The stage was set by Thomas A.
Depner MD, University of
California–Davis, Sacramento,
California, who summarized the controversy
as a “denominator dilemma” that has
far-reaching consequences in terms of
both how we envision uremic toxins being
generated and how dialysis should be
prescribed. To normalize the expression of
dialysis dose so that all patients are given
the same dose, some adjustment for body
size is logical. For native kidney function,
the glomerular filtration rate scales to surface
area or, more precisely, to the 0.75
power of body mass. This is consistent with
the universal scaling law that dictates the
scaling of several body functions and
anatomical structures in animals ranging
in size from mice to elephants, to the
power of 0.75. Patients maintained with
hemodialysis appear to receive more dialysis,
measured as K x t, if they are smaller.
For larger patients, understandably, it is
more difficult to provide high Kt/V, so they
receive, on average, less dialysis than
smaller patients when the dose is factored
for V or body weight. For these same
patients, however, size appears to have a
positive effect on patient outcome.
Therefore, when V is applied as a denominator
for scaling dose, the data are confusing.
Depner challenged the discussants,
Glen Chertow, MD, and Frank Gotch, MD,
to help clarify this dilemma so that practicing
nephrologists in the audience could
interpret data rationally and prescribe
hemodialysis more appropriately.
Glen Chertow, MD, University of
California, San Francisco, California,
agreed with the position that Kt/V is flawed
and Kt should not be scaled to V. He began
his discussion with an example of a patient
who weighed 112 kg and was dialyzed
“inadequately” by current standards,
despite 5 hours of dialysis, 3 days/week.
When asked to add a fourth day of dialysis,
the patient refused because he felt well.
Previous studies dating back to 1993
showed a reversed “J”-shape relationship
between mortality and Kt/V, meaning that
the highest doses appeared to be associated
with slightly worse outcome. It was also
clear from earlier data collected and interpreted
by Lowrie, that V is an independent
risk factor for mortality, larger patients
faring better than smaller patients. Dr.
Chertow’s explanation for the reversed “J”
phenomenon was that by dividing one
good parameter (K x t) by another factor
that also has a good effect on outcome (V),
we lose the significance of both. The
strongest correlate with outcome in Dr.
Chertow’s and Dr. Lowrie’s data was K x t,
which did not exhibit the reversed “J” rela-tionship
to mortality. In addition, when V
was added as a denominator, the correlation
with mortality diminished and
became insignificant in some patient
groups, such as African-American males.
Since it did not seem logical that different
thresholds for uremic toxicity should exist
for different patient sizes and different ethnic
origins, Dr. Chertow questioned the
validity of V as a denominator for dialysis
dosing, arguing that larger patients, like his
example patient, are probably better dia-lyzed
than our yardstick (Kt/V) indicates
and, conversely, that smaller patients are
probably underdialyzed.
Frank Gotch, MD, Medical Systems
Consultant, San Francisco, California, disagreed
with Dr. Chertow’s arguments and
began with a plea not to throw out the
baby with the bath water. He agreed that V
appears to be an independent risk factor
but criticized cross-sectional studies that
appear to eliminate the importance of V.
He noted that, even as far back as the
National Cooperative Dialysis Study
(NCDS) in the late 1970s and early 1980s, a
correlation of outcome with size was
observed. He cited recently published
USRDS data showing that Kt/V correlates
with mortality independent of V, and that
when a correction is made for the independent
effect of V, the correlation between
Kt/V and mortality is stronger. He agreed
that multiple factors confound the observational
outcome data and that the current
NIH-sponsored Hemodialysis (HEMO)
Study would likely shed more light on the
issue. Meanwhile, he cautioned nephrologists
to continue to factor dialysis doses by
V, but to be aware that there are independent
risks in smaller patients that require
attention to patient nutrition, and perhaps
other unknown factors that increase mortality
risk in this group.
