Introduction: Human enterovirus D68 (EV-D68), which belongs to enteroviruses of the small RNA family, is a type of enterovirus that can cause acute respiratory tract infection and central nervous system diseases. This study systematically analysed and summarized EV-D68 antibody studies in databases and identified the seropositivity rates of different regions, ages, and sexes. Methods: Meta-analysis was performed using STATA 16.0 software. I2 and Q tests were used to analyse the heterogeneity of the included studies. Meta-regression analysis was performed for different groups, and Egger’s linear regression analysis was used to evaluate publication bias. Results: The results of multiple studies indicated that the serological prevalence range of EV-D68 antibody was 17.78–96.69%. The results of the meta-analysis showed that the seropositivity rate of EV-D68 antibody was 76% (95% confidence interval [CI]: 67–84%), among which that of the Chinese population was 74% (95% CI: 61–86%) and that of other countries was 79% (95% CI: 65–91%). At the same time, a subgroup analysis was conducted. The seroprevalence of EV-D68 antibody was related to age but not sex or region. Conclusion: The seropositivity rate was lower in the below 5-year age group; however, it gradually increased with age. The results of this study showed that EV-D68 infection was widespread in the population, and the current clinical infection situation could not reflect the actual epidemic situation of the virus, among which children under 5 years old were vulnerable to infection, which should be given greater attention for epidemic prevention and control.

Enterovirus D68 (EV-D68) belongs to human enterovirus D, which is a single-stranded positive-strand RNA virus that is not enveloped [1]. The genome of EV-D68 only has an open reading frame that encodes the production of four capsid proteins and seven nonstructural proteins [1]. Unlike most enteroviruses, EV-D68 is mainly isolated from respiratory tract samples, and its biological and molecular characteristics are more similar to those of rhinoviruses [2].

EV-D68 infection can cause symptoms of respiratory tract infection to varying degrees, which can usually manifest as fever, cough, asthma, sore throat, and general discomfort; however, severe cases can be further complicated with bronchitis, pneumonia, and asthma [3]. In addition, EV-D68 infection can cause serious neurological complications and even death [4, 5]. Recently, acute flaccid myelitis (AFM) due to EV-D68 infection has been found in Switzerland, China, the USA, France, and Japan [6‒10]. Furthermore, EV-D68 cases with concomitant neural dysfunction have also been reported [11].

EV-D68 was first found and reported in 1962 in CA, USA, and a small number of cases have continued to be reported since then [12]. In 2014, the largest outbreak of EV-D68 in the history of the USA occurred, spanning a total of 2,287 cases that were confirmed throughout the year, including 14 deaths among children, which caused widespread global concern [12]. Since then, cases or outbreaks of EV-D68 have been reported in many regions globally, making it a new target for respiratory virus surveillance [13, 14]. The World Health Organization has identified EV-D68 as a new potential public health risk factor and called on countries to strengthen surveillance, prevention, and control [15]. In recent years, epidemiological data related to EV-D68 have gradually increased [16‒19]. Overall, there is a lack of systematic and comprehensive surveillance and documentation of EV-D68 prevalence globally, and latent infection of EV-D68 is not being taken seriously. Acute flaccid paralysis caused by EV-D68 infection is increasing, and there is currently a lack of effective drugs and vaccines against the virus [6, 20, 21]. Therefore, it is very important to further strengthen the understanding of virus serology for disease prevention.

Serological investigation of EV-D68 antibody levels in the human population serves as an effective method to clarify the prevalence of EV-D68 [22]. In this study, a comprehensive analysis of published EV-D68 serological studies further clarified the dynamic changes in antibodies and susceptibility to the virus in individuals of different races, regions, ages, and sexes. On the one hand, it helps us systematically and comprehensively understand the epidemic situation of the virus. On the other hand, it also provides a theoretical basis for the staff of disease prevention and control departments to develop effective prevention and immunization programs and to provide a reference for the effective prevention and control of EV-D68-related diseases.

Search Strategy

Systematic Evaluation and Meta-analysis (PRISMA) guidelines were followed throughout the study design process (online suppl. Table S1; for all online suppl. material, see https://doi.org/10.1159/000531853). This study systematically analysed serological studies related to EV-D68 published prior to February 2022. Two Chinese databases (CNKI and Wanfang) and three English databases (PubMed, Embase, and Cochrane Library) were selected for literature retrieval (online suppl. Table S2). To ensure the integrity of retrieval and analysis, we used manual retrieval for each database.

The following free words or subject words were used when searching: (“Enterovirus D68” OR “Human Enterovirus D” OR “Enterovirus 68” OR “EV-D68” OR “Enterovirus 68, Human” OR “Human Enterovirus 68”) AND (“Seroepidemiologic Studies” OR “Study, Seroepidemiologic” OR “Seroprevalence” OR “Seroprevalences” OR “antibody”).

Inclusion and Exclusion Criteria

The relevant literature was screened with reference to the following criteria: inclusion criteria included retrospective or cross-sectional studies published before February 2022. In this paper, the specific data of EV-D68 antibody were determined by specific experiments. The exclusion criteria were as follows: the document was a conference paper or review, not a research paper; the literature data were unknown or wrong, and valid and useful data could not be provided.

Literature Screening and Data Extraction

We first screened the valid literature by referring to the previous inclusion and exclusion criteria and then conducted an intensive reading of the valid literature. At the same time, valid information in the literature was collected in a standardized table, including the author, collection time and place of serum samples, detection methods and criteria, detection and selection of virus strains and genotypes, age and sex of the study population. Meanwhile, we classified the methods used in this study, and the criteria for “random sampling,” “conditional sampling,” and “physical examination” methods are shown in previous studies [23]. Seropositivity is defined as the proportion of EV-D68 serum antibody-positive individuals to total serum samples in the sample to be tested. To reduce the analysis error, we also arranged two researchers (S.Y. and S.L.) to screen the title, abstract, and text of the literature and exclude the literature that did not meet the statistical and analytical requirements. Then, the text was intensively read by two researchers (S.Y. and S.L.) to determine the valid literature and extract valid data according to the requirements. If two researchers disagree on some aspects of the same literature, we will arrange a third researcher (R.Q.) to carry out statistical analysis on the literature and data (Table 1; online suppl. Table S3).

Table 1.

Basic information of the literature included in this analysis

First authorPublication yearCountryLocationSample sizeNo. of positiveSeroprevalence rate (%)Assay methodPositive thresholdAge rangeGroup factorsLanguageAHRQ scores
Xiang 2017 China Beijing 393 355 90.33 NTA 1:8 0–93 years Age ENG 
Sun 2018 China Jiangsu 385 228 59.00 NTA 1:8 1 month–15 years Age ENG 
Sun 2018 China Jiangsu 177 87 49.15 NTA 1:8 6–35 months Age, gender ENG 
Hu 2019 China Taiwan 124 83 66.94 NTA 1:8 2–5 years Age, gender ENG 
Lee 2020 China Taiwan 920 612 66.52 NTA 1:8 0–49 years Age, gender, region ENG 
Liu 2020 China Beijing 1,050 971 92.48 NTA 1:8 0–70 years Age, gender ENG 
Lin 2021 China Xiamen 515 422 81.90 ELISPOT 1:16 5 months–83 years Age, gender, regions CHN 
Takaya 2018 Japan Sendai 90 16 17.78 PCR, NTA 1:4 25 days–13 years Age, gender ENG 
Kadji 2020 Japan Sendai 103 94 91.26 NTA log2 ≥3 0–6 years Age, gender ENG 
Chan 2021 Malaysia Kuala Lumpur 450 342 76.00 NTA 1:16 1–89 years Age, gender ENG 
Karelehto 2019 Netherlands Dutch 242 234 96.69 NTA 1:8 0–79 years Age, gender ENG 
Kamau 2019 UK Unknown 1,082 894 82.62 NTA 1:16 0.5–40 years, >40 years Age ENG 
Harrison 2019 USA Kansas 436 394 90.37 NTA 1:8 2–81 years Age, gender ENG 
First authorPublication yearCountryLocationSample sizeNo. of positiveSeroprevalence rate (%)Assay methodPositive thresholdAge rangeGroup factorsLanguageAHRQ scores
Xiang 2017 China Beijing 393 355 90.33 NTA 1:8 0–93 years Age ENG 
Sun 2018 China Jiangsu 385 228 59.00 NTA 1:8 1 month–15 years Age ENG 
Sun 2018 China Jiangsu 177 87 49.15 NTA 1:8 6–35 months Age, gender ENG 
Hu 2019 China Taiwan 124 83 66.94 NTA 1:8 2–5 years Age, gender ENG 
Lee 2020 China Taiwan 920 612 66.52 NTA 1:8 0–49 years Age, gender, region ENG 
Liu 2020 China Beijing 1,050 971 92.48 NTA 1:8 0–70 years Age, gender ENG 
Lin 2021 China Xiamen 515 422 81.90 ELISPOT 1:16 5 months–83 years Age, gender, regions CHN 
Takaya 2018 Japan Sendai 90 16 17.78 PCR, NTA 1:4 25 days–13 years Age, gender ENG 
Kadji 2020 Japan Sendai 103 94 91.26 NTA log2 ≥3 0–6 years Age, gender ENG 
Chan 2021 Malaysia Kuala Lumpur 450 342 76.00 NTA 1:16 1–89 years Age, gender ENG 
Karelehto 2019 Netherlands Dutch 242 234 96.69 NTA 1:8 0–79 years Age, gender ENG 
Kamau 2019 UK Unknown 1,082 894 82.62 NTA 1:16 0.5–40 years, >40 years Age ENG 
Harrison 2019 USA Kansas 436 394 90.37 NTA 1:8 2–81 years Age, gender ENG 

NAT, neutralizing antibody test; ENG, English; CHN, Chinese.

Quality Assessment

During literature quality evaluation, the evaluation criteria of the Agency for Healthcare Research and Quality (AHRQ) were referred to for various studies. This standard contains 11 items, and the literature obtained by effective screening is evaluated according to relevant standards. Refer to previous studies for specific evaluation criteria and grades [23].

Statistical Analysis

In this study, Stata 16.0 software was used for meta-analysis of literature data, and the seropositivity rate and 95% confidence interval (CI) were used for statistics. In the analysis of heterogeneity, if p ≥ 0.1 and I2 ≤50%, it indicates that there is homogeneity among studies, so the fixed-effect model was selected for statistical analysis. Conversely, if p < 0.1 and I2 >50%, indicating heterogeneity among various studies, a random-effects model was selected for data analysis [24]. At the same time, when the heterogeneity is large, subgroup analysis should be used to further explore the source of heterogeneity, and different age, sex, and sampling methods were used as grouping factors. In this study, sensitivity analysis was used to evaluate the stability of the meta-analysis results, and Egger’s test was used to evaluate publication bias.

Study Search Results

Based on PRISMA guidelines, we conducted a systematic search of the EV-D68 serological literature in five databases (Fig. 1; online suppl. Table S2). Further screening was conducted according to preset inclusion and exclusion criteria. After removing duplicate and invalid literature, we obtained 13 valid research studies and carried out systematic statistical analysis on the valid information in the literature [25‒37]. These 13 studies included a total of 5,967 samples, of which 4,732 were positive for EV-D68 neutralizing antibody, with the total positive rate of serum EV-D68 antibody ranging from 17.78 to 96.69%. Among the 13 studies, 5 were from mainland China, 2 were from Taiwan, 2 were from Japan, 1 was from Malaysia, 1 was from the Netherlands, 1 was from the UK, and 1 was from the USA (Table 1). We evaluated the quality of the literature by referring to AHRQ standards. Eight of the included studies were judged to be of high quality, and five were of medium quality.

Fig. 1.

Flowchart of the literature selection process.

Fig. 1.

Flowchart of the literature selection process.

Close modal

Seroprevalence of EV-D68 Antibody

The total seropositivity rate was analysed in groups according to age, sex, sampling method, seasonality, language of publication, and race to clarify the correlation between each factor and the seropositivity rate (Table 2). The forest diagram of the total seropositive rate is shown in Figure 2, and the total analysis results are shown in Table 2. The seroprevalence of EV-D68 antibody was found to be 76% (95% CI: 67–84%) in the overall population and 74% (95% CI: 61–86%) in the Chinese population (Table 2). Moreover, the age-based subgroup analysis indicated that the seroprevalence of EV-D68 antibody was 52% (38–65%) in the <1 age group, 45% (23–69%) in the 1–3 age group, 61% (95% CI: 48–74%) in the 0–5 age group, and 88% (95% CI: 77–95%) in the 5 and older age group (Fig. 3). Meanwhile, the subgroup analysis based on sex indicated that the seroprevalence of EV-D68 antibody was 60% (95% CI: 33–85%) in males and 61% (95% CI: 33–85%) in females (Fig. 4).

Table 2.

Seroprevalence in different groups

VariableRef (n)nEventTransformation methodSeroprevalence (%)95% CI (%)I2 (%)Heterogeneity (p value)
Total 13 5,967 4,732 Freeman-Tukey double arcsine 0.76 0.67, 0.84 98.34 <0.001 
Age 
 <1 year 692 403 0.52 0.38, 0.65 92.4 <0.001 
 1–3 years 871 439 Freeman-Tukey double arcsine 0.45 0.23, 0.69 97.96 <0.001 
 <5 years 11 2,111 1,242 Freeman-Tukey double arcsine 0.61 0.48, 0.74 97.22 <0.001 
 >5 years 3,216 2,839 Freeman-Tukey double arcsine 0.88 0.77, 0.95 98.28 <0.001 
Gender 
 Female 1,234 729 Freeman-Tukey double arcsine 0.61 0.33, 0.85 98.89 <0.001 
 Male 1,301 731 Freeman-Tukey double arcsine 0.60 0.28, 0.87 99.20 <0.001 
Sampling method 
 Physical examination 450 342 Freeman-Tukey double arcsine 0.76 0.72, 0.80 
 Random sampling 3,135 2,609 Freeman-Tukey double arcsine 0.83 0.72, 0.91 98.04 <0.001 
 Conditional sampling 2,382 1,781 Freeman-Tukey double arcsine 0.71 0.53, 0.87 98.72 <0.001 
Seasonality 
 Season 2,015 1,388 Freeman-Tukey double arcsine 0.67 0.50, 0.82 98.18 <0.001 
 Non-season 177 87 0.492 0.418, 0.565 
 Unknown 3,775 3,257 Freeman-Tukey double arcsine 0.88 0.81, 0.93 96.28 <0.001 
Publication language 
 English 12 5,452 4,310 Freeman-Tukey double arcsine 0.76 0.66, 0.85 98.48 <0.001 
 Chinese 515 422 Freeman-Tukey double arcsine 0.82 0.78, 0.85  
Ethnicity 
 Chinese 3,564 2,758 Freeman-Tukey double arcsine 0.74 0.61, 0.86 98.61 <0.001 
 Others 2,403 1,974 Freeman-Tukey double arcsine 0.79 0.65, 0.91 98.18 <0.001 
VariableRef (n)nEventTransformation methodSeroprevalence (%)95% CI (%)I2 (%)Heterogeneity (p value)
Total 13 5,967 4,732 Freeman-Tukey double arcsine 0.76 0.67, 0.84 98.34 <0.001 
Age 
 <1 year 692 403 0.52 0.38, 0.65 92.4 <0.001 
 1–3 years 871 439 Freeman-Tukey double arcsine 0.45 0.23, 0.69 97.96 <0.001 
 <5 years 11 2,111 1,242 Freeman-Tukey double arcsine 0.61 0.48, 0.74 97.22 <0.001 
 >5 years 3,216 2,839 Freeman-Tukey double arcsine 0.88 0.77, 0.95 98.28 <0.001 
Gender 
 Female 1,234 729 Freeman-Tukey double arcsine 0.61 0.33, 0.85 98.89 <0.001 
 Male 1,301 731 Freeman-Tukey double arcsine 0.60 0.28, 0.87 99.20 <0.001 
Sampling method 
 Physical examination 450 342 Freeman-Tukey double arcsine 0.76 0.72, 0.80 
 Random sampling 3,135 2,609 Freeman-Tukey double arcsine 0.83 0.72, 0.91 98.04 <0.001 
 Conditional sampling 2,382 1,781 Freeman-Tukey double arcsine 0.71 0.53, 0.87 98.72 <0.001 
Seasonality 
 Season 2,015 1,388 Freeman-Tukey double arcsine 0.67 0.50, 0.82 98.18 <0.001 
 Non-season 177 87 0.492 0.418, 0.565 
 Unknown 3,775 3,257 Freeman-Tukey double arcsine 0.88 0.81, 0.93 96.28 <0.001 
Publication language 
 English 12 5,452 4,310 Freeman-Tukey double arcsine 0.76 0.66, 0.85 98.48 <0.001 
 Chinese 515 422 Freeman-Tukey double arcsine 0.82 0.78, 0.85  
Ethnicity 
 Chinese 3,564 2,758 Freeman-Tukey double arcsine 0.74 0.61, 0.86 98.61 <0.001 
 Others 2,403 1,974 Freeman-Tukey double arcsine 0.79 0.65, 0.91 98.18 <0.001 
Fig. 2.

