The diagnostic value of nasal microbiota and clinical parameters in a multi-parametric prediction model to differentiate bacterial versus viral infections in lower respiratory tract infections

Abstract

Background The ability to accurately distinguish bacterial from viral infection would help clinicians better target antimicrobial therapy during suspected lower respiratory tract infections (LRTI). Although technological developments make it feasible to rapidly generate patient-specific microbiota profiles, evidence is required to show the clinical value of using microbiota data for infection diagnosis. In this study, we investigated whether adding nasal cavity microbiota profiles to readily available clinical information could improve machine learning classifiers to distinguish bacterial from viral infection in patients with LRTI. Results Various multi-parametric Random Forests classifiers were evaluated on the clinical and microbiota data of 293 LRTI patients for their prediction accuracies to differentiate bacterial from viral infection. The most predictive variable was C-reactive protein (CRP). We observed a marginal prediction improvement when 7 most prevalent nasal microbiota genera were added to the CRP model. In contrast, adding three clinical variables, absolute neutrophil count, consolidation on X-ray, and age group to the CRP model significantly improved the prediction. The best model correctly predicted 85% of the ‘bacterial’ patients and 82% of the ‘viral’ patients using 13 clinical and 3 nasal cavity microbiota genera (Staphylococcus, Moraxella, and Streptococcus). Conclusions We developed high-accuracy multi-parametric machine learning classifiers to differentiate bacterial from viral infections in LRTI patients of various ages. We demonstrated the predictive value of four easy-to-collect clinical variables which facilitate personalized and accurate clinical decision-making. We observed that nasal cavity microbiota correlate with the clinical variables and thus may not add significant value to diagnostic algorithms that aim to differentiate bacterial from viral infections.

Differentiating viral and bacterial infections: A machine learning model based on routine blood test values

“…Ramgopal et al derived an ML model to predict serious bacterial infections in young febrile infants [ 65 ]. Lien et al [ 66 ] have recently shown that an ML model using only complete blood count with differential leukocyte count data achieved similar performance as procalcitonin data in bacteremia prediction and Li et al [ 67 ], showed in 293 patients with suspected lower respiratory tract infections and/or sepsis that the most predictive variable was CRP. New experimental methods for viral/bacterial classification also build on ML methods; a nice example is the use of ML on infrared spectroscopy to diagnose inaccessible infections [ 68 , 69 ].…”

Bacterial Proteases as Potentially Exploitable Modulators of SARS-CoV-2 Infection: Logic from the Literature, Informatics, and Inspiration from the Dog

“…While the host PAB machinery alone provides factors that may influence trends in COVID-19 vulnerability and resistance, it is important to recall that human genetics are only one determinant of our pathophysiologically relevant proteome. A crucial component of human biochemical complexity and diversity emerges from microbial complementarity, with distinct microbiomes in upper respiratory [159][160][161][162][163], lower respiratory [115,[164][165][166][167], and gastrointestinal tracts [166][167][168][169][170] all having tangibly different effects on viral outcomes. Among the various biochemical effects associated with microbial proteases are known modulations of the innate immune system including IFNγ, several interleukins, and others [171], plus proteolytic stabilization of viral capsids in a manner that may bolster virion persistence outside of hosts in moist environments [172].…”