-3.6), p=0.003. That’s important to me as a parent.

Among the 150 children who underwent surgical management, 14 (9%) had intraoperative complications that included ulnar nerve injuries, hardware issues, fragmentation of the bony fragment. Seven required more surgery afterwards for these complications, and another 17 required surgery for screw or wire removal.

There were 5 (2%) complications in the 184 children who underwent non-surgical management, with one having a complication related to the cast and 3 who had to undergo delayed fixation of the bone fragments.

The surgical treatment option was also more expensive with a mean additional cost of £2435 per patient (95% CI; £1812 to £3057).

We want to give a shout-out to the involvement of patients and the public in many aspects of the study. The authors report that patients and the public were involved in design, conduct, oversight and dissemination. They helped select the outcomes that mattered the most and defined what constituted a “meaningful” treatment effect. They helped develop the patient-facing materials and aspects of the qualitative study. Some of the details of their contributions are described in each respective section of the manuscript. What a fantastic example of how more research should be done! Involve the stakeholders; not only can they help make the research more relevant and robust, but they can also be allies in knowledge dissemination.

Unmasked Trial:

This was an unblinded trial with patient-reported outcomes. The children, families, and treating clinicians knew the treatment allocation, and the primary outcome was a subjective functional score rather than an independently measured objective outcome. This matters because expectations about surgery or casting could influence reported function, pain, or return to activities. The central trial team and statisticians were masked during data cleaning, but there were no independent blinded clinician outcome assessors. Unmasked trials with subjective outcomes do not automatically invalidate a trial, but they do introduce potential biases, such as performance bias and observer bias.

Clinical vs Statistical Significance:

In the intention-to-treat analysis, we observed a p-value just above the typical cutoff of 0.05. In the per-protocol analysis, the p-value was less than 0.05.

But who cares? The patients and families do.

Thanks to their involvement, they had helped set the upper limit threshold for a clinically important difference to be at least 4 points. No matter what kind of analysis you do, the mean difference did not come close to that threshold.

Patient/Family vs Surgeon Preferences:

Among the patients included in this study, there were 5 in which there was no clinician equipoise. In four of those patients, the treating surgeon opted for surgical management.

In comparison, there were 161 children who met eligibility, but the families declined to participate. Of those children and families, 131 had a treatment preference, with 104 (~80%) wanting non-surgical care. This could introduce selection bias.

We also saw this with some of the protocol deviations of which there were 26 treatment crossovers. Twenty patients in the surgical care group ended up declining surgery due to parent or child decisions.

The ITT analysis is the better primary analysis because it preserves randomization, but it estimates the effect of a treatment policy rather than the pure biological effect of receiving surgery. The per-protocol analysis slightly favoured surgery statistically, but the authors correctly noted it was prone to selection bias and did not reach clinical significance.

Subgroup Analysis: Elbow Dislocation:

There still remains some uncertainty around managing the children with an elbow dislocation. Typically, surgical management was preferred for this group. Children with elbow dislocation had poorer functional outcomes, with an average difference of -2.88 (95% CI -5.36 to -0.4), p=0.023. This difference appears to be less pronounced by 12 months. We can see this in Supplemental Figure 2b.

However, the authors acknowledge that the study was not powered to detect these differences.

Potential Harms and Benefits:

Many interventions and decisions we make in medicine require balancing the potential harms and benefits. We want to maximize the benefits while minimizing the harms. The study based on this data suggests that there are minimal overall benefits from surgical management of displaced medial epicondyle fractures of the elbow in children. The function, return to activity, pain scores, and quality of life were very similar between groups.

There were more potential harms in the surgical treatment group, which included nerve injury, hardware issues, and intraoperative difficulties. A portion of the children undergoing surgery also had to undergo surgery again for hardware removal.

Comment on the Authors’ Conclusion Compared to the SGEM Conclusion: We generally agree with the authors’ conclusion. This trial makes it hard to justify routine fixation for displaced medial epicondyle fractures when function is the same, surgery increases downstream procedures/complications, and it costs more. However, there is still room for nuance in identifying the children who may develop poorer outcomes from non-operative treatment.

SGEM Bottom Line: Don’t be too quick to call the OR for kids with an isolated displaced medial epicondyle fracture. The best available evidence shows routine surgical fixation does not provide a patient-important improvement in function at one year compared with non-surgical care, but it does add surgical risk, more procedures, and cost.

Case Resolution: You consult the orthopedist who discusses both options surgical vs non-surgical care with the family. You cover the potential harms and benefits of both approaches and engage in shared decision making. The family opts for non-surgical care.

Clinical Application: This study should nudges routine practice toward non-surgical care for children aged 7 to 15 years of age with an isolated displaced medial epicondyle fracture who match the trial population. In the ED, that means analgesia, careful neurovascular and ulnar nerve assessment, reduction of any elbow dislocation, repeat imaging to ensure the fragment is not trapped in the joint, immobilization, and orthopaedic follow-up.

Do not apply this trial to children with an incarcerated fragment, open fracture, complex intra-articular injury, medial condyle fracture, additional major fractures, persistent dislocation, gross instability, or significant neurovascular compromise. Those remain orthopaedic decision points. For athletes, the trial included many sports injuries, but it was not powered to answer every elite-athlete subgroup question, so shared decision-making with pediatric orthopaedics is still appropriate.

For the patients treated non-operatively, close and timely follow up in orthopaedic clinic is important.

 What do I tell the patient/family? It appears that your daughter has a small break in the bone around her elbow. There are two potential approaches to treatment. The first option would be surgical. The other option would be to immobilize it with a cast. There’s newer data suggesting that surgery does not meaningfully improve arm function, return to activity, or quality of life compared with casting. But in certain circumstances, it still may be appropriate. Let’s talk more about what you would prefer.

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.

Other orthopedic SGEM episodes: