Abstract
BACKGROUND: The Korean National Healthcare-associated Infections Surveillance System (KONIS) monitors multidrug-resistant (MDR) bacterial infections in intensive care units (ICUs). However, simultaneously monitoring hundreds of ICUs remains challenging. Our study aimed to visualize the trends of MDR gram-negative bacterial infections in ICUs monitored by KONIS. METHODS: We evaluated KONIS data from 137 ICUs (2006-2011) and from 368 ICUs (2017-2022). Pneumonia, urinary tract infection, and bloodstream infection caused by Klebsiella pneumoniae, Pseudomonas aeruginosa, and Acinetobacter baumannii were analyzed. Transformation was employed to convert the infection rate graphs of each ICU into arrows. The length and angle of the arrows reflect changes in carbapenem susceptibility and infection rate, respectively. ICUs are categorized into red (rapid shift from susceptible to resistant bacteria and increased infection rate), yellow (slow shift from susceptible to resistant bacteria and decreased infections rate), and green (shift from resistant to susceptible bacteria) groups. The proportional changes in each ICU category were compared during the first and last five years of the study periods. RESULTS: For K. pneumoniae, the proportion of red category ICUs increased (0% to 17%, p-value 0.586), while the proportions of yellow (33.3% to 7%, p-value 0.288) and green category ICUs (66.6% to 36%, p-value 0.290) decreased. For P. aeruginosa, the proportions of red (12% to 27%, p-value 0.016) and green category ICUs (38% to 46%, p-value 0.358) increased, while the proportion of yellow category ICUs decreased (8% to 2%, p-value 0.043). For A. baumannii, the proportions of red (19% to 14%, p-value 0.649) and yellow category ICUs (5% to 1%, p-value 0.187) decreased, while the proportion of green category ICUs increased (19% to 72%, p-value <0.001). CONCLUSIONS: Graph transformation allowed the observation of MDR Gram-negative bacterial infection trends in ICUs. Further studies should aim to confirm whether our arrow indicators are useful for infection control and in identifying factors for reducing infections.