Abstract
Human cytomegalovirus (CMV) transmitted through breast milk poses fatal risks to preterm infants. However, current molecular assay systems often do not accommodate breast milk samples. In this study, we evaluated the analytical and clinical performance of the measurement procedure of CMV load in breast milk utilizing the Cobas CMV test on the Cobas 6,800 system. This was enabled by incorporating a simple independent sample preparation procedure before the application of samples on the automated assay system. Clinical data from electronic medical records were retrospectively analyzed. Breast milk samples from mothers of preterm infants born before 33 weeks of gestation were screened for CMV using the automated assay system. CMV positivity rates in breast milk and neonatal samples and the CMV transmission rate were calculated. Furthermore, to validate the analytical accuracy of the overall measurement procedure with newly obtained residual breast milk samples, the linearity of the measurement procedure was assessed, and a simplified sample preparation method was validated against a conventional method. The CMV positivity rates in maternal breast milk and neonatal samples were 57.8 and 5.2%, respectively. The CMV transmission rate through breast milk was 7.7%. No significant differences in gestational age or birth weight were found between the CMV-negative and CMV-positive neonates. The linearity of the procedure was observed within a range of 1.87-4.73 log IU/mL. The simplified sample preparation method had an equivalent or even improved CMV detection sensitivity than the conventional method. Incorporating a simple independent sample preparation procedure effectively resolved any potential issues regarding the application of breast milk on the automated assay system. Our approach contributed to reduced vertical transmission of CMV by providing a convenient and reliable method for the monitoring of breast milk CMV positivity for clinicians.