The Epidemiological and Clinical Characteristics of Hand, Foot, and Mouth Disease in Hangzhou, China, 2016 to 2018.

Hand, foot, and mouth disease (HFMD) is most frequently caused by several serotypes of human enterovirus (EV) including Enterovirus 71 (EV-A71), coxsackievirus A16 (CV-A16), or other types of EV. The aim of this study was to determine the epidemiological characteristics of HFMD and to describe the epidemiologic characteristics of HFMD among severe and mild cases. We collected 4760 HFMD cases in Hangzhou from 2016 to 2018. Specimens from these cases were collected and tested for EV-A71, CV-A16, CV-A6, CV-A10, CV-A2, and CV-A5 by reverse transcriptase polymerase chain reaction. From 2016 to 2018, the prevalence of HFMD was seasonal each year. Among the 4760 probable HFMD cases, 3559 cases were confirmed (74.8%), including 426 cases of EV-A71 infections (8.9%), 249 cases of CV-A16 infections (5.2%), and 2884 cases of other EV infections (60.6%). The percentage of other EV infections was more than 80%, which increased year by year. Random selection of samples for detection of other EV infections in 2017 and 2018, among the 1297 cases, showed there were 835 (64.4%) cases of CV-A6 infections, 177 (13.6%) cases of CV-A10 infections, 100 (7.7%) cases of CV-A2 infections, 40 (3.1%) cases of CV-A5 infections, 3 (0.02 %) cases of mixed infections, and 11.0% untyped EV infections. Preschool children were still the primary population susceptible to HFMD. In severe cases, EV-A71 infection was the main cause. Characterizing the epidemiology and the relationship between severe and common cases of HFMD would provide relevant evidences for the prevention and treatment of HFMD.


Introduction
Hand, foot, and mouth disease (HFMD), a common childhood illness, is increasingly recognized as a significant health issue worldwide. HFMD poses a significant threat to public health in China and was classified as a C-class notifiable communicable disease in China in 2008. Clinical manifestations of HFMD are diverse and complex. Main symptoms include vesicular rashes on palms, soles, mouth, and/or buttocks with or without fever and central nervous system (CNS) involvement in severe cases. 1 HFMD is caused by human enteroviruses (EVs). Enterovirus A71 (EV-A71) and coxsackievirus A16 (CV-A16) are the most common pathogens causing HFMD in children. [2][3][4][5] EV-A71-induced HFMD is often benign and self-limiting, but it also can be associated with neurologic complications. These complications include encephalitis, aseptic meningitis, transverse myelitis, cerebellar ataxia, and Guillain-Barre syndrome. [6][7][8][9][10] CV-A6 and CV-A10 are the most common serotypes for mild HFMD worldwide and in China since 2010. [11][12][13][14][15][16][17][18] We found that the number of HFMD cases has increased since the nationwide HFMD outbreak in China in 2008; HFMD has been the leading cause of significant mortality in China since 2013 to present, according to the report of National Health Commission of the People's Republic of China.
In this study, we collected specimens from 4760 patients clinically diagnosed with HFMD in Hangzhou Children's Hospital from January 2016 to December 2018. And we aim to describe the epidemiological characteristics and the relationship between severe and mild HFMD, which would provide relevant evidences for the prevention and treatment of HFMD.

Study Participants
A total of 4760 specimens (1384 stool specimens, 3315 throat swab specimens, and 61 cerebrospinal fluid specimens) were collected from children clinically diagnosed with HFMD in Hangzhou Children's Hospital from January 2016 to December 2018. HFMD cases were clinically diagnosed according to the Ministry of Health Diagnostic Criteria (2018 edition). 19 Children who displayed vesicular rashes on the hands, feet, oral mucosa, or buttock in epidemic seasons were clinically diagnosed with HFMD. In this study, 783 children with serious complications, including CNS complications (such as encephalitis, meningitis, and brain stem encephalitis) and/or cardiorespiratory failure, were considered as having severe HFMD.
The main clinical manifestations of CNS complications are headache, vomiting, irritability, lethargy, nuchal rigidity, and myoclonus. The main clinical manifestations of cardiorespiratory failure are pulmonary edema or pulmonary hemorrhage. Children without the serious complications mentioned above were classified as having mild HFMD. This study and the use of clinical samples were approved by the Medical Ethics Committee of Hangzhou Children's Hospital and all experiments were performed in accordance with relevant guidelines and regulations.

