Prediction of Congenital Cytomegalovirus Infection in High-Risk Pregnant Women

Transcription

1 Clinical Infectious Diseases MAJOR ARTICLE Prediction of Congenital Cytomegalovirus Infection in High-Risk Pregnant Women Kenji Tanimura, 1 Shinya Tairaku, 1 Yasuhiko Ebina, 1 Ichiro Morioka, 2 Satoshi Nagamata, 1 Kana Deguchi, 1 Mayumi Morizane, 1 Masashi Deguchi, 1 Toshio Minematsu, 3 and Hideto Yamada 1 1 Department of Obstetrics and Gynecology, and 2 Department of Pediatrics, Kobe University Graduate School of Medicine, Kobe; and 3 Research Center for Disease Control, Aisenkai Nichinan Hospital, Miyazaki, Japan Background. This prospective study aimed to determine maternal clinical, laboratory, and ultrasound findings that effectively predict the occurrence of congenital cytomegalovirus (CMV) infection (CCI) in high-risk pregnant women. Methods. Three hundred CMV immunoglobulin (Ig) M positive pregnant women were enrolled. The maternal clinical and laboratory findings, including serum CMV IgM and IgG; IgG avidity index (AI); antigenemia assay (C7-HRP); polymerase chain reaction (PCR) for the detection of CMV-DNA in the maternal serum, urine, and uterine cervical secretion; and prenatal ultrasound findings, were evaluated. To determine predictive factors for the occurrence of CCI, logistic regression analyses were performed. Results. In 22 of the 300 women, CCI was confirmed using PCR for CMV-DNA in newborn urine. Univariate analyses demonstrated that the presence of maternal flu-like symptoms, presence of ultrasound fetal abnormalities, serum titers of CMV IgM, positive results for C7-HRP, CMV IgG AI <40%, and positive PCR results in the uterine cervical secretion were statistically associated with the occurrence of CCI. Multivariable analysis revealed that the presence of ultrasound fetal abnormalities (odds ratio [OR], 31.9; 95% confidence interval [CI], ; P <.001) and positive PCR results in the uterine cervical secretion (OR, 16.4; 95% CI, ; P <.001) were independent predictive factors of CCI in CMV IgM-positive women. Conclusions. This is the first prospective cohort study to suggest that the presence of CMV-DNA in the maternal uterine cervical secretion and ultrasound fetal abnormalities are predictive of the occurrence of congenital CMV infection in high-risk pregnant women. Keywords. cytomegalovirus; congenital infection; PCR for CMV-DNA; pregnancy; ultrasound examination. Human cytomegalovirus (CMV) is the most common cause of congenital viral infection in humans. In the United States, an estimated children are born annually with congenital CMV infection, resulting in an estimated 8000 children with long-term sequelae, including hearing difficulty and mental retardation [1]. The annual economic costs of caring for these affected children are estimated at $1 to $2 billion [1, 2]. It has been demonstrated that early diagnosis using polymerase chain reaction (PCR) for CMV-DNA in newborn urine samples and early intervention with antiviral drugs may improve neurological outcomes in symptomatic congenital CMV-infected infants [3, 4]. However, universal screening of newborns for congenital CMV infection is not recommended. Therefore, prenatal Received 27 May 2016; editorial decision 2 October 2016; accepted 18 October 2016; published online October 20, Correspondence: H. Yamada, Department of Obstetrics and Gynecology, Kobe University Graduate School of Medicine, Kusunoki-cho, Chuo-ku, Kobe , Japan (yhideto@ med.kobe-u.ac.jp). Clinical Infectious Diseases 2017;64(2): The Author Published by Oxford University Press for the Infectious Diseases Society of America. All rights reserved. For permissions, journals.permissions@oup.com. DOI: /cid/ciw707 detection of mothers and newborns at high risk for congenital CMV infection is important. When pregnant women have primary CMV infection during the first trimester, approximately 25% of their fetuses will be infected [5]. Therefore, infants born to mothers with primary CMV infection