Advertisement

A prospective, comparative study of severe neurological and uncomplicated hand, foot and mouth forms of paediatric enterovirus 71 infections

Open AccessPublished:April 21, 2017DOI:https://doi.org/10.1016/j.ijid.2017.04.005

      Abstract

      Objectives

      In this study, we document the clinical characteristics and investigated risk factors for uncomplicated and severe forms of EV-A71 disease in Cambodian children.

      Methods

      From March to July 2014 inclusive, all patients with suspicion of EV-A71 infection presenting to Kantha Bopha Hospitals in Phnom Penh and Siem Reap and confirmed by the Virology Unit at the Institut Pasteur du Cambodge were prospectively enrolled in this study. Throat swabs, rectal swabs and serum samples were collected from all consecutive patients with suspected EV-A71 infection. In addition, CSF was also collected from patients with suspected EV-A71 associated encephalitis. A total of 122 patients (29 with uncomplicated disease and 93 with severe disease) with confirmed EV-A71 infection with all available demographic and clinical data for clinical classification and further analysis were included in the study.

      Results

      In this prospective EV-A71 study in Cambodia, we confirmed the previously reported association of male gender and absence of mouth or skin lesions with severe disease. We also highlighted the strong association of neutrophils in blood, but also in CSF in patients with pulmonary oedema. More importantly, we identified new putative nutrition-related risk factors for severe disease.

      Conclusions

      EV-A71 is an important cause of encephalitis in the Asia-Pacific region. Further studies to determine the risk factors associated with severe EV-A71 disease are needed.

      Keywords

      Introduction

      Enterovirus 71 (EV-A71) is a genotype within the species Enterovirus A, genus Enterovirus, family picornaviridae. The virus is one of the most common aetiologies of hand, foot and mouth disease (HFMD) in children. EV-A71 associated disease is usually mild with children typically recovering within 4-6 days. However, the virus has been associated with fulminant disease during large outbreaks in many parts of the World, including Bulgaria (
      • Shindarov L.M.
      • Chumakov M.P.
      • Voroshilova M.K.
      • Bojinov S.
      • Vasilenko S.M.
      • Iordanov I.
      • et al.
      Epidemiological, clinical, and pathomorphological characteristics of epidemic poliomyelitis-like disease caused by enterovirus 71.
      ), Hungary (
      • Nagy G.
      • Takátsy S.
      • Kukán E.
      • Mihály I.
      • Dömök I.
      Virological diagnosis of enterovirus type 71 infections: experiences gained during an epidemic of acute CNS diseases in Hungary in 1978.
      ), Malaysia (
      • Chan L.G.
      • Parashar U.D.
      • Lye M.S.
      • Ong F.G.
      • Zaki S.R.
      • Alexander J.P.
      • et al.
      Deaths of children during an outbreak of hand, foot, and mouth disease in Sarawak, Malaysia: clinical and pathological characteristics of the disease.
      ), Taiwan (
      • Huang C.-C.
      • Liu C.-C.
      • Chang Y.-C.
      • Chen C.Y.
      • Wang S.T.
      • Yeh T.F.
      Neurologic complications in children with enterovirus 71 infection.
      ), Singapore (
      • Chong C.-Y.
      • Chan K.-P.
      • Shah V.A.
      • Ng W.Y.M.
      • Lau G.
      • Teo T.E.-S.
      • et al.
      Hand, foot and mouth disease in Singapore: a comparison of fatal and non-fatal cases.
      ), and China (
      • Li J.
      • Chen F.
      • Liu T.
      • Wang L.
      MRI findings of neurological complications in hand-foot-mouth disease by enterovirus 71 infection.
      ,
      • Liu N.
      • Xie J.
      • Qiu X.
      • Jia L.
      • Wu Z.
      • Ma Y.
      • et al.
      An atypical winter outbreak of hand, foot, and mouth disease associated with human enterovirus 71, 2010.
      ,
      • Huang W.-C.
      • Shih W.-L.
      • Yang S.-C.
      • Yen T.Y.
      • Lee J.T.
      • Huang Y.C.
      • et al.
      Predicting severe enterovirus 71 infection: Age, comorbidity, and parental behavior matter.
      ). A meta-analysis has estimated case-fatality rates for hospitalized cases of HFMD associated with EV-A71 at 1.7% (
      • Zhao Y.Y.
      • Jin H.
      • Zhang X.F.
      • Wang B.
      Case-fatality of hand, foot and mouth disease associated with EV-A71: a systematic review and meta-analysis.
      ).
      In 2012, an outbreak of EV-A71 in Cambodia associated with cardiovascular collapse and pulmonary oedema gained international attention following the deaths of nearly 100 young children over a short period (
      • Duong V.
      • Mey C.
      • Eloit M.
      • Zhu H.
      • Danet L.
      • Huang Z.
      • et al.
      Molecular epidemiology of human enterovirus 71 at the origin of an epidemic of fatal hand, foot and mouth disease cases in Cambodia.
      ). Although this was the first EV-A71 epidemic reported in Cambodia, retrospective seroepidemiological testing showed that there was widespread circulation of the virus for at least a decade prior to the outbreak (
      • Horwood P.F.
      • Andronico A.
      • Tarantola A.
      • Salje H.
      • Duong V.
      • Mey C.
      • et al.
      Seroepidemiology of human enterovirus 71 infection among Cambodian children.
      ). The pathogenetic and epidemiological mechanisms of how this virus intermittently causes large deadly outbreaks remains unclear.
      We report factors associated with clinical characteristics of 122 Cambodian pediatric patients during an EV-A71 outbreak in 2014. We focus on environmental, clinical and biological risk factors for severe neurological and/or pulmonary disease associated with confirmed EV-A71 infection.

