Search

The various radiologic methods used in the detection of hepatic lesions in the oncologic patient are discussed. The advantages and disadvantages of each imaging technique as well as a new diagnostic approach are also discussed. With the rapid technological advances made in cross-sectional imaging, the authors believe that ultrasound and, whenever possible, computed tomography, should be used as the screening method.

A series of 16 patients with adrenal masses were biopsied percutaneously under computed tomography (CT) guidance with 18- to 22-gauge modified Chiba needles. Adrenal adenomas, cysts, metastases, melanoma, and adrenal hemorrhage were identified. Of nine oncologic patients, four had adrenal metastases, while five had other nonmalignant adrenal masses. Thus, an adrenal mass in an oncologic patient is not always metastases. No complications occurred. The diagnostic evaluation of an adrenal mass in selected cases should include CT-guided percutaneous aspiration as a safe and reliable alternative to open surgical biopsy. CT-guided biopsy can be performed as an outpatient procedure, avoiding the cost of hospitalization and the morbidity of surgery.

Results of percutaneous needle biopsies were evaluated retrospectively in 58 patients in whom a diagnosis of lymphoma was suspected. The biopsy specimen was diagnostic in 94% of the 36 patients with lymphoma, 20 of whom had recurrent disease and 16 of whom had newly diagnosed lymphoma. Sufficient tissue was obtained in 94% of these positive biopsy specimens to allow histologic subtyping of the lymphoma. Immunohistochemical studies performed on seven of the biopsy specimens allowed immunologic subclassification into B-cell and T-cell types of lymphoma. Our results suggest that the percutaneous needle biopsy is a useful and reliable tool in the diagnosis and classification of lymphoma.

The results of radionuclide (RN) liver scans, computed tomography (CT), and ultrasonography (US) were compared in 163 patients. Thirty-eight patients had all three studies, while ten were examined by CT and RN liver scans. One hundred fifteen patients had only US and RN studies. Radionuclide liver scanning demonstrated more false positive and negative studies than CT or US. Also, CT and US demonstrated more areas of metastasis during a single examination than RN liver scans. Ultrasonography displayed roughly the same accuracy of CT when a technically adequate examination was obtained. However, US was hampered by technically inadequate studies in 19% of 153 patients because of interfering intestinal gas. Computed tomography proved the most accurate and reliable modality in 48 patients.

The accuracy of chest computerized tomography (CT) in detecting malignant hilar and mediastinal involvement by squamous cell carcinoma of the lung is examined. The preoperative chest CT scans of 74 patients with pathologically proven squamous cell lung carcinoma were prospectively and retrospectively reviewed. Criteria for the diagnosis of malignant hilar involvement were nonvascular mass enlarging the hilum; local alteration of hilar contour; adenopathy greater than 1 cm; thickened posterior wall of the bronchus intermedius and distal upper lobe bronchi; and bronchial displacement, compression, and obstruction. Criteria for the diagnosis of malignant mediastinal involvement were confluence of tumor with the mediastinum, altered contour of the azygoesophageal recess, thickened posterior wall of the proximal main stem bronchi, and mediastinal adenopathy greater than 1 cm. Calcified hilar and mediastinal nodes were considered benign. Our results, corrected for reader error, were 92% sensitive, 92% specific, and 96% accurate in the hilum and 95% sensitive, 77% specific, and 82% accurate in the mediastinum. These data support a significant role for chest CT in the preoperative staging of non small cell lung carcinoma.

The radiologic approach to the pancreatic mass is discussed. The efficacy of the various diagnostic methods is also discussed. Cross-sectional imaging techniques have greatly improved the accuracy of diagnosis of pancreatic lesions and are less invasive than either angiography or ERCP. Computed tomography is more reliable than sonography and should be the initial diagnostic procedure in any patient who is suspected of having a pancreatic neoplasm.

One hundred and eight ultrasonographic examinations of the abdomen and pelvis in 89 patients with malignant melanoma were reviewed. Forty of these patients were found to have metastatic disease demonstrated by this new diagnostic modality. The varied ultrasonographic manifestations of melanoma are illustrated. Its use in staging and evaluating therapy is stressed.

A retrospective analysis was performed on 102 patients who were examined by both ultrasound (US) and computed tomography (CT) for known or possible carcinoma of the pancreas. In 38% of the patients, ultrasonography was unsatisfactory due to overlying interfering intestinal gas or ascites, whereas only 2% of CT studies were unsatisfactory due to technical considerations. In comparing the two modalities, CT was found to be more accurate (96% CT vs. 84% US), and this became more significant when nondiagnostic studies were considered in evaluating accuracy (95% CT vs. 54% US). Because of the findings in this analysis, CT is recommended as the initial diagnostic imaging modality for the evaluation of possible pancreatic neoplasm.