3207jum_online.qxp:Layout 1
6/19/13
2:08 PM
Page 1157
ORIGINAL RESEARCH
Evaluation of the Vascular Architecture of Focal Liver Lesions Using Micro Flow Imaging Hong Yang, MD, PhD, Guang-Jian Liu, MD, Ming-De Lu, MD, DMSc, Hui-Xiong Xu, MD, PhD, Xiao-Yan Xie, MD, PhD Objectives—To identify the vascular architecture of focal liver lesions using micro flow imaging and compare it with characteristics on contrast harmonic imaging during the arterial phase. Methods—Micro flow imaging and contrast harmonic imaging were performed in 118 patients with various focal liver lesions: hepatocellular carcinoma (n = 70), metastasis (n = 19), intrahepatic cholangiocarcinoma (n = 3), lymphoma (n = 1), hemangioma (n = 17), and focal nodular hyperplasia (n = 8). The vascular architecture of the lesions on micro flow imaging was evaluated by 2 investigators independently to reveal 6 patterns (types I– VI). Enhancement characteristics on contrast harmonic imaging were also evaluated.
Received August 27, 2012, from the Department of Medical Ultrasonics, First Affiliated Hospital of Guangxi Medical University, Guangxi, China (H.Y.); Institute of Diagnostic and Interventional Ultrasound, Sun Yat-Sen University, Guangzhou, China (G.-J.L., M.-D.L., H.-X.X., X.-Y.X.); and Departments of Medical Ultrasonics (G.-J.L., H.X.X., X.-Y.X.) and Hepatobiliary Surgery (M.D.L.), First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China. Revision requested September 19, 2012. Revised manuscript accepted for publication November 12, 2012. Address correspondence to Ming-De Lu, MD, DMSc, Institute of Diagnostic and Interventional Ultrasound, Sun Yat-Sen University, Department of Hepatobiliary Surgery, First Affiliated Hospital of Sun Yat-Sen University, 58 Zhongshan Road, 2, 510080 Guangzhou, China. E-mail:
[email protected] Abbreviations
CT, computed tomography; MI, mechanical index; MRI, magnetic resonance imaging doi:10.7863/ultra.32.7.1157
Results—Inter-reader agreement for delineating the vascular architecture was higher on contrast harmonic imaging (κ = 0.856) than micro flow imaging (κ = 0.613). On micro flow imaging, the vascular patterns of hepatocellular carcinomas were types I (28.6%), II (65.7%), and III (5.7%). On contrast harmonic imaging, 44 of 70 (62.9%) hepatocellular carcinomas showed chaotic vessels, of which 40 were type II and 4 were type II. The vascular patterns of metastases were types IV (78.9%), I (10.5%), and II (10.5%). Typical rim enhancement was identified in 57.9% of metastases on contrast harmonic imaging, and all were type IV. The vascular patterns of focal nodular hyperplasia were types VI (87.5%) and I (12.5%). Typical spoked wheel arteries were identified on contrast harmonic imaging in 2 focal nodular hyperplasia cases. The vascular patterns of hemangiomas were types V (94.1%) and II (5.9%). Typical peripheral nodular enhancement was identified in 88.2% of hemangiomas on contrast harmonic imaging, and all were type V. The χ2 test revealed that differences in vascular architecture between the lesions were significant on micro flow imaging (P < .001). Conclusions—Micro flow imaging permitted detailed delineation of the vascular architecture of focal liver lesions. Hepatocellular carcinoma, metastasis, focal nodular hyperplasia, and hemangioma showed characteristic vascular architecture. Key Words—contrast-enhanced sonography; focal liver lesions; micro flow imaging; vascular architecture
W
ith the development of sonographic contrast agents and contrast-specific imaging techniques, contrast-enhanced sonography has overcome the limitations of color and power Doppler imaging for displaying the parenchymal microvasculature.1 Contrast harmonic imaging is a nonlinear contrastenhanced sonographic mode that uses low–mechanical index (MI) imaging to minimize contrast microbubble disruption and prolong
©2013 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2013; 32:1157–1171 | 0278-4297 | www.aium.org
3207jum_online.qxp:Layout 1
6/19/13
2:08 PM
Page 1158
Yang et al—Vascular Architecture of Focal Liver Lesions on Micro Flow Imaging
contrast enhancement.2 Depending on the contrast agent and nonlinear harmonic imaging mode, dynamic focal liver lesion enhancement patterns are visualized during 3 continuous vascular phases.1 Intranodular vascular morphologic characteristics are also identified in some focal liver lesions during the early arterial phase and can help diagnosis, such as spoked wheel arteries for diagnosis of focal nodular hyperplasia and chaotic vessels for diagnosis of malignancy. Micro flow imaging is a flash-replenishment technique that uses high-power (high-MI) flash pulses to destroy contrast microbubbles, followed by low-power (low-MI) contrast-enhanced sonography to show contrast replenishment in vessels and tissues.3 A composite image depicting the vascular architecture is constructed by maximumintensity capture of temporal data in consecutive lowpower images. Micro flow imaging is a product of Toshiba Medical Systems Co, Ltd (Tokyo, Japan), although there are comparable techniques made by other manufacturers.3 SonoVue (Bracco SpA, Milan, Italy) is a sulfur hexafluoride– filled microbubble contrast agent that is stabilized by phospholipids. The microbubbles have a diameter of less than 8 μm (mean, 2.5 μm)4,5 and may be insonated with a high acoustic power output (high MI) to produce extensive microbubble destruction or a low acoustic power output (low MI) to elicit microbubble resonance with production of harmonic frequencies.2,6 Our previous research showed that this method was sensitive for showing vascular morphologic characteristics within tumors, and delineated the vascular architecture of hepatocellular carcinomas as cotton, shrubbery, and deadwood patterns.7 This present study aimed to further identify the vascular architecture of focal liver lesions on micro flow imaging and to compare it with the characteristics on contrast harmonic imaging during the arterial phase.
