Welcome to the Hitachi Medical Systems America, Inc. MRI Anatomy and Positioning Series. Over the coming months, we will be offering teaching modules to allow users of Hitachi MRI scanners to advance their positioning skills and review the anatomy that should be seen on some common MRI exams. Our intention is to discuss and review the anatomy that is most often seen, and the positioning that is most often used in your MRI studies. Good positioning skills are needed to ensure the best possible image quality for your studies.
In this third module, we will discuss the anatomy and positioning of the major abdominal organs, to include the liver, the pancreas, and the kidneys. We will discuss their anatomy and physiology, as well as their arterial and venous systems, nerve supply, and the relevance of MRI in the characterization and diagnosis of diseases of these organs. We will also discuss the benefits of dynamic contrast-enhanced exams and the use of more advanced procedures to further enhance our imaging of these abdominal organs.
Within our modules, we will offer suggestions as to appropriate RF coils to be used for various MRI exams. The RF coils that are recommended for abdominal imaging are part of the standard group of coils that are furnished with your magnet. Regardless of the RF coil that is being used, every attempt should be made to route the coil cable(s) in a manner that will avoid contact with the patient.
We will also discuss the use of the various pads that are furnished with our MRI systems (trough pads, table pads, accessory pads, coil cable pads, etc.). It is important to use the various pads that are provided to assist in eliminating, or at least minimizing, the amount of each patient’s skin-to-skin, skin-to-bore, or skin-to-cable contact. Reducing the amount of each of the aforementioned contacts reduces the patient’s chances of thermal injury. Please refer to the MR Patient Warming Prevention Plan published by Hitachi Medical Systems America, Inc. for more information concerning the prevention of patient warming.
When imaging the abdominal organs, the use of breath-holds or respiratory gating is necessary, due to the constant motion of the abdomen from patient respirations. Faster imaging techniques typically allow for scan performance in a single breath-hold. Expiration is preferred, as the position of the organs (at least the kidneys) remains more constant on expiration, as compared to inspiration. The Breath-hold field is located on the Dynamic tab on the Altaire, AIRIS Elite, and AIRIS II systems. Once the Breath-hold field is set to On, a Timing field becomes available, where the technologist can insert a pause before the First acquisition, or before Every acquisition. The technologist can perform breath-holds manually, by giving instructions to the patient each time using the microphone. Some systems have the Auto Voice option, which is selected in the Voice field on the Others tab. The specific breathing instructions and their timing can be selected in the Auto Voice Set Tool. These instructions will be automatically relayed to the patient when both the Breath-hold field and the Voice field are selected.
On the Oasis, Echelon, and Echelon OVAL systems, breath-holds can be acquired by setting the Wait mode field to ON. This field is found under the Scan Control section of parameters. Breath-hold instructions can be given manually, by speaking to the patient over the microphone each time, or Auto Voice may be used. The Auto Voice Setting window can be found under the System Settings launcher button. Breathing instructions and timing settings can be selected and saved with specific names. Saved selections are displayed in a dropdown listing for the Auto Voice field, which is also found in the Scan Control section.
If your patient is unable to perform breath-holds, or your scan time precludes the use of breath-holds, respiratory gating can be used. The patient’s expirations trigger the sequence acquisitions. The respiratory bellows must be positioned and secured on the patient before scanning begins. The respiratory waveform is monitored to acquire the average respiratory rate. The average number of respirations is then input in the Beat Rate field under the Gating section of parameters, or on the Gating tab. Numerous additional Gating parameters can be manipulated.
Respiratory gating equipment for the Altaire, AIRIS II, and AIRIS Elite systems consists of a respiratory sensor, respiratory belt, and respiratory sensor tubing, as seen in Figure 1. Briefly observe the patient’s breathing to determine proper placement of the respiratory belt and sensor. The belt must be tight enough so the sensor can react to the patient’s breathing, but not too tight as to constrict the patient’s breathing. One end of the respiratory sensor tubing is plugged into the respiratory sensor, and the other end is plugged into the monitoring module on the MRI system. The Waveform window should be open to monitor the respiratory waveform before scanning begins (Figure 2). This is done to ensure proper placement of the respiratory gating equipment on the patient, as well as to ensure proper working order of the equipment, resulting in a strong, steady signal.
Respiratory gating equipment for the Oasis and Echelon systems consists of respiratory bellows, a respiratory belt, and a respiratory hose, as seen in Figure 3. Briefly observe the patient’s breathing to determine proper placement of the respiratory bellows and belt. The belt must be tight enough so the bellows can react to the patient’s breathing, but not too tight as to constrict the patient’s breathing. The respiratory hose is plugged into the monitoring module on the MRI system. The Waveform window should be open to monitor the respiratory waveform before scanning begins (Figure 4). This is done to ensure proper placement of the respiratory gating equipment on the patient, as well as to ensure proper working order of the equipment, resulting in a strong, steady signal.
Respiratory gating equipment for the Echelon OVAL system consists of respiratory bellows, respiratory sensor tubing, and a wPPU Wireless Module. A battery from the Wireless Module Battery Charger should be inserted into the back of the wPPU Wireless Module (Figure 5), and the battery and communication statuses confirmed on the front of the Module. The respiratory sensor tubing should be attached to the respiration connector of the wPPU Wireless Module (Figure 6). Briefly observe the patient’s breathing to determine proper placement of the respiratory bellows. A belly band can be used to hold the respiratory bellows in place. The band must be tight enough so the bellows can react to the patient’s breathing, but not too tight as to constrict the patient’s breathing (Figure 7). The respiration sensor tubing should lie along the body axis, and the wPPU Wireless Module should be placed in a secure position outside the FOV. The Waveform can be monitored on the WIT Monitor on the gantry (Figure 8), or on the Waveform window on the console before scanning begins (Figure 9). This is done to ensure proper placement of the respiratory gating equipment on the patient, as well as to ensure proper working order of the equipment, resulting in a strong, steady signal.
TIGRE is a grouping of scan parameters that results in a fast T1-weighted 3D RSSG sequence with fat suppression. It is available on the Echelon, Oasis, and Echelon OVAL systems. TIGRE delivers high spatial resolution, as well as high temporal resolution, and is outstanding when used for dynamic, contrast-enhanced imaging of the abdomen (Figure 10). The use of a segmented fat sat pulse maintains fat suppression throughout the scan. TIGRE also uses an echo allocation method called TPEAKS (Triggered Peak Enhancing Artery K-space filling Sequence). This method involves the filling of k-space in an elliptic centric fashion, which ensures the consistent capturing of the critical arterial phase of contrast enhancement.
When positioning a patient for an abdominal exam in an open MRI system, it is important to consider the choices you have for coil selection and patient setup. Positioning that will result in both the coil and the anatomy at isocenter in all three planes will have a positive impact on the image quality of your study.
Components of positioning to consider include coil selection, table pads and accessory pads, and proper laser light centering. Coil selection will be influenced by the patient’s size and body habitus. The extensive inventory of table and accessory pads should be used for proper centering, as well as patient comfort and stability (Figure 11). The use of trough pads and thick or thin table pads are keys to achieving coronal centering of both the patient and the coil. When scanning on open MRI systems, it is extremely important to center the anatomy of interest in the laser lights (Figure 12) in all three directions: head-to-foot (axial or transverse plane), right-to-left (sagittal plane), and anterior-to-posterior (coronal plane).
The coil of choice for a study of the liver, pancreas, or kidneys is the RAPID body coil. Fit the base of the coil in the table trough. The longitudinal center mark on the coil will be centered at the midline of the table. The horizontal center mark on the coil should be between the marking on the patient table and the end of the table nearest the magnetic field. This will allow for sufficient table travel while still achieving isocenter positioning. The patient should be positioned supine on the base portion of the coil, with their midline aligned with the longitudinal center mark on the coil. A point approximately three fingers below the patient’s xiphoid process, at the lower costal margin, should be aligned with the horizontal centering mark on the coil. Coil pads, as well as table pads and sponges, can be added to or removed from the coil to maintain the patient in the center of the coil in the coronal plane. Coronal centering is especially important for MRI of the kidneys, as these organs are retroperitoneal, and lie posterior to the liver and pancreas. The upper portion of the coil is then placed on the base and pushed firmly into place to lock the coil (Figures 17 and 18). The patient should now be centered in all three planes - centered on the middle of the abdomen in the coronal and axial planes, and centered midline in the sagittal plane.
The flexible body coils can be used for abdominal scanning to accommodate larger patients. However, they do not have RAPID capabilities, and should not be used with protocols that are labeled as “RAPID”. Trough and/or table pads should be placed on the table to help center the flexible coil in the coronal plane. The horizontal center of the flexible body coil should be between the marking on the patient table and the end of the table nearest the magnetic field. This will allow for sufficient table travel while still achieving isocenter positioning. The longitudinal center mark on the bottom of the coil should be aligned with the sagittal laser light. The patient should be positioned supine on the coil, with their midline aligned with the longitudinal center mark on the coil. A point approximately three fingers below the patient’s xiphoid process, at the lower costal margin, should be aligned with the horizontal centering mark on the coil. Depending on the patient’s body habitus, table and/or accessory pads may have to be adjusted to maintain coronal centering. Close the flexible body coil around the patient and secure the latches. Accessory pads can be placed between the flex coil and the patient to maintain the anterior portion of the coil in a level position, which will minimize stress on the latches (Figure 19). Table straps may be used to further secure the flexible coil. The patient should now be centered in all three planes - centered on the middle of the abdomen in the coronal and axial planes, and centered midline in the sagittal plane (Figure 20).
For those patients that cannot be accommodated in the abovementioned coils, the Oasis and Altaire systems each have an integrated Transmit/Receive Body coil. This coil is not recommended for use for abdominal imaging on a routine basis, and is not RAPID compatible. Specific parameter changes should be made before using the integrated coil. The patient can be positioned on the table either head first or feet first. However, the patient’s upper abdomen must be positioned between the marking on the patient table and the end of the table nearest the magnetic field to allow for sufficient table travel to achieve isocenter positioning. Trough and/or table pads are placed on the table as needed, depending on the patient’s size and body habitus. The patient’s midline should be centered in the sagittal laser light, and their upper abdomen should be centered in the axial laser light.
Both the RAPID Body coils and the Flexible Body coils allow for patient positioning for upper abdomen exams in either the head first or feet first position. Keep in mind that the horizontal center marking on the coil must remain between the marking on the patient table and the end of the table nearest the magnetic field to allow for sufficient table travel to achieve isocenter positioning. The large knee bolster cushion can be placed under the patient’s knees for added comfort. It is preferable to have the patient positioned with their arms over their head, with no interlocking of their fingers. If this position cannot be tolerated, the patient can place their arms at their sides, outside of the coil. Positioning pads or sponges should be placed under the patient’s arms to avoid direct contact with the patient table or coils. This position may allow an increase in artifacts from aliasing, and protocol adjustments may need to be made.
The 1.5T Echelon system comes equipped with a variety of trough and table pads, as well as various positioning pads and sponges (Figure 21). These positioning aids can be used to support the position of both the patient and the coil, as well as to keep your patient comfortable and secure. The magnet bore of the Echelon does not permit lateral patient table movement. However, accurate patient and coil positioning, and correct use of the sagittal and axial laser lights (Figure 22) to center the patient’s anatomy ensures that high quality images will be acquired through isocenter scanning.