      Methods

      Clinical methods

      During an EV-A71 disease outbreak in Cambodia from March to July 2014 inclusive, all patients with confirmed EV-A71 infection presenting at either of two Kantha Bopha Hospitals (one in Cambodia’s southeastern capital Phnom Penh and the other in Siem Reap in the north-west of the country) were enrolled in the study. The clinical definitions used throughout the study are listed in Table 1. Patients with uncomplicated EV-A71 HFMD were compared to patients with severe forms of infection, including isolated encephalitis without pulmonary oedema (ECP) and pulmonary oedema with or without neurological involvement (PO). Demographic and clinical data were recorded for further analysis on a case report form (CRF) that was specifically designed for the study. Clinical outcome was not available for the study and patients were not followed-up.
      Table 1Clinical definitions used in this study.
      • Suspected EV-A71 infection was defined as an acute illness with either systemic (e.g., fever), respiratory, neurological or skin signs and symptoms suggestive of viral infection.
      • Confirmed EV-A71 infection was defined as a suspected EV-A71 infection plus the isolation of EV-A71 virus or molecular detection of EV-A71 RNA in a rectal swab, throat swab, serum or cerebrospinal fluid sample.
      • Uncomplicated Hand Foot Mouth Disease patients (HFMD) were defined as patients with a confirmed viral infection presenting with the classical signs and symptoms of HFMD and without neurological symptoms.
      • Skin lesions were defined as vesiculo-papular or maculo-papular rash (mostly present over the hands, soles and/or buttocks).
      • Mouth lesions were defined as oral ulceration usually observed on anterior tonsillar pillars, soft palate, buccal mucosa, or uvula.
      • Neurological symptoms were defined as Glasgow Coma Scale (GCS) score (or modified GCS in less than 2 years old) less than 14 or clinical rhombencephalitis or limb weakness.
      • Neurological involvement was defined as neurological symptoms and/or CSF white cell count >5/mm3 and/or abnormal brain imaging (computed tomography scan or magnetic resonance imaging) suggestive of encephalitis.
      • Pulmonary oedema was defined as respiratory symptoms and bilateral alveolar congestion on chest radiography.
      • Encephalitis patients (ECP) were defined as patients with neurological involvement without pulmonary oedema.
      • Pulmonary oedema patients (PO) were patients with pulmonary oedema with or without neurological involvement.
      • ECP and PO were termed severe patients.

      Virological methods

      Throat swabs, rectal swabs and serum samples were collected from all patients presenting to Kantha Bopha Hospitals with suspected EV-A71 infection. In addition, cerebrospinal fluid (CSF) was also collected from patients with suspected EV-A71 associated encephalitis. Nucleic acids were extracted from all samples using the QIAamp Viral RNA Minikit (Qiagen, Hilden, Germany), as outlined in the manufacturer’s instructions. EV-A71 infections were detected by testing all clinical samples using an EV-A71 specific qRT-PCR assay (
      • Khanh T.H.
      • Sabanathan S.
      • Thanh T.T.
      • Thoa le P.K.
      • Thuong T.C.
      • Hang V.t.
      • et al.
      Enterovirus 71-associated hand, foot, and mouth disease, Southern Vietnam, 2011.
      ) or culture isolation of EV-A71 from original clinical material in Vero E6 cells. EV-A71 infection was confirmed if PCR or culture was positive in any of the tested samples. Severe EV-A71 cases with neurological symptoms were confirmed by the detection of EV-A71 by qRT-PCR or culture from CSF or blood; and/or detection of EV-A71 in both throat and rectal swabs, as previously recommended (
      • Jain S.
      • Patel B.
      • Bhatt G.C.
      Enteroviral encephalitis in children: clinical features, pathophysiology, and treatment advances.
      ). Laboratory test results were integrated in the light of clinical findings to provide diagnosis (Figure 1).
      Figure 1
      Figure 1Flow chart showing enrolment into the study.

      EV-A71 phylogenetic analysis

      EV-A71 isolates were randomly selected from throughout the outbreak period for sequence analysis of the VP-1 region. Amplicons (∼500 bp) of the VP-1 region were produced using RT-PCR (
      • Nix W.A.
      • Oberste M.S.
      • Pallansch M.A.
      Sensitive, seminested PCR amplification of VP1 sequences for direct identification of all enterovirus serotypes from original clinical specimens.
      ) for EV-A71 isolates associated with uncomplicated HFMD (n = 6) and severe EV-A71 infection (n = 5). The amplicons were sequenced at a commercial facility (Macrogen, South Korea) by Sanger method. Contiguous sequences were assembled using CLC Workbench (CLC bio) and compared to representative EV-A71 sequences downloaded from the NCBI GenBank database. Neighbour-joining trees were constructed with MEGA5 (
      • Tamura K.
      • Peterson D.
      • Peterson N.
      • Stecher G.
      • Nei M.
      • Kumar S.
      MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods.
      ) and bootstrap values were calculated and expressed as a percentage from 1,000 replicates.

      Data collection and analysis

      Data was collected at the bedside using a specifically designed CRF that was focused on socio-demographic variables as well as general and neurological clinical findings. Demographic and clinical data were collected for age, sex, hospital of inclusion, province, fever, peak body temperature, and neurological and extra neurological symptoms and signs at presentation. While in hospital, bodyweight, heart rate, blood pressure, body temperature, mental status, general and neurological examination and treatment were assessed. Weight-for-age z scores were computed for all patients and compared with the World Health Organization child growth standards (www.who.int/childgrowth).
      Laboratory data including blood-cell count and differentiated white-blood-cell count, haemoglobin, platelet count, creatinin, liver enzymes, and blood glucose were collected from all patients. Cell count, differentiated cell count, glucose, and protein were also collected from the cerebrospinal fluid of patients with neurological symptoms. Chest radiography, brain computed tomography, and brain magnetic resonance imaging scans were reviewed by radiologists and recorded. Data was entered using standardized paper forms and then exported into an Excel® spreadsheet.

      Statistical analysis

      Data was analysed with the SPSS statistical package (Version 22; IBM Corp. Armonk, NY: USA). The Student’s t-test for continuous variables and Mann-Whitney test and Chi 2 test were used for categorical data. Fisher’s exact test was used adjunctively if the expected values were less than 5. Bonferroni correction was used as a correction for multiple comparison analysis. Binary logistic regression analysis was used to examine the multivariate-adjusted odds ratios for risk factors that were significant (p value of less than 0.05) in the univariate analysis.
      Factors significantly associated with ECP or PO patients in the univariate analysis were included in the multivariate analysis. Forward stepwise regression was used for identifying the optimum variable for the model. A p value of less than 0.05 was considered statistically significant.

      Ethics

      Ethical clearance was obtained from the Cambodian National Ethics Committee for Human Research before testing commenced. All parents/guardians of sick children who participated in this study provided written informed consent; and clinical CRFs were anonymized to protect the identity of patients.

