Effect of cerebral circulatory arrest on cerebral near-infrared spectroscopy in pediatric patients

BACKGROUND/AIMS: The aim was to investigate whether cerebral transcutaneous near-infrared spectroscopy (NIRS) or two-site NIRS is a suitable monitoring tool to detect or confirm a cerebral circulatory arrest in pediatric intensive care unit (PICU) patients.

METHODS: Prospective single-center pediatric observational study. Simultaneous NIRS measurements over forehead (cNIRS, crS02) and kidney (rNIRS, rrSO2), at the same time, the cardiac output were determined by transthoracic echocardiography. Area under the curve (AUC) in the receiver-operating curve (ROC) was analyzed for NIRS regarding cerebral circulatory arrest.

RESULTS: There were two groups of patients (weight 2.1-73 kg): Group A: patients with intact cerebral perfusion (n = 36). Group B: patients with cerebral circulatory arrest (n = 8) proven by Doppler ultrasound scan or perfusion scintigraphy. There was no difference in cardiac output between the groups. PICU mortality for Group A was 3/36 (8.3%), for Group B 8/8, (100%). Mean cNIRS values were significantly higher with 68.92 (SEM = 2.54, SD = 15.25) in Group A compared with 34.63 (SEM = 5.36, SD = 15.15) in Group B (P < 0.001). ROC analysis for cNIRS detecting cerebral circulatory arrest was significant (AUC 0.948, 95% confidence interval 0.876-1.000, SE = 0.037, P < 0.001). Discrimination was optimal at 46 for cNIRS, at 36.5 for the difference rNIRS-cNIRS and at 0.5646 for the quotient cNIRS/rNIRS. The probability of a cerebral circulatory arrest was 77.8% (cNIRS) and 87.5% (combinations of cNIRS and rNIRS) at these cutoffs.

CONCLUSIONS: cNIRS did detect cerebral circulatory arrest with high sensitivity. Specificity was, however, not high enough to confirm a cerebral circulatory arrest.