There is a fundamental gap in understanding the maternal and neonatal effects of antenatal corticosteroid (ACS) administration in women with threatened preterm birth (PTB) who have diabetes. Since the initial discovery of ACS for neonatal benefit in 1972, more than 40 randomized controlled trials have been performed evaluating its efficacy. However, none of these trials have included women with T2DM, and there is limited data among women with gestational diabetes. While ACS have been shown to reduce neonatal morbidity associated with PTB in non-diabetic women, the side effects of ACS (maternal hyperglycemia and fetal hyperinsulinemia) may mitigate the neonatal benefit of ACS in women with diabetes. Before neonatal benefit of ACS can be evaluated in this population, the first step is to optimize maternal glycemic control after ACS. Previous studies evaluating maternal hyperglycemia after ACS have been limited by small sample size, retrospective study design, or insufficient glucose data. Use of continuous glucose monitoring (CGM) in a randomized clinical trial provides a unique opportunity to overcome these challenges. Our long-term goal is to improve maternal and child health among women with diabetes as an independently funded clinical researcher. The research objectives of this proposal are to test the efficacy of three treatment strategies at achieving maternal glycemic control after ACS and evaluate the association between maternal glycemic control and neonatal outcomes. Our central hypothesis is that treatment with a continuous insulin infusion will improve maternal glycemic control, which is key to improving neonatal outcomes, but at the cost of less patient satisfaction and more health resource utilization. This hypothesis will be tested by pursuing the following specific aims: 1) Test the efficacy of three treatment strategies (addition of sliding scale insulin, up-titration of home insulin, and continuous insulin infusion) at achieving maternal glycemic control after ACS and 2) Quantify the association between maternal glycemic control after ACS and neonatal morbidity. Completion of these aims will determine the optimal strategy to achieve maternal glycemic control after ACS and inform a larger, multicenter trial to improve neonatal outcomes among women with diabetes and threatened PTB.

Eligible, consenting women with medication-treated gestational or pregestational type 2 diabetes receiving ACS at 23 0/7 - 36 5/7 weeks' gestation for threatened PTB will be enrolled and randomized in a 1:1:1 ratio to receive 1) addition of sliding scale insulin to their home insulin regimen, 2) up-titration of their home insulin regimen, or 3) discontinuation of their home insulin regimen and initiation of a continuous insulin infusion. The randomization sequence will be created by the study statistician in a permuted block design, and assignments will be centrally allocated using the Research Electronic Data Capture (REDCap) application. Study investigators will be masked to the randomization sequence and varying block sizes.

Regardless of treatment group, all women will receive Dexcom G6 Professional CGM for 10 days. The device will be applied by research staff or if desired, by the patient herself, under direct supervision by research staff. The CGM sensor will be applied to the patient's abdomen away from skin folds, where there is hair, near the waistband or areas of scarring, tatoos, irritation or open wounds. An additional waterproof adhesive will be applied to help prevent the sensor from being dislodged and finally the transmitter will be attached. The patient will be provided a handout with instructions for care. The Professional CGM device will ensure both patients and providers are masked to CGM data, which will be used for research purposes only as CGM is not readily available to guide insulin titration in most clinical settings. Capillary blood glucose testing, the method routinely used in clinical practice, will be used to guide insulin titration for each treatment group. Women assigned to the sliding scale insulin group will continue their home insulin regimen and receive supplemental insulin as needed for post-prandial hyperglycemia based on capillary blood glucose testing 4 times daily (fasting and 1-hour post-prandial) for 5 days after ACS. Women assigned to the up-titration of home insulin group will have their home insulin dosages increased based on an algorithm with capillary blood glucose testing 4 times daily and additional sliding scale insulin as needed for post-prandial hyperglycemia. For example, if a woman's baseline insulin regimen includes a total of 70 units of daily insulin (NPH 30 units in the morning (qAM), NPH 10 units at bedtime (qPM), and Aspart 10 units with each meal), then on day 2 after ACS she will receive 50% more or 105 units total (NPH 45 units qAM, NPH 15 units qPM, and Aspart 15 units with each meal). Women assigned to the continuous insulin infusion group will have their home insulin discontinued and receive a continuous insulin infusion based on hourly capillary blood glucose testing for 5 days after ACS. Given the high acuity of insulin infusion management these women will be monitored on L&D, but women in the other two treatment groups may be managed on L&D or on the Antepartum unit at the discretion of the primary provider. Regardless of treatment group, all women will be allowed to eat as long as it is deemed safe by the primary provider. If women assigned to receive the sliding scale insulin group or the up-titration of home insulin group are made nil per os (NPO) at any time after enrollment, they will be placed on the continuous insulin infusion per protocol with hourly capillary blood glucose testing. Once a diet is resumed, they will be switched back to the insulin algorithm.

Upon completion of the study intervention, all women will complete the Diabetes Treatment Satisfaction Questionnaire (DTSQ) to assess satisfaction with their insulin treatment strategy. The DTSQ is one of the most widely used treatment questionnaires as it is internationally validated and approved by the World Health Organization and International Diabetes Federation and available in over 100 different languages. The questionnaire is composed of 8 questions, each of which are scored on a scale ranging from 0 ("very dissatisfied or inconvenient") to 6 ("very satisfied or convenient"). The first section assesses treatment satisfaction and includes 6 questions that ask about 1) satisfaction with treatment, 2) flexibility, 3) convenience, 4) understanding of diabetes, 5) recommend treatment to others, and 6) willingness to continue. The second section consists of 2 questions that assess the burden of hyper- and hypoglycemia. Overall treatment satisfaction is measured by the sum of the scores on the first 6 questions, and a higher score indicates higher satisfaction (maximum score of 36). The DTSQ is particularly well-suited for use in this study because it is able assess treatment satisfaction regardless of the specific treatment method, and it is easy to answer without placing a large burden on patients.

Other than glycemic management during the 5 days after ACS, antenatal care (fetal testing, maternal laboratory evaluation, timing and mode of delivery) will be at the discretion of the primary obstetric provider. Umbilical cord blood will be collected at delivery and stored at the University of Alabama at Birmingham (UAB) for analysis at the conclusion of the trial. After birth, all neonates born to women with diabetes have a heelstick performed to measure capillary blood glucose as part of standard of care. Additional neonatal care after birth will be at the discretion of the primary neonatal provider.

Comprehensive baseline maternal data and maternal and neonatal outcomes will be abstracted from the UAB electronic medical record and Professional CGM devices. Additionally, measures of health resource utilization will be collected such as duration of time on labor and delivery, number of capillary blood glucose tests, treatments administered such as insulin, intravenous fluids, and dextrose for hypoglycemia.

Inclusion Criteria:

• Gestational or pregestational type 2 diabetes mellitus treated with daily insulin injection(s) or oral hypoglycemic agents such as metformin
• Hospitalized for antenatal corticosteroid administration in anticipation of preterm birth
• Gestational age 23 0/7 weeks - 36 5/7 weeks
• Maternal age 18-50

Exclusion Criteria:

• Planned delivery < 72 hours after 1st dose of antenatal corticosteroids
• More than 16 hours after 1st dose of antenatal corticosteroids
• Major fetal anomaly
• Triplet or higher order multiple gestation

The primary aim of the study is an intervention unique to pregnancy. Male and female neonates will be included.