Increased propofol consumption with later anesthesia start times in sedated gastrointestinal endoscopy: Insights from regression and machine learning models

Qiong Lan¹Zhuonan Sun¹Tian Wang¹Zhuya Huang²Dengyang Han¹Taotao Liu¹Hua Zhang¹Ye Wang¹Rui Zhang¹Binlong Li¹Ning Yang¹Yinyin Qu¹Huili Liu^1*Mao Xu¹

• ¹Peking University Third Hospital, Beijing, China
• ²Beijing University of Posts and Telecommunications School of Electronic Engineering, Beijing, China

Background: Chronopharmacology is an important but underexplored aspect of propofol administration. Despite the implementation of propofol administration models, none have yet incorporated temporal variables. This study aims to investigate the impact of temporal variations on propofol administration during sedated gastrointestinal (GI) endoscopy. Moreover, we aim to develop regression models to predict manually-controlled propofol administration that integrate temporal variables. Methods: This prospective single-center cohort study enrolled patients undergoing sedated GI endoscopy. For analysis, patients were categorized into 4 groups based on the anesthesia start time: Group 1 (8:00-10:00), Group 2 (10:00-12:00), Group 3 (13:00-15:00), and Group 4 (15:00-17:00). Perioperative characteristics and propofol doses were compared across groups. Correlation analysis was conducted to evaluate the relationship between propofol dose and the anesthesia start time. Subsequently, linear regression models were developed for manually-controlled propofol administration. Results: A total of 146 cases were included in the statistical analysis. Significant differences were found for all parameters related to propofol dose across the 4 different groups, including induction dose, maintenance dose, total dose and these doses per kilogram per hour. Furthermore, there were positive correlations between the anesthesia start time and all parameters. In the linear regression models, the induction dose equation incorporated the anesthesia start time, age and weight as variables. The model of the maintenance dose per kilogram per hour included the anesthesia start time, duration and weight as variables. Conclusions: The results suggest that propofol dose increases with later anesthesia start times. Therefore, further clinical administration of propofol should incorporate a heightened consideration of temporal factors. Trial registration: This prospective study has been registered in the Chinese Clinical Trial Registry (Registration date: December 3, registry number ChiCTR2400093328).