Forest plots for the seroprevalence of EV-D68 antibody in the overall population.

Fig. 2.

Forest plots for the seroprevalence of EV-D68 antibody in the overall population.

Close modal
Fig. 3.

Forest plots for the seroprevalence of EV-D68 antibody in different age groups.

Fig. 3.

Forest plots for the seroprevalence of EV-D68 antibody in different age groups.

Close modal
Fig. 4.

Forest plots for the seroprevalence of EV-D68 antibody among people in different genders.

Fig. 4.

Forest plots for the seroprevalence of EV-D68 antibody among people in different genders.

Close modal

Sensitivity Analysis and Publication Bias

Meta-regression analysis indicated that seasonality and age markedly influenced the heterogeneity of the meta-analysis results (p < 0.05) (Table 3). The combined rate of seroprevalence was strongly affected by the eighth study, and the combined effect after its removal changed from 0.74 (0.70, 0.82) to 0.79 (0.73, 0.86) (online suppl. Fig. S1). According to the Egger tests, the publication bias plot was drawn. Publication bias is the phenomenon that a paper with a positive result (a study with statistically significant results) is more likely to be accepted and published than a paper with a negative result (a study with no statistically significant results) in a similar study. The Egger method and Begg method were used to evaluate publication bias in the meta-analysis literature. The Egger method uses linear regression to test whether the intercept is zero. The intercept of this study is not zero, suggesting the possibility of the presence of publication bias (Fig. 5).

Table 3.

Results of meta-regression

CovariateCoefficient95% CItp valueAdjusted R2 (%)
Age 
 1–3 years 36.97 
 <1 year 0.0591517 −0.1853366, 0.30364 0.49 0.624  
 <5 years 0.1336648 −0.0804125, 0.3477421 1.28 0.212  
 >5 years 0.3873577 0.166248, 0.6084674 3.58 0.001  
Gender 
 Female −11.00 
 Male −0.0122661 −0.4147161, 0.3901839 −0.07 0.947  
Sampling method 
 Physical examination −16.44 
 Random sampling 0.0533797 −0.4891908, 0.5959503 0.22 0.831  
 Conditional sampling −0.0652324 −0.5957734, 0.4653086 −0.27 0.790  
Seasonality 
 Unknown  32.98 
 Season −0.202109 −0.4381287, 0.0339107 −1.91 0.085  
 Non-season −0.374289 −0.8264419, 0.0778639 −1.84 0.095  
Publication language 
 Chinese −10.74 
 English −0.079054 −0.5785678, 0.4204599 −0.35 0.734  
Ethnicity 
 Others −9.62 
 Chinese −0.0420323 −0.3146522 0.2305875 −0.34 0.741  
CovariateCoefficient95% CItp valueAdjusted R2 (%)
Age 
 1–3 years 36.97 
 <1 year 0.0591517 −0.1853366, 0.30364 0.49 0.624  
 <5 years 0.1336648 −0.0804125, 0.3477421 1.28 0.212  
 >5 years 0.3873577 0.166248, 0.6084674 3.58 0.001  
Gender 
 Female −11.00 
 Male −0.0122661 −0.4147161, 0.3901839 −0.07 0.947  
Sampling method 
 Physical examination −16.44 
 Random sampling 0.0533797 −0.4891908, 0.5959503 0.22 0.831  
 Conditional sampling −0.0652324 −0.5957734, 0.4653086 −0.27 0.790  
Seasonality 
 Unknown  32.98 
 Season −0.202109 −0.4381287, 0.0339107 −1.91 0.085  
 Non-season −0.374289 −0.8264419, 0.0778639 −1.84 0.095  
Publication language 
 Chinese −10.74 
 English −0.079054 −0.5785678, 0.4204599 −0.35 0.734  
Ethnicity 
 Others −9.62 
 Chinese −0.0420323 −0.3146522 0.2305875 −0.34 0.741  

The first line of every covariate represented reference; adjusted R2 was used to indicate the degree of heterogeneity explained by study characteristics.

Fig. 5.

Egger’s publication bias plot for the seroprevalence of EV-D68 antibody among people.

Fig. 5.

Egger’s publication bias plot for the seroprevalence of EV-D68 antibody among people.

