Trends and drivers of multidrug-resistant bacteria incidence in 59 Chilean intensive care units, 2015-2024: a Bayesian hierarchical analysis.
retrospective_cohort · Level III
Where this comes from
- Record sourced from PubMed, PMID 42005089.
- Also identified by DOI 10.1016/j.lana.2026.101467 and PMC identifier 13089062.
- Licence recorded as CC BY.
- The licence permits redistribution, so the abstract is shown in full and the full text is available from the publisher.
Abstract
Antimicrobial resistance (AMR) is a major threat in intensive care units (ICUs). Evidence on determinants of multidrug-resistant (MDR) infections in ICUs remains limited. We aimed to assess temporal, institutional, and antibiotic-use drivers of MDR incidence across 59 Chilean ICUs across 40 hospitals (2015-2024). We conducted an ecological time-trend analysis using data from the Collaborative Group on Bacterial Resistance. MDR incidence density rates (IDRs) in 1000 patient-days comprised methicillin-resistant <i>Staphylococcus aureus</i> (MRSA), vancomycin-resistant Enterococcus spp. (VRE), extended-spectrum β-lactamase (ESBL)-producing <i>Klebsiella pneumoniae</i> and <i>Escherichia coli</i>, carbapenem-resistant Enterobacterales (CRE), <i>Pseudomonas aeruginosa</i> (CRPA), <i>Acinetobacter baumannii</i> (CRAB), and carbapenemase-producing Enterobacterales (CPE). IDRs were modelled using three-level Bayesian hierarchical regressions, accounting for repeated annual measures within hospital-pathogen pairs and differences between hospitals. Models included hospital infrastructure, infectious disease specialist hours, antimicrobial stewardship (AMS) programmes, socioeconomic variables, and antibiotic use (cephalosporins, quinolones, carbapenems; in DDDs/1000 bed-days). Between 2015 and 2024, MDR incidence declined by 21% (1.82-1.44 per 1000 patient-days), driven by reductions in CRPA (4.8-1.9), MRSA (3.2-1.0), and VRE (1.4-0.9). CRE declined modestly (2.7-1.7), while CPE increased from 0 to 1.3 after 2017. Adult ICUs and public hospitals had higher IDRs than paediatric and private units. In adjusted models, quinolone use was associated with higher MDR incidence (β = 0.08, 95% CI 0.03-0.14; p = 0.004), as was carbapenem use (β = 0.06, 0.03-0.09; p < 0.0001). Each additional hour of infectious disease specialist coverage per 100 bed-days reduced MDR incidence by ∼2% (β = -0.02, -0.03 to -0.01; p = 0.023). MRSA increased with quinolones, while CRE and CRPA increased with carbapenems. MDR incidence in Chilean ICUs remains high and driven by quinolone and carbapenem use. Strengthening AMS and specialist oversight, alongside stricter prescribing, could reduce burdens. No funding.