Scripties UMCG - Rijksuniversiteit Groningen
English | Nederlands

Neonatal Care: Can pain in neonates be detected by cerebral oxygenation and electrocerebral activity

(2016) Vries, R. de (Rosanne)

Although interest in neonatal pain has risen in recent years and more and more negative long
term consequences become known, the detection and quantification of neonatal pain still
difficult. In this study we tried to examine the efficacy of the lower border and bandwidth of
the aEEG signal and the regional oxygen saturation (rSO2) and fractional tissue oxygen
extraction (FTOE) as measured by near-infrared spectroscopy (NIRS) in pain detection. The
3 goals were to examine the changes of rSO2, FTOE, lower border and bandwidth by 1) a
painful stimulus, 2) analgesic medications and 3) to examine the effect of a painful stimulus
on the neurovascular coupling. In 17 neonates, we registered 28 skin breaking procedures,
that were part of routine care. Using a paired samples t-test we found a significant lower rSO2
(p<.05) and higher FTOE (p<.05) in the minute after the skin breaking event, compared to the
minute before, but not in the 2, 5, 10, 20 and 60 minutes intervals around the skin breaking
event. There were no significant changes in heart rate, respiratory rate, blood pressure and
arterial oxygen saturation in the 1, 10, 20, 60 minute interval, nor were there any changes in
lower border and bandwidth in the 10, 20 and 60 minute interval. There was a significant
trend of a lower FTOE in the morphine group compared to both the sucrose and no
medication group. The proportion of intact neurovascular coupling was significantly higher in
the 10 minutes before the skin breaking event compared to the 10 minutes after, but no
differences in other intervals were seen. We conclude that a higher FTOE and lower rSO2, in
the absence of changes in peripheral arterial oxygen saturation and heart rate, indicates
cerebral processing of the painful stimulus in the first minute. Further research is needed to
see whether these results can be replicated with a larger sample size. Another relevant
question for future research is whether and how this subclinical pain detected by NIRS should be treated.

To top