首页|卡泊芬净血药浓度预测模型对血液科真菌感染患者预测准确性的验证与评价

卡泊芬净血药浓度预测模型对血液科真菌感染患者预测准确性的验证与评价

扫码查看
目的 研究卡泊芬净(CPFG)血药浓度影响因素,构建预测模型,并验证该模型的预测效果,为血液科真菌感染患者的个体化用药提供参考.方法 选取2021年3月至2022年6月入住天津医科大学总医院血液科,使用CPFG进行抗真菌治疗的75例患者为研究对象,进行CPFG血药浓度监测,探讨CPFG血药浓度的影响因素,并据此构建预测模型.用Hosmer-Lemeshow(H-L)检验模型的拟合优度,另选取30例患者作为验证组,通过受试者工作特征(ROC)曲线验证模型的预测效果.结果 患者0.5、9、24 h的平均血药浓度分别为(12.54±4.38)、(6.80±2.76)、(4.13±2.16)μg·mL-1,平均 AUC0-24h为(152.05±57.60)µg·mL-1·h,2 例患者 AUC0-24h低于参考值(98 µg·mL-1·h).相关性分析结果显示,性别与0.5 h血药浓度呈现相关性(P<0.05),与其余2个时间点血药浓度及AUC0-24h间无相关性(P>0.05).体质量和白蛋白(Alb)浓度与0.5、9、24 h血药浓度、AUC0-24 h呈现相关性(P<0.05),其余指标与各时间点血药浓度和AUC0-24 h无相关性(P>0.05).多因素分析结果显示,患者0.5 h血药浓度影响因素为性别、Alb浓度及体质量,9和24 h血药浓度以及AUC0-24h的影响因素为Alb浓度和体质量(P<0.05).相关性分析结果显示,日剂量与CPFG 0.5、9、24 h的血药浓度和AUC0-24h均具有正相关性(P<0.05).多因素分析结果显示,日剂量也为CPFG血药浓度影响因素之一(P<0.05).ROC曲线显示,该模型预测能力较好.结论 体质量、Alb与CPFG血药浓度及药时曲线下面积显著相关,可据此预防规避卡泊芬净在临床使用中的风险.
Validation and evaluation of the predictive accuracy of the caspofungin blood concentration prediction model in patients with fungal infections in the haematology department
Objective To study the factors influencing the blood concentration of caspofungin(CPFG),construct a prediction model,and validate the predictive effect of the model,so as to provide reference for the individualised dosing of patients with fungal infections in haematology.Methods Seventy-five patients admitted to the Department of Haematology,General Hospital of Tianjin Medical University,who were treated with CPFG for antifungal therapy during the period of March 2021 to June 2022 were selected as the study subjects,and CPFG blood concentration monitoring was carried out to explore the influencing factors of CPFG blood concentration and to construct a prediction model accordingly.Hosmer-Lemeshow(H-L)was used to test the goodness-of-fit of the model,and another 30 patients were selected as the verification group,and the predictive effect of the model was verified by the receiver's operating characteristics(ROC)curve.Results The mean blood concentrations of the patients at 0.5,9 and 24 h were(12.54±4.38),(6.80±2.76),(4.13±2.16)μg·mL-1,and the mean AUC0-24h were(152.05±57.60)µg·mL-1·h.AUC0-24h was lower than the reference value(98 μg·mL-1·h)in two patients.The results of correlation analysis showed that gender showed a correlation with 0.5 h blood concentration(P<0.05),and there was no correlation with the rest of the two time points blood concentration and AUC0-24h(P>0.05).Body weight and albumin(Alb)concentration showed correlation with 0.5,9,24 h blood drug concentration and AUC0-24 h(P<0.05),and the rest of the indicators showed no correlation with blood drug concentration and AUC0_24h at each time point(P>0.05).The results of multifactorial analysis showed that the factors influencing the patients'0.5 h blood concentration were gender,Alb concentration and body weight,and the factors influencing the 9 and 24 h blood concentration and AUC0-24h were Alb concentration and body weight(P<0.05).Correlation analysis showed that the daily dose was positively correlated with the plasma concentration of CPFG at 0.5,9 and 24 h and AUC0-24h(P<0.05).The results of multivariate analysis showed that the daily dose was also one of the influencing factors of the plasma concentration of CPFG(P<0.05).ROC curve shows that the model has good prediction ability.Conclusion Body weight and Alb are significantly associated with CPFG blood concentrations and area under the drug-time curve,which can be used as a basis for preventive risk avoidance.

caspofunginfungal infectionsblood concentrationinfluencing factorspredictive modelling

谢栋、毕重文、段蓉、王一浩、袁恒杰、李正翔

展开 >

天津医科大学总医院药剂科,天津 300052

天津医科大学总医院血液科,天津 300052

卡泊芬净 真菌感染 血药浓度 影响因素 预测模型

白求恩医学科学研究基金

B21028FN

2024

中国临床药理学杂志
中国药学会

中国临床药理学杂志

CSTPCD北大核心
影响因子:1.91
ISSN:1001-6821
年,卷(期):2024.40(12)
  • 3