Conclusions
From all four neuromarkers studied, only GFAP was demonstrated to be a good diagnostic and prognostic factor in cyanotic CHD patients. NSE had only prognostic value.
Methods
Pediatric patients with congenital heart defects (CHDs) undergoing elective cardiac surgery were included. Neurodevelopmental functioning was assessed preoperatively and 4-6 months postoperatively using the Denver Developmental Screening Test II. Blood samples were collected preoperatively and postoperatively. During surgery, regional cerebral tissue oxygen saturation was monitored using near-infrared spectroscopy (NIRS).
Results
Forty-two patients were enrolled and dichotomized into cyanotic and non-cyanotic groups based on peripheric oxygen saturation. Nineteen patients (65.5%) had abnormal developmental scores in the non-cyanotic group and eleven (84.6%) in the cyanotic group. A good diagnostic model was observed between NIRS values and GFAP in the cyanotic CHD group (AUC = 0.7). A good predicting model was observed with GFAP and developmental scores in the cyanotic CHD group (AUC = 0.667). A correlation was found between NSE and developmental quotient scores (r = 0.09, p = 0.046). Conclusions: From all four neuromarkers studied, only GFAP was demonstrated to be a good diagnostic and prognostic factor in cyanotic CHD patients. NSE had only prognostic value.