Close modal

EV-D68 is a type of enterovirus D that circulates worldwide. EV-D68 infection can cause a variety of symptoms, such as fever, cough, wheezing, and body aches [38]. Numerous studies have shown that EV-D68 infection can cause upper respiratory tract infection as well as AFM, which may even lead to death in severe cases [39, 40]. Recently, the EV-D68 epidemic has posed a substantial threat to public health. Moreover, antibodies, an important effector molecule of the humoural immune response, are very important in the prevention and control of EV-D68 infection. Detailed analysis of age-specific seroprevalence of EV-D68 antibody is able to clarify the risk profile of EV-D68 infection in different age groups while assisting health authorities in determining the implementation of vaccine protocols. Various serological studies have already been performed in many countries. Testing for levels of neutralizing antibodies against EV-D68 is of great importance in assessing the epidemic trend of the virus and taking preventative measures. In this study, the serological results of EV-D68 were summarized to further explore the influencing factors of seropositivity to provide a basis for disease prevention, control, and vaccination decision-making.

This study evaluated 13 published studies, showing that the positive rate of EV-D68 antibody was relatively high, with 74.4% (95% CI: 66.9–82.0%) in the general population and 72.6% (95% CI: 61.6–83.7%) in the Chinese population. In another systematic study, the researchers found that the prevalence of previous EV-D68 infection was 66.3% (95% CI: 40.0–88.2), while the overall prevalence of current EV-D68 infection was 4% in apparently healthy individuals and patients with acute respiratory infection, AFM, and asthma-related illness [21]. Both the prevalence rate and severe disease rates were much lower than the seropositivity rate of EV-D68 in this study, indicating that the virus may spread silently, and serological data may better reflect the regional epidemic situation of the virus. Xiang et al. [25] also found that the earliest EV-D68 infection in China was found in 2006. In subsequent years, more cases were found in different populations and different regions. The statistical results showed that EV-D68 infection was prone to recessive infection, and the actual clinical statistics could not reflect the prevalence of the virus. Serological investigation in the general population can more scientifically assess the prevalence of the virus and the immunity of the population, which is of great significance for public health managers and decision-makers to formulate effective prevention and control measures.

The analysis revealed that the seroprevalence of the EV-D68 antibody gradually increased from 1 to 5 years old, in which the group over 5 years of age and adults were all noted to have a high level of antibodies. In a separate study on maternal and infant antibody protection, Sun et al. [26] recruited 20 pregnant women and tested their maternal and newborn antibody levels, in which the antibody-positive rate of EV-D68 in neonates was shown to decrease from 100% at birth to 10% at 1 year of age, with the lowest geometric mean titre of the antibody noted in 8-month-old infants. In addition, the study tracked a gradual increase in the seropositive rate of EV-D68 among children from 10% at the age of 1 year to 92% at ages 11–15 years [26]. Another study of serum samples collected in the UK in 2006 and 2016 found that the lowest seropositivity rate was found among children aged 0.5–1 years, and the seropositivity rate of EV-D68 among adults was close to 100% [36]. These results suggested that neonates can obtain high titres of EV-D68 antibodies from their mothers, although the neonates’ serum levels gradually decreased as they aged. Another study also demonstrated that the seropositive rate of EV-D68 in neonates was 32%; however, it was only 18% in one-year-old children. After the first year of life, the positive rate gradually increased with age, reaching 100% in the 16- to 49-year-old group [29]. Therefore, children approximately 1 year of age are susceptible to EV-D68 infection. In China, children between the ages of 3 and 6 begin to study alongside each other in kindergarten, which increases the risk of EV-D68 exposure and infection. Children with both dominant and negative infections have shown significant increases in antibody levels. Another study examined the transmission of EV-D68 among children in a kindergarten class in Taipei, Taiwan, between 2006 and 2008 [28]. The study recruited 190 children aged between 2 and 5 years and found that the seroprevalence increased from 19% (25/130) to 67% (83/124) over the course of the study. The high seroconversion rate of preschool children with EV-D68 (73%) suggests that preschool children are vulnerable to EV-D68 infection; hence, effective measures should be taken to prevent it. Furthermore, active preventive measures should be taken in kindergarten classes or nurseries to reduce transmission and avoid severe complications of EV-D68 infection. Lee et al. [29] also found that age, larger family size, and kindergarten/day care attendance were the most important risk factors associated with EV-D68 seropositivity among preschool children. The study found a fairly high seropositive rate of EV-D68 in children under 15 years of age and 100% in adults. No significant difference in the positive rate of EV-D68 in different countries was found in this study, indicating that EV-D68 is widely prevalent globally. There are many individuals with latent infection in the general population, although only a few severe cases have been reported. Although some studies on this virus exist, the epidemiology, serology, and related mechanisms of this virus remain quite limited. Accordingly, a more comprehensive surveillance and management system should be established to improve the prevention and control of EV-D68.