RNA Extraction and Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
Clinical specimens were collected and stored at −80°C for extraction of genomic RNA. RNA was extracted by nucleic acid automatic extraction instrument (NP968 Tian-Long Company, Xian, China), operated strictly according to the operating instructions of the nucleic acid extraction kit. The detection of EV/EV-A71/ CV-A16 was performed in an ABI 7500 system via a commercial one-step real-time RT-PCR assay kit (Shuo-Shi Company, Jiangsu, China). The real-time RT-PCR was conducted under the following conditions: 30 minutes at 45°C, 5 minutes at 94°C, and then followed by 40 cycles of 10 seconds at 94°C and 40 seconds at 55°C.
To download multiple gene sequences including CV-A6, CV-A10, CV-A2, and CV-A5 from the National Center for Biotechnology Information database, the specific primers and probes of CV-A6, CV-A10, CV-A2, and CV-A5 subtypes for fluorescent quantitative RT-PCR assay were designed with Primer Express 3.0 software, and the primers and the probes' sequences were validated by Blast. PCR primers were synthesized by Shanghai ShengGong Gene Co Ltd.
The RT-PCR mixture for each tube consisted of 5 µL viral RNA, 12.5 µL one-step RT-PCR buffer, 0.5 µL Takara EX Taq HS, 0.5 µL RT enzyme, 1.6 µL specific primer mixture, and 4.9 µL nuclease-free water, up to a final volume of 25 µL. The PCR was conducted under the following conditions: 30 minutes at 50°C, 5 minutes at 95°C, and then followed by 40 cycles of 15 seconds at 94°C and 45 seconds at 55°C.

Statistical Analysis
All statistical analyses were performed using SPSS version 19.0, and χ 2 test was used to compare the distribution of EV-A71, CV-A16, and other EV-positive rates for the groups categorized by age and gender. Count data were compared in percentage. All statistical tests were 2 sided, and P < .05 was considered statistically significant. convenience, we defined EV-positive but both EV-A71 and CV-A16 negative cases as "other EV infections cases." Figure 1 shows that although HFMD cases occurred throughout the year, we observed the number of cases decreased between January and March, and obviously increased between May and July in the last 3 years. Two seasonal peaks were observed in the total HFMD cases in 2016 and 2017. The first and larger peak was between May and July reflecting the summer HFMD epidemic. The second peak was between October and November. Interestingly, the second peak was not obvious and the time was advanced to September in 2018, which is different from the previous 2 years.

Characteristics of the Clinical Data From HFMD Patients
The distribution of HFMD causative pathogens in mild cases showed that other EV infections cases predominated from 2016 to 2018, accounting for 88.2% (2594/2940). However, EV-A71-positive cases predominated in severe cases from 2016 to 2018, accounting for 47.7% (295/619), which was a little higher than the proportion of other EV-positive cases. Further analysis showed that the severe cases with EV-A71 positive accounted for 69.2% of the total EV-A71-positive cases (295/426). EV-A71 was more significantly associated with severe HFMD compared with CV-A16 (13.7%, 34/249) and other EV infections cases (10.1%, 290/2884; P < .05; Table 1).
The gender distribution ratio of male to female cases was 1.66:1 in 3559 EV-positive cases. There are 2223 males, consisting of 276 EV-A71, 169 CV-A16, and 1778 other EV infections cases, and there are 1336 females, consisting of 150 EV-A71, 80 CV-A16, and 1106 other EV infections cases. There is no gender difference in the rate of infection of different EV (χ 2 = 4.901, P = .086). Age distribution showed that children younger than 5 years of age accounted for more than 90%. In total cases, the children aged between 1 and 2 years had the highest incidence of HFMD. The average ages for EV-A71, CV-A16, and other EV infections were 2.5, 2.3, and 1.8 years, respectively. The mean age of patients with other EV infections was younger than that of those with EV-A71 or CV-A16 infections. The average ages for EV-A71 infections were 2.3, 2.5, and 4.1 years in 2016, 2017, and 2018, respectively. The mean age of patients with EV-A71 infections tended to be older year by year. In this study, EVs were detected in throat swab and feces, and there was no significant difference in the detection rate of EVs in throat swab and feces (χ 2 = 0.127, P = .722). No EV was detected in cerebrospinal fluid specimens in our study (Table 2).

Novel Epidemic Patterns in HFMD Etiology
As per our laboratory surveillance system, the annual proportions of other EVs with both non-EV-A71 and non-CV-A16 were 69.5%, 80.4%, and 92.9% from 2016 to 2018, which increased year by year. However, the annual proportions of EV-A71 decreased year by year in the last 3 years, and especially in 2018 only 13 cases were detected (Table 1). Other EVs replaced EV-A71 and CV-A16 as the major serotype during 2016 to 2018, accounting for more than 80% of EV-positive HFMD cases. Regular fluctuations in EV-A71 infection proportions were observed with one large EV-A71 infection peak each in 2016 and 2017 lasting from May to July, while the other EV infection peak occurred in autumn and winter accounting for more than 80% of these cases in these 2 years. Interestingly, there was no EV-A71 infection peak in 2018. CV-A16 distributed evenly throughout the year in the 3 years (Figure 1).