      Results

      Patient characteristics

      During the five-month period, 148 patients with laboratory confirmed EV-A71 infection were enrolled. Of these 117 (79.1%) patients were recruited in Phnom Penh and 31 (20.9%) were recruited in Siem Reap hospital sites. The recruited patients originated from 17 out of 25 of Cambodia’s provinces, mainly from the three most populated provinces in Cambodia: Phnom Penh (25.9%), Kandal (14.2%) and Kampong Cham (12.2%).
      The mean age at onset of disease was 28.6 months (range 5 months to 15 years; median 20 months). The majority (n = 133; 90%) of the patients included in the study were less than five years old. The male:female ratio was 1.5:1. Patients had a low median weight for age adjusted z score of −1.11 (range −4.05 to 2.40; mean −0.94; IQR 2.39) (Table 2), compared with the World Health Organization child growth standards.
      Table 2Statistical analyses of the demographic and clinical characteristics associated with uncomplicated and severe cases of enterovirus 71 infections.
      CharacteristicsHFMD n = 29N
      N: the number of patients with available data.
      Encephalitis (ECP) n = 35N
      N: the number of patients with available data.
      Pulmonary oedema (PO) n = 58N
      N: the number of patients with available data.
      p value HFMD vs ECPp value ECP vs PO
      Demographic
       Male gender17 (58.6)10 (28.6)18 (31)0.0150.802
       Age, months21 [12–72]2920 [5–168]3519 [5–96]580.9620.221
       Province11 (37.9)5 (14.3)10 (17.2)0.0030.052
        Phnom Penh7 (24.1)2 (5.7)9 (15.5)
        Kandal0 (0)6 (17.1)1 (1.7)
        Battambang11 (37.9)22 (62.9)37 (63.8)
        Other78.6 [5.8–231]29108 [2.90–302]3586 [2.9–315]570.0220.734
       Mean distance to hospital, Km2 [0–3]183 [1–7]283 [0–31]490.0060.690
      History of disease
       Fever23 (79.3)34 (97.1)58 (100)0.0230.196
       Duration of fever2 [2–4]223 [1–16]333 [1–10]530.0100.329
       Cough4 (13.8)15 (42.9)32 (55.2)0.0110.25
       Sputum0 (0)2 (5.7)14 (24.1)0.1910.023
       Dyspnea2 (6.9)8 (22.9)46 (79.3)0.080<0.001
       Vomiting3 (10.3)21 (60)44 (75.9)<0.0010.106
       Abdominal pain0 (0)4 (11.4)10 (17.2)0.0600.448
       Diarrhea0 (0)6 (17.1)10 (17.2)0.0190.99
       Mouth ulcer26 (89.7)11 (31.4)15 (25.9)<0.0010.562
       Duration of mouth lesions2 [1–3]263 [2–5]112 [1–4]110.0430.401
       Skin lesion28 (96.6)16 (45.7)13 (22.4)<0.0010.019
       Duration of skin lesion2 [1–3]283 [1–6]162.5 [2–4]100.0280.623
       Headache1 (3.4)10 (28.6)13 (22.4)0.0080.505
       Neck stiffness0 (0)3 (8.6)3 (5.2)0.1060.518
       Limb weakness0 (0)8 (22.9)9 (15.5)0.0060.375
       Seizure0 (0)7 (20)34 (58.6)0.011<0.001
       Coma0 (0)4 (11.4)16 (27.6)0.0600.066
      Examination findings
       Weight, Kg11 [7.5–21]299.6 [7–30]3410 [6–20]580.2840.390
       Weight Age adjusted z score−0.5 [−2.87–1.58]28−1.51 [−3.25–0.81]29−1.26 [−4.05–1.96]530.0040.597
       Temperature, °C37 [36.5–40]2938.3 [36.5–40]3438.3 [35–40]580.0070.828
       Pulse, bpm120 [108–180]29124 [100–160]32140 [60–190]440.154<0.001
       Respiratory rate per min40 [32–70]2940 [30–60]3160 [24–120]430.400<0.001
       Glasgow Coma Scale15 [14–15]2915 [3–15]3211 [3–15]55<0.0010.015
       Oral lesions27 (93.1)10 (28.6)14 (24.1)<0.0010.636
       Skin lesions28 (96.6)14 (40)14 (24.1)<0.0010.106
       Abnormal chest examination27 (93.1)30 (90.9)16 (28.6)0.752<0.001
       Enlarged liver0 (0)5 (14.3)15 (25.9)0.0580.188
       Brainstem clinical involvement0 (0)6 (19.4)8 (14.8)0.0130.587
       Paraplegia0 (0)1 (2.9)7 (12.1)0.3590.125
      Investigation findings
       Haemoglobin, g/dL11.8 [9.8–13.2]1610.5 [7.9–12.9]3510.6 [6.4–19.1]550.0040.497
       Mean corpuscular volume, fl71 [51–80]1766 [55–83]3566.5 [42–83]540.0360.762
       White cells count, ×109 cells/L11.7 [7.1–22.1]1817.4 [6.7–46.8]3519.6 [7.1–54.3]560.0010.029
