Comparison of volume of blood processed on haemodialysis adequacy measurement sessions vs regular non-adequacy sessions.

Background: Knowledge that adequacy measures such as the urea reduction ratio (URR) or Kt/V(urea) are being measured on haemodialysis may influence the behaviour of patients or staff such that the treatment may be better on those days. This study therefore tested the hypothesis that mean volume of blood processed (VBP), utilized as a surrogate for adequacy, is higher on adequacy measurement days than non-measurement days.

Methods: Patients were identified who had been on haemodialysis over the preceding 8 months. Primary outcome was the difference in the mean VBP (in litres) on URR measurement compared with non-URR measurement days (DeltaVBP(U)(-N)). Univariate and multivariate correlates of mean VBP and DeltaVBP(U)(-N) were also determined.

Results: Eighty-nine patients were identified who met inclusion and exclusion criteria. Linear regression demonstrated a weak relationship between VBP and URR (r=0.24, P<0.02). This relationship was much stronger when VBP was adjusted for patient weight (mean VBP/weight; r=0.78, P<0.0001). The overall mean VBP was 87.4 l (+/-1.2 l) and the average DeltaVBP(U)(-N) was 1.1 l (+/-0.3 l) (P=0.001). Twenty per cent of patients had a clinically relevant DeltaVBP(U)(-N) of >3.6 l. Patients with a graft or fistula had a significantly higher DeltaVBP(U)(-N) than patients with a tunnelled catheter.

Conclusions: This study demonstrates that the average VBP is less on non-URR than on URR measurement days; this difference was clinically important in >20% of patients. Univariate analysis indicated that the use of a fistula or graft correlated with a higher DeltaVBP(U)(-N). This implies that our current method of assessing dialysis adequacy does systematically overestimate the average delivered dose of dialysis in a subset of patients.