Teardrop superior mesenteric vein: CT sign for unresectable carcinoma of the pancreas.

OBJECTIVE: Our objective was to investigate whether a tethered, teardrop-shaped superior mesenteric vein (SMV) is a reliable CT indicator of unresectable adenocarcinoma of the head of the pancreas. MATERIALS AND METHODS: CT scans of 92 patients with high suspicion for pancreatic head adenocarcinoma were retrospectively reviewed by two radiologists who were unfamiliar with the patients' outcomes. The reviewers were asked to assess whether the teardrop SMV sign was present or not; agreement was reached by consensus. Teardrop SMV was considered absent in patients with an obstructed vessel. RESULTS: Of 92 patients, 30 had a normal pancreas without a teardrop SMV. A mass in the head of the pancreas was seen in all 62 patients with cancer. Of these 62 patients, 30 (seven with teardrop SMV) were deemed to have inoperable disease by standard CT or clinical criteria. The remaining 32 patients underwent surgery; only 15 of these 32 had successful pancreatoduodenectomies. No patient with resectable tumor had an unequivocal teardrop SMV sign. In 17 patients (13 with teardrop SMV), resection of the tumor could not be accomplished because of vascular encasement (n = 12) or metastasis (n = 5). Added to conventional signs, teardrop SMV significantly increased CT's sensitivity (from 60% to 91%) and accuracy (from 79% to 95%) without significantly changing its specificity (from 100% to 98%) for resectability of pancreatic head cancer. CONCLUSION: The teardrop SMV is a reliable sign for predicting unresectability of adenocarcinoma of the head of the pancreas and can significantly contribute to preoperative planning.

A SPECT imaging study of MRI white matter hyperintensity in patients with degenerative dementia.

We investigated the correlation between cortical perfusion and white matter hyperintensities on magnetic resonance images (MRI) of patients with dementia. The study included 40 subjects, each of whom had undergone both MRI and single-photon emission computed tomography (SPECT) studies as part of their diagnostic evaluation for degenerative dementia. Two neuroradiologists rated the MRI films for severity of periventricular white matter changes on a 0-5 point scale and severity of subcortical white matter changes on a 0-4 point scale. Twelve regions of interest from association cortex were sampled for the semiquantitative analysis of SPECT images. No relationship was found between these global MRI ratings and semiquantitative or qualitative SPECT findings. Dementia severity as measured by the Mini-Mental State Examination and the Clinical Dementia Rating was significantly correlated with SPECT, whereas age was significantly correlated with MRI ratings, particularly in the periventricular regions. These data support the view that cortical SPECT abnormalities are not associated with global MRI abnormalities in the subcortical and periventricular regions of patients with a clinical picture of degenerative dementia.