ORIGINAL ARTICLE Cerebral blood flow in temporal lobe epilepsy: a partial volume correction study Giampiero Giovacchini & Robert Bonwetsch & Peter Herscovitch & Richard E. Carson & William H. Theodore Received: 5 February 2007 / Accepted: 25 May 2007 / Published online: 1 September 2007 # Springer-Verlag 2007 Abstract Purpose Previous studies in temporal lobe epilepsy (TLE) have shown that, owing to brain atrophy, positron emission tomography (PET) can overestimate deficits in measures of cerebral function such as glucose metabolism (CMR glu ) and neuroreceptor binding. The magnitude of this effect on cerebral blood flow (CBF) is unexplored. The aim of this study was to assess CBF deficits in TLE before and after magnetic resonance imaging-based partial volume correc- tion (PVC). Methods Absolute values of CBF for 21 TLE patients and nine controls were computed before and after PVC. In TLE patients, quantitative CMR glu measurements also were obtained. Results Before PVC, regional values of CBF were signif- icantly (p<0.05) lower in TLE patients than in controls in all regions, except the fusiform gyrus contralateral to the epileptic focus. After PVC, statistical significance was maintained in only four regions: ipsilateral inferior temporal cortex, bilateral insula and contralateral amygdala. There was no significant difference between patients and controls in CBF asymmetry indices (AIs) in any region before or after PVC. In TLE patients, AIs for CBF were significantly smaller than for CMR glu in middle and inferior temporal cortex, fusiform gyrus and hippocampus both before and after PVC. A significant positive relationship between disease duration and AIs for CMR glu , but not CBF, was detected in hippocampus and amygdala, before but not after PVC. Conclusion PVC should be used for PET CBF measure- ments in patients with TLE. Reduced blood flow, in contrast to glucose metabolism, is mainly due to structural changes. Keywords Temporal lobe epilepsy . Partial volume correction . Cerebral blood flow . Positron emission tomography . Cerebral glucose metabolism Introduction Positron emission tomography (PET) studies of cerebral metabolic rate for glucose (CMR glu ) and cerebral blood flow (CBF) have been used extensively for presurgical evaluation of temporal lobe epilepsy (TLE) patients refractory to medical treatment. Interictal CBF, as mea- sured by PET with [ 15 O]water or single-photon emission tomography (SPECT) flow tracers, appears to be less accurate for identification of epileptogenic foci than CMR glu measured with [ 18 F]fluorodeoxyglucose ([ 18 F] FDG). Reduced interictal CBF was detected in temporal regions in 30–40% of patients, compared with 70–90% for CMR glu ; deficits in perfusion were smaller than for glucose metabolism [1–10]. Some studies attributed more Eur J Nucl Med Mol Imaging (2007) 34:2066–2072 DOI 10.1007/s00259-007-0499-x G. Giovacchini (*) University Milano-Bicocca, Via Olgettina, 60, 20132 Milan, Italy e-mail: giovacchini.giampiero@hsr.it R. Bonwetsch : W. H. Theodore Clinical Epilepsy Section, National Institute of Neurological Diseases and Strokes, Bethesda, MD, USA P. Herscovitch PET Department, Clinical Center, National Institutes of Health, Bethesda, MD, USA R. E. Carson Yale PET Center, New Haven, CT, USA