Enteroviruses (EVs) are well-known causes of sepsis in neonates and infants. In recent years, the extent to which parechoviruses may be contributing to neonatal and infant morbidity and mortality has begun to emerge.1–3
EVs and human parechoviruses (HPeVs) are nonenveloped, single-stranded, positive-sense RNA viruses and members of the Picornavirus family. They are common causes of neonatal and infant sepsis, worldwide.
EVs exist as multiple serotypes, subdivided into various genus, including echoviruses, Coxsackie A and B viruses and the numbered EVs. HPeVs exist in at least 17 genotypes, of which genotypes 1–6 are most commonly found in humans, with genotype 3 being most commonly responsible for sepsis in neonates and infants.
Whereas most episodes of EV and HPeV neonatal and infant sepsis are self-limiting, more severe illness can occur, and there are current concerns regarding longer term sequelae, particularly in HPeV infections, where there is more significant neurologic involvement. Previous studies have found that the clinical presentation of the 2 viruses are often indistinguishable.1,3
Our diagnostic virology laboratory has only recently (since mid 2014) introduced routine testing for parechoviruses as part of our neonatal and infant sepsis workup. We examined the demographics, laboratory results and clinical notes for pediatric patients admitted with sepsis with laboratory-confirmed human EV or HPeV infections of the cerebrospinal fluid (CSF), during February 2014 to August 2017.
All cases from February 2014 to August 2017 were selected on the basis of a positive CSF polymerase chain reaction result for either EV or HPeV RNA, using assays previously described elsewhere,2,4 which were performed as part of the routine workup for neonates or infants admitted with suspected sepsis. Additional samples were taken depending on the degree of clinical illness, including ethylenediaminetetraacetic acid blood, rectal swabs or stool samples and various respiratory samples (nasopharyngeal aspirates, throat swabs and bronchoalveolar lavages). For each patient, their laboratory values were extracted from the laboratory database, and clinical notes were reviewed.
This study was performed as part of a pediatric departmental audit which aimed to ensure that all EV- and HPeV-infected pediatric patients had received appropriate follow-up after discharge for sepsis, where there was laboratory-confirmed EV or HPeV infection of the CSF. Formal ethics approval was not required.
Clinical indices examined included age at presentation, length of stay, fever, rash, seizures, respiratory difficulty, feeding problems, antimicrobial use and admission to high dependency unit (HDU) or intensive care unit (ICU). Laboratory values compared included C-reactive protein, white cell counts (WCC), liver function tests, CSF profile (glucose, protein and cell counts) and radiologic investigations, where available.
Clinical and laboratory characteristics were compared between patients with HPeV and EV infection. Continuous variables were presented as mean and standard deviation (or median and interquartile range, if not normally distributed) and compared with Student’s t test (or Wilcoxon test). Binary variables were presented as frequency and percentages and compared with the Fisher exact test. Multivariable analysis for risk ratios (RRs) comparing HPeV infection with EV infection as the reference was estimated using log-Binomial regression model.
There were no statistically significant differences in age or sex of the children affected by EV versus HPeV central nervous system infections, but there appeared to be a difference in range, with EV often affecting older children than HPeV (interquartile range: 29–102 days for EV and 25.5–61 days for PeV).
Of a total of 163 cases, there were 131 EV (ie, 7 Coxsackie A, 18 Coxsackie B, 46 echoviruses, with 60 EVs that could not be typed further) and 32 HPeV infections (Table 1). All HPeV infections were caused by HPeV genotype 3. Of the EV cases, 73% (95 cases) were in children younger than 90 days (3 months), whereas greater than 90% (30 cases) of HPeV cases were in children younger than 90 days (3 months).
Cases of EV meningitis showed 3 peaks of activity each year with the most significant being in November to December. In contrast, HPeV had only 1 significant outbreak during 2 months in summer 2016 (Fig., Supplemental Digital Content 1, https://links.lww.com/INF/D223). There was no difference in the mean age or sex of the children affected by EV or PeV, although there was a difference in range, with EV meningitis affecting some older children.
A greater number of abnormal indices were found with HPeV than for EVs, with a greater likelihood of admission to HDU/ICU (P = 0.004) and a higher rate of persistent symptoms (ie, fever, irritability and feeding problems, P < 0.05). Compared with children infected with EV, children with HPeV were more likely to have an abnormally low WCC (leukopenia) (56% HPeV vs. 14% EV, P < 0.001) and an abnormally low lymphocyte count (lymphopenia) (91% HPeV vs. 39% EV, P < 0.001) (Table, Supplemental Digital Content 2, https://links.lww.com/INF/D224).
In contrast, EV cases were more likely to have a high WCC in the CSF (6% HPeV vs. 50% EV, P < 0.001) (Table, Supplemental Digital Content 3, https://links.lww.com/INF/D225). In the adjusted (log-Binomial regression) analysis, the HPeV cases were over 5 more times more likely to have lymphopenia than EV cases (RR = 5.11; 95% confidence interval: 1.53–17.05; P = 0.008), with EV cases being marginally more likely to have a higher CSF WCC (RR = 0.22; 95% confidence interval: 0.05–0.92; P = 0.038) (Table, Supplemental Digital Content 4, https://links.lww.com/INF/D226).
