Recirculation reassessed: the impact of blood flow rate and the low-flow method reevaluated.

The recognition that during hemodialysis the blood urea nitrogen level in a peripheral vein differs substantially from that in an artery has made invalid the conclusions of prior studies based on the standard (peripheral vein) means of calculating recirculation. The increase in recirculation associated with an increase in blood flow may be artifactual, as may the reported inaccuracy of the two-needle low blood flow technique for measuring recirculation. Using a stop-flow technique to measure recirculation, a study was performed to reassess these issues. Sixteen patients were each studied once within the first 30 minutes of a standard hemodialysis session (mean, 3.78 hour) at routine blood flow (mean, 419 mL/min) and at 200 mL/min. Dialyzer afferent and efferent blood line samples were obtained, as were afferent samples after sequential, rapid reductions in blood flow to 200, 50, and 0 mL/min. The final two samples were obtained within 30 seconds of reducing blood flow to less than 200 mL/min. Blood flow had a significant (P = 0.03) although clinically minor effect on recirculation (mean, 13.9% at 419 mL/min v 10.2% at 200 mL/min). The low blood flow technique yielded the same results as the stop flow technique (stop flow technique 13.9% v low blood flow technique 13.2% at 419 mL/min, P = 0.76). We conclude (1) that the increase in recirculation associated with increased blood flow is substantially less than has been reported previously and is minimal in a well-functioning access, and (2) that a modified low blood flow (50 mL/min) technique for measuring recirculation yields results equivalent to those obtained using the stop flow technique.