Relationship between enlargement of the lateral ventricle and periventricular leukomalacia in infants.

AIM

To examine if we could predict periventricular leukomalacia (PVL) from the area of the lateral ventricle (LV).

METHODS

Six neonates in whom cystic PVL could be detected by magnetic resonance imaging (MRI) but not by ultrasound (US) were termed the 'invisible group'. Six neonates in whom cystic PVL could be detected by MRI and US were termed the 'visible group'. Eleven neonates in whom cystic PVL could not be detected by MRI or US were termed the 'control group'. The ratio of LV to head circumference (HC) was calculated as the area of LV (cm²)/HC (cm) × 100. Receiver operating characteristic (ROC) curve analysis was carried out to find the cutoff value.

RESULT

There were no significant differences among the three groups with respect to gestational age, birthweight, postnatal age and HC. The ratio of LV to HC in the control group was a median value of 0.38, it was 0.79 in the invisible group, and 0.96 in the visible group. The ratio was significantly higher in the visible group (P < 0.001) and in the invisible group (P < 0.05) than in the control group. This ratio was low in the two infants who had PVL only in the collateral trigone. The ROC curve suggested a cutoff value of 0.6 (sensitivity 79.17%, specificity 100%) to suspect PVL.

CONCLUSION

We may need to suspect PVL in infants whose lateral ventricle is enlarged even if cystic PVL is not detected by ultrasound. PVL present only in the collateral trigone needs to be evaluated using cerebral MRI.