       Neutrophils count, ×109 cells/L5.36 [3.08–13.48]179.51 [0.47–28.22]3512.22 [0–35.84]550.0030.028
       Lymphocytes count, ×109 cells/L4.75 [2.62–8.0]174.18 [1.51–16.38]355.06 [0.5–23.80]550.6890.211
       Platelets count, ×109 cells/L336 [162–575]18482 [67–743]35468 [86–874]550.0120.970
       Creatinine, μmol/L38.5 [13–97]1835 [3.5–75]3538.5 [1–105]520.1040.129
       AST, IU/L37 [25–93]1843 [11.4–175]3452 [11–365]530.9230.095
       ALT, IU/L18.5 [15–74]1826.5 [11–119]3422 [10–549]540.0460.574
       Blood glucose, mmol/L4.9 [1.3–16]185 [2–14.5]329.1 [1.1–88]530.2620.000
       CSF white cells count, ×109 cells/L4 [4–4]133 [2–549]3561 [2–927]420.3330.539
       CSF lymphocytes, %NA056.5 [32–71]2039 [30–68]22NA0.023
       CSF neutrophils, %NA044 [29–68]2060.5 [32–70]22NA0.037
       CSF proteins, g/LNA00.31 [0.12–0.75]350.32 [0.1–14]40NA0.778
       CSF glucose, mmol/L4.4 [4.4–4.4]14 [2–35.2]355.3 [1.75–98]420.722<0.001
       Abnormal chest X Ray1 (5.6)2 (5.9)58 (100)0.962<0.001
       Abnormal brain CT scan0 (0)1 (100)5 (83.3)NA0.659
       Abnormal brain MRI0 (0)20 (100)14 (93.3)NA0.241
       MRI brainstem involvement0 (0)8 (22.9)5 (8.6)0.0060.069
      The round brackets contain percentage; the square brackets contain vales range.
      NA: not available.
      a N: the number of patients with available data.
      A total of 122 (82%) patients were enrolled in the study with all available data for clinical classification and further analysis. The median time to hospitalization for these children was two days (range from 0 to 31 days; mean 3.2 days). Twenty-nine (19.6%) patients were classified as uncomplicated HFMD patients and 93 patients were classified with severe EV-A71 disease, including 35 (23.6%) who had ECP and 58 (39.2%) with PO (with or without neurological involvement in 52 and 6 cases respectively).
      Overall, 84 (56.8%) patients had either mouth or skin lesions on admission examination. Typical clinical course began with cough and fever 3.2 ± 1.4 days on average before admission. Skin and mouth lesions then appeared 2.4 ± 0.82 days before admission, and in severe cases dyspnea and vomiting would begin 2.2 ± 1.23 days before admission, followed by coma and convulsions 1.6 ± 0.99 days before admission. Lumbar puncture was performed in all ECP patients, in 48 (82.8%) PO patients and in 1 (3.4%) HFMD patient. Chest X-ray was available for 117 (79.1%) patients. Seven (4.7%) patients underwent a brain CT scan and 36 (24.4%) patients had a brain MRI.
      Overall, rectal swabs had the highest detection rate of EV-A71 by PCR or culture (100/122; 82.0%), followed by throat swabs (95/122; 77.9%), serum samples (59/122; 48.4%) and CSF samples (26/93; 28.0%) (Table 3).
      Table 3Virological detection rates of enterovirus 71 from different clinical samples.
      DetectionMild diseaseSevere disease
      HFMD (n = 29)ECP (n = 35)PO (n = 58)Total (n = 93)
      CSF PCRNA7 (20.0%)17 (29.3%)24 (25.8%)
      CSF cultureNA5 (14.3%)3 (5.2%)8 (8.6%)
      CSF total
      total values are the number of samples positive by PCR OR culture.
      0/1 (0%)8 (22.9%)18 (31.0%)26 (28.0%)
      Serum PCR15 (51.7%)14 (40.0%)28 (48.3%)42 (45.2%)
      Serum culture1 (3.4%)3 (8.6%)1 (1.7%)4 (4.3%)
      Serum total
      total values are the number of samples positive by PCR OR culture.
      15 (51.7%)16 (45.7%)28 (48.3%)44 (47.3%)
      Throat PCR25 (86.2%)25 (71.4%)45 (77.6%)70 (75.3%)
      Throat culture10 (34.5%)7 (20.0%)20 (34.5%)27 (29.0%)
      Throat total
      total values are the number of samples positive by PCR OR culture.
      25 (86.2%)25 (71.4%)45 (77.6%)70 (75.3%)
      Rectal PCR27 (93.1%)27 (77.1%)45 (77.6%)72 (77.4%)
      Rectal culture11 (37.9%)7 (20.0%)15 (25.9%)22 (23.7%)
      Rectal total
      total values are the number of samples positive by PCR OR culture.
      27 (93.1%)27 (77.1%)46 (79.3%)73 (78.5%)
      a total values are the number of samples positive by PCR OR culture.

