Patterns of Dysglycemia Identified by Continuous Glucose Monitoring among Critically Ill Children in Malawi and Bangladesh

Abstract Dysglycemia is a critical metabolic disturbance associated with mortality in acutely ill children, yet its burden may be underrecognized in low-income settings due to reliance on single point-of-care measurements. Using continuous glucose monitoring (CGM), we aimed to characterize glucose patterns in acutely ill children of different anthropometric status. Methods Children aged 2-23 months admitted with acute illness were prospectively recruited from two hospitals in Bangladesh and Malawi. Clinical data were collected, and interstitial glucose was monitored for 48 hours using the Dexcom G4 Platinum system. Glucose excursions and variability were analyzed and associated with anthropometric status. Results Of 93 enrolled children, 88 had sufficient CGM data: 21 not wasted (NW), 22 moderately wasted (MW), and 45 with severe malnutrition (SAM; 29 severe wasting [SW], 16 edematous malnutrition [EM]). Low-glucose excursions were detected in 8 (38%) children with NW, 11 (50%) with MW, 12 (41%) with SW, and 10 (63%) with EM. While not confirmed hypoglycemia, these low-glucose excursions were longer and more frequently below severe thresholds in children with EM. Hyperglycemic excursions occurred in 31% of children and were longer in children with SAM compared to NW (median 41 vs. 23 min, p<0.0001). Overall, 35% of children maintained euglycemic profiles, while others exhibited marked glucose variability. Conclusion CGM revealed frequent glucose instability among acutely ill children, with patterns varying across anthropometric groups. When interpreted cautiously, CGM may serve as a research tool to detect dysglycemia and assess response to therapeutic or nutritional interventions in critically ill children in low-resource settings.