Objective: Data is provided on the outcome of children with isolated c
ongenital hydrocephalus, hydrocephalus accompanied by spina bifida or
encephaloceles, and neonatal hydrocephalus. This information should he
lp those who have to inform parents of affected fetuses how hydrocepha
lus and accompanying defects are likely to affect their child in the l
ong term. Methods: One hundred sixty-one infants with hydrocephalus we
re followed up for 3-14 years. Congenital hydrocephalus was due to aqu
educt stenosis in 45, Arnold-Chiari malformation and myelomeningoceles
in 40, Dandy-Walker complexes in 8, and hydrocephalus and encephaloce
les in 6. Acquired hydrocephalus was due to intraventricular hemorrhag
es in 60 premature infants, and neonatal meningitis in 2. Ventriculo-p
eritoneal shunts were inserted. Mortality rates, head sizes, shunt rev
ision rates, physical defects, seizure rates, and types of schools att
ended were determined. Results: Only aqueduct stenosis caused head siz
es larger than normal after 3 years of age. No child with hydrocephalu
s and a myelomeningocele was grossly mentally retarded. Mental develop
ment of children with Dandy-Walker complexes and aqueduct stenosis was
the same; a similar proportion attended mainstream schools, schools f
or the educationally subnormal, and training centers. Children with en
cephaloceles had a poor prognosis compared with other categories; fewe
r attended mainstream schools, and more were mentally and physically h
andicapped. Intraventricular hemorrhages were the commonest cause of h
ydrocephalus; the outlook was not as favorable as that of aqueduct ste
nosis, Arnold-Chiari malformations, and Dandy-Walker complexes; they h
ad higher death rates, poorer mainstream school attendance, and higher
seizure rates. Conclusions: This series was highly selective; to be i
ncluded, children had to have survived gestation and birth, and be ref
erred by pediatricians. Even so, it provides information for parents s
eeking advice about similarly affected fetuses and neonates.