单侧原发性醛固酮增多症患者术后高血压完全治愈预测模型构建
DOI:
作者:
作者单位:

作者简介:

陈涛,男,博士,主任医师,硕士研究生导师。 中华医学会内分泌分会免疫内分泌学组委员,中国医疗保健国际 交流促进会中老年医疗保健分会委员,四川大学华西医院肾上腺 疾病诊治中心秘书等。 主要研究方向:内分泌高血压 / 肾上腺疑 难重症、低钾血症、糖尿病 / 脂代谢紊乱的诊治。

通讯作者:

基金项目:

四川省科技厅重点研发项目 (编号: 2023YFS0033);国家重点研发计划(编号: 2021YFC2501601);四 川大学华西医院临床孵化项目重大项目(编号:2021HXFH008)


Construction of a predictive model for complete cure of hypertension after surgery in patients with unilateral primary aldosteronism
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    目的 探讨单侧原发性醛固酮增多症(原醛症)患者术后高血压完全缓解的影响因素,并构建预测模型。 方法 纳入 2018 年 1 月至 2021 年 12 月在四川大学华西医院住院并接受腹腔镜肾上腺切除术的 128 例单侧原醛症患者。 根据术 后高血压是否完全治愈,将患者分为完全治愈组(n = 70)和未治愈组(n = 58)。 收集术前临床及生化数据,采用 LASSO-Logistic 回归分析影响术后高血压完全治愈的独立因素,并通过列线图将模型可视化。 结果 128 例患者中,70 例(54. 7% )术后高血 压完全治愈。 LASSO-Logistic 回归分析显示,性别、高血压持续时间、肌酐和尿酸是术后高血压完全治愈的独立预测因素。 基 于这些变量构建的列线图预测模型在训练集和验证集中的 AUC 分别为 0. 75 和 0. 88,具有较好的鉴别能力。 校正曲线显示模 型拟合良好,决策曲线分析(DCA)表明,预测概率为 0. 1 ~ 1 的范围内具有良好的临床适用性。 结论 基于 LASSO Logistic 回 归分析构建的单侧原醛症术后高血压完全治愈预测模型具有较高的预测准确性和临床应用价值,有助于术后高血压治愈的 个体化管理。

    Abstract:

    Objective To explore the influencing factors for complete resolution of hypertension after surgery in patients with unilateral primary aldosteronism (PA) and to construct a predictive model. Methods One hundred and twenty eight patients with unilateral PA who underwent laparoscopic adrenalectomy in our hospital from January 2018 to December 2021were included. Based on whether hypertension was completely cured after surgery, the patients were divided into a complete cure group (n = 70) and a non-cure group (n = 58). Clinical and biochemical data before operation were collected. LASSO logistic regression was used to analyze the independent predictors of complete resolution of hypertension after surgery. The predictive model was visualized through nomogram. Results There were 70 patients (54. 7% ) with complete resolution of hypertension after surgery. LASSO logistic regression analysis showed that gender, duration of hypertension, serum creatinine and uric acid were independent predictors for complete resolution of hypertension after surgery. A nomogram-based predictive model was constructed. The model showed good discriminatory ability in both training and validation sets. Their AUC values were 0. 75 and 0. 88, respectively. Calibration curves indicated good model fit. Decision curve analysis (DCA) indicated that prediction probabilities of this model within the range of 0. 1 to 1 had good clinical applicability. Conclusions A predictive model for complete resolution of hypertension after surgery in patients with unilateral PA is constructed based on LASSO logistic regression analysis. It has higher predictive accuracy and clinical application value. It can support individualized management for hypertension resolution.

    参考文献
    相似文献
    引证文献
引用本文

陈彦希a,彭建越,谭璐a,张婷婷b,任艳a,朱育春c,沈思魁c,陈涛a.单侧原发性醛固酮增多症患者术后高血压完全治愈预测模型构建[J].实用医院临床杂志,2025,22(1):17-21

复制
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:2024-11-05
  • 最后修改日期:2024-11-12
  • 录用日期:
  • 在线发布日期: 2025-01-27
文章二维码