Discussion
In the past few decades, HFMD has contributed to several large outbreaks and thousands of HFMD fatal cases worldwide, which has been a growing public health problem. [20][21][22] After the first report of HFMD in Shanghai in 1981, alternating epidemics of EV-A71 and CV-A16 occurred in the following years. Our previous research revealed that EV-A71 and CV-A16 represented the 2 major epidemic EV serotypes in Hangzhou between 2012 and 2013. However, the predominant types of EVs have changed since 2013, and the positive rates of the EV-positive HFMD cases that were both non-EV-A71 and non-CV-A16 have grown rapidly. The other EVs replaced EV-A71 and CV-A16 as the major serotype in Hangzhou between 2014 and 2015. This study continues to analyze the characteristics of HFMD epidemic that occurred in Hangzhou in 2016, 2017, and 2018.
In this study, we reported that other EVs (which is CV-A6) had replaced EV-A71 as the primary viral serotype responsible for HFMD onset in Hangzhou between 2016 and 2018. Concurrently, the same trend was observed in other cities near Hangzhou in China. 18,23 The cases of EV-A71 decreased year by year in the past 3 years, and compared with 2016 and 2017, only 13 cases of EV-A71 were detected in 2018, which meant a significant decrease in the proportion. Therefore, 1297 cases randomly selected of the other EVs were tested in 2017 and 2018. There were 835 (64.4%) cases of CV-A6 infections and 177 (13.6%) cases of CV-A10 infections in 2017 and 2018. Less mixed infections were detected among different types in this study. The cases of CV-A6 infections also increased dramatically and CV-A6  became one of the main causative agents of HFMD in Hangzhou. CV-A10 also overtook CV-A16 to become the second most common viral serotype. Hangzhou is the capital city of Zhejiang province, and its high population mobility made it a HFMD-prone area and likely more conducive to virus' faster evolution. Some epidemiologists suggest 3 primary factors were involved in the regionally increased CV-A6 infection rates and decreased EV-A71 infection rates. First, after a long period of the EV-A71 epidemic, elevated antibody and thus immunity levels against EV-A71 in the population might impart a selection pressure for other EV serotypes to arise, because antibodies against EV-A71 and CV-A16 cannot protect the susceptible population from the other serotypes. 24 Second, the popularity of EV-A71 vaccination may play a role in this change in recent years. Mass EV-A71 vaccination is not expected to substantially reduce the total number of HFMD cases because the vast majority of HFMD cases are mild and more than half of the mild cases were due to CV-A16 and other EV serotypes infections. However, most of severe and death HFMD cases could be prevented by EV-A71 vaccination. 25 Third, CV-A6 might have transformed into a more virulent strain to escape human immune protection. Yoshitomi's study showed that the phylogenetic analysis of the entire VP1 gene revealed the diversity of the prevalent CV-A6 between 2013 and 2017 in Fukuoka. The CV-A6 strains were classified into 7 genetic clades (A-G) and subgroups of clade A (subclades A1-A4) based on the entire VP1 sequences. The phylogenetic analysis revealed that all the CV-A6 strains detected in Fukuoka between 2013 and 2017 were classified into clade A (A3 and A4), which were different from Hong Kong (D4 and D5). 22,26 The emergence of CV-A6 strains with high genetic diversity was believed to be a factor associated with the epidemics. More studies including whole genome sequencing of the epidemic strains are warranted to determine whether these variations affect the epidemic incidence of CV-A6. We think more attention should also be paid to the other EV serotypes, such as CV-A6 and CV-A10, in the future surveillance and control of HFMD in Hangzhou and even in Zhejiang province. We speculate that the year of 2018 may become a turning point in the EV serotypes causing HFMD, and we should pay more attention to this phenomenon. Additionally, Hangzhou shared similar seasonal epidemic patterns of EV-predominant serotypes with other southern cities in China. Our data showed 2 seasonal peaks in 2016 and 2017 in the total HFMD cases. Similar finding was reported in Shenzhen, Wuxi, and Hong Kong. [26][27][28] In these 3 districts, the season alternations from spring to summer and from autumn to winter are relatively mild, which might be in favor of the spread of certain EV viral serotypes. This is consistent with the results of some studies that found the number of patients with HFMD is affected by environmental factors such as temperature, humidity, sunshine time, and so on. 29 In this study, we reported that other EV infection peaks mainly occurred during October to December in 2016 and 2017. Di's study also found CV-A6 infection mainly occurred during the autumn and winter, and the outbreak of CV-A6 infection might contribute to the second peak of the HFMD. 30 In 2018, the main infection was other EV infection in Hangzhou, and there was no EV-A71 infection peak, which might also be the reason why the second peak was not obvious in 2018. CV-A16 displayed a low detection rate throughout the year.
In our study, we investigated whether the etiological spectrum of mild and severe HFMD changed. EV-A71 was still the leading serotype responsible for severe HFMD cases, while the other EV infection accounted for an increase in severe cases. Between 2016 and 2018, other EV infection cases accounted for 46.8% of EV-positive specimens in severe HFMD cases in Hangzhou. This proportion was much higher than the result of our study in the previous several years. This may be due to the change in predominant types of the prevalent EV serotype distributions throughout the year. In our study, the severe case rate of EV-A71 infections (69.2% of the total EV-A71-positive cases) was significantly higher than that of CV-A16 infections (13.7%, 34 of 249) and other EV infections (10.1%, 290 of 2884), which suggested that EV-A71-infected patients had still a greater chance of developing severe cases of HFMD than other EV serotype-infected patients. However, the pathogenic mechanism of EV-A71 severity and fatality remained unclear. Li et al's report suggested that the different genetic background of host innate immunity may lead to different clinical outcomes. 31 They found that RIG-1 rs3739674 and RIG-1 rs9695310 polymorphisms are associated with EV-A71 HFMD risk. RIG-1 rs3739674 and TLR3 rs5743305 polymorphisms are associated with disease severity. 31 Previous reports showed that the levels of serum interleukin-6, B-type natriuretic peptide, hyperglycemia, and leukocytosis in children with severe neurological symptoms caused by EV-A71 infection significantly increased, causing a systemic inflammatory factor storm and leading to heart and lung failure, which might indicate a strong relationship between pro-inflammatory cytokines and severity in EV-A71 infection. 32,33 Statistical analysis of the gender distribution of all cases in the present study showed more infection in males than in females in our study. Male predominance of HFMD was observed, possibly because of higher physical activity and less attention to hygiene of boys. However, there was no gender difference in the ratio of each different EV infection. Statistical analysis of the age in the present study showed that children younger than 5 years were the most susceptible to HFMD, which is consistent with previous reports. 34 We found the age of patients with other EV infection was younger than those with EV-A71 or CV-A16 infection. With the popularity of other EV serotypes, the smallaged infants become the focus of the prevention and treatment of HFMD in Hangzhou. We also found that the average age of EV-A71 infection gradually increased from 2016 to 2018, especially in 2018, indicating that the protection of EV-A71 vaccine in lower age groups has been obviously effective. No EV was detected in cerebrospinal fluid specimens in our study. Relevant reports elsewhere also showed that the detection rate of EV nucleic acid in cerebrospinal fluid was lower than in throat swab, feces, or bleb fluid. 35

