FilmArray Meningitis/Encephalitis (ME) panel in the diagnosis of bacterial meningitis

Biofire FilmArray Meningitis/Encephalitis panel for the aetiological diagnosis of central nervous system infections: A systematic review and diagnostic test accuracy meta-analysis

“…In the meta-analysis of "all bacteria" with reference test 1 (16 studies/6183 patients), 10,17−22,24−27,29−33 we obtained combined sensitivity and specificity of 89¢5% (95%CI 81¢1−94¢4), and 97¢4% (95%CI 94−98¢9), respectively. With reference test 2 (15 studies/5524 patients), 10,17,[29][30][31][32][33][18][19][20][21][22]24,26,27 z Reference standard item for reference test 1 for bacterial detection was considered as low risk only when it was clear that the cultures samples were taken before the antimicrobial treatment.…”

“…The 19 studies included 11,351 patients. 10 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 Four studies enrolled only children , 18 , 19 , 27 , 28 two included only adults, and the rest included both. Six studies 10 , 17 , 18 , 26 , 29 , 32 used a laboratory method to resolve the disagreements between FA/ME and reference test 1, and the rest used clinical information or CSF cytochemical parameters to resolve disagreements.…”

“…Four studies 10 , 17 , 31 , 33 were judged as unclear RoB in the patient selection domain due to uncertainty in the definition of suspected CNS infection. Five studies 20 , 21 , 23 , 33 , 34 were judged as high RoB in the index test criterion due to unclear handling of the sample, storage (recently collected, stored or frozen) or processing of the FA/ME. Six studies were classified as low RoB in the reference test domain, 10 , 17 , 18 , 22 , 26 , 29 and the remaining were judged as high RoB since it was not clear whether lumbar puncture and CSF sample collection occurred before antimicrobial therapy, and they did not use alternative molecular tests for bacteria.…”

“…Six studies were classified as low RoB in the reference test domain, 10 , 17 , 18 , 22 , 26 , 29 and the remaining were judged as high RoB since it was not clear whether lumbar puncture and CSF sample collection occurred before antimicrobial therapy, and they did not use alternative molecular tests for bacteria. Seventeen and two studies were judged as of low and unclear RoB, respectively, in the flow and timing domain 20 , 21 …”

“…x Chi2-test and corresponding p-value to assess presence of statistical heterogeneity. y Defined by the authors as >10 CSF cells in one study (21) and with no definition in the other study (24). z We defined it as studies with more than 70% of patients with previous antimicrobial therapy.…”