Seroprevalence of ToRCH Pathogens in Southeast Asia

ToRCH is the acronym for several pathogens associated with pregnancy complications and sequelae in the unborn or newborn child. Particularly primary infections during pregnancy are associated with increased risk. Seroprevalence data of ToRCH infections are important, especially in countries with weak disease surveillance systems, to estimate immunity and vaccination levels, as well as exposure rates and thus the risk of infection during pregnancy. A systematic literature search spanning a 30-year time period was conducted to identify serosurveys on ToRCH pathogens in Southeast Asia. The 96 identified studies showed that some pathogens were well studied, while only limited data were available for others. Studies from the better-developed countries of the region were more abundant. Moreover, seroprevalence data were often limited to a certain geographical region within the country or to certain cohorts, there was an evident lack of recent serosurveys, and the study quality was often not adequate. Well-designed and area-wide serosurveys of ToRCH pathogens are clearly warranted. If combined with risk factor analysis, these studies may guide the development and implementation of effective measures for infection prevention, especially during pregnancy. In addition, educational programs for health care workers and for pregnant women during antenatal care are urgently needed.


Introduction
ToRCH is the acronym for several pathogens associated with pregnancy complications. ToRCH pathogens include Toxoplasma gondii (T. gondii), others (such as Varicella zoster virus-VZV and Primate erythroparvovirus 1-B19V), Rubella virus (RV), Cytomegalovirus (CMV), and Herpes simplex virus (HSV) [1,2] (Table 1). Primary infected pregnant women are mostly asymptomatic or show only mild symptoms. However, transplacental, perinatal or postnatal transmission of the pathogens can severely affect the unborn or newborn child. Sequelae include preterm birth, anomalies, stillbirth, fetal growth restriction, organ injuries, and chronic postnatal infections [3][4][5]. Usually, latent infections or virus reactivations of the mother do not pose a threat to the unborn child [6].
The global impact of congenital infections remains largely unknown. In developing countries, the burden is thought to be higher than in industrialized countries, because of coinfections of the mother or mother and child malnutrition [4]. In addition, inefficient or inexistent disease surveillance and knowledge gaps among healthcare workers (HCW) lead to underreporting and misdiagnosis [4,7]. Data on acute ToRCH infections in the included countries are rare. Disease surveillance systems for RV are established in all studied countries and for congenital rubella syndrome (CRS) in all but Malaysia [8,9], while surveillance for VZV infections seems to exist only in Malaysia and Thailand [10].  [3,17,18] In several Southeast Asian countries, the neonatal mortality rates (NMR) still rank among the highest worldwide. The upper-middle-income countries Malaysia and Thailand had NMR of 4.3 and 5.0, the lower-middle-income countries Cambodia and Laos reported NMR of 14.4 and 22.7 and only the high-income country Singapore reported a very low NMR of 1.1 per 1000 live births in 2018 [19]. It is likely that deaths resulting from congenital ToRCH infections as well as their complications (including pneumonia, pre-term birth, sepsis, and congenital abnormalities) contribute considerably to the high NMR [20].
Presence of immunoglobulin (Ig) G antibodies against a specific pathogen indicates past infection or vaccination. Consequently, IgG serosurveys represent an efficient approach to estimate immunity and vaccination levels, as well as exposure rates. When combined with socio-demographic and behavioral data, such studies allow the identification of risk factors (RFs) for infection and of vulnerable population groups, as well as the need for introducing new vaccines [21]. However, since most serosurveys are conducted in developed countries, the situation in developing countries is poorly understood [22].
To get an overview over the current situation in Southeast Asia, a systematic review of ToRCH serosurveys was conducted. Focus was put on Cambodia, Laos, Myanmar, Malaysia, Singapore, Thailand, and Vietnam, because of their geographic proximity and their social and economic disparities. Based on the available seroprevalence data, we assessed the risk of pregnant women to become infected and identified RFs for past infection.

