Estimating intracranial volume using intracranial area in healthy children and those with childhood status epilepticus

Background
Correcting volumetric measurements of brain structures for intracranial volume (ICV) is important in comparing volumes across subjects with different ICV. The aim of this study was to investigate whether intracranial area (ICA) reliably predicts actual ICV in a healthy pediatric cohort and in children with convulsive status epilepticus (CSE).

Methods
T1-weighted volumetric MRI was performed on 20 healthy children (control group), 10 with CSE with structurally normal MRI (CSE/MR-), and 12 with CSE with structurally abnormal MRI (CSE/MR+). ICA, using a mid-sagittal slice, and the actual ICV were measured.

Results
A high Spearman correlation was found between the ICA and ICV measurements in the control (r = 0.96; P < 0.0001), CSE/MR− (r = 0.93; P = 0.0003), and CSE/MR+ (r = 0.94; P < 0.0001) groups. On comparison of predicted and actual ICV, there was no significant difference in the CSE/MR− group (P = 0.77). However, the comparison between predicted and actual ICV was significantly different in the CSE/MR+ (P = 0.001) group. Our Bland–Altman plot showed that the ICA method consistently overestimated ICV in children in the CSE/MR+ group, especially in those with small ICV or widespread structural abnormalities.

Conclusions
After further validation, ICA measurement may be a reliable alternative to measuring actual ICV when correcting volume measurements for ICV, even in children with localized MRI abnormalities. Caution should be applied when the method is used in children with small ICV and those with multilobar brain pathology.