目的 探讨儿童肱骨髁上骨折临床愈合的现状及影响因素，建立回归方程，预估骨折愈合时间。 方法 采用回顾性研究，选取2021年6月1日至2023年5月30日在重庆医科大学附属儿童医院骨科接受手术治疗的肱骨髁上骨折儿童为研究对象，收集相关资料，包括一般情况、诊疗记录、检查检验结果、复诊信息、骨痂形成及骨折临床愈合时间等。使用SPSS 23.0进行单因素、多因素分析，建立骨痂形成和骨折临床愈合的线性回归方程。 结果 ①本研究共调查251例患儿，月龄13~161个月，平均月龄（69.49±2.16）；男性153例（60.96%），女性98例（39.04%）；伸直型229例（91.23%），屈曲型22例（8.76%）；骨痂形成（距骨折）（17.21±5.73） d，骨痂形成（距手术）（13.66±5.68） d，骨折临床愈合（距骨折）（46.67±13.52） d，骨折临床愈合（距手术）（41.24±12.98） d。②骨痂形成与等待手术时间和总蛋白值有关，骨折临床愈合与月龄（X₁）、骨折发生在秋季（X₂）、等待手术时间（d）（X₃）、纤维蛋白原（X₄）、手术时长（min）（X₅）、白蛋白、骨折分型有关；③采用多重线性回归可预测骨痂形成时间的5.5%或6.2%，骨折临床愈合时间的20.8%。儿童肱骨髁上骨折愈合时间（距离骨折发生时间点）Y=22.274+0.118X₁-5.233X₂+0.047X₃+3.139X₄+0.058X₅，R²=0.225，调整后R²=0.208，德宾-沃森系数=1.953，F=13.528，P=0.001。 结论 儿童肱骨髁上骨折发生在每年7~9月，患儿年龄越小，骨折临床愈合时间越短；通过缩短等待手术时间、降低纤维蛋白原值、缩短手术时间等方式可缩短骨折临床愈合时间。利用回归方程可初步预测骨折愈合时间，协助医患安排复诊和取出装置，避免长期石膏固定和多次就诊。

Objective To investigate the clinical healing status of supracondylar fracture of the humerus in children and related influencing factors，to establish a regression equation，and to predict the time to fracture healing. Methods A retrospective study was conducted among the children with supracondylar fracture of the humerus who underwent surgical treatment in Department of Orthopedics，Children’s Hospital of Chongqing Medical University，from June 1，2021 to May 30，2023，and related data were collected，including general status，diagnosis and treatment records，examination results，follow-up information，time to callus formation，and time to clinical fracture healing. SPSS 23.0 was used for univariate and multivariate analyses to establish a linear regression equation for callus formation and clinical fracture healing. Results A total of 251 children，aged 13-161 months，were enrolled in this study，with a mean age of （69.49±2.16） months，and there were 153 boys（60.96%） and 98 girls（39.04%）. There were 229 children with extension-type fracture（91.23%） and 22 with flexion-type fracture（8.76%）. The mean callus formation time was （17.21±5.73） days from fracture and （13.66±5.68） days from surgery，and the clinical fracture healing time was （46.67±13.52） days from fracture and （41.24±12.98） days from surgery. Callus formation was associated with the waiting time for surgery and total protein，and clinical fracture healing was associated with age in months（X₁），onset of fractures in autumn（X₂），waiting time for surgery（days） （X₃），fibrinogen （X₄），time of operation（minutes）（X₅），albumin，and fracture type. The multiple linear regression equation could predict 5.5% or 6.2% of callus formation time and 20.8% of clinical fracture healing time. The healing time of supracondylar fracture of the humerus in children（from the onset of fracture） Y=22.274＋0.118X₁-5.233X₂+0.047X₃+3.139X₄+0.058X₅（R²=0.225，adjusted R² =0.208，Durbin-Watson coefficient=1.953，F=13.528，P=0.001）. Conclusion Supracondylar fracture of the humerus in children often occurs in July to September，and the younger the age，the shorter the time to clinical fracture healing. The methods，such as shortening the waiting time for surgery，reducing fibrinogen，and shortening the time of operation，can help to shorten the time to clinical fracture healing. The regression equation can preliminarily predict the time to fracture healing to assist doctors and patients in arranging follow-up visits and the removal of devices，thereby avoiding long-term plaster fixation and multiple visits.