A prospective cohort study on glucose variability and clinical outcomes in comatose children due to acute central nervous system infections admitted in the pediatric intensive care unit

Abstract

Background: Pediatric acute central nervous system (CNS) infections are associated with severe neuromorbidity. This study aimed to study the effect of glucose variability on clinical outcomes in comatose children due to acute CNS infections admitted in pediatric intensive care unit (PICU). Subjects and Methods: A prospective cohort study enrolled comatose children aged 1 month to 12 years due to acute CNS infection. Within 6 h, continuous glucose monitoring was started (Freestyle Libre Pro, Abbott). The unit practice was targeting blood glucose (BG, mg/dL) of <140–145. The hyperglycemic index was calculated to estimate the relative time spent above BG of >126, >140, >180, >200, and <60. Glucose variability was defined as BG fluctuation, with both hypoglycemia (<60) and hyperglycemia (>126). The primary outcome was new-onset organ dysfunction. The secondary outcomes were organ support, length of mechanical ventilation, hospital (including PICU) stay, and 90-day composite poor outcome (mortality or severe neurodisability). Results: Total BG values measured were 27,792 from 66 patients (mean [standard deviation (SD)] 421.1 [212.6] values per patient). The mean (SD) BG was 103.2 (37.7) (minimum: 42.1; maximum: 228.8). The new-onset organ dysfunction has occurred in 83.3% (n = 55/66), and no difference was noted among normoglycemic and abnormal glycemic groups (84.4% vs. 80.9%; relative risk = 1.09, 95% confidence interval: 0.67–1.76). The median (interquartile range) PICU stay (days) was higher in the normoglycemic group (7, 5–14 vs. 4, 3.5–8.5; P = 0.014). No difference was noted in other outcomes. Conclusions: Glucose variability was not significantly associated with new-onset organ dysfunction and poor outcome in comatose children due to acute CNS infections.