Antibiotic resistance, bacterial transmission and improved prediction of bacterial infection in patients with antibody deficiency

Background

Antibody-deficient patients are at high risk of respiratory tract infections. Many therefore receive antibiotic prophylaxis and have access to antibiotics for self-administration in the event of breakthrough infections, which may increase antimicrobial resistance (AMR).

Objectives

To understand AMR in the respiratory tract of patients with antibody deficiency.

Methods

Sputum samples were collected from antibody-deficient patients in a cross-sectional and prospective study; bacteriology culture, 16S rRNA profiling and PCR detecting macrolide resistance genes were performed. Bacterial isolates were identified using MALDI-TOF, antimicrobial susceptibility was determined by disc diffusion and WGS of selected isolates was done using Illumina NextSeq with analysis for resistome and potential cross-transmission. Neutrophil elastase was measured by a ProteaseTag immunoassay.

Results

Three hundred and forty-three bacterial isolates from sputum of 43 patients were tested. Macrolide and tetracycline resistance were common (82% and 35% of isolates). erm(B) and mef(A) were the most frequent determinants of macrolide resistance. WGS revealed viridans streptococci as the source of AMR genes, of which 23% also carried conjugative plasmids linked with AMR genes and other mobile genetic elements. Phylogenetic analysis of Haemophilus influenzae isolates suggested possible transmission between patients attending clinic.In the prospective study, a negative correlation between sputum neutrophil elastase concentration and Shannon entropy α-diversity (Spearman's ρ = -0.306, P = 0.005) and a positive relationship with Berger-Parker dominance index (ρ = 0.502, P < 0.001) were found. Similar relationships were noted for the change in elastase concentration between consecutive samples, increases in elastase associating with reduced α-diversity.

Conclusions

Measures to limit antibiotic usage and spread of AMR should be implemented in immunodeficiency clinics. Sputum neutrophil elastase may be a useful marker to guide use of antibiotics for respiratory infection.

Evaluation of Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry for Molecular Typing of Acinetobacter baumannii in Comparison with Orthogonal Methods

Colonization and subsequent health care-associated infection (HCAI) with Acinetobacter baumannii are a concern for vulnerable patient groups within the hospital setting. Outbreaks involving multidrug-resistant strains are associated with increased patient morbidity and mortality and poorer overall outcomes. Reliable molecular typing methods can help to trace transmission routes and manage outbreaks. In addition to methods deployed by reference laboratories, matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) may assist by making initial in-house judgments on strain relatedness. However, limited studies on method reproducibility exist for this application. We applied MALDI-TOF MS typing to A. baumannii isolates associated with a nosocomial outbreak and evaluated different methods for data analysis. In addition, we compared MALDI-TOF MS with whole-genome sequencing (WGS) and Fourier transform infrared spectroscopy (FTIR) as orthogonal methods to further explore their resolution for bacterial strain typing. A related subgroup of isolates consistently clustered separately from the main outbreak group by all investigated methods. This finding, combined with epidemiological data from the outbreak, indicates that these methods identified a separate transmission event unrelated to the main outbreak. However, the MALDI-TOF MS upstream approach introduced measurement variability impacting method reproducibility and limiting its reliability as a standalone typing method. Availability of in-house typing methods with well-characterized sources of measurement uncertainty could assist with rapid and dependable confirmation (or denial) of suspected transmission events. This work highlights some of the steps to be improved before such tools can be fully integrated into routine diagnostic service workflows for strain typing. IMPORTANCE Managing the transmission of antimicrobial resistance necessitates reliable methods for tracking outbreaks. We compared the performance of MALDI-TOF MS with orthogonal approaches for strain typing, including WGS and FTIR, for Acinetobacter baumannii isolates correlated with a health care-associated infection (HCAI) event. Combined with epidemiological data, all methods investigated identified a group of isolates that were temporally and spatially linked to the outbreak, yet potentially attributed to a separate transmission event. This may have implications for guiding infection control strategies during an outbreak. However, the technical reproducibility of MALDI-TOF MS needs to be improved for it to be employed as a standalone typing method, as different stages of the experimental workflow introduced bias influencing interpretation of biomarker peak data. Availability of in-house methods for strain typing of bacteria could improve infection control practices following increased reports of outbreaks of antimicrobial-resistant organisms during the COVID-19 pandemic, related to sessional usage of personal protective equipment (PPE).