肾性继发性甲状旁腺功能亢进患者甲状旁腺切除术后复发的预测模型

Prediction model of recurrence after parathyroidectomy in secondary hyperparathyroidism

目的:定量评估肾性继发性甲状旁腺功能亢进症(SHPT)患者接受甲状旁腺切除术(parathyroidectomy,PTX)治疗后复发的风险。方法:收集中国人民解放军北部战区总医院2017年6月-2019年5月期间接受PTX的168例患者的临床资料,通过赤池信息准则(Akaike information criterion,AIC)筛选因素、列线图的形式构建预测模型,由2019年6月-2021年9月期间接受PTX治疗的158例患者组成验证集,在区分度、一致性和临床实用性3个方面对该模型进行外部验证。结果:本研究构建的预测模型包含6个变量,分别是透析方式、异位甲状旁腺、术后1 d及术后1个月的全段甲状旁腺激素水平(iPTH)、切除甲状旁腺数目以及术后1 d的血磷,该模型外部验证的C指数为0.992,Calibration曲线的P值为0.886:1,决策曲线也显示该模型评估效果良好。结论:本研究构建的预测模型有助于判断SHPT患者行PTX术后是否复发,对其进行个体化预测。

Objective: To quantitatively evaluate the risk of recurrence in patients with secondary hyperparathyroidism after parathyroidectomy.Methods: The clinical data of 168 patients who underwent parathyroidectomy(PTX)from June 2017 to May 2019 were collected.The prediction model was constructed by using Akaike information criterion(AIC)to screen factors.A total of 158 patients treated with PTX from June 2019 to September 2021 were included in the validation set to conduct external validation of the model in three aspects of differentiation,consistency and clinical utility.Results: The prediction model we constructed includes different dialysis methods,ectopic parathyroid gland,the iPTH level at one day and one month after surgery,the number of excisional parathyroid and postoperative blood phosphorus.The C index of external validation of this model is 0.992 and the P value of the Calibration curve is 0.8861.The decision curve analysis also shows that the evaluation effect of this model is perfect.Conclusion: The prediction model constructed in this study is useful for individualized prediction of recurrence after PTX in patients with secondary hyperparathyroidism.