Literature Search
A systematic literature review was conducted. The database PubMed was searched in January 2019 for studies published between 1 January 1989 and 31 December 2018. The search terms included Name of the pathogen (abbreviation as well as full name), Name of the country (different spellings if applicable) and the Year range (1 January 1989 to 31 December 2018) (Figure 1a). After removing duplicates, the titles of all identified studies were screened for their relevance. Only titles suggesting that ToRCH seroprevalence data were obtained were kept. If the abstracts of the selected articles met the inclusion criteria (Figure 1b), the full texts were screened if available. In a second step, the references quoted in these articles were checked and articles citing the retained studies were retrieved with Google Scholar to identify missed articles. The loose inclusion and exclusion criteria (Figure 1b) allowed drawing a representative picture of data availability and quality in the study area.
Presence of immunoglobulin (Ig) G antibodies against a specific pathogen indicates past infection or vaccination. Consequently, IgG serosurveys represent an efficient approach to estimate immunity and vaccination levels, as well as exposure rates. When combined with socio-demographic and behavioral data, such studies allow the identification of risk factors (RFs) for infection and of vulnerable population groups, as well as the need for introducing new vaccines [21]. However, since most serosurveys are conducted in developed countries, the situation in developing countries is poorly understood [22].
To get an overview over the current situation in Southeast Asia, a systematic review of ToRCH serosurveys was conducted. Focus was put on Cambodia, Laos, Myanmar, Malaysia, Singapore, Thailand, and Vietnam, because of their geographic proximity and their social and economic disparities. Based on the available seroprevalence data, we assessed the risk of pregnant women to become infected and identified RFs for past infection.

Literature Search
A systematic literature review was conducted. The database PubMed was searched in January 2019 for studies published between 1 January 1989 and 31 December 2018. The search terms included Name of the pathogen (abbreviation as well as full name), Name of the country (different spellings if applicable) and the Year range (1 January 1989 to 31 December 2018) (Figure 1a). After removing duplicates, the titles of all identified studies were screened for their relevance. Only titles suggesting that ToRCH seroprevalence data were obtained were kept. If the abstracts of the selected articles met the inclusion criteria (Figure 1b), the full texts were screened if available. In a second step, the references quoted in these articles were checked and articles citing the retained studies were retrieved with Google Scholar to identify missed articles. The loose inclusion and exclusion criteria (Figure 1b) allowed drawing a representative picture of data availability and quality in the study area.

Data Processing
Overall and age-distributed seroprevalence, as well as 95% Confidence Intervals (95% CI) were extracted from the studies. If not available, a free online sample size calculator [23] was applied to estimate the 95% CI (https://www.surveysystem.com/sscalc.htm, 11 January 2019) using the given sample size and overall seroprevalence. If studies included more than one study cohort, the seroprevalence of each cohort was considered separately if it met the inclusion criteria ( Figure 1b). If only age-distributed seroprevalence

Data Processing
Overall and age-distributed seroprevalence, as well as 95% Confidence Intervals (95% CI) were extracted from the studies. If not available, a free online sample size calculator [23] was applied to estimate the 95% CI (https://www.surveysystem.com/ sscalc.htm, 11 January 2019) using the given sample size and overall seroprevalence. If studies included more than one study cohort, the seroprevalence of each cohort was considered separately if it met the inclusion criteria ( Figure 1b). If only age-distributed seroprevalence was available, the overall seroprevalence was calculated by dividing the number of seropositive participants by the total number of participants. Some studies reported combined seroprevalence for IgG and IgM. If possible, the IgG seroprevalence was calculated as mentioned above. If the differentiation between IgG and IgM data was not feasible, the overall IgM/IgG seroprevalence was shown as IgG seroprevalence but tagged accordingly. Since only asymptomatic cohorts without acute infections were considered (Figure 1b), the IgM/IgG seroprevalence largely reflects IgG seroprevalence. Finally, to obtain an estimate of the national seroprevalence of a pathogen, the seroprevalence range for the country was specified when there were more than three studies available in the country. If the same data on seroprevalence of a pathogen was reported in multiple studies, the data was considered only once.

Quality Criteria
The quality of the studies was evaluated by applying criteria of a recent review [24] after slight adaptation (Table S1). Quality criteria were, for example, information about study location and sample size, as well as sample size calculation and inclusion of ethical consideration and reporting bias. For each quality criteria fulfilled, a point was attributed to the study with a maximum of 17 achievable points. The quality criteria were not considered to be inclusion or exclusion criteria but were only used to assess the quality of research.