during pregnancy are at high risk for the occurrence of congenital CMV infection. Maternal serum CMV immunoglobulin (Ig) M antibody is often tested to identify primary infection. However, true primary CMV infection is determined in only 20% 25% of pregnant women with positive results for CMV IgM [6]. This is because CMV IgM may persist for 6 9 months following primary infection [7] or may be detected during latent reactivation [6]. The serum CMV IgG avidity index (AI) is used as a confirmatory test to determine primary infection during pregnancy [8]. In addition, fetal CMV infection occurs in 1% 2.2% of pregnant women with reactivation of a latent virus or reinfection with a new strain of CMV [9, 10]. It is well known that ultrasound fetal abnormalities, including ventriculomegaly, microcephaly, intracranial calcification, fetal growth restriction (FGR), pleural effusion, ascites, hepatosplenomegaly, and periventricular/intestinal high Prediction of Congenital CMV Infection CID 2017:64 (15 January) 159

2 echogenicity, are closely associated with fetal CMV infection [11 14]. PCR assay for CMV-DNA in the amniotic fluid is highly sensitive and specific for fetal CMV infection, though amniocentesis is essentially an invasive procedure with a risk of membrane rupture [15]. On the other hand, it has been found that virus isolation in maternal urine or uterine cervical secretions is a poor indicator of congenital CMV infection [16] and that results of PCR and antigen assays in maternal blood fail to correlate with a risk of congenital infection or the disease severity [17, 18]. However, no one has determined whether a positive PCR result for CMV-DNA in maternal urine or uterine cervical secretion is predictive of congenital CMV infection. Our aim in this prospective cohort study was to determine maternal clinical, laboratory, and ultrasound findings that effectively predict the occurrence of congenital CMV infection among pregnant women with positive results for CMV IgM. METHODS Study Design and Participants The Kobe University Hospital Institutional Review Board approved this study. Pregnant women who visited or were referred to the university hospital from April 2009 to February 2016 were enrolled. They underwent blood screening for CMV IgG and CMV IgM with informed consent until 22 weeks of gestation (GW). When pregnant women were referred from another hospital or clinic beyond 22 GW, they underwent the antibody screening at the first visit. Approximately 1 2 weeks later, pregnant women who had positive tests for both CMV IgG and CMV IgM received a series of examinations for CMV infection after written informed consent was obtained. The workup for congenital CMV infection was performed in all newborns and stillbirths but not for induced abortion. Procedures Serological tests for CMV IgM (negative, <0.8; borderline, ; and positive, 1.2 index) were performed using enzyme immunoassay kits (Denka Seiken, Tokyo, Japan). Pregnant women positive for CMV IgM underwent clinical interviews concerning the presence of maternal fever or flu-like symptoms during pregnancy. The peripheral white blood cell count, serum C-reactive protein (CRP) measurements (positive, >0.3 mg/dl), and serum levels of liver enzymes, including alanine aminotransferase (ALT; normal, 34 IU/L), aspartate aminotransferase (AST; normal, 31 IU/L), and γ-glutamyl transpeptidase (GGT; normal, 30 IU/L) were determined. Additionally, serum CMV IgG measurements (Siemens Healthcare Diagnostics, Tokyo, Japan) (negative, 230; borderline, ; and positive, 240), serum CMV IgG avidity measurements (Aisenkai Nichinan Hospital, Miyazaki, Japan), and plasma C7-HRP assay (CMV antigen test; TEIJIN TFB, Tokyo, Japan) (positive, 1 positive cell/slide) were performed. The serum CMV IgG AI was measured as described previously [19]. A cutoff value for IgG AI of <40% was used for the statistical analyses [20]. Simultaneously, real-time PCR assays for the detection of CMV-DNA in the maternal