The coil of choice for a study of the liver, pancreas, or kidneys is the RAPID Torso/Body coil. Trough and/or table pads are placed on the table as needed, depending on the patient’s size and body habitus. The lower portion of the coil should be positioned in the middle of the table on top of the pads. The exact placement of the coil on the table, and the direction it faces, are based on the patient’s preference to enter the scanner head first or feet first. The patient should be in a supine position on the lower portion of the coil, with a point approximately three fingers below the patient’s xiphoid process, at the lower costal margin, aligned with the horizontal center of the coil. The patient’s feet should be pointed in the same direction as the coil cables and plugs. The upper portion of the coil is then placed over the patient, aligned with the lower portion of the coil, and secured with the Velcro® straps (Figure 24). The midline of the patient’s body should be aligned with the longitudinal center of the upper portion of the coil. The large knee bolster cushion can be placed under the patient’s knees for added comfort. It is preferable to have the patient positioned with their arms over their head, with no interlocking of their fingers. If this position cannot be tolerated, the patient can place their arms at their sides, outside of the coil. Positioning pads or sponges should be placed under the patient’s arms to avoid direct contact with the patient table or coils. This position may allow an increase in artifacts from aliasing, and protocol adjustments may need to be made.
This integrated coil is the coil of choice for very large patients, as well as high-anxiety or claustrophobic patients. These patients can be scanned without a coil being placed over or around their bodies. However, use of the T/R body coil in this manner is not recommended for routine day-to-day scanning, and this coil is not RAPID compatible. Specific parameter changes should be made before using the integrated coil. The patient can be positioned on the table either head first or feet first. Trough and/or table pads are placed on the table as needed, depending on the patient’s size and body habitus. The patient’s midline should be centered in the sagittal laser light, and their upper abdomen should be centered in the axial laser light.
The 1.5T Echelon OVAL system incorporates the WIT (Workflow Integrated Technology) RF coils for upper abdominal scanning. Eight element and twelve element WIT Spine coils are integrated in the table. The WIT Spine coils are used in combination with the eight element WIT Torso coil to provide uniform, volumetric coverage over a large field of view, which is essential for dynamic abdominal scans. RAPID is available for use with the WIT coils. The Echelon OVAL system incorporates table pads that are always in place between the Posterior WIT spine coils and the patient’s body. Numerous positioning pads and sponges are furnished and should be used for patient comfort and safety.
These three coils are the coil combination of choice for upper abdominal imaging on the Echelon OVAL system. The Posterior WIT spine coils are plugged directly into the table. The WIT Spine 8 coil should remain in the center of the table, with the capability to move the WIT Spine 12 coil either above or below the WIT Spine 8 coil, depending on the patient’s preference to enter the scanner head first or feet first. Table pads fit directly on top of the WIT Spine coils, so there is not direct contact between these coils and the patient. The patient is then positioned on top of the table pads, in either a head first or feet first manner, with their midline in the center of the table. The WIT Torso coil is positioned on top of the patient, with the midline of the coil centered on the patient’s midline. The horizontal center of the coil should be positioned at a point approximately three fingers below the patient’s xiphoid process, at the lower costal margin. The WIT Torso coil is then secured to the Posterior WIT Spine coils using the Velcro® straps (Figure 27). The large knee bolster cushion can be placed under the patient’s knees for added comfort. It is preferable to have the patient positioned with their arms over their head, with no interlocking of their fingers. If this position cannot be tolerated, the patient can place their arms at their sides, outside of the coil. Positioning pads or sponges should be placed under the patient’s arms to avoid direct contact with the patient table or coils. This position may allow an increase in artifacts from aliasing, and protocol adjustments may need to be made.
This integrated coil is the coil of choice for very large patients, as well as high-anxiety or claustrophobic patients. These patients can be scanned without a coil being placed over or around their bodies. However, use of the T/R body coil in this manner is not recommended for routine day-to-day scanning, and this coil is not RAPID compatible. Specific parameter changes should be made before using the integrated coil. The patient can be positioned on the table either head first or feet first. Trough and/or table pads are placed on the table as needed, depending on the patient’s size and body habitus. The patient’s midline should be centered in the sagittal laser light, and their upper abdomen should be centered in the axial laser light.
The main job of the liver is to filter blood from the digestive tract before passing it to the rest of the body. The liver may be responsible for up to 500 separate functions, usually in combination with other systems and organs. It detoxifies chemicals and metabolizes drugs, secreting bile that ends up back in the intestines while doing so. The liver makes proteins that are important for blood clotting and other functions. It is involved in carbohydrate metabolism, protein metabolism (synthesis and degradation), and amino acid synthesis. In lipid metabolism, the liver is involved in cholesterol synthesis, the production of triglycerides (fats), and the synthesis of the bulk of lipoproteins. The liver breaks down insulin and other hormones, and converts ammonia to urea (urea cycle). It also stores glucose, in the form of glycogen, as well as vitamins A, D, B12, K, iron, and copper. With all of these functions to perform, it is easy to understand how liver disease can have such a debilitating effect on the entire body.Anatomy
The liver is the largest gland in the body, and the second largest organ in the body (the skin is the largest organ). It occupies the entire upper right quadrant of the abdominal cavity. The liver is wedge-shaped when seen from the side, with a rounded upper border, and a thin, sharp inferior border. The base of the liver is to the right, and the apex towards the left. The upper surface of the liver can typically be percussed (gently tapped for diagnostic purposes) at the level of the fifth intercostal space. The liver weighs approximately 3 pounds when healthy, but can be over 22 pounds when diseased, as with chronic cirrhosis.
From an anterior perspective, the liver is divided into a large right lobe and a smaller left lobe by the falciform ligament (Figure 28). The posterior or visceral surface has two additional lobes between the right and left lobes, which are the caudate lobe superiorly, and the quadrate lobe inferiorly (Figure 29). The caudate and quadrate lobes are divided by the transverse fissure, or porta hepatis. The ligamentum venosum and ligamentum teres serve to divide the caudate and quadrate lobes from the large left lobe. The right sagittal fossa separates the right lobe of the liver from the caudate and quadrate lobes.
The liver is enveloped in visceral peritoneum, which is a thin, double-layered membrane that serves to reduce friction with other organs. In many areas, this membrane is referred to as a “ligament”, which is a recognizable surface landmark, but not at all related to the anatomic ligaments found in various joints. The falciform ligament is the only functional ligament, attaching the anterosuperior surface of the liver to the anterior abdominal wall and the diaphragm. The free edge of the falciform contains the ligamentum teres hepatis, or round ligament, which is attached to the inferior surface of the liver. The falciform is also continuous with the anterior layer of the coronary ligament, located on the superoposterior surface of the liver. The anterior and posterior layers of the coronary ligament unite on their ends to form the right and left triangular ligaments, respectively (Figure 30). Between the two leaves of the coronary ligament, and to the right of the inferior vena cava, is the bare area of the liver. This area is not enclosed in visceral peritoneum, but instead lies flush against the fascia covering the diaphragm. The bare area is clinically important because it represents a site where infection can spread from the abdominal cavity to the thoracic cavity. The posterior layer of the coronary ligament continues to the right as the hepatorenal ligament. The hepatorenal pouch is the area below the posterior layer of the coronary ligament and the right triangular ligament over the right kidney. The hepatogastric and hepatoduodenal ligaments help to connect the liver to the lesser omentum that covers the lesser curvature of the stomach and the first part of the duodenum (Figure 31).
Liver cells are divided into polyhedron-shaped lobules, with sinusoids that converge on a central vein. The central veins are tributaries of larger veins, which merge to form three hepatic veins (right, middle, and left) at the posterior superior aspect of the liver. The three hepatic veins carry blood out of the liver, joining the inferior vena cava just inferior to the diaphragm. At the corners of the lobules, a triad is formed by hepatic artery and portal vein branches, and bile ducts (Figure 32). The hepatic artery branches supply the cells, the portal vein branches feed into the sinusoids, and the bile ducts drain the bile ductules. Liver cells discharge products into the sinusoids (except bile), while absorbing nutrients and non-nutrients from the sinusoids. Liver cells are quite busy - they store and release proteins, carbohydrates, lipids, iron, and certain vitamins (A, D, E, K); they detoxify many harmful ingested substances; they manufacture urea from amino acids, and bile from pigments and salts. Bile is an alkaline compound which aids in digestion via the emulsification of lipids. It is discharged from liver cells into surrounding bile canaliculi (small canals), which merge to form bile ductules. The ductules join the bile ducts, which are part of the triad of the liver lobules mentioned above. Bile is brought out of the liver by the right and left hepatic ducts, which merge at the porta hepatis to form the common hepatic duct (Figure 33). The common hepatic duct joins with the cystic duct from the gall bladder to form the common bile duct, which travels to the pancreas.
The liver is a highly vascular organ with a unique dual blood supply (approx.1500 mL/min.) from both the proper hepatic artery (20- to 40-percent) and the portal vein (60- to 80-percent). The common hepatic artery comes off the celiac trunk at the trifurcation with the splenic artery and the left gastric artery. It gives off the gastroduodenal artery, and continues on as the proper hepatic artery, staying to the left of the bile duct and in front of the portal vein. In the hepatic hilum, the common hepatic artery divides in a Y-shaped manner into the right and left hepatic arteries (Figure 34). The right hepatic artery ascends behind the common hepatic duct. The cystic artery, which supplies the gall bladder, usually branches off the right hepatic artery.
The portal system of veins is named as such because it transports nutrients and other molecules from the first capillary network in the intestines (the organs being drained) directly to the second capillary network (sinusoids) of the liver without going through the heart first. The portal venous system has no valves. The portal vein is formed from a union of the superior mesenteric vein and the splenic vein. It collects blood from the GI tract, the spleen, and the pancreas. The portal vein ascends behind the common bile duct and the hepatic artery proper, and divides in a T-shaped manner into the right and left portal vein branches in the hepatic hilum. After the blood in the portal veins travels through the sinusoids of the liver, it is conveyed to the hepatic veins, and on to the inferior vena cava.
The right, middle, and left hepatic veins were mentioned previously as the vessels that drain blood from the liver to the inferior vena cava. They also serve to further divide the liver into segments or sections (Figure 35). The right hepatic vein divides the right lobe into anterior and posterior sections. The middle hepatic vein demarcates the true right and left lobes of the liver. The left hepatic vein divides the left lobe into medial and lateral sections, with the medial section also known as the quadrate lobe. The fissure for the ligamentum teres also serves to separate the medial and lateral sections. The caudate lobe (Number 1 in Figure 35) is a separate structure, and receives blood flow from both the right and left-sided vascular branches.
The hilum or porta hepatis is the central area where the common bile duct, hepatic portal vein, and the hepatic artery proper enter the liver (Figure 36). The duct, vein, and artery divide into right and left branches and supply the functional right and left lobes of the liver.
Blood tests are used to diagnose normal liver function, as the liver may only produce symptoms of disease after extensive damage has already been done. Hepatitis, or inflammation of the liver, usually results from viruses such as hepatitis A, B, or C. Inflammation can also have non-infectious causes, such as heavy drinking, drugs, allergic reactions, and obesity. Hemochromatosis is a hereditary disease that causes the accumulation of iron in the body, eventually leading to liver damage. Primary sclerosing cholangitis causes inflammation and scarring in the bile ducts in the liver. Primary biliary cirrhosis is a rare disorder that slowly destroys the bile ducts in the liver, leading to permanent liver scarring, also known as cirrhosis. There are multiple causes of long-term damage to the liver that may result in cirrhosis, including viral hepatitis and alcoholism. These conditions can cause the death of liver cells; the dead cells are then replaced by fibrous tissue that forms in the liver. Cirrhosis may lead to a decrease in the superoinferior span of the liver, as well as caudate lobe hypertrophies (Figure 37). Over time, jaundice and portal hypertension may result. Jaundice, an accumulation of bile pigment in the blood stream, can be caused by the obstruction of the bile duct system, which is part of the anatomy of the liver.
