Patients with indication will receive MDI treatment with at least one basal insulin injection plus 2 to 3 prandial insulin injection.

Time above range (TAR) and time below range (TBR) calculated by CGM data at 24, 28 and 34 weeks of gestation.

Mean amplitude of glucose excursion (MAGE), coefficient of variation (CV) and standard deviation (SD) calculated by CGM data recorded at 24, 28 and 34 weeks of gestation.

The area under curve (AUC) of peripheral blood glucose within 24 hours before delivery: (a) >7.8 mmol/l or 140 mg/dl (b)>6.7 mmol/l or 120 mg/dl (c) <3.5 mmol/L or <63 mg/dl (d) <2.8 mmol/L or <50 mg/dl.

(a)Episodes of severe hypoglycemia requiring assistance. (b)Episodes of mild-moderate episodes of hypoglycemia < 3.5mmol/L (mild) and < 2.8 mmol/L (moderate) from patients' SMBG data or from CGM data defined as AUC <3.5 mmol/L or AUC less than or equal to 2.8 mmol/L for 20 minutes duration. (c) Nocturnal hypoglycemia defined as glucose <3.5 (mild) and <2.8 (moderate) by SMBG or CGM between the hours of 23.00-07.00.

The total daily insulin dosage at randomization, 4 weeks after randomization, and 24, 28, 34 weeks of gestation as well as 6 weeks of postpartum.

Hypertension during pregnancy (up to 42 weeks of gestation): Incidence of worsening of chronic hypertension, gestational hypertension, preeclampsia.

Incidence of caesarean section (primary and total)

Gestational weight gain: Absolute and relative weight gain at 24, 28, 34 weeks of gestation and 6 weeks of postpartum comparing to baseline (4-8 weeks of gestation at the time of enrollment)

Length of hospital stay including admission for delivery and for other obstetric situations during pregnancy.

Infant birthweight (at birth): Infant birthweight>90th centile using customized growth curves; infant birthweight<10th centile using customized growth curves; infant birthweight≥4000g or ≤2500g.

Including miscarriage, stillbirth and neonatal death (≤28 days of life).

Infant Outcomes (at birth): The incidence of preterm delivery (<37 weeks and early preterm <34 weeks) and post-term delivery (>42 weeks).

Infant Outcomes (at birth): Apgar score

The incidence of birth injury, shoulder dystocia, neonatal hypoglycemia with intravenous dextrose.

Incidence of hyperbilirubinemia, respiratory Distress Syndrome (RDS), NICU admission > 24 hours.

The combined adverse outcomes of infants including miscarriage, stillbirth and neonatal death; Neonatal birth injury, shoulder dystocia, neonatal hypoglycemia (requiring intravenous glucose infusion), neonatal jaundice, neonatal respiratory distress syndrome (RDS) and more than 24 hours of treatment in NICU.

The score of World health organization (WHO)-5 physical and mental health index acquired from patients.

The WHO-5 scale (1998 version) was used to measure the patient's quality of life. The scale contains 5 items which initial points is 0-25. The initial points are multiplied by 4 to obtain the percentage points, ranging from 0 to 100，which are used to monitor possible changes in physical and mental health. 0 represents the worst possible quality of life, 100 represents the best possible quality of life, and a difference of 10% indicates significant changes.

The score of Self-manage behavior scale acquired from patients. Diabetes self-care behaviors includes a range of activities (e.g., eating diabetic food, exercising, glucose monitoring and taking medicine), which were evaluated by the Summary of Diabetes Self-Care Activities (SDSCA). SDSCA, compiled by Toobert in 2000, is an 8-likert scale containing 11 items belongs to 6 dimensions, could estimate general diet, special diet, physical activity, glucose monitoring, foot care and medicine compliance of patients with T2DM. The total scores range from 0 to 77, higher score means better diabetes self- care behaviors.

The score of self-efficacy scale for diabetes mellitus acquired from patients. Self-Efficacy for Diabetes (SED), a 5-Likert scale including 9 items, is used to measure patients' self-efficacy. The average score is 1-5 points, and the higher the average score, the higher the self-efficacy level of participants. The effectiveness and internal consistency of the Chinese version of SED are reliable, and the load factors of each factor are between 0.579-0.922.

The score of Self-rated Anxiety Scale (SAS) acquired from patients. Self-Rating Anxiety Scale (SAS) is used to measure the degree of anxiety in patients. There are 20 items in total. The scores are divided into 4 levels, including 5 (items 5, 9, 13, 17, 19) reverse scoring items and 15 positive scoring items. Add the scores of the 20 items to get the rough score, then multiply it by 1.25 and take the integer part to get the standard score. The higher the standard score, the more serious the degree of anxiety.

The score of Self-rated Depression Scale (SDS) acquired from patients. The Self-Rating Depression Scale (SDS) is used to measure the degree of depression in patients. There are 20 items in total, including 10 (items 2, 5, 6, 11, 12, 14, 16, 17, 18, and 20) reverse scoring items, and 15 positive scoring items. The depression severity index can be calculated by the cumulative score of each item divided by 80, which could reflect the degree of depression. The index range is 0.25-1.0, and the higher the index, the more severe the degree of depression.