serum, urine, and uterine cervical secretion (SRL, Tokyo, Japan) (positive, copies/ml) were performed. To collect cervical specimens, a cotton-tipped swab (JCB Industry, Tokyo, Japan) was placed in the uterine cervical canal and rotated several times for about 10 seconds. The swab were removed and placed into a specimen container that contained 1 ml of sodium dodecyl sulfate and ethylenediamine tetraacetic acid disodium salt. The cervical specimen was stored at 4 C until measurement. Ultrasound examinations were performed by obstetricians (K. T., S. T., and H. Y.) using the Voluson E8 Expert system (GE Healthcare, Milwaukee, Wisconsin) or the ProSound Alpha 7 system (Hitachi-Aloka Medical, Tokyo, Japan) until 22 GW. Pregnant women referred from another hospital or clinic beyond 22 GW received ultrasound examinations for fetal CMV infection at the first visit. The following were defined as ultrasound findings suggestive of congenital CMV infection: FGR, ventriculomegaly, microcephaly, intracranial calcification, pleural effusion, ascites, hepatosplenomegaly, and hyperechoic bowel of the fetus. FGR was defined as an estimated fetal body weight less than mean 1.5 standard deviation for gestational age. The presence of congenital infection was determined by a positive PCR result for CMV-DNA in the urine of newborns within 2 weeks of age. If the pregnancy ended in stillbirth, congenital infection was confirmed by PCR tests for CMV-DNA in the umbilical cord blood or immunohistochemistry of placental tissues. Statistical Analyses Maternal clinical characteristics, laboratory findings, and prenatal ultrasound findings among cases with congenital CMV infection and cases without infection were compared. Differences between the 2 groups were analyzed using the Mann Whitney U test, Fisher exact test, or χ 2 test. Statistical significance was considered to be present at P value <.05. A stepwise approach was used to determine clinical, laboratory, and ultrasound findings that effectively predicted the occurrence of congenital CMV infection. To avoid overfitting in multivariable logistic regression analyses, the number of variables in the final model of multivariable analyses was restricted to a maximum of 10% of the case number. Variables with P values <.05 in the final model of multivariable logistic regression analyses were determined to be independent predictive factors of congenital CMV infection. All statistical analyses were performed using SAS software, version 9.1 (SAS Institute Inc., Cary, North Carolina). 160 CID 2017:64 (15 January) Tanimura et al

3 RESULTS A flow diagram of study participants is shown in Figure 1. During the study period, 3057 pregnant women visited or were referred to the university hospital with no information about CMV infection. A total of 258 of the 3057 pregnancies ended in spontaneous miscarriages, and 277 women were referred from another hospital or clinic because they had borderline or positive results for CMV IgM. A total of 3076 women underwent blood screening for CMV IgG and CMV IgM, and 301 women with positive results for both CMV IgG and CMV IgM received a series of examinations for CMV infection described in the Procedures section (Figure 1). One pregnancy ended in induced abortion. Therefore, the remaining 300 pregnant women with positive results for both CMV IgG and CMV IgM were evaluated. Congenital CMV infection was confirmed in 22 cases, including 12 newborns with symptomatic infection, 7 newborns with asymptomatic infection, and 3 stillbirths with asymptomatic infections. Nine newborns with symptomatic congenital CMV infection received antiviral therapies. Table 1 shows the clinical characteristics of the22 pregnancies with congenital CMV infection and the 278 without congenital infection. GW at maternal CMV examinations in the congenital infection group (median 25 GW, range 12 35) was later than that in the no infection group (18 GW, 6 36; P <.001). However, GW at delivery in the congenital