Introduction

Rubella virus usually causes a mild self-limited fever and rash in children and adults. Congenital rubella syndrome (CRS) has serious consequences, such as miscarriage, stillbirth, and severe birth defects in infants, resulting from rubella virus infection during pregnancy, especially during the first trimester [20].

The main goal of rubella immunization is to prevent CRS [3, 20]. The incidence of CRS in most developed countries has greatly decreased since the implementation of rubella vaccination [2, 6, 9]. Effective rubella vaccination programs were implemented in the USA in the 1960s and resulted in the elimination of CRS in those countries since 2010 [10].

However, rubella vaccine has not yet been implemented in many developing countries [3]. Indonesia is among those countries that have not introduced a rubella vaccine into the national immunization program. Yogyakarta, a province of Indonesia with a population of 3.45 million [12], is an endemic area for rubella cases. Therefore, a large number of pregnant women are infected with rubella, and their children consequently suffer from CRS. In this study, we conducted a prospective surveillance study of infants aged <1 year with suspected CRS. This study aimed (1) to estimate the incidence of CRS in Indonesia, (2) describe the clinical features of CRS at our referral hospital, and (3) pilot a CRS surveillance system to be extended to other hospitals.

Materials and methods

This cross-sectional study was conducted between September and December 2013 to identify and describe CRS cases among infants aged <1 year who were hospitalized during those surveillance time at Dr. Sardjito Hospital, Yogyakarta. The Institutional Review Board for Human Research of the Faculty of Medicine, Universitas Gadjah Mada/Dr. Sardjito Hospital, approved this study (KE/FK/902/EC). Written informed consent was obtained from all parents for this study.

The infants were recruited from otolaryngology, neurology, cardiology, growth and development, and ophthalmology departments, private outpatient clinics, and pediatric wards. Classification of CRS cases in this study was based on the WHO case definition (Table 1) [21]. Definition of CRS cases included the following categories: suspected, laboratory confirmed, clinically compatible, and epidemiologically linked. The clinical criteria of CRS consisted of the presence of ≥two clinical features from group A, or one feature from group A and ≥one feature from group B in the following lists. Group A comprised sensorineural hearing impairment, congenital heart disease, pigmentary retinopathy, cataract(s), and congenital glaucoma. Group B comprised purpura, splenomegaly, microcephaly, developmental delay, meningoencephalitis, radiolucent bone disease, and jaundice with onset within 24 h of birth. Clinical examination of these cases was performed by a pediatric neurologist at the hospital. We also involved specialists from different units of the hospital, such as ophthalmologists, cardiologists, and otolaryngologists. Infants who presented with hearing impairment, congenital heart defects, congenital cataracts, or all of these defects were enrolled in the study. The exclusion criteria consisted of patients who presented with congenital defects that were not compatible with the definition of CRS.

Table 1 CRS case definitions for surveillance purposes [21]
Full size table

Data were collected using a questionnaire that included patients’ information, clinical characteristics, family history, and laboratory findings. Infants underwent an auditory brainstem response examination. Laboratory confirmation of CRS cases was based on the WHO case definition [21]. WHO laboratory criteria for confirmation of suspected CRS cases include the following: detection of rubella IgM antibody, or sustained rubella IgG antibody levels as determined on at least two occasions between 6 and 12 months of age in the absence of receipt of rubella vaccine; or detection of rubella virus (e.g., nucleic acid detection by RT-PCR or isolation of rubella virus) in an appropriate clinical sample (best results are from throat swabs, but nasal swabs, blood, urine, or cerebrospinal fluid specimens are also acceptable) [21].

Investigation for specific IgM antibodies was conducted using Vidas® RUB IgM immunoassay test kits (bioMérieux Indonesia, Jakarta). Furthermore, infants aged ≥6 months were also tested for rubella IgG antibodies using Vidas® RUB IgG immunoassay test kits. Descriptive data, such as frequency, were analyzed using STATA 11 (StataCorp LP, College Station, TX, USA). Laboratory tests were performed in the same hospital.

