ACCESS RECIRCULATION

Normally blood flow through an AV access averages about 1L/min. The blood pump which normally
routes a portion of this flow through the dialyzer, usually is set to take a flow of 350-500 ml/min. Because
flow through the vascular access normally exceeds the demand of the blood pump, usually all of the blood
coming into the blood pump is coming from the access upstream to the needle insertion site. The urea
concentration of blood entering the dialyzer is the same as that in the upstream access, and there is no
access recirculation (assuming of course, that the access needle have not been placed too close to one
another, and that the arterial and venous needle positions havenot been inadverently reversed). In a failing
AV graft or fistula, flow through the access can decrease markedly, to 350-500 ml/min or slower. In such
circumstances, part of the flow leaving the dialyzer reverses flow through the access and reenters the
dialyzer. Then the dialyzer inlet blood becomes admixed, or “diluted”, with dialyzer oulet blood. This
phenomenon is called access recirculation.

A. Impact of access recirculation on dialysis adequacy. When access recirculation occurs, the
urea concentration in the blood entering the dialyzer may be reduced by 5%-40% or more. The
amount of urea removed in the dialyzer is equal to the volume of blood cleared × dialyzer inflow
urea concentration. Although dialyzer clearance remains unchanged, the amount of urea removed
is because the concentration of urea entering the dialyzer inlet is reduced. In patients with access
recirculation, if blood at the end of dialysis is drawn from the dialyzer inlet blood line, the urea
level in this blood will be lower than that in the patient’s upstream blood. Hence, the apparent
postdialysis SUN (serum urea nitrogen) will be artifactually low, and the URR (urea reduction
ratio) and, consequently, the spKt/V both will be overestimated.
B. Avoiding the impact of access recirculation on URR or spKt/V by slowing the blood flow or
by stopping dialysate flow at the end of dialysis prior to blood sampling. To ensure that bllod
being sampled reflects patient blood, one needs to slow the blood pump to a flow rate (e.g.,
100ml/min) that is assuredly below the access flow rate for a short period of time (10-20
seconds). Lowering the blood flow stops the backward flow of blood from the dialyzer outlet to
inlet and now all blood entering the arterial needle is upstream blood. The length of the slow-flow
period depends on the dead space between the tip of the arterial needle and the sampling port
(usually about 9ml in most adult blood lines). A 10-20 second period of 100ml/min flow should
be sufficient to allow the column of nonadmixed blood to reach the sampling port in most blood
lines. Postdialysis blood shoukd always be drawn after a short slow-flow period for this reason.
Merely stopping the blood pump prior to drawing the sample at the end of dialysis does not
prevent this problem, as the admixed blood in the inlet blood line is simply “frozen” in place. A
sample taken from the inlet blood line after stopping the pump still reflects admixed blood.
Another clever method of avoiding this problem is to just shut off the dialysate flow for 3 minutes
at the end of the dialysis (or put dialysate flow into bypass) while letting the blood flow go full
tilt. After 3 minutes, the SUN level in the blood leaving the dialyzer is similar to that going in,
and so the inlet SUN level now reflects the SUN level in the patient’s blood.

Reference: Hand book of Dialysis 5TH edition by John T. Daugirdas, Peter G. Blake and Todd S.
Ing.
This occurs when blood that has just been dialysed returns directly to the dialyser inlet. It is usually
caused by retrograde blood flow within a fistula or graft, when flow through AVF or AVG falls below
350-500 ml/min, or when venous blood is drawn up as arterial blood through a dual-lumen catheter.

Measured using a two-needle urea-based technique (i.e. using the dialysis needles only) but non-urea-
based dilution methods increasingly common. The three-needle peripheral vein method overestimates
recirculation unpredictably, and requires additional venepuncture (for arterial, venous, and peripheral
venous samples).

Recirculation >10% (urea method) or >5% (dilution method) requires further investigation (by
venography).

Recirculation does not occur unless access flow rate is less than the dialyser blood pump flow rate, and is
a marker of venous stenosis.

Two-needle measurement of recirculation

 Perform test after 30 min of dialysis with UF switched off.

 Take arterial (A) and venous (V) blood samples from the access lines.
 Reduce blood flow rates to 120 ml/min for 10s then switch off the pump.
 Clamp arterial line above sampling port and take systemic arterial sample (S) from the arterial
line.
 Resume dialysis.
 Measure urea in arterial, venous, and systemic sample (A,V, and S).
Recirculation=S-A/S-V × 100
This methos is less accurate than dilution techiques, but easy to perform and reliable. Dilution
methods are increasingly used though as more dialysis machines come equipped with appropriate
technology.

US dilution techniques for measurement of recirculation

Recirculation can be maeasured using US measurement of the dilution of saline (e.g., Transonic device)
or by temperature change,as for measuring flow but with blood lines not reversed. A bolus of saline
injected into the V line if there is significant recirculation.

Oxford Handbook of Dialysis 3RD Edition by Jeremy Levy, Julia Morgan and Edwina Brown.