It can be seen from the selection of virus strains that the selection of virus strains varies greatly in different studies (online suppl. Table S3). Among them, one study selected the prototype strain of EV-D68 (Fermon [GenBank accession No. AY426531]), 6 studies selected the clinical isolates, and the remaining 6 studies used both the prototype strain of EV-D68 and the clinical isolates. At the same time, these virus strains belong to different genotypes. The selection of prototype strains is more comparable between different studies, but as the virus circulates and mutates widely, the selection of circulating clinical isolates can more accurately reflect the actual prevalence of the virus. In specific studies, antibody reactivity may vary among virus strains. Karelehto et al. [35] found that EV-D68 neutralizing antibodies have a serological positive rate of 94.4–95.0% against Fermon EV-D68 strains and 44.4–68.4% against clinical isolates in children under 1 year of age. Children (1–10 years) and young adults (11–20 years) had markedly lower geometric mean titres of neutralizing antibodies against the Fermon strain than against the clinical EV-D68 isolate [35]. In another study, four EV-D68 strains were used to test for neutralizing antibodies: Fermon (the prototype strain) and three clinical isolates (14-18949/B1, 14-18952/B2, and 14-18953/A2) [37]. The positive rates of neutralizing antibodies were Fermon (100%), 14-18949/B1 (100%), 18952/B2 (89%), and 14-18953/A2 (97%) in 436 serum samples [37].

At the same time, we refined the seropositive rate of different years in the references and drew the seropositive forest map of different years (online suppl. Fig. S2). In 2014, a large-scale epidemic of EV-D68 broke out in the USA, and at the same time, EV-D68 was widely popular in many countries around the world, causing widespread concern [41]. We used 2014 as the cut-off point to compare seropositivity rates in the years before and after 2014. The results showed that the positive rate of serum samples in 2014 and after 2014 was significantly higher than that before 2014. This result is consistent with the results of several other studies. Chan et al. [34] investigated the seroepidemiology of EV-D68 in Kuala Lumpur, Malaysia, before and after the 2014 US outbreak. The results showed that in all age groups, the positive rate of EV-D68 in serum samples after September 2014 was higher than that before September 2014, and the positive rate was lowest in children aged 1–3 years. Liu et al. [30] conducted a retrospective analysis on the positive rate of EV-D68 in a normal population in Beijing, China, and found that the positive rate in 2014–2016 was higher than that in 2012. A separate study in the Netherlands also showed higher seroprevalence rates in 2015–2016 than in serum samples taken in 2006–2007 [35].

This meta-analysis has certain limitations. First, different laboratories used different strains of the virus during testing, leading to differences in the sensitivity and positive rates. Second, half of the included studies were from China. Therefore, the results of the analysis cannot accurately reflect the seroprevalence of EV-D68 antibody in the whole population. Due to influences in study region, health status of the study population, age range, sampling method, and time, the meta-analysis has great heterogeneity, and the results of meta-regression cannot fully explain the source of heterogeneity. Due to insufficient information on sex, age, and antibody titres provided by certain studies, the data for the subgroup analysis were also insufficient. There were many retrospective and cross-sectional studies in this study, and a number of studies used previously preserved serum samples. Antibody titres may decrease with the passage of storage time, leading to deviations in the determination results. According to the prevalence of the virus, it may be meaningful to conduct serological analyses in different years. However, due to the limited sample size, this study did not conduct further detailed analysis in years.

Overall, this study showed that EV-D68 is widely prevalent worldwide and that the seroprevalence of EV-D68 antibody was related to age but not to sex or region. The seropositivity rate was lower in the below 5-year age group; however, it gradually increased with age, and children approximately 1 year of age were susceptible to EV-D68 infection.

The authors acknowledge all studies involved in the meta-analysis for providing original data.

An ethics statement is not applicable because this study is based exclusively on published literature.

The authors declare no conflicts of interest relevant to this study.

This work was supported by Natural Science Foundation of Hubei Province (2018CFB254) and Wuhan COVID-19 Emergency Research Project (EX20D04).

Yingying Shi and Lu Shi designed the project. Yingying Shi, Qinqin Ran, and Xiaochen Wang performed statistical analysis. Yingying Shi, Qinqin Ran, Xiaochen Wang, and Lu Shi interpreted and wrote the manuscript and did systematic search and literature review. All authors approved the final manuscript.

All data generated or analysed during this study are included in this article and its online supplementary material files. Further enquiries can be directed to the corresponding author.

