Objectives: Sleep characteristics have been used for prediction of neuro-developmental outcome. The aim of our study was to evaluate the influence of morphine and midazolam on the development of SWC in newborns > 32 weeks' gestational age after major non-cardiac surgery.

Study design: This prospective aEEG study included infants > 32+0 weeks' gestation admitted to the Neonatal Intensive Care Unit at The Royal Children's Hospital in Melbourne who were undergoing major non-cardiac surgery. The BrainZ Monitor (BRM2, Version 8.0, BrainZ Instruments, New Zealand) was applied post-operatively. The time of onset and quality of SWC and the maximum levels of morphine and midazolam as predictors of time to SWC were then assessed.

Results: Forty-seven eligible infants were included. Emergence of SWC was observed at a mean of 13 hours post-surgery. The maximum dose of morphine or midazolam was not predictive of time to SWC.

Conclusions: Despite high doses of continuous infusions of morphine and midazolam SWC was observed on aEEG in neonates > 32 weeks' gestational age soon after major non-cardiac surgery. The main type of aEEG background pattern was not affected by the maximum dose of either morphine or midazolam. Abnormalities in aEEG in post-surgical patients are not always drug related.

There is increasing evidence of neurodevelopmental impairment in survivors of non-cardiac surgery (1,2,3,4). Psychosocial maldevelopment and emotional problems have been reported to occur more commonly in children with repaired abdominal wall defects, and routine assessment of survivors of non-cardiac surgery has been recommended throughout later childhood (5,6). These reports also emphasize the need for further study of children who undergo major non-cardiac surgery in the neonatal period so that the neurological substrate for subsequent abnormal neurodevelopment can be better understood.

Amplitude-integrated electroencephalography (aEEG) is being increasingly used to monitor cerebral activity at the bedside in the neonatal intensive care unit (NICU) (7). Widely used scoring systems for aEEG incorporate consideration of background activity, presence of sleep wake cycling (SWC) and seizures (8). There are several aEEG reports indicating that analgesic and sedative agents routinely used in neonates suppress amplitude and therefore alter background activity (9,10,11,12,13) and it is believed that interpretation of aEEG in sedated patients is unreliable. Most of these studies have been conducted investigating pre-term infants or infants with hypoxic-ischaemic encephalopathy.

Sleep characteristics have been reported as predictors of neurodevelopmental outcome (14,15,16) and there is general agreement that indices of stability in SWC over time are positively correlated with improved clinical outcomes. SWC is described as smooth sinusoidal variations, mostly in the minimum amplitude. Phases with broader bandwidth represent discontinuous background activity during quiet sleep, whilst the phases with narrower bandwidth correspond to the more continuous activity during wakefulness and active sleep. SWC develops with increasing maturation of the child. From 31 to 32 weeks' gestational age, quiet sleep periods are clearly discernable in the aEEG trace as distinct periods with increased bandwidth. At term-equivalent age these periods represent trace alternant electroencephalography (EEG) patterns (17). The average duration of quiet sleep periods is 24 to 28 minutes for infants between 32 and 36 weeks' gestational age. These periods are slightly longer at night but are otherwise relatively stable and are not affected by incubator covers or developmental care interventions (18). The internal trigger for SWC is located in the brainstem and the presence of SWC in healthy term newborns is considered a sign of brain stem integrity (19,20). In newborns with hypoxic-ischemic encephalopathy sleep organization can be altered and the presence of SWC on aEEG is considered to be a good prognostic sign (21). Timing of onset of SWC has been shown to predict neurodevelopmental outcome based on whether the SWC returns before or after 36 hours of a presumed hypoxic insult (22).

There are several reports indicating that analgesic and sedative agents routinely used in neonates may lead to amplitude depression on aEEG. However, to our knowledge there are currently no reports of aEEG studies and SWC in newborn infants requiring major non-cardiac surgery during the neonatal period. Studying background activity and emergence of SWC on aEEG in neonates requiring major non-cardiac surgery may provide useful information about the relationship of analgesia and sedation and cerebral function in the post-operative infant. Knowledge of the effects of drugs that may modify cerebral activity is important to the interpretation of aEEG's in critically ill neonates requiring high doses of analgesics and sedatives.

The aim of this prospective observational aEEG study was to describe the influence of analgesic and sedative medication on background pattern and the development of SWC in newborn infants born > 32 weeks' gestation after major non-cardiac surgery.

Inclusion Criteria:

• > 32 weeks' gestational age
• abdominal surgery

Exclusion Criteria:

• < 32 weeks' gestational age
• neurologically compromised

newborns > 32 weeks' gestational age post abdominal surgery