关键词: 糖尿病；消化系统；腹部大手术；术后恢复质量；潜在剖面分析

Abstract: Objective: To explore the latent profile and influencing factors of early postoperative recovery quality in patients with diabetes mellitus undergoing major abdominal surgery for digestive system diseases, and to provide insights for the development of intervention strategies. Methods: Convenience sampling was employed to select 370 inpatients with diabetes mellitus who underwent major abdominal surgery for digestive system diseases from two tertiary grade A general hospitals in Zhenjiang and Wuhu from February to July 2025 as the research participants. The General Information Questionnaire, the Chinese version of the Quality of Recovery-40 Questionnaire (QoR-40), the Diabetes Specific Quality of Life Scale (DSQLS), the Hospital Anxiety and Depression Scale (HADS), and the Social Support Rating Scale (SSRS) were used for investigation. Latent profile analysis was performed on the early postoperative recovery quality of the participants, and Logistic regression analysis was conducted to explore the influencing factors of different classes. Results: The early postoperative recovery quality of diabetic patients after major abdominal surgery for digestive system diseases could be divided into three distinct latent classes: high-quality physical and mental balance group (39.46%), moderate-quality pain-dominant group (36.76%), and low-quality physical and mental distress group (23.78%). Age, operation duration, postoperative complications, duration of diabetes mellitus, preoperative blood glucose control, social support, and anxiety and depression levels were independent influencing factors of class membership (all P<0.05). Conclusion: Three distinct types of early postoperative recovery quality exist in diabetic patients undergoing major abdominal surgery for digestive diseases. Medical staff should implement targeted interventions according to these influencing factors.

Key words: diabetes mellitus; digestive system; major abdominal surgery; postoperative recovery quality; latent profile analysis