Background
Hand, foot and mouth disease (HFMD) is a common pediatric illness [
1]. It is characterized by 3–4 days of fever and then the development of vesicles on the palmar and plantar skin, buccal mucosa and tongue [
2‐
4]. This illness as itself is mild and self-limited [
5]. While in some cases, accompanied with other neurological complications, HFMD may also lead to severe outcomes and even death [
6‐
8]. Children under 10 years old especially those less than 5 are the most susceptible population for reasons that are not clearly understood [
2,
9‐
11].
The first viral agent identified for HFMD is Coxsackievirus A 16 (CA16) [
12]. Belonging to the picornaviridae family, this virus, associated with Enterovirus 71 (EV71), is responsible for nearly all of the HFMD epidemics in Southeast Asia [
13,
14]. In Japan, CA16 was reported to be a prevalent type, while EV71 recurrent every 3 years [
10,
15]. In Singapore, from 2001 to 2008, the predominant EV strains isolated from HFMD cases were CA16 and EV71 [
16,
17]. In China, EV71 and CA 16 are also frequently reported to co-circulate and cause HFMD, although the major etiologic agent identified from hospitalized cases was EV71 [
18,
19].
Many large HFMD outbreaks with severe and fatal outcomes in Southeast Asia have recently been described [
4,
20]. In Taiwan, the most severe HFMD outbreak occurred in 1998, resulting in 405 severe neurological cases and 78 deaths [
21,
22]. In China, the recent outbreaks of HFMD were initiated from 2007 in Shandong province. Then a national widespread epidemic occurred in 2008, with more than 176,000 reported human cases [
23]. Contrary to HFMD outbreaks in other Asian countries, the epidemics in China were more lasting [
24]. Unprecedented HFMD outbreaks occurred in the following years. By the end of 2010, a total of 3,419,149 cases and 1384 fatal cases were reported [
25,
26]. It is said that the recently occurred HFMD epidemics have become a serious public health problem in Southeast Asia, especially in China.
Currently,only a few studies have been conducted to investigate the human immunity to HFMD in China [
27,
28]. The seroepidemiology of EV71 infection before, during and shortly after the epidemics could pave ways for prophylactic intervention strategies. In addition, several EV71 vaccine candidates are being developed in mainland China from 2008 [
29‐
31]. And some of them are at various stages of clinical development [
32,
33]. To better establishing the immunization program against EV71 and CA16 infection, seroepidemiological surveillance is urgently needed. In this study, we conducted a cohort study in Guangdong province, China. Children less than 10 years old were enrolled. The levels of EV71 and CA16 specific antibodies in children aged between 1 and 9 years were evaluated and compared for 3 years (2007–2009).
Discussion
In humans, humoral immunity with neutralizing antibodies played central roles in anti viral infection [
35‐
37]. Recent vaccine studies in animal model also demonstrated the protective effects of neutralizing antibody in EV71 challenge [
38,
39]. In addition, surveillance data in other countries also suggested a causative role for the decreased herd immunity in susceptible population during HFMD outbreaks [
40]. To date, few surveillance data were reported on seroepidemiology of EV71 and CA16 in China. In this study, the neutralizing antibodies against EV71 and CA16 among children in Guangdong were investigated. Seroprevalence changes before, during and shortly after the national widespread epidemic in 2008 were described.
In China, HFMD was first reported in Shanghai in 1981, and since then, sporadic cases of HFMD in other provinces including Beijing, Fujian, Shandong, Hubei, Jilin were reported in 1980s. Recently the epidemic situation of HFMD in China tended to be more serious. In 2007, nearly 40,000 HFMD cases were reported in Shandong and over 10,000 cases were recorded in large cities such as Beijing and Shanghai [
41]. However, little reports have described the HFMD situations in Guangdong in 2007. Our results suggested moderate EV71 infections occurred among the children in Guangdong before the large HFMD epidemic in 2008. The overall seroprevalence of EV71 NA in 2007 was similar to those in 2005 in Guangdong, as well as those in 2006–2007 in Lu’an [
42,
43]. About half of children less than 10 years old had no neutralizing antibody against EV71. Compared to EV71, significant higher positive rate of neutralizing antibody against CA16 in 2007 was identified. This was contrary to a previous study, which suggested more common EV71 infections in South China in 2005 [
42]. The high CA16 seroprevalence in 2007 indicated that frequent asymptomatic and/or unrecognized CA16 infections have occurred before and in this year. In 2008, among 936 laboratory-confirmed cases, EV71 was found in 59% and CA16 in 26% [
19]. Considering the protective effects of neutralizing antibody in viral infection, the epidemiology surveillance data in 2008, to some degree, had verified our results. The seroprevalence of EV71 and CA16 in this year could serve as HFMD trends predictors for next year.