infection group (36 GW, 21 40) was earlier than that in the no infection group (39 GW, 22 41; P <.001). The birth weight in the congenital infection group (2192 g, ) was less than that in the no infection group (2966 g, ; P <.001). Prenatal clinical, laboratory, and ultrasound findings are shown in Table 2. The number of women who had positive results for C7-HRP and CMV-DNA in serum, urine, and uterine cervical secretions was 5, 0, 19, and 27, respectively. The Figure 1. Flow diagram of the study participants. During the study period, 3076 pregnant women underwent blood screening for cytomegalovirus (CMV) immunoglobulin (Ig) G and CMV IgM, and 301 women with positive results for both CMV IgG and CMV IgM received a series of examinations for CMV infection. One pregnancy ended in induced abortion. The remaining 300 pregnant women with positive results for CMV IgG and CMV IgM were evaluated. Congenital CMV infection was confirmed in 22 cases, including 12 newborns with symptomatic infection, 7 newborns with asymptomatic infection, and 3 stillbirths with asymptomatic infections. Abbreviations: CRP, C-reactive protein; PCR, polymerase chain reaction; WBC, white blood cell. Prediction of Congenital CMV Infection CID 2017:64 (15 January) 161

4 Table 1. Clinical Characteristics of 300 Women With Positive Results for Cytomegalovirus Immunoglobulin M Characteristic Congenital Infection n =22 No Congenital Infection n = 278 P Value Age, y 29 (19 38) 32 (16 44) 0.07 Gravidity 1 (0 7) 1 (0 9) 0.9 Parity 1 (0 3) 1 (0 6) 0.3 Percentage of referral Weeks of gestation at maternal examination 25 (12 35) 18 (6 36) <0.001 Weeks of gestation at delivery 36 (21 40) 39 (22 41) <0.001 Birth weight, g 2192 ( ) 2966 ( ) <0.001 Data are expressed as the median (range) or %. median (range) GW of positive PCR results for the uterine cervical secretion in 11 cases with congenital CMV infection was 25 GW (range, 13 34), and no cases had samplings at delivery. GW of negative PCR results for the uterine cervical secretion in the other 11 cases with congenital CMV infection was 26 GW (12 35). There was no statistical difference in GW between the 2 groups. The proportion of pregnant women with ultrasound fetal abnormality (P <.001), positive C7-HRP result (P <.001), and positive PCR result in the uterine cervical secretion (P <.001) in the congenital infection group was higher than those in the no infection group. However, the proportion of pregnant women with maternal fever or flu-like symptoms, elevated white blood cell counts (>12,000/µL), a positive test for CRP, elevated liver enzymes (ALT, AST, or GGT), and positive PCR result in urine was not significantly different between the 2 groups. No women had positive PCR result in the serum. The serum titers of CMV IgM in the congenital infection group (median 5.5, range, index) were higher than those in the no infection group (2.1, index; P <.001; Table 2). The CMV IgG AI in the congenital infection group (35.5%, 0.9% 77.8%) was lower than that in the no infection group (61.3%, 3.1% 94.3%; P <.005). The serum titers of CMV IgG were not significantly different between the 2 groups. Supplemental Figure 1 shows the CMV IgG AI of all pregnant women according to weeks of gestation at maternal examination. Fifty-nine (19.7%) of the 300 women had an AI of <40%. Thirteen (59.1%) of the 22 women with congenital CMV infection had an AI of <40%. Table 3 shows ultrasound fetal abnormalities found in all pregnant women. Eighteen women had at least 1 ultrasound finding suggestive of congenital CMV infection, including FGR (n = 15), ventriculomegaly (n = 7), microcephaly (n = 1), intracranial calcification (n = 1), pleural effusion or ascites (n = 5), hepatosplenomegaly (n = 5), and fetal hyperechoic bowel (n = 4). The presence of congenital CMV infection was confirmed in 11 of the 18 newborns. In this study, ventriculomegaly, microcephaly, intracranial calcification, and hepatosplenomegaly were detected only in fetuses with