Impaired delivery of hemodialysis prescriptions: an analysis of causes and an approach to evaluation.

Serial kinetic modeling is commonly used in hemodialysis to assess the adequacy of dialysis. A variety of problems lead to declining Kt/V in previously stable patients. These include noncompliance, vascular access recirculation, and dialyzer dysfunction. The purpose of this study was to find the relative frequencies of these problems in a group of patients undergoing routine hemodialysis. Simultaneous urea kinetic modeling and access recirculation were tested during 3 consecutive months. The baseline Kt/V was defined as the average of each patient's Kt/V values obtained during the previous 4 mo. A clinically important fall in Kt/V was defined as a decline of > or =0.2 if the baseline Kt/V was > or =1.2, or a decline of > or =0.1 if the baseline Kt/V was <1.2. Ninety-three of 375 (25%) sessions met the criteria for a significant decline in urea kinetic modeling. The baseline Kt/V in this group was 1.33 +/- 0.20 (mean +/- SEM) and declined to 1.02 +/- 0.18 in the abnormal month (P < 0.05). In 42% of instances with a decline of Kt/V, reduced blood processing due to a lower blood flow or shorter time than prescribed was responsible. Recirculation of >12% was found in 25% of sessions with a decrease in Kt/V. These patients most often had access dysfunction or reversed needles. The remaining one-third of patients with decreases in Kt/V had no problem identified, and subsequent monthly kinetic modeling results returned to baseline. These results suggest that analysis of falling urea kinetic modeling results should include a careful review of the dialysis record for reductions in prescribed time or blood flow rates followed by vascular access testing. If these evaluations are unrevealing, urea kinetic modeling results usually return to baseline in the next month.