Optic pathway gliomas (OPG) are found in about 15% of patients with ne
urofibromatosis Type 1 (NF1). The natural history of OPG is not yet we
ll documented. Treatment in cases with growing tumors is still controv
ersial. Twenty-one patients with NF-1 and OPG, diagnosed over a 20-yea
r period, and followed neuroradiologically and ophthalmologically for
at least two years, were reevaluated. The diagnosis of OPG tvas made a
t a mean age of 7.1 years (range 0-14.5 years); six children were asym
ptomatic, 15 were symptomatic. The mean follow-up was 9.0 years (2.0-1
8.5 years). In eight initially operated or biopsied patients (three op
tic nerve and five chiasmal gliomas) tumor regrowth was found in one p
atient without progression on subsequent follow-up. Improvement of vis
ual acuity occurred in one child after operation of a large suprasella
r tumor and deterioration in one patient after biopsy of a chiasmal gl
ioma. The neuroradiological follow-up of the 13 not-operated and not-r
adiated patients (four optic nerve and nine chiasmal gliomas) was stab
le in 10, progressive in three, resulting in visual loss in one patien
t. In 11 children (52%) a second tumor outside the optic pathway was f
ound at a mean age of 4.0 years after the diagnosis of an OPG. Until n
ow they are mostly asymptomatic. Second site tumors were operated in t
wo children because of rapid tumor growth, one child died of a brainst
em tumor. OPG are a frequent complication in children with NF-1, appea
ring within the first decade. No definite correlation between the size
of tumor and the visual function could be shown. Deterioration of vis
ual function correlated more or less with growth of OPG but stable vis
ion did not exclude it. OPG, once detected, are often stable over year
s and only exceptionally lead to blindness after the diagnosis. Theref
ore our present attitude is a ''wait and see'' policy in patients with
stable vision, with follow-up of noninvasive investigations (NW and V
EP) at initially yearly intervals. In case of visual deterioration we
would consider treatment with carboplatin. We are concerned about the
large proportion of additional cerebral tumors in our series. They may
be responsible for a deleterious prognosis in some patients with NF-1
and OPG.