Dissertation committee:
Αθανάσιος Μίχος, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Χριστίνα Κανακά- Gantenbein, Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Σουλτάνα Σιαχανίδου, Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Βασιλική Συριοπούλου, Ομότιμη Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Βασιλική Σπούλου, Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Φλώρα Μπακοπούλου, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Εμμανουηλ Ζουμάκης, Επίκουρος Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Summary:
Introduction: Central nervous system (CNS) infections are caused by bacteria, viruses, fungi and parasites and are a serious disease that can cause multiple and long-term consequences for the patient's health and even death. Early diagnosis and treatment can be crucial for the subsequent development of the patient's health. Molecular epidemiology data of CNS infections, especially viral ones, available for the pediatric population are limited.
Purpose: The aim of the present study was: 1) to investigate the importance of syndromic diagnosis for the rapid detection of CNS infections in children using multiplex PCR in cerebrospinal fluid (CSF). In addition, to what extent it can contribute to the reduction of antibiotic administration days, the reduction of hospitalization days as well as the reduction of hospitalization costs. 2) the molecular detection, genotyping and phylogenetic analysis of enterovirus and parechovirus strains associated with CNS infections or septic syndrome in children and neonates. In addition, the correlation of the genotype of the viruses with the clinical and epidemiological characteristics and the severity of the disease.
Patients and method: During the first period of the study, a prospective cohort study was carried out in previously healthy children 0-16 years old with clinical suspicion of meningitis or encephalitis who were hospitalized at the "Agia Sophia" Children's Hospital during the period 04/2018-04/2019. The study included children in whom, at the discretion of the attending physician, a CSF sample was obtained and > 15 cells/mm3 were found. CSF was tested with the multiplex PCR syndromic method for CNS infection FilmArray-ME which tests for the following microorganisms: CMV, Enterovirus, HSV 1/2, HHV-6, Η. parechovirus, VZV, Cryptococcus neoformans/gattii, E.coli K1, H.influenzae, L.monocytogenes, N. meningitidis, S. agalactiae (GBS), and S. pneumoniae. For each child screened by a syndromic diagnostic method, an age-matched control child was selected, whose CSF was tested by individual PCRs for specific microorganisms at the discretion of the attending physician. The CSF of all children was examined by standard microbiological methods. Final results compared between case and control group for outcome, length of hospital stay, duration of antimicrobial use, duration of acyclovir use, and total hospital cost. During the second period 5/2019-12/2021 of the study, all CSF samples of children with suspected CNS infection were tested with a syndromic diagnostic method of the CSF without limitation in relation to the number of cells in the CSF. In children whose CSF samples were found positive for enteroviruses (EV) or HpeV, stool and pharyngeal samples were collected and performed genotyping and phylogenetic analysis.
Results: The population of children who met the inclusion criteria for the cohort study was 142 children (71 in the syndromic diagnosis group). A pathogen was detected in 37/71 (52.1%) children using multiplex PCR and in 16/71 (22.5%) in the control group (P value < 0.001). In cases of aseptic meningitis, virus was detected in 27/61 (44.2%) and in 11/66 (16.7%) in the control group (P value < 0.001). Median length of stay in patients and controls with aseptic meningitis was 5 IQR (4-8) and 8 IQR (6-10) days, respectively (P value < 0.001). The median duration of antimicrobials in patients and controls was 4 IQR (2-5.7) and 7 IQR (5-10) days respectively (P value < 0.001). Hospitalization costs were estimated in patients and controls to be €1042 (932–1372) and €1522 (1302–1742), respectively (P value < 0.001). In the second period when all children were screened by multiplex PCR regardless of CSF cells,out of the 330 children tested, 75 (22.7%) had EV infection and 6/330 (1.8%) HPV infection. The median age of children with CNS EV infection was 2 months (IQR: 1–60) and 44/75 (58.7%) of them were boys. There was a seasonal distribution of EV infections, with most cases detected between June and September (38/75.50.7%). The predominant EV genotypes were CV-B5 (16/45, 35.6%), E30 (10/45, 22.2%), E16 (6/45, 13.3%), and E11 (5/45, 11 .1%). No previously unpublished EV serotypes associated with unusual neurologic manifestations were found, and all children recovered without apparent acute sequelae. All cases of HPeV infection were identified in infants under 3 months of age, and in this age group HPeV accounted for 11.6% (6/51) of cases of viral meningitis. The median age of HPeV-positive infants was 0.6 months (IQR: 0.5–1.6) and 83.3% (5/6) were boys. Neonates (6/6) presented with febrile illness, 3/6 presented with sepsis-like illness, 3/6 with rash, and 1/6 with convulsions. In 3/6 neonates empiric antibiotic treatment was discontinued after detection of HPeV while the neonate with sepsis-like disease received intravenous γ-globulin. Genotyping yielded 3 previously unpublished recombinant strains.
Conclusions: In conclusion, in our study, the use of syndromic multiplex PCR had a significant benefit in the clinical management of children with CNS infection in terms of the reduction of antimicrobial drugs used, the total duration and cost of hospitalization. The molecular study of EVs and HPeVs provided data on the epidemiology and genetic structure of the genotypes circulating in our country.