MR imaging of the liver is important for characterization of metastases or primary liver tumors, as well as identification of benign lesions. In MR imaging, spoiled gradient echo sequences are typically ideal for T1-weighting (Figure 38). In-phase sequences also give strong T1-weighting, and can typically be acquired in one breathhold. They can also minimize effects from magnetic susceptibility. FSE or single shot techniques, often combined with fat suppression, are the most commonly used T2-weighted sequences. Patients with chronic liver disease (i.e., cirrhosis, hepatitis B and C) are at higher risk for developing hepatocellular carcinoma (HCC), which is the most common primary liver cancer. Cholangiocarcinoma, the second most common primary malignant liver tumor, is an adenocarcinoma that originates in the epithelium of the biliary ducts (Figure 39). Its origin can be in secondary intrahepatic ducts, or in the main right and left hepatic ducts closer to the hilum. Primary liver cancers can spread by draining into the hepatic lymph nodes at the porta hepatis, as well as by moving into the lymph nodes in the hepatoduodenal ligament.
Due to its high amount of vascularity, the liver is one of the most frequently involved organs as a site of metastasis, second only to the lymph nodes. Hepatic metastasis is the most frequently found malignant lesion of the liver. Metastatic spread to the liver is most common from tumors of the colon, lungs, pancreas, and stomach (Figure 40). Most metastases are hypovascular and present as hypoattenuating nodules in relation to the hepatic parenchyma in the portal-venous phase, with heterogeneous or ring-shaped contrast-enhancement.
Common benign liver lesions, such as hemangiomas, focal nodular hyperplasia (FNH), and adenomas, benefit from MR imaging for lesion detection as well as lesion characterization. Hemangiomas are the most frequent benign tumor of the liver. They typically display on MR imaging with well-defined margins, and hyperintense signal on T2-weighted sequences (Figure 41). Focal nodular hyperplasia (FNH) is the second most frequent benign hepatic tumor. It is defined as a nodule with apparently normal hepatocytes that occurs in a liver with normal histology. It is more common in women and young patients, sometimes in association with hepatic hemangiomas. Vascular malformation and/or vascular injury are suggested as probable mechanisms for the development of focal nodular hyperplasia. It often presents with a hypointense lesion on T1-weighted images, and a subtle hyperintensity on T2-weighted images (Figure 42). In a large majority of FNH lesions, a central scar can be identified with a higher intensity signal than the rest of the lesion on T2-weighted images. Hepatocellular adenomas are rarer, and usually found in women with a history of long-term use of oral contraceptives.
A liver lobe can be removed from a live donor for transplant. Intraoperative ultrasound is often used during transplant procedures to delineate blood vessels and bile ducts. The liver is the only human organ capable of natural regeneration of lost tissue. Starting with as little as 25-percent of the original liver, a whole liver can be regenerated. This is not true regeneration, but rather compensatory growth. A removed lobe does not regrow; liver growth is measured as a restoration of original function, not original form. In true regeneration, both original function and form are restored.
MR of the liver is highly dependent on the administration of contrast, especially when the detection and characterization of focal lesions are the main issues. Dynamic MRI with multiple phases (arterial, portal, and late phase) after IV contrast administration offers important information (Figure 43). Arterial phase imaging is critical for the early detection of hepatocellular carcinoma, as arterially enhancing lesions greater than 1cm are likely to be malignant. In addition, malignant lesions are typically T2 hyperintense. Normal gadolinium chelate contrasts offer only a brief window of time (seconds) for imaging before the contrast diffuses to the interstitial space of both healthy liver and liver lesions, reducing the contrast gradient necessary for easy lesion detection. When combined with MRCP, liver MRI is very useful in the evaluation of patients with hepatic and biliary disease. Hepatobiliary contrast agents are available that are specific for liver uptake and excretion via the biliary system. Normal hepatocytes will take up the paramagnetic substance in the contrast agent (i.e., manganese, gadolinium). Diseased liver tissue will not include hepatocytes, or may include those whose function is disturbed. The end result is an increase in signal of healthy liver tissue on T1 weighted sequences, and no increase in signal in liver lesions. Hepatobiliary contrast media are quickly excreted through the biliary system, allowing better opacification of the biliary ducts and the gall bladder. Superparamagnetic iron oxides are another type of liver imaging contrast agent, which may increase the accuracy of detection of small metastases. These particles accumulate in the reticuloendothelial system of the liver, darkening the healthy liver tissue in T2 weighted images. In altered tissue with either reduced or no reticuloendothelial activity, the contrast agent concentration is low or non-existent, which improves the overall liver to lesion contrast. One drawback is that the reticuloendothelial-targeted agents remain in the body longer, which could increase possible side effects.
The following are HMSA suggestions for liver imaging. Always check with your radiologist for his/her imaging preferences. Liver imaging typically involves the use of breath holds or respiratory gating to eliminate artifacts caused by abdominal motion during breathing. Dynamic contrast-enhanced studies of the liver should be considered a routine part of a comprehensive MR exam of the liver, in combination with unenhanced T1- and T2-weighted images.
Axial slices of the liver can be set up using coronal and sagittal images or scanograms. Slices should be prescribed in a sufficient number to cover the area from the diaphragm to the inferior margin of the liver, as seen in Figure 44. Some references recommend coverage to include the dome of the liver superiorly, and the aortic bifurcation inferiorly.
Coronal slices of the liver can be set up using axial and sagittal images or scanograms. Slices should be prescribed in a sufficient number to cover the area from the posterior abdominal muscles to the anterior abdominal wall. The slices should include the area from the symphysis pubis to the diaphragm, as seen in Figure 45.
Sagittal scans are not typically used in liver imaging.
Saturation pulses or “presats” may be placed superiorly and inferiorly to eliminate flow motion from the aorta and inferior vena cava. In addition, presats placed superiorly may help eliminate gastric motion, and presats placed inferiorly may help eliminate bowel motion.
MRI has emerged as the superior modality for liver lesion characterization and detection, due in part to the use of dynamic scanning techniques. Dynamic MR scanning of the liver involves the use of a bolus injection of contrast, with precise timing of imaging in order to acquire images in the arterial, portal venous, and equilibrium phases. Malignant and benign lesions can occur concurrently in the same patient, which makes correct characterization of these lesions in each specific phase of utmost importance for therapeutic decisions.
To capture the arterial phase, imaging should optimally take place between 20-30 seconds after the start of the injection. The hepatic arteries and often, the main portal vein branches, will be opacified, and show high signal on T1-weighted images (Figure 46). Hepatic veins and the liver parenchyma should not be enhanced. Some hypervascular metastases and primary hepatocellular carcinomas have blood supply only from the hepatic artery, making this phase of imaging crucial for their detection.
The portal venous phase occurs approximately 55-60 seconds after the bolus contrast injection. The liver parenchyma, as well as both the hepatic and portal veins, should show enhancement. Hypovascular tumors and metastases can be well visualized, as they will appear as hypointense lesions relative to the enhancing liver. This phase also offers excellent visualization of the hepatic and portal vein branches, and easier depiction of the segments of the liver, which is beneficial for surgeons. The portal venous phase displays the greatest differences in liver-to-lesion signal intensity, as the greatest enhancement of liver parenchyma occurs at this time.
The equilibrium phase occurs between 3 and 5 minutes after the contrast injection. The gadolinium chelate has now diffused out of the vascular compartment into the interstitial space of the normal liver parenchyma and tumors. The majority of tumors are less visible in this phase. However, cholangiocarcinomas will have increased conspicuity on delayed images, due to their prominent interstitial space, which accumulates more contrast than the adjacent normal liver. Hemangiomas will usually fill in, and some extrahepatic diseases are well visualized, due to their slow accumulation of gadolinium. Some lesions with delayed contrast uptake become isointense or hyperintense 12-15 minutes after the contrast injection. This results from contrast washout from normal liver tissue with concomitant contrast pooling in the diseased liver.
Dynamic liver examinations on the Hitachi Oasis, Echelon, and Echelon OVAL can be performed with breath hold instructions or with Auto Voice, and the Stopwatch feature. These examinations are explained in the How-to Manuals for each system. The FLUTE, or Fluoro Triggered Exam feature can also be used with dynamic liver scanning for more precise timing, especially for the arterial phase of imaging.
The pancreas is essential to the proper processing of food, from digestion through the absorption, storage, and utilization of nutrients. It is a dual function gland with both endocrine and exocrine functions, secreting both hormones and enzymes. The hormones secreted into the bloodstream by the endocrine gland include insulin and glucagon, which regulate the level of glucose in the blood. The pancreas also secrete the hormone somatostatin, which prevents the release of insulin and glucagon. The exocrine gland of the pancreas secretes inactive digestive enzymes into a network of ducts. These ducts join the main pancreatic duct, which runs the length of the pancreas. From the pancreatic duct, the enzymes move into the bile duct, and on to the duodenum, where they are activated. These enzymes help break down carbohydrates, fats, proteins, and acids in the duodenum. The exocrine tissue also secretes bicarbonate to neutralize stomach acid in the duodenum.Anatomy
The name “pancreas” comes from the Greek, meaning “all flesh”, appearing as a pale, rubbery gland in the epigastric and left hypochondriac regions of the abdomen. It is approximately 6 inches long, and tapers as it extends across the back of the abdomen and behind the stomach (Figure 47). The right side of the organ, called the head, is its widest part, and lies in the curve of the duodenum, which is the first section of the small intestine (Figure 48). The pancreas connects to the duodenum through the pancreatic duct, a small tube that runs the length of the gland. The uncinate process emerges from the lower part of the head, lying deep to the superior mesenteric vessels. The neck of the pancreas is the constricted portion between the head and the body. The body lies behind the stomach, extending slightly upward, and leading to the tail, which is the narrow end of the pancreas on its left end. The tail lies in contact with the spleen, and runs in the lienorenal, or splenorenal, ligament. The tail is the only intraperitoneal portion of the pancreas. Due to the deep location of the pancreas, tumors are rarely palpable. Often times, patients will have no symptoms of pancreatic problems until a tumor grows large enough to interfere with the function of nearby structures, such as the stomach, duodenum, liver, or gall bladder.
The duct of Wirsung is the main pancreatic duct that extends from the tail of the organ to the major duodenal papilla, or ampulla of Vater (Figure 49). The duct is typically a width of 2 mm. at the tail, widening to approximately 4 mm. at the pancreatic head. It is close, and almost parallel, to the distal common bile duct. The pancreatic and common bile ducts combine to from a common duct channel prior to approaching the duodenum. The common bile duct, which is the union of the cystic duct and common hepatic duct, is approximately 8 cm. long. It descends behind the first part of the duodenum, deep to or through the head of the pancreas. The common bile duct and pancreatic duct join, and together form the hepatopancreatic ampulla and sphincter in the wall of the second part of the duodenum. It is through this ampulla that the common bile duct expels its bile, and the pancreatic duct expels its pancreatic juice to aid in digestion. The hepatopancreatic ampulla is typically located at the major duodenal papilla (ampulla of Vater), which is a nodular protuberance on the medial wall of the duodenum. The muscles that surround the ends of the common bile duct (biliary sphincter) and pancreatic duct (pancreatic sphincter) fuse at the level of the ampulla of Vater, and are then known as the sphincter of Oddi. The dynamic structure of the sphincter of Oddi allows for relaxation and contracture to change the dimensions of the ampulla of Vater, thereby controlling the amounts of bile and pancreatic juices that flow into the duodenum. Approximately 70-percent of people have an accessory dorsal pancreatic duct, known as the duct of Santorini. This dorsal duct may communicate with the main pancreatic duct, but the degree of communication between the dorsal and ventral ducts will vary.
The exocrine and endocrine functions of the pancreas are both vital for proper “feeding” of body tissues. The external secretions resulting from the exocrine functions of this gland are necessary for digestion and absorption of foodstuffs. The endocrine function secretes internal hormones that organize the use or storage of nutrients after their intake to the bloodstream.
