Microbiome on the Bone-Anchored Hearing System: A Prospective Study

Abstract

The bone-anchored hearing system (BAHS) has evolved to a common treatment option for various types of hearing revalidation. The BAHS consists of an implant in the skull that breeches the skin. Soft tissue reactions are a common complication associated with BAHS and are generally poorly understood. This study aims to investigate the influence of BAHS and associated skin reactions around the implant. A total of 45 patients were prospectively followed from implantation up to at least 1 year. Swabs were obtained at baseline, 12 weeks follow-up and during cases of inflammation (Holgers score ≥2). The microbiota was assessed using IS-proTM, a bacterial profiling method based on the interspace region between the 16S–23S rRNA genes. Detection of operational taxonomic units, the Shannon Diversity Index, sample similarity analyses and Partial Least Squares Discriminant Analysis (PLS-DA) were employed. Staphylococcus epidermidis, Streptococcus pneumoniae/mitis, Propionibacterium acnes, Staphylococcus capitis, Staphylococcus hominis, Bifidobacterium longum, Haemophilus parainfluenzae, Lactobacillus rhamnosus, Bordetella spp., Streptococcus sanguinis, Peptostreptococcus anaerobius, Staphylococcus aureus, Lactococcus lactis, Enterobacter cloacae, and Citrobacter koseri were the most commonly found bacterial species. S. pneumoniae/mitis was significantly more often observed after implantation, whereas P. acnes was significantly less observed after implantation compared with baseline. The relative abundance of S. epidermidis (17%) and S. aureus (19.4%) was the highest for the group of patients with inflammation. The Shannon Diversity Index was significantly increased after implantation compared with pre-surgical swabs for Firmicutes, Actinobacteria, Fusobacteria, Verrucomicrobia (FAFV), but not for other phyla. When combining all phyla, there was no significant increase in the Shannon Diversity Index. The diversity index was similar post-surgically for patients experiencing inflammation and for patients without inflammation. With a supervised classifier (PLS-DA), patients prone to inflammation could be identified at baseline with an accuracy of 91.7%. In addition, PLS-DA could classify post-surgical abutments as non-inflamed or inflamed with an accuracy of 97.7%. This study shows the potential of using IS-proTM to describe and quantify the microbiota associated with the percutaneous BAHS. Furthermore, the results indicate the possibility of an early identification of patients susceptible to adverse skin reaction following implantation. Both S. aureus and S. epidermidis should be considered as relevant bacteria for BAHS-associated inflammation.