Results
In total, 96 studies were eligible (see Figure 2 based on Moher et al. [25]): positive participants by the total number of participants. Some studies reported combined seroprevalence for IgG and IgM. If possible, the IgG seroprevalence was calculated as mentioned above. If the differentiation between IgG and IgM data was not feasible, the overall IgM/IgG seroprevalence was shown as IgG seroprevalence but tagged accordingly. Since only asymptomatic cohorts without acute infections were considered ( Figure  1b), the IgM/IgG seroprevalence largely reflects IgG seroprevalence. Finally, to obtain an estimate of the national seroprevalence of a pathogen, the seroprevalence range for the country was specified when there were more than three studies available in the country. If the same data on seroprevalence of a pathogen was reported in multiple studies, the data was considered only once.

Quality Criteria
The quality of the studies was evaluated by applying criteria of a recent review [24] after slight adaptation (Table S1). Quality criteria were, for example, information about study location and sample size, as well as sample size calculation and inclusion of ethical consideration and reporting bias. For each quality criteria fulfilled, a point was attributed to the study with a maximum of 17 achievable points. The quality criteria were not considered to be inclusion or exclusion criteria but were only used to assess the quality of research.

Results
In total, 96 studies were eligible (see Figure 2 based on Moher et al. [25]):  The studies contained 124 seroprevalence rates among different cohorts. The quality criteria scores ranged between 5 and 17 points with a mean of 10.6 points and with a higher score for more recent studies from 2010 to 2018 (9.2 vs. 12.7). The criteria "Overall result given" was fulfilled most often (n = 92), while only 14 studies provided the least fulfilled criteria "Sample size calculation done" (Table S1).

Toxoplasma Gondii
A high proportion of studies reported T. gondii seroprevalence (42/96). Five of these studies reported seroprevalence rates in more than one study population ( Table 2).   Two studies from Cambodia investigated women of childbearing age (seroprevalence 5.8% [27]) and adults (13.1% [26]).
For Laos, the only available study dated back to the early 1990s and the seroprevalence was relatively low (15.3%) in the general healthy population [28].
The two regionally limited studies from Myanmar targeted pregnant women and the seroprevalence ranged from 30.2% [29] to 31.7% [48].
The two studies from Singapore showed a higher rate in the HIV-infected cohort (23.7%) [50] than in pregnant women (17.2%) [49].
Low seroprevalence rates of T. gondii antibodies were reported in the three studies from Vietnam, irrespective of study populations (range: 4.2% [66] to 11.2% [67]). The highest seroprevalence in Vietnam was reported in pregnant women [67]. Most studies were from the South of Vietnam, only one study included a population from the North [66].
For 15 of the reported seroprevalences, no study year was available. The remaining studies were conducted between 1992 [52] and 2015 [48] with no notable change in seroprevalence over time. The studies on T. gondii seroprevalence included different populations with the majority focusing on healthy pregnant women (15/42) or HIV-infected individuals (8/42), with the seroprevalence mostly being higher among HIV-infected people. While the large majority (28/42) of the studies used a commercial ELISA kit, other detection methods (e.g., Sabin-Feldman Dye Test or Immunofluorescent antibody test) were used especially in studies published before 2000. Similar seroprevalence rates were found using the different test methods in the respective countries and study populations.

Risk Factors for Seropositivity
Only 21 studies included an assessment of RFs for seropositivity to T. gondii antibodies. Nine studies conducted in Laos, Malaysia or Thailand reported that higher age (e.g., older than 30 years [47,61] or 40 years [42,46]) was statistically significant associated with T. gondii seropositivity. This RF did not seem to be restricted to a specific study population. Identified RFs were often indirectly linked to age such as parity [29,48] and marital status [38].

