| Objective As a new method of ultrasonic imaging, ultrasonic elast- ography(UE) is developing rapidly in recent years, which can provide a new method for noninvasive diagnosis of hepatic diseases. In our study, we use elasticity grade and strain ratio method on real-time UE for diagnosis of focal liver lesions, to explore the value of real-time UE in differential diagnosis of benign and malignant focal liver lesions.Methods From June 2010 to September 2010, in the First Affiliated Hospital of Guangxi Medical University, seventy-nine patients with 91 lesio- ns which were diagnosed as focal liver lesions by conventional ultrasonog- raphy or other imaging technology recruited in this study. All patients were examined by real-time UE before surgery or treatment. For each patient, liver sonography was performed with Hitachi HI-Vision Preirus ultrasound scanner which equipped with real-time tissue elastography(RTE) technology and quan- titative analysis software. When the patient in the supine or left lateral position, the liver was multi-sectionally scanned with conventional ultrasonography first until the target lesion was found, observed the lesion size, shape, border, internal echo and blood flow, then the maximum diameter and depth of the lesion were measured and the site of it was recorded. And then we switched to the UE mode, selected EUP-C532 convex-array probe, with transducer freq- uency 4-MHz to 8-MHz, we puted the probe on intercostal or costal margin and moved the region of interest(ROI) around the lesion, selected section as far as possible to avoid intrahepatic large vessels, ensured that the size of ROI was about 2 ~ 3 times the size of the lesion, if the lesion was too large to ensure this, took a part of the lesion in ROI. In double width condition, the two-dimen- sional image and the elastic image were observed, the probe was manipulated with relatively slight pressure to obtain images that were suitable for analysis, the pressure and compression frequency indicating bar was displayed as 2 to 3. When obtained stable elastic image, freezed and storaged the map, analyzed the map for lesions hardness grading according to the Itoh five-point scoring system first (Grade A: the entire lesion and its surrounding area were evenly shaded in green; Grade B: the lesion had a mosaic pattern of green and blue, mainly in green; Grade C: the lesion had a mosaic pattern of green and blue, mainly in blue; Grade D: the entire lesion was blue, but its surrounding area was not included; Grade E: the entire lesion and its surrounding area were blue), then started with a Strain Ratio measurement button of this instrument: selecting a representative region of interest from lesion and the average strain was expressed as A. A corresponding region of interest of adjacent liver tissue of the same depth was then selected and the average strain was expressed as B. The resultant strain ratio was calculated according to the equation strain ratio = B/A, which reflected the property of stiffness of the lesion. The pathology results or clinical diagnosis as the standard, using statistical software to analyse whether there were significant differences of elasticity grades and strain ratio between benign and malignant lesions. The optimal cut-off point of them for differentiating benign and malignant lesions were determined using the recei- ver operating characteristic(ROC) curve and the diagnostic accuracy of them were evaluated using area under the curve(AUC) respectively.Results (1) UE examination revealed: The compression deformation of benign lesions with soft texture was large, and that of malignant lesions with hard texture was small. Elasticity grade method showed that the elasticity grad- es of benign lesions group were statistically lower than that of malignant lesi- ons group (P<0.05). An elasticity grade of A~C was observed in benign lesi- ons group, while a grade of C~E in malignant lesions group. Strain ratio meth- od showed that the strain ratio of benign lesions group was statistically lower than that of malignant lesions group (P<0.05). The strain ratio range of benign lesions group was from 0.44 to 7.01, the median ratio was 1.75, and that of malignant lesions group was from 2.17 to 52.33 and 6.79 respectively. (2) The optimal cut-off point: Through the analysis of ROC curve, the optimal cut-off point of elasticity grade was grade C(Youden index = 0.465), and that of strain ratio was 3.47(Youden index = 0.759). Lesions whose grade≥C (C~E) were diagnosed as malignant, and whose grade |
PDF Full Text Request