Meningitis gone viral: Description of the German Echo Wave 2013

BACKGROUND: Aseptic meningitis epidemics may pose various health care challenges. OBJECTIVE: We describe the German enterovirus meningitis epidemics in the university hospital centers of Düsseldorf, Cologne and Berlin between January 1st and December 31st, 2013. RESULTS: A total of 72 enterovirus (EV-positive) meningitis cases were detected in our multicenter cohort, corresponding to 2.1% of all EV-positive cases which were voluntarily reported within the National Enterovirussurveillance (based on investigation of patients with suspected aseptic meningitis/encephalitis and acute flaccid paralysis) by physicians within this period of time. Among these 72 patients, 38 (52.8%) were echovirus-positive (18 pediatric and 20 adult cases, median age 18.5 years). At the same time, 45 aseptic meningitis cases in our cohort were excluded to be due to enteroviral infection (EV-negative). Three EV-negative patients were tested positive for varicella zoster virus (VZV) and one EV-negative patient for herpes simplex virus 2. Hospitalization was significantly longer in EV-negative cases. Cerebrospinal fluid analysis did not reveal significant differences between the two groups. After discharge, EV-meningitis resulted in significant burden of sick leave in our pediatric cohort as parents had to care for the children at home. CONCLUSION: Voluntary disease-specific surveillance, such as provided by the National Enterovirussurveillance in our study may be a valuable tool for epidemiological research.

paralysis) by physicians within this period of time. Among these 72 patients, 38 (52.8%) were echovirus-positive (18 pediatric and 20 adult cases, median age 18.5 years). At the same time, 45 aseptic meningitis cases in our cohort were excluded to be due to enteroviral infection (EV-negative). Three EV-negative patients were tested positive for varicella zoster virus (VZV) and one EV-negative patient for herpes simplex virus 2.
Hospitalization was significantly longer in EV-negative cases. Cerebrospinal fluid analysis did not reveal significant differences between the two groups. After discharge, EVmeningitis resulted in significant burden of sick leave in our pediatric cohort as parents had to care for the children at home. CONCLUSION: Voluntary disease-specific surveillance, such as provided by the National Enterovirussurveillance in our study may be a valuable tool for epidemiological research.

Background
Periodic aseptic meningitis epidemics can be a challenge in patient-and health care. A large retrospective analysis of a US-American cohort revealed that in 21% of cases the etiology of aseptic meningitis remains unknown1. Aseptic meningitis is defined2 by an inflammation of the leptomeninx in which the causative agent cannot be identified by cerebrospinal fluid culture3. Viruses are the most common causes of this disease1,3. Most viral meningitides are caused by enteroviruses4. These belong to the picornaviridae consisting of species A-D, and comprise the subtypes of coxsackieviruses, polioviruses, echoviruses and certain enterovirus serotypes. The infection occurs usually via fecal-oral route, but also infestation of the respiratory tract and a droplet infection are possible.
Interestingly, in previous studies, adults were more likely to be tested positive for herpes simplex virus and West Nile virus while children were more likely to be tested positive for enterovirus (EV)5. Therefore, multiple studies have been conducted in order to better understand this phenomenon: In a Danish nation-wide prospective observational study between 1st of January 2015 to 30th of June 2016 an unfavorable outcome of viral meningitis was documented in 17% of all patients6. According to a UK study, the infection rates of viral meningitis are mainly driven by an EV predominance of echovirus 307. EV meningitis epidemics in Shandong, (People's Republic of China, 20148) and Finland (2009 and 20109) and the clinical pattern of viral central nervous system (CNS) infections in Italy10 have previously been characterized: EV-positive patients presented with fever, nausea and vomiting, were most likely to be children, and had no clear gender predominance. In addition to antipyretic and analgesic therapy for temperature control, pleconaril, an inhibitor of enteroviral replication, is available. However, it was not approved in Europe, as a placebo-controlled double-blind study demonstrated only a moderate effect of the drug11. In particularly severe cases, administration of immunoglobulins also had a positive effect on the course of the disease12 . Cerebrospinal fluid (CSF) and clinical features of EV-positive meningitis patients in Germany13 and the differences in adult and pediatric EV-positive meningitis patients in Switzerland14 have already been analyzed, but there is still a paucity of data describing the differences in EVpositive meningitis and EV-negative meningitis patients.