Varicella Zoster Virus
Of the 96 identified studies, 11 investigated VZV IgG seroprevalence (Table 3).  There were no studies on VZV seroprevalence from Cambodia and Myanmar and the only article from Laos reported a seroprevalence of 95.0% in HCW from northern and central provinces with high rates (94.1%) already in the 15-to 24-year-olds [68].
The only study from Malaysia reported a seroprevalence of 86.6% among HIV-infected patients in Kuala Lumpur [69]. Likewise, for the only study from Vietnam, 99.0% of HIV-infected drug users from Ho Chi Minh City had anti-VZV antibodies [65].
All studies in the region reported VZV IgG seroprevalence rates exceeding 50% ( Table 3). The earliest study was conducted in 1994 [73] and the most recent was published in 2015 [72]. Irrespective of the limited amount of data, seroprevalence rates did not vary over the years in comparable study populations. Most studies (8/11) included HCW or the general population with HCW showing the highest VZV IgG seroprevalence.
Regardless of the study year, all identified studies used enzyme immunoassays (EIA) for detection of VZV IgG antibodies.

Risk Factors for Seropositivity
Several studies suggested that VZV IgG seroprevalence increases with age. In fact, the lowest seroprevalence was found among children aged less than 4 years and 1-to 6-year-olds (11.3% [74] and 34.5% [71], respectively). Moreover, five of the studies from Singapore and Thailand identified increasing age as a statistically significant RF for VZV IgG seropositivity [71][72][73][74][75].
The highest seroprevalence rates were found in HCW [68,72] and in HIV-infected individuals [65,69]. As for T. gondii, some studies suggested ethnicity as RF. Singaporean studies identified highest seroprevalence values in participants with Chinese ethnicity and lowest in participants with Indian ethnicity [71,72].
One study suggested a higher seroprevalence in the more temperate regions of the North of Thailand than in the more humid regions in the South and an influence of population density with notably lower VZV IgG seroprevalence in rural compared to urban settings in the South [74].

Primate Erythroparvovirus 1
Six out of 96 studies investigated anti-B19V IgG seroprevalence (Table 4). The latter study included blood donors, undergraduate students and various immunocompetent patients recruited between 1999 and 2000 in Kuala Lumpur and found an overall seroprevalence of 37.6% [78].
Half of the studies (3/6) were from Thailand, where B19V IgG seroprevalence rates ranged between 10.9% [82] and 20.2% [80]. Two studies focused on younger cohorts, while one study included children as well as blood donors up to the age of 51 years [80].
In the two studies from Singapore, seroprevalence rates of 16.2% and 30.0% were reported in a healthy study population [79] and in pregnant women [49], respectively.
Recent studies were missing. Before 2000, B19V IgG seroprevalence did not exceed 40% [78]. Due to the small amount of data, no regional trends could be identified. Studied populations included mainly healthy persons such as students, blood donors or a general healthy population [78][79][80]82]. Three studies focused on various healthy individuals [79,80,82] and two on vulnerable study populations (i.e., pregnant women and immunocompromised children [49,81]). Regardless of the study year or the country, all studies used EIAs for the detection of anti-B19V IgG.

Risk Factors for Seropositivity
Due to differing study designs and the limited number of studies, it was difficult to extract trends or identify RFs.
In Singapore, highest rates were found in pregnant women and lowest rates in healthy individuals. Whether there is a statistically significant difference in seroprevalence between different age groups remains unknown. In one study, seropositivity increased with age, but remained below 70% even in older participants [78].

Rubella Virus
Eighteen studies covered anti-RV IgG seroprevalence ( Table 5). Four of these studies investigated seroprevalence rates for multiple study populations.

Seroprevalence by Country
All studied countries currently use rubella-containing vaccine (RCV) for infants [98]. The only study from Cambodia was conducted before the introduction of RCV in the country [83,99] and revealed an average seroprevalence of 73.3% among women aged 15 to 39 years [83].
In the three studies from Laos, anti-RV IgG seroprevalence ranged from 43.6% [84] to 86.2% [68] among children and HCW. The earliest study was conducted before the introduction of the RCV [84]. Seroprevalence in children showed an increase from 43.6% in 6-to 12-year old [84] to 90.2% in an age group of 5-to 14-year old children [85].
The only study from Vietnam was done before introduction of RCV [97,99] and found an overall seroprevalence of 71.1% in pregnant women [97].
The studies used various types of immunoassays, but seroprevalence results did not seem to vary greatly between these assays.