The exocrine function of the pancreas results in a pancreatic juice that contains enzymes capable of breaking down the three major components of food-carbohydrates, fats, and proteins. The pancreas is divided into units called acini (acinus is Latin for berry). Each acinus is spherical, with enzyme-secreting cells surrounding a central space (Figure 50). Every enzyme-secreting cell synthesizes all the pancreatic enzymes. The enzymes pass into the center of the acini, entering the narrowest ducts of the branching secretory system. The enzymes pass by larger and larger ducts, eventually reaching the single pancreatic duct, or duct of Wirsung. The cells that line the duct system secrete water and bicarbonate ions, adding them to the enzymes. Consequently, the final pancreatic juice is alkaline. The volume of juice (approximately 2000 ml per 24-hour day) that is secreted is precisely enough to neutralize the acid contents of the stomach as they both enter the duodenum. Pancreatic enzymes are most effective when the contents of the duodenum are at a neutral pH. Chronic disease of the exocrine component of the pancreas results in deficiency of these pancreatic enzymes, giving rise to poor absorption of foodstuffs. The most conspicuous feature of malabsorption is the excretion of fats in feces.
External pancreatic secretion is regulated by hormones, primarily secretin and cholecystokinin. The homeostasis that results when alkaline pancreatic juices mix with acidic gastric juices causes the release of the hormone secretin from cells in the walls of the duodenum. Secretin passes into the bloodstream and stimulates the production of water and bicarbonate ions from the duct system of the pancreas (Figure 51). A greater volume of acid gastric juice passing into the duodenum triggers a greater volume of bicarbonate-rich juice provided by the pancreas to keep the duodenal contents neutral. The pH balance is maintained, and the pancreatic enzymes can perform at their most effective level. The hormone cholecystokinin, which means “gall bladder mover”, is also synthesized in the walls of the duodenum, and is released in response to the presence of amino acids and fatty acids, as partly-digested food starts arriving from the stomach. Cholecystokinin passes around the circulation and causes enzyme secretion by the pancreatic acinar cells. This enzyme secretion increases the ability of the pancreatic juice to break down more fats and proteins. Cholecystokinin also causes contraction of the gall bladder, which provides bile that promotes the absorption of fatty acids and glycerol. Bile from the liver, as well as pancreatic juices, are necessary for the absorption of fat. The common bile duct and the pancreatic duct that bear these fluids converge, so the fluids enter the duodenum together.
The endocrine functions of the pancreas result in the internal secretion of the main hormones insulin and glucagon. Both hormones are necessary for regulation of the storage, release, and utilization of the fuels for metabolism. Insulin lowers the blood sugar, while glucagon raises it. Both hormones are also important in the body’s handling of the nutrients that are derived from fats, proteins, and carbohydrates.
Among the acini and ducts that secrete enzymes in the pancreas are small clumps of endocrine tissue that secrete hormones, but do not connect with a duct system (Figure 53). These small islands, called islets of Langerhans, make up a small fraction of the bulk of the pancreas, but have a vital function (Figure 52). These islets have two main types of endocrine cells-one that produces glucagon (alpha cells) and one that produces insulin (beta cells). The term insulin comes from the Latin word “insula”, which means an island. The two hormones have opposite effects on the level of glucose in the blood. In the liver, these two hormones have contrary effects on the balance between storage of glucose and its release into the blood. They also have opposite effects on the new formation of glucose from amino acids. In fatty tissue, they have opposing effects on storage vs. release of fuels. Insulin facilitates the uptake and usage of glucose by body tissues, especially muscle. Decreased insulin secretion, or a decrease in the numbers or activity of insulin receptors leads to glucose intolerance and/or diabetes mellitus. A third type of endocrine cells, the delta cells, produce somatostatin. It inhibits the release of both insulin and glucagon by the pancreas. Somatostatin also inhibits the release of gastrin by the gastric mucosa, and the release of secretin by the intestinal mucosa.
The insulin:glucagon, or I:G ratio, is important and variable. The balancing act between insulin and glucagon maintains the blood glucose level for supplying the brain, matches the nutrient supply to the immediate needs of the body’s tissues, and stores the surplus. Insulin takes precedence to facilitate the uptake and storage of nutrients when they flood into the blood after digestion of a meal, so the I:G ratio is high. Glucagon promotes the release of glucose and fatty acids into the circulating blood when the use of fuels for energy are at their peak during muscular work, so the I:G ratio is rather low. Glucagon is also of major importance during fasting. The levels of glucose and amino acids in the blood supplying the pancreas cause a response from the secretory cells to regulate these counter balancing secretions; i.e., a rise in blood glucose affects cell membrane receptors on beta cells, resulting in an increase in the synthesis and extrusion of insulin. The islet cells secrete additional hormones, such as gastrin and pancreatic polypeptide. Their interactions with glucagon and insulin are current research topics.Arteries and Veins of the Pancreas
Numerous arteries with sources springing from the celiac and superior mesenteric arteries supply the pancreas (Figure 54). The superior pancreaticoduodenal artery (from the gastroduodenal artery), and the inferior pancreaticoduodenal artery (from the superior mesenteric artery), run in the groove between the pancreas and the duodenum and supply the head of the pancreas. Pancreatic branches of the splenic artery supply the neck, body, and tail. The largest of these branches is the arteria pancreatica magna, whose occlusion is rare, but fatal.
The extensive capillary networks of the pancreas are drained by tributaries of the hepatic portal vein, which conducts secreted hormones from the pancreatic islets to the liver and beyond for general circulation. The neck and body of the pancreas drain into the splenic vein (Figure 55). The pancreatic head drains into the superior mesenteric and portal veins.
The lymph nodes of the pancreas are distributed along the major vascular pathways (Figure 56). Lymph nodes are tiny bean-shaped organs that can be found throughout the body. They help the body fight infection. Lymph is drained via the splenic, celiac, and superior mesenteric lymph nodes. Lymph node involvement around a tumor is part of the TNM staging system. The letter “T” with a letter or number describes the size and location of the tumor; the letter “N” with a letter or number stands for (lymph) node involvement; the letter “M” with a letter or number is used for metastasis. The TNM system can be used with additional staging information to determine whether or not a tumor resection is a viable option.
In addition to the hormonal and chemical mechanisms for regulating the endocrine and exocrine functions of the pancreas, the autonomic nervous system prepares the pancreas to deal with food that is on its way. Branches from the network of autonomic nerves send signals related to events in the stomach and duodenum. Parasympathetic fibers from the vagus nerve stimulate enzyme secretion in response to eating, before the meal reaches the duodenum. Nerves that supply the hormone-secreting cells are from both the sympathetic and parasympathetic systems, which respectively inhibit and promote insulin release, with the reverse action on glucagon. These actions are in accord with nutrient mobilization from body stores during exercise and stress, and, on the other hand, the need for storage after digestion of a meal.MRI and Diseases of the Pancreas
There are a large number of diseases and conditions that affect the pancreas. Type 1 diabetes, formerly called juvenile diabetes or insulin-dependent diabetes, occurs when the body’s immune system attacks and destroys the pancreas’ insulin-producing cells. This results in lifelong insulin injections to control blood sugar. Type 2 diabetes, formerly called adult-onset or noninsulin-dependent diabetes, is the most common form, and occurs when the pancreas loses its ability to appropriately produce and release insulin. The body becomes resistant to the insulin, and blood sugar rises. Diabetes can be a symptom of pancreatic cancer, and may also increase the risk of pancreatic cancer development for those with long-standing adult-onset diabetes. Cystic fibrosis is a genetic disorder that affects multiple systems, usually including the lungs and pancreas. Digestive problems and diabetes often accompany cystic fibrosis, so pancreatic enzymes can be given orally to replace those that the malfunctioning pancreas is no longer making. Enlargement of the pancreas may be normal, or may be an anatomic abnormality. It is a condition that should be monitored by a physician. MRI may be helpful in detecting or excluding a neoplasm in patients with indeterminate pancreatic enlargement (Figure 57). A fatty abnormality may show increased signal on T1-weighted images, decreased signal on fat suppressed images, and phase cancellation on out-of-phase images. MRI may allow for noninvasive characterization of a fatty abnormality, and the ability to distinguish these abnormalities from cancer.
Pancreatitis is a condition in which the pancreas becomes inflamed and damaged by its own digestive chemicals. This can result in swelling and death of pancreatic tissue. The cause of most pancreatitis is unknown, but alcohol and gallstones may contribute to this condition.
Patients with pancreatic cancer often develop chronic pancreatitis, as inflammation of the pancreas can be related to obstruction of the duct by a pancreatic mass. Pancreatitis and pancreatic cancer have similar MR imaging features: both display as hypointensities on unenhanced T1-weighted fat suppressed images, and both show delayed enhancement on dynamic T1-weighted fat suppressed images with gadolinium (Figure 59). A patient that presents with pancreatitis without an appropriate clinical history should prompt further investigation for an underlying mass. The characteristics of pancreatic duct dilation may suggest chronic pancreatitis or pancreatic cancer as the cause. Pancreatic cancer can present with gland enlargement, abnormal enhancement, and peripancreatic stranding, which all mimic acute pancreatitis. Additional findings, such as retroperitoneal lymphadenopathy, may suggest an underlying malignancy. A pancreatic pseudocyst may be discovered after a bout of pancreatitis (Figure 58). This is a fluid-filled cavity that may resolve spontaneously, or may require surgical drainage.
Pancreatic cancer is one of the most deadly of all cancers, and is the fourth leading cause of cancer death for both men and women. Five-year survival rates are near 25-percent if the cancer is surgically removed while it is still small and has not spread to the lymph nodes. There are currently no reliable screening tests for pancreatic cancer. Symptoms can be vague and easily confused with other diseases, resulting in this cancer being in an advanced stage at the time of discovery. In many cases, the patient is considered “unresectable” at the time of diagnosis.
Pancreatic cancer commonly metastasizes to the lymph nodes, liver, and peritoneum. People at greatest risk for pancreatic cancer include those with 2 or more relatives with pancreatic cancer, cigarette smokers, patients with chronic pancreatitis, people of Ashkenazi Jewish descent, those over age 50, and people with BRCA2, p16, or STK11 gene mutations. Those with diets high in meats, cholesterol fried foods, and nitrosamines may increase their pancreatic cancer risk, while those with diets high in fruits and vegetables may reduce their risk of this cancer.
There are at least 20 different types of tumors under the umbrella of cancer of the pancreas. The pancreas has many different types of cells, and each can give rise to a different type of tumor. In the vast majority of cases, the term “cancer of the pancreas” refers to a primary cancer. Primary cancers are then sub-divided as to whether or not they show endocrine differentiation, which will impact treatment and outcome. The most common type of pancreatic cancer comes from the cells that line the pancreatic duct, and is known as ductal adenocarcinoma of the pancreas.
On MRI, normal pancreas show high signal intensity on unenhanced T1-weighted fat suppressed images because of the presence of acinar proteins. The normal pancreas will show homogeneous intense enhancement in the early arterial phase, becoming isointense to the liver on more delayed enhanced sequences. Conversely, pancreatic adenocarcinoma is hypointense to the normal pancreas on T1-weighted fat suppressed images, and shows decreased enhancement on arterial phase images (Figure 60). The cancer will show progressive enhancement on delayed sequences. These MRI features are related to the fibrotic nature of the tumor. Imaging in the arterial phase yields the greatest conspicuity of pancreatic cancer, which appears hypointense compared with adjacent normal pancreas. The superior soft-tissue contrast found with MRI makes this imaging modality quite useful in the detection and evaluation of subtle non-contour-deforming pancreatic masses. A mass that is too small to deform the pancreatic contour may appear on an MRI as a hypointense mass within the enhancing pancreas on gadolinium-enhanced fat suppressed images. Secondary signs of an underlying mass include focal pancreatic atrophy or pancreatic duct dilation. Dilation of both the common bile duct and the pancreatic duct (double duct sign) may result from either benign or malignant causes, but is most commonly associated with cancer (Figure 61). T2-weighted sequences and an MRCP (Magnetic Resonance CholangioPancreatography) are both valuable when assessing the pancreaticobiliary ducts. MRCP is also valuable when characterizing an obstruction and its location and causes, such as stones or masses (Figure 62).