CMV infection. Univariate logistic regression analyses demonstrated that the presence of maternal fever or flu-like symptoms (odds ratio [OR], 2.5; 95% confidence interval [CI], ; P <.05), the Table 2. Comparison of Prenatal Clinical, Laboratory, and Ultrasound Findings Maternal Clinical, Laboratory, and Ultrasound Findings Congenital Infection n = 22 No Congenital Infection n = 278 P Value Maternal fever or flu-like symptoms, % White blood cell >12,000/µL, % Positive for C-reactive protein, % Elevation of liver enzymes, % Ultrasound fetal abnormality, % <0.001 CMV IgG 11,834 ( ,898) ( ) 0.1 CMV IgM, index 5.5 ( ) 2.1 ( ) <0.001 Positive for antigenemia assay <0.001 (C7-HRP), % CMV avidity index, % 35.5 ( ) 61.3 ( ) <0.005 Positive polymerase chain reaction results for CMV In the serum, % 0 0 Not applicable In the urine, % In the uterine cervical secretion, % <0.001 Data are expressed as the median (range) or %. Abbreviations: CMV, cytomegalovirus; Ig, immunoglobulin. 162 CID 2017:64 (15 January) Tanimura et al

5 Table 3. Ultrasound Fetal Abnormalities Associated With Congenital Cytomegalovirus Infection Group and Case Number Fetal Growth Restriction Ventriculo- megaly Microcephaly Intracranial Calcification Pleural Effusion or Ascites Hepatospleno- megaly Hyperechoic bowel Congenital infection n = 11 1 X 2 X X 3 X X X X 4 X 5 X 6 X X X 7 X X X X 8 X X X X 9 X X X X 10 X X 11 X X X X No congenital infection n = 7 12 X X 13 X 14 X 15 X 16 X 17 X 18 X X indicates the presence of the index ultrasound finding; indicates the absence of the index ultrasound finding presence of ultrasound fetal abnormality (OR, 35.2; 95% CI, ; P <.001), the serum titers of CMV IgM (OR, 1.3; 95% CI, ; P <.001), positive results for C7-HRP (OR, 15.3; 95% CI, ; P <.001), CMV IgG AI <40% (OR, 8.2; 95% CI, ; P <.001), and positive PCR results in the uterine cervical secretion (OR, 18.0; 95% CI, ; P <.001) were statistically associated with the occurrence of congenital CMV infection (Table 4). Multivariable logistic regression analyses of the 2 factors with the lowest P values in univariate analyses revealed ultrasound fetal abnormalities (OR, 31.9; 95% CI, ; P <.001) and positive PCR results for the uterine cervical secretion (OR, 16.4; 95% CI, ; P <.001) were significant independent factors for the prediction of congenital CMV infection in women with positive results for CMV IgM (Table 4). If multivariable logistic regression analyses were performed for 6 factors with P <.05 in univariate analyses, the same 2 factors the presence of ultrasound fetal abnormalities (OR, 37.2; 95% CI, ; P <.001) and positive PCR results for the uterine cervical secretion (OR, 5.2; 95% CI, ; P <.05) were selected as significant independent factors for the prediction of congenital CMV infection. However, the viral load of CMV (median , range copies/ml) in the cervical secretion from 11 women who had positive PCR results and congenital infection was not significantly Table 4. Results of Univariate and Multivariable Logistic Regression Analyses of Predictive Factors for Congenital Cytomegalovirus Infection Univariate Analysis Multivariable Analysis Factors OR (95% CI) P Value OR (95% CI) P Value Maternal fever or flu-like symptoms 2.5 ( ) <0.05 Ultrasound fetal abnormality 35.2 ( ) < ( ) <0.001 CMV IgG 1.0 ( ) 0.08 CMV IgM, index 1.3 ( ) <0.001 Positive for antigenemia assay (C7-HRP) 15.3 ( ) <0.001 CMV IgG avidity index <40% 8.2 ( ) <0.001 Positive polymerase chain reaction results for CMV In the urine 2.7 ( ) 0.14 In the uterine cervical secretion 18.0 ( ) < ( ) <0.001 Abbreviations: CI, confidence interval; CMV, cytomegalovirus; Ig, immunoglobulin; OR, odds ratio. Prediction of Congenital CMV Infection CID 2017:64 (15 January) 163