      Comparison of HFMD and ECP patients

      By definition, ECP patients had a higher rate of neurological signs and symptoms (headache, limb weakness and convulsion), abnormal neurological examination (lower mean GCS and higher rate of brainstem clinical involvement) and MRI brainstem involvement. In univariate analysis, compared to HFMD cases, ECP patients were significantly more often male, and more often resided outside Phnom Penh and had a longer median distance from home to the hospital. ECP patients also had significantly lower weight-for-age adjusted z score (−1.51 [−3.25–0.81] vs −0.5 [−2.87–1.58]; p = 0.004) (Table 2).
      ECP patients had a longer prodromal phase before admission, including a higher percentage of fever, cough and digestive prodromes (diarrhea or vomiting). They also had higher median fever on examination. They had significantly lower rate of mouth or skin lesions and, when they appeared, a longer duration of skin or mouth lesions. ECP patients had lower haemoglobin titers, lower mean corpuscular volume, higher neutrophil and platelets counts, and ALT value.

      Comparison of ECP and PO patients

      By definition, PO patients had higher rate of sputum, dyspnea, abnormal chest examination or chest X-Ray and higher respiratory and pulse rates compared to ECP patients (Table 2). Blood pressure at admission, however, did not differ within groups. Compared to ECP patients, PO patients were less likely to have skin lesions and had a higher frequency of neurological signs (lower GCS score and higher frequency of convulsion). They had higher white cells count, especially neutrophil count and higher median blood glucose level (9.1 [1.1–88] mmol/L vs 5 [2–14.5] mmol/L, p = 0.000). PO patients were also more likely to have elevated PNN counts in CSF. There was no significant difference in clinical or MRI brainstem involvement between PO patients with neurological involvement and ECP patients.

      Multivariate analysis

      In logistic regression, PNN count (OR = 2.57 CI95[1.25–5.31], p = 0.011), and haemoglobin titers (OR = 0.181, CI95[0.037–0.881], p = 0.034) were significantly associated with ECP. Time to hospitalization didn’t reach statistical significance (OR = 3.065, CI95[0.846–11.11], p = 0.088). Blood PNN count (OR = 2.57 CI95[1.25–5.31], p = 0.011) and CSF neutrophil differential count (OR = 1.079, CI95[1.011–1.151], p = 0.021) were significantly associated with PO patients.

      Phylogenetic analysis

      Sequence comparison and phylogenetic analysis of representative EV-A71 strains associated with uncomplicated HFMD and severe cases found no differences in the VP-1 region (Figure 2). There was no discernible clustering of EV-A71 strains according to severity of illness. All of the viruses analyzed were EV-A71 C4a strains, which were previously associated with the 2012 outbreak in Cambodia. The 2014 viruses clustered closest to EV-A71 strains associated with the Cambodian EV-A71 epidemic in 2012; and Vietnamese strains from 2011-2013.
      Figure 2
      Figure 2Phylogenetic tree based on VP-1 sequences of enterovirus 71 strains, generated in MEGA5 (
      • Ooi M.H.
      • Wong S.C.
      • Podin Y.
      • Akin W.
      • del Sel S.
      • Mohan A.
      • et al.
      Human enterovirus 71 disease in Sarawak, Malaysia: a prospective clinical, virological, and molecular epidemiological study.
      ). The numbers next to the branches indicate the percentage of 1000 bootstrap replicates that support each phylogenetic branch. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site.
      ▲ EV-A71 strains sequenced in the present study associated with severe illness.
      △ EV-A71 strains sequenced in the present study associated with uncomplicated illness.

