Antimicrobial resistance (AMR) represents a major threat to global health, with critically ill patients in inten- sive care units (ICUs) particularly vulnerable. In Kin- shasa, Democratic Republic of Congo, both pediatric and adult ICUs face a high prevalence of multidrug-resistant pathogens, including Escherichia coli, Klebsiella pneu- moniae, Staphylococcus aureus, extended-spectrum beta-lactamase (ESBL)-producing, and carbapenem-re- sistant strains. Limited laboratory capacity, weak antimi- crobial stewardship, widespread empirical antibiotic use, and suboptimal infection control exacerbate the problem, contributing to prolonged hospital stays, higher morbid- ity and mortality, and increased healthcare costs. Innova- tive approaches are urgently needed to improve detec- tion, management, and prevention of resistant infections in resource-limited settings. Artificial intelligence (AI) offers a transformative solution by integrating clinical, microbiological, and environmental data to predict re- sistant infections, guide individualized therapy, detect outbreaks in real time, and support stewardship programs. Evidence from pediatric ICUs shows reduc- tions in inappropriate antibiotic use and improved clinical outcomes, while predictive models in adult ICUs can an- ticipate sepsis and early mortality. Successful implemen- tation in Kinshasa requires adaptation to local epidemiol- ogy, staff training, high-quality data systems, and inte- gration into existing workflows. Combining AI with ro- bust antimicrobial stewardship and infection control measures could enhance patient care, optimize antibiotic use, and inform public health strategies, offering a sus- tainable approach to mitigating AMR in critically ill pop- ulations in low-resource environments.
Mayemba et al. (Sun,) studied this question.