Introduction

Sensorineural hearing loss (SNHL) is a common clinical entity in newborns [1,2] and children [3,4]. Pediatric SNHL is an etiologically heterogeneous condition caused by a plethora of genetic [5–7] and environmental factors [8–12]. Recent advances in molecular genetics have revolutionized the assessment armamentarium of pediatric SNHL, enabling us to ascertain the etiology in 40–60% of the children with SNHL [13,14]. However, the clinical significance and contribution of the environmental factors that might lead to pediatric SNHL in these children largely remains unclear [8,11,12,15,16].

Among these environmental factors, congenital cytomegalovirus (cCMV) infection is the leading cause of pediatric SNHL [17] and neurodevelopmental disability [18] in developed countries. The importance of identifying cCMV infection as the etiology of SNHL in newborns has become clinically relevant with the availability of oral antiviral agents that may prevent the progression of cCMV-related SNHL [19]. Furthermore, children with cCMV infection are at risk for progressive SNHL that may not be present until several years of age, at a time when the golden period for hearing-loss rehabilitation has passed [20–22].

The prevalence and clinical characteristics of cCMV-infected newborns have been reported in several Western series [17,23]. However, there is still a paucity of such reports in East Asia, a populated region with rapid economic development. In the present study, we aimed to investigate the prevalence of cCMV infection in newborns from an urban region of East Asia and the clinical characteristics of cCMV-positive children and their mothers.

Materials and methods

Results

During the study period, 3,273 neonates were born at the Taipei City Hospital Fuyou Branch (Fig 1). Of these, the parents of 1,532 neonates agreed to undergo a newborn CMV screening for their children. The CMV screening was positive in seven newborns (0.46%), including one girl and six boys (Table 1). CMV infection was confirmed in all seven newborns by isolation of CMV from saliva and/or urine. All seven CMV-positive newborns passed the NHS. Of the other 1,525 infants who were negative for CMV, 25 (1.6%) failed the initial NHS and three (0.2%) were subsequently confirmed to have unilateral or bilateral SNHL. Among the 1,741 infants who did not undergo CMV screening, 64 (3.7%) failed the initial NHS and 16 (0.9%) were subsequently confirmed to have SNHL. In total, the prevalence of neonatal cCMV infection was 0.46%, with 0.74% for males and 0.14% for females. There was no statistically significant difference in the prevalence between the sexes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flow diagram of the recruitment and the outcome of the newborn hearing screening (NHS) and cCMV screening in 3,273 newborns in Taipei, Taiwan.

NTUH denotes National Taiwan University Hospital.

https://doi.org/10.1371/journal.pone.0248801.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Prevalence of cCMV infection in the 1,532 newborns.

https://doi.org/10.1371/journal.pone.0248801.t001

The characteristics of the seven CMV-positive newborns are presented in Table 2. The newborns of cases 2 and 3 were given birth by a cesarean section (2/7, 28.6%), while the other five newborns were given birth through a normal spontaneous delivery (5/7, 71.4%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Characteristics of the 7 newborns with cCMV infection.

https://doi.org/10.1371/journal.pone.0248801.t002

All seven newborns with cCMV infection were asymptomatic at birth, and none of them were admitted to a neonatal intensive care unit or developed hyperbilirubinemia that required a specific treatment, including phototherapy or plasmapheresis (Table 2). The newborns of cases 2 and 4 were lost to follow-up for unknown reasons. The other five newborns were followed up at the National Taiwan University Hospital for a mean duration of 14.4±6.3 months. None of these five newborns developed hearing or other symptoms during the follow-up period.

Table 3 presents a comparison of the maternal/neonatal characteristics between newborns with and without cCMV infection. The mothers of the CMV-positive neonates had a slightly higher gravidity (2.4 ± 1.4 vs. 2.1 ± 1.2, respectively) and parity (2.0 ± 1.2 vs. 1.6 ± 0.7, respectively) than those whose newborn babies were screened negative for CMV; yet, the difference did not reach statistical significance.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Comparison of maternal/neonatal characteristics between subjects with and those without cCMV infection.

