Studies on the Value of Diffusion-Weighted MR Imaging in the Early Prediction of Periventricular Leukomalacia

Periventricular leukomalacia (PVL) is the most common cerebral injury in premature infants. While cranial ultrasonography and conventional magnetic resonance imaging (MRI) offer little for its early diagnosis, still they are much favored at present. Based on the pathologic mechanism of ischemic cerebral injury in PVL, we attempt to elucidate the effects of diffusion-weighted MR imaging (DWI) for the early diagnosis of PVL. Twelve cases of PVL in premature infants confirmed by MRI examinations were included in the current study. All cases underwent DWI in addition to conventional MR imaging (T1-weighted [T1W] and T2-weighted [T2W]) 7 days, 2 weeks, and 4 similar to 8 weeks after birth, respectively. Initial DWI examination (carried out at an average of 4.5 days after birth) revealed symmetrical and diffuse high signal in bilateral periventricular areas in all cases, while conventional MRI demonstrated normal signal. DWI examination of 2 weeks later showed irregular inhomogeneous signals in cerebral white matter, while conventional MRI revealed hyperintensity on T1WI and slight hypointensity on T2WI in corresponding areas. Four weeks later, DWI revealed various intensities of cystic low signals beside lateral ventricles, while conventional MRI showed hypointensity on T1WI and hyperintensity on T2WI in corresponding lesion, ie, change in cystic PVL. Four months later, conventional MRI evidenced gradual shrink and disappearance of cystic cavity, reduction in cerebral white matter, and ventricle enlargement. Symmetrical, diffuse hyperintensity in both lateral periventricular white matter on DWI was the earliest sign of PVL in premature infants. Imaging abnormalities provided by DWI had a close correlation with the results obtained by conventional MRI performed at the advanced stage of PVL, which suggests that DWI may be an important sequence for the early evaluation of cerebral white matter injury as well as for the prediction of PVL occurrence in premature infants. J Neuroimaging 2009;19:13-18.