Risk Factors for Seropositivity
In contrast to the other pathogens, most studies reported RFs for susceptibility to RV rather than for seropositivity.
As expected, several studies conducted before introduction of RCV found that younger age groups [83,84,97] were more likely to be susceptible to RV infection, while studies conducted after RCV introduction found that seroprevalence rates decreased with age [68,85,86,91].
Singaporean studies identified Malay ethnicity or permanent residency in Singapore as RF for susceptibility to RV [91,92]. One study identified occupation as a "worker" as a statistically significant RF for susceptibility [87], while others reported missing history of vaccination [87,88] or not being included in the Supplementary Immunization Activity (SIA) of 2011 in Laos [85] as RFs.

Cytomegalovirus
Overall, 14 studies on CMV seroprevalence were identified (Table 6).   [100] with the lowest rate found among pregnant women [101]. In contrast to the other pathogens, the study locations covered not only Kuala Lumpur, but also the surrounding area as well as other regions.
In the two studies from Singapore, the seroprevalence values were 87.0% [49] and 96.8% [50], with the lowest seroprevalence also reported for pregnant women [49].
One study from Vietnam conducted in 1996 included HIV-infected drug users in Ho Chi Minh City and showed an overall seroprevalence of 100% [65].
Seroprevalence was high in all countries and except for one study investigating Thai blood donors [104], rates were always above 70%. Most studies provided data for the years between 1990 and 2000 and only two studies were conducted later [50,101]. Most studies involved pregnant women (6/14) or blood donors, as representatives of a healthy population (6/14). For one study, no detection method was mentioned [50].

Risk Factors for Seropositivity
Few RFs for seropositivity to CMV were identified. Although some studies noted an increase of seropositivity with age, this difference was not statistically significant [49,102]. One study in Thailand identified female sex as RF for seropositivity and related it to a stronger role of women in childcare [106].
Similar to studies of other pathogens from Singapore, ethnicity was identified as RF for seropositivity (Non-Singaporean ethnicity) [49].

Herpes Simplex Virus
In total, 19 studies were identified for HSV. The studies included seven seroprevalence rates for HSV-1, 16 for HSV-2 and one for unclassified HSV (Table 7).   No studies on either HSV-1 or HSV-2 seroprevalence were identified for Cambodia, Laos, or Myanmar and only one study from Kuala Lumpur, Malaysia, reporting seroprevalence rates of 70.7% and 53.9% for HSV-1 and HSV-2, respectively, in adult HIV-infected individuals [69].
Two studies were identified for Singapore with seroprevalence rates of 55.8% [109] and 76.7% [108] for HSV-1, and of 28.5% [109] and 79.0% [108] for HSV-2. The studies included attendees of a sexual infection clinic that were either sex workers or general population [108,109]. In every age group of the general population, HSV-1 exceeded HSV-2 seroprevalence [109]. In contrast, in sex workers, the predominating HSV type depended on the age group [108] with HSV-2 seroprevalence exceeding HSV-1 seroprevalence in the age groups of 30 years or older.
Seven studies were identified for Thailand and the seroprevalence for HSV-1 and HSV-2 was between 56.5% [113] to 91.0% [112] and 14.9% [110,111] to 80.0% [52], respectively. HSV-1 rates tended to be higher than HSV-2 rates, e.g., a study population of male army conscripts showed an HSV-1 seroprevalence of 77.0% and a lower HSV-2 seroprevalence of 14.9% [110,111]. The highest HSV-1 rate (91.0%) was reported among female sex workers, who had a lower HSV-2 rate (75.6%) [112]. Surprisingly, the highest HSV-2 rate of 80.0% was reported in a population of pregnant women [52]. The study cohorts were exclusively recruited in Bangkok and in the North of Thailand.
Nine studies were identified for Vietnam and the only reported HSV-1 seroprevalence was 98.0% [114] while HSV-2 seroprevalence ranged between 2.0% [122] and 30.8% [117,118]. Almost all studies covered only HSV-2 seroprevalence and in contrast to other countries, HSV-2 serosurveys from Vietnam focused especially on healthy women and sex workers.
Overall, HSV-1 seroprevalence seemed to be similarly high in all countries, generally above 55%, irrespective of the study population.
When comparing HSV-1 to HSV-2 seroprevalence in similar study populations, HSV-1 seroprevalence seemed to be generally higher than HSV-2 seroprevalence. Most studies (11/19) focused on high-risk groups such as sex workers and HIV-infected individuals. For HSV-1 IgG seroprevalence, there was only one study focusing on the general population [114]. For HSV-2 IgG seroprevalence, only four studies included healthy study populations [52,117,118,122]. HSV-2 IgG seroprevalence of above 50.0% was reported in HIV-infected individuals [69,115] and Thailand reported the highest HSV-2 seroprevalence overall. HSV-2 seroprevalence of sex workers in Thailand and Singapore and of HIVinfected individuals ranged between 75.6% [112] and 79.0% [108]. In contrast, sex workers in Vietnam showed a lower HSV-2 seroprevalence of 24.9% [121] to 27.7% [120].
One study did not report the method used [121]. Most studies (15/19) used ELISA for antibody detection and three studies used immunoblots [110][111][112]. One study used ELISA for detection and western blot as confirmation [122]. Studies using immunoblot tended to be older than studies using other methods and all recent studies used ELISA for IgG seroprevalence investigation.