Pancreatic neuroendocrine tumors, or islet cell tumors, are rare neoplasms of the pancreas that originate from the islets of Langerhans. They are further classified as functioning or non-functioning tumors. Functioning tumors release hormones into the bloodstream, while non-functioning tumors form a mass, but do not cause symptoms by releasing hormones. Islet cell tumors usually grow slower than pancreatic cancers, and can be cured if treated early enough. These types of rare tumors may be found in patients with a familial genetic syndrome, such as Multiple Endocrine Neoplasia Type 1. Examples of islet cell tumors include insulinomas, glucagonomas, and gastrinomas. Insulinomas arise from islet beta cells, and are the most common functioning neuroendocrine tumors of the pancreas that are responsible for typical hypoglycemic symptoms. They produce large quantities of insulin that is released into the bloodstream, causing a dramatic lowering of blood sugar. More than 90 percent of insulinomas are benign, intra-pancreatic solitary tumors. Most insulinomas are quite small and difficult to localize; 55 to 70 percent are less than 1.5 cm (Figure 63). Surgery is the only curative treatment, with pre-op imaging recommended to include MRI for exact localization and surgical strategy. Small insulinomas usually have low signal on T1-weighted images (especially if fat suppressed), and high signal on T2-weighted images. They will typically show strong enhancement in the arterial phase, with prolonged enhancement relative to normal pancreas in the delayed phase. The timing and degree of enhancement can be highly variable, so enhancement may be hard to demonstrate. Fibrosis in pancreatic tumors may reduce blood flow by compressing the arterial supply, leading to a lack in tumor enhancement. DWI and ADC sequences have an important role in the detection and characterization of insulinomas (Figure 64). DWI detects changes in the molecular diffusion of water in biologic tissues, while the degree of water motion can be quantified by ADC. Restricted diffusion translates into high signal intensity on DWI and decreased signal on ADC maps. Reduced ADC is observed with most malignancies.
Glucagonomas are very rare tumors of the islet cells of the pancreas, and are usually malignant (60-percent). This cancer tends to spread and get worse, often times having already spread to the liver by the time of diagnosis. Glucagonomas cause the islet cells to produce too much of the hormone glucagon, which results in high blood sugar, or hyperglycemia. High blood sugar levels can cause metabolic problems and tissue damage (Figure 65). Glucagonomas are usually accompanied by skin lesions that are scaly, erythematous, and hyperpigmented. The cause of glucagonomas is unknown, but genetic factors may play a role in some cases. Preferred treatment is surgery, which will only cure about 20-percent of people with this tumor. If this tumor is found only in the pancreas, and the surgery to remove it is successful, patients have a 5-year survival rate of about 85-percent.
Gastrinomas (Zollinger-Ellison syndrome) involve a small tumor in the pancreas or small intestine that produces a high level of the hormone gastrin in the blood. High levels of gastrin cause the production of an excess of stomach acid. Gastrinomas can occur as single tumors, or as multiple small tumors. Approximately one half to two thirds of single gastrinomas are malignant, often spreading to the liver and nearby lymph nodes (Figure 66). Medications are the first choice for treatment of gastrinomas, specifically those called proton pump inhibitors. These drugs reduce acid production by the stomach, thereby promoting the healing of stomach and small intestine ulcers, and relieving abdominal pain. Surgery to remove a single gastrinoma may be performed, if there is no evidence of metastasis. Unfortunately, the cure rate for gastrinomas remains low, even with early diagnosis and surgery to remove this tumor. Gastrinomas tend to grow slowly, so patients can live for years after the tumor is discovered. Acid-suppressing medications help to control the symptoms of excess acid production.
Another type of pancreatic tumor is an intraductal papillary mucinous neoplasm (IPMN). These tumors form cysts within the pancreatic ducts, both the main duct and its branches (Figure 67). They produce a thick fluid, or mucin, from the tumor cells, in amounts large enough that mucin may be seen oozing from the ampulla of Vater. Branch duct IPMN is typically less aggressive than main duct IPMN. 70-percent of main duct neoplasms harbor pre-cancerous or invasive cancerous tumors. Many of these tumors are found by accident in patients that are having imaging examinations done for other reasons. The patients are either asymptomatic, or have symptoms that are common to multiple diseases. Intraductal papillary mucinous neoplasms are fairly common, especially amongst the elderly (age 80 and above). They can transform from benign tumors to invasive malignant tumors if left untreated. IPMNs have a 95-percent cure rate if there is no invasive cancer at the time of resection. Patients that are asymptomatic, with branch duct IPMN, can be monitored using imaging modalities every 3-6 months. Guidelines to maintain these patients in “monitor” status include no neoplasm greater than 3 cm, no main duct dilatation, and no solid masses.
Metastases from pancreatic cancer are usually found in the liver, peritoneum and omentum, lymph nodes, and blood vessels. The detection of liver metastasis is critical for the proper staging of pancreatic carcinoma, as it could result in the patient being unresectable. MRI is valuable for further delineating small lesions, such as cysts, hemangiomas, or metastases, as they influence both the patient’s work-up and prognosis (Figure 68). Liver metastases are minimally hypointense on T1-weighted images, and isointense to moderately hyperintense on T2-weighted images. Although primary tumors are generally hypovascular, a range of hypovascular to hypervascular liver metastases can be seen with perilesional enhancement after IV gadolinium (Figure 69).
The detection of peritoneal and omental metastases is critical, as it could prevent the patient from undergoing unnecessary surgery. The peritoneum is often involved in metastatic pancreatic carcinoma. MRI may be more sensitive than CT for detecting peritoneal enhancement. Lymph node metastasis commonly involves the peripancreatic and porta hepatis lymph nodes (Figure 70).
Vascular invasion by pancreatic cancer can be a contraindication for surgical resection. Encasement of the celiac, hepatic, or superior mesenteric arteries precludes surgical intervention (Figure 71). Limited involvement of the portal vein or superior mesenteric vein with a pancreatic tumor may be treated through a surgical bypass. Extensive involvement of major venous structures is a contraindication for surgery.
A variety of tests and procedures are available to aid in the diagnosis of pancreatic problems or disease. Elevated levels of amylase and lipase in blood tests can be suggestive of pancreatitis. MRCP (Magnetic Resonance CholangioPancreatography) uses MRI to visualize the biliary and pancreatic ducts. Ultrasound is another imaging modality that is often used to view the pancreas and ducts. ERCP (Endoscopic Retrograde CholangioPancreatography) uses a flexible endoscopic tube that is inserted down the esophagus, through the stomach, and into the duodenum. A physician can inject contrast into the duodenal papilla to check the functioning of the common bile duct and pancreatic duct. This more invasive procedure may be used when the ducts are compromised due to tumors, gallstones, pancreatitis, infection, scarring, etc. Procedures to treat narrowed areas or blockages can be performed during the ERCP examination. Surgical resection for pancreatic cancer is often done using the Whipple procedure. This complex surgery involves the removal of the head of the pancreas, the gall bladder, the first section of the small intestine (duodenum), and sometimes a small part of the stomach. Contraindications for surgical resection include liver and peritoneal metastasis, distant lymph node metastasis, arterial encasements, and greater than 50-percent encasement of major venous structures. Transplantation of the pancreas may be performed for patients with diabetes or cystic fibrosis. The transplanted organ can sometimes cure the diabetes. Islet cell transplantation is an experimental procedure that involves the harvesting of insulin-producing cells from the donor’s pancreas, and their transplantation into a patient with Type 1 diabetes (Figure 72). This procedure could potentially be a cure for Type 1 diabetes. A development in the fight against pancreatic cancer comes from the University of California-Davis. Their studies show that pancreatic cancer cells need the amino acid arginine in order to be able to divide and multiply. By artificially reducing the levels of arginine, researchers have been able to reduce pancreatic cancer cell development by 50-percent. Rather than killing cells, which occurs with chemotherapy, they are removing key building blocks that cancer cells need to function.
The following are HMSA suggestions for pancreas imaging. Always check with your radiologist for his/her imaging preferences. Pancreas imaging typically involves the use of breath holds or respiratory gating to eliminate artifacts caused by abdominal motion during breathing. Examinations of the pancreas often include an MRCP, or Magnetic Resonance CholangioPancreatography, a non-invasive technique used to evaluate the hepatobiliary tree and pancreatic duct.
Axial slices of the pancreas can be set up using coronal and sagittal images or scanograms (Figure 73). Slices should be prescribed to include the diaphragm superiorly, and should extend below the most inferior margin of the liver, as the pancreas lie angled posterosuperiorly.
Coronal slices of the pancreas can be set up using axial and sagittal images or scanograms. Slices should be prescribed in a sufficient number to cover the area from the posterior abdominal muscles to the anterior abdominal wall. The slices should include the area from the symphysis pubis to the diaphragm, as seen in Figure 74.
Sagittal scans are not typically used in imaging of the pancreas.
Saturation pulses or “presats” may be placed superiorly and inferiorly to eliminate flow motion from the aorta and inferior vena cava. In addition, presats may be used to eliminate gastric and bowel motion, as the head of the pancreas lies in close proximity to the stomach and duodenum.
Magnetic Resonance CholangioPancreatography, or MRCP, is a non-invasive technique used to evaluate the hepatobiliary tree and pancreatic duct. An MRCP may be performed for a variety of reasons: to evaluate bile duct obstruction, to determine benign and/or malignant causes of biliary dilatation, to detect cholangiocarcinoma, to determine if there is a post-op bile duct injury, to determine if a patient with recurrent pancreatitis has stones or strictures, to detect pancreatic cancer, and to detect parenchymal changes due to pancreatitis. Preparation of patients may be site and radiologist specific. Some references recommend that the patient be NPO after midnight, and allowed only water, in order to keep the gall bladder distended. Others recommend the use of an oral negative contrast, which could be blueberry or pineapple juice, or Gastromark, which is given to the patient when they arrive in the scan room. The fruit juices contain manganese, which reduces the signal intensities from fluids in the gastrointestinal tract.
The sequences that are run for MRCP scans are typically heavily T2-weighted, in order to emphasize the signal from the fluids in the common bile and pancreatic ducts, which are bile and pancreatic juices, respectively. The fluid in these ducts will appear bright against the darker tissues that surround them. Typical protocols will include coronal sequences that include the common bile duct as it passes through the head of the pancreas anteriorly to the porta hepatis. Single-shot sequences are performed as breath-holds, while the 3D sequences are respiratory gated. MRCP protocols typically include coronal radial slices and/or a coronal 3D slab. The coronal radial sequence results in multiple images oriented radially around the pancreatic head (Figure 75). Additional radial images may be required that are oriented along the pancreatic body and tail. The 3D sequence results in images that can be put through MIP post-processing to yield multi-dimensional images of the entire biliary tree and the pancreatic ducts (Figure 76).
