Contrast-enhanced Ultrasound For Evaluation Of Superficial Lymphadenopathy

BACKGROUNDThe detection or exclusion of lymph node metastasis in patients with tumors is of special importance because not only the patient's prognosis but also the therapeutic concept is directly influenced by the nodal status. Conventional ultrasonography including gray-scale imaging(B-mode), color-coded Doppler sonography(CCDS) and power Doppler(PD) have shown to serve as a promising and accurate tool in differentiating diagnosis of superficial lymphadenopathy by detecting the characteristics of morphology and angioarchitecture such as the longitudinal-transverse diameter ratio, margin, intranodal echogenicity, echogenic hilus and vascular pattern. However, malignant nodes may be the same shape as benign nodes when they are in the early stage of involvement. In metastatic as well as benign enlarged lymph nodes, a Doppler signal is not always detectable. Micrometastases are frequently discovered in the no-enlarged lymph nodes, and more than half of the nodes that had a maximum transverse diameter between 2 and 4 mm were found no vascularity. In some studies, a relatively lower sensitivity of vascular pattern was found in distinguishing reactive from lymphomatous nodes. This is probably due to the high incidence of hilar vascularity in lymphomatous nodes, which resulted in a high false-negative rate.Some studies exploited the introduction of the first-generation air-based contrast agent (such as SH U 508A, Levovist) to improve the Doppler sonography signal. But the first-generation ultrasound contrast agent is not enough stable and always destroyed with high MI (mechanical index) imaging methods, and there are many color Doppler imaging artifacts, so the microcirculation can not be delineated.Recently, new ultrasound techniques, such as pulse inversion harmonic imaging, have been developed that are extremely sensitive to non-linear effects of US interaction with microbubble contrast agents. With the second-generation ultrasound contrast agents, such as SonoVue, real-time evaluation of the lymph nodal perfusion can be carried out at a low MI. Different lymphadenopathy have different pathological characteristics, especially tumor tissue has peculiar tumor microcirculation. Real-time gray-scale contrast-enhanced ultrasound for superficial lymph nodes can differentiate benign from malignant nodes better by detecting the distribution and biologic effect of microbubble in the nodes, and may help to choose appropriate lymph node for biopsy. It may also be useful in early assessing the response of lymph nodes to radiochemotherapy. The objective of this study was to assess the enhancement of echogenicity and the characteristics of the time-intensity curves by using real-time gray-scale contrast-enhanced ultrasound for superficial lymph nodes.Materials and methodsFrom March to December 2005 43 consecutive patients(26 women, 17 men; age range: 18 to 71 years, mean 51 years) with enlarged superficial cervical, axillary, or inguinal lymph nodes clinically suspected to be malignant were prospectively examined by ultrasound immediately. Of the 43 patients, 33 had metastatic nodes, 8 had reactive hyperplasia, and 2 had lymphoma. In the 33 patients with metastatic nodes, the primary tumors were in the nasopharynx(n=12),breast(n=19), lung(n=1), thyroid(n=1). Atotal of 75 enlarged lymph nodes were found on sonography. At the time of the US examination, none but one patient with lymphoma had received treatment.Scanning was performed by one experienced radiologist blinded to the final diagnosis with a commercially available scanner (Acuson Sequoia 512) using an electronically focused near-field linear array transducer with a 8-to 15-MHZ bandwidth and provided with contrast pulse sequences(CPS) contrast imaging and Axius? Auto-Tracking Contrast Quantification(ACQ) software.In the first step, all patients were examined using conventional B-mode sonography. Echo texture, margin sharpness, longitudinal-transverse diameter ratio, and the location of the hilus were used as parameters. In the second step, patients were examined using color Doppler sonography. The vascular pattern was evaluated. Nodes having an inhomogeneous echo, blurred margins, absent or aberrant hilus, the longitudinal-transverse diameter ratio less than 2, and the preservation of peripheral or mixture vessels were classified as malignant. If the vascular pattern and at least one B-mode criterion were classified as malignant, the node was judged to be malignant and vice versa. In the nodes in which no vessels could be identified, the criteria of B-mode sonography were