Abstract: Objective：To investigate the association of 24 h urinary glucose between blood glucose fluctuation and islet function. Methods：A crosssectional survey of 124 inpatients with type 2 diabetes was conducted, the clinical data were collected, dynamic blood glucose monitoring system (CGMS) was applied to assess blood glucose fluctuation, the pancreatic β cell secretion function was calculated, aiming to investigate the association of 24 h urinary glucose between blood glucose fluctuation and islet function. Results：The average of the renal threshold for glucose (RTG) was 11.8 mmol/L and of which 75% were elevated (＞10 mmol/L). 24 h urinary glucose was 30.54 g. Correlation analysis demonstrated that 24 h urinary glucose was positively associated with fasting blood glucose (FBG), 2 h postprandial blood glucose (2 h PBG), the mean blood glucose (MBG), fasting plasma (preprandial) glucose area under the curve (AUC2), postprandial glucose area under the curve (AUC1) and glomerular filtration rate (eGFR), there was a significantly negative relationship between 24 h urinary glucose and the mean amplitude of glucose excursions (MAGE) and RTG , but not associated with course of disease, gender, age of patient, body mass index (BMI), blood pressure、absolute means of daily differences (MODD), the pancreatic β cell secretion function and HbA1c. Multiple stepwise regression analysis showed that 24 h urine sugar was correlated with MAGE, MBG, FBG, eGFR, RTG and AUC2 (absolute value of β’ was -0.668，0.437，0.148，0.116，-0.107, and 0.086, respectively; P＜0.05). Conclusions：The average of the renal threshold for glucose in patients with type 2 diabetes was elevated. The 24 h urinary glucose was related to blood glucose level, blood glucose fluctuations assessed by CGMS , eGFR and RTG, among which, MAGE influenced 24 h urinary glucose most. MAGE may be the most predictive risk factor for 24 h urinary glucose.