(P18-027-23) Assessing Glycemic Patterns in Response To Exercise in Pregnancies With Obesity – A Feasibility Study

Objectives: To assess the feasibility of collecting longitudinal data via continuous glucose monitoring (CGM) technology in singleton pregnancies with obesity during either a physical activity intervention or normal standard of care.

Methods: Ancillary to a randomized parallel-arm physical activity intervention study conducted at Arkansas Children’s Nutrition Center (ACNC, NCT02125149), we are implementing CGM in 24 sedentary women with BMI ≥30 kg/m², ≥18 years of age, recruited within weeks 11-13 of gestation; without chronic conditions such as type I or II diabetes mellitus or hypertension; and with obtained health care provider clearance to participate. Consented participants are wearing a CGM sensor (FreeStyleLibrePro^®, Abbot Diabetes Care, Inc., Alameda, CA) for 2 weeks within their 1^st, 2^nd and 3^rd trimester, aligned with RCT study visits. Participants re-consent and opt-in prior to each CGM period. CGM data is collected in single-blind manner, preventing participant reactivity. CGM data is processed using the cgmanalysis package (v2.7.6) in R (v4.1.2; R Core Team 2021).

Results: Eight women have thus far consented to the study; with a 100% enrollment and re-consent rate since ancillary study inception. There has been one adverse event: temporary bleeding at application site without consequential sensor removal or aberrant CGM readings. Of the 11 recorded CGM periods, 9 provide 1248 readings (13 full 24h periods), 1 provides 1152 readings and 1 provides 1056 readings, due to sensor replacement (accidental removal). Sensor cv is ≤ 20% for 7/11 CGM readings, with 4 ranging 21-28%. CV > 20% are explained by sensor readings spent below 54 mg/dL (75 – 1635 min per CGM period) in these 4 individuals. CGM-period mean glucose values are 87.5 mg/dL ± 16.9 (Control) vs 94.0 mg/dL ± 16.5 (Exercise). Those with aberrant (low) CGM readings show median values of 76 – 78 with estimated A1c values of 4.3 and 4.4, respectively. Though premature for statistical analysis, descriptive statistics indicate glycemic variation to be lowest around 5am and highest around 3pm.

Conclusions: CGM application in pregnancies with obesity is feasible, adds low participant burden and yields interpretable results. Data from this CGM system must be screened for aberrant readings below 54 mg/dL, before inclusion into analytic dataset.

Funding Sources: Supported by the ACRI, the ABI & the CCOP (under the NIH)