中国血液净化

中国血液净化 ›› 2023, Vol. 22 ›› Issue (12): 896-899.doi: 10.3969/j.issn.1671-4091.2023.12.003

• 临床研究 • 上一篇    下一篇

维持性血液透析患者皮肤瘙痒影响因素分析及其列线图预测模型构建

陆耀玲   仲丽丽   周 芳   顾红叶   

  1. 223002 淮安，1淮安市第二人民医院(徐州医科大学附属淮安医院)血液净化中心
  • 收稿日期:2023-06-19 修回日期:2023-09-19 出版日期:2023-12-12 发布日期:2023-11-30
  • 通讯作者: 仲丽丽 E-mail:haeyzll@163.com
  • 基金资助:
    江苏省卫生健康委员会科研项目(H2019062)

Analysis of the influencing factors for skin itching in uremic hemodialysis patients and construction of a column chart model for prediction analysis

U Yao-ling, ZHONG Li-li, ZHOU Fang, GU Hong-ye   

  1. Blood Purification Center, Huai'an Second People's Hospital (Huai'an Hospital affiliated to Xuzhou Medical University), Huai'an 223002, China
  • Received:2023-06-19 Revised:2023-09-19 Online:2023-12-12 Published:2023-11-30
  • Contact: 223002 淮安，1淮安市第二人民医院(徐州医科大学附属淮安医院)血液净化中心 E-mail:haeyzll@163.com

PDF

45

摘要： 目的  探讨维持性血液透析(maintenance hemodialysis，MHD)患者皮肤瘙痒影响因素，构建预测模型。 方法  选取2020年7月—2022年12月于淮安市第二人民医院(徐州医科大学附属淮安医院)收治的162例MHD患者根据是否发生皮肤瘙痒分为皮肤瘙痒组(96例)和无皮肤瘙痒组(66例)。收集患者临床资料，采用Logistic回归分析影响MHD患者皮肤瘙痒的危险因素；利用R软件构建列线图模型，通过校准曲线验证模型的一致性，通过ROC曲线的曲线下面积评估模型的区分度。 结果  162例MHD患者中皮肤瘙痒有96例，发生率为59.26%；单因素分析结果显示：MHD患者皮肤瘙痒组的糖尿病史(χ2=5.569，       P=0.018)、甲状旁腺激素(t=13.510，P＜0.001)、β2微球蛋白(t=11.836，P＜0.001)、血磷(t=10.478，P＜0.001)、C反应蛋白水平(t=13.221，P＜0.001)与无皮肤瘙痒组比较差异具有统计学意义；多因素Logistic回归分析结果显示：β2微球蛋白(OR=1.451，95% CI：1.217～1.731，P＜0.001)、血磷(OR=7.455，   95% CI:1.194～46.533，P=0.032)、C反应蛋白(OR=10.929，95% CI:4.086～29.234，P＜0.001)；列线图预测模型具有较好的区分度(ROC曲线下面积为0.977，95% CI:0.956～0.998)和一致性(Hosmer-Lemeshow拟合优度检验χ2=10.065，P=0.261)。 结论  本研究以β2微球蛋白、血磷、C反应蛋白3项危险因素构建预测尿毒症血液透析患者发生皮肤瘙痒风险的列线图模型，区分度及一致性较好，可用于临床实践。

关键词: 血液透析, 皮肤瘙痒, 列线图

Abstract: Objective  To explore the influencing factors for skin itching in uremic hemodialysis patients and to construct a column chart model for prediction analysis.  Methods  A total of 162 uremic hemodialysis patients admitted to Huai'an Second People's Hospital (Huai'an Hospital affiliated to Xuzhou Medical University) from July 2020 to December 2022 were recruited as the study subjects. They were grouped into a skin itching group (96 cases) and a non-skin itching group (66 cases) based on the symptom of skin itching. Clinical data were collected from the patients. Logistic regression was applied to analyze the risk factors for skin itching in uremic hemodialysis patients. R software was applied to construct a column chart model, and calibration curve was applied to verify the consistency of the mode. The area under the ROC curve was applied to evaluate the predictive value and discrimination ability of the model.  Results  Ninety-six cases (59.26%) of the 162 uremic hemodialysis patients had pruritus. Univariate analysis showed that diabetes history (χ2=5.569, P=0.018), parathyroid hormone (t=13.510, P<0.001), β2 microglobulin (t=11.836, P<0.001), serum phosphorus (t=10.478, P=0.001) and C-reactive protein (t=13.221, P<0.001) were significantly different between skin itching group and non-skin itching group. Multivariate logistic regression showed that β2 microglobulin (OR=1.451, 95% CI: 1.217～1.731, P<0.001), serum phosphorus (OR=7.455, 95% CI: 1.194～46.533, P=0.032), and C-reactive protein (OR=10.929, 95% CI: 4.086～29.234, P<0.001) were the independent risk factors for itching in uremic hemodialysis patients. The nomogram prediction model had better differentiation (area under ROC curve=0.977, 95% CI:0.956～0.998) and consistency (Hosmer-Lemeshow goodness of fit test, χ2=10.065, P=0.261) abilities.  Conclusion   This study constructs a column chart model for predicting the risk of skin itching in uremic hemodialysis patients based on three risk factors: β2 microglobulin, serum phosphorus, and C-reactive protein. This model has better discrimination and consistency abilities, and is expected to become an important tool for predicting the risk of skin itching in uremic hemodialysis patients.

Key words: Hemodialysis, Skin itch, Column chart

中图分类号: