Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1.1 Introduction

Ultrasonography (US) is usually the first imaging modality chosen for the primary evaluation of the pancreas. The pancreatic gland can almost always be visualized by US. Even though there are well-known and sometimes over-emphasized limitations, the pancreatic gland can be adequately visualized by using correct US techniques, imaging and settings. Conventional US is a noninvasive and relatively low cost imaging method which is widely available and easy to perform. Tissue harmonic imaging (THI) and Doppler imaging are well known technologies that provide significant complementary information to the conventional method, playing an important role in the diagnosis and staging of pancreatic diseases. In recent decades, new interesting US methods have been developed focused on the evaluation of mechanical strain properties of tissues, such as elastography and sonoelasticity. Acoustic radiation force impulse (ARFI) imaging is a promising new US method that allows the evaluation of mechanical strain properties of deep tissues with the potential to characterize tissue without the need for external compression. Contrast-enhanced ultrasonography (CEUS) advances the accuracy of this first line examination by characterizing focal solid and cystic lesions and providing an accurate real-time evaluation of macroand microcirculation in and around a focal mass.

The aim of this chapter is to describe the US imaging methods and implementations now available for studying the pancreatic gland. Advantages and disadvantages of the US imaging methods are also mentioned. US approaches, such as transabdominal, endoscopic, laparoscopic and intraoperative procedures will be accurately illustrated in a dedicated chapter.

1.2 Conventional Imaging

Conventional US is a well-known, relatively low cost noninvasive imaging method which is widely available and easy to perform compared to computed tomography (CT) and magnetic resonance imaging (MRI), modalities which are usually used as second-line examinations. It is also free from side effects (i.e. lack of ionizing radiation) or contraindications, so is largely applicable also in young people. Two other important aspects are its real-time and multiplanar capabilities [1]. According to the literature, the pancreatic gland can almost always be visualized by US, even though in some cases this can be difficult due to the limited contrast between the pancreas and surrounding fat [2, 3]. In some overweight patients the visualization of the gland may also be difficult or unfeasible, despite several attempts. Examining the patient in different positions, such as erect or supine, with upleft or upright rotation, with suspended inspiration or expiration, may be suitable for achieving better pancreatic visualization. In the presence of abundant gas distension of the digestive tract, moving the transducer and applying compression can be useful to displace the bowel loops and visualize the pancreatic gland [2, 4]. Filling the stomach with degassed water (100–300 mL) or simethicone- water mixture may be used as a last option to improve US visualization of the pancreas since air bubble that cause artifacts will also be introduced into the stomach and a filled stomach is less compressible.

The US examination of the pancreas requires the use of multifrequency transducers (at least from 3 to 4 MHz) to study the entire gland with the proper frequencies for any depth (Fig. 1.1). The anatomic location, the bodysize of the patient and the respiration phase may influence the depth of the pancreas, which is not a completely fixed retroperitoneal gland (see Chapter 6). Conventional US utilizes the same frequency bandwidth for both the transmitted and the received signal. The choice of frequency is mainly based on a compromise between the spatial resolution, which depends on the wavelength, and higher frequencies, which provide higher spatial resolution but which suffer greater tissue attenuation [5]. The basic US wave is a simple sinusoidal wave with a spectrum characterized by a single line and just one frequency of energy (f0), also called the fundamental frequency or first harmonic. Furthermore, new technologies based on both the amplitude and the phase information of the return echo (e.g. coherent image formation, Acuson, Siemens) to create images are able to produce images with more information and detailed resolution [2].

Fig. 1.1 a–d
figure 1

Pancreas. a B-mode image (4.0 Mhz). b Vascular enhancement image (4.0 Mhz). c Spatial compound image (4.0 Mhz). d Harmonic compound image (3.0 Mhz)

Full size image

The US study should be performed after a minimum fast of 6 hours to improve the visualization of the pancreas, creating the best situation for the evaluation of the gland. Through transverse, longitudinal and angled oblique scan planes (multiplanar view), the entire pancreatic gland should be recognizable. Beginning with the patient in the supine position, the probe should be slightly moved to the right of the midline to visualize the head and neck of the pancreas descending a little above the umbilical line for the uncinate process. To adequately study the body and tail of the pancreas the operator should move the transducer to the left of the midline with the end (right part) of the probe rotated slightly cranially. This positioning obviously reflects the most common location of the pancreatic gland, with the head at a more caudal plane than the tail [1]. The left lateral approach may also be useful for the evaluation of the pancreatic tail, which can be visualized between the spleen and the left kidney (see Chapter 6).

