Association between ultrafiltration variability and clinical outcomes in patients undergoing hemodialysis

Our study seeks to access the impact of ultrafiltration volume (UFV) variability on clinical outcomes in patients undergoing maintenance hemodialysis (HD) using a population-based cohort. This study employed a retrospective design using data derived from the national HD quality assessment initiative in Republic of Korea, which was combined with health insurance claims records (<i>n</i> = 50,583). To assess UFV variability, a linear regression model was fitted for each individual across the six measurements, and the residual standard deviation from this model was calculated. Based on this metric, patients were stratified into four quartiles representing increasing levels of UFV variability (Q1 to Q4). The 5-year survival rates for patients in Q1, Q2, Q3, and Q4 were 68.3%, 67.9%, 66.4%, and 65.8%, respectively. Multivariable analysis revealed that the hazard ratio (HR) for all-cause mortality was the highest in the Q4 group. Additionally, a spline curve using the multivariable model indicated that an increase in UFV variability, based on a median of 0.44 L/session, was linked to all-cause mortality. Multivariable Cox regression indicated that the Q4 group had a higher HR for cardiovascular events or atrial fibrillation compared to the Q1 and Q2 groups. Additionally, the Q1 group had the lowest HR for dementia among the four groups. Our study demonstrated an association between high UFV variability and various clinical outcomes, particularly all-cause mortality and dementia. These findings suggest that UFV variability could serve as a useful supplementary indicator for predicting prognosis, in addition to UFV or ultrafiltration rate.