Preoperative carbohydrate drink intake attenuate insulin resistance. However, carbohydrate loading may compromise blood glucose control in patients with diabetes. There is some evidence that oral carbohydrate loading may be safe in patients with type 2 diabetes. Therefore, the investigators aim to evaluate whether preoperative carbohydrate intake affects insulin resistance and glycemic variability in patients with diabetes. Fifty patients scheduled for total knee arthroplasty will be divided into carbohydrate (n=25) and control (n=25) groups. Randomly selected patients of the carbohydrate group are given 400ml of 12.8 g/100 ml carbohydrate beverage 2-3 hours before their scheduled operation. In contrast, patients in the control group are fasted from water 2 hours before surgery according to standard protocol.

Inclusion Criteria:

• 1. Patients undergoing total knee arthroplasty

Exclusion Criteria:

• 1. Patients with contraindications to spinal anesthesia (local infection, clotting disorders, anatomical abnormalities, sepsis, etc.)
• 2. Patients with gastroesophageal reflux disease, gastric emptying disorders, inflammatory bowel disease, or previous treatment for intra-abdominal cancer
• 3. Patients with chronic renal disease or severe cardiovascular disease
• 4. HbA1c >69 mmol/mol or BMI >30 kg/m2
• 5. A duration of ≥5 hours between consumption of CHO and initiation of surgery.
• 6. The subject is a foreigner or illiterate