      Discussion

      In this prospective study, comparative analysis showed uncomplicated HFMD and ECP patients differed significantly regarding gender, province of residence, weight-for-age adjusted z score, time to admission, fever, mouth and skin lesions, haemoglobin, mean corpuscular volume, neutrophil and platelets counts. ECP and PO patients also differed regarding skin lesions, neurological signs, neutrophil count in blood or CSF and median blood glucose level. Provinces outside Phnom Penh, distance to hospital reference center, weight-for-age adjusted z score, haemoglobin, mean corpuscular volume and thrombocytosis were associated with severe forms of EV-A71 infection.
      Patients experiencing severe EV-A71 infection had a significantly lower percentage of mouth or skin lesions. In 29 fatal cases of EV-A71 infections, one of the largest case series previously reported, oral ulcers and extremity rash were reported in 66% and 62% of patients respectively (
      • Chan L.G.
      • Parashar U.D.
      • Lye M.S.
      • Ong F.G.
      • Zaki S.R.
      • Alexander J.P.
      • et al.
      Deaths of children during an outbreak of hand, foot, and mouth disease in Sarawak, Malaysia: clinical and pathological characteristics of the disease.
      ), contrasting with the lower rate of skin or mucosal lesions among PO patients (24%) in our study. The absence of history or clinical findings of mouth or skin lesion have already been associated with severe disease (
      • Chong C.-Y.
      • Chan K.-P.
      • Shah V.A.
      • Ng W.Y.M.
      • Lau G.
      • Teo T.E.-S.
      • et al.
      Hand, foot and mouth disease in Singapore: a comparison of fatal and non-fatal cases.
      ,
      • Ooi M.H.
      • Wong S.C.
      • Podin Y.
      • Akin W.
      • del Sel S.
      • Mohan A.
      • et al.
      Human enterovirus 71 disease in Sarawak, Malaysia: a prospective clinical, virological, and molecular epidemiological study.
      ). This association could be simply related to a bias in patient recruitment. Indeed, by an implicit definition, non-severe symptomatic patients included in our study went to hospital because they had skin or mucosal lesions, whereas patients with neurological or respiratory symptoms were hospitalized regardless of their dermatological status. Furthermore, EV-A71 related rash has been reported as often subtle and scant, especially in younger patients with natural tanned skin, and could also have been missed because of delayed hospitalization (
      • Chan L.G.
      • Parashar U.D.
      • Lye M.S.
      • Ong F.G.
      • Zaki S.R.
      • Alexander J.P.
      • et al.
      Deaths of children during an outbreak of hand, foot, and mouth disease in Sarawak, Malaysia: clinical and pathological characteristics of the disease.
      ,
      • Zhang Q.
      • MacDonald N.E.
      • Smith J.C.
      • Cai K.
      • Yu H.
      • Li H.
      • et al.
      Severe enterovirus type 71 nervous system infections in children in the Shanghai region of China: clinical manifestations and implications for prevention and care.
      ). No known mutations associated with skin or neurotropism (
      • Ishimaru Y.
      • Nakano S.
      • Yamaoka K.
      • Takami S.
      Outbreaks of hand, foot, and mouth disease by enterovirus 71. High incidence of complication disorders of central nervous system.
      ,
      • Hsueh C.
      • Jung S.M.
      • Shih S.R.
      • Kuo T.T.
      • Shieh W.J.
      • Zaki S.
      • et al.
      Acute encephalomyelitis during an outbreak of enterovirus type 71 infection in Taiwan: report of an autopsy case with pathologic, immunofluorescence, and molecular studies.
      ) were found in the VP1 region, other genes were not sequenced. A final hypothesis could be that, in a particular population, skin or dermatological lesions reflect the host’s ability to mount an efficient immune response against EV-A71, as reported in other viral infections, such as parvovirus B19 infection (
      • Bültmann B.D.
      • Sotlar K.
      • Klingel K.
      Parvovirus B19.
      ).
      Our findings confirm previously reported risk factors for severe forms of EV-A71 infection, including male gender (
      • Zhang Q.
      • MacDonald N.E.
      • Smith J.C.
      • Cai K.
      • Yu H.
      • Li H.
      • et al.
      Severe enterovirus type 71 nervous system infections in children in the Shanghai region of China: clinical manifestations and implications for prevention and care.
      ), delayed time to hospitalization (
      • Fang Y.
      • Wang S.
      • Zhang L.
      • Guo Z.
      • Huang Z.
      • Tu C.
      • et al.
      Risk factors of severe hand, foot and mouth disease: A meta-analysis.
      ), high or prolonged fever (
      • Ooi M.H.
      • Wong S.C.
      • Podin Y.
      • Akin W.
      • del Sel S.
      • Mohan A.
      • et al.
      Human enterovirus 71 disease in Sarawak, Malaysia: a prospective clinical, virological, and molecular epidemiological study.
      ,
      • Zhang Q.
      • MacDonald N.E.
      • Smith J.C.
      • Cai K.
      • Yu H.
      • Li H.
      • et al.
      Severe enterovirus type 71 nervous system infections in children in the Shanghai region of China: clinical manifestations and implications for prevention and care.
      ,
      • Ooi M.H.
      • Wong S.C.
      • Mohan A.
      • Podin Y.
      • Perera D.
      • Clear D.
      • et al.
      Identification and validation of clinical predictors for the risk of neurological involvement in children with hand, foot, and mouth disease in Sarawak.
      ,
      • Chang L.Y.
      • Lin T.Y.
      • Hsu K.H.
      • Huang Y.C.
      • Lin K.L.
      • Hsueh C.
      • et al.
      Clinical features and risk factors of pulmonary oedema after enterovirus-71-related hand, foot, and mouth disease.
      ) and vomiting (
      • Chong C.-Y.
      • Chan K.-P.
      • Shah V.A.
      • Ng W.Y.M.
      • Lau G.
      • Teo T.E.-S.
      • et al.
      Hand, foot and mouth disease in Singapore: a comparison of fatal and non-fatal cases.
      ,
      • Ooi M.H.
      • Wong S.C.
      • Podin Y.
      • Akin W.
      • del Sel S.
      • Mohan A.
      • et al.
      Human enterovirus 71 disease in Sarawak, Malaysia: a prospective clinical, virological, and molecular epidemiological study.
      ,
      • Chang L.Y.
      • Lin T.Y.
      • Hsu K.H.
      • Huang Y.C.
      • Lin K.L.
      • Hsueh C.
      • et al.
      Clinical features and risk factors of pulmonary oedema after enterovirus-71-related hand, foot, and mouth disease.
      ). A clear and well-documented overrepresentation of males exists in many infectious diseases, with an apparent influence of testosterone and X chromosome on immune responses and disease progression (
      • Libert C.
      • Dejager L.
      • Pinheiro I.
      The X chromosome in immune functions: when a chromosome makes the difference.
      ,
      • Bernin H.
      • Lotter H.
      Sex bias in the outcome of human tropical infectious diseases: influence of steroid hormones.
      ). However, a recent meta-analysis of 19 studies investigating the risk factors for severe EV-A71 disease found no association with gender (
      • Fang Y.
      • Wang S.
      • Zhang L.
      • Guo Z.
      • Huang Z.
      • Tu C.
      • et al.
      Risk factors of severe hand, foot and mouth disease: A meta-analysis.
      ).
      Our study did not confirm other risk factors for CNS disease identified following large previous EV-A71 outbreaks. Young age (
      • Ooi M.H.
      • Wong S.C.
      • Podin Y.
      • Akin W.
      • del Sel S.
      • Mohan A.
      • et al.
      Human enterovirus 71 disease in Sarawak, Malaysia: a prospective clinical, virological, and molecular epidemiological study.
      ,
      • Zhang Q.
      • MacDonald N.E.
      • Smith J.C.
