Abstract: Objective To explore the causes of liver function injury associated with elemene emulsion injection in the real world. Methods The clinical data on 4 421 patients who had used elemene emulsion injection at least once in 21 big hospitals nationwide between January 1, 2003 and July 31, 2019 were collected from the integrated data warehouse of the large hospital information system (HIS) established by the Institute of Clinical Basic Medicine of Chinese Medicine of the Chinese Academy of Traditional Chinese Medicine. Twenty-seven patients with abnormal liver function were included. Through SOMTE sampling, 27 patients with liver function impairment and 108 patients in the control group were finally obtained. All the factors that affected normal liver function in the 135 cases were processed by Logistic regression, weighted Logistic regression of gradient elevator tendency scores without covariates and weighted Logistic regression of gradient elevator tendency scores with covariates. Results Logistic regression showed that there were 4 factors (P<0.05), namely, multiple secondary malignant tumors, combination with furosemide 7 days before administration, combined use with lidocaine (sirocarine), and vitamins. Further analysis using the tendency score method showed that multiple secondary malignant tumors had an impact on liver function. The use of furosemide 7 days before elemene emulsion injection might affect liver function. Whether lidocaine combined with elemene emulsion injection would lead to liver function injury needs to be studied, and vitamins combined with elemene emulsion injection would not affect liver function. Conclusion There is no compelling evidence that the use of elemene emulsion injection will lead to liver function injury. For more evidence, rigorous randomized controlled experiments are required.

Key words: elemene emulsion injection, propensity score, liver function injury, the real world