6 different from that (median , range copies/ml) in 16 women who had positive PCR results and no congenital infection. The diagnostic accuracy of the 2 independent factors for the prediction of congenital CMV infection is shown in Table 5. DISCUSSION To our knowledge, this cohort study is the first to demonstrate that ultrasound fetal abnormalities and positive PCR results in the uterine cervical secretion are independent factors for the prediction of congenital CMV infection in pregnant women with positive results for CMV IgM. It has been demonstrated that the presence of ultrasound fetal abnormalities is associated with congenital CMV infection in pregnant women who have positive or borderline results for serum CMV IgM [12]. No studies have prospectively assessed CMV-DNA using PCR in the uterine cervical secretion as a predictive factor of congenital CMV infection. PCR assay of the cervical secretion in addition to ultrasound examinations is much safer than amniocentesis followed by PCR assay of the amniotic fluid. These results provide useful information for clinical practitioners who treat pregnant women at high risk for congenital CMV infection. It is well known that ultrasound fetal abnormalities, including ventriculomegaly, microcephaly, intracranial calcification, FGR, pleural effusion, ascites, hepatosplenomegaly, and periventricular/intestinal high echodensities, are predictive of symptomatic congenital CMV infection [11 14]. Likewise, the presence of ultrasound fetal abnormalities was identified as a predictive factor of congenital CMV infection in the present study. Among 300 pregnant women with positive results for CMV IgM, ultrasound fetal abnormalities of ventriculomegaly, microcephaly, intracranial calcification, and hepatosplenomegaly were found at the time of maternal examination only in fetuses with CMV infection. These ultrasound abnormalities, rather than FGR, appear to be more specific to fetal CMV infection. Since general obstetricians, but not ultrasound specialists, performed ultrasound examinations in the present study, routine ultrasound examinations during pregnancy might be useful to detect the above-mentioned fetal abnormalities associated with congenital CMV infection. Transmission from mothers to newborns can occur perinatally via cervical and vaginal secretions [21], and the existence of CMV in the genital tract is associated with risk of congenital CMV infection [22, 23]. However, it has been reported that virus isolation in maternal urine or uterine cervical secretions during the first 2 trimesters is a poor indicator for intrauterine CMV transmission or the severity of fetal/neonatal damage [16]. Results of PCR and antigen analyses using maternal blood also failed to correlate with risk of congenital infection or severity [17, 18]. In the present study, which enrolled 300 pregnant women with positive results for CMV IgM, no pregnant women tested positive for CMV-DNA in the blood, while positive results for antigen (C7-HRP) in the blood was found to be statistically associated with congenital infection by univariate analyses. This prospective cohort study demonstrated for the first time, using uni- and multivariable analyses, that a positive PCR result in the uterine cervical secretion was predictive of congenital CMV infection. Viral excretion in the uterine cervix during pregnancy is presumed to represent reactivation of latent virus rather than primary infection [24]. A previous study demonstrated that 7.7% of 993 Japanese healthy women at low risk for CMV infection tested positive for CMV-DNA in the vaginal secretion during the first trimester [25]. In the present study, 27 (9.0%) of 300 pregnant women with positive results for both CMV IgG and CMV IgM had positive PCR results in the uterine cervical secretion. This rate was similar to the 7.7% in low-risk pregnant women. However, the frequency of CMV-DNA in the cervical secretion in women with fetal CMV infection was as high as 50%, but only 5.8% in women without fetal infection, showing a relatively high sensitivity and positive predictive value for congenital infection. CMV-DNA in the amniotic fluid might leak into the genital tract, resulting in an increased rate of positive