Locatelli F et al for the Italian Cooperative Dialysis Study
Group
Effects of different membranes and dialysis technologies on
patient treatment tolerance and nutritional parameters
Kidney Int
(Oct) 50:1293-1302 1996

Whether or not dialysis with complement activating membranes has any
adverse effects is a controversial topic. In the chronic setting,
there are non-randomized data suggesting that dialysis against
unsubstituted cellulose is associated with a higher mortality risk.
Earlier this year, Parker et al published in KI the results of a
randomized trial suggesting that dry weight is better maintained, as
is serum albumin, when patients are assigned to a synthetic low flux
membrane vs. to unsubstituted cellulose (KI 49:551-556, 1996). The
present study by Locatelli and colleagues, which was a multicenter,
randomized trial, sought to identify a membrane effect on nutritional
outcomes and on intradialytic adverse events.
In 41 centers, patients were randomly allocated to either Cuprophan or
low flux polysulfone. In 30 centers where high flux dialysis was
done, random allocation was to 4 groups: Cuprophan, low flux
polysulfone, high flux polysulfone, or hemodiafiltration with high
flux polysulfone. 279 patients were enrolled in the two arm study,
and 205 in the 4-arm study. There was a low overall proportion of
patients with diabetic nephropathy (about 6%). Kt/V was about 1.3 in
each group.
To make a long story short, absolutely no differences could be found
in nutritional parameters over a 24 month follow up period. No
difference in body weight, albumin, serum transferrin, and no
difference in intradialytic hypotensive episodes. If anything, there
were trends for increasing body weight and and for fewer intradialytic
hypotensive episodes with Cuprophan. Predialysis beta-2
microglobulin levels were reduced in the high flux groups as expected.
Comment: The trial was of good size, multicenter and
randomized in nature, and appears to have been carefully carried out.
It suggests that the nutritional effects found in the Parker trial may
have been due to some fluke (e.g., a high asymmetric dropout rate),
and suggests that one should be very cautious before ascribing
mortality effects in non-randomized trials to the dialysis membrane,
where some associated but unquantified factor may be causing the
effect. Truly randomized trials in dialysis of this size are very
rare. This is a very important study. (John T. Daugirdas, M.D.,
University of Illinois at Chicago)