Continuous Glucose Monitoring in Enterally Fed Children with Severe Central Nervous System Impairment.

Abstract

This prospective, observational study included nineteen patients (median (25-75 pc) age: 12.75 (6.17-15.55) years) with permanent CNS damage (Gross Motor Function Classification System V) receiving long-term tube enteral feeding, recruited from two paediatric university nutritional treatment centres. Patients with acute conditions and diagnosed diabetes were excluded. The nutritional status and nutritional support were analysed in all the inpatients in accordance with a uniform protocol. Using the CGM system (Medtronic iPro2), glycaemic curves were analysed, and in addition, HbA1C levels were determined in fourteen patients. CGM results were analysed using GlyCulator2.0. Statistical analysis was performed using the Statistica version 11 software (StatSoft Inc. Tulsa, OK, USA). More than half (11/19; 58%) of the patients were undernourished (BMI < 3 pc for age and gender), with the stature age being significantly lower than calendar age (5 (4.5-9) vs. 12.75 (6.17-15.55) years; p = 0.0010). The actual caloric intake was 50 (37.7-68.8) kcal/kg (median; 25-75 pc). In patients fed using the bolus method, the number of calories consumed per day was statistically significantly higher than in children subjected to a continuous feeding supply (56.00 (41.00-75.00) vs. 33.40 (26.70-50.00) kcal/kg BW (body weight; p = 0.0159). Decreases in blood glucose levels below the alarm level (<70 mg/dL) were recorded in fifteen patients (78.9%), including two patients with episodes of clinically significant hypoglycaemia (<54 mg/dL). The minimum and maximum glycaemic values recorded in any individual CGM records were 67 mg/dL (median) (minimum: 41 mg/dL; maximum: 77 mg/dL) and 146 (minimum: 114 mg/dL; maximum: 180 g/dL), respectively, for the entire recording. The maximum percentage of glycaemic concentrations > 140 mg/dL (TAR 140) recorded overnight in children with BMI ≥ 3 amounted to 1.6% vs. 0% in undernourished patients (TAR 140: 0.0 (0.00-1.6%) vs. 0% (0.00-0.0%; p = 0.0375); the percentage of glycaemic concentrations <70 mg/dL in the entire recording was comparable (0.77% (0.13-2.2%) vs. 1.8% (0.5-14.4%) vs. p = 0.2629). There was a positive correlation between the mean daily glucose recorded using the CGM method and patients' BMI z-scores (R = 0.48, p = 0.0397). No statistically significant relationship was demonstrated between the occurrence of alarm hypoglycaemia events in the CGM records and undernutrition expressed by BMI z-scores (OR = 1.50 (95%CI: 0.16-13.75), the type of diet (for commercially manufactured OR = 0.36 (95%CI: 0.04-3.52), and the modality of diet delivery (for bolus feeding OR = 2.75 (95%CI: 0.28-26.61). In children with chronic OU damage, enteral feeding is associated with a risk of hypoglycaemia, but further studies involving a larger number of patients are needed, and CGM might be a useful tool to estimate the metabolic adequacy of enteral nutritional support in terms of glucose control.