There is an association between congenital heart disease (CHD) and neurodevelopmental delay (42%). Brain injury can be minimized by adequate perfusion control based on NIRS (6% Vs 26%): a low NIRS index under 40% is a good predictor of brain lesion. Here, monitoring must be foolproof and easy to understand. Assessed parameters must imply the opportunity of an early intervention by the physician. Alas, NIRS is assessed peri-operatively. Our aim was to find an earlier predictor that could be used to assess brain vulnerability pre-operatively.

We prospectively enrolled 14 children (mean±SD: age=2.2±3.3 y.o.; weight=9.4±9 Kg) who underwent cardiopulmonary bypass for a CHD repair surgery. Cerebral oxygen saturation (rSo2) was monitored by NIRS. We calculated the ΔNIRS as de difference between NIRS level before the procedure and during the CPB. We compared the rSO2 data with the physiologic brain status assessed by a computed electroencephalogram (EEG spectral analysis). This later provided four spectral index (alpha, beta, delta, theta).

The NIRS decreased during CPB. ΔNIRS value was 24,21% (±16.28) .Two patients had a NIRS impairment below 40% (14.3%). There was no correlation between NIRS and either the spectral index alpha, delta or theta. Linear regression based on Beta index shows two significant relationships. The preoperative beta spectral index is correlated with the ΔNIRS (P=0,03*^{1* p value less than 0.05 - significant.} ) and with the minimum NIRS level during the CPB (P=0.01).

Preoperative Beta spectral index is therefore an early predictor of brain vulnerability. A high Beta index preoperatively could predict a low NIRS index during CPB and therefore a possibility of brain injury. That was a preliminary study and our findings need to be validated onto a wider panel of patients. Beta index is probably could be linked to the CHD type too. Nevertheless, beta index could help to design targeted procedure for brain protection adapted to each type of patient.