Risk Factors for Seropositivity
Most studies did not identify RFs for HSV-1 seropositivity, except for one where participants of higher age were statistically significant more likely to be seropositive [109].
For anti-HSV-2 IgG seropositivity, higher age was identified as statistically significant RF in Thailand, Singapore, and Vietnam in different study populations [108,111,116,117]. In line with this, one study identified age below 20 as a statistically significant RF for anti-HSV-2 IgG negativity [120]. Several studies in Thailand also identified young age at first sexual contact as statistically significant RF [111,117]. High frequency of sexual contact with female sex workers, as well as young age at first contact with female sex workers were statistically significant RFs linked to an increased anti-HSV-2 IgG seropositivity rate [111].
As mentioned, study populations of sex workers showed high HSV seroprevalence rates, regardless of the subtype [108,109]. Studies reported that sex workers were statistically significant more likely to be seropositive for HSV-2 when they already worked 9 years or longer in the sector [108] or when co-infected with other sexually transmissible diseases [112,120].

Discussion
This review provides an overview about serosurveys of ToRCH pathogens conducted in Southeast Asia during a 30-year time period. While some pathogens were well studied (e.g., T. gondii), only limited data were available for others, such as VZV and B19V. The majority of the studies was conducted in the better-developed countries of the region, namely Malaysia, Singapore, and Thailand. In contrast, there was a general paucity of studies from Cambodia, Laos, Myanmar and Vietnam. Moreover, seroprevalence data were often limited to a certain geographical region within the country or to certain cohorts. There was an evident lack of recent serosurveys: many studies date back to the early 2000's or late 1990's. Finally, the study quality (Table S1) was often poor, irrespective of the studied pathogen or the study country, but the quality of more recent articles seemed slightly better. General information such as study location and year, test method and specimen type used and overall seroprevalence result were provided by most of the studies. In contrast, quality criteria such as calculating a confidence interval, discussing bias, or presenting a sample size calculation, were met by less than one third of the studies. Welldesigned and nation-wide studies on IgG seroprevalence of ToRCH pathogens are urgently needed as there are no surveillance systems for ToRCH pathogens except for RV and VZV [8][9][10]. Combined with RF analysis, these studies may lead to recommendations for infection prevention, especially during pregnancy. Studies investigating the causes of neonatal death and an analysis of the data in relation to seroprevalence information would contribute to our understanding in how far the prevalence of ToRCH pathogens and the resulting risk of infection during pregnancy influence NMR. Serosurveys about vaccine-preventable infections inform about the need to introduce a new vaccine, about the benefits and weaknesses of vaccination programs and are essential to identify target groups for SIAs.
The T. gondii studies demonstrated a wide disparity in seroprevalence rates ranging from 2.6% in healthy Thai women [60] to 59.7% in the general Malay population [44]. This disparity can be explained by differing cultural or behavioral practices, as well as a wide variability regarding study design, target cohort and laboratory testing. Contact to cats, consumption of undercooked or raw meat and usage of unclean water [11,124] were identified as RFs in some studies from Thailand and Malaysia [29,39,40,53,59,61,64].
Other studies suggested high age as RF for past infection [28,29,[40][41][42]46,47,61,64]. This was also observed in developed countries such as Germany [125]. The seroprevalence data found, suggest that the circulation of the parasite is higher in Malaysia, Thailand, and Myanmar than in the other countries. Women of reproductive age with the associated risky behaviors may therefore be more at risk of primary infection with the pathogen in these countries. Although preventive measures exist [126], their implementation is thwarted by knowledge gaps, which are not limited to developing countries [127,128]. Indeed, knowledge about T. gondii was limited among pregnant women from the Philippines, Malaysia, and Thailand [129]. While screening for T. gondii serostatus during antenatal care is widely discussed in the literature [12], it is so far implemented in only a few countries such as Austria and France [11]. In low resource-countries, improving the training of HCW, who can pass on crucial information during antenatal care sessions, seems more appropriate as a first approach to reduce the risk of primary infection during pregnancy.
