A Nine-Year Analysis of WHO Critical Priority Pathogens from the Tunisian AMR Surveillance System
Itani, D.; Philips, L. T.; Kotb Tolba, S.; Achour, W.; Smaoui, H.; Thabet, L.; Zribi, M.; Foster-Nyarko, E.; Holt, K. E.; Boutiba-Ben Boubaker, I.
Show abstract
BackgroundAntimicrobial resistance (AMR) surveillance is essential for quantifying and monitoring the burden of AMR among World Health Organization (WHO) priority pathogens. We analysed Tunisian AMR surveillance system (TARSS) data across five sentinel hospitals from 2014 to 2022. MethodsWe conducted a retrospective isolate-level analysis for Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Acinetobacter spp. Temporal, ward, and specimen associations were quantified using multivariable logistic regression models. Sex and age categories were explored in secondary models due to missingness. Temporal trends were assessed using Cochran-Armitage test, and co-resistance was summarised for third-generation cephalosporin and carbapenem phenotypes. We also evaluated temporal dynamics of 3GCR and CR profiles. ResultsA total of 35,525 E. coli, 14,325 K. pneumoniae, 9,679 P. aeruginosa, and 5,597 Acinetobacter spp. were reported to TARSS between 2014 and 2022. Mean annual MDR prevalence was high for Acinetobacter spp. (85.1%), moderate for K. pneumoniae (45.5%) and for P. aeruginosa (27.1%), and lower for E. coli (17.5%). Adjusted models indicated increased odds of resistance to several antibiotics, whereas E. coli showed decreased odds. Intensive care unit (ICU) and blood isolates were associated with higher odds of resistance in all pathogens. ConclusionThis nine-year multi-hospital analysis reveals a high prevalence of AMR across the four WHO priority pathogens, settings, and specimen types, with increasing resistance for some pathogen-antibiotic combinations. The higher odds of clinically important resistance amongst ICU and blood isolates support the use of ward-level antibiograms and stratified stewardship and infection prevention measures.
Matching journals
The top 10 journals account for 50% of the predicted probability mass.