Abstract
BACKGROUND: The optimal organisational model for delivering continuous renal replacement therapy (CRRT) to critically ill patients remains a subject of investigation. OBJECTIVE: This study compared the impact of a kidney intensive care unit (KICU) healthcare-led model with a traditional haemodialysis nurse-led model on key clinical process indicators and patient outcomes during CRRT. METHODS: In this retrospective observational study, patients requiring CRRT admitted to the KICU of a tertiary hospital between January 2022 and December 2023 were selected. Patients were divided into two groups based on the treatment leadership: the KICU group, where CRRT was led by intensive care unit supervising physicians and nurses certified in critical care blood purification techniques, and the control group, led by supervising physicians and haemodialysis nurses. Data were collected via the hospital information system and nursing records. After propensity score matching, 1180 patients were included in each group. Supplementary multivariable regression analyses were performed to provide adjusted effect estimates. Outcomes included treatment interruption rate, catheter-related infection rate, complication rate, actual delivered effluent dose, catheter insertion to treatment initiation time and patient satisfaction. RESULTS: The KICU group demonstrated a significantly lower overall complication rate (1.8% vs 3.7%, p = 0.004) and a higher rate of achieving the prescribed effluent dose (87.7% vs 81.2%, p < 0.001). Time to CRRT initiation was shorter in the KICU group (23.75 ± 2.43 vs 30.60 ± 6.36 minutes, p < 0.05). Patient satisfaction was also higher (p < 0.05). No significant differences were found in treatment interruption or catheter-related infections. The benefits of the KICU model on these key outcomes remained statistically significant in multivariable-adjusted analyses. CONCLUSION: Continuous renal replacement therapy management by a dedicated KICU multidisciplinary team was associated with a lower complication rate, improved delivery of the prescribed dialysis dose, faster treatment initiation and higher patient satisfaction compared with a traditional model. These findings suggest that a structured, healthcare-led model can optimise the CRRT process for critically ill patients. Further multicentre studies are warranted to confirm these results.