VZV seroprevalence was rather high in all studied countries and ranged between 52.8% in young Thai adults [74] and 99.0% in middle-aged intravenous drug users in Vietnam [65]. As of October 2020, VZV vaccination is not integrated in the national immunization program of any of the studied countries [98]. However, it is recommended for susceptible HCW in Singapore and Malaysia [130,131]. Consequently, high IgG seroprevalence indicates most likely high virus circulation levels. Several studies suggested that infection occurs in late adolescence or early adulthood [71,74,75] and earlier in temperate climates [132]. Previous studies indicated that temperate weather conditions might favor virus transmission, explaining earlier immunity in these countries [133,134]. Others suggested that a tropical humid climate interferes with virus stability and that virus transmission patterns differ between the different virus variants. Supposedly because of lower population densities, virus circulation in rural settings of tropical countries is low [132]. While the high seroprevalence rates in adults suggest that the risk of primary infection during pregnancy is small, vaccination of vulnerable groups such as women of reproductive age, people with chronic diseases, children and HCW should be further discussed.
B19V was the least studied ToRCH-pathogen in the region. The lack of recent data makes it difficult to make well-founded statements about the infection risk during pregnancy. However, the low seroprevalence ranging between 10.9% in Thai undergraduate students and 37.6% in the general Malay population [78] suggests low infection rates and therefore a limited risk of women during pregnancy.
In contrast to B19V, many studies covered the vaccine-preventable RV infections. The main goal of rubella vaccination is the prevention of CRS [135]. However, studies showed that immunization coverage needs to be ≥ 80.0% to avoid an increased risk of CRS due to a phenomenon called paradoxical shift, when virus circulation is not interrupted and women enter reproductive age without vaccine-induced or natural immunity [135,136]. As of October 2020, all countries included in this review have introduced RCV into their routine vaccination schedule [99]. Although Singapore, Thailand, and Malaysia have provided RCV since more than 30 years [88,137,138], serosurveys suggest that more than 10% of the population remains susceptible to RV infection. Some studies from South East Asia reported seroprevalence rates below 90.0% in children [90,[92][93][94], and below 85.0% in adults older than 35 years [90,91,93]. Even a more recent study from 2018 found a seroprevalence below 90.0% in some older age groups [96]. These studies indicate some vaccination gaps and suggest that some population groups were targeted neither by the national immunization programs, nor by SIAs, but the risk of infection in these countries can be considered to be low. Cambodia, Vietnam, Laos, and Myanmar introduced RCV only in 2011 or later [85,99,[139][140][141][142]. Reports of the pre-vaccination era provide an estimate of virus circulation in unvaccinated populations. Here, more than 25% were susceptible to RV and only older age groups reached seroprevalence rates of more than 80% [83,97]. Infections were commonly acquired in childhood as studies reporting age-distributed data showed seroprevalence rates of 50% or above already at age 11 and older [83,84,97]. Except for a study from Laos [85], there were no studies investigating seroprevalence after RCV introduction and thus the success of the vaccination programs is largely unknown. The Lao study showed that particularly young women benefited from RCV introduction [85].
Despite the vaccination, local outbreaks of rubella have been described in recent years, which were most likely due to above mentioned immunity gaps [8,9]. Nevertheless, rubella and CRS cases have declined in Southeast Asia [8,9]. Expanding the target age groups in SIAs could be a reasonable approach to reduce the risk of infection in all women of reproductive age. Additionally, serosurveys remain important to monitor immunity gaps and to identify susceptible populations to prevent rubella outbreaks.