An accurate US study of the pancreas consists of the evaluation of the morphology, size, contour and echotexture of all the portions of the gland, the latter being comparable to the normal liver. The main pancreatic duct and the common bile duct, together with the main peri-pancreatic vascular structures, such as the celiac, superior mesenteric, hepatic and splenic arteries and the portal, superior mesenteric and splenic veins should be assessed. Lastly, the evaluation of the adjacent organs, in particular the liver, is always required for a complete study.

As reported in the literature, conventional B-mode US has a high sensitivity in detecting focal pancreatic disease due to differences in acoustic impedance between diseases and surrounding parenchyma. The teardrop sign, which is highly suggestive of vascular encasement in the presence of a neoplastic lesion, can only be detected in B-mode, which is also able to identify a dilation of the main pancreatic duct, parenchymal or ductal calcifications and potentially present peri-pancreatic fluid collections with great confidence [2].

Technical developments in recent years have led to image fusion, which is now currently available. This technology may help in diagnostic and interventional procedures by making the comparison between US and other imaging modalities more immediate. In interventional pancreatic procedures the advantages of US guidance, such as its dynamism and the possibility of innumerable manual scanning planes, would be maintained and it would also overcome the technical limitations of the technique, such as tympanites and obesity, through the simultaneous visualization of the previously acquired CT images matched and synchronized with the US images.

1.3 Harmonic Imaging

Tissue harmonic imaging (THI) is a well know technology that improves conventional US by providing images of higher quality [5–7]. While conventional US utilizes the same frequency bandwidth for both the transmitted and the received signal (f0), THI uses low frequency for the transmitted signal and higher harmonic frequencies for the received signal. In other words, by using a Gaussian shaped transmit pulse the harmonic component can be separated from the returning echo without overlapping with fundamental reflections. In fact, nonlinear harmonic frequencies, generated by propagation of the US wave through the tissue, occur as whole-numbered multiples of the fundamental or transmitted sonographic frequency [5]. Therefore, the waveform changes compared to the basic US wave, resulting in a distorted wave with a complex form owing to the presence of both the fundamental and multiple harmonic frequencies [8].

THI takes advantages of nonlinear harmonic frequencies to correct the defocusing effects and to extensively reduce artifacts caused by low amplitude pulses [8]. As a consequence, THI produces images with improved lateral resolution by reducing side-lobe artifacts and improved signal-to-noise ratio compared with conventional US, thus resulting in an enhanced overall image quality [9]. The primary advantage is fewer artifacts in cavities, such as vascular structures, which can therefore be better evaluated. There are also advantages in fluid-solid differentiation, with the finely detailed depiction of anatomy such as the main pancreatic duct [7]. The physical basis depends on three main factors: (1) the contraction of the width of the harmonic wave; (2) the reduction of side-lobe artifacts; and (3) a received signal free of the original frequency transmitted.

Lateral resolution mostly depends on the width of the US wave. Since nonlinear harmonic waves are narrower than the fundamental, they also have lower sidelobe levels, thus improving lateral resolution which is most evident in fluid-filled structures (Fig. 1.2). The signal- to-noise ratio is consequently enhanced, with higher contrast resolution, resulting in images characterized by brighter tissues and darker cavities (e.g. main pancreatic duct, vascular structures, cystic lesions). Therefore, a narrow-bandwidth low-frequency pulse is transmitted, a filter automatically processes the received signal, and only the returning echo, characterized by high-frequency harmonic signal is used to generate the image.

Fig. 1.2 a,b
figure 2

Pancreatic mucinous cystic neoplasm. Better definition of the cystic wall and intralesional septa moving from conventional US (a) to harmonic US (b) imaging

Full size image

THI has been incorporated in all state-of-the-art systems. By pushing the specific button on the US scanner, the receiver automatically is regulated on a frequency higher than the fundamental, with little or no overlap between them, and all the components that are in the transmitted pulses are rejected. Harmonic band filtering and phase inversion are the two main methods used for the generation of harmonic images [8]. In harmonic band filtering, there is little or no overlap between the transmitted and received pulses, but through a highpass filter to the received signal, just the higher harmonic frequencies should be used. However, to separate them a fine bandwidth of the fundamental transmitted frequency must be selected and, as a consequence, decreased spatial resolution is the result. The same processes are also applied to the receiver, with a consequent decrease in contrast resolution [10]. These shortcomings can be overcome with the phase inversion method. This uses two sequential pulses, the second of which is phase reversed, and is able to remove the fundamental frequency by electronically storing the reflected signal following the first pulse and adding it to the second one, leaving only the harmonic waves [8]. The disadvantages are that the frame rate is halved and motion artifacts can occur.

The pancreatic examination requires the use of the same multifrequency curved array transducers (at least from 3 to 4 MHz) used for conventional US. Typically, the frequency setting consists of a transmitted frequency of 2.0 MHz and a received frequency of 4.0 MHz (second harmonic). The examination protocol is similar to that reported above for conventional US.

As reported in the literature, an accurate pancreatic THI examination is characterized by a higher sensitivity than conventional B-mode US regarding the detection of focal solid and cystic pancreatic lesions [8, 11]. THI is able to more clearly delineate lesion margins as well as internal solid components of a mass with more confidence [7]. Compared to conventional US, THI provides a higher soft tissue differentiation, allowing both the detection of even small lesions with little changes in echogenicity with respect to the surrounding parenchyma and the identification of calcifications [11, 12]. Moreover, other important advantages consist of the ability to clearly study deep structures and overweight patients, due to the rejection of low-amplitude pulses which generate artifacts in the conventional examination [8]. In a nutshell, in the study of the pancreas and compared to conventional B-mode US, THI can increase both spatial and contrast resolution, providing an enhanced overall image quality, better lesion conspicuity, and advantages in fluid-solid differentiation, thus achieving a better detection of pancreatic cancer.

1.4 Compound and Volumetric Imaging

State-of-the-art systems provide images with high detail resolution owing to both amplitude and phase information of the return echo and compound technology. Compounding is able to improve contrast and spatial resolution in the B-mode image (Fig. 1.1), reducing the intrinsic acoustic noise of US imaging (speckle) by generating several independent frames of data and then averaging them [2]. There are different types of compounding technology available, such as frequency compounding and spatial compounding (Fig. 1.1).

The introduction of volumetric image acquisition, which maintains the real-time and multiplanar capabilities of conventional US, opens up new clinical opportunities for a more complete evaluation of the pancreatic gland [1]. Volumetric US imaging is a relatively new technique based on the acquisition of a volume dataset of anatomic structures (Fig. 1.3). Automated volumetric imaging is able to overcome the low reproducibility of the previous volume freehand sweep acquisition, owing to the possibility of a standardized and objective acquisition during the study. The whole volume of a region of interest is automatically acquired during a breath hold of a few seconds without moving the probe (Fig. 1.4). With the volumetric electromechanical transducers, such as 4D3C (GE Healthcare, Waukesha, WI, USA), the acquisition is related to the internal movement of the piezoelectric elements inside the probe with an angle of acquisition from 40° to 60°. Therefore the entire volume is uniformly and automatically acquired, and then reviewed and studied by means of different applications: volume review for reviewing the whole volume acquired to obtain a virtual scan of the pancreas; tomographic imaging for allowing the multiplanar vision of the region of interest; volume rendering for allowing the volumetric visualization of a pancreatic lesion. Moreover, when studying a pancreatic mass the evaluation of the involvement of the peri-pancreatic vessels can be improved by using multiplanar reconstruction (Fig. 1.5). In general, the correct application of these new technologies in the US study of the pancreas results in a conventional imaging of the gland with very high spatial and contrast resolution.

Fig. 1.3
figure 3

Solid focal pancreatic lesion. Volumetric imaging of a solid focal hypoechoic (arrow) pancreatic head lesion

Full size image
Fig. 1.4
figure 4

Pancreatic mucinous cystic neoplasm. Volumetric imaging of a cystic pancreatic mass completely included in the automated acquisition scan

Full size image
Fig. 1.5
figure 5

Solid focal pancreatic lesion. Sagittal views of a solid focal hypoechoic (arrow) pancreatic head lesion after automated volumetric acquisition scan

Full size image

1.5 Doppler Imaging

Doppler imaging is a well-known technology that advances and completes the conventional US examination, providing significant complementary information about the vascular structures. Since its high sensitivity in evaluating flow in all the main peri-pancreatic arterial (i.e. celiac, superior mesenteric, hepatic and splenic arteries) and venous (i.e. portal, superior mesenteric and splenic veins) structures, together with its increased sensitivity in recognizing smaller intrapancreatic and intratumoral vessels, this technology plays an important role in diagnosing and staging pancreatic diseases [6, 13].

