A new observational study in France has concluded that while some types of antibiotic-resistant bacteria spread by contact between infected patients and visitors, other mechanisms may be just as important in the spread of certain other common bacteria. The authors suggest that more than just strict hand hygiene is needed to contain these organisms.
The study was published in the journal PLOS Computational Biology on the 30th May 2019.
The emergence of bacteria which are resistant to multiple drugs is becoming an ever-increasing threat for patients treated in hospitals and other healthcare facilities. Such bacteria often produce enzymes called extended-spectrum beta-lactamases (ESBL) which make them invulnerable to a broad range of antibiotics. To prevent and contain the spread of such strains, we must know more about how these are spread from one patient to another.
The current study made use of wearable sensors to track contacts, called Close Proximity Interactions (CPIs), between 329 patients for eight weeks over a four-month period. The sensors were handed out to hundreds of patients and healthcare staff in the hospital. They were fitted with RFID tags to help detect contact patterns.
Patient interactions at distances below 1.5 meters were captured at intervals of 30 seconds. All patients were also screened weekly for the presence of ESBL-producing Klebsiella pneumoniae (K. pneumoniae) and Escherichia coli (E. coli).
Klebsiella pneumoniae (K. pneumoniae) and Escherichia coli (E. coli) are high-priority organisms for new antibiotic development. They cause one- fifth of all infections transmitted during healthcare visits or stays, and over half of all urinary tract infections associated with health care.
In fact, the latter bacterium causes a fifth of all healthcare-associated infections by itself, according to a WHO study. The presence of blood-borne infection with an ESBL-producing strain increases the risk of death by 63% compared to non-ESBL-producing strains.
What did the researchers find?
In the current study, the highest number of existing and new infections with ESBL E. coli and ESBL K. pneumoniae were found in the geriatric and neurologic wards, respectively. No transmission candidate could be traced in just under half of infections. The remaining episodes of new infection during the study period were analyzed for contact patterns.
Most ESBL K. pneumoniae infections were found in a single ward while resistant E. coli quickly spread through the facility. The latter was imported 8 times more frequently compared to the former and showed multiple resistance profiles.
The scientists suggest that this indicates a community origin for most of these ESBL E. coli infections. This could also explain why no transmission candidate was found in the contact network in several cases since several patients might have been colonized in the community before arriving at the hospital.
Other explanations for not being able to find contacts through the sensor network include antibiotic exposure unmasking latent infection, or plasmid-mediated acquisition of infection with E. coli which had previously acquired resistance to antibiotics.
Not all infections are spread by human contact
About 90% of ESBL-producing K. pneumoniae was spread to uninfected patients by direct or indirect contact with the infected patient, but only 54% of ESBL-positive E. coli. Thus, enforcing strict handwashing guidelines could help prevent transmission in 9 out of 10 ESBL K. pneumoniae infections.
However, other steps, like using disinfectants on potentially contaminated environmental surfaces surrounding infected patients, or using the right type of antibiotics, would be required to keep ESBL K. pneumonia from spreading. These findings agree with previous research which shows that ESBL K. pneumoniae spreads 3 times as easily as ESBL E. coli.
E. coli is a heterogeneous organism which has several strains capable of rapid spread within a community. This makes long-term care facilities a potential reservoir for multidrug-resistant strains and a potent means to disseminate these organisms through healthcare facilities and the community at large.
The study also demonstrates that wearable sensors can be used to understand how multidrug resistant bacterial strains spread.
Of course, the study had its limitations, as isolates were classified based only on phenotypic resistance patterns. False positives are possible with the use of CPIs as the standard of interactions.