Computed tomography (CT) was performed in 38 patients with 41 benign cystic teratomas of the ovary and two patients with malignant transformation. CT depicted all tumors. The presence of fat in 40 of 43 cases (93%), tooth or calcification in 24 of 43 (56%), Rokitansky protuberance in 35 of 43 (81%), tufts of hair in 28 of 43 (65%), and a fat-fluid level in five of 43 (12%) allowed a definite diagnosis of ovarian cystic teratoma in 42 of 43 cases (98%). In the two cases of malignancy, single large (greater than 10 cm) plugs (with uptake of contrast medium in one) with a cauliflower appearance and an irregular border forming an obtuse angle with the inner wall of the cyst suggested malignant transformation. In three cases of benign cystic teratoma, a mucinous tumor (one benign, one borderline, one malignant) arising in the same ovary was seen at pathologic examination but was only diagnosed with the help of CT in two of three cases. Thickening of the tube was noted in two cases of torsion of the adnexa. CT findings were compared with findings at radiography of the abdomen and hysterosalpingography in 30 cases, ultrasound in 31, and magnetic resonance imaging in three. This study demonstrated that CT was the best procedure for imaging cystic teratomas of the ovary.

No abstract

Metastatic peritoneal implants were assessed preoperatively with computed tomography (CT) in 38 patients with ovarian tumors. In the 106 biopsy specimens of gross peritoneal implants and the 118 random biopsy specimens obtained from these patients, metastatic deposits were detected in 27 of 38 (71%) patients and in 104 biopsy sites. CT depicted metastatic lesions in 17 of 27 (63%) patients and in 63 of 104 (61%) biopsy sites. The three sites most commonly involved were the right subphrenic region, the greater omentum, and the pouch of Douglas. The usefulness of CT in detecting lesions depended mainly on the location of the implant and the presence of adjacent ascites, rather than on lesion size.

No abstract

A computed tomographic method was used to assess the pattern of abdominal fat distribution in normal males and females at different abdominal levels. The method permitted site specific calculations of total body volume (TA), total fat volume (TF), subcutaneous fat volume (SF), and intraabdominal fat volume (IF) in each computed tomography scan. The ratio of TF/TA, SF/TF and IF/TF were calculated for the L1, L3, and L5 vertebral levels. Regression analysis of IF versus SF, SF versus TF, IF versus TF, TF versus TA, and TF versus body mass index and age were calculated. A significant linear correlation between the measured variables TA, SF, IF, and TF and between TF and body mass index was found for virtually all correlations attempted at all scanned levels. Females had a higher total fat volume and greater percentage of subcutaneous fat at all levels. Males accumulated more fat intraabdominally than subcutaneously at the L1 and L3 levels. The male-female differences were greatest at L1 and the ratio SF/IF statistically significant at the L1 and L5 levels. Our results demonstrate that computed tomography can noninvasively quantify abdominal fat distribution at various sites. There is an inherent difference in abdominal fat distribution between males and females that is not related to weight. The distribution of body fat in males and females varies markedly from level to level.