Children with sickle cell anemia (SS) have an increased risk for cerebral v
asculopathy with stroke (CVA) and cognitive impairment. The present study e
xamines the extent to which adding positron emission tomography (PET) to ma
gnetic resonance imaging (MRI) can improve the detection of cerebral vascul
opathy. Whereas MRI has been the prime modality for showing anatomical lesi
ons, PET excels at assessing the functional metabolic state through glucose
utilization 2-deoxy-2 [F-18] fluoro-D-glucose (FDG) and microvascular bloo
d flow ([O-15]H2O). Forty-nine SS children were studied. Among them, 19 had
clinically overt CVA, 20 had life-threatening hypoxic episodes or soft neu
rologic signs, and 10 were normal based on neurological history and examina
tion. For the entire sample of 49 subjects, 30 (61%) had abnormal MRI findi
ngs, 36 (73%) had abnormal PET findings, and 44 (90%) showed abnormalities
on either the MRI or the PET or both. Of the 19 subjects with overt CVA, 17
had abnormal MRI (89%), 17 had abnormal PET (89%), and 19 (100%) had eithe
r abnormal MRI or PET or both. Among the 20 subjects with soft neurologic s
igns, 10 (50%) had abnormal MRI, 13 (65%) had abnormal PET, and 17 (85%) ha
d abnormal MRI and/or PET. Six (60%) of the 10 neurologically normal subjec
ts had abnormal PET. Among the 30 subjects with no overt CVA, 25 (83%) demo
nstrated imaging abnormalities based on either MRI or PEI or both, thus, si
lent ischemia. Lower than average full-scale intelligence quotient (FSIQ) w
as associated with either overt CVA or silent ischemic lesions. Four subjec
ts who received chronic red blood cell transfusion showed improved metaboli
c and perfusion status on repeat PET scans. In conclusion, (1) the addition
of PET to MRI identified a much greater proportion of SS children with neu
roimaging abnormalities, particularly in those who had no history of overt
neurologic events. (2) PET lesions are more extensive, often bihemispheric,
as compared with MRI abnormalities. (3) PET may be useful in management as
a tool to evaluate metabolic improvement after therapeutic interventions,
and (4) the correlation of PET abnormalities to subsequent stroke or progre
ssive neurologic dysfunction requires further study. (C) 1999 by The Americ
an Society of Hematology.