Serum procalcitonin and other biologic markers to distinguish between bacterial and aseptic meningitis.

OBJECTIVE: To identify the biologic tests that best distinguish between bacterial and aseptic meningitis in an emergency department (ED). STUDY DESIGN: All children hospitalized for bacterial meningitis between 1995 and 2004 or for aseptic meningitis between 2000 and 2004 were included in a retrospective cohort study. Predictive values of blood (C-reactive protein, procalcitonin [PCT], white blood cell [WBC] count, neutrophil count) and cerebrospinal fluid (CSF) findings (protein, glucose, WBC count, neutrophil count) available in the ED were determined. Tests with the best predictive value were identified by using univariate and multivariate analyses and ROC curves comparison. RESULTS: Among the 167 patients included, 21 had bacterial meningitis. The CSF gram-stain and bacterial antigen test had 86% and 60% sensitivity rates, respectively. PCT (>/=0.5 ng/mL) and CSF protein (>/=0.5 g/L) were the best biologic tests, with 89% and 86% sensitivity rates, 89% and 78% specificity rates, adjusted odds ratios of 108 (95% CI, 15-772) and 34 (95% CI, 5-217), and areas under the ROC curves of 0.95 and 0.93, respectively. CONCLUSION: PCT and CSF protein had the best predictive value to distinguish between bacterial and aseptic meningitis in children.

RT-nested PCR for the detection of enterovirus in biological samples from patients with suspected enteroviral infections

In this study, we have tested a reverse transcription (RT) nested polymerase chain reaction (nPCR) for detection of enterovirus (EV) RNA in cerebrospinal fluid (CSF), serum samples, and conjunctival swabs (CS) from patients with suspected enterovirus infections. A specific 113-bp fragment was amplified using primers designed based on 5 non coding region of the enterovirus genome. The enterovirus RT-nPCR was able to detect 0.001 plaque forming unit (pfu)/ml. Since no PCR product was detected in each of the CSF, CS and serum samples from patients with proven-non-enterovirus viral infections, this method was found to be specific. EV RNA was detected in all 30 culture-confirmed CSF samples and yielded positive results in 5 out of 7 additional cases of culture-negative CSF samples with other evidences of enterovirus infection. Overall, EV RNA was detected in 95 of the patients with clinical diagnosis of viral central nervous system (CNS) disease and confirmed enterovirus infection. Furthermore, we were able to detect EV RNA in 24 (47) out of 51 CSF samples from patients with clinical diagnosis of viral CNS disease and negative laboratory evidence of viral infection. The percentage of positive EV RNA detection in paired CSF and serum samples from 11 patients with an enterovirus isolate in CSF was 100 (11 of 11) and 73 (8 of 11), respectively. In addition, EV-specific IgM was detected in 64 (7 of 11) of the sera tested. The method was also tested against 136 samples of CS from patients with clinical diagnosis of acute hemorrhagic conjunctivitis. Ninety nine of them resulted positive (73), while only 27 (20) had been positive for viral culture. In summary, our study shows the importance of enterovirus RT-nPCR for the diagnosis of enterovirus associated disease in different kind of biological samples and different types of diseases.(AU)

RT-nested PCR for the detection of enterovirus in biological samples from patients with suspected enteroviral infections

In this study, we have tested a reverse transcription (RT) nested polymerase chain reaction (nPCR) for detection of enterovirus (EV) RNA in cerebrospinal fluid (CSF), serum samples, and conjunctival swabs (CS) from patients with suspected enterovirus infections. A specific 113-bp fragment was amplified using primers designed based on 5' non coding region of the enterovirus genome. The enterovirus RT-nPCR was able to detect 0.001 plaque forming unit (pfu)/ml. Since no PCR product was detected in each of the CSF, CS and serum samples from patients with proven-non-enterovirus viral infections, this method was found to be specific. EV RNA was detected in all 30 culture-confirmed CSF samples and yielded positive results in 5 out of 7 additional cases of culture-negative CSF samples with other evidences of enterovirus infection. Overall, EV RNA was detected in 95 of the patients with clinical diagnosis of viral central nervous system (CNS) disease and confirmed enterovirus infection. Furthermore, we were able to detect EV RNA in 24 (47) out of 51 CSF samples from patients with clinical diagnosis of viral CNS disease and negative laboratory evidence of viral infection. The percentage of positive EV RNA detection in paired CSF and serum samples from 11 patients with an enterovirus isolate in CSF was 100 (11 of 11) and 73 (8 of 11), respectively. In addition, EV-specific IgM was detected in 64 (7 of 11) of the sera tested. The method was also tested against 136 samples of CS from patients with clinical diagnosis of acute hemorrhagic conjunctivitis. Ninety nine of them resulted positive (73), while only 27 (20) had been positive for viral culture. In summary, our study shows the importance of enterovirus RT-nPCR for the diagnosis of enterovirus associated disease in different kind of biological samples and different types of diseases.

RT-nested PCR for the detection of enterovirus in biological samples from patients with suspected enteroviral infections.

