Comparison of a subcutaneous versus intravenous insulin protocol for managing hyperglycemia following antenatal betamethasone in women with diabetes: a pilot randomized controlled trial

Abstract

Introduction

Evaluate the safety and efficacy of a subcutaneous insulin (SC-I) versus intravenous insulin (IV-I) protocol for optimizing maternal blood glucose levels (BGLs) post-betamethasone administration.

Methods

Randomized controlled in-patient pilot study in pregnant women with diabetes, excluding type 1 diabetes, receiving betamethasone ≥24 weeks’ gestation. Interventions were stratified SC-I and IV-I protocols, titrated to hourly BGLs (IV-I) or predicted maternal hyperglycemia and 2–4 hourly BGLs (SC-I). Primary outcome was percentage at-target BGL 4.0–8.0 mmol/L over 48 h post-betamethasone. Secondary outcomes were rates of maternal hyperglycemia (>8.0 mmol/L), hypoglycemia (<4.0 mmol/L) and neonatal hypoglycemia (≤2.5 mmol/L).

Results

19 women (3 with type 2 diabetes [T2DM], 4 with gestational diabetes [GDM]-diet, 12 GDM-insulin) were randomized to a SC-I (n = 13) or IV-I (n = 6) protocol in a 9-month period. There was a non-significant trend for higher mean percentage at-target BGLs with SC-I vs IV-I (87% vs 81%; p = .055); this was significant when the cohort was restricted to women with GDM (89% vs 81%; p = .04). Maternal hyperglycemia (85% vs 100%; p = .31) and hypoglycemia (54% vs 33%; p = .41) were not significantly different, but there were no BGLs <3.8 mmol/L with IV-I (vs 4 women with SC-I; p = .13). The rate of neonatal hypoglycemia was not different between groups.

Conclusion

A SC-I or IV-I protocol controls maternal BGLs following betamethasone, but SC-I appears safe and minimizes labor intensive IV-I in GDM. An adequately powered RCT to assess superiority of SC-I is planned.

Keywords:

• Preterm delivery
• glucocorticoids
• diabetes
• insulin
• neonatal hypoglycemia

Acknowledgments

The authors would like to thank Royal Prince Alfred Hospital Departments of Women & Babies and Endocrinology, Dr Joel Lasschuitt and Dr Jaime Lin for their support and to ADIPS for their ADIPS-Novo Nordisk Education Research Grant.

Disclosure statement

None.

Additional information

Funding

This research was funded by an Australasian Diabetes in Pregnancy Society (ADIPS)/NovoNordisk Education Research Grant.