Reliability of Radiographic Union Score and correlation of clinical outcomes in children operated for supracondylar humerus fracture: A prospective study

To the editor,

We read with interest the above publication by Yüzügüldü et al. ¹ Despite the prospective study design, the follow-up appointments were at widely spread intervals which invalidates the authors’ creation of the follow-up groups, since the “2-week” follow-up was done at a median of 17.1 days (13–21 days) and the “4-week” follow-up at a median of 33.8 days (26–39 days). This shows that the authors ¹ removed the vast majority of K-wires in the fifth and sixth weeks and not in the fourth week as they had claimed in their discussion. Yüzügüldü et al. ¹ did not consider linking age with the Radiographic Union Score for tibial fractures (RUST), with a 1-year-old having a faster healing potential compared to a 12-year-old and should therefore achieve a higher RUST earlier. RUST might also be affected by the supracondylar humerus fracture (SCHF) patterns as described in the Johns Hopkins classification, ² with oblique SCHFs potentially healing quicker than transverse fractures and low fractures quicker than high fractures because of larger fracture surfaces and a higher cancellous/cortical bone ratio.

Yüzügüldü et al.’s ¹ first follow-up appointment is too late to perform a re-manipulation if a fracture has re-displaced because it is unlikely for the bone segments to be mobile anymore at this stage. We recommend that children be followed up at 6–8 days after a closed/open reduction or casting in situ for obtaining repeat X-rays to assure adequate alignment. This would also allow early detection of wound problems. Using RUST as a tool to decide when to remove wires and casts would potentially require multiple cast removals, since the casts would obscure the radiographic signs of bone healing, but cast removals can be stressful for children. Yüzügüldü et al. ¹ claimed that clinical findings can be used to determine bone union, but this is contradicted by the wide ranges of movement (ROM) seen at all three follow-up stage groups, which would not allow for reliable link between a specific amount of ROM to RUST. We could also not identify a study in the literature which examined the recovery of elbow function over time following SCHFs such as that conducted by Bernthal et al. ³ for lateral condyle fractures of the humerus, who reported that the mean relative arc of motion was 44% of the non-injured elbow at the time of cast removal, 84% at 12 weeks, 90% at 24 weeks, and 97% at 48 weeks. Until 18 weeks after the injury, patients treated with closed or open reduction had a significantly reduced ROM compared to those treated nonoperatively. ³

There is an inconsistency between the Flynn scores reported by Yüzügüldü et al. ¹ and the provided ROM data. McKay et al. ⁴ reported a mean elbow flexion of 146° and mean extension of 7° for non-injured elbows of children 3–9 years of age. Yüzügüldü et al. ¹ claimed that most of their patients had already a normal or near normal ROM at the third follow-up at a median 61 days and that only three patients (5.2%) had an unsatisfactory Flynn score but based on McKay et al.’s ⁴ data at least 29 (50%) of the study patients ¹ had an unsatisfactory Flynn score.

Yüzügüldü et al. ¹ reported that the modified Gartland classification defines a type-2 SCHF as having an intact posterior cortex but the modified classification introduced the type-2a which has a sagittal deformity with an intact posterior cortex and the type-2b which is malrotated or translated and does not have an intact posterior cortex, which suggests that the authors ¹ might have included all type-2b fractures in the type-3 group.

Rutarama and Firth ⁵ reported routine K-wire removal at 3 weeks after closed reduction and percutaneous K-wire fixation of type-3 SCHFs under the age of 14 years and encouraged return to normal activities as pain allowed.

In conclusion, Yüzügüldü et al. ¹ concluded that RUST scores can be used to provide data-based predictions of union for splint and pin removal without providing any data and information to guide the reader on how this could be achieved in clinical practice, especially with the authors ¹ not having considered patient age and the Johns Hopkins classification types. Yüzügüldü et al.’s ¹ study period ended February 2020 and the article was submitted February 2024 so that the authors ¹ had ≥3 years to test the functionality of RUST in their practice but it seems that this was either not done or had failed, since the authors did not share such information, indicating that it is probably impractical and nonbeneficial to use RUST to judge the length of cast immobilization and timing of K-wire removal. We continue to remove casts and K-wires at 3–4 weeks after the start of treatment in the vast majority of children, which depends on the age of the children and clinic availability, and is in line with Rutarama and Firth’s ⁵ practice.

Supplemental Material

sj-pdf-1-cho-10.1177_18632521251370896 – Supplemental material for Reliability of Radiographic Union Score and correlation of clinical outcomes in children operated for supracondylar humerus fracture: A prospective study

Supplemental material, sj-pdf-1-cho-10.1177_18632521251370896 for Reliability of Radiographic Union Score and correlation of clinical outcomes in children operated for supracondylar humerus fracture: A prospective study by Andreas Rehm, Luke Granger, Rebecca J Worley, Felix Morriss, Eve McMahon and Elizabeth Ashby in Journal of Children's Orthopaedics