Skip to main content
SMFM 2024 Pregnancy Meeting Main banner
Final Program


Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Additional registration fee required
Additional registration fee required
Faculty have requested this content not be shared on social media
Faculty have requested this content not be shared on social media
CME Credit Offered
CME Credit Offered
Poster Icons
Foundation Awardee
Foundation Awardee
No Social Media
No Social Media
Back

Oral Concurrent Session 2 - Diabetes

Oral Concurrent Sessions

(32) Classifying patients based on weekly continuous glucose monitoring metrics and their association with pregnancy outcomes

Tuesday, February 13, 2024
3:15 PM – 3:30 PM  
Location: Potomac Ballroom C, Ballroom Level
Objective: Continuous glucose monitoring (CGM) data is complex and can be difficult to interpret. Using machine learning techniques, we sought to 1) identify groups of patients based on CGM metrics, and 2) compare patient characteristics and outcomes across groups. 

Study Design: Retrospective cohort study of patients with type 1 or type 2 diabetes who used real-time CGM during pregnancy and delivered at a tertiary center (2019-2023). CGM data were collapsed into weekly glycemic summaries related to centrality, spread, excursions, and circadian cycle patterns. Principal component analysis and k-means clustering were used to identify 5 discrete groups, and patients were assigned to the group that best represented their CGM patterns. Finally, patient characteristics and outcomes were compared across groups.

Results: We identified five glucotype groups using 2040 weeks of CGM data from 145 patients: 1) well-controlled and sub-optimally controlled with 2) nocturnal hyperglycemia, 3) daytime hyperglycemia, 4) early morning hyperglycemia, and 5) high variability with fasting hypoglycemia and daytime hyperglycemia (Figure). Patients in groups 1 and 2 were more likely to be older and have type 2 diabetes whereas groups 3, 4 and 5 more likely to be younger and have type 1 diabetes (p< 0.01 for both). Groups 1, 2 and 5 were more likely to be non-Hispanic Black and groups 3 and 4 non-Hispanic White (p< 0.05). Insurance and parity were similar between groups (p=0.44 and 0.87). With regards to outcomes, groups 2 and 5 were most likely to have NICU admission and neonatal hypoglycemia (p< 0.01 for both). Groups 2 and 4 were most likely to have preterm birth, with group 5 next (p< 0.01). Preeclampsia and cesarean delivery did not differ (p=0.08 and 0.40).

Conclusion: Weekly CGM metrics successfully classified pregnant patients into discrete glucotype groups integrating glycemic excursions and circadian cycle patterns. Nocturnal hyperglycemia and high glycemic variability with hypo- and hyperglycemia should be avoided as these patients appear most likely to experience adverse outcomes.