Patients And Methods
We conducted a retrospective chart review study at the Departments of Neurology of the Heinrich-Heine University Düsseldorf, the University Hospital of Cologne, the Charité -Universitätsmedizin Berlin and the Department of General Pediatrics, Neonatology and Pediatric Cardiology of the Heinrich-Heine University Düsseldorf searching for all patients with aseptic CNS infection in 2013. The study was approved by the ethics committee, University of Düsseldorf (registry number 4423). We used ICD-10 codes to identify cases of interest. As such, priority was given to the ICD-10 keys A87 and G02 (Table 1).
However, as patient data may not have been in the categories listed above due to less precise encryption despite manifest illness, a wider query was additionally performed to identify all patient data encoded as A85-A89 (Other viral encephalitis, not elsewhere classified; Unspecified viral encephalitis; Viral meningitis; Other viral infections of central nervous system, not elsewhere classified; Unspecified viral infection of central nervous system, Table 1) and G02-05 (Meningitis in other infectious and parasitic diseases classified elsewhere; Meningitis due to other and unspecified causes; Encephalitis, myelitis and encephalomyelitis; Encephalitis, myelitis and encephalomyelitis in diseases classified elsewhere, Table 1). Virological testing was performed in a standardized manner by the National Reference Laboratory for Poliomyelitis and Enteroviruses. Case definition After the above-mentioned identification of patient data, the patient records were individually evaluated to exclusively select cases of aseptic meningitis defined according Furthermore, we identified 45 enterovirus-negative cases (16 females, 29 males, median age 36 (28; 48.5) years). Among EV-negative cases, three were related to varicella zoster infection, one to herpes simplex 2 and no virus could be identified in the other patients.
EV-positive meningitis cases peaked in July/August (Figure 1). Analysis of the CSF parameters cell count (EV-positive: 81 (12; 205) cells/µl, EV negative 67 (17.5; 185.25) cells/µl, Figure 2A) and total protein (EV-positive 0.52 (0.35; 0.68) g/l, EV-negative 0.53 (0.36, 0.78) g/l, Figure 2B) in EV-positive and EV-negative patients revealed no significant difference (Mann-Whitney U test). 2 EV-positive and 1 EV-negative CSF samples could not be evaluated due to a blood contamination. Analysis of the number of nights spent in hospital by the patients revealed that hospitalization was significantly longer in EVnegative (6 (3; 13) nights) than in EV-positive cases (3 (1; 5) nights, adj. p < 0.01, Mann-Whitney U test, Figure 3). No fatal cases occurred. In children, parents reported of mild complaints after discharge like headache, backache and fatigue for 0 to 7 days (median 2 days; Düsseldorf cohort). After discharge, one of the parents had to care for the children at home (0 to 5 days, median 3 days; Düsseldorf cohort).
Adults vs. children A significantly higher ratio of children was found in the enteroviruspositive cohort, compared to the enterovirus negative group (38/72 vs. 4/45, Chi-Squaretest, adj. p < 0.001). Children (both EV-positive and EV-negative) had a shorter period of hospitalization (adj. p < 0.001) and lower CSF protein levels than adults (adj. p < 0.001).
There was no significant difference of CSF cell counts (Mann-Whitney U test, respectively).
An exclusive analysis of either adults or children did not reveal significant differences between EV-positive and EV-negative patients regarding age, duration of inpatient stay, CSF cell count and CSF total protein (Mann-Whitney U test, respectively). Numerical data of abovementioned comparisons are provided in Table 2.
Predictors of hospitalization periods In general, the duration of inpatient stay correlated with age (Spearman's Rho correlation coefficient 0.418, p < 0.001), CSF total protein (Spearman's Rho correlation coefficient 0.319, p < 0.001), and the delay from symptom onset to lumbar puncture (Spearman's Rho correlation coefficient 0.232, p = 0.023). For nominal variables, enterovirus status (η = -0,32) correlated with the duration of inpatient stay. In contrast, echovirus status, gender, and location (Neurological center the patient was treated) did not show a relevant correlation with the duration of the inpatient stay (| η | < 0,3). RKI database query An RKI database query (retrieved from https://survstat.rki.de/) revealed a total of 3455 EV-positive cases in 2013. 1242 of these cases had positive EV-PCR testing and 672 cases were positive for echovirus-30. Therefore, our study includes 2.1% of the reported EV-positive cases in Germany.

Discussion
The prevalence of enteroviral meningitis is high worldwide (estimated 75,000 cases annually in the United States)11, with the highest rates of infection in the summer and early autumn months (June, July and August; Figure 1). Notably, coxsackie and echoviruses are identified as responsible serotypes. In addition, enteroviruses are also highly neurotropic and are associated with meningitis, meningoencephalitis, poliomyelitislike anterior myelitis, and Guillain-Barré syndrome17,18. With age, the incidence of enteroviral meningitis decreases. Therefore, the incidence is highest in infants and toddlers19, which was also the case in our cohorts (Table 2). In adult patients, the disease generally necessitates inpatient treatment for several days20. When children are affected, one parent may be incapacitated for a certain period of time to care for the child. In both cases, the disease may be associated with a temporary inability to work (either patient or parent). Because of the high number of cases per year, considerable costs arise for society due to the loss of work and the necessary resources for medical treatment21, although meningitis caused by enteroviruses usually has a relatively benign course. To the best of our knowledge, the economic burden for society due to EV-positive meningitis has not been determined so far; and our data also provide just a limited insight, since we analyzed the duration of inpatient stay but did not assess any further inability to work.
Our data indicate that the course of EV-positive meningitis is predominantly benign, and that hospitalization time was significantly shorter in EV-positive, compared to EV-negative cases. This was also the case, when we did not consider the above-mentioned meningitis cases that received specific antiviral therapy (varicella zoster virus, herpes simplex virus 2). Moreover, a higher ratio of affected children and young adults were found in EVpositive cases. This could be explained by affected parents of young, diseased children.
Hence, earlier convalescence in EV-positive groups may be explained by differences of age between both groups rather than different courses of the disease in general. Despite the generally excellent outcome of aseptic meningitis, there are rare instances of complicated courses that may lead to persistent neurological disability or even death22-25. Strategies for the systematic containment of endemic diseases are focused on ensuring hygienic measures to prevent the spread of viruses, as some weeks after illness, virulent pathogens can still be excreted via the feces. In agreement with previous studies, enterovirus infections were detected as the most common cause for an aseptic meningitis in our cohort, driven by a high prevalence of echoviruses (75%). Indeed, our study demonstrates that voluntary reporting of diseases such as in this case can be an effective tool to better understand epidemiological details of certain diseases: The mean age of EVpositive patients in our centers was 15 (3.25; 32.75)

Consent for publication
Due to the retrospective manner of this study, patient consent was waived by the ethics committee.
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.