LAUSR

To the Editor: It has been a great pleasure to read this article by Ramo et al1 in your esteemed journal. The optimal treatment and classification of pediatric lateral humeral condyle fractures remain controversial. The authors reviewed the Song classification, treatment, and outcome of 736 pediatric patients with lateral condyle fractures, and found that the Song classification with high interobserver and intraobserver reliability. This study can be used as important evidence for determining the risk for failure of nonoperative treatment and guiding treatment outcomes. Nonetheless, I have some confusion about this article and hope to be addressed for clarification. It is well known that stage 3 of Song classification indicated a fracture gap that was as wide laterally as it was medially and ≤2mm of displacement.2 However, we have learned the sentence of “required displacement to be <2mm on all 4 views for Song 1 or Song 2 and that displacement of >2mm on any of the 4 views be classified as Song 3 to 5” in theMethods section from the article. It seems that the authors believe that the fracture displacement of stage 3 of Song classification should be >2mm. This is inconsistent with our existing knowledge. Was it the modified Song classification that the authors used in this study? Also, Song classification is based on the degree of fracture displacement and the degree of articular cartilage damage.2 Especially in stage 1 to 3 of Song classification, displacement of these fractures is < 2mm, and the classification can be determined by the integrity of the articular cartilage. However, in this study, the surgeon performed an arthrogram in only 49 patients and most fractures were classified by radiographs alone. Our experience is that when there is no technology to determine the integrity of articular cartilage and only by radiographs, the interobserver reliability of classification is not high.3 I would like to know how did the observers have such a high interobserver reliability of classification in the absence of arthrography and other techniques to determine the integrity of articular cartilage. Finally, I am puzzled that there are 6 observers in the abstract but only 5 observers in the rest of the article. Thanks to the authors for reporting their excellent study, and I have learned a lot from this article. I look forward to discussing with the authors on the confusion I mentioned above.