Two-dimensional ultrasound measurements vs. magnetic resonance imaging-derived ventricular volume of preterm infants with germinal matrix intraventricular haemorrhage

Post-haemorrhagic ventricular dilatation can be measured accurately by MRI. However, two-dimensional (2-D) cranial US can be used at the bedside on a daily basis. To assess whether the ventricular volume can be determined accurately using US. We included 31 preterm infants with germinal matrix intraventricular haemorrhage. Two-dimensional cranial US images were acquired and the ventricular index, anterior horn width and thalamo-occipital distance were measured. In addition, cranial MRI was performed. The ventricular volume on MRI was determined using a previously validated automatic segmentation algorithm. We obtained the correlation and created a linear model between MRI-derived ventricular volume and 2-D cranial US measurements. The ventricular index, anterior horn width and thalamo-occipital distance as measured on 2-D cranial US were significantly associated with the volume of the ventricles as determined with MRI. A general linear model fitted the data best: ∛ventricular volume (ml) = 1.096 + 0.094 × anterior horn width (mm) + 0.020 × thalamo-occipital distance (mm) with R${}^2$ = 0.831. The volume of the lateral ventricles of infants with germinal matrix intraventricular haemorrhage can be estimated using 2-D cranial US images by application of a model.