Reply to the correspondence letter: “Congenital CMV: are we treating too many?”

Abstract

Dear Editor, I read with interest the comments of LI Gonzalez-Granado regarding the criteria for selecting infants suffering from symptomatic congenital cytomegalovirus (CMV) infection for antiviral treatment. In our case series of infants [1], we chose to treat only infants with clinical or ultrasound findings related to central nervous system (CNS) involvement: microcephaly, choioretinitis, sensorineural hearing loss (SNHL) or brain ultrasound findings. Of the 23 infants included in the study, 19 had more than one site of CMV damage. Two out of the four infants had only one site SNHL. In the other two patients, ultrasound findings included brain calcification. The issue of lenticulostriated vasculopathy (LSV) as a sign of CNS involvement is still unclear; however, in our study, LSV was detected on brain ultrasound in only four infants. All four also had other signs of congenital CMV infection: chorioretinitis in two, moderate and severe SNHL in three, and microcephaly in one. We feel, as do others [4,5], that LSV in infants with congenital CMV is a sign of CNS involvement. We recently completed a study dealing with this issue. Based on our data, we conclude that LSV is a good predictor for SNHL in infants with congenital CMV.Most data about of the incidence of LSV were taken from hospital records of infants hospitalized in neonatal intensive care units due to either prematurity or serious medical problems [2,3]. Therefore we cannot say at this point that “LSV is a normal finding on brain ultrasound during the neonatal period” .W e agree that most infants with LSV do not suffer from congenital infection but those who have had a congenital infection such as CMV, findings of LSV is probably a subtle sign of CNS involvement.