Impact of sevoflurane anesthesia on brain oxygenation in children younger than 2 years

To assess the impact of sevoflurane and anesthesia-induced hypotension on brain oxygenation in children younger than 2 years.

Inhalational induction with sevoflurane is the most commonly used technique in young children. However, the effect of sevoflurane on cerebral perfusion has been only studied in adults and children older than 1 year. The purpose of this study was to assess the impact of sevoflurane anesthesia on brain oxygenation in neonates and infants, using near-infrared spectroscopy.

Children younger than 2 years, ASA I or II, scheduled for abdominal or orthopedic surgery were included. Induction of anesthesia was started by sevoflurane 6% and maintained with an expired fraction of sevoflurane 3%. Mechanical ventilation was adjusted to maintain an endtidal CO2 around 39 mmHg. Brain oxygenation was assessed measuring regional cerebral saturation of oxygen (rSO2 c), measured by NIRS while awake and 15 min after induction, under anesthesia. Mean arterial pressure (MAP) variation was recorded.

Hundred and ninety-five children were included. Anesthesia induced a significant decrease in MAP (-27%). rSO2 c increased significantly after induction (+18%). Using children age for subgroup analysis, we found that despite MAP reduction, rSO2 c increase was significant but smaller in children ≤ 6 months than in children >6 months (≤ 6 months: rSO2 c = +13%, >6 months: rSO2 c = +22%; P < 0.0001). Interindividual comparison showed that, during anesthesia at steady-state with comparable CMRO2, rSO2 c values were significantly higher when MAP was above 36 mmHg. And the higher the absolute MAP value during anesthesia was, the higher the rSO2 c was. We observed a rSO2 c variation ≤ 0 in 21 patients among the 195 studied, and the majority of these patients were younger than 6 months (n = 19). No increase or decrease of rSO2 c during anesthesia despite reduction of CMRO2 can be explained by a reduction of oxygen supply. Using the ROC curves, we determined that the threshold value of MAP under anesthesia, associated with rSO2 c variation ≤ 0%, was 39 mmHg in all the studied population (AUC: 0.90 ± 0.02; P < 0.001). In children younger than 6 months, this value of MAP was 33 mmHg, and 43 mmHg in children older than 6 months.

Despite a significant decrease of MAP, 1 MAC of sevoflurane induced a significant increase in regional brain oxygenation. But subgroup analysis showed that MAP decrease had a greater impact on brain oxygenation, in children younger than 6 months. According to our results, MAP value during anesthesia should not go under 33 mmHg in children ≤6 months and 43 mmHg in children >6 months, as further changes in MAP, PaCO2 or hemoglobin during anesthesia may be poorly tolerated by the brain.