While conventional US is based on short pulses of US, Doppler signals derive from both continuous and pulsed waves and are mostly due to scattering from red blood cells. Some special methods have been developed for Doppler study. Continuous-wave technique, which is very sensitive to small vessels, enables measurements of a wide velocity range, but is unable to obtain information about the source of the Doppler signal because any moving object produces a signal. To overcome this shortcoming, the pulsed-wave technique, which is based on the pulse length and the duty cycle, enables the selective measurement of the wave speed at precise locations in the beam, even though the exact source of the Doppler signal remains difficult to determine because an image of the subsurface anatomy is not reported and is prone to false velocity indications (i.e. aliasing). The real advance in the application of Doppler technology is duplex Doppler imaging. This is more complex and expensive as it combines both previous techniques, but it does enable the precise location of the signal; image and both peak velocity and velocity distribution are provided in real-time together with indications of the sample size. Lastly, color-flow Doppler imaging, which combines both anatomic and velocity data, provides qualitative and quantitative information adding velocity information to the conventional images as color data: red represents blood moving toward the transducer, whereas blue represents blood moving away. Variation of the velocity is also reproduced as a different color intensity. Typically, the lighter the color is, the higher the velocity (i.e. aliasing in the presence of improper velocity range) [14].

Doppler technology has been incorporated in all state-of-the-art systems. The pancreatic examination requires the use of the same multifrequency curved array transducers (at least from 3 to 4 MHz) used for conventional US and is based on an adequate visualization of the gland and of the targeted vascular structures at Bmode US. Color gain and velocity settings are tuned to provide good color filling of the vascular structures avoiding the generation of artifacts [15]. Typically, the frequency setting varies from 1 to 4 MHz, mostly depending on two factors: first, the targeted vascular structures, since lower frequencies allow an adequate evaluation of the peri-pancreatic main vessels owing to their higher penetration, while higher frequencies allow the evaluation of smaller vessels characterized by slower flows or vascular structures in thin patients whose pancreas is less deep; second, the patient’s habitus. An accurate velocity measurement requires: (1) a correct angle between the vessel, the Doppler angle and the axis of the US beam, which should be as small as possible to generate signals with high signal-to-noise ratios; (2) the gate has to be located in the vessel center, with a size as small as possible; and (3) a correct angle for the velocity measurement has to be chosen, usually less than 60°. High-pass filters are used to reduce the influence of vessel wall and other non-vascular movements [14]. The examination protocol is similar to that reported above for conventional US.

Doppler technology implements conventional US in studying vascular structures, providing useful anatomic information and an accurate evaluation of patency (color-power study) and blood flow (color-Doppler study). At color-power imaging, a patent vessel of course appears colored. The color study offers an adequate evaluation of large vessels, providing information about the direction of flow, but it is dependent on the angle and is potentially affected by aliasing due to the difficulty in separating background noise from true flow in slow-flow states. Smaller vascular structures are better identified by the power study, which along with being relatively angle independent and unaffected by aliasing is characterized by higher signal persistence with better definition of vessel margins. However, it also suffers from increased movement artifacts and is unable to demonstrate flow direction or to estimate flow velocity [16]. Moreover, both technologies may provide useful information about the vascular network of focal lesions which may be present. Therefore, spectral waveform changes in peri-pancreatic vessels may depend on the effect of pancreatic diseases on the vascular structures [13].

As reported in the literature, an accurate pancreatic Doppler examination is based on the evaluation of all peripancreatic, intrapancreatic and intratumoral vessels. The most important applications are the identification of the vascular nature of an anechoic lesion (Fig. 1.6) detected at conventional US (i.e. pseudoaneurysm) and the differentiation between resectable (Fig. 1.7) and nonresectable (Fig. 1.7) pancreatic tumor (i.e. localized aliasing with reverse flow, mosaic pattern and accelerated flow velocity are detected at the site of stenosis, while parvus et tardus flow is observed downstream from an infiltrated tract) [17–19] with a reported accuracy of 85– 90.5% [19]. As well described in the literature, a locally advanced pancreatic mass is defined by the extended invasion of a main arterial or venous vessel, by the encasement of a main arterial structure and/or by the occlusion of a main venous structure [19, 20]. Splenic arterial or venous encasement is not a contraindication for surgical resection [6]. If both a dilation of small peripancreatic veins and a tumor surrounding three quarters of a main vessel lumen allow the diagnosis of a vascular infiltration, while the teardrop sign, due to a tumor surrounding more than a half but less than three quarters of a main vessel lumen is highly suggestive of vascular encasement, a simple contiguity (less than a half of the vessel circumference) between tumor and vessel does not necessary correspond to vascular invasion [20].

