Predicting dialyzer fiber blocking is hard due to high intrapatient variability and limited utility of thrombin generation markers.

To date, no reliable blood-based or instrumental parameters are available to quantify dialyzer fiber blocking during hemodialysis. The only accurate method remains micro-computed tomography (µCT) imaging of the hemodialyzer post-dialysis, a labor-intensive process currently limited to research settings. Although several cross-sectional studies have employed µCT to evaluate hemodialyzer performance and anticoagulation strategies, intrapatient variability of this method has not been systematically examined. This study addressed that gap by enrolling ten chronic hemodialysis patients, each undergoing two series of three identical dialysis sessions: one with full anticoagulation and one with only 25% of the standard anticoagulation dose. µCT imaging after each session revealed important intrapatient variability in fiber blocking, comparable in magnitude to interpatient variability. These findings underscore the need for caution when interpreting cross-sectional assessments of dialyzer clotting and highlight the importance of accounting for intrapatient variability in the design and sample size estimation of future studies on clotting. Additionally, among the parameters derived from whole blood thrombin generation (WBTG) tests, only lag time and time to peak demonstrated correlations with fiber patency. However, due to their limited predictive accuracy at the individual level, even these promising markers fall short of providing clinically useful predictors for dialyzer fiber blocking.