Furthermore, we estimated the incidence of CRS by dividing the number of laboratory-confirmed CRS cases with the number of newborns in Yogyakarta Province from September to December 2013.

Results

In this study, we identified and described CRS cases among Indonesian infants. From September to December 2013, we evaluated 55 suspected cases of CRS in infants, of whom five participants were excluded for the following reasons: two cases of Down syndrome, one case of isolated patent ductus arteriosus, and two infants whose parents refused to participate in the study. Of 50 participants, 24, seven, three, one, one, seven, and eight infants were recruited from the departments of neurology, cardiology, growth and development, ophthalmology, and otolaryngology, private outpatient clinics, and pediatric wards, respectively. Three participants were excluded because of no serological test results. A total of 47 infants remained for further analysis.

The majority (89.4%) of infants were aged ≥1 month old, including 25 boys and 22 girls (Table 2). According to the WHO case definition, 11/47 (23.4%), 9/47 (19.1%), and 27/47 (57.5%) infants were classified as laboratory-confirmed, clinically compatible, and discarded CRS, respectively (Table 3).

Table 2 Characterization of congenital rubella syndrome cases in Yogyakarta, Indonesia
Full size table
Table 3 Clinical and laboratory findings of Indonesian infants with congenital rubella syndrome
Full size table

Furthermore, 10/47 (21.3%) infants showed IgM-positive results for rubella and 1/47 (2.1%) had sustained rubella IgG antibody levels as determined on two occasions. Moreover, there were four IgM-negative (at birth) infants and five IgG-positive (at 6–11 months old) infants with suspected CRS who did not undergo a second test. Therefore, we further classified these infants as clinically compatible cases (Table 3). Most (63.6%) laboratory-confirmed cases of CRS were in the 1–5 month age group of infants, of whom 3/11 (27.3%) were boys and 8/11 (72.7%) were girls (Table 2). From September to December 2013, the number of newborns in Yogyakarta Province was 16,569 [13]. Therefore, the estimated incidence of CRS in Yogyakarta, Indonesia during the study period was 1:1500.

The most common clinical manifestation among CRS cases was hearing impairment (100%), followed by congenital cataracts (72.7%), microcephaly (72.7%), and congenital heart defects (45.5%). Other clinical features of CRS cases included hepatosplenomegaly (18.2%) and global developmental delay (9.1%) (Table 2).

Analysis of the maternal history showed that the median age of the mothers was 27 years (range, 21–34 years). None of the mothers had been vaccinated against rubella, and only five (33.3%) mothers had febrile rashes during their pregnancy. However, we could not classify the mothers’ infection according to the WHO definition of rubella cases because of a lack of information on the mothers’ infections.

Discussion

In this study, we showed that the incidence of CRS in Indonesian infants was high, and 23.4% of the patients during our short period of surveillance were laboratory-confirmed CRS. The reason for this finding may be that rubella vaccination has not been implemented in the national immunization program in Indonesia. There are no data on the incidence of CRS in Indonesia. However, notably, the incidence of CRS estimated in this study was solely based on one tertiary hospital in Yogyakarta Province. Therefore, our estimate may not reflect the incidence of CRS in Indonesia.

Our method of surveillance will be implemented in other hospitals as a basis for implementation of rubella vaccine in Indonesia. Our hospital is one of the tertiary referral hospitals in Indonesia for Yogyakarta and the South of Central Java region. The Indonesian government is considering rubella vaccine in the national immunization program in the near future.