used as mentioned earlier.Finally, the gray-scale contrast-enhanced sonography was performed after intravenous injection of the US contrast agent microbubble (SonoVue, Bracco, Milan, Italy)as a bolus at a 2.4mL dose, followed by 5mL of normal saline flush. A low frame rate (7MHZ)and a low mechanical index (MI=0.1～0.2) were used. Each exam lasted about 2 min following bolus injection. The cine clips were acquired for each nodes and digitally stored on the US unit, they were reviewed retrospectively on screen. Readers subjectively assessed in consensus node contrast enhancement. The regions-of interest (ROI) was placed in the parenchyma near the capsula,and the time-intensity curve for each node was acquired and analyzed by ACQ. The arrival time (AT, the time that the contrast "arrives" in the image can be considered as when the image changes significantly), time-to peak (TTP, the interval from time zero to the image where the intensity in the ROI is a maximum.), peak intensity (PI, the value of the maximum intensity in the ROI), rate parameter (|3, the rate constant of the wash in and is expressed in units of I/seconds.), Goodness of Fit (GOF, the correlation coefficient between the measured data and the fitted values at the same points)were evaluated. GOF values range from 0 to 1. GOF values near 1 indicate that the fitted curve accurately represents the data. Low values of GOF indicate that the fit parameters should not be highly valued. GOF values must be lager than 0.7 in our study.In all patients the final diagnosis was made by means of histological analysis of resected specimens. The sensitivity, specificity, and accuracy were calculated for conventional and contrast-enhanced sonography in differentiation between benign and malignant lymph nodes. The u-test was used. In the quantitative analysis, the results were expressed as means ± standard (s) and analyzed with Student's t-test. Significance was declared at the P<0.05 level.Results1, The enhancement was classed as follows: type I , intense homogeneous enhancement; type II, intense but inhomogeneous enhancement, with perfusion defects in the echogenic hilus; type III, intense but inhomogeneous enhancement, with perfusion defects in the nodal parenchyma; type IV, scarce intranodal enhancement. The study showed enhancement of type I in 16 of 28(57%) benign lymph nodes, type II in 10 of 28 (36%)benign lymph nodes; type III in 30 of 39(77%) metastases, type IV in 8 of 39(21%) metastases; type I in 5 of 8 lymphomatous nodes, type III in 3 0f 8 lymphomatous nodes.2, The nodes with enhancement of type I and II were classified as benign, with enhancement of type III and IV were classified as malignant. The sensitivity, specificity, and accuracy of conventional techniques in differentiation between benign and malignant lymph nodes were 70%, 86%,and 75% versus 87%, 93%,and 89% for contrast-enhanced ultrasound. The increase in correct diagnosis was significant (P<0.05) when conventional ultrasound was tested against contrast-enhanced ultrasound.3, The time-intensity curves appeared quick-up and quick-down type in the benign and lymphoma groups, but quick-up and slow-down in the metastatic group.4, There were no significance in the AT and TTP between benign and metastatic groups (both P>0.05); but the PI in benign group was significantly higher than metastatic group(P<0.05).The AT and TTP in lymphoma were significantly lower than those in the benign (P<0.001 and P<0.05) and metastatic (both P<0.001) groups; and the PI in lymphoma were significantly higher than those of the benign and metastatic groups (both P<0.001).Conclusions1, Type I and II mostly represent enhancement patterns of benign lymph nodes, type III and IV mostly represent enhancement patterns of malignant lymph nodes.2, The pretreatment lymphomatous nodes mostly show enhancement of type I, the post-treatment lymphomatous nodes mostly show enhancement of type II.3,The time-intensity curve is well established criterion which represent the perfusion of nodes. The curves appeared quick-up and quick-down type in the benign and lymphoma groups, but quick-up and slow-down in the metastatic group.4, There were no significance in th AT and TTP between benign andmetastatic groups; but the PI in benign group was significantly higher than metastatic group.The AT and TTP in lymphoma were significantly lower than those in the benign and metastatic groups; and the PI in lymphoma were significantly higher than those of the benign and metastatic groups.5, With gray-scale contrast-enhanced ultrasound, we can do better in differentiations of superficial lymphadenopathy than conventional sonography.