Liver Ultrasound PDF
Document Details
Uploaded by Deleted User
Martha Moon Larson, John S. Mattoon, Yuri Lawrence, Rance K. Sellon
Tags
Summary
This document discusses ultrasonography of the liver, including indications, complications, and technique in veterinary medicine. It covers the importance of systematic evaluation of the liver parenchyma, as well as portal and hepatic veins. The document also outlines considerations for sampling the liver and post-biopsy monitoring.
Full Transcript
9 Liver Martha Moon Larson, John S. Mattoon, Yuri Lawrence, Rance K. Sellon Ultrasonography is an essential imaging too...
9 Liver Martha Moon Larson, John S. Mattoon, Yuri Lawrence, Rance K. Sellon Ultrasonography is an essential imaging tool for evaluating the liver. A consistently grow bacterial organisms than liver samples.9 Such needs complete examination requires a systematic evaluation of the hepatic could, in some cases, limit the value of ultrasound-guided biopsies, parenchyma, portal and hepatic veins, and the gallbladder and biliary and interpretation of such biopsy results should be carefully consid- system. The most common indications for hepatic ultrasonography ered in light of other diagnostic (laboratory, imaging) results.1,2,6 are hepatomegaly or masses in the liver region, suspected primary or Laparoscopic biopsies, when possible, offers a balance of adequate metastatic hepatic neoplasia, ascites, and unexplained increases in sample size, visualization of the liver and gallbladder, and noninvasive- hepatic enzymes. There are many other reasons that could justify ness (compared with laparotomy). imaging of the liver and biliary system; a more comprehensive list of Hemorrhage is the most concerning complication of hepatic bi- indications for ultrasound of the hepatobiliary system is provided in opsy, and assessment of coagulation parameters (platelet numbers, Box 9.1. Efforts should be made, when possible, to exclude diseases prothrombin time [PT] and partial thromboplastin time [PTT]) are such as protein-losing enteropathy and protein-losing nephropathy, advised before liver biopsy.7 Nevertheless, there is not yet information pancreatitis, hyperadrenocorticism and “atypical” hypoadrenocorti- that would indicate specific coagulation abnormalities that would pose cism, and others that may have features suggestive of hepatobiliary an increased risk of post-biopsy hemorrhage. The exception would disease before focusing on the liver, particularly if moving toward liver seem to be thrombocytopenia; available studies do hint at an increased sampling (and especially biopsy). risk of bleeding post-biopsy in the face of thrombocytopenia (,80,000/ Ultrasound is used to direct percutaneous liver sampling (aspirates uL), especially in dogs.10 A recent paper investigating the risk of bleed- or biopsies) or follow treatment response and progression of disease ing after liver biopsy in cats found no correlation between platelet for inflammatory or neoplastic conditions. There are important con- numbers, PT, PTT, and risk of post-biopsy bleeding.11 Extrapolating siderations to take into account when deciding how to sample the liver. from recommendations in people and existing veterinary data, a panel One study of 70 dogs with necropsy-obtained liver samples found that of veterinarians proposed that dogs were at high risk of bleeding if the sampling at least two liver lobes identified the underlying liver disease packed cell volume (PCV) was less than 30%, the platelet count less in 98% of the cases1; nevertheless, discordant interpretations between than 50,000/uL, the PT and PTT more than 1.5 the upper limit of normal, the two lobes is possible, making correlation of histopathology results the fibrinogen less than 100 mg/dL, the von Willebrand factor activity with other clinical data critical to biopsy interpretation. The number (when considering biopsies in predisposed breeds) less than 50% of of liver lobes that would have to be sampled using ultrasound-guided the standard, and the buccal mucosal bleeding time greater than approaches has not been established but would likely be at least two. In 5 minutes.7 In some patients, coagulation abnormalities can be cor- another study, 53% and 50% of paired-needle (18-gauge) and wedge- rected or improved with vitamin K supplementation, platelet transfu- liver biopsies in dogs and cats, respectively, were discordant.2 Cytologic sion, desmopressin acetate, or administration of fresh frozen plasma. and histologic