Page 84 - 《中国药房》2025年10期
P. 84
吡拉西坦相关血小板减少症的危险因素分析及其风险预测模型
建立
Δ
2
1 #
3
1
黄天敏 ，陆星名 ，郑 媚 ，郭贵宗 ，陆 欣 ，罗艺林 ，杨映霞 （1.广西医科大学第一附属医院药学部，南宁
1＊
1
4
530021；2.广西壮族自治区宾阳县人民医院药学部，南宁 530400；3.广西医科大学教务处，南宁 530021；
4.百色市食品药品认证审评中心，广西 百色 533000）
中图分类号 R969.3；R979.9 文献标志码 A 文章编号 1001-0408（2025）10-1226-06
DOI 10.6039/j.issn.1001-0408.2025.10.13
摘 要 目的 分析吡拉西坦相关血小板减少症的危险因素，并建立其风险预测模型。方法 回顾性收集广西医科大学第一附属
医院2021年1月－2023年12月使用吡拉西坦治疗住院患者的电子病历信息，汇总性别、年龄、基础疾病、联合用药、实验室数据
等临床资料。根据是否发生血小板减少症将患者分为发生组和未发生组，比较两组患者临床资料的差异。通过单因素/多因素
Logistic回归分析确定独立危险因素，以列线图可视化展示，并构建风险预测模型；采用受试者工作特征（ROC）曲线、Bootstrap内
部验证、校准曲线评估模型的预测效能。结果 共纳入 224 例患者，其中未发生组 196 例、发生组 28 例，血小板减少症发生率为
12.50%。单因素Logistic回归分析结果显示，发生组联用三联及以上抗菌药物的患者比例和血肌酐水平均显著高于未发生组，血
红蛋白水平显著低于未发生组（P＜0.05）。多因素Logistic回归分析结果显示，联用三联及以上抗菌药物、低血红蛋白水平、高血
肌酐水平是吡拉西坦相关血小板减少症的独立危险因素（P＜0.05）。所构建的风险预测模型为LogitP＝－1.114+1.256×联用三
联及以上抗菌药物－0.017×血红蛋白水平+0.009×血肌酐水平。该模型ROC曲线的曲线下面积（AUC）为0.757，最佳截断值为
0.474；Bootstrap内部验证的ROC曲线的AUC为0.733；未校准曲线、偏倚校准曲线均与参考曲线接近。结论 联用三联及以上抗
菌药物、低血红蛋白水平、高血肌酐水平是吡拉西坦相关血小板减少症的独立危险因素；所建风险预测模型具有良好的预测效能。
关键词 吡拉西坦；血小板减少症；危险因素；风险预测模型；合理用药
Analysis of risk factors for piracetam-associated thrombocytopenia and the establishment of risk prediction
model
2
HUANG Tianmin ，LU Xingming ，ZHENG Mei ，GUO Guizong ，LU Xin ，LUO Yilin ，YANG Yingxia（1. Dept.
1
1
1
3
4
1
of Pharmacy， the First Affiliated Hospital of Guangxi Medical University， Nanning 530021， China；2. Dept. of
Pharmacy， Binyang County People’s Hospital of Guangxi Zhuang Autonomous Region， Nanning 530400，
China；3. Office of Academic Affairs， Guangxi Medical University， Nanning 530021， China；4. Food and Drug
Certification Evaluation Center of Baise， Guangxi Baise 533000， China）
ABSTRACT OBJECTIVE To analyze the risk factors contributing to piracetam-associated thrombocytopenia and develop a
predictive model for risk prediction. METHODS The electronic medical record information of inpatients treated with piracetam was
collected retrospectively from the First Affiliated Hospital of Guangxi Medical University from January 2021 to December 2023，
including gender， age， underlying diseases， combined medication， and laboratory data， etc. Patients were divided into the
occurrence group and the non-occurrence group according to whether thrombocytopenia occurred， and the differences in clinical
data between the two groups were compared. The independent risk factors were determined through univariate/multivariate Logistic
regression analysis. A nomogram was drawn to visually present the independent risk factors， and a risk prediction model was
constructed. The predictive efficacy of the model was evaluated using the receiver operating characteristic （ROC） curve， Bootstrap
internal validation and calibration curve. RESULTS A total of 224 patients were included， among which 196 cases were in the non-
occurrence group and 28 cases in the occurrence group. The incidence of thrombocytopenia was 12.50%. The results of the
univariate Logistic regression analysis showed that the
Δ 基金项目 广西壮族自治区卫生健康委员会自筹经费科研课题 proportion of patients using three or more combined antibiotics
（No.Z20191021） and the level of serum creatinine in the occurrence group were
＊第一作者 副主任药师，硕士。研究方向：临床药学。E-mail：
significantly higher than those in the non-occurrence group，
46327916@qq.com
# 通信作者 副主任药师，硕士。研究方向：医院药学。E-mail： while the level of hemoglobin was significantly lower （P＜
19830727@qq.com 0.05）. The results of the multivariate Logistic regression
· 1226 · China Pharmacy 2025 Vol. 36 No. 10 中国药房 2025年第36卷第10期
