Normal anatomy and limitations in CT interpretation of lymph node disease

Normal lymph node size in the mediastinum: a retrospective study in two patient groups

The size and number of normal mediastinal lymph nodes seen on computed tomography were studied retrospectively in 110 patients from the Royal Marsden Hospital, Sutton, and St Bartholomew's Hospital, London. The mediastinum was divided into four areas. Lymph nodes were seen in 81% of patients. The number of patients with nodes in a specific area varied from 28% in the right superior mediastinum to 46% in the pretracheal retrocaval space. Seventeen per cent of the areas studied contained one node, 11% contained two or three nodes and 4% had three nodes. Fewer nodes were seen in patients with sparse mediastinal fat. Sixty-seven per cent of patients had nodes measuring 5 mm or less in maximum transverse diameter and 21% of patients had one or more nodes of 6-10 mm. Only a single calcified node measured more than 10 mm. We conclude that the majority of nodes are 5 mm or less in diameter and that almost all normal nodes measure 10 mm or less.

Predictive value of computerized tomography in the presurgical evaluation of primary carcinoma of the cervix

To determine the predictive value of abdominal-pelvic CT scan in assessing pelvic and paraaortic node metastases in untreated cervical carcinoma, the preoperative CT scans of 61 patients were compared with the gross and microscopic surgical findings. Although 75% of enlarged pelvic nodes on CT contained metastases, and 97% of patients with negative pelvic nodes had negative CT scans (specificity = 97%), histologically positive pelvic nodes were often missed on CT scan (sensitivity = 25%). For paraaortic nodes the CT scan had a specificity of 100% and a sensitivity of 67%. No paraaortic nodes were enlarged in stages I-IIA (0/51). CT scans are recommended for routine evaluation of paraaortic nodes in stages II-IV, but are not warranted in pretreatment evaluation of the pelvic nodes in stage I.

Lymphangiography, ultrasonography, and computed tomography in Hodgkin's disease and non-Hodgkin's lymphoma

Findings from histologic analysis, lymphangiography, ultrasonography, and computed tomography were reviewed for 54 cases of Hodgkin's disease and 18 cases of non-Hodgkin's lymphoma. All patients were classified as clinical stage 1 or 2 disease at the time of the imaging studies. The ultrasound and computed tomography studies identified only 30 to 40% of the truly positive patients. This low sensitivity contrasts with lymphangiography, which identified 95% of the truly positive Hodgkin's disease patients and 70% of the patients with abdominal spread of non-Hodgkin's lymphoma. Many errors in interpretation were attributable to location and distribution of disease in these patients and the differing patterns of spread in Hodgkin's disease and non-Hodgkin's lymphoma. The evidence indicates that when findings are positive on computed tomography or ultrasound, no other study is necessary. In stage 1 or 2 lymphoma, non-Hodgkin's patients with negative noninvasive findings and all Hodgkin's patients, regardless of their noninvasive findings, should undergo lymphangiography.

CT diagnosis of venous pseudotumors in the chest and abdomen

Appearances simulating neoplastic masses were found in five patients undergoing computed tomographic (CT) scanning of the chest and abdomen for a variety of indications. In each case the appearance was shown to be due to dilated collateral venous channels in association with portal hypertension or interruption of the inferior vena cava. The vascular origin of the masses was confirmed by CT scanning during intravenous bolus injection of contrast. Angiographic confirmation was obtained in three subjects. Enlarged collateral veins should be considered in the differential diagnosis of appearances suggesting mediastinal or abdominal lymphadenopathy at CT scanning.