302 - Continuous Glucose Monitoring in Infants with Intestinal Failure Reveals Variable Hypoglycemia Burden During Cycled Parenteral Nutrition

Publication Number: 3291.302

Jessica L. Ruiz, Boston Children's Hospital, Cambridge, MA, United States; Alexandra Carey, Harvard Medical School, Needham, MA, United States; Megan Gray, Baxter Healthcare, Quincy, MA, United States; Lisa Asaro, Boston Children's Hospital, Boston, MA, United States; Rachael Levy, Boston Children's Hospital, Boston, MA, United States; Allison S. Bernique, Boston Children's Hospital, Boston, MA, United States; Michael Agus, Harvard Medical School, Boston, MA, United States; Christina M. Astley, Boston Children's Hospital, Boston, MA, United States

Background: Infants receiving cycled parenteral nutrition (PN) are at risk of hypoglycemia during PN off periods given persistent endogenous insulin action and insulin-glucose mismatch. Continuous glucose monitoring (CGM) captures clinically-actionable glucose trends that may be missed with intermittent glucose checks.

Objective: We used CGM to characterize the severity and timing of trough glucose during PN off periods in infants with intestinal failure (IF) initially transitioning to cycled PN as part of usual care.

Design/Methods: This prospective, observational study enrolled hospitalized infants age 2-18 months weighing >4 kg with IF and without dysglycemia-predisposing comorbidities/medications. Study participants wore blinded CGM devices for up to 30 days when clinically transitioning from continuous to cycled ( < 24 hr/day) PN. Per standard of care, PN was typically tapered off in 2 steps over 60 minutes, and point-of-care glucose was monitored at 30 minutes and halfway through the PN off period. Daily trough glucose was defined as the lowest CGM reading during each PN off period. Minimum trough glucose was the lowest trough glucose per subject across all days. For readings below the CGM reporting limit (40 mg/dL), the last measured value was selected as the trough. Trough glucose was characterized relative to 60 mg/dL, the hypoglycemia threshold used in our IF PN practice.

Results: The study population (N=9) was 33% female, 2.8-8.1 months, 4.5-8.0 kg and with a mean (standard deviation) of 16.1 (6.8) cycling days and 3.2 (1.4) daily hours off PN per participant. During PN off periods, minimum trough glucose averaged 52.3 (11.6) mg/dL and the daily trough glucose averaged 70.3 (13.9) mg/dL (Figure 1). Six participants (67%) ever experienced trough glucose < 60 mg/dL and the proportion of daily trough glucose < 60 mg/dL ranged from 0-73.9% per participant (Figure 2). Among daily trough glucose < 60 mg/dL, 28.6% occurred during the 30 minutes off PN and the remaining 71.4% were distributed across the remaining time off PN (Figure 3).

Conclusion(s): CGM in infants with IF during PN cycling revealed two-thirds of infants with any trough glucose < 60 mg/dL, occurring predominantly beyond 30 minutes into the PN off period. Standard monitoring at 30 minutes may miss clinically significant hypoglycemia. Future analyses will assess the relationship between glucose infusion rate, clinical features and trough glucose, accounting for CGM lag time. This research informs hypoglycemia risk in infants with PN dependence due to IF and will help optimize monitoring strategies during PN cycling.

Figure 1
Fig1_850_850.jpegBox-and-whisker plots of minimum trough glucose (left) and daily trough glucose (right) CGM readings during PN off periods across study participants. Black horizontal lines inside the boxes denote median values, whiskers outside the boxes extend to the smallest/largest values no further than 1.5 × interquartile range. The dashed orange horizontal line indicates glucose of 60 mg/dL, our clinical hypoglycemia threshold.

Figure 2
Fig2_1050_850.jpegProportion of daily troughs below 60 mg/dL across study participants. Bars represent the proportion for each individual participant, arranged in descending order from the highest to the lowest proportion.

Figure 3
Fig3_1050_850.jpegTiming of troughs below 60 mg/dL across all possible PN off periods. Main graph represents the timing of all troughs in 1-hour increments. Inset graph represents the timing of troughs that occurred during the first hour of the PN off period, in 15-minute increments.