Universal screening for congenital Cytomegalovirus infection using saliva from the neonate

Background

Congenital Cytomegalovirus infection (cCMV) is the most common intrauterine infection in developed countries. The diagnostic gold standard is detection of CMV in neonatal urine. Considering difficulties in urine sample collection, saliva is an alternative specimen providing supportive material for early diagnosis and timely intervention.

Objectives

Aim of this study was to evaluate a universal screening program using saliva as target specimen, and urine as confirmatory material.

Results

Between May 2019 and November 2024, 5168 newborns were screened for CMV shedding in saliva, 45 neonates were tested positive. From 44 neonates a urine sample was examined. Twelve out of forty-four neonates had concordant results (prevalence = 0,232 %) with five newborns showing CMV related symptoms. In saliva, the difference between the Ct-value of the cCMV negative and the cCMV positive group was significant (38.59 vs. 28.56, p-value < 0.0001). At a Ct-value cut-off of 38.8 sensitivity was 100 %, specificity 53.1 %, positive predictive value 44 %, and negative predictive value 100 %. ROC analysis yielded an AUC value of 0.930. The mean virus load in urine was 31,266,663 copies/ml.

Discussion

This study reports a prevalence of 0,232 % for cCMV infection in the catchment area of the hospital Steyr. PCR based CMV detection in saliva is a good screening tool, but one must be aware of false positive results, arising from possible contamination. In our cohort CMV positive vaginal or cervical secretion seems more likely being the source of contamination than CMV positive breast milk. The high false positive rate in saliva makes confirmatory testing in urine samples mandatory.