Kapoian T, Miller AJ, Sherman RA
Validation of a revised slow/stop flow recirculation method
Am Soc Nephrol
J Am Soc Nephrol (abstract) (Sep) 7:1409 1996

Although there is much discussion as to how to draw the postdialysis BUN sample in terms of rebound, all agree that a method must be used which eliminates the possibility that recirculated blood is being sampled. Otherwise, in patients with severe access recirculation (about 3-5% of the population, usually), the postdialysis BUN will be contamined with dialyzer outlet blood, and will underestimate the true postdialysis BUN. This can result in a severe overestimation of delivered Kt/V or URR.

In the past, to eliminate access recirculation (AR), the suggested strategy was to slow the pump to 50-100 ml/min for about 15-30 sec. Recirculation occurs usually when access flow (Qa) falls below the extracorporeal blood flow (Qb). Under such conditions, the blood pump begins to pull in blood from the downstream limb of the access. As a corollary, to stop recirculation, one needs to simply reduce Qb so it is less than Qa. A pump flow of 50-100 ml/min is adequate for this purpose. Blood samples are usually drawn from the sampling port, which almost always (in US blood lines) is about 8-10 ml downstream from the tip of the vascular access needle. To clear this deadspace, one should ideally pass double the volume of blood through the line (20 ml) before stopping the pump completely and obtaining the sample. Sherman concluded that, at a flow of 120 ml/min, or 2 ml/sec, 10 sec would be sufficient to clear the "deadspace" in the line of recirculated blood. Waiting longer than necessary is problematic, because the A/V urea gradient established during dialysis begins to close rapidly as soon as the blood pump is slowed. As such, an early increase in access urea level is seen within about 10-15 sec of slowing the blood pump.

Sherman et al used this 10 sec slow flow (120 ml/min) method in 50 patients, and also measured AR by ultrasound dilution (UD). Access flow was also measured. In 7/50 patients, AR by ultrasound dilution (UD) was 33%, Qa by UD averaged 254 ml/min, and AR by this new urea method was 27%. Since Qb was 500 ml/min, these patients by rights should have true AR. In the remaining 43 patients, AR by UD was 2%, Qa by UD averaged 870 ml/min, and recirculation by slow flow averaged 0%. It appeared that even with this method, due to variability in urea measurements, an AR of up to 10% by the urea method should be considered as a negative test.

Comment: This is a nice piece of work, especially as Qa was also measured. It suggests that even a 20 sec slow flow period may be too long in measuring recirculation, although the waiting period depends on the slow flow Qb and also on the deadspace volume. The waiting period would theoretically be less than 10 sec if one draws the sample from the line attached to the needle segment (deadspace only a few mls). It is very important to STOP THE PUMP after the slow flow period prior to sampling. This freezes the desired sample in the dialysis tubing, and prevents further changes in the composition of blood to be sampled during the delay associated with obtaining the sample. See the FAQ on HDCN regarding this topic for more information. (John T. Daugirdas, M.D., University of IL at Chicago)

To go back use the BACK button on your browser.
Otherwise click on the desired link to this article below:
Am Soc Nephrol
Basic hemodialysis : Vascular Access: graft/fistula