425 - Neonatal glycemic control and visual short-term memory at 12 months in very preterm infants: evidence from early continuous glucose monitoring

Poster Number: 425
Publication Number: 425.223

Silvia Guiducci, University of Padova: Department of Women' and Children Health, Padova, Veneto, Italy; Sabrina Brigadoi, Department of Developmental and Social Psychology, Università degli Studi di Padova, Italy, Padova, Veneto, Italy; Maria Cusinato, University of Padova, Padova, Veneto, Italy; Giulia Res, University of Padova, Toronto, ON, Canada; Livia Avagliano, University Study of Padua, Cava de' Tirreni, Campania, Italy; Federica Savio, Università degli studi di Padova, Padova, Veneto, Italy; Elena Priante, Neonatal Intensive Care Unit, Women's and Children's Health Department of Padova, Padova, Veneto, Italy; Eugenio Baraldi, Dipartimento Salute Donna e Bambino, Padova, Veneto, Italy; Alfonso Galderisi, University of Padova, Padova, Veneto, Italy

Background: Neonatal glycemic control has been associated with executive function development, with a debated role for the impact of hypoglycemia, hyperglycemia, and glucose variability. Recently, subcutaneous continuous glucose monitoring (CGM) has been proven to safely provide, even in preterms, continuous data regarding the time of exposure to different glucose ranges.

Objective: To evaluate the relationship between percentage time below 72mg/dL and above 144mg/dL during CGM worn for 5 days after birth in very preterm neonates and visual short-term memory (VSTM), a precursor of executive functions at 12-month of corrected age (CA). 

Design/Methods: This is an interim analysis from the “BabyGlucolight study” (NCT04347590) conducted at the University of Padova. We enrolled preterm infants born at the Neonatal Intensive Care Unit of Padua ≤32 weeks gestational age (GA) or birthweight ≤1500 g (BW). Follow-up was performed at 12 months CA using Bayley Scales of Infant and Toddler Development III (BSID-III) and the simultaneous stream change detection task, measuring VSTM (Fig.1). Visual preference for the “non-changing” or the “changing” stream of stimuli was expressed as the percentage time spent looking at that stream. A preference for the “changing” stream would indicate a reliable VSTM at the given set size.

Results: Data were available for 9 patients: 3 females, GA 30±2 and BW M 1456g±493. At 12 months of CA, BSID-III score was within the normal range for all domains: cognition (102,22±11,49), language (97,11±8,75) and motor (97,78±6,32), without statistically significant relation with CGM metrics. For the VSTM, we observed a visual preference for the “changing” stream at both set sizes (55% set size 1; 52% set size 2), which was however beneath the expected value for 12 months old children. Furthermore, we found a negative relationship between the percentage time >144mg/dL and the visual preference for the “changing” stream, especially at set size 2 (ρ -0.698; p-value 0.05). Conclusion(s): Our preliminary data support the hypothesis that time spent in mild hyperglycemic range ( >140mg/dL) may affect long-term executive function in vey preterm infants.
Fig. 1. Simultaneous stream change detection task.In this task, the infant is presented with a black screen with two white rectangles displayed on both sides of fixation. In the set size 1 condition, a single stimulus is presented within each rectangle. For each block of trials, lasting approximately 20 s, the stimulus in one hemifield is kept constant (“non changing” stream), whereas the stimulus in the other hemifield is varied at each trial (“changing” stream). In the set size 2 condition, two stimuli are presented within each rectangle and, at each trial, one of stimuli in the “changing” stream will be different than the one in the previous trial. Set size and side of “changing” stream are randomly varied during the experiment, which lasts until infants keep attention to the monitor.