High levels of CMV circulation have been reported in Africa, South America, and Asia [143]. Also in Southeast Asia, high seroprevalence rates were reported throughout all study populations ranging between 52.4% in healthy adults [104] and 100.0% in pregnant women [52,106] or intravenous drug users [65]. The large majority of the studies reported high seroprevalence rates (>70%) already in young age groups, between 15 to 30 years old [49,102]. Consequently, CMV infection is also in this area frequently acquired at young age and most adults, including pregnant women, are IgG positive. However, anti-CMV antibodies do not protect from reinfection and reinfection during pregnancy can lead to congenital infection [144]. In high prevalence settings, reinfection was even considered to be responsible for most cases of congenital CMV [145,146]. Yet, a Singaporean study reported that only 20% of pregnant women had heard of CMV [147], suggesting that knowledge and awareness are low despite the high prevalence. This is also true for countries outside of the investigated region [128,148,149] and is of concern since CMV infection is considered to be the most common congenital infection worldwide [150]. Educational programs for HCW as well as for pregnant women seem to be urgently needed to reduce the risk of acquiring a CMV infection during pregnancy.
HSV-1 is another virus for which the global burden is high [151]. Seroprevalence rates in South East Asia ranged from 55.8% in 15-to 80-year-old Singaporeans [109] to 98.0% in Vietnamese 18 to 55 year old women [114]. Little age-stratified information was available suggesting acquisition of immunity during late adolescence or young adulthood since high levels of seropositivity were already found in individuals younger than 20 years [108,109]. This finding implies that women of reproductive age are largely protected from primary infection.
Interestingly, for most countries, especially Thailand and Vietnam, more studies on HSV-2 than on HSV-1 were available, probably due to the strong influence of sex tourism in these countries. Seroprevalence of anti-HSV-2 IgG among individuals with risky sexual behavior ranged between 24.9% [121] and 79.0% [108]. Among individuals without risky sexual behavior, the seroprevalence rates were mostly below 20% but ranged between 2.0% in Vietnamese married women [122] and 80.0% in Thai pregnant women [52]. This indicates that women without risky sexual behavior are most likely not immune against HSV-2, but are also not at risk of developing genital ulcers during pregnancy due to reactivation. While transplacental transmission of HSV-1 or -2 is rather rare, mother-tochild transmission during vaginal delivery is more common [17]. RFs for seropositivity reported from South-East Asia included having multiple sexual partners, history of other sexually transmitted diseases, female sex, low educational level, belonging to minority ethnic groups and start of sexual activity at an early age [18,152]. To be able to assess virus exposure and to identify RFs, more seroprevalence surveys are needed and awareness raising campaigns should target in particular sex workers. Moreover, HSV screening should be offered during pregnancy.
Although the present review provides a comprehensive overview of current knowledge of ToRCH pathogens in the region, there are some limitations. Geographical restrictions, different study cohorts and test methods hampered result comparability and the suboptimal quality of many studies raises concerns related to data reliability. However, despite these constraints, the review provides insights into pathogen distribution, identifies immunity gaps and susceptible populations in the region, allows a risk-estimation of primary infections during pregnancy and provides guidance for future research.

Conclusions
Women of childbearing age in Southeast Asia are susceptible to many ToRCH pathogens. The paucity of reliable information for several pathogens and the often low quality of the studies warrant comprehensive nationwide serosurveys including pregnant women but also the general population. The data could serve as a basis to evaluate and improve current prevention measures. To raise knowledge and awareness of the risks posed by ToRCH pathogens both in HCW and in pregnant women, is an important first step to prevent fetal loss and congenital malformations.

Conflicts of Interest:
The authors declare no conflict of interest.