cause mortality rates were not significantly different between
patients with residual urea clearance #0.24 ml/min and those
with residual urea clearance .0.24 ml/min (P= 0.24) [17], which
is not consistent with our results. There are a number of possible
explanations for this discrepancy.
First, it should be noted that the HEMO study only included
patients with residual urea clearance ,1.5 ml/min/35L of urea.
Because the impact of high-flux dialysis on mortality may be less
apparent in patients with greater residual renal function, the
exclusion of patients with greater residual renal function may be a
confounding factor in comparing the impact of high-flux dialysis
on mortality according to residual renal function.
Additionally, the HEMO study included HD patients in which
dialyzers were reused. Although the relationship between reuse of
the dialyzer and effectiveness of removal of middle molecules has
been controversial, reuse of dialyzers may be associated with
structural damage of the membrane and a reduced permeability to
middle molecules [18,19]. Therefore, reuse of dialyzer also may be
a confounder to determine the impact of dialyzer membrane flux
on mortality.
Another large randomized controlled trial, the MPO study,
showed that there was no significant difference in mortality
between high- and low-flux dialysis in the whole cohort [9]. In
subgroup analysis, the MPO study showed a survival benefit with
high-flux dialysis in patients with serum albumin level #4 mg/dl,
while there was no significant difference in mortality between
high- and low-flux dialysis in patients with serum albumin level .
4 mg/dl. Data on interaction between residual renal function and
the type of flux intervention with respect to all-cause mortality
were not shown in the MPO study. However, the MPO study
provides some interesting clues on the impact of residual renal
function on the relationship between high flux dialysis and
mortality. First, patients with serum albumin level #4 mg/dl in
the MPO study had longer duration of follow-up than the patients
with serum albumin level .4 mg/dl because the study protocol
was amended during the course of the study [9]. The longer
duration of follow-up in patients with serum albumin level #
4 mg/dl may explain the relationship between survival benefit
with high-flux dialysis and residual renal function. Long duration
of dialysis may cause accumulation of toxic middle molecules and
decrease residual renal function to remove them. Therefore, in the
MPO study, the patients with serum albumin level #4 mg/dl
could have benefited more from the removal of toxic middle
molecules with high-flux dialysis than the patients with serum
albumin level .4 mg/dl because of the longer follow up. Second,
the survival benefit of high-flux dialysis in the patients with serum
albumin level #4 mg/dl was evident only after about 12 months
of follow-up period, possibly when the residual renal function was
lost. These findings of the MPO study therefore support the results
of our study.
Our findings have a number of clinical implications. The
European Best Practice Guideline (EBPG) relating to dialyzer
membrane permeability recommends that the use of synthetic
high-flux membranes should be considered to delay long-term
complications of HD therapy [20]. The EBPG suggests the specific
indication for high-flux dialysis to reduce dialysis-related amyloidosis,
to improve control of hyperphosphatemia, to reduce the
increased cardiovascular risk, and to improve control of anemia
[20]. Our findings support the evidence for use of high-flux dialysis
membrane and further contribute to the indications established for
high-flux dialysis therapy in HD patients without residual renal
function.
Our study has several limitations. First, the design of our study
was not a randomized, controlled study but rather was a
prospective observational study. The prescription of the dialyzer
might be influenced by the results of previous study for the
membrane flux on mortality such as MPO study. Accordingly,
some baseline characteristics between the high-flux dialysis group
and low-flux dialysis group differed in our study, indicating
potential selection bias. In addition, the median follow-up period
of 31 months was relatively short. Finally, despite the multicenter
nature of the study, our cohort consisted of solely Korean patients.
Therefore, it is uncertain whether our results can be generalized to
other ethnic groups with HD treatment.
In conclusion, we found that HD using high-flux dialysis
membranes had survival benefit in patients with 24 h-residual
urine volume ,100 ml, but not in patients with 24 h-residual
urine volume $100 ml. These findings suggest that dialyzer
membrane flux impacts survival differently according to residual
renal function. Thus, high-flux dialysis rather than low-flux dialysis
might be considered in HD patients without residual renal
function.