The kidneys are very sophisticated re-processing machines. Each day, they process about 200 quarts of blood in order to sift out about 2 quarts of waste products and extra water. The wastes and extra water become urine, which flows to the bladder through the ureters. The bladder stores urine until it is released during urination. Wastes in the blood come from the normal breakdown of active tissues, such as muscles, as well as from food. The body uses food for energy and self-repairs. After the body has taken what it needs from food, the wastes are sent to the blood. If the kidneys did not remove the wastes, they would build up in the blood and damage the body. Besides the filtration and excretion of metabolic waste products, such as urea and ammonium, the kidneys also regulate necessary electrolytes, and the fluid and acid-base balance. They measure out chemicals like sodium, potassium, and phosphorus, and release them back into the blood to return to the body. In this way, the kidneys regulate the body’s levels of these substances. The right balance of these substances is necessary for life. The kidneys also reabsorb glucose and amino acids, and release three important hormones - erythropoietin, renin, and calcitriol. Erythropoietin (or EPO) stimulates the bone marrow to make red blood cells. Renin regulates blood pressure, and is part of the renin-angiotensin-aldosterone system, which controls the reabsorption of water and maintains the intravascular volume. Calcitriol is the active form of Vitamin D, which helps maintain calcium for our bones, and a normal chemical balance in the body as a whole.Anatomy
The kidneys are two bean-shaped structures, weighing about five and a quarter ounces (150g) in men and four and three quarter ounces (135g) in women. They are approximately 10-12 cm in length, 5-7 cm in width, and 2-3 cm in thickness. The kidneys are paired retroperitoneal structures, normally located between the transverse processes of the T12-L3 vertebrae. The left kidney is typically somewhat more superior in position than the right, due to the large presence of the liver (Figure 77). The upper poles of the kidneys are normally oriented medially and posteriorly when compared to the lower poles. Superiorly, the suprarenal, or adrenal glands sit adjacent to the upper pole of each kidney. The descending portion of the duodenum abuts the medial aspect of the right kidney. On the left, the greater curvature of the stomach can drape over the superomedial aspect of the kidney. The tail of the pancreas may extend to overlie the left renal hilum. The spleen is anterior to the upper pole of the left kidney, and is connected to it by the splenorenal (or lienorenal) ligaments. Inferiorly, the colon rests anteriorly to the kidneys on both sides. Posteriorly, the diaphragm covers the upper third of each kidney, while the 12th ribs commonly cross the upper poles. The kidneys sit over the psoas muscles medially, and over the quadratus lumborum muscles laterally.
The kidneys are divided into outer and inner portions, known as the renal cortex and renal medulla, respectively. The renal cortex is the outer portion, between the renal capsule and the renal medulla, where ultrafiltration occurs, and where the hormone erythropoietin is produced (Figure 78). In an adult, it forms a continuous smooth outer zone with a number of projections, called cortical columns, which extend down between the medulla’s renal pyramids. The cortex contains the renal corpuscles and the renal tubules, except for portions of the loop of Henle that descend into the medulla. The cortex also contains blood vessels and cortical collecting ducts.
The renal medulla is the innermost part of the kidney, and is split into a number of sections known as renal pyramids. The medulla contains the structures of the nephrons that are responsible for maintaining the salt and water balance of blood, including the loop of Henle. The renal medulla also aids in the reabsorption of water.
The functional units of the kidneys are called nephrons (Figure 80). Each kidney has approximately one million nephrons. The initial blood filtering components of the nephrons are the renal corpuscles, which include the glomerulus and the Bowman’s capsules (Figure 79). Blood from afferent glomerular arterioles (leading blood towards the glomerulus) passes through a juxtamedullary apparatus to the glomerulus. The glomerulus is a network of capillaries that filters blood across the Bowman’s capsules into a proximal convoluted tubule. The glomerulus keeps normal proteins and cells in the bloodstream, allowing extra fluid and wastes to pass through. (Proteins do not normally pass through the glomerular filter because of their large size. Proteins only appear in the filtrate or urine when a disease process has affected the glomerular capsule or the proximal and distal tubules of the nephron.) When the filtrate reaches the proximal convoluted tubule in the renal cortex, glucose and various electrolytes along with water are reabsorbed, while the filtrate passes through. Meanwhile, blood that was filtered at the glomerulus passes into the efferent glomerular arterioles (leading blood away from the glomerulus), and descends into the renal pyramid (Figure 81). From the proximal convoluted tubules, the filtrate passes through the loop of Henle, which has descending and ascending portions, all of which are located in the renal medulla. The filtrate then flows back into the renal cortex and exits the nephron via the distal convoluted tubule. The distal tubules of many individual nephrons converge onto a single collecting duct, and drain their filtrate fluid. Numerous collecting ducts then join to form several hundred papillary ducts. Approximately thirty papillary ducts combine per renal papilla, which are the tips of the renal pyramids that point to the center of the kidney. At each renal papilla, the contents of the papillary ducts drain into the minor calyces, which typically number from nine to twelve. The minor calyces converge into three or four major calyces, which drain to the center of the kidney known as the renal pelvis. The remainder of the collecting system involves the passage of urine from the renal pelvis to the ureter at the ureteropelvic junction, and on to the bladder.
There are many natural variations in the location of the kidneys, as well as in the kidney’s collecting system drainage. The kidneys are typically located retroperitoneally, but may be in an ectopic location, such as the pelvis, when they do not ascend properly. They may also be malrotated or fused, as in horseshoe kidneys, where the inferior poles of the kidneys are fused giving them a U-shaped configuration (Figure 82).
Other fusion anomalies, such as crossed-fused ectopia, result in both kidneys being located on the same side. This anomaly may be associated with some pathological conditions, such as hydronephrosis and ureteropelvic joint obstruction. However, it can also remain completely asymptomatic and undiscovered until it is diagnosed in an imaging study.
Variants in drainage of the collecting system often involve duplication anomalies. More than a single collecting system may form and drain separately into the bladder (complete duplication), or it may join at some point proximally before draining into a single orifice in the urinary bladder (partial duplication).Renal Arteries and Veins
The kidneys receive approximately 20-percent of the cardiac output. Their blood supply comes from the paired renal arteries, located at the level of L2. The renal vessels enter into the renal hilum, the passageway into the kidney, with the renal vein located anteriorly, and the renal artery and renal pelvis located posteriorly (Figure 83).
The first branch from the renal artery is the inferior suprarenal artery (Figure 84). The renal artery then branches off into five segmental branches. The posterior segmental artery supplies most of the posterior kidney, with the exception of the lower pole. The anterior branches include the superior segmental artery, the anterior superior segmental artery, the anterior inferior segmental artery, and the inferior segmental artery. These segmental arteries branch into interlobar arteries, which travel in a parallel fashion in between the major calyces. The interlobar arteries branch further into the arcuate arteries, which run within the renal cortex across the bases of the renal pyramids. The arcuate arteries radiate into the interlobular arteries, which extend into the renal cortex to become afferent arterioles, then peritubular capillaries, and on to efferent arterioles (Figure 85). Some of the terminal branches of the interlobular arteries become perforating radiate arteries that supply the renal capsule. The renal pelvic and superior ureteric branches also originate from the renal artery and supply the upper portion of the collecting system.
The renal vein is generally anterior to the renal artery at the hilum, and drains the kidneys in a similar distribution. The left renal vein is longer, as it must cross the midline, and travel under the origin of the superior mesenteric artery to reach the inferior vena cava (Figure 86). The left gonadal vein drains into the left renal vein inferiorly, while the left suprarenal vein drains into the superior aspect of the left renal vein at approximately the same level. Posteriorly, the left second lumbar vein also drains into the left renal vein. On the right side, the renal vein and the gonadal vein drain separately and directly into the inferior vena cava.
Renal vein thrombosis can result from trauma to the abdomen or back, stricture, or a tumor that obstructs the renal vein. It is also a common complication of other glomerulopathic diseases. Clotting in a renal vein results in renal congestion, engorgement, and the possibility of infarction (Figure 87). Renal vein thrombosis may affect both kidneys, and can occur in an acute or chronic form. Chronic thrombosis usually impairs renal function, causing nephrotic syndrome. Abrupt onset of thrombosis that causes extensive damage may precipitate rapidly fatal renal infarction. Thrombosis that affects only one kidney or has a more gradual progression allows time for the development of collateral circulation, and the possibility of preservation of renal function.
Anatomic variations in renal vasculature occur in 25- to 40-percent of patients. Supernumerary, or accessory renal arteries are the most common arterial variation, with most branches supplying the lower pole of the kidney. These extra arteries may pass anterior to the inferior vena cava and over the ureteropelvic junction, and may be associated with or cause obstruction of the ureteropelvic junction. Persistence of a right subcardinal vein anterior to the ureter can lead to a retrocaval ureter, which can also cause obstruction.Renal Lymphatics
Renal lymphatic drainage parallels the venous drainage system. After leaving the renal hilum, primary lymphatic drainage on the left is into the left lateral aortic lymph nodes, including nodes anterior and posterior to the aorta, between the inferior mesenteric artery and the diaphragm. Right side lymph drainage flows into the right lateral caval lymph nodes.Renal Nerves
The kidneys receive autonomic nerve supply via both the sympathetic and parasympathetic portions of the nervous system (Figure 88). Sympathetic nervous innervation arises from the spinal cord between the levels of T8-L1. These nerves follow a plexus of nerves that run with the arteries. Activation of the sympathetic system causes vasoconstriction of the renal vessels. Parasympathetic innervation arises from the renal branches of the tenth cranial nerve, which is the vagus nerve. When stimulated, parasympathetic nerves cause vasodilation.
Diseases of the kidneys usually attack the nephrons, causing them to lose their filtering capacity. This can occur quickly, due to injury or poisoning, but most diseases destroy the nephrons slowly and silently. Damage may only become apparent after years or decades. Most diseases will attack both kidneys simultaneously. The two most common causes of kidney disease are diabetes and high blood pressure. Those with a family history of kidney problems are also at risk for kidney disease.
Diabetic kidney disease, or diabetic nephropathy, occurs when diabetes prevents the body from using glucose as it normally should. Glucose acts like a poison when it stays in the blood instead of breaking down. Nephrons are damaged by this unused blood glucose. This disease can be delayed or prevented by keeping blood glucose levels down. Certain high blood pressure medications, such as angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) can also slow or delay the progression of diabetic kidney disease.
Hypertensive nephropathy occurs when high blood pressure damages the small blood vessels in the kidneys. Damaged vessels cannot filter the wastes from blood as they should. Persons with diabetes or reduced kidney function should strive to maintain their blood pressure at 130/80 or lower. ACE inhibitors and ARBs (mentioned previously as blood pressure medications) protect the kidneys more than other hypertension medications.
Glomerular disease can result from other diseases, such as autoimmune diseases, infection-related diseases, and sclerotic diseases. These diseases can attack the tiny blood vessels, or glomeruli, in the kidneys. Early signs of glomerular disease are proteinuria and hematuria, too much protein in the urine, and blood in the urine, respectively. Glomerular disease can slowly destroy kidney function. Treatments include immunosuppressive drugs or steroids to reduce inflammation and proteinuria, depending on the specific disease. Blood pressure control is important with glomerular disease, as well as all kidney diseases.
Inherited and congenital kidney diseases can be diagnosed during pregnancy, or perhaps not until adulthood. Polycystic kidney disease is a genetic disorder in which cysts slowly replace much of the mass of the kidney, reducing kidney function, and eventually leading to kidney failure (Figure 89). Kidney disease in children may manifest as high blood pressure, a low number of red blood cells (anemia), proteinuria, or hematuria. With the advances in diagnostic imaging technology, diseases can be diagnosed before symptoms even appear in children. Many people are born with only one kidney, but they are able to lead normal healthy lives. Each year, thousands of people make the charitable choice to donate a kidney for transplantation purposes.
Kidney disease can be brought on by trauma, or poisons. Some over the counter pain medications can be poisonous to the kidneys if taken regularly over long periods of time. Acute renal failure typically occurs from an injury to the kidneys, a large amount of blood loss, dehydration, a blockage in the urinary tract, drugs, or poisons. Acute renal failure is usually reversible. Chronic kidney disease, or chronic renal insufficiency, may occur slowly and silently, with a gradual loss of kidney function. Patients may develop permanent kidney failure, and are at a high risk of death from a stroke or heart attack. Signs of chronic kidney disease include urinating more or less often than normal, feeling tired, loss of appetite, nausea and vomiting, swelling in hands or feet, feeling itchy or numb, darkened skin, muscle cramps, and drowsiness or difficulty concentrating. Those with end stage renal disease are at total or nearly total and permanent kidney failure. The body fills with extra water and waste products, a condition called uremia. The patient’s hands and feet may swell. They may feel tired and weak, because the body needs clean blood to function properly. Untreated uremia can lead to seizures or coma, and ultimately results in death. These patients need dialysis or transplantation to stay alive. In cases of transplantation, a living related donor is considered the best match. If the diseased kidneys are not causing high blood pressure or infection, they may be left in place.