      • Cai K.
      • Yu H.
      • Li H.
      • et al.
      Severe enterovirus type 71 nervous system infections in children in the Shanghai region of China: clinical manifestations and implications for prevention and care.
      ,
      • Chang L.Y.
      • Lin T.Y.
      • Hsu K.H.
      • Huang Y.C.
      • Lin K.L.
      • Hsueh C.
      • et al.
      Clinical features and risk factors of pulmonary oedema after enterovirus-71-related hand, foot, and mouth disease.
      ), home care (
      • Fang Y.
      • Wang S.
      • Zhang L.
      • Guo Z.
      • Huang Z.
      • Tu C.
      • et al.
      Risk factors of severe hand, foot and mouth disease: A meta-analysis.
      ) and neurological signs (headache, history of lethargy, limb weakness) (
      • Zhang Q.
      • MacDonald N.E.
      • Smith J.C.
      • Cai K.
      • Yu H.
      • Li H.
      • et al.
      Severe enterovirus type 71 nervous system infections in children in the Shanghai region of China: clinical manifestations and implications for prevention and care.
      ,
      • Ooi M.H.
      • Wong S.C.
      • Mohan A.
      • Podin Y.
      • Perera D.
      • Clear D.
      • et al.
      Identification and validation of clinical predictors for the risk of neurological involvement in children with hand, foot, and mouth disease in Sarawak.
      ,
      • Chang L.Y.
      • Lin T.Y.
      • Hsu K.H.
      • Huang Y.C.
      • Lin K.L.
      • Hsueh C.
      • et al.
      Clinical features and risk factors of pulmonary oedema after enterovirus-71-related hand, foot, and mouth disease.
      ) have been reported. Viral factors such as genogroup (
      • Ooi M.H.
      • Wong S.C.
      • Podin Y.
      • Akin W.
      • del Sel S.
      • Mohan A.
      • et al.
      Human enterovirus 71 disease in Sarawak, Malaysia: a prospective clinical, virological, and molecular epidemiological study.
      ) or VP1 amino acid substitution (
      • Liu N.
      • Xie J.
      • Qiu X.
      • Jia L.
      • Wu Z.
      • Ma Y.
      • et al.
      An atypical winter outbreak of hand, foot, and mouth disease associated with human enterovirus 71, 2010.
      ) have also been studied. Co-infection and antibody responses have not been shown to be related to severe forms (
      • Ooi M.H.
      • Wong S.C.
      • Podin Y.
      • Akin W.
      • del Sel S.
      • Mohan A.
      • et al.
      Human enterovirus 71 disease in Sarawak, Malaysia: a prospective clinical, virological, and molecular epidemiological study.
      ,
      • Yang C.
      • Deng C.
      • Wan J.
      • Zhu L.
      • Leng Q.
      Neutralizing antibody response in the patients with hand, foot and mouth disease to enterovirus 71 and its clinical implications.
      ). In multivariate meta-analysis, male gender, residing in rural areas and absence of mouth ulcers at examination were not significantly associated with severity (
      • Fang Y.
      • Wang S.
      • Zhang L.
      • Guo Z.
      • Huang Z.
      • Tu C.
      • et al.
      Risk factors of severe hand, foot and mouth disease: A meta-analysis.
      ).
      Our results are conflicting regarding typical CSF in EV-A71 CNS infection. In one study in children with EV-A71 meningitis, CSF showed a neutrophil predominance in 64% of cases and >200 WBC/mm3 in 25% (
      • Goldberg F.
      • Weiner L.B.
      Cerebrospinal fluid white blood cell counts and lactic acid dehydrogenase in enterovirus type 71 meningitis.
      ) In a case series of 29 children with EV-A71 fatal infections, lymphocytic predominance was reported in 11/12 patients with pleiocytosis on lumbar puncture (
      • Chan L.G.
      • Parashar U.D.
      • Lye M.S.
      • Ong F.G.
      • Zaki S.R.
      • Alexander J.P.
      • et al.
      Deaths of children during an outbreak of hand, foot, and mouth disease in Sarawak, Malaysia: clinical and pathological characteristics of the disease.
      ). Our results confirm previous findings on the association between EV-A71 related pulmonary oedema and high neutrophil count or hyperglycemia (
      • Chang L.Y.
      • Lin T.Y.
      • Hsu K.H.
      • Huang Y.C.
      • Lin K.L.
      • Hsueh C.
      • et al.
      Clinical features and risk factors of pulmonary oedema after enterovirus-71-related hand, foot, and mouth disease.
      ,
      • Lin T.-Y.
      • Chang L.-Y.
      • Hsia S.-H.
      • Huang Y.C.
      • Chiu C.H.
      • Hsueh C.
      • et al.
      The 1998 enterovirus 71 outbreak in Taiwan: pathogenesis and management.
      ) but also highlights the higher rate of neutrophilic CSF in PO patients and lymphocytic CSF in encephalitis patients .
      Interestingly, PO patients in our study did not differ significantly in terms of clinical or MRI brainstem lesions or dysautonomic symptoms, compared to ECP patients. Our findings suggest that PO pathophysiology in Cambodian children may be more closely related to cytokinic inflammatory cardiac dysfunction rather than to neurogenic pulmonary oedema.
      Our study has important limitations, including the lack of outcome data for patients. Although many surrogate markers of malnutrition were identified as putative risk factors for severe disease, only low haemoglobin titres were independently associated with encephalitis in multivariate analysis. Lack of power and some degree of variable colinearity could explain the lack of differences across groups in multivariate analysis. This study contained potential biases in the recruitment of patients due to the fact that parents of rural origin are less likely to travel long distances to bring their children to the hospital for mild disease such as uncomplicated HFMD, hence the overrepresentation and late presentation of severe illness among children of rural origin. Indeed, there was a negative correlation between distance to hospital and GCS on admission (Pearson −2.56, p = 0.004), suggesting people from remote areas tend to come to hospital for more severe symptoms. However, there was no correlation between distance to hospital and weight adjusted z score (Pearson = −0.06, p = 0.43), suggesting malnutrition may be an independent key factor for severe disease. In addition, the significant association between rash and mouth ulcerations with mild illness may be an artefact due to these factors being included in the definition for uncomplicated HFMD. Finally, the possibility that co-infection with other viral pathogens contributed to the clinical characteristics of children infected with EV-A71 was not explored in this study. However, previous studies suggest that co-infections with other viruses rarely occur in EV-A71 associated encephalitis or HFMD cases (
      • Le V.T.
      • Phan T.Q.
      • Do Q.H.
      • Nguyen B.H.
      • Lam Q.B.
      • Bach V.
      • et al.
      Viral etiology of encephalitis in children in southern Vietnam: results of a one-year prospective descriptive study.
      ,
      • Sapkal G.N.
      • Bondre V.P.
      • Fulmali P.V.
      • Patil P.
      • Gopalkrishna V.
      • Dadhania V.
      • et al.
      Enteroviruses in patients with acute encephalitis, uttar pradesh, India.
      ,
      • Yang F.
      • Du J.
      • Hu Y.
      • Wang X.
      • Xue Y.
      • Dong J.
      • et al.
      Enterovirus coinfection during an outbreak of hand, foot, and mouth disease in Shandong, China.
      ).
      In this prospective EV-A71 outbreak study in Cambodia, we confirmed the previously reported association of male gender and absence of mouth or skin lesions with severe disease. We also highlighted the strong association of neutrophils in blood but also in the CSF of patients with pulmonary oedema. More importantly, we identified new putative malnutrition-related risk factors of severe disease that warrant further investigations, including a more detailed analysis of nutritional status.

