F83. Full 10–20 EEG application in hospitalised infants is not associated with an increase in stress hormone levels

Abstract

Introduction Neonatal EEG monitoring in pre-term and full-term infants offers a window onto neurological functioning, and full 10/20 electrode coverage captures important regional organisation. Nevertheless, 10/20 EEG application involves a relatively high level of handling and could be a potential stressor, especially in those infants who are already exhibiting higher stress levels. Salivary cortisol reflects activity of the hypothalamic pituitary axis: increased cortisol concentration indicates increased acute stress. To assess the impact of 10/20 EEG set-up on systemic stress levels in hospitalised infants, we measured salivary cortisol immediately before EEG application, and approximately 15 min after application (median time between swabs: 46 min), based on the latency of stressor-induced changes in cortisol concentration (Ramsay and Lewis, 2003). Methods Subjects included 33 un-sedated infants with gestational age 26 + 4 − 41 + 6 weeks + days (median 36 + 4), corrected age 32 + 3 − 47 + 6 weeks + days (median 37 + 5), and postnatal age 0.5–95 days (median 5 days). Cortisol concentration pre- and post-EEG application was compared with paired Wilcoxon tests. Statistical significance was set at 0.05. Results Median cortisol concentration pre-EEG application was 0.27 μ g/dL (range 0.09–2.91), and post-EEG application was 0.37 μ g/dL (range 0.03–1.74), values similar to those found in previous studies. There was no difference between cortisol concentration pre- and post-EEG application (p = .201). In particular, this was also the case for vulnerable sub-groups (corrected age: pre-term, n = 13; ward location: special care or high dependency unit, n = 14) (p ⩾ .432). Finally, we investigated whether baseline stress level influenced the effect of EEG application by splitting infants into those with higher pre-EEG cortisol concentration ( ⩾ .25 μ g/dL, n = 17) and lower pre-EEG cortisol concentration ( μ g/dL, n = 16). EEG application was associated with a decrease in cortisol concentration in infants who had a higher concentration at baseline (p = .023; median 0.31 μ g/dL decrease, which is a median 41% decrease in the pre-EEG value), and no change in infants who had a lower cortisol concentration at baseline (p = .120). Conclusion In this study we have shown that EEG application does not necessarily increase stress levels in hospitalised infants. Our findings are reassuring and support our clinical impression (based on behavioural observation, and review of heart rate and pulse oximetry acquired by the neonatal unit monitors where available) that EEG application can be well tolerated by both pre-term and full-term infants. Interestingly, we show that EEG application is actually associated with decreased stress levels in infants with higher cortisol concentration pre-EEG application. In line with this, tactile stimulation of the head and body is associated with decreased stress behaviours in neonates (Hernandez-Reif et al., 2007).