Renal cell carcinomas account for 3-percent of all malignancies in adults. Almost 50-percent of these cancers are detected incidentally, and as many as 85-percent of suspicious renal lesions turn out to be malignant. Features of renal lesions that may indicate potential malignancy include size, the presence of calcifications, the distribution of calcifications within the lesion, wall thickness, the presence of septa in cases of cystic lesions, inhomogeneity of the lesion, and enhancement after contrast. In one study of 186 renal tumors, all tumors over 7cm were malignant, while 80-percent of those under 3 cm were malignant as well. Solid renal lesions that are less than 7 cm in size can have many differential diagnoses. Cysts and angiomyolipomas can typically be positively identified as benign with high confidence. MRI is often performed after a renal lesion is identified on ultrasound or CT for further characterization. The main goal in the evaluation of renal lesions is to differentiate and separate surgical lesions (renal cell carcinomas) from those that are non-surgical (cysts, angiomyolipomas). Most simple cysts are easily recognizable, but complicated or multiloculated cysts need more attention in order to differentiate them from cystic carcinomas.
CT is usually used for renal cell carcinoma staging purposes, which are based on the degree of tumor spread beyond the kidney. The extent of tumor spread into the renal vein and IVC are important pieces of information for surgeons. IVC involvement is seen in 4- to 10-percent of renal cell carcinomas.
MRI may play a role in the decision to perform a partial nephrectomy vs. a total nephrectomy, thereby preserving some kidney function. The ideal tumor for a partial nephrectomy is smaller than 3 cm, confined to the parenchyma of the kidney, and has a peripheral location. Another favorable sign for a partial nephrectomy is the presence of a pseudocapsule around the renal tumor, as this is a sign of a lack of perinephric fat invasion. These pseudocapsules are usually well seen on MRI, appearing as a hypointense rim around the tumor on both T1- and T2-weighted sequences. They are usually best seen on the T2 sequences, and often seen on contrast enhanced gradient echo images also.Laboratory Tests for Renal Disease
There are a variety of blood and urine tests that can be performed to detect kidney disease. Healthy kidneys take wastes out of the blood, but leave protein. Impaired kidneys may fail to separate a blood protein, called albumin, from the wastes. Small amounts of albumin will then leak into the urine, causing a condition called microalbuminuria, which is a sign of deteriorating kidney function. As kidney function worsens, the amount of albumin and other proteins in the urine increases, leading to a condition called proteinuria. Laboratories can then measure and calculate a protein-to-creatinine or albumin-to-creatinine ratio. Creatinine is a waste product in the blood created by the normal breakdown of muscle cells during activity. Healthy kidneys can take creatinine out of the blood, and put it into the urine to leave the body. Kidneys that are not working well allow creatinine to build up in the blood.
MRI technologists that are performing contrast-enhanced exams should be familiar with the term “GFR”, but may not fully understand what the GFR measurement tells us about kidney function.
The Glomerular Filtration Rate (GFR) is a calculation of how efficiently the patient’s kidneys are filtering wastes from the body, and corresponds to the percentage of kidney function that is available. Current GFR measurements may actually be eGFR, or estimated GFR values, based on the patient’s creatinine reading, age, and values assigned for sex and race. The lab test for creatinine measures how many milligrams of creatinine in one deciliter of blood (mg/dL). Each lab has its own normal range, usually between 0.6 to 1.2 mg/dL. A slight elevation in creatinine is a sign that the kidneys are not working at full strength. One formula equates a creatinine level of 1.7 mg/dL for men and 1.4 mg/dL for women to 50-percent of normal kidney function. Creatinine levels are quite variable, and can be affected by diet, so the eGFR value is accepted as a more accurate way to determine reduced kidney function. The National Kidney Foundation has determined different stages of chronic kidney disease, based on the eGFR value. An eGFR of 90 or above is considered normal. An eGFR below 60 for three months or more puts a patient in the chronic kidney disease category. As kidney function declines, the risk of complications rises. An eGFR between 30 and 59 is considered to be a moderate decrease in kidney function. Hormones and minerals can be thrown out of balance, leading to anemia and weak bones. This level can be treated with medications, and better food choices. An eGFR between 15 and 29 means there is a severe reduction in kidney function. Patients at this level will typically need treatment for the complications of chronic kidney disease, and would require dialysis. Kidney failure occurs at the lowest eGFR values, typically less than 15. The kidneys do not work well enough to sustain life, and these patients need renal dialysis or a kidney transplant.
A blood test for Blood Urea Nitrogen (BUN) is another way to determine kidney function. The blood carries proteins to cells throughout the body. After the cells use the proteins, the remaining waste product is returned to the blood as urea, which is a compound that contains nitrogen. Healthy kidneys take urea out of the blood and put it in the urine. Kidneys that are not working well leave the urea in the blood. A deciliter of normal blood contains 7-20mg of urea. A BUN reading of more than 20mg/dL may mean that the kidneys are not working at full strength. An elevated BUN can also result from dehydration and heart failure.MRI and Renal Disease
Although imaging requirements will differ between institutions and radiologists, recommendations for renal MRI protocols should minimally consist of T1 and T2-weighted pre-contrast images, including T1-weighted images with fat saturation, followed by rapidly acquired breath-hold T1-weighted dynamic post-gadolinium images. A fast imaging technique is required due to the respiratory motion of the kidneys. Scans performed in a single breath-hold on expiration are ideal, as the position of the kidneys remains more constant on expiration as compared to inspiration. Respiratory triggering is recommended if the sequences are too long.
Axial T2-weighted FSE sequences with fat suppression are useful for characterizing cysts and intraparenchymal abscesses, for evaluating hydronephrosis, as well as for the detection of solid lesions. Axial T1-weighted gradient echo in and out of phase sequences are useful for lesion characterization. Many solid renal lesions are hypointense compared to renal parenchyma on T1-weighted sequences, but lesions that contain melanin, hemorrhage, macroscopic fat, or cysts with high protein content may show hyperintense signal. Out of phase T1 GE sequences can be used to prove the presence of small amounts of fat. Coronal 3D fast gradient echo sequences with fat suppression are often performed as the delayed contrast-enhanced sequences to visualize renal venous anatomy, to analyze tumor thrombus, and to evaluate the extent of tumor in perinephric fat.
Four distinct phases of contrast enhancement can be identified in dynamic renal imaging after IV gadolinium injection, including a capillary (cortical) phase, early tubular phase, ductal phase, and excretory phase. These phases may be captured with imaging immediately after the contrast injection, followed by repeat images at 1-2 minutes, 3 minutes, and 5-10 minutes post injection. Immediate post-injection images capture the arterial phase of renal enhancement, as well as the capillary phase, which demonstrates corticomedullary differentiation. The 1-2 minute and the 3 minute images capture the nephrogram phase, where renal parenchyma is maximally enhanced, and the 5-10 minute images capture the excretory phase. Each of these phases may be acquired in a single breath-hold interval, which minimizes image degradation from respiratory motion.
The T1-shortening effects of gadolinium produce a visibly increased T1-signal intensity. In the arterial phase, increased renal cortical signal intensity may be observed as early as 10-20 seconds after contrast injection, with maximum intensity reached between 20-50 seconds post injection. This can be explained by the heavy arterial perfusion of the renal cortex relative to the remaining renal parenchyma. The renal cortex signal intensity decreases slowly and constantly. Gadolinium is filtered completely in the glomerulus, so it is diluted in the extravascular spaces.
The renal medulla shows signal intensity changes that are quite different from those in the cortex, with increased signal intensity observed 10-20 seconds later in the medulla as compared to the cortex. Maximum medullary signal intensity reaches values that are quantitatively equal to those in the cortex. Thirty to forty seconds after the first visible signs of perfusion, the medullary signal intensity sharply decreases to pre-contrast levels. When gadolinium is dilute, T1 shortening occurs, creating high-signal intensity urine. When gadolinium is at a high concentration (secondary to water resorption in the proximal convoluted tubules, loop of Henle, and collecting tubules), gadolinium induces signal loss, leading to low-signal intensity urine.
The main MRI feature that indicates the potential malignancy of a renal tumor is enhancement after IV gadolinium, which differentiates a lesion from a cyst. MRI has been able to detect lesions as small as 1 cm, and offers reliable differentiation between solid and cystic masses when breath-hold and fat-saturation techniques are combined with IV contrast. Enhancement may be difficult to measure on MR, and can be done subjectively, by subtraction, or by quantitative assessment. The quantitative assessment method involves calculating the relative enhancement, which is the signal intensity increase after contrast administration compared to the signal intensity before contrast. The relative enhancement peak is at its maximum between 2-4 minutes after contrast injection. Malignant lesions have a threshold of 15-percent relative signal intensity enhancement after IV gadolinium administration. This method has a very high sensitivity rate when used to detect renal cell carcinoma in renal lesions. Classic renal cell carcinoma appears as an irregular mass with ill-defined margination from normal renal parenchyma (Figure 90). Immediate arterial phase post-gadolinium images demonstrate heterogeneous enhancement that reflects the hypervascularity of the tumor. Delayed images show diminished enhancement of the lesion, as contrast washout from the hypervascular tumor is displayed against a background of contrast-retaining renal tubules. Routine imaging in multiple post-contrast phases is necessary, as small, homogeneously enhancing tumors may be difficult to distinguish from normal cortex on immediate post-contrast images. In addition, renal cell carcinomas can be hypovascular, and may display with less enhancement than the surrounding renal parenchyma. The various subtypes of renal cell carcinomas may be visualized differently in MRI. Clear cell carcinomas represent 88-percent of all renal cell carcinomas. They may show a loss of signal intensity on out of phase images versus in phase images, due to intracellular lipids. Papillary carcinoma and chromophobe carcinoma are less commonly seen (10-percent and 2-percent of cases, respectively.) Fat-containing renal cell carcinoma is rare, but easy to confuse with a benign angiomyolipoma. Most of the reported cases of renal cell carcinoma that contained macroscopic fat also contained intratumoral calcifications, which are rare in angiomyolipomas. Therefore, the presence of calcifications in a lesion with macroscopic fat may serve as a warning of renal cell carcinoma. Ultrasounds and CT should also be reviewed in these cases, as MRI has a low accuracy for detecting calcifications. MRI with contrast is also beneficial when trying to differentiate between a bland thrombus and a tumor thrombus, as enhancement of the thrombus indicates a tumor, while lack of enhancement indicates a clot.
In Figures 91 through 93, the images are identified as displaying clear cell renal cell carcinoma (Figure 91), papillary renal cell carcinoma (Figure 92), and chromophobe renal cell carcinoma (Figure 93). These images were obtained from a study in which the percentage of signal intensity change found in different phases of dynamic renal sequences was being examined for its usefulness in the indication of tumor subtypes. It was determined that a hypervascular pattern of enhancement in the corticomedullary phase indicated clear cell renal cell carcinoma, and a hypovascular pattern of enhancement in this same phase was indicative of papillary renal cell carcinoma.
Renal cortical cysts are the most common renal masses in the adult patient. MR characteristics of simple cysts include sharp margination from renal parenchyma, absent signal on T1-weighted images, homogeneously high signal on T2-weighted images, and no enhancement in any phase after gadolinium injection (Figure 94). Less than 5-percent of renal cell carcinomas are cystic, but the use of MRI in their evaluation may show more septa, increased wall thickness, or increased enhancement when compared to CT (Figure 95).
