基于倾向性评分的苦碟子注射液对肾功能影响的临床实效研究

该文旨在了解真实世界中苦碟子注射液对肾功能的影响,为临床用药安全提供参考。从中国中医科学院中医临床基础医学研究所建立的18家大型三甲医院信息系统(HIS)数据库中选取年龄在18~80岁的使用苦碟子注射液(暴露组,24 225例)和未使用苦碟子注射液(非暴露组,14 191例)的患者,通过倾向性评分法平衡混杂因素,采用经典Logistic回归、GBM加权倾向评分Logistic回归、带协变量的GBM倾向评分加权Logistic回归3种方法以及敏感性分析对肾功能的影响进行研究。结果显示,2组导致血肌酐(SCr)发生异常的可能性无显著性差异(3种分析方法的P=0.94,0.679,0.834),导致尿素氮(BUN)异常的可能性也无显著性差异(3种分析方法的P=0,0.045,0.164)。因此,基于现有数据,未发现苦碟子注射液对肾功能有损害。该研究为回顾性研究,且基于真实世界,仍可能存在未考虑到的混杂因素及潜在偏倚,因此需进一步重点监测苦碟子注射液对肾功能是否存在损害,以确保临床用药安全。

Clinical effect of Kudiezi injection on renal function based on propensity score

To explore the effect of Kudiezi injection on renal function in the real world, in order to provide the basis for the clinical medication safety. Patient aged between 18-80 were selected from 18 large hospitals information system (HIS) databases established by clinical research institute for basic traditional Chinese medicine of China academy of Chinese medical sciences. The patients who were treated with Kudiezi injection (24 225 cases) were defined as the exposed group, whereas those who were not treated with Kudiezi injection (14 191 cases) were defined as the non-exposed group. The propensity score method was used to balance the confounding factors. Classic logistic regression, GBM weighted propensity score logistic regression, GBM propensity score weighted logistic regression with covariate and sensitivity analysis were adopted to study the effect of Kudiezi injection on renal function. The results showed no significant difference in the possibility in abnormality in serum creatinine (Scr) (P=0.940, 0.679, 0.834) and urea nitrogen (BUN) (P=0, 0.045, 0.164) between both groups. Therefore, the existing data indicated no damage of Kudiezi injection on renal function. Because this study is a retrospective study based on the real world, there may be unknown confounding factors and potential bias. Therefore, further studies shall be conducted to monitor whether Kudiezi injection causes damage on renal function, in order to ensure the clinical medication safety.