      Funding

      This study was supported by the Second Health Sector Support Program, Ministry of Health of the Kingdom of Cambodia, and the Asian Development Bank.
      Ethical approval
      Ethical clearance was obtained from the Cambodian National Ethics Committee for Human Research before testing commenced. All parents/guardians of sick children who participated in this study provided written informed consent; and clinical case report forms were anonymized to protect the identity of patients.

      Conflict of interest

      Philippe Buchy is currently an employee of GSK Vaccines but this position has no link with the work presented here. The other authors declare no conflict of interest.

      Acknowledgments

      The authors thank the patients and parents who participated in this study. We also thank the clinical staff from Kantha Bopha hospitals.

      References

        • Bültmann B.D.
        • Sotlar K.
        • Klingel K.
        Parvovirus B19.
        N Engl J Med. 2004; 350: 2006-2007
        • Bernin H.
        • Lotter H.
        Sex bias in the outcome of human tropical infectious diseases: influence of steroid hormones.
        J Infect Dis. 2014; 209: S107-S113
        • Chan L.G.
        • Parashar U.D.
        • Lye M.S.
        • Ong F.G.
        • Zaki S.R.
        • Alexander J.P.
        • et al.
        Deaths of children during an outbreak of hand, foot, and mouth disease in Sarawak, Malaysia: clinical and pathological characteristics of the disease.
        Clin Infect Dis. 2000; 31: 678-683
        • Chang L.Y.
        • Lin T.Y.
        • Hsu K.H.
        • Huang Y.C.
        • Lin K.L.
        • Hsueh C.
        • et al.
        Clinical features and risk factors of pulmonary oedema after enterovirus-71-related hand, foot, and mouth disease.
        Lancet. 1999; 354: 1682-1686
        • Chong C.-Y.
        • Chan K.-P.
        • Shah V.A.
        • Ng W.Y.M.
        • Lau G.
        • Teo T.E.-S.
        • et al.
        Hand, foot and mouth disease in Singapore: a comparison of fatal and non-fatal cases.
        Acta Paediatr. 2003; 92: 1163-1169
        • Duong V.
        • Mey C.
        • Eloit M.
        • Zhu H.
        • Danet L.
        • Huang Z.
        • et al.
        Molecular epidemiology of human enterovirus 71 at the origin of an epidemic of fatal hand, foot and mouth disease cases in Cambodia.
        Emerg Microbes Infect. 2016; 5: e104
        • Fang Y.
        • Wang S.
        • Zhang L.
        • Guo Z.
        • Huang Z.
        • Tu C.
        • et al.
        Risk factors of severe hand, foot and mouth disease: A meta-analysis.
        Scand J Infect Dis. 2014; 46: 515-522
        • Goldberg F.
        • Weiner L.B.
        Cerebrospinal fluid white blood cell counts and lactic acid dehydrogenase in enterovirus type 71 meningitis.
        Clin Pediatr (Phila). 1981; 20: 327-330
        • Horwood P.F.
        • Andronico A.
        • Tarantola A.
        • Salje H.
        • Duong V.
        • Mey C.
        • et al.
        Seroepidemiology of human enterovirus 71 infection among Cambodian children.
        Emerg Infect Dis. 2016; 22: 92-95
        • Hsueh C.
        • Jung S.M.
        • Shih S.R.
        • Kuo T.T.
        • Shieh W.J.
        • Zaki S.
        • et al.
        Acute encephalomyelitis during an outbreak of enterovirus type 71 infection in Taiwan: report of an autopsy case with pathologic, immunofluorescence, and molecular studies.
        Mod Pathol Off J U S Can Acad Pathol Inc. 2000; 13: 1200-1205
        • Huang C.-C.
        • Liu C.-C.
        • Chang Y.-C.
        • Chen C.Y.
        • Wang S.T.
        • Yeh T.F.
        Neurologic complications in children with enterovirus 71 infection.
        N Engl J Med. 1999; 341: 936-942
        • Huang W.-C.
        • Shih W.-L.
        • Yang S.-C.
        • Yen T.Y.
        • Lee J.T.
        • Huang Y.C.
        • et al.
        Predicting severe enterovirus 71 infection: Age, comorbidity, and parental behavior matter.
        J Microbiol Immunol Infect. 2014; (pii: S1684-1182(14)00243-6. [Epub ahead of print])https://doi.org/10.1016/j.jmii.2014.11.013
        • Ishimaru Y.
        • Nakano S.
        • Yamaoka K.
        • Takami S.
        Outbreaks of hand, foot, and mouth disease by enterovirus 71. High incidence of complication disorders of central nervous system.
        Arch Dis Child. 1980; 55: 583-588
        • Jain S.
        • Patel B.
        • Bhatt G.C.
        Enteroviral encephalitis in children: clinical features, pathophysiology, and treatment advances.
        Pathog Glob Health. 2014; 108: 216-222
        • Khanh T.H.
        • Sabanathan S.
        • Thanh T.T.
        • Thoa le P.K.
        • Thuong T.C.
        • Hang V.t.
        • et al.
        Enterovirus 71-associated hand, foot, and mouth disease, Southern Vietnam, 2011.
        Emerg Infect Dis. 2012; 18: 2002-2005
        • Le V.T.
        • Phan T.Q.
        • Do Q.H.
        • Nguyen B.H.
        • Lam Q.B.
        • Bach V.
        • et al.
        Viral etiology of encephalitis in children in southern Vietnam: results of a one-year prospective descriptive study.
        PLoS Negl Trop Dis. 2010; 4: e854
        • Li J.
        • Chen F.
        • Liu T.
        • Wang L.
        MRI findings of neurological complications in hand-foot-mouth disease by enterovirus 71 infection.
        Int J Neurosci. 2012; 122: 338-344
        • Libert C.
        • Dejager L.
        • Pinheiro I.
        The X chromosome in immune functions: when a chromosome makes the difference.
        Nat Rev Immunol. 2010; 10: 594-604
        • Lin T.-Y.
        • Chang L.-Y.
        • Hsia S.-H.
        • Huang Y.C.
        • Chiu C.H.
        • Hsueh C.
        • et al.
        The 1998 enterovirus 71 outbreak in Taiwan: pathogenesis and management.
        Clin Infect Dis Off Publ Infect Dis Soc Am. 2002; 34: S52-S57
        • Liu N.
        • Xie J.
        • Qiu X.
        • Jia L.
        • Wu Z.
        • Ma Y.
        • et al.
        An atypical winter outbreak of hand, foot, and mouth disease associated with human enterovirus 71, 2010.
        BMC Infect Dis. 2014; 14: 123
        • Nagy G.
        • Takátsy S.
        • Kukán E.
        • Mihály I.
        • Dömök I.
        Virological diagnosis of enterovirus type 71 infections: experiences gained during an epidemic of acute CNS diseases in Hungary in 1978.
        Arch Virol. 1982; 71: 217-227
        • Nix W.A.
        • Oberste M.S.
        • Pallansch M.A.
        Sensitive, seminested PCR amplification of VP1 sequences for direct identification of all enterovirus serotypes from original clinical specimens.
        J Clin Microbiol. 2006; 44: 2698-2704
        • Ooi M.H.
        • Wong S.C.
        • Podin Y.
        • Akin W.
        • del Sel S.
        • Mohan A.
        • et al.
        Human enterovirus 71 disease in Sarawak, Malaysia: a prospective clinical, virological, and molecular epidemiological study.
        Clin Infect Dis Off Publ Infect Dis Soc Am. 2007; 44: 646-656
        • Ooi M.H.
        • Wong S.C.
        • Mohan A.
        • Podin Y.
        • Perera D.
        • Clear D.
        • et al.
        Identification and validation of clinical predictors for the risk of neurological involvement in children with hand, foot, and mouth disease in Sarawak.
        BMC Infect Dis. 2009; 9: 3
        • Sapkal G.N.
        • Bondre V.P.
        • Fulmali P.V.
        • Patil P.
        • Gopalkrishna V.
        • Dadhania V.
        • et al.
        Enteroviruses in patients with acute encephalitis, uttar pradesh, India.
        E merg Infect Dis. 2009; 15: 295-298
        • Shindarov L.M.
        • Chumakov M.P.
        • Voroshilova M.K.
        • Bojinov S.
        • Vasilenko S.M.
        • Iordanov I.
        • et al.
        Epidemiological, clinical, and pathomorphological characteristics of epidemic poliomyelitis-like disease caused by enterovirus 71.
        J Hyg Epidemiol Microbiol Immunol. 1979; 23: 284-295
        • Tamura K.
        • Peterson D.
        • Peterson N.
        • Stecher G.
        • Nei M.
        • Kumar S.
        MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods.
        Mol Biol Evol. 2011; 28: 2731-2739
        • Yang C.
        • Deng C.
        • Wan J.
        • Zhu L.
        • Leng Q.
        Neutralizing antibody response in the patients with hand, foot and mouth disease to enterovirus 71 and its clinical implications.
        Virol J. 2011; 8: 306
        • Yang F.
        • Du J.
        • Hu Y.
        • Wang X.
        • Xue Y.
        • Dong J.
        • et al.
        Enterovirus coinfection during an outbreak of hand, foot, and mouth disease in Shandong, China.
        Clin Infect Dis Off Publ Infect Dis Soc Am. 2011; 53: 400-401
        • Zhang Q.
        • MacDonald N.E.
        • Smith J.C.
        • Cai K.
        • Yu H.
        • Li H.
        • et al.
        Severe enterovirus type 71 nervous system infections in children in the Shanghai region of China: clinical manifestations and implications for prevention and care.
        Pediatr Infect Dis J. 2014; 33: 482-487
        • Zhao Y.Y.
        • Jin H.
        • Zhang X.F.
        • Wang B.
        Case-fatality of hand, foot and mouth disease associated with EV-A71: a systematic review and meta-analysis.
        Epidemiol Infect. 2015; 143: 3094-3102