diagnosis does not always agree.3 Importantly, there can After biopsy, patients should be closely monitored; in the feline study, be microscopic evidence of disease in a liver that appears ultrasono- some cats did not reach the lowest PCV until 24 hours after biopsy.11 graphically normal.4 Therefore, ultrasound is not a reliable means of The diagnostic value and method of acquisition of liver biopsies establishing a cytologic or histologic diagnosis of liver disease.1-5 should be considered carefully if needed in patients with coagulation In addition to having samples from different lobes, having a suffi- abnormalities. cient number of portal triads is also important for accurate assessment of liver biopsies. In one study, samples that had 3 to 12 portal triads were considered of adequate diagnostic utility, and all biopsy tech- TECHNIQUE niques (needle biopsy, laparoscopic or laparotomy liver biopsies) were A complete abdominal examination is recommended for each animal, able to provide a sufficient number of portal triads.6 Based on human although only hepatic imaging is discussed in this chapter. Hair should criteria for liver biopsies, recent recommendations are that optimal be clipped from the 10th intercostal space caudally over the ventral samples for assessment of hepatitis should include at least 12 to 15 por- abdomen and halfway down the lateral abdominal walls. Acoustic tal triads.7 In addition to routine histopathology, sufficient samples coupling gel is applied liberally to the skin and worked into the stubble should be obtained to permit copper stains of liver sections and copper of the hair coat. Alcohol can also be used to clean and wet the skin quantitation; inadequate sample sizes run the risk of erroneous mea- surface. surement of copper concentration.8 Some clinicians advocate cultur- Gas in the stomach is a barrier to successful ultrasound examination ing liver samples, although one study found culture of bile to more of the liver. Therefore feeding or procedures that result in struggling or 355 356 Small Animal Diagnostic Ultrasound BOX 9.1 Some Specific Indications for Ultrasound of the Hepatobiliary System Hepatic encephalopathy Acute liver enzyme elevation Hepatomegaly Hypercholesterolemia Fever of unknown origin Abnormal bile acids Abdominal pain Hyperbilirubinuria (especially in cats) Icterus Hypoalbuminemia Hematemesis Hypocholesterolemia Melena Hypoglycemia Hepatocutaneous syndrome Decreased blood urea nitrogen Immune mediated disease Decrease in liver size Cranial organomegaly Hypercalcemia Abdominal effusion Sternal lymphadenomegaly Chronic, especially unexplained, liver enzyme elevations of any magnitude Metastatic pulmonary disease aerophagia should be avoided. The animal should be restrained on a Evaluation of the liver is begun by placing the transducer in the table in dorsal or lateral recumbency (Fig. 9.1; see also Chapter 4, subxiphoid position and angling the sound beam craniodorsally to Fig. 4.8, A). Positioning the patient to redistribute gastric fluid and gas cranioventrally (cranial to stomach) in a midsagittal plane (see or using the intercostal windows can aid in liver evaluation when gas is Fig. 9.1, A, B; see also Fig. 4.8, Aa). Scans of the liver are then made by present. Tranquilization is used as needed. sweeping the beam from left to right through the entire liver (see A 5.0- to 8.0-MHz microconvex transducer works well for ultra- Fig. 4.8, Ab, Ac). The transducer should remain in the subxiphoid sound examination of the liver in most medium to large dogs. Higher position as much as possible during the entire sweep. The beam is frequencies (8 to 12 MHz) can be used in small dogs and cats. In large angled dorsally or ventrally in successive sweeps to make sure the or giant breed dogs, a 3.0-MHz transducer may be necessary to pene- entire liver is imaged. The beam is then oriented in a transverse plane, trate to the dorsal aspects of the liver. scanning the liver from ventral to dorsal. Successive sweeps are made A B Fig. 9.1 Positioning for ultrasound examination of the liver. A, Dorsal recumbency with subxiphoid positioning for liver examination. B, Left lateral recumbency with the transducer placed subcostal on midline. C, Left lateral recumbency with the transducer placed for a transverse view at the 10th to 12th right intercostal space. C CHAPTER 9 Liver 357 by angling the transducer to the left and right of midline to make sure The ultrasonographic identification of liver lobes and of