Materials and Methods Patients From October 2004 to November 2006, 118 patients admitted to our hospital with histopathologically (n = 81) or clinically (n = 37) proven focal liver lesions were retrospectively included in the study. All patients underwent micro flow imaging and contrast harmonic imaging. Informed consent was obtained from all patients after the nature and process of the procedures were fully explained. The study was approved by the Ethics Committee of the hospital. There were 86 men and 32 women with a mean age ± SD of 47.8 ± 11.8 years (range, 21–79 years). Seventy-nine patients (66.9%) had a single nodule, and
1158
39 (33.1%) had multiple nodules. In patients with multiple nodules, the largest and most conspicuous nodule was selected for evaluation. A total of 118 nodules were evaluated; their characteristics on non–contrast-enhanced grayscale sonography are presented in Table 1. The diagnosis of hepatocellular carcinoma in 61 patients was confirmed via histopathologic examination after surgery or sonographically guided percutaneous biopsy. The remaining 9 diagnoses were confirmed via imaging and laboratory data (ie, typical manifestations on contrast-enhanced computed tomography [CT] or magnetic resonance imaging [MRI] in combination with serum α-fetoprotein levels ≥400 ng/mL). The diagnosis of liver metastasis was confirmed via percutaneous core needle biopsy in 6 of 19 patients; the remaining 13 diagnoses were proven via clinical data (histopathologic results of the primary tumors, newly detected and continuously enlarged lesions, and typical CT or MRI findings). Liver metastases were secondary to colon adenocarcinoma (n = 7), gastric carcinoma (n = 3), renal cancer (n = 3), pancreatic carcinoma (n = 1), esophageal adenocarcinoma (n = 1), adrenal pheochromocytoma (n = 1), extrahepatic cholangiocellular carcinoma (n = 1), Vater papillary carcinoma (n = 1), and endometrial carcinoma (n = 1). The hemangiomas were proven via histopathologic examination (n = 2) or a typical contrast-enhanced CT or MRI appearance with no change in the lesion size on followup imaging for at least 6 months (n = 15). All cases of intrahepatic cholangiocarcinoma, hepatic lymphoma, and focal nodular hyperplasia were proven via histopathologic examination with the specimens obtained by sonographically guided percutaneous biopsy or surgical resection. Equipment The sonographic equipment used in this study was an Aplio XV scanner (Toshiba Medical Systems Co, Ltd) equipped with a 375BT convex transducer with a frequency range of 1.9 to 6.0 MHz. Contrast-specific software, including both contrast harmonic imaging and micro flow imaging, was incorporated in the system. Contrast Agent The contrast agent used was SonoVue. Through a 20gauge intravenous cannula (Venflon; Becton Dickinson, Helsingborg, Sweden), SonoVue was injected into the antecubital vein in a bolus fashion, followed by a flush of 5 mL of a 0.9% sodium chloride solution. Each patient received 2 injections: 1 for micro flow imaging (1.2 mL of SonoVue) and the other for contrast harmonic imaging (2.4 mL of SonoVue).
J Ultrasound Med 2013; 32:1157–1171
3207jum_online.qxp:Layout 1
6/19/13
2:08 PM
Page 1159
Yang et al—Vascular Architecture of Focal Liver Lesions on Micro Flow Imaging
Table 1. Characteristics of 118 Focal Liver Lesions on Grayscale Sonography Lesion Hepatocellular carcinoma Metastasis Intrahepatic cholangiocarcinoma Focal nodular hyperplasia Hemangioma Lymphoma
n
Size, cma
Hyper
70 19 3 8 17 1
5.7 ± 3.0 (1.4–14.9) 4.5 ± 2.7 (1.1–8.5) 7.3 ± 5.7 (2.5–13.6) 6.1 ± 4.6 (1.7–14.6) 4.6 ± 3.0 (1.4–12.7) 2.9
20 7 0 1 7 0
Echogenicity Iso Hypo 9 0 0 2 0 0
26 9 2 4 7 1
Mixed 15 3 1 1 3 0
aData are expressed as mean ± SD (range).
Sonographic Examinations Sonographic examinations were performed by a single investigator with at least 5 years of experience in contrastenhanced sonography of the liver. The investigator who performed the sonographic examinations was blinded to the results of all previous imaging examinations. Before administration of the contrast agent, baseline sonography was performed to scan the liver thoroughly. All lesions were located less than 12 cm from the body surface. When the target lesion was located, imaging settings such as gain, depth, and focal zone were optimized. Generally, a single focal zone was placed at the bottom of the lesion to be assessed. The maximum diameter of each lesion was measured, and the liver segment location was determined. Color or power Doppler imaging was then performed to evaluate the intralesional vascularity, and Doppler spectral analysis of intralesional vessels was performed to identify pulsatile arterial or continuous venous flow. Contrast-enhanced sonography was then performed twice in each patient. The timer was activated promptly at the beginning of contrast agent administration, and the entire process was recorded and stored on the hard drive contained within the scanner. As in previous reports,1 contrast enhancement of the lesions was divided into 3 phases: arterial (8–30 seconds after contrast agent injection), portal (31–120 seconds after injection), and late (121–300 seconds after injection). First, the contrast harmonic imaging mode was initiated. The imaging settings, such as gain, depth, and focus, were readjusted to visualize the lesion with great conspicuity. Only a single focus was placed at the bottom of the lesion. Machine parameter settings for contrast harmonic imaging were as follows: acoustic power, 0.2%; MI,
