Bacterial colonization and succession in a newly opened hospital

Abstract

The microorganisms that inhabit hospitals may influence patient recovery and outcome, although the complexity and diversity of these bacterial communities can confound our ability to focus on potential pathogens in isolation. To develop a community-level understanding of how microorganisms colonize and move through the hospital environment, we characterized the bacterial dynamics among hospital surfaces, patients, and staff over the course of 1 year as a new hospital became operational. The bacteria in patient rooms, particularly on bedrails, consistently resembled the skin microbiota of the patient occupying the room. Bacterial communities on patients and room surfaces became increasingly similar over the course of a patient's stay. Temporal correlations in community structure demonstrated that patients initially acquired room-associated taxa that predated their stay but that their own microbial signatures began to influence the room community structure over time. The α- and β-diversity of patient skin samples were only weakly or nonsignificantly associated with clinical factors such as chemotherapy, antibiotic usage, and surgical recovery, and no factor except for ambulatory status affected microbial similarity between the microbiotas of a patient and their room. Metagenomic analyses revealed that genes conferring antimicrobial resistance were consistently more abundant on room surfaces than on the skin of the patients inhabiting those rooms. In addition, persistent unique genotypes of Staphylococcus and Propionibacterium were identified. Dynamic Bayesian network analysis suggested that hospital staff were more likely to be a source of bacteria on the skin of patients than the reverse but that there were no universal patterns of transmission across patient rooms.

Figure 1. Alpha and Beta Diversity of Hospital Sample Types

(A) Average alpha diversity of sample types based on Faith’s phylogenetic diversity (x-axis) and the Shannon diversity index (y-axis). Bars indicate standard error of the mean. (B) Heat map of beta diversity relationships between sample types based on the median weighted UniFrac distance between pairwise comparisons. Sample groups are clustered based on similarity in beta diversity patterns and median distances within individual sample types are highlighted in black along the diagonal.

Figure 2. Heat map of PC Space Correlations between Sample Types

Within- and between-environment comparisons are differentiated by color scheme.

Figure 3. Variability in Patient Room Microbiota

(A) Scatter plot of the percent of 16S reads in the “core microbiome” for the 8 rooms sampled weekly, with the core definition on the x-axis and the percent of reads in the core on the y-axis. Points represent the 8 individual rooms while the trend line is a moving average of the data. (B) Heat map of the predictive accuracy of SourceTracker models that used samples taken from the first day of a patient’s stay (source sample) to predict which patient a day 2 sample was taken from (sink sample).

Figure 4. Effect of Environmental Factors on Microbial Transmission

(A) Heat maps of the correlation between environmental factors and the weighted UniFrac distance between samples taken from the same room on the same day. (B) Seasonal change in the distances between the hands and noses of nurses working on the same floor. Trend lines are a smoothed moving average of the data. (C) Correlations between environmental factors and the distances of hand and nose samples for nurses working on the same floor on the same day. Color scheme is as in (A).