In this study, we have tested a reverse transcription (RT) nested polymerase chain reaction (nPCR) for detection of enterovirus (EV) RNA in cerebrospinal fluid (CSF), serum samples, and conjunctival swabs (CS) from patients with suspected enterovirus infections. A specific 113-bp fragment was amplified using primers designed based on 5' non coding region of the enterovirus genome. The enterovirus RT-nPCR was able to detect 0.001 plaque forming unit (pfu)/ml. Since no PCR product was detected in each of the CSF, CS and serum samples from patients with proven-non-enterovirus viral infections, this method was found to be specific. EV RNA was detected in all 30 culture-confirmed CSF samples and yielded positive results in 5 out of 7 additional cases of culture-negative CSF samples with other evidences of enterovirus infection. Overall, EV RNA was detected in 95% of the patients with clinical diagnosis of viral central nervous system (CNS) disease and confirmed enterovirus infection. Furthermore, we were able to detect EV RNA in 24 (47%) out of 51 CSF samples from patients with clinical diagnosis of viral CNS disease and negative laboratory evidence of viral infection. The percentage of positive EV RNA detection in paired CSF and serum samples from 11 patients with an enterovirus isolate in CSF was 100% (11 of 11) and 73% (8 of 11), respectively. In addition, EV-specific IgM was detected in 64% (7 of 11) of the sera tested. The method was also tested against 136 samples of CS from patients with clinical diagnosis of acute hemorrhagic conjunctivitis. Ninety nine of them resulted positive (73%), while only 27 (20%) had been positive for viral culture. In summary, our study shows the importance of enterovirus RT-nPCR for the diagnosis of enterovirus associated disease in different kind of biological samples and different types of diseases.

Variáveis clínicas e laboratoriais para diagnóstico diferencial entre meningites asséptica e piogênica em crianças / Laboratorial and clinical variables to the differential diagnosis between aseptic and pyogenic meningitis in children

Foram estudadas informaçöes obtidas do prontuário de 573 crianças com idade entre 1 mês e 15 anos e diagnóstico de meningite, internadas no Hospital Couto Maia, na cidade de Salvador-Bahia, no período de janeiro-1990 a dezembro-1992. Crises epilépticas, diminuiçäo do nível de consciência e rigidez de nuca foram mais frequentes no grupo com meningite bacteriana. Glicorraquia menor que 45 mg/dL, proteinorraquia igual ou superior a 140 mg/Dl e celularidade liquórica superior a 600 cels/mm3 mostraram-se preditores de meningite piogênica. A análise de curva ROC foi utilizada para estabelecer o melhor ponto de corte nas medidas liquóricas de celularidade, proteínas e glicose capaz de predizer meningite bacteriana. Os resultados encontrados enfatizam que informaçöes clínicas, obtidas com a anamnese e o exame neurológico, e liquóricas simples, definidas pelos níveis de proteína, glicose e análise da celularidade, podem ser utilizadas como medidas de grande acurácia para diferenciar meningite piogênica de asséptica em crianças.

[Laboratory and clinical variables in the differential diagnosis of aseptic and pyogenic meningitis in children]. / Variáveis clínicas e laboratoriais para o diagnóstico diferencial entre meningites Asséptica e piogênica em crianças.

We reviewed the charts of 573 children with a final diagnosis of pyogenic or aseptic meningitis, who were hospitalized in a large reference hospital for the treatment of infectious disease, from January 1990 to December 1992. Seizures, decreased consciousness, nuchal rigidity were more frequent in bacterial than in aseptic meningitis. A cerebrospinal fluid (CSF) glucose level lower than 45 mg/dL, a protein level equal or greater than 140 mg/dL and cell count greater than 600/mm3 were predictors of pyogenic meningitis. Receiver operating characteristic curve analysis was used to assess the best point in CSF measures of leukocytes, glucose and protein that could predict bacterial meningitis. These results suggest that clinicians should differentiate bacterial from aseptic meningitis in children with greater accuracy utilizing only clinical and simple CSF data.

Serum C-reactive protein in childhood meningitis in countries with limited laboratory resources: a Chilean experience.

Quantitative C-reactive protein (CRP) was determined sequentially by nephelometry and photometry from a finger prick serum sample in 67 children with bacterial meningitis (BM) and 16 children with aseptic meningitis (AM). The initial mean CRP value of 180 mg/liter in children with BM differed significantly from the 12 mg/liter found in those with AM (P less than 0.001). In BM a slow descent instead of rapid normalization or a secondary increase in sequential CRP values were early indicators of complications during recovery, such as resistance to the antibiotic. A significant difference in the mean CRP values between uneventful and complicated courses of BM was observed from the fourth day on (P less than 0.001). The measurements obtained with nephelometry correlated reliably with the more widely available photometry (r = 0.99). Easily performed rapid CRP determinations can considerably improve the quality of care in meningitis patients, especially in those situations where facilities for performing bacterial cultures or antibiotic susceptibility testing are not available.