Fig. 1.6 a,b
figure 6

Pseudoaneurysm. Cystic lesion (asterisk) in the pancreatic tail at conventional imaging (a) in patient with chronic pancreatitis with final diagnosis of pseudoaneurysm at Doppler study (b)

Full size image
Fig. 1.7 a–d
figure 7

Pancreatic mass resectability. a Schematic representation of a resectable pancreatic head mass. b US detection of a resectable hypoechoic mass (arrow) of the pancreatic head. c Pancreatic head solid mass infiltrating the superior mesenteric vein at conventional imaging and confirmed at Doppler study. d Pancreatic head solid mass infiltrating the superior mesenteric artery at conventional imaging and confirmed at Doppler study

Full size image

Some new technologies have been developed: wideband Doppler, which improves both spatial and temporal resolution of the color-Doppler signal with decreased artifacts [13]; power-like flow systems such as B-flow (General Electric) and e-flow (Aloka) imaging which are able to suppress tissue clutter and improve sensitivity to directly visualize blood reflectors and consequently provide images characterized by better spatial resolution [13]; color flow imaging (CFI), mostly used to image the blood movement through arteries and veins, but also to represent the motion of solid tissues [21]. The weak signals from blood echoes are enhanced and correlated with the corresponding signals of the adjacent frames to suppress non-moving tissues. The remaining aspects of the data processing are essentially the same as in conventional grey-scale imaging. In comparison with Doppler techniques these new US flow imaging modalities are not affected by aliasing and have the advantages of a significantly lower angle dependency and better spatial resolution with reduced overwriting. As a consequence, evaluation of vessel profiles is markedly improved (Fig. 1.8).

Fig. 1.8
figure 8

Superior mesenteric artery. Doppler based US imaging of superior mesenteric artery shows flow only inside the lumen of the artery with a perfect detection of the arterial wall (arrow)

Full size image

Other new Doppler-based technologies are able to improve image quality, owing to the immediate identification of the vascular structures in B-mode. For example, Clarify Vascular Enhancement (Acuson, Siemens) enables image optimization by enhancing the B-Mode display with information derived from power-Doppler, clearly differentiating vascular anatomy from acoustic artifacts and surrounding tissue (Fig. 1.9). In studying the pancreas, the resulting images can immediately appear diagnostic or more informative.

Fig. 1.9
figure 9

Small solid focal pancreatic lesion. Doppler based US imaging of a very small solid focal hypoechoic (arrow) pancreatic lesion in the pancreatic body

Full size image

1.6 Elastography Imaging

In recent decades, new and interesting US techniques have been developed focused on the evaluation of mechanical strain properties of tissues. The noninvasive analysis of tissue stiffness immediately received major interest, owing to a revolutionary approach in the study of focal and diffuse diseases able to provide a new diagnostic tool. Tissue stiffness has long been an asset in physical palpation for clinicians and surgeons. Since the introduction of these new technologies, it has become a new and useful technique for radiologists able to complement other traditional data when making a diagnosis.

The first imaging techniques developed to image tissue elasticity consisted of elastography, the static US approach [22], and sonoelasticity, the dynamic US approach [23]. In elastography, the longitudinal stress and strain of superficial tissues can be estimated by tracking tissue motion mainly derived from external mechanical compression applied by the US probe [24]. In sonoelasticity, externally applied vibrations at low amplitude (less than 0.1 mm displacement) and low frequencies (10– 1000 Hz) are used to induce oscillations within tissues and this motion is detected by Doppler US [25]. Through a color or grey scale map, a qualitative evaluation of the elastic properties of tissues is provided. As a consequence, isoechoic lesions which are undetectable at conventional US often might be identified at elastography and sonoelasticity imaging, owing to their altered vibration response. US elastography and sonoelasticity have been implemented as simple add-ons alongside conventional US scanners or as dedicated units. Transient US elastography utilizes a displacement wave generated by a piston or acoustic force which provides the stress to the tissue, without producing an image, but only numeric data of the tissue stiffness. This has mainly been used in the evaluation of diffuse liver diseases [26].

As widely reported in the literature, several clinical applications have been studied: for diagnostic purposes and biopsy targeting in breast and prostate; to differentiate benign from malignant nodules in the thyroid gland; to differentiate benign from malignant lymph nodes [27–30]; and in the evaluation of liver fibrosis [31].

Elastography has the same problems as B-mode sonography. The stress propagating into a tissue is in fact attenuated by tissues, causing it to spread into other directions from the primary incidental direction and to interact with a boundary between two media of different elastic properties, with potential distraction.

A more recent elastographic technique called acoustic radiation force impulse (ARFI) imaging has been developed [32, 33]. This new promising US method enables the evaluation of mechanical strain properties of deep tissues without the need for external compression. It produces a high intensity push pulse to displace the tissue and lower intensity pulses for imaging. The physical basis depends on the evaluation of the transverse wave spread away from the target tissue. There are two basic types of wave motion for mechanical waves, most widely used in US testing: longitudinal or compression waves and transinverse or shear waves. Whereas the particle displacement is parallel to the direction of wave propagation in a longitudinal wave, in a transverse wave the particle displacement is perpendicular to the direction of wave propagation. In other words, if compression waves can be generated in liquids as well as solids, shear waves are not effectively propagated in gas or fluids owing to the absence of a mechanism for driving motion perpendicular to the sound beam. Transverse waves are also relatively weak when compared to longitudinal waves, since they are usually generated using some of the energy from longitudinal waves. As is well known, sound travels at different speeds in different materials, mostly because elastic constants are different for different media. Young’s modulus deals with the velocity of a longitudinal wave, while the shear modulus deals with the velocity of a shear wave.