1.
Elrick
MJ
,
Pekosz
A
,
Duggal
P
.
Enterovirus d68 molecular and cellular biology and pathogenesis
.
J Biol Chem
.
2021
;
296
:
296100317
.
2.
Foster
CB
,
Coelho
R
,
Brown
PM
,
Wadhwa
A
,
Dossul
A
,
Gonzalez
BE
.
A comparison of hospitalized children with enterovirus d68 to those with rhinovirus
.
Pediatr Pulmonol
.
2017
;
52
(
6
):
827
32
.
3.
Reiche
J
,
Bottcher
S
,
Diedrich
S
,
Buchholz
U
,
Buda
S
,
Haas
W
.
Low-level circulation of enterovirus d68-associated acute respiratory infections, Germany, 2014
.
Emerg Infect Dis
.
2015
;
21
(
5
):
837
41
.
4.
Kreuter
JD
,
Barnes
A
,
McCarthy
JE
,
Schwartzman
JD
,
Oberste
MS
,
Rhodes
CH
.
A fatal central nervous system enterovirus 68 infection
.
Arch Pathol Lab Med
.
2011
;
135
(
6
):
793
6
.
5.
Nguyen-Tran
H
,
Park
SW
,
Messacar
K
,
Dominguez
SR
,
Vogt
MR
,
Permar
S
.
Enterovirus d68: a test case for the use of immunological surveillance to develop tools to mitigate the pandemic potential of emerging pathogens
.
Lancet Microbe
.
2022
;
3
(
2
):
e83
85
.
6.
Bigi
S
,
Ramette
A
,
Barbani
MT
,
Bieri
A
,
Hoffmann
A
,
Aebi
C
.
Acute flaccid myelitis in Switzerland: association with enterovirus d68
.
Swiss Med Wkly
.
2023
;
153
:
15340045
.
7.
Gong
L
,
Wang
Y
,
Zhang
W
,
Chen
C
,
Yang
X
,
Xu
L
.
Acute flaccid myelitis in children in zhejiang province, China
.
Front Neurol
.
2020
;
11
:
11360
.
8.
Maloney
JA
,
Mirsky
DM
,
Messacar
K
,
Dominguez
SR
,
Schreiner
T
,
Stence
NV
.
Mri findings in children with acute flaccid paralysis and cranial nerve dysfunction occurring during the 2014 enterovirus d68 outbreak
.
AJNR Am J Neuroradiol
.
2015
;
36
(
2
):
245
50
.
9.
Lang
M
,
Mirand
A
,
Savy
N
,
Henquell
C
,
Maridet
S
,
Perignon
R
.
Acute flaccid paralysis following enterovirus d68 associated pneumonia, France, 2014
.
Euro Surveill
.
2014
;
19
(
44
):
20952
.
10.
Chong
PF
,
Kira
R
,
Mori
H
,
Okumura
A
,
Torisu
H
,
Yasumoto
S
.
Clinical features of acute flaccid myelitis temporally associated with an enterovirus d68 outbreak: results of a nationwide survey of acute flaccid paralysis in Japan, august-december 2015
.
Clin Infect Dis
.
2018
;
66
(
5
):
653
64
.
11.
Hixon
AM
,
Frost
J
,
Rudy
MJ
,
Messacar
K
,
Clarke
P
,
Tyler
KL
.
Understanding enterovirus d68-induced neurologic disease: a basic science review
.
Viruses
.
2019
;
11
(
9
):
821
.
12.
Holm-Hansen
CC
,
Midgley
SE
,
Fischer
TK
.
Global emergence of enterovirus d68: a systematic review
.
Lancet Infect Dis
.
2016
;
16
(
5
):
e64
75
.
13.
Midgley
SE
,
Benschop
K
,
Dyrdak
R
,
Mirand
A
,
Bailly
JL
,
Bierbaum
S
.
Co-circulation of multiple enterovirus d68 subclades, including a novel b3 cluster, across europe in a season of expected low prevalence, 2019/20
.
Euro Surveill
.
2020
;
25
(
2
):
1900749
.
14.
Xiang
Z
,
Wang
J
.
Enterovirus d68 and human respiratory infections
.
Semin Respir Crit Care Med
.
2016
;
37
(
4
):
578
85
.
15.
Mehand
MS
,
Al-Shorbaji
F
,
Millett
P
,
Murgue
B
.
The who r&d blueprint: 2018 review of emerging infectious diseases requiring urgent research and development efforts
.
Antiviral Res
.
2018
;
159
:
15963
7
.
16.
Levy
A
,
Roberts
J
,
Lang
J
,
Tempone
S
,
Kesson
A
,
Dofai
A
.
Enterovirus d68 disease and molecular epidemiology in Australia
.
J Clin Virol
.
2015
;
69
:
69117
121
.
17.
Peci
A
,
Winter
AL
,
Warshawsky
B
,
Booth
TF
,
Eshaghi
A
,
Li
A
.
Epidemiology of enterovirus d68 in ontario
.
PLoS One
.
2015
;
10
(
11
):
e0142841
.
18.
Srinivasan
M
,
Niesen
A
,
Storch
GA
.
Enterovirus d68 surveillance, st. Louis, Missouri, USA, 2016
.
Emerg Infect Dis
.
2018
;
24
(
11
):
2115
7
.
19.
Esposito
S
,
Zampiero
A
,
Ruggiero
L
,
Madini
B
,
Niesters
H
,
Principi
N
.
Enterovirus d68-associated community-acquired pneumonia in children living in milan, Italy
.
J Clin Virol
.
2015
;
68
:
6894
6
.
20.
Vogt
MR
,
Wright
PF
,
Hickey
WF
,
De Buysscher
T
,
Boyd
KL
,
Crowe
JE
Jr
.
Enterovirus d68 in the anterior horn cells of a child with acute flaccid myelitis
.
N Engl J Med
.
2022
;
386
(
21
):
2059
60
.
21.
Fall
A
,
Kenmoe
S
,
Ebogo-Belobo
JT
,
Mbaga
DS
,
Bowo-Ngandji
A
,
Foe-Essomba
JR
.
Global prevalence and case fatality rate of enterovirus d68 infections, a systematic review and meta-analysis
.