the hepatic the entire liver is imaged. The caudate liver lobe is evaluated along the and portal venous vascular anatomy has been described in detail and caudal, dorsal aspect of the right costal arch. If the liver is enlarged, compared with angiographic and postmortem examination findings extended evaluation of the right and left craniolateral abdomen should in normal dogs.13-17 Knowledge of the venous anatomy helps identify be included. The liver can also be imaged through an intercostal specific regions of the liver on ultrasound scans (see Fig. 9.2). window, scanning through the 10th to 12th right and left intercostal Portal veins are identified by their bright, echogenic walls. Hepatic spaces (see Fig. 9.1, C). This is especially useful in patients who have veins lack these prominent walls and appear as anechoic tubular struc- a small liver, excessive stomach gas, or a deep-chested conformation. tures. The distinct composition of hepatic vein walls results in more of The right intercostal view is frequently required to optimally view the a specular reflector, becoming hyperechoic only when the sound beam caudal vena cava, portal vein, common bile duct, and lymph nodes in is perpendicular to the vessel wall.18 Because of its composition, the the region of the porta hepatis.12 A sector transducer is necessary to portal vein wall appears hyperechoic from many angles. Hepatic veins fit in the small contact area between ribs and is placed transversely can be traced to their termination in the caudal vena cava at the level (parallel to the ribs) at about the 11th intercostal space, 5 to 10 cm of the diaphragm (Fig. 9.3). The hepatic veins to the left medial and ventral to the spine. The porta hepatis is usually located at this level, lateral lobes, quadrate lobe, and right medial lobe are best located in a but one may have to “walk” the transducer to a cranial or caudal transverse plane from a subxiphoid location. These veins converge to intercostal space, and slide dorsally to ventrally, to find the best enter the left ventral caudal vena cava at a single location near the dia- window. If lung gas is encountered, the transducer should be moved phragm. The quadrate hepatic vein, located to the left of the gallbladder, one or two intercostal spaces caudally. If the right kidney is seen, the joins the right medial hepatic vein, located to the right of the gallbladder. transducer should be moved cranially. The porta hepatis is located The left lobe hepatic vein forms from the confluence of the left medial when the liver and the cross-sectional views of the aorta, caudal vena and left lateral portions farther to the left. More caudally, the right cava, and portal vein are seen in a single window. lateral hepatic (right hepatic) vein and right caudate hepatic (middle After evaluation of the dorsal intercostal window, the transducer is hepatic) vein enter the caudal vena cava separately. These veins are moved cranially and ventrally, toward the sternum. The gallbladder more difficult to see, especially from the ventral abdomen. The caudate can be visualized from this window, along with the cystic duct and left vein enters slightly more caudally than the right lateral hepatic vein, hepatic and portal vein branches. and both are best traced from the caudal vena cava by a right lateral approach through the 9th to 11th intercostal spaces in a longitudinal ANATOMY plane cranial to the right kidney. Portal vein distribution differs slightly from that of the hepatic Landmarks veins (see Figs. 9.2 and 9.3). The left portal vein, best viewed in a trans- The liver is bounded cranially by the diaphragm, ventrally by falciform verse plane from a subxiphoid or a ventral right intercostal location, is fat, and caudally by the right kidney on the right, the stomach cen- seen caudal and dorsal to the neck of the gallbladder between the gall- trally, and the spleen on the left. The curvilinear echogenic line seen bladder and hepatic veins. The left portal vein branches into the left cranial to liver on sonograms is commonly referred to as the dia- lateral and left medial portions slightly caudoventral to the gallbladder. phragm but is really the interface between the diaphragm and lung. The left lateral portal vein can be traced farther peripherally into the This echogenic line is referred to as the diaphragm throughout this liver than the left medial portal vein. The portal quadrate vein arises chapter for convenience. The actual diaphragm is only seen when there