ARFI imaging has been incorporated in only a few US systems, and all papers present in the literature at this moment describe the application of the Siemens ACUSON S2000scanner (Siemens, Erlanger, Germany). The pancreatic examination requires the use of the same multifrequency curved array transducers (at least from 3 to 4 MHz) used for conventional US. A single transducer is used both to generate radiation force and to track the resulting displacements. Pushing the specific button on the US scanner, the transducer is automatically regulated on the THI imaging, with a received frequency of 4.0 MHz. On a traditional harmonic US image, the target region of interest (ROI) is selected utilizing a box with fixed dimensions of 1 × 0.5 cm, able to descend at a maximum depth of 5.5 cm (8 cm in the most recent scanner). The box has to be completely included in the target tissue (i.e. organ in cases of diffuse diseases or lesion in cases of focal diseases), taking care not to comprise any fluid structures, such as vessels or ducts. Once the target ROI has been correctly located, the patient should maintain a proper suspended inspiration or expiration, to minimize motion artifacts. Pushing a specific button on the US scanner, acoustic push pulses are then transmitted. The push pulse is characterized by short duration (less than 1 msec) and runs immediately on the right side of the target ROI. Owing to its very high speed, it is minimally and not significantly influenced by the structures encountered through the path away from the transducer up to the box. The acoustic beam is able to generate localized, micron-scale displacements in the selected ROI proportional to the tissue elasticity. As a consequence, detection waves of lower intensity (1:100) are generated. The shear waves produced, which run away perpendicular to the acoustic beam, are measured. The speed of the shear waves reflects the tissue elasticity, being dependent on the elasticity modulus that is mainly related to the resistance offered by the tissue to the wave propagation, and is proportional to the tissue stiffness: the stiffer a tissue is, the higher the shear wave speed it generates [34]. As a result, according to the interaction between waves and transducer previously selected by the operator, the response may be reported as qualitative or quantitative information (Fig. 1.10). The qualitative response consists of a grey scale map of the previously selected ROI, characterized by a lack of anatomic details, but with high contrast resolution, in which a bright shade corresponds to soft tissue, while a dark shade represents stiff tissue. The implementation of ARFI imaging able to provide this kind of response is called Virtual Touch tissue imaging. Obviously, this new advance could play an important role in the presence of focal disease. The quantitative response consists of a numeric wave velocity value, expressed in m/s, which derives from multiple measurements automatically made by the system for the previously selected ROI. It provides objective and reproducible data regarding the shear wave speed: the stiffer a tissue is, the higher the shear wave speed. The implementation of ARFI imaging able to provide this numerical response is called Virtual Touch tissue quantification and can be applied both in the presence of focal and diffuse disease [35].

Fig. 1.10
figure 10

Pancreas. Acoustic radiation force impulse (ARFI) US imaging with virtual touch quantification shows normal shear wave velocity in the normal pancreas of a healthy volunteer

Full size image

The most significant advantages of ARFI technology over previous elastographic techniques are: (1) its in tegration into a conventional US system, thus allowing the visualization of B-mode, color-Doppler mode and ARFI images with the same equipment; (2) the consequent selection of an ROI in the target tissue on a conventional US image; (3) the subsequent possibility of precisely studying target lesions during a real-time visualization at conventional US; (4) the opportunity to also study deep tissues, since there is no need for external compression; and (5) the objective quantification of the tissue stiffness expressed as a numeric value, by Virtual Touch tissue quantification. There are nonetheless some important limitations: (1) the fixed box dimensions of the target ROI, while less important in cases of diffuse disease, could be significantly limiting in cases of focal lesions; and (2) a high sensitivity to movement artifacts, such as lack of suspended respiration or heart motion.

The US examination should be performed after a minimum fast of 6 hours to improve the visualization of the pancreas, creating the best situation for the evaluation of the gland. The good visualization of the target tissue at conventional US is a mandatory condition for performing the ARFI examination.

