Predictive Role of F2-Isoprostanes as Biomarkers for Brain Damage after Neonatal Surgery

Figures

• 

  Figure 1. Time distribution sampling points.

• 

  Figure 2. Perioperative data of oxidative stress biomarkers. (a) Data of plasma NPBI at each time point. Dotted line indicates 2.3 umol/L NPBI, and normal values of plasma NPBI are below this cutoff value. (b) Data of plasma F2-isoprostane at each time point. Dotted line indicates 60 pg/mL F2-isoprostane, and normal values of plasma F2-isoprostane are below this cutoff value. (c) Data of urinary F2-isoprostane at each time point. Dotted line indicates 1.3 pg/mg creatinine F2-isoprostane, and normal values of urinary F2-isoprostane are below this cutoff value.

• 

  Figure 3. Examples of parenchymal and nonparenchymal brain injury. (a) Coronal T1-weighted image: cortical infarction and subdural haemorrhage; (b) coronal T1-weighted image: white matter lesion; (c) coronal T1-weighted image: white matter lesion; (d) susceptibility weighted image: multiple punctate cerebellar lesions; (e) diffusion-weighted image: thalamic infarction; (f) susceptibility weighted image: cerebellar haemorrhage.

• 

  Figure 4. A significant difference in plasma F2-isoprostane between “no injury” and “parenchymal injury” was found using the Mann–Whitney U test with post hoc Bonferroni correction (U = 11.0, p < 0.01, and r = −0.70). ∗Significant difference between parenchymal injury and no injury.

• 

  Figure 5. Receiver operating characteristic (ROC) curve analysis for NPBI at 72 hours after surgery. The area under the curve indicates that NPBI at 72 hours after surgery allows to differentiate neonates with nonparenchymal injury from no injury. The area under the curve was 0.785 (95% confidence interval: 0.595–0.975), with 66.7% specificity and 90.9% sensitivity. 1.34 μM/l or higher was identified as predictive threshold to have nonparenchymal injury.