Necrosis observed on CT enhancement is of prognostic value in soft tissue sarcoma.

Fifty-one patients with deep-seated soft tissue sarcoma of the extremities and trunk wall were examined with contrast-enhanced CT for presence of nonenhanced tumor areas (CT necrosis). After a median follow-up time of 3 years, 19 of the 41 patients with CT necrosis had developed metastases, compared to none of the 10 patients who had tumors without CT necrosis. Tumors with CT necrosis were larger than tumors without, but in tumors of similar size, absence of CT necrosis was a favorable prognostic sign.     

Preliminary evidence for a role of apolipoprotein E alleles in identifying haemodialysis patients at high vascular risk

Conventional risk factors have very low predictive power in identifying haemodialysis patients at high risk of vascular accidents. A role for apolipoprotein E isotypes was looked for in a small, but rigorously defined, cohort of longterm haemodialysis patients. In individuals with high vascular risk, as identified by higher common carotid intima/media thickness, we found an excess of apolipoprotein E4 alleles. This preliminary result requires confirmation in large patient cohorts.      

System for prostate brachytherapy and biopsy in a standard 1.5 T MRI scanner.

A technique for transperineal high-dose-rate (HDR) prostate brachytherapy and needle biopsy in a standard 1.5 T MRI scanner is demonstrated. In each of eight procedures (in four patients with intermediate to high risk localized prostate cancer), four MRI-guided transperineal prostate biopsies were obtained followed by placement of 14-15 hollow transperineal catheters for HDR brachytherapy. Mean needle-placement accuracy was 2.1 mm, 95% of needle-placement errors were less than 4.0 mm, and the maximum needle-placement error was 4.4 mm. In addition to guiding the placement of biopsy needles and brachytherapy catheters, MR images were also used for brachytherapy treatment planning and optimization. Because 1.5 T MR images are directly acquired during the interventional procedure, dependence on deformable registration is reduced and online image quality is maximized.     

Hodgkin disease: CT of the thymus

The computed tomography (CT) scans in two groups of patients with Hodgkin disease were reviewed to determine the frequency of thymic enlargement. In 50 CT scans from 50 patients with evidence of thoracic disease on CT scans who were examined for primary staging, the thymus was enlarged in 15 of 50 (30%). Fifty CT scans were obtained from 44 patients at the time of 50 separate episodes of known or suspected relapse. Relapse occurred in the mediastinum in 12 episodes, lung parenchyma in five, and both sites in one. Thymic enlargement thought to be due to involvement by disease was present in seven of 18 (38%). Mediastinal disease was associated with thymic enlargement in all but one patient in whom a thymic cyst developed after radiation therapy. Differentiation of thymic enlargement from enlarged superior mediastinal lymph nodes was easily made in all but two patients. Thymic enlargement in the absence of lymph node enlargement may indicate a different disease, since isolated Hodgkin disease of the thymus is uncommon. Primary thymic tumor should be considered initially, whereas after treatment, rebound hyperplasia of the thymus may be the cause of enlargement.     

Solitary bronchioloalveolar carcinoma: CT criteria

The computed tomographic (CT) scans of 30 patients with solitary bronchioloalveolar carcinoma were reviewed. Common features at CT included the peripheral or subpleural location of a pulmonary mass (25 cases), pseudocavitation (18 cases), heterogeneous attenuation (17 cases), irregular margins forming a star pattern (22 cases), and pleural tags (21 cases). Using these CT criteria, four independent observers attempted to identify cases of bronchioloalveolar carcinoma from a larger sample of lung cancers and benign lesions by categorizing a series of test cases into four probability categories. Although the bronchioloalveolar carcinomas were correctly ranked in the two highest probability categories 75% of the time (in 45 of 60 cases), there was considerable overlap with other lung lesions, particularly with adenocarcinoma and large cell undifferentiated carcinoma. However, even though the typical features of bronchioloalveolar carcinoma are not invariable or highly specific, they are characteristic enough to suggest the diagnosis.