The objective of therapeutic plasma exchange(1) is to remove plasma fr
om toxic substances, either autoantibody, alloantibody, immune complex
, monoclonal protein or toxin. Plasma exchange can also act through re
plenishment of a specific plasma factor. Pathophysiology was the first
rationale for the use of plasma exchange. Its first indication was in
deed the hyperviscosity syndrome,(1) on the single basis of pathogenes
is. However, pathogenesis alone does not convince physicians to use su
ch a costly and potentially unsafe treatment. Therefore, randomized cl
inical trials have been conducted for the last 15 years to ascertain t
he effectiveness and tolerance of plasma exchange, especially in neuro
logical diseases. In myasthenia gravis, the recommended use of plasma
exchange was mainly based on the theoretical argument of elimination o
f anti-acetylcholine receptor antibodies. However, even if clinical im
provement was observed after plasma exchange, this was also observed i
n myasthenia gravis associated with no circulating antibodies. In cont
rast, numerous randomized clinical trials have been initiated in the G
ullain-Barre syndrome, although the pathogenesis of the disease is unk
nown. Nevertheless, the large number of trials assessing plasma exchan
ge in the last 15 years, especially in neurological. diseases, explain
s the increased indications for plasma exchange in the national base o
f the French register of plasmapheresis.(2) The demonstrated clinical
benefit of plasma exchange through randomization may be an initial ste
p in the understanding of disease. Plasma exchange probably does not s
imply act through removal of toxic substances, but also, as high doses
of intravenous immune globulins,(3) through immunomodulation. This co
uld at least explain the competition and the complementarity of these
two treatments. The objective of this paper is to present the main ind
ications for plasma exchange in neurological diseases, complementing t
he conclusions of the consensus conference that met in 1996.(4) We foc
us on the assessments through randomized clinical trials, though uncon
trolled studies in rare diseases are also reported. Copyright (C) 1996
Elsevier Science Ltd.