RESUMO
AIM: To evaluate the accuracy of water enema computed tomography (CT) for predicting the location of endometriosis in patients with contraindications to magnetic resonance imaging (MRI), focusing on rectosigmoid lesions and having laparoscopic and histological data as the reference standard. MATERIALS AND METHODS: Thirty-three women (mean age 33.4 ± 3.1 years) suspected of having deep pelvic endometriosis underwent 64-row CT and video laparoscopy within 4 weeks. Two radiologists blinded to the clinical data evaluated the CT images obtained after colonic retrograde distension using water as the contrast medium, and a comparison with laparoscopic and histological findings was performed. CT sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy were calculated. The radiation dose to patients was estimated. Cohen's weighted kappa (κ) test was used to evaluate the interobserver agreement. RESULTS: In 23 out of 33 patients (69%) intestinal implants were found at surgery and pathological examinations. CT confirmed the diagnosis of rectosigmoid endometriosis in 20 out of 23 implants. Three nodules located on the proximal sigmoid colon (two serosal lesions and one infiltrating the muscularis layer) with a diameter of less than 1 cm were not diagnosed. CT sensitivity, specificity, PPV, NPV, and accuracy values were 87, 100, 100, 77, and 91%, respectively. The mean effective dose estimate was 6.30 ± 1.7 mSv. Almost perfect agreement between the two readers was found (k = 0.84). CONCLUSION: Water enema CT can play a role in the diagnosis of bowel endometriosis and represents another accurate potential tool for video laparoscopic approaches, especially in patients for whom MRI is contraindicated.
Assuntos
Endometriose/diagnóstico , Doenças do Colo Sigmoide/diagnóstico , Adulto , Contraindicações , Enema/métodos , Feminino , Humanos , Laparoscopia/métodos , Imageamento por Ressonância Magnética , Variações Dependentes do Observador , Estudos Prospectivos , Sensibilidade e Especificidade , Tomografia Computadorizada por Raios X/métodos , Cirurgia Vídeoassistida/métodos , Água , Adulto JovemRESUMO
INTRODUCTION: We studied the CT patterns of abdominal carcinoid tumors. MATERIAL AND METHODS: Eight patients with carcinoid tumors were examined with CT. The tumors were in the ileum (2 patients), duodenum (1), colon (1), and mesentery (4). Pathologic confirmation was obtained at surgery in all patients. The symptoms were the carcinoid syndrome in 2 patients, abdominal pain in 4, jaundice in 1 patient and a right lower quadrant mass in 1. All patients were examined with(out) i.v. contrast agent administration; 6 patients received oral contrast material and 2 were submitted to water enema, to enhance visualization of the distal ileum and colon. RESULTS: CT identified the tumor in all patients but correctly defined its site in 7 cases only. CT showed a small mass in the distal common bile duct in 1 case, which postoperative histology diagnosed as a malignant carcinoid tumor of the duodenum. The CT findings of carcinoid tumors were a rounded mesenteric mass displacing bowel loops in 4 cases, an infiltrating colonic tumor in 1 case, an apparently intracholedochal mass in 1, a large necrotic mesenteric mass in 1 case. CT showed lymph node involvement in 2 cases, but no liver metastases were found. A carcinoid tumor was diagnosed in 4 patients based on the typical CT finding of a mesenteric mass with radiating soft-tissue density bands resulting in a stellate pattern. No preoperative diagnosis was possible in some other cases because the CT patterns were aspecific and mimicked those of other lesions, namely of tumor of the distal common bile duct, adenocarcinoma of the right colon, submucosal tumor of the distal ileum, large necrotic mesenteric mass. CONCLUSIONS: CT is a useful tool in the detection of carcinoid tumors. The correct diagnosis can be made based on the characteristic CT finding of a rounded mesenteric mass. In our experience, however, carcinoid tumors may present with other CT patterns mimicking those of other tumors.