Targeted Screening with the Use of Clinical Risk Factors for Detecting Congenital Cytomegalovirus Infection in Newborns: A Prospective Multicenter Cohort Study.

Abstract

Congenital cytomegalovirus infection (cCMV) is one of the most common congenital infections. This study aimed to evaluate the diagnostic performance of targeted screening with the use of clinical risk factors for cCMV. A total of 3063 pregnant women and their 3139 newborns were enrolled. Six clinical findings consisting of maternal fever or flu-like symptoms during pregnancy (fever/flu-like symptoms), hospitalization for threatened miscarriage or preterm labor before 34 weeks of gestation, preterm delivery before 34 weeks of gestation, fetal ultrasound abnormalities, small for gestational age (SGA), and refer results of automated auditory brainstem response screening (AABR refer) were defined as cCMV risk factors before participant registration. All newborns underwent urine cytomegalovirus polymerase chain reaction tests within one week of birth. The predictive accuracy of these six risk factors was analyzed. Nine (0.29%) of the three thousand one hundred and thirty-nine newborns were diagnosed with cCMV, having at least one of the six risk factors. Logistic regression analysis identified fever/flu-like symptoms (odds ratio (OR), 7.5; 95% CI, 1.9-30.3), fetal ultrasound abnormalities (OR, 17.9; 95% CI, 4.4-72.8), SGA (OR, 6.8; 95% CI, 1.8-25.6), and AABR refer (OR, 75.5; 95% CI, 19.7-289) as significant risk factors. The predictive accuracy of the targeted screening for cCMV, when at least one of the six risk factors was present, yielded 100% sensitivity (95% CI, 55.5-100) and 70.7% specificity (95% CI, 69.1-72.3), with a Youden index of 0.707. When at least one of the four significant risk factors was present, 100% sensitivity (95% CI, 55.5-100) and 81.2% specificity (95% CI, 79.8-82.6) with the maximum Youden index of 0.812 were achieved. In conclusion, targeted screening with the use of clinical risk factors in mothers and their newborns could effectively identify cCMV.