[Dynamic computed tomography in the characterization of focal hepatic lesions].
暂无分享,去创建一个
The main goal of our study was to test dynamic CT capability to characterize focal liver lesions. We examined 57 patients: 6 were affected with focal nodular hyperplasia (FNH), 19 with hepatocellular carcinoma (HCC), 1 with a regenerating nodule on cirrhosis; 14 patients had metastases, 3 focal fatty infiltration, 1 a necrotic nodule, 1 a non-Hodgkin's lymphoma, 1 a cysto-adeno-cholangiocarcinoma and 11 hemangiomas. All lesions were identified with US and the diagnosis was confirmed with the gold standard technique--that is, biopsy or surgery, and red blood cell SPECT for hemangiomas. All lesions were studied with a CT multiphase protocol consisting of a single-level dynamic phase followed by an incremental dynamic phase and finally by a delayed phase to study prolonged and delayed enhancement. Single-level dynamic bolus CT requires an injection of 60 ml nonionic contrast agent administered with a power injector into a cubital vein, at a rate of 5 ml/s. Scanning begins 10 seconds after the injection and consists of 6 series of 2 scans each; each scan lasts 2 seconds and is obtained during the same respiratory apnea, with a 5-second interscan pause. In this phase, 12 scans 5 mm thick are obtained, lasting 24 seconds in all, with pauses lasting 25 seconds--in all, 49 seconds. The next phase is the dynamic incremental scanning, to study the whole liver: this phase requires a 50-ml contrast agent injection at a rate of 4 ml/s, followed by 70 ml at a rate of 1 ml/sec, using 5 mm slice thickness and 8 mm scan interval. This results in 16 scans, beginning 20 seconds after the injection, with a scan time of 2 seconds and 4 seconds of interscan delay, 92 seconds in all. In the last phase, scanning begins 5 minutes after the injection, with a maximum delay of 10-15 minutes. Enhancement variations in both the lesions and the surroundings parenchyma, as related to time, were collected together with morphological data. Time density curves were grouped according to histologic classification and red blood cells SPECT findings; the curves were analyzed with the regression analysis. The results were obtained by analyzing a series of equations describing the different densities of the lesion and the surrounding parenchyma at fixed time intervals, integrated with morphological data, and then comparing the groups of lesions with each other. The regression analysis of the density curves and of the morphological data allowed us to correctly differentiate the 4 most frequent types of lesions--that is, hemangioma, HCC, FNH and metastasis--in 89% of the patients.