CMV-DNA in the uterine cervical secretion of women with fetal CMV infection. Another potential explanation is that CMV might be transmitted to the amniotic fluid and the fetus through the genital tract by ascending infection. The CMV IgG AI in women with fetal CMV infection was significantly lower than that in women without fetal infection. Univariate analyses selected an IgG AI <40% as a predictive factor of congenital infection. However, an IgG AI <40% was not a significant predictive factor (P =.07) if multivariable logistic regression analyses were performed for 6 factors with P <.05 in univariate analyses. The present study used a cutoff value of CMV IgG AI <40% for pregnant women, which had been determined by receiver operating characteristic analyses and found to be useful for Table 5. Diagnostic Accuracy of 2 Factors for the Prediction of Congenital Cytomegalovirus Infection Factors Sensitivity, % Specificity, % Positive Predictive Value, % Negative Predictive Value, % Accuracy, % Ultrasound fetal abnormality Positive polymerase chain reaction results for cytomegalovirus in the uterine cervical secretion CID 2017:64 (15 January) Tanimura et al

7 the prediction of congenital CMV infection [20]. In the present study, 8 of the 22 women who had congenital CMV infection underwent avidity measurements beyond 28 GW, and 6 of the 8 women showed a CMV IgG AI of 40%. A cutoff value of CMV IgG AI <40% for the diagnosis of primary CMV infection in pregnancy might not be useful during the third trimester. There are some potential limitations associated with our study. Kobe University Hospital has a maternal fetal center where pregnant women at high risk for fetal CMV infections are often referred from other hospitals and clinics. Indeed, 171 of 301 women with positive results for CMV IgM enrolled in the present study were referred from another hospital or clinic due to positive results for CMV IgM, thus the range of gestational age at IgG avidity was broad. Twelve of 22 cases with congenital CMV infection were symptomatic. These might have partially influenced the results of the present study. The effective methodology of universal screening for fetal CMV infection, such as maternal serological tests, has not been established. The results of our study suggest that maternal universal screening of CMV IgG and CMV IgM followed by ultrasound examinations and PCR assays for CMV-DNA in the cervical secretion only for women with positive CMV IgM may be useful and cost effective in determining pregnancies with higher risks of fetal CMV infection. Amniocentesis for the prenatal diagnosis should be when for pregnancies that are at higher risks of fetal CMV infection. Most clients will want information that can be provided by performing noninvasive procedure before they give consent for invasive procedures. Prenatally, neonatologists can identify high-risk neonates who need workup for congenital infection. The early assessment and detection may lead to accurate diagnosis as well as early commencement of antiviral therapy. Supplementary Data Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the author to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the author, so questions or comments should be addressed to the author. Notes Acknowledgments. We are grateful for the participation of the patients and the care provided by staff at Kobe University Hospital. We thank the clinical and laboratory personnel who supported this study at Kobe University Hospital and Aisenkai Nichinan Hospital. Financial support. This work was supported in part by grants-in-aid from the Ministry of Health, Labour and Welfare of Japan (H23-Jisedai- Ippan-001 and H25-Jisedai-Shitei-003). Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed. References 1. In: Stratton KR, Durch JS, Lawrence RS. Vaccines for the 21st Century: A Tool for Decisionmaking. Washington, DC, Dobbins JG, Stewart JA, Demmler GJ. Surveillance of congenital cytomegalovirus disease, Collaborating Registry Group. MMWR CDC Surveill Summ 1992; 41: Kimberlin DW, Lin CY, Sánchez PJ, et al; National Institute of Allergy and Infectious Diseases Collaborative Antiviral Study Group. Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial. J Pediatr 2003; 143: Nishida K, Morioka I, Nakamachi Y, et al. Neurological outcomes in symptomatic congenital cytomegalovirus-infected infants after introduction of newborn urine screening and antiviral treatment. Brain Dev 2016; 38: Bodéus M, Hubinont C, Goubau P. Increased risk of cytomegalovirus transmission in utero during late gestation. Obstet Gynecol 1999; 93: Revello MG, Fabbri E, Furione M, et al. Role of prenatal diagnosis and counseling in the management of 735 pregnancies complicated by primary human cytomegalovirus infection: a 20-year experience. J Clin Virol 2011; 50: Lazzarotto T, Guerra B, Gabrielli L, Lanari M, Landini MP. Update on the prevention, diagnosis and management of cytomegalovirus infection during pregnancy. Clin Microbiol Infect 2011; 17: Lagrou K, Bodeus M, Van Ranst M, Goubau P. Evaluation of the new architect cytomegalovirus immunoglobulin M (IgM), IgG, and IgG avidity assays. J Clin Microbiol 2009; 47: Fowler KB, Stagno S, Pass RF, Britt WJ, Boll TJ, Alford CA. The outcome of congenital cytomegalovirus infection in relation to maternal antibody status. N Engl J Med 1992; 326: Kenneson A, Cannon MJ. Review and meta-analysis of the epidemiology of congenital cytomegalovirus (CMV) infection. Rev Med Virol 2007; 17: Guerra B, Simonazzi G, Puccetti C, et al. Ultrasound prediction of symptomatic congenital cytomegalovirus infection. Am J Obstet Gynecol 2008; 198:380.e Sonoyama A, Ebina Y, Morioka I, et al. Low IgG avidity and ultrasound fetal abnormality predict congenital cytomegalovirus infection. J Med Virol 2012; 84: Enders G, Bäder U, Lindemann L, Schalasta G, Daiminger A. Prenatal diagnosis of congenital cytomegalovirus infection in 189 pregnancies with known outcome. Prenat Diagn 2001; 21: Malinger G, Lev D, Lerman-Sagie T. Imaging of fetal cytomegalovirus infection. Fetal Diagn Ther 2011; 29: Nigro G, Adler SP, La Torre R, Best AM; Congenital Cytomegalovirus Collaborating Group. Passive immunization during pregnancy for congenital cytomegalovirus infection. N Engl J Med 2005; 353: Lazzarotto T, Guerra B, Lanari M, Gabrielli L, Landini MP. New advances in the diagnosis of congenital cytomegalovirus infection. J Clin Virol 2008; 41: Lazzarotto T, Gabrielli L, Lanari M, et al. Congenital cytomegalovirus infection: recent advances in the diagnosis of maternal infection. Hum Immunol 2004; 65: Revello MG, Zavattoni M, Sarasini A, Percivalle E, Simoncini L, Gerna G. Human cytomegalovirus in blood of immunocompetent persons during primary infection: prognostic implications for pregnancy. J Infect Dis 1998; 177: Ebina Y, Minematsu T, Morioka I, et al. Rapid increase in the serum cytomegalovirus IgG avidity index in women with a congenitally infected fetus. J Clin Virol 2015; 66: Ebina Y, Minematsu T, Sonoyama A, et al. The IgG avidity value for the prediction of congenital cytomegalovirus infection in a prospective cohort study. J Perinat Med 2014; 42: Ross SA, Boppana SB. Congenital cytomegalovirus infection: outcome and diagnosis. Semin Pediatr Infect Dis 2005; 16: Ho M. Epidemiology of cytomegalovirus infections. Rev Infect Dis 1990; 12(Suppl 7):S Murph JR, Souza IE, Dawson JD, et al. Epidemiology of congenital cytomegalovirus infection: maternal risk factors and molecular analysis of cytomegalovirus strains. Am J Epidemiol 1998; 147: Shen CY, Chang SF, Chao MF, et al. Cytomegalovirus recurrence in seropositive pregnant women attending obstetric clinics. J Med Virol 1993; 41: Tanaka K, Yamada H, Minami M, et al. Screening for vaginal shedding of cytomegalovirus in healthy pregnant women using real-time PCR: correlation of CMV in the vagina and adverse outcome of pregnancy. J Med Virol 2006; 78: Prediction of Congenital CMV Infection CID 2017:64 (15 January) 165