https://doi.org/10.1371/journal.pone.0248801.t003

Discussion

In this study, we demonstrated that the prevalence of cCMV infection is approximately 0.46% in Taipei, a typical populous city in East Asia. The rate is slightly lower than that in other populations. In developed countries, the prevalence of cCMV infection ranged from 0.58% to 0.70% [17,26,27]. In developing countries, such as Mexico, Nigeria, and Gambia, the prevalence of cCMV infection was higher than that in our study, ranging from 0.9% to 5.4% [28–32]. Several factors have been proposed to account for the differences in the prevalence of cCMV infections in newborns between the developed and developing countries, including the testing methods [26], sampling methodology [26], selection bias [26], and seroprevalence rate in the general population [31,33–35], and ethnicity [26,36,37].

Noticeably, in a prior study in 1996, the prevalence of cCMV infection in Taiwanese newborns was found to be 1.8% [38]. The significantly lower prevalence in the present study might reflect the improved public health conditions, including better prenatal care in Taiwan over the past two decades. Interestingly, a recent study in China also reported a low cCMV infection prevalence (0.7%, n = 10.933) [33] as compared to that in three other large-scale studies in Portuguese (1.05%, n = 3,600) [34], Brazilian (1.1%, n = 8,047) [35], and Turkish (1.9%, n = 944) [31] populations, although the seroprevalence of CMV is commensurately high among the four populations. The authors attributed the lower cCMV infection in the Chinese population to the lower exposure of pregnant women to young children than in the other populations as a result of China’s unique one-child policy [33]. In our study, we also observed a higher parity number in the mothers with CMV-positive newborns (2.0 ± 1.2) than in those with CMV-negative newborns (1.6 ± 0.7). It is thus likely that the low fertility rate, that is, the number of children a woman is expected to have during her childbearing years, might also contribute to the low cCMV infection rate in the Taiwanese newborns in the present study. Since 2008, one of the lowest fertility rates (0.895–1.27) among any territory in the world has been recorded in the Taiwanese population (https://eng.stat.gov.tw/public/data/dgbas03/bs2/yearbook_eng/y005.pdf).

The screening tool for cCMV infection is also crucial when the prevalence of cCMV infection is determined and compared across populations. It has been demonstrated that real-time PCR assays on saliva samples collected from live-born newborns performed within 2 days from birth, as performed in the present study, enables high sensitivity and specificity for identifying a cCMV infection in newborns [39,40]. In contrast, real-time PCR assay on dry bloodspot samples had a low sensitivity, restraining its value as a screening test for cCMV infection [41]. Our earlier work also showed a lower rate of cCMV infection (0.17%, 3/1,716) in a similar population when PCR on dried blood spot (DBS) specimens was used as a screening method [25].

Of note, although all seven newborns who screened positive for cCMV infection passed the hearing screening at birth and none of them developed a hearing impairment during the follow-up period, a regular and continual check-up on the hearing level and viral status is necessary in these infants, as it has been documented that a certain proportion of infants with asymptomatic cCMV infection might develop a hearing impairment or other sequelae later on, even as late as at the adolescent age [22].

The strength of this study is that all subjects were recruited from a community-based hospital, which offered an unbiased estimation of the representative prevalence of cCMV infection among newborns in Taipei. All the CMV PCR and confirmatory CMV isolation tests were fully supported by research grants and were hence free to parents. This further prevented selection biases brought about by socioeconomic status. The composition of newborns in the present study is different from that recruited from medical centers [26], where the recruitment of subjects could be biased to a study population with a higher risk of disease propensity.

However, some limitations of this study deserve discussion. First, the current study included a sample size of 1,532 newborns and an average follow-up period of 14.4 months. A larger series with a longer follow-up period will possibly disclose the prevalence and clinical outcomes of cCMV infection with greater precision. Second, the notion that low cCMV infection prevalence is associated with low fertility rates needs to be tested in other populations with low fertility rates. Further investigations in other East Asian populations with a low fertility rate, such as the Korean and Japanese populations, are highly expected.

Conclusion

The prevalence of cCMV infection in newborns of Taipei, Taiwan is 0.46%, which is slightly lower than that in other populations and that previously reported in the Taiwanese population. The relatively low prevalence in this study might be attributed to the improved public health and prenatal care system and the decreased fertility rate in Taiwan.

Supporting information