Conclusion
Other EV replaced EV-A71 and became the predominant causative agent of HFMD in Hangzhou, China, from 2016 to 2018. Other EV serotypes such as CV-A6 and CV-A10 have recently surfaced in China, and they have become the prominent serotypes leading to HFMD in Hangzhou. Thus, monitoring the long-term serotypes' responses to EV strains such as CV-A6 and CV-A10 should be emphasized in the future, as the EV-A71 vaccine has been successfully applied in Hangzhou and other regions in China. Further research on effective multivalent vaccines is needed to prevent and control the outbreak and epidemic of HFMD.

Author Contributions
JW: Conceptualized and designed the study, coordinated and supervised data collection, contributed in preliminary data analysis, and drafted the initial manuscript. JZ: Collected the data and carried out the laboratory data analysis. GX, SZ, BL: Supervised the laboratory techniques and outcome assessment. YC: Critically reviewed and revised the manuscript. YW: Conceptualized and designed the study and critically reviewed and revised the manuscript. All authors approved the final manuscript as submitted and agreed to be accountable for all aspects of the work.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The design of this study and collection of data were supported by the Zhejiang Medical and Health Project (2014KYA186); analysis of data of this study was supported by the Zhejiang Natural Science Foundation Project (LY13H040007); and interpretation of data was supported by the Zhejiang Province Science and Technology Project (2017C33044) and the National Key Programs for Infectious Diseases of China (2017ZX10103008).