Color Doppler EUS evaluates the vascularity of pancreatic tumors
[31]. Contrast-harmonic EUS (CH-EUS) is a technique that allows real-time visualization of microvasculature and pancreatic parenchyma perfusion, without Doppler-related artifacts, showing to be particularly useful in differentiating PDAC from other solid pancreatic lesions
[31][76][77][78]. The development of microbubble-based contrast agents, together with advances in ultrasound technology, has led to improve imaging of fine vascular structures and visualization of microflow patterns within target lesions
[76]. The most commonly used second-generation ultrasound contrast agents are composed of microbubbles of perfluoro gas entrapped into a lipid shell (e.g., Sulfurhexafluoride, Sonovue™, Bracco, Milan, Italy; Perfluorobutane; Sonazoid™, Daiichi-Sankyo, Tokyo, Japan; GE Healthcare, Milwaukee, WI, USA; not available in the USA and Europe). After having identified the target lesion, the second-generation contrast agent is injected as a bolus through an antecubital vein, using a 20-gauge catheter, followed by a 10/20 mL saline solution flush. The examination of the lesion lasts a minimum of 45 s after contrast agent injection
[76][77]. The definition of ‘hypoenhancement’ is a pattern of enhancement in which the echo intensity of the lesion is lower than that of the surrounding pancreatic tissue on CH-EUS. It has been demonstrated that the hypovascular pattern predicts malignancy with a sensitivity of 92–96% and accuracy of 82–95%
[79]. CH-EUS also provided greater diagnostic sensitivity (83.3%) for differentiating small PDAC (<2 cm) from other tumors when compared with power Doppler EUS (11%) and contrast-enhanced CT (50%)
[80][81]. Kitano et al.
[82] demonstrated the superiority of CH-EUS over other imaging modalities for diagnosing small PDAC (<2 cm), with sensitivity and specificity of 91.2% and 94.4% for CH-EUS and 70.6% and 91.9% for CT, respectively. On the other hand, hyper or isoenhancement is a strong negative predictor of PDAC. PanNENs are usually hyperenhancing (
Figure 3). In the same study
[82], the authors observed that hyperenhanced lesions were diagnosed as PanNETs with a sensitivity of 79% and specificity of 99%. PanNENs are a heterogeneous group of malignancies with different biological characteristics. Tumor size, grading, and the Ki67 proliferation index are known predictors of malignancy. Histologically, microvessel density is inversely correlated to tumor grading. The evaluation of microvascularization with CH-EUS has shown to have a high diagnostic value to predict malignancy with a sensitivity, specificity, and accuracy of 90.5%, 90%, and 90.2%, respectively
[83]. The usefulness of CH-EUS in predicting malignancy of PanNENs has been evaluated in a recent retrospective study
[84]. The authors included 92 patients with PanNENs and observed that heterogeneous enhancement on CH-EUS was able to predict malignancy with a sensitivity, specificity, PPV, and NPV of 93.3%, 93.5%, 87.5%, and 96.7%, respectively. A similar diagnostic yield was also shown in small (<2 cm) GI tumors. Palazzo et al.
[77] demonstrated in a study how tumors with heterogeneous enhancement at an early arterial phase were likely to be aggressive. NPV value of CH-EUS for tumor aggressiveness was more than 95%. Diagnostic values were particularly high in the clinically relevant subgroup of G1/G2 non-functioning tumors without preoperative metastasis showing how CH-EUS should have a role in algorithm management in the choice between surgery versus the “wait and see” strategy and between oncology surgery versus ablation technique without lymph node excision. The differential diagnosis between benign and malignant lesions appears to be the main field of application of CH-EUS. CH-EUS may also predict PanNENs aggressiveness as tumors with heterogeneous enhancement have fewer vessels and more fibrosis which are the features associated with more aggressive tumors
[84][85]. Concerning pancreatic solid neoplasms, the accuracy of CH-EUS seems comparable to that of EUS-TA. Moreover, the concomitant use of both CH-EUS and EUS-TA may prove to have additive value in increasing the overall accuracy by overcoming the false negative results of each individual technique. EUS-TA is not going to be replaced by CH-EUS; however, CH-EUS may contribute to the selective use of EUS-TA
[80]. As a result of the reported advantages above, CH-EUS is the imaging modality of choice to discriminate PanNENs from PDAC at the first clinical assessment, particularly in the case of small tumors
[76]. Further technical refinements and novel substances are likely to improve its diagnostic potential in the near future. Of particular interest, a quantitative assessment of CH-EUS has been proposed. Using dedicated software, temporal change of the echo enhancement intensity can be measured inside a region of interest (ROI) set with the maximum possible size in the center of the mass. As a result, a time-intensity curve can be achieved and different parameters evaluated (e.g., echo intensity change from baseline to peak, time for peak enhancement, the velocity of contrast imaging from baseline to peak, echo intensity reduction rate)
[86]. In a study including 91 patients with different pancreatic diseases, the diagnostic accuracy based on the time-intensity curve was 88% and rose to 94.7% when considered in combination with EUS B-mode
[86]. Moreover, Takada et al., in a study including 26 patients with PanNENs, demonstrated a diagnostic accuracy of time-intensity curve analysis close to 100% to differentiate G1/G2 PanNETs from G3 PanNECs
[85].