Our study showed different results from previous studies of developed countries [2, 6, 9, 14, 16, 22]. The incidence of CRS in most developed countries has greatly decreased since the implementation of rubella vaccination [2, 6, 9]. However, our study showed similar results to studies from Asian countries [13, 15, 17]. Rubella IgG antibodies were detected in 74% of hearing-impaired children in Bangladesh [15]. By screening with real-time PCR, rubella virus RNA was detected in throat swabs and placental tissues in all cases (100%) of fetuses/newborns with congenital cataracts in Vietnam [13]. Bangladesh and Vietnam have not yet implemented rubella vaccine in the national immunization program [13, 15]. Only a few Asian countries have implemented rubella vaccination into national immunization programs. Therefore, rubella still remains poorly controlled in many countries in Asia, especially in the Southeast Asian continent [18]. Furthermore, according to WHO surveillance data regarding the incidence of rubella in 2012, Indonesia was one of the countries that reported most of the rubella cases in the Asia-Pacific region [9]. Sudan is one of the developing countries that have not implemented a rubella vaccination program. Interestingly, in Sudan, surveillance only detected 7.1% of laboratory-confirmed CRS among 98 infants with suspected and clinically confirmed CRS [1]. The most likely explanation for this finding is that the observed defects in those infants were caused by other pathogens involved in congenital disorders, such as Toxoplasma gondii, cytomegalovirus, or Herpes simplex virus [4]. In accordance with those findings, Villagra et al. suggested performing an integrated surveillance for CRS, with screening for TORCH pathogens to strengthen CRS surveillance and avoid missing rubella cases [19]. Furthermore, in our study, some infants had clinical CRS, but this was not confirmed by laboratory criteria. The most likely explanation for this finding is that the affected infants had infections or pathology other than rubella infection, such as T. gondii, cytomegalovirus, or H. simplex virus [4].

Notably, a 4-month period is too short for prospective surveillance for a condition such as CRS. We conducted surveillance during this short period because of a limitation of funding. Additionally, this surveillance was a pilot study for CRS surveillance in other hospitals in Indonesia.

The ELISA test, which is used to identify virus-specific IgM and/or IgG antibodies, is a popular method for diagnosing CRS in developing countries because of its simplicity and reliability. Almost 100% of infected infants aged younger than 3 months will show rubella-specific IgM. However, this antibody gradually decreases to less than 50% by 12 to 18 months of age [17]. In our study, 9/10 (90%) infants showed rubella-specific IgM antibodies at <6 months of age, while only 1/10 (10%) infants showed rubella-specific IgM antibodies at ≥6 months of age (Table 3). A limitation of our study is that IgM-negative infants (at birth) with suspected CRS did not undergo a second test at 1 month old or shortly after. Furthermore, a diagnosis of CRS based only on the presence of rubella IgG antibodies should be carefully determined because the test does not differentiate between maternal-induced immunity and acquisition of infection during early gestation. A diagnosis of CRS based solely on rubella IgG may only be confirmed if the antibodies persist beyond 4–6 months old in infants [8]. Additionally, the presence of IgG antibodies to qualify as laboratory-confirmed CRS would potentially allow an infant who acquired rubella postnatally and who had IgG antibodies as a consequence, to be falsely identified as having CRS. However, clinical manifestation of postnatal rubella infection is usually a mild and self-limited disease, but CRS has severe consequences, such as miscarriage, stillbirth, and severe birth defects [11]. Additionally, postnatally, rubella-infected infants show a rubella-specific IgG response from 7 to 10 days after onset of a rash, whereas infants with CRS demonstrate high/increasing rubella-specific IgG levels in the first year of life [5].

Our results are similar to previous studies, which showed that hearing impairment was the most frequent defect of CRS [7, 9]. However, Zimmerman et al. showed that congenital heart defects were the most common clinical manifestations in CRS cases [22]. Clinical diagnosis of CRS is difficult, particularly in infants with single and mild defects. Early diagnosis of CRS is essential for prompt intervention for specific impairments and also to prevent further dissemination of the virus because infants with CRS might shed the virus for long periods. Therefore, the ELISA test is important for confirming CRS, especially in single and mild deficit cases [17].

In conclusion, the number of laboratory-confirmed CRS cases among Indonesian infants is high. Furthermore, hearing impairment is the most common clinical feature found in infants with CRS. Our findings indicate the importance of implementation of rubella vaccination in the national immunization program in Indonesia. Our results also suggest that a CRS hospital-based surveillance program should be conducted in other hospitals in Indonesia.