Molecular diagnosis of polymicrobial brain abscesses with 16S rDNA-based next generation sequencing

Abstract Objectives Brain abscesses lead to high mortality despite antibiotic and surgical treatment. Identification of causative bacteria is important to guide antibiotic therapy, but culture-based methods and molecular diagnostics by Sanger sequencing of 16S PCR products are hampered by antibiotic treatment and the often polymicrobial nature of brain abscesses. We have applied 16S rRNA-based next generation sequencing (NGS) for metagenomic analysis of intracranial (brain and epidural) abscess and meningitis samples. Methods 79 samples from 54 patients with intracranial abscesses or meningitis were included. DNA was subjected to 16S PCR. Amplicons were analyzed with the Illumina MiSeq system, sequence reads were blasted versus the NCBI 16S bacterial database and analyzed using MEGAN software. Results were compared to Gram-staining, culture and Sanger-sequencing. Results The NGS workflow was successful for 51 intracranial (46 brain and 5 epidural) abscess and 9 meningitis samples. Inclusion of (mono)-bacterial meningitis samples allowed to establish a cut-off criterion for exclusion of contaminating sequences. A total of 86 bacterial taxa were identified in brain abscesses by NGS, with Streptococcus intermedius and Fusobacterium nucleatum as most prevalent species, whereas Propionibacterium and Staphylococcus spp. were associated with epidural abscesses. NGS identified two or more bacterial taxa in 31/51 intracranial abscesses, revealing the polymicrobial nature of these infections and allowing to discriminate up to 16 bacterial taxa per sample. Conclusion These results extend earlier studies showing that NGS methods expand the spectrum of bacteria detected in brain abscesses and demonstrate that the MiSeq platform is suitable for metagenomic diagnostics of this severe infection.