As reported in the literature, the mean wave velocity value obtained in the healthy pancreas (Fig. 1.10) is about 1.40 m/s [6, 35]. An accurate pancreatic US examination consists of the application of both qualitative and quantitative implementations of ARFI technology, whenever possible, to assess the concordance of the results. Different focal and diffuse diseases that alter the tissue stiffness should be characterized by different shades and wave velocity values. For example, since pancreatic ductal adenocarcinoma is a firm mass which is stiffer than the adjacent parenchyma (see also Chapter 8) owing to the presence of fibrosis and marked desmoplasia, it should appear as a dark shade with higher values (Fig. 1.11).

Fig. 1.11 a,b
figure 11

Pancreatic ductal adenocarcinoma. a Acoustic radiation force impulse (ARFI) US imaging shows a solid mass in the pancreatic body appearing black (asterisk) and therefore stiff at virtual touch imaging. b Acoustic radiation force impulse (ARFI) US imaging shows a solid mass in the pancreatic body with very high value of shear wave velocity at virtual touch quantification and therefore stiff

Full size image

According to the physical principles of the shear waves, ARFI imaging has been tested in the study of solid tissues. However, fluids in vivo, and as a consequence pancreatic cystic lesions, can be markedly different and different responses at ARFI technology might be expected. The qualitative evaluation should give a bright shade, while as recently reported in the literature, it seems that the quantitative study usually gives non numeric values in serous cystadenoma (see also Chapter 9), which contains a simple fluid, and mainly numeric values in mucinous tumors (Fig. 1.12), which contain a more complex content [36, 37].

Fig. 1.12 a,b
figure 12

Pancreatic mucinous cystic neoplasm. Acoustic radiation force impulse (ARFI) US imaging of a cystic mass with numerical value of shear wave velocity at virtual touch quantification of the fluid content

Full size image

Since its recent introduction, few data regarding the usefulness of ARFI technology in the study of pancreatic diseases are available in the literature. However, it seems to be potentially able to allow tissue characterization by imaging and may constitute a feasible alternative to invasive needle-biopsy in the future.

1.7 Contrast-enhanced Ultrasound

Contrast-enhanced ultrasonography (CEUS) is a relatively recent implementation of conventional US which significantly advances the accuracy of this first line examination in characterizing focal solid and cystic diseases. The administration of microbubbles allows an accurate evaluation of macro- and microcirculation, in and around a focal mass, giving more detailed and advanced results than the color-Doppler study thanks to its high spatial, contrast and temporal resolution. This new technology has been widely used to study hepatic diseases and also more recently applied in the study of the pancreas, giving promising results in diagnosis and staging of pancreatic diseases already detected at conventional US [6, 38].

The introduction of US contrast agents goes back some decades and their effects during cardiac catheterization were first described at the end of the 1960s. Today their use has been approved in Europe, Asia and Canada, but the Food and Drug Administration in the United States has not yet approved their application for non-cardiac use. Only the administration in pregnancy and pediatrics is off label. Some recommendations exist, especially for second generation contrast agents filled with sulfur-hexafluoride: they are not recommended in patients with recent acute coronary syndrome, unstable angina, recent acute heart attack, recent coronary artery intervention, acute or class III or IV chronic heart failure or severe arrhythmias. No interactions with other drugs have been reported and only rarely some subtle and usually transient adverse reactions have been described, such as tissue irritation and cutaneous eruptions, dyspnea, chest pain, hypo- or hypertension, nausea and vomiting. No severe effects have been described in humans to date [39, 40].

US contrast agents consist of microbubbles, characterized by a diameter that ranges from 2 to 6 microns, a shell of biocompatible materials, such as proteins, lipids or biopolymers and a filling gas, such as air or gas with high molecular weight and low solubility (e.g. perfluorocarbon or sulfur hexafluoride). Their small diameter allows their passage through the pulmonary district, thus microbubbles are exhaled during respiration 10–15 minutes after injection, while the components of the shell are metabolized or filtered by the kidney and eliminated by the liver. Shell and gas influence the time of circulation and acoustic behavior of microbubbles. The thin shell ranges from 10 to 200 nm and allows the passage through the pulmonary district with a consequent systemic effect and a more prolonged contrast effect. The filling gas produces a vapor concentration inside the microbubbles higher than the surrounding blood, increasing their stability in the peripheral circulation [38, 41].

Both the shell and the filling gases have been changed over the years, passing from first generation contrast media to second generation agents. The first generation contrast media were characterized by a stiff shell (denatured albumin) and air as filling gas. The stiff shell allows more stability in the peripheral blood, with a reduction in non-linear behavior. Therefore, as the microbubbles have a short half-life because they are easily destroyed, their US response depends on the echogenicity and the concentration. The second generation contrast media are both more stable and resistant. They are characterized by a flexible shell (phospholipids), which allows the prevalence of nonlinear behavior, and filling gas other than air. Their US response consists of the generation of nonlinear harmonic frequencies, since at low acoustic power of insonation (about 30–70 kPa), the degree of microbubble expansion is greater than its compression [41].

