To assess the risk of persistent (>7 days) increases in brain parenchymal echogenicity in preterm infants and their association with known obstetric risk factors.Case–control study of prospectively collected data.A University hospital in Northern Italy.Eighty-five singleton infants between 24 and 34 weeks of gestation with a cranial ultrasonographic diagnosis of persistently increased parenchymal echogenicity without development of cystic degeneration, and 170 control infants with negative cranial ultrasonographic findings.A comparison of the prevalence of selected obstetric risk factors between infants with persistent echo-dense lesions and negative controls.Odds ratios of persistent echo-dense lesions including first-degree interactions between variables.After adjusting for birthweight, logistic regression analysis showed that the only factor associated with an increased risk of persistent brain echo-dense lesions in infants was multiple courses of antenatal steroids (OR= 2.14, 95% CI= 1.11–4.15,P= 0.024). In this group, the risk of persistent echo-dense lesions was particularly high in: (i) mothers receiving dexamethasone rather than betamethasone (Pvalue for interaction= 0.015) and (ii) after expectant management of pre-eclampsia or intrauterine growth retardation (Pvalue for interaction= 0.03).Multiple doses of antenatal steroids, especially dexamethasone, could influence the prevalence of persistent increases in brain parenchymal echogenicity in preterm infants. [ABSTRACT FROM AUTHOR]