先天性马蹄足术后复发的影响因素及预测模型的建立

The factors influencing postoperative recurrence in congenital clubfoot and construction of a prediction model

摘要:

目的 基于列线图模型分析先天性马蹄足术后复发的影响因素。方法 回顾性分析2019年12月—2021年12月嘉兴市妇幼保健院和南通市妇幼保健院收治的153例先天性马蹄足患儿的临床资料，所有患儿接受矫治手术，根据术后2年内是否复发，将患儿分为复发组（59例）和未复发组（94例）。收集两组患儿性别、年龄、患侧、家族疾病史、术前Dimeglio分型、术前是否石膏矫形、是否行跟腱切断、术后支具依从性等基本资料，术前通过Pirani评分评估患儿病情，并利用X射线测量其横位距跟角角度、侧位胫骨跟骨角度。通过多因素一般Logistic回归模型筛选先天性马蹄足患儿术后复发的影响因素，构建列线图预测模型，并评价模型的拟合效果。结果 复发组术前Pirani评分、Dimeglio Ⅳ型、有家族疾病史、术前未进行石膏矫正、术后支具使用依从性差的占比均高于未复发组（P <0.05），而横位距跟角角度、侧位胫骨跟骨角度均小于未复发组（P <0.05）。多因素一般Logistic回归分析结果显示，有家族疾病史［O^R=8.997（95% CI：1.961，41.270）］、Dimeglio Ⅳ型［O^R=4.835（95% CI：1.462，15.986）］、术前Pirani评分高［O^R=153.681（95% CI：7.907，2 987.012）］、术后支具依从性较差［O^R=5.418（95% CI：1.628，18.029）］为影响患儿术后复发的危险因素（P <0.05）；术前行石膏矫正［O^R=0.347（95% CI：0.133，0.907）］、侧位胫骨跟骨角度大［O^R=0.061（95% CI：0.010，0.375）］、横位距跟角角度大［O^R=0.255（95% CI：0.089，0.732）］是影响患儿术后复发的保护因素（P <0.05）。校准曲线的平均绝对误差为0.027，表明模型的预测准确度高，趋近于理想曲线。结论 家族遗传史、术前矫治情况、病情严重程度、治疗依从性均是影响先天性马蹄足患儿术后复发的重要因素。

关键词:先天性马蹄足;术后复发;多因素分析;列线图模型

Abstract:

Objective To analyze the factors influencing the postoperative recurrence of congenital clubfoot using a nomogram model.Methods A retrospective analysis was conducted on clinical data of 153 children with congenital clubfoot treated at Jiaxing Maternal and Child Health Hospital and Nantong Maternal and Child Health Hospital from December 2019 to December 2021. All children underwent corrective surgery and were divided into a recurrence group (59 cases) and a non-recurrence group (94 cases) based on whether they experienced a recurrence within two years after surgery. Data including sex, age, affected side, family history of diseases, preoperative Dimeglio classification, preoperative orthosis using plaster casting, Achilles tendon lengthening, and postoperative bracing compliance were collected. Preoperative disease severity was assessed using the Pirani scoring system, and X-ray examinations were performed to measure the transverse talocalcaneal angle and the lateral tibiocalcaneal angle. Multivariable Logistic regression analysis was used to identify factors influencing postoperative recurrence in children with congenital clubfoot, and a nomogram prediction model was constructed and its fit was evaluated.Results The recurrence group had higher Pirani scores and higher proportions of patients with Dimeglio type IV clubfoot, family history of diseases, no preoperative orthosis using plaster casting, and poor postoperative bracing compliance compared to the non-recurrence group (P < 0.05). The transverse talocalcaneal and lateral tibiocalcaneal angles in the recurrence group were smaller than those in the non-recurrence group (P < 0.05). Multivariable Logistic regression analysis indicated that family history of diseases [O^R = 8.997 (95% CI: 1.961, 41.270) ], Dimeglio type IV clubfoot [O^R = 4.835 (95% CI: 1.462, 15.986) ], high preoperative Pirani scores [O^R = 153.681 (95% CI: 7.907, 2987.012) ], and poor postoperative bracing compliance [O^R = 5.418 (95% CI: 1.628, 18.029) ] were risk factors for postoperative recurrence (P < 0.05), and that preoperative orthosis using plaster casting [O^R = 0.347 (95% CI: 0.133, 0.907) ], larger lateral tibiocalcaneal angle [O^R = 0.061 (95% CI: 0.010, 0.375) ], and larger transverse talocalcaneal angle [O^R = 0.255 (95% CI: 0.089, 0.732) ] were protective factors against postoperative recurrence (P < 0.05). The calibration curve's mean absolute error was 0.027, indicating high prediction accuracy and an ideal curve fit.Conclusions Family history, preoperative treatment, severity of the condition, and treatment compliance are significant factors influencing postoperative recurrence in children with congenital clubfoot.

Key words:congenital clubfoot;postoperative recurrence;multivariable analysis;nomogram model