Other laboratory and clinical indices, including overall length of stay, did not differ significantly between the EV or HPeV cases. There was no significant difference in whether or not EV or HPeV cases received antibiotic (98.4% vs. 100%, respectively, P = 0.999) and/or acyclovir (37.1% vs. 34.4%, respectively, P = 0.839) treatment (Table, Supplemental Digital Content 2, https://links.lww.com/INF/D224).
Relatively few patients were deemed to require longer term follow-up. Of the total number of cases, 80% of children did not require any follow-up at 1 year post infection. At 1 year post infection, 3% of children were receiving follow-up by ophthalmology, with no abnormalities detected. Sixteen children (~10%) attended a routine hearing check, but none had any detectable sensorineural hearing loss. Only 3 patients (<2%) were reported as having had any developmental delay problems on admission: 1 child had delayed speech and manipulative skills—both of which resolved by 1 year post infection. Another child had some speech delay at 1 year follow-up, and 1 child had gross motor delay (despite a normal magnetic resonance imaging). As 2 of the 3 children with developmental problems had uncomplicated, short inpatient stays, this might suggest that their viral infections were not direct causes of this. However, this does not completely exclude this etiologic possibility.
Infections by EVs and HPeVs are well-documented causes of neonatal and infant sepsis. However, relatively few studies have compared the severity of clinical illness caused by these viruses within the same pediatric population within the same season.5
Here, we demonstrate differences in presentation and severity of these 2 viruses, with HPeV cases having a higher likelihood of having persistent fevers (P < 0.05), irritability or feeding problems (P < 0.05), leukopenia, lymphopenia and requiring admission to pediatric HDU or ICU units than children with EV infections. These findings are consistent with those reported from other studies.6–8 In addition, more specifically, Cabrerizo et al9 also noted a higher CSF pleocytosis in EV versus HPeV infections, as found in this study.
In our population, more children 30–90 days old (n = 19; 58%) were infected with HPeV than neonates (n = 11; 33%). Some studies have found children over the age of 2 months8 or 3 months9 were unaffected by HPeV, whereas 21% of our cases (7 patients) were diagnosed in children over 2 months old, with 2 cases being in a 4 months and 6 months old, respectively. Routine testing for HPeV in all children with febrile rash illness and sepsis may reveal a higher number of older children infected with HPeV.
Although some previous studies have found pediatric HPeV and EV infections clinically indistinguishable,1,3 anecdotally, in our pediatric population, nurses who worked with children involved in our recent HPeV outbreak2 reported that they were able to distinguish which children had HPeV rather than EV, before any laboratory confirmation, on their clinical presentation alone. These HPeV cases were noted to be generally more irritable and persistently unconsolable, tachychardic and pyrexial than the more frequently encountered annual, seasonal EV cases with which the nurses were very familiar.
The main limitation of this study is related to the infrequent and sporadic approach to the longer term follow-up of these EV- and HPeV-infected patients, as individual clinical teams were left to decide on whether patients being admitted under them warranted such follow-up. This was mostly based on the individual patient’s clinical course during their admission, as well as prior experiences of the lead pediatrician concerned, rather than any local consensus guidelines.
At present, clinical guidelines do not differentiate between the management of children presenting with EV versus HPeV infections.3 This study demonstrates that differences in the severity of clinical illness can be seen between the HPeV and EV central nervous system infections, with a greater degree of severity in HPeV cases. Further studies are required to clarify and confirm these findings, which may then lead to more practical clinical guidelines for the immediate and longer term management and follow-up of these patients.
1. Verboon-Maciolek MA, Krediet TG, Gerards LJ, et al. Severe neonatal parechovirus infection and similarity with enterovirus infection. Pediatr Infect Dis J. 2008;27:241245.
2. Tang JW, Holmes CW, Elsanousi FA, et al. Cluster of human parechovirus infections as the predominant cause of sepsis
in neonates and infants, Leicester, United Kingdom, 8 May to 2 August 2016. Euro Surveill. 2016;21
3. Drysdale SB, Kelly DF. Fifteen-minute consultation: enterovirus meningitis and encephalitis—when can we stop the antibiotics? Arch Dis Child Educ Pract Ed. 2017;102:6671.
4. Holmes CW, Koo SS, Osman H, et al. Predominance of enterovirus B and echovirus 30 as cause of viral meningitis in a UK population. J Clin Virol. 2016;81:9093.
5. Walters B, Peñaranda S, Nix WA, et al. Detection of human parechovirus (HPeV)-3 in spinal fluid specimens from pediatric patients in the Chicago area. J Clin Virol. 2011;52:187191.
6. Sharp J, Harrison CJ, Puckett K, et al. Characteristics of young infants in whom human parechovirus, enterovirus or neither were detected in cerebrospinal fluid during sepsis
evaluations. Pediatr Infect Dis J. 2013;32:213216.
7. Sano K, Hamada H, Hirose S, et al. Prevalence and characteristics of human parechovirus and enterovirus infections in febrile infants. Pediatr Int. 2018;60:142147.
8. Harvala H, Calvert J, Van Nguyen D, et al. Comparison of diagnostic clinical samples and environmental sampling for enterovirus and parechovirus surveillance in Scotland, 2010 to 2012. Euro Surveill. 2014;19:pii: 20772.
9. Cabrerizo M, Trallero G, Pena MJ, et al.; study group of “Enterovirus and parechovirus infections in children under 3 years-old, Spain” PI12-00904. Comparison of epidemiology and clinical characteristics of infections by human parechovirus vs. those by enterovirus during the first month of life. Eur J Pediatr. 2015;174:15111516.