In March 2008, a serious HFMD epidemic with 20 deaths broke out in Fuyang City, Anhui Province, and subsequently spread quickly in almost all provinces of China [
44]. After that, the Ministry of Health of China classified HFMD as one of the category “C” notifiable diseases. According to the Guangdong HFMD web-based surveillance system, a total of 47,660 HFMD cases were reported in 2008. And this number almost doubled in 2009 (92,998 reported cases), was five-fold in 2010 (230,978 reported cases) [
41]. To correlate the level of immune protection to the incidence rate of HFMD cases, seroprevalence of EV71 and CA16 NA in and after the 2008 were also investigated. We didn’t find significant higher positive rate of EV71 and CA16 in and after 2008. Previous study suggested that HFMD presents a seasonal pattern every 2–3 years [
45]. Together with the increased incident rates of HFMD cases in 2009–2010, our results suggested that the peak of recent HFMD epidemic cycle in Guangdong was not in year 2008 but in the years 2009 and 2010. Consistent with these, Yu, et al. identified that children in Lu’an had significant higher seropositive rate in 2010 when compared to that before 2008 [
43]. The ensuing HFMD epidemics from 2008 to 2010 largely increased the exposed chance of viral infection and thus the seropositive rate of viral NA in children.
EV71 and CA16 were highly diverse in the nucleotide sequences of structure proteins which serve as major antigen in host immune response. Li et al. used VP1s and VP4s as antigens to detect of serum antibodies against EV71 and CA16 [
46]. Their results suggested immunological reaction to VP1 and VP4 of both EV71 and CA16 was different. The study from the clinical patient also indicated that individuals with or without prior EV71neutralizing antibody showed a similar incidence of non-71 Enterovirus infection [
47]. EV71 and CA16 were the most common causes of HFMD diseases in recent epidemics in China [
44]. To evaluate the immune protection level against HFMD, we also calculated the proportion of individuals that were positive for both the EV71 and CA16. In total, about 30.0% of all tested individuals had antibodies both to EV71 and to CA16 before and during the 2008 epidemic, while this rate greatly reduced to 20.0% in year 2009. The large number of susceptible individuals may be partly responsible for the large outbreak of HFMD in Guangdong in 2010. Similar with reports in other countries, the dates in our study also proved that individual susceptible probability depends on the protective level of neutralizing antibodies. Epidemic occurred when herd immunity decreased.
As a common febrile illness of early childhood, HFMD occurred mainly among children ≤5 years old. In Guangdong, a total of 47,660 and 92,998 cases have been reported to the provincial surveillance system in 2008 and 2009 respectively. The number of children ≤5 years old accounted for the largest proportion (from 87.5% to 93.3%) [
48]. These were highly consistent with the age related seroprevalence trends in our study. In our results, both EV71 and CA16 NA rise with age among children less than 5 years and reach a plateau thereafter [
34,
49]. We also observed significant reduction at 2 years group, the most susceptible population that had the highest incidence rates of HFMD in 2008 and 2009 epidemic year in Guangdong. In 2007, the lowest seroprevalence was identified in 1 year old children. While for children in 2 year old group, slightly high positive rate was shown. Epidemic waves before and in this year could not be traced. Whether this different seroprevalence trend in 2007 provided an accurate picture, or it just was the result of very small number of cases in 1 year group remains to be determined by future investigations.
The GMT for EV71 and CA16 was also analyzed. All the seropositive individuals in three years tested showed relative high CA16 NA levels (Figure
2). Unlike a previous report conducted in Jiangsu province China, which showed low antibody titers against CA16 in children aged 2–15 years, the high GMT observed in this study suggested common CA16 infections in Guangdong province [
27]. In addition, our data also suggested that most of the EV71 infections are acquired when they were less than 5 years old, while for CA16, mainly occurred among 1–3 years groups (Figure
3), because the highest antibody titers were observed in these age groups and neutralizing antibody titers were high at early stage of infection. GMT for both the EV71 and CA16 were all declined among children 5 years or older, which account for about 10% of the total reported HFMD cases [
48]. The relative low level of EV71 and CA16 NA in these less susceptible populations indicated a critical role for the long-lasting immunity rather than immunity level in protecting against viral infection.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
YL and KC drafted the manuscript. LW, SJ, LJ, ZH, GD, MC, ZW and XH carried out the laboratory tests, LH, ZY and LJ performed the statistical analyses. All authors read and approved the final manuscript.