Perioperative drug-related rare adverse reactions are often characterized by acute onset,high risk,and unpredictable,involving complex physiological and genetic factors.Currently,most perioperative pharmacovigilance relies on clinical monitoring and real-time data reporting by medical teams.However,due to scattered data and the masking of symptoms by anesthetics and pain medications,it is difficult to predict rare adverse reactions accurately and promptly.This paper systematically reviews the latest advancements in the integration of digital and intelligent technologies across various fields.Based on real-world data,our research team had leveraged digital-intelligence fusion technologies to deeply integrate big data with artificial intelligence,thereby constructing a standardized anesthesia-specific database.This enabled dynamic monitoring of vital signs,individualized risk prediction,and comprehensive analysis of multimodal data in real-world studies,providing an innovative solution for perioperative pharmacovigilance.The aim of this paper is to enhance the personalization and intelligence of perioperative drug safety management,thereby offering more effective protection for patient medication safety during the perioperative period.

Perioperative drug-related rare adverse reactions are often characterized by acute onset,high risk,and unpredictable,involving complex physiological and genetic factors.Currently,most perioperative pharmacovigilance relies on clinical monitoring and real-time data reporting by medical teams.However,due to scattered data and the masking of symptoms by anesthetics and pain medications,it is difficult to predict rare adverse reactions accurately and promptly.This paper systematically reviews the latest advancements in the integration of digital and intelligent technologies across various fields.Based on real-world data,our research team had leveraged digital-intelligence fusion technologies to deeply integrate big data with artificial intelligence,thereby constructing a standardized anesthesia-specific database.This enabled dynamic monitoring of vital signs,individualized risk prediction,and comprehensive analysis of multimodal data in real-world studies,providing an innovative solution for perioperative pharmacovigilance.The aim of this paper is to enhance the personalization and intelligence of perioperative drug safety management,thereby offering more effective protection for patient medication safety during the perioperative period.

Objective To investigate the feasibility and safety of OLV anesthesia about infant lung resection of CCAM by video-assisted thoracicscopy. Methods Endo-tracheal intubation was performed after 43 CCAM infants had undergone rapid intravenous induction. One side of lungs was ventilated by injecting 4 ~ 6 mmHg CO2 for the construction of artificial pneumothorax, and the side lung was compressed forming OLV. SpO2, ECG, MAP, PETCO2, T, PaO2, PaCO2, bleeding volume and urine volume were monitored. The numerical value of SpO2, PaO2, HR, MAP, PETCO2, and PaCO2 were recorded at scheduled intervals. Results Compared with 5min after induction,the PaO2,HR and MAP of the infants significantly reduced; the PETCO2 and PaCO2 significantly increased at OLV at 10 min and 60 min. Compared with OLV at 10 min, the PaO2, PETCO2 significantly increased at OLV 60 min. Conclusion Appropriate respiratory management and drug usage are feasible and safe for infant surgery of CCAM by video-assisted thoracicscopy.