Angiomyolipomas are the most common benign tumors of the kidney, composed of blood vessels, smooth muscle, and fat. The ability of MR to positively differentiate angiomyolipomas is especially important for patients with tuberous sclerosis, as this benign tumor develops in about 80-percent of these patients, and they are already at an increased risk of developing renal cell carcinoma. Imaging characteristics of angiomyolipomas depend on the proportion of each component of the lesion. A typical angiomyolipoma has high fat content relative to vascular content, creating a lesion with high non-contrast signal on T1-weighted images (that diminishes with fat saturation), and that is hypovascular on post-contrast imaging (Figures 96, 97). Angiomyolipomas with the opposite characteristics (low fat and high vascular content) can be difficult to diagnose, as they appear hypervascular, and have an enhancement pattern similar to that of renal cell carcinoma. This diagnosis can be further confused in the rare cases of renal cell carcinomas that contain fat, as mentioned above.
The MR signal intensity of the cysts involved in polycystic kidney disease, or the acquired cystic disease of dialysis, varies depending on the presence of blood products of different ages, as well as on the presence or absence of infection. Patients suffering from these diseases have an increased risk of renal cell carcinoma. Even with today’s advanced techniques, CT, MRI and ultrasound can all have difficulties when it comes to differentiating between infected, hemorrhagic and malignant cysts, due in part to the small size and high number of these cysts.
Dynamic MR of the kidney can be used as a functional analysis tool in cases of both acute and chronic ischemia. Imaging of this type can provide relative quantification of renal blood flow by comparing signal intensity increases between the kidneys following contrast injection. Both acute and chronic ischemia can result in diminished contrast enhancement of the affected kidney relative to the normal one, though appropriate corticomedullary differentiation persists in both. The acutely ischemic kidney will appear enlarged, while the chronically ischemic kidney will appear small and shrunken.
MRI is a valuable tool in the analysis of filling defects of the renal collecting system. Calculi are the most common filling defects and, regardless of their calcium content, appear as signal voids on all MRI sequences. Dilute gadolinium in the renal collecting system creates high-signal intensity urine on T1-weighted images collected between 2 and 30 minutes after contrast injection, which allows for excellent contrast between the signal-void stones and high-signal urine. Neoplasia and blood clots are similarly well demonstrated with dilute gadolinium. Dynamic imaging is superior for depicting complications of renal calculi, namely renal obstruction. During acute obstruction, the affected kidney is enlarged, with resultant retained parenchymal contrast (Figure 98). Subsequent imaging typically shows a prolonged nephrogram phase, and diminished corticomedullary differentiation. In cases of chronic obstruction, the affected kidney is small, with decreased perfusion, which leads to decreased cortical enhancement. The resultant increased transit time of the contrast agent leads to the delayed nephrogram. Renal cortical scarring and thinning that result from chronic reflux are best appreciated on immediate postcontrast images, when corticomedullary differentiation is maximal.
MRI is an important pre-operative test for patients that are potential living kidney donors. Surgeons must be informed of the arterial and venous vasculature of the donor kidney, the presence of accessory vessels, abnormal vessel locations (i.e., extrahilar branching, retrocaval position of vessels), abnormal collecting systems, or the presence of cysts or tumors. MRI imaging should focus on arterial and venous imaging, and standard parenchymal imaging (Figure 99). Gadolinium is excreted by the kidneys, so the concentration of contrast in the renal veins is lower than that found in the renal arteries. A MIP post-processing procedure can be performed to better delineate the renal veins.
Research is ongoing concerning the uses of MRI in relation to kidney function. Arterial spin labeling is being considered for use in the assessment of renal perfusion. MR techniques may be used eventually to measure glomerular filtration rates. DWI is being considered for the characterization of different abnormalities. Studies are being performed using BOLD (Blood Oxygenation Level-Dependent) MRI on 3T systems for the assessment of renal lesions without the use of IV contrast. The BOLD method uses the paramagnetic properties of deoxyhemoglobin as an endogenous contrast agent. Deoxyhemoglobin is found in the draining veins after oxygen has been unloaded. Lesions with higher concentrations of deoxyhemoglobin have shown greater rates of signal loss. The amount of deoxyhemoglobin among renal lesions may be different because the vascularity or perfusion of the renal lesions may differ. Further research may lead to improved characterization of renal lesions using the BOLD method, especially in cases where IV contrast is not advised.
The following are HMSA suggestions for kidney imaging. Always check with your radiologist for his/her imaging preferences. Kidney imaging typically involves the use of breath holds or respiratory gating to eliminate artifacts caused by abdominal motion during breathing. Scans performed in a single breath-hold on expiration are ideal, as the position of the kidneys remains more constant on expiration as compared to inspiration. Respiratory triggering is recommended if the sequences are too long. Dynamic contrast-enhanced studies of the kidneys should be considered a routine part of a comprehensive MR exam of the kidneys, in combination with unenhanced T1- and T2-weighted images, and fat suppression techniques.
Axial slices of the kidneys can be set up using coronal images or scanograms. Slices should be prescribed in a sufficient number to cover the area from the inferior margin of the kidneys to the superior aspect of the adrenals, as seen in Figure 100.
Coronal slices of the kidneys can be set up using axial images or scanograms. Slices should be prescribed in a sufficient number to cover the area from the posterior abdominal wall past the region of the IVC and aorta, as seen in Figure 101.
Sagittal scans are not typically used in imaging of the kidneys.
Saturation pulses or “presats” may be placed superiorly and inferiorly to eliminate flow motion from the aorta and inferior vena cava. In addition, anterior presats may be used to eliminate gastric and bowel motion.
VASC-ASL stands for Veins and Arteries Sans Contrast - Arterial Spin Labeling. This is a non-contrast MRA technique that uses a Selective IR pulse with a respiratory-gated RF Fatsat 3D BASG sequence. Arterial Spin Labeling allows you to “tag” inflowing blood, while nulling the background tissue signal, so the resultant signal is even brighter than that achieved with VASC alone. This method enhances vessel patency to aid in the diagnosis of stenosis or occlusion.
VASC-ASL can be performed with or without subtraction, with the non-subtraction technique offering a time-savings. The Selective IR pulse that is used in ASL is positioned to affect only a “select” area of anatomy, similar to a preset. Using the non-subtraction method, the Selective IR pulse (gold band) is positioned to cover the slice slab and the vessels that are distal or “downstream” from the target vessels, as seen in Figure 102.
VASC-ASL can also be performed using a subtraction technique, in which images are acquired with the Selective IR pulse turned off and turned on. When the Selective IR (SIR) pulse is turned on, it is meant to suppress the target vessels. After subtraction is performed, the resulting image shows only the target vessels, as seen in Figure 103.
The subtracted VASC-ASL images can then be loaded into a MIP task, and post-processed as a Sliding MIP, as seen in Figure 104.
When discussing MRI of the abdomen, our focus is typically on the upper abdominal organs, including the liver, pancreas, and kidneys. However, we must also include discussion of the small bowel, as it is positioned amongst the abdominal organs, and can be imaged in MR through the use of enterography and enteroclysis. Cross-sectional imaging techniques have found an increasing role in the evaluation of suspected small bowel disorders. MRI is able to provide exquisite anatomic, functional, and real-time information without the need for ionizing radiation. This is especially important for patient’s suffering from Crohn’s disease, as they will be subjected to multiple follow-up imaging exams. MRI is also advantageous due to its ability to provide dynamic information regarding bowel distention and motility, its superior soft-tissue contrast, and its relatively safe intravenous contrast agent profile.
There are two methods commonly used for MR examinations of the small bowel. MR enterography involves the oral administration of large volumes of fluid (1350-2000ml) that contains enteric contrast agents. Enteroclysis involves the administration of the enteric contrast through nasojejunal intubation. Although it is reported that enteroclysis offers superior bowel distention, it may not be offered at all facilities, and may not be suitable for all patients.
There are many enteric contrast agents available, with their classifications based on the signal intensity they produce on T1 and T2-weighted images. Positive contrast agents (gadolinium chelates, manganese ions, some food substances such as milk and vegetable oil) produce high signal intensities, and are beneficial for demonstrating wall thickening on T1-weighted images. They are also useful for assessing the progress of contrast material through the bowel. Negative contrast agents produce low signal intensities, reducing the signal intensity of the bowel lumen, which improves the contrast of high signal intensity inflammation in the bowel wall on T2-weighted images. Negative contrast agents also improve the conspicuity of interloop abscesses, which otherwise may be mistaken for fluid-filled bowel. One drawback to negative contrast agents is that there are few commercially available in the US. The biphasic contrast category includes the largest number of contrast agents. Low signal intensity on T1-weighted images improves the contrast between the bowel lumen and hyperenhancing wall inflammation or masses following the administration of intravenous contrast. The characteristics of enteric contrast agents can be quite variable, depending on the concentration and type of solution they are mixed with.
MR enterography findings that are suggestive of active inflammation include bowel wall thickening and hyperenhancement, ulcerations, increased mesenteric vascularity, and perienteric inflammation (Figure 105). Complications of Crohn’s disease are well depicted, and may include penetrating disease and small-bowel obstruction (Figure 106). Pathologic findings may be demonstrated without the administration of IV contrast; however, IV contrast may be helpful in detecting areas of hyperenhancement that are suggestive of active Crohn’s inflammation, as well as in the detection of hyperenhancing masses. The use of spasmolytics (i.e., glucagon) may be helpful to reduce bowel peristalsis and motion artifacts. It is especially important to keep artifacts at a minimum for the fast gradient echo sequences that should be performed after IV contrast is given.
Patients may be positioned supine or prone for enterography exams. Reports from literature suggest that the prone position may facilitate the elevation and separation of small-bowel loops from the pelvis, and reduce the area to be imaged. Imaging should be performed in both the axial and coronal planes, as some strictures may be more apparent on one projection than another. It is recommended that a single FOV should encompass the entire small bowel on coronal acquisitions (Figure 107).
Enterography protocols will most likely be site and radiologist specific. One of the benefits of performing these examinations using MRI is that sequences can be repeated to capture multiple discrete vascular phases, to reassess abnormal bowel segments, or to improve image quality without the consequences of ionizing radiation. A variety of highly-recommended sequences are mentioned in the literature. T2-weighted SS FSE sequences are considered critical for characterizing the cause of bowel wall thickening, as areas of high signal intensity are seen in the presence of active inflammation, while lower signal intensity is seen with fibrostenotic disease. Areas of edema and luminal dilatation should also be visible. Fat suppression added to these sequences will improve the conspicuity of high signal intensity bowel wall inflammation as well as perienteric fat (Figure 108). Steady state free precession sequences (BASG) performed in the coronal plane in multiple phases with free breathing can be used to depict bowel motility, which is advantageous when trying to distinguish between fixed and transient segments of narrowing. The images can be displayed as a cine loop to assess bowel motility, to exclude or confirm fixed stenosis and segmental dilatation, and to detect adhesions. The high image contrast is helpful for assessing mesenteric vascularity and lymphadenopathy. Coronal fat-suppressed 3D T1-weighted gradient echo sequences with breath-holds are recommended before and after IV contrast, with the arterial phase typically evident 25 seconds after the contrast administration. Additional 3D coronal acquisitions offer the opportunity to have several complete volumetric data sets within two minutes after the contrast injection that may include periods of peak bowel wall enhancement, and at least one set that may be motion free. Post-contrast T2-weighted axials with fat suppression can be used to assess the bowel wall and surrounding tissues for fluid and edema. Delayed contrast-enhanced axials are advantageous for multiplanar correlation.
One of the difficulties mentioned in the literature concerning MR enterography is that protocols and reporting methods have not as yet been standardized. This step will be necessary for more systematic assessments of various small bowel diseases, treatment regimens, and outcome measurements for clinical trials.
This concludes the Abdominal Imaging module of the Hitachi Medical Systems America’s MRI Anatomy and Positioning Series. You must complete the post-test for this activity in order to receive your continuing education credit.