Analysis of Factors Influencing Downtime During Continuous Renal Replacement Therapy in Critically Ill Patients and Construction of a Prediction Model

BACKGROUND: Prolonged CRRT downtime compromises treatment efficacy and worsens outcomes. Daily downtime > 20% exacerbates acidosis and is associated with 28-day mortality. AIMS: This study aimed to develop and validate a bedside-accessible nomogram for early identification of treatment days at risk of suboptimal CRRT downtime control in critically ill patients. STUDY DESIGN: This retrospective cohort study was conducted in one Chinese ICU (January 2019-December 2023). The binary outcome was cumulative downtime > 2.4 h per standardised treatment day. Variables included demographics, laboratory parameters, CRRT circuit and other treatment factors. GEE identified predictors; the model was displayed as a nomogram and assessed by AUC, calibration and decision curve analysis. RESULTS: A total of 145 patients contributing 595 CRRT treatment days were included. Total treatment time was 14 280 h and cumulative downtime was 1725.07 h. Mean daily downtime was 2.90 ± 2.72 h (12.08% ± 11.35% of treatment time). Suboptimal CRRT downtime control occurred on 232 treatment days (39.0%). Multivariable GEE analysis identified five independent risk factors for suboptimal downtime control: daily number of filter replacements (OR = 2.629, 95% CI 1.822-3.794, p < 0.001), agitation (OR = 2.331, 95% CI 1.041-5.218, p = 0.040), plasma exchange (OR = 4.654, 95% CI 1.042-20.782, p = 0.044), catheter dysfunction (OR = 10.528, 95% CI 2.939-37.710, p < 0.001) and out-of-unit transport for procedures (OR = 7.563, 95% CI 2.663-21.484, p < 0.001). The nomogram yielded AUCs of 0.789 (95% CI 0.726-0.841), 0.813 (95% CI 0.725-0.897) and 0.835 (95% CI 0.768-0.896) in the training, internal validation and external validation sets, respectively. CONCLUSIONS: Daily number of filter replacements, agitation, plasma exchange, catheter dysfunction and out-of-unit transport are independent risk factors for suboptimal CRRT downtime control. This nomogram showed satisfactory discrimination, calibration and net clinical benefit in internal and external validation, offering ICU nurses a rapid bedside risk assessment tool. RELEVANCE TO CLINICAL PRACTICE: ICU nurses can use this nomogram at CRRT initiation to identify high-risk patients and prioritise bedside procedures, sedation, catheter care and filter longevity to reduce downtime.