Multidrug-Resistant Klebsiella pneumoniae Strains in a Hospital: Phylogenetic Analysis to Investigate Local Epidemiology

Abstract

Multidrug-resistant Klebsiella pneumoniae is a significant healthcare challenge that particularly affects vulnerable patients through opportunistic nosocomial infections. Surveillance is crucial for monitoring the prevalence of these infections. Eighty-four KPC K. pneumoniae strains (2019-2022) were collected from patients admitted in Fondazione Policlinico Universitario Campus Bio-Medico. Strains were identified by MALDI-TOF and tested for antimicrobial susceptibility, and gene amplification was performed to identify the different blaKPC variants. Phylogenetic reconstructions were carried out using Bayesian methods. Additionally, to create a Bayesian skyline plot (BSP), additional analyses were conducted, running a simulation of 100 million generations under a Bayesian skyline model along with the uncorrelated log-normal relaxed clock model. To identify potential subgroups within genetic clusters and evaluate genetic variability among sequences, principal coordinate analysis (PCoA) was performed. In total, 84 Klebsiella pneumoniae isolates were classified as multidrug-resistant (MDR), characterized by resistance to three or more antibiotic classes, including carbapenems, and testing positive for KPC gene presence, and were included in the study. The Bayesian evolutionary tree for K. pneumoniae showed strongly supported branches but no genetic structure related to sampling dates or hospital departments. Phylogenetic analysis revealing a 73-year evolutionary span of K. pneumoniae strains. PCoA analysis identified three genetic outliers from 2022 and one from 2021, indicating higher genetic distances. The Bayesian skyline plot revealed increased genetic variability peaking at the end of 2019, followed by stabilization from early 2020 onward, with no significant changes in genetic variability thereafter. Overall, the study found no genetic structure correlating with sampling date or hospital department, suggesting significant variability in pathogen introduction during the pandemic. The increase in multidrug-resistant K. pneumoniae was linked to the influx of severe COVID-19 cases, prolonged hospitalizations, and heightened broad-spectrum antibiotic use, which likely facilitated resistance development and transmission amidst altered infection control practices.

Keywords: Klebsiella pneumoniae; MDR; epidemiology investigation; phylogenetic analysis.

Figure 1

Phylogenetic reconstruction of KPC Klebsiella pneumoniae strains. Time-scaled maximum clade credibility phylogenetic tree of n = 82 sequences. Node values are expressed in years and represent the divergence time of the related clade. All nodes are considered highly supported presenting values of posterior probabilities upper to 0.9.

Figure 2

Principal coordinate analysis of KPC Klebsiella pneumoniae strains. The cumulative variability explained by the first three axes amounts to 89.81% (axis 1: 60.42; axis 2: 21.61; axis 3: 7.78). (A) The groups were set a priori in accordance with the sampling years (2019 12 samples, 2020 19 samples, 2021 25 samples, and 2022 28 samples). (B) The groups were set a priori in accordance with the location of the department from which they were isolated (ground floor 40 samples, first floor 12 samples, second floor 5 samples, third floor 21 samples, fourth floor 4 samples, and another building 1 sample).

Figure 3

Bayesian skyline plot of KPC Klebsiella pneumoniae strains. The genetic variability and viral effective size (y-axis) are shown as a function of time (x-axis). The solid area represents the 95% high posterior density (HPD) region. Values of time are expressed in years before the most recent sample (i.e., 2022).