Prognostic Insights from Longitudinal Multicompartment Study of Host-Microbiota Interactions in Critically Ill Patients

Abstract

Critical illness can disrupt the composition and function of the microbiome, yet comprehensive longitudinal studies are lacking. We conducted a longitudinal analysis of oral, lung, and gut microbiota in a large cohort of 479 mechanically ventilated patients with acute respiratory failure. Progressive dysbiosis emerged in all three body compartments, characterized by reduced alpha diversity, depletion of obligate anaerobe bacteria, and pathogen enrichment. Clinical variables, including chronic obstructive pulmonary disease, immunosuppression, and antibiotic exposure, shaped dysbiosis. Notably, of the three body compartments, unsupervised clusters of lung microbiota diversity and composition independently predicted survival, transcending clinical predictors, organ dysfunction severity, and host-response sub-phenotypes. These independent associations of lung microbiota may serve as valuable biomarkers for prognostication and treatment decisions in critically ill patients. Insights into the dynamics of the microbiome during critical illness highlight the potential for microbiota-targeted interventions in precision medicine.

Keywords: biomarkers; critical illness; dysbiosis; microbiome; precision medicine.

Figure 1.. Ecological features of dysbiosis in three body compartments in critically ill patients.

A. Samples from critically ill patients had significantly lower alpha diversity (Shannon index) compared to corresponding healthy control samples in each compartment (p<0.001), with further decline of Shannon index over time in longitudinal samples (p<0.001). B. Baseline samples from critically ill patients had markedly significant differences in beta diversity from healthy controls (permutational analysis of variance [permanova] p-values <0.001). C-E. Taxonomic composition comparisons with the limma package showed high effect sizes and significance thresholds (threshold of log2-fold-change [logFC] of centered-log-transformed [CLR] abundances >1.5; Benjamini-Hochberg adjusted p-value<0.05) showed depletion for multiple commensal taxa in critically ill patients samples, with significant enrichment for Staphylococcus in oral and lung samples, and Anaerococcus and Enterococcus in gut samples (significant taxa shown in red in the volcano plots). F. Lung samples had lower bacterial burden compared to oral and gut samples by 16S qPCR (all p<0.001). G. Oral and lung samples had higher compositional similarity (Bray-Curtis indices) compared to lung and gut samples in the baseline and middle interval (p<0.001). H-I: Taxonomic comparisons between compartments revealed that no specific taxa were systematically different between oral and lung microbiota (H), whereas in gut-lung comparisons, lung communities were enriched for typical respiratory commensals (e.g. Rothia, Veillonella, Streptococcus) and gut communities for gut commensals (e.g. Bacteroides, Lachnoclostridium, Lachnospiraceae) (I).

Figure 2:. Longitudinal analysis of bacterial composition showed a progressive loss of obligate anaerobes in oral and lung communities as well as enrichment for recognized respiratory pathogens in all three compartments.

Top Panels (A-B): Relative abundance barplots for oral, lung and gut samples with classification of bacterial genera by oxygen requirement into obligate anaerobes (anaerobes), aerobes, facultative anaerobes, microaerophiles, genera of variable oxygen requirement and unclassifiable. Comparisons of relative abundance for the three main categories of bacteria (obligate anaerobes, aerobes and facultative anaerobes) by follow-up interval (baseline, middle and late). Data in boxplots (B) are represented as individual values with median values and interquartile range depicted by the boxplots with comparisons between intervals by non-parametric tests. Bottom Panels (C-D): Relative abundance barplots for oral, lung and gut (F) samples with classification of bacterial genera by plausible pathogenicity into oral commensals, recognized respiratory pathogens and “other” category. Comparisons of relative abundance for these categories of bacteria by follow-up interval (baseline, middle and late) in boxplots (D).

Figure 3:. Unsupervised clustering approaches revealed differences in bacterial alpha diversity and composition in three body compartments of critically ill patients.

Panels A-D demonstrate bacterial Dirichlet Multinomial Mixture (DMM) modeling results for each compartment separately. DMM clusters had significant differences in alpha diversity (A) and composition (obligate anaerobe abundance in shown in panel B and pathogen abundance shown in panel C), with cluster 3 in each compartment showing very low Shannon Index and enrichment for pathogens (Low-Diversity cluster). Oral and lung cluster assignments were strongly associated (Odds ratio for assignment to the Low-Diversity cluster: 9.74 (5.61–17.29), p<0.0001), whereas lung and gut cluster assignments were less strongly but significantly associated (panel D).

Figure 4:. Lung bacterial and bacterial-fungal clusters strongly predicted 60-day survival independent of clinical predictors, organ dysfunction severity and host-response subphenotypes.

A-C: Kaplan-Meier curves for 60-day survival from intubation stratified by oral (A), lung (B) and gut (C) bacterial DMM clusters. The Low-Diversity lung DMM cluster was independently predictive of worse survival (adjusted Hazard Ratio = 2.51 (1.26–4.98), p=0.008), following adjustment for age, sex, history of COPD, immunosuppression, severity of illness by sequential organ failure assessment (SOFA) scores and host-response subphenotypes. The Lung bacterial-fungal SNF cluster with high pathogen and C. albicans abundance (cluster 1) was independently predictive of worse survival (D), whereas the oral and gut bacterial-fungal SNF clusters (D, F) did not impact survival.

Figure 5:. Lung and Gut Microbiota Associations with COVID-19 Severity in Two Independent Cohorts.

A. Application of the dysbiosis index in lung (ETA) microbiota profiles in the UPMC-COVID cohort classified subjects in three clusters, with significant differences in Shannon index and bacterial load by 16S qPCR. B. The low diversity cluster in lung samples from UPMC-COVID subjects was significantly associated with higher plasma levels of sTNFR1 and Ang-2. C-D. Application of the dysbiosis index models in lung (sputum or ETA) and gut (stool) samples in the MGH-COVID cohort classified subjects in three clusters, with significant differences in Shannon index and anaerobe abundance between clusters. E-F: Cluster assignments in the MGH cohort were strongly associated with clinical severity for lung samples only. Membership in the Low-Diversity cluster in the lungs was associated with an odds ratio of 8.77 (1.75–61.74) for severe disease (black belt connecting the Low-Diversity cluster and Severe Disease perimetric zones in the chord diagram). Gut clusters were not significantly associated with clinical severity of COVID-19 pneumonia.