Carbon Dioxide and Glucose Affect Electrocortical Background in Extremely Preterm Infants

OBJECTIVES: To investigate if PaCO(2) and plasma glucose levels affect electrocortical activity. METHODS: Ours was an observational study of 32 infants with a gestational age of 22 to 27 weeks. We performed simultaneous single-channel electroencephalogram (EEG) and repeated blood gas/plasma glucose analyses during the first 3 days (n = 247 blood samples with corresponding EEG). Interburst intervals (IBIs) and EEG power were averaged at the time of each blood sample. RESULTS: There was a linear relationship between PaCO(2) and IBI; increasing PaCO(2) was associated with longer IBIs. One day after birth, a 1-kPa increase in PaCO(2) was associated with a 16% increase in IBI in infants who survived the first week without severe brain injury. EEG power was highest at a PaCO(2) value of 5.1 kPa and was attenuated both at higher and lower PaCO(2) values. Corrected for carbon dioxide effects, plasma glucose was also associated with IBI. Lowest IBI appeared at a plasma glucose level of 4.0 mmol/L, and there was a U-shaped relationship between plasma glucose level and EEG with increasing discontinuity at glucose concentrations above and below 4.0 mmol/L. CONCLUSIONS: Both carbon dioxide and plasma glucose level influenced EEG activity in extremely preterm infants, and values considered to be within normal physiologic ranges were associated with the best EEG background. Increasing EEG discontinuity occurred at carbon dioxide levels frequently applied in lung-protection strategies; in addition, moderate hyperglycemia was associated with measurable EEG changes. The long-term effects of changes in carbon dioxide and glucose on brain function are not known. Pediatrics 2011; 127: e1028-e1034