Several contrast-specific software applications have been developed for CEUS examination, even though the most promising techniques are phase and amplitude modulation. Pulse inversion is the most common phase modulation technique [42], while power modulation is a well-known amplitude modulation software application [41]. Cadence contrast pulse sequencing (CPS) is a more advanced combined phase and amplitude modulation technique [38, 43].

The CEUS examination should be performed after an accurate conventional US of the pancreas with the evidence of a focal or diffuse pancreatic disease [44]. The pancreatic examination requires the use of the same multifrequency curved array transducers (at least from 3 to 4 MHz) used for conventional US. Nowadays, second generation contrast agents are used. Harmonic microbubblespecific software applications are required to filter all the background tissue signals so only vascularized structures related to the harmonic responses of the microbubbles are visualized after injection. The dual screen should be used to adequately and continuously compare B-mode and contrast images. Focus and depth should be regulated simultaneously in both images and low acoustic US pressures should be selected (mechanical index less than 0.2). The examination protocol and technique are similar to those reported above for conventional US.

The dynamic evaluation begins immediately after the intravenous administration of a 2.4-mL bolus of microbubble contrast agent. Since the pancreatic blood supply is exclusively arterial, the enhancement of the gland begins almost together with the arteries. Enhancement of the pancreatic gland begins almost at the same time as aortic enhancement. After this early phase (arterial/pancreatic; from 10 s to 30 s), as with other dynamic imaging modalities there is a second phase, the venous phase (from 30 to approximately 120 s) defined by hyperechogenicity within the spleno-mesenteric-portal venous axis. The late phase (about 120 s after injection) is defined by hyperechogenicity of the hepatic veins.

US specific contrast agents have a purely intravascular distribution without any interstitial phase, so they differ from all contrast media used during CT and MRI examinations [45]. Moreover, CEUS with second generation contrast media enables real-time evaluation of target tissues, with high spatial, temporal and contrast resolution. Unlike other imaging modalities, as reported above, only vascularized structures are visible after the administration of microbubbles (see Chapter 11). Therefore, compared to conventional US and other imaging modalities, pancreatic CEUS is better able to differentiate between solid and cystic lesions (Fig. 1.13), characterize focal masses and provide a clear differentiation between remnant tissue, fibrosis and necrosis [44]. Moreover, the CEUS examination covers an important role in evaluating the resectability or non-resectability of a focal mass [46], together with Doppler imaging for the assessment of the relationship between the tumor and the adjacent main vessels, and during the late phase to exclude the presence of liver metastases.

Fig. 13 a–b
figure 13

Pancreatic intraductal papillary mucinous neoplasm. a Pseudosolid appearance of the pancreatic head lesion at conventional US resulting hypoechoic (arrow) but avascular with cystic appearance at CEUS (b). The cystic nature (arrow) of the lesion is confirmed at MRI (c)

Full size image

Some new applications of CEUS have been developed: the use of CEUS enhancement as a prognostic factor, both in the diagnostic workup and in the follow-up of patients. In fact, as reported in the literature, in the presence of focal pancreatic lesions, the accurate description of the enhancement pattern at CEUS is mandatory for a prompt prognostic evaluation. The association between intratumoral microvessel density (MVD) and tumor aggressiveness has already been proven [46]. The use of microbubbles as a vehicle for targeted therapies is an interesting future possibility [47]. Moreover, the development of new software applications for the perfusion study has been recently improved. Some papers have reported the qualitative, subjective evaluation of the enhancement pattern of different pancreatic tumors studied at CEUS [48], while other studies have described the potential quantitative evaluation of the CEUS enhancement, derived from the offline evaluations of different pancreatic tumors [46, 49]. More recently, a few US systems have been developed to quantitatively evaluate the enhancement at CEUS, based on either the video intensity analysis or the raw data analysis, which are able to immediately achieve repeatable results comparable to those derived from perfusion CT examinations [50].

An accurate pancreatic CEUS examination should be performed only after an adequate conventional US and consists of a real-time continuous observation of the target tissue (pancreatic gland in cases of diffuse disease or focal lesion already detected at conventional US in cases of focal disease [51]) during all the dynamic phases after the administration of microbubbles. At the end, in all cases a liver study during the late phase should be performed (Fig. 1.14).

Fig. 1.14 a,b
figure 14

Liver metastatic pancreatic adenocarcinoma. a At US no focal liver lesion was detected. b At late phase of contrastenhanced US a hypoechoic solid metastatic lesion (arrow) was detected in the left lobe of the liver

Full size image