Effectiveness of an information-driven targeted isolation strategy in reducing healthcare-associated infections caused by multidrug-resistant gram-negative bacteria in the intensive care unit: A prospective uncontrolled before-after study

BACKGROUND: Cross-transmission risk of multidrug-resistant Gram-negative bacteria (MDR-GNB) is substantially higher in intensive care units (ICUs). Broad patient isolation is resource-intensive, prompting this study to assess whether a validated MDR-GNB healthcare-associated infection (HAI) risk calculator-guided targeted preemptive isolation could reduce cross-transmission. METHODS: A prospective uncontrolled before-after study was conducted across four adult ICUs in two tertiary hospitals in Urumqi, with two 3-month phases separated by a 2-week washout period. Standard precautions were applied in the control phase, while the intervention phase used the risk calculator for targeted preemptive isolation of high-risk patients. The primary outcome was ICU-acquired MDR-GNB infection (ICUAI) incidence. RESULTS: Among a total of 2,101 ICU admissions, 1091 patients met the inclusion criteria and were included in the analysis. The rate of MDR-GNB healthcare-associated infection (HAI) was significantly lower during the intervention periods (2 out of 435, 0.46%) than during the control periods (13 out of 656, 1.98%; p = 0.035). CONCLUSION: Despite the relatively short duration of this trial, the use of an MDR-GNB HAI risk calculator to guide the preemptive isolation of high-risk patients was both feasible and effective in reducing transmission. TRIAL REGISTRATION NO: ChiCTR2500113287, retrospectively registered.