PLoS Negl Trop Dis
.
2022
;
16
(
2
):
e0010073
.
22.
Vogt
MR
,
Crowe
JE
Jr
.
Current understanding of humoral immunity to enterovirus d68
.
J Pediatr Infect Dis Soc
.
2018
7
suppl_2
S49
53
.
23.
Shi
Y
,
Chen
P
,
Bai
Y
,
Xu
X
,
Liu
Y
.
Seroprevalence of coxsackievirus a6 and enterovirus a71 infection in humans: a systematic review and meta-analysis
.
Arch Virol
.
2023
;
168
(
2
):
37
.
24.
Barendregt
JJ
,
Doi
SA
,
Lee
YY
,
Norman
RE
,
Vos
T
.
Meta-analysis of prevalence
.
J Epidemiol Community Health
.
2013
;
67
(
11
):
974
8
.
25.
Xiang
Z
,
Li
L
,
Ren
L
,
Guo
L
,
Xie
Z
,
Liu
C
.
Seroepidemiology of enterovirus d68 infection in China
.
Emerg Microbes Infect
.
2017
;
6
(
5
):
e32
.
26.
Sun
S
,
Gao
F
,
Hu
Y
,
Bian
L
,
Wu
X
,
Su
Y
.
A cross-sectional seroepidemiology study of ev-d68 in China
.
Emerg Microbes Infect
.
2018
;
7
(
1
):
99
.
27.
Sun
SY
,
Gao
F
,
Hu
YL
,
Bian
LL
,
Mao
QY
,
Wu
X
.
Seroepidemiology of enterovirus d68 infection in infants and children in jiangsu, China
.
J Infect
.
2018
;
76
(
6
):
563
9
.
28.
Hu
YL
,
Huang
LM
,
Lu
CY
,
Fang
TY
,
Cheng
AL
,
Chang
LY
.
Manifestations of enterovirus d68 and high seroconversion among children attending a kindergarten
.
J Microbiol Immunol Infect
.
2019
;
52
(
6
):
858
64
.
29.
Lee
JT
,
Shih
WL
,
Yen
TY
,
Cheng
AL
,
Lu
CY
,
Chang
LY
.
Enterovirus d68 seroepidemiology in taiwan, a cross sectional study from 2017
.
PLoS One
.
2020
;
15
(
3
):
e0230180
.
30.
Liu
Y
,
Gong
C
,
Luo
M
,
Zhang
T
,
Li
M
,
Shen
L
.
Seroepidemiology of enterovirus d68 in a healthy population in beijing, China, between 2012 and 2017: a retrospective study
.
J Med Virol
.
2021
;
93
(
6
):
3524
31
.
31.
Yu
L
,
Rui
Z
,
Yuanyuan
W
,
Longfa
X
,
Zhichao
Y
,
Hongwei
Y
.
Seroepidemiological survey of enterovirus d68 among healthy population in xiamen city, China
.
Chin J Virol
.
2021
;
93
(
6
):
3524
31
.
32.
Metoki
T
,
Okamoto
M
,
Suzuki
A
,
Kitaoka
S
,
Miyabayashi
H
,
Rokugo
Y
.
Concurrent community transmission of enterovirus d68 with human rhinoviruses and respiratory syncytial virus among children in sendai, Japan
.
Pediatr Infect Dis J
.
2018
;
37
(
5
):
394
400
.
33.
Kadji
F
,
Nishimura
H
,
Okamoto
M
,
Sato
K
,
Ohmiya
S
,
Ito
H
.
Fluctuations in antibody titers against enterovirus d68 in pediatric sera collected in a community before, during, and after a possible outbreak
.
Jpn J Infect Dis
.
2020
;
73
(
1
):
55
7
.
34.
Chan
YF
,
Sam
IC
,
Nayan
E
,
Tan
XH
,
Yogarajah
T
.
Seroepidemiology of enterovirus d68 infection in kuala lumpur, Malaysia between 2013 and 2015
.
J Med Virol
.
2021
;
94
(
6
):
2607
12
.
35.
Karelehto
E
,
Koen
G
,
Benschop
K
,
van der Klis
F
,
Pajkrt
D
,
Wolthers
K
.
Enterovirus d68 serosurvey: evidence for endemic circulation in The Netherlands, 2006 to 2016
.
Euro Surveill
.
2019
;
24
(
35
):
1800671
.
36.
Kamau
E
,
Harvala
H
,
Blomqvist
S
,
Nguyen
D
,
Horby
P
,
Pebody
R
.
Increase in enterovirus d68 infections in young children, United Kingdom, 2006-2016
.
Emerg Infect Dis
.
2019
;
25
(
6
):
1200
3
.
37.
Harrison
CJ
,
Weldon
WC
,
Pahud
BA
,
Jackson
MA
,
Oberste
MS
,
Selvarangan
R
.
Neutralizing antibody against enterovirus d68 in children and adults before 2014 outbreak, Kansas city, Missouri, USA(1)
.
Emerg Infect Dis
.
2019
;
25
(
3
):
585
8
.
38.
Schuster
JE
,
Selvarangan
R
,
Hassan
F
,
Briggs
KB
,
Hays
L
,
Miller
JO
.
Clinical course of enterovirus d68 in hospitalized children
.
Pediatr Infect Dis J
.
2017
;
36
(
3
):
290
5
.
39.
Pfeiffer
HC
,
Bragstad
K
,
Skram
MK
,
Dahl
H
,
Knudsen
PK
,
Chawla
MS
.
Two cases of acute severe flaccid myelitis associated with enterovirus d68 infection in children, Norway, autumn 2014
.
Euro Surveill
.
2015
;
20
(
10
):
21062
.
40.
Hellferscee
O
,
Treurnicht
FK
,
Tempia
S
,
Variava
E
,
Dawood
H
,
Kahn
K
.
Enterovirus d68 and other enterovirus serotypes identified in south african patients with severe acute respiratory illness, 2009-2011
.
Influenza Other Respir Viruses
.
2017
;
11
(
3
):
211
9
.
41.
Tan
Y
,
Hassan
F
,
Schuster
JE
,
Simenauer
A
,
Selvarangan
R
,
Halpin
RA
.
Molecular evolution and intraclade recombination of enterovirus d68 during the 2014 outbreak in the United States
.
J Virol
.
2016
;
90
(
4
):
1997
2007
.