from the portal vein itself or branches from the left medial portal vein. is adjacent peritoneal or pleural fluid, or both. Falciform fat lies ventral This is in contrast to the hepatic venous system, in which the quadrate to the liver, and in the dog it has a variable echogenicity that may be hepatic vein usually joins the right medial hepatic vein. The right me- isoechoic, hypoechoic, or hyperechoic to liver parenchyma. In the cat, dial portal vein arises from the portal vein to enter the right medial falciform fat is typically isoechoic to hyperechoic to liver parenchyma. lobe to the right of the gallbladder. Portal vein branches to the right lateral and caudate lobes are more difficult to identify and cannot be Lobation and Vascular Anatomy seen in the same plane as the left portal vein. These branches can be The liver is divided into the left, right, quadrate, and caudate lobes seen either from the ventral abdomen or from a right intercostal (Fig. 9.2). The left lobe lies to the left of midline and is divided into the approach through the 9th to 11th intercostal spaces. The right lateral left lateral and left medial sublobes; the right lobe lies to the right of and caudate portal vessels usually arise together from the portal vein. midline and is divided into the right medial and right lateral sublobes. The intrahepatic branches of the hepatic arteries are not normally The quadrate lobe lies in the median plane and is partially fused to the seen within the liver. Nevertheless, the main hepatic artery may occa- right medial sublobe. The gallbladder is located between these two sionally be identified extending cranioventrally and to the right, off of lobes and, along with the portal vein, divides the liver into the left and the celiac artery. The hepatic artery lies dorsal to the portal vein and right lobes. The caudate lobe consists of the caudate and papillary pro- common bile duct, especially on a right lateral intercostal window. It cesses, which are connected by a short isthmus bounded ventrally by is best distinguished from the common bile duct or portal veins by the portal vein and dorsally by the caudal vena cava. The papillary Doppler ultrasonography. process, located more centrally, is sometimes mistaken for the body or left limb of the pancreas as the transducer is fanned caudally through Size the liver and stomach in transverse image planes. The cranial pole of Determination of liver size is based on subjective assessment. Radio- the right kidney is situated in the renal fossa of the caudate process of graphic criteria of liver margin extension beyond the costal arch may the caudate lobe. The divisions between liver lobes are not well seen be used for ultrasound assessment. Breed and body conformation ultrasonographically except when peritoneal fluid is present. The must be taken into account. For example, deep-chested dogs may branches of the hepatic artery and bile ducts are also poorly seen appear to have smaller livers than other breeds even though there is within the liver. Nevertheless, the location of hepatic arteries can no clinical or biochemical evidence of liver disease. Cats have less be determined by their characteristic arterial signals using Doppler variability because of more uniform body size, but liver size determi- interrogation. nation can sometimes be confounded by isoechoic falciform fat. 358 Small Animal Diagnostic Ultrasound Hepatic ducts Right Medial Quadrate Left Medial Hepatic artery Right Lateral Celiac artery Bile duct GB Left Lateral Cranial Cranial mesenteric pancreatico- artery duodenal artery Caudate Right PV gastroepiploic CVC A artery B 10 9 8 7 1 2,3 4 5 6 C D Fig. 9.2 Hepatic schematics of the arterial, venous (portal and hepatic) and the biliary system super- imposed over the hepatic parenchyma. A, Diagrammatic representation of major anatomic relationships in the liver region showing the aorta and hepatic arterial supply (in red), caudal vena cava and hepatic veins (in blue), portal veins (in purple), and biliary tract (in green). B, Diagrammatic representation of the major hepatic and portal venous branches to the liver lobes. CVC, Caudal vena cava; GB, gallbladder; PV, portal vein. C, Sagittal scan planes 1 to 6 corresponding to ultrasound images in Fig. 9.3, A to F. D, Transverse scan planes 7 to 10 corresponding to ultrasound images in Fig. 9.3, G to J. RK GB L 6 6 D A B CHD LPV PV LHV HV 6 D 6 CVC 8 A C D LPV LPV LHV LHV S L 6 D L 6 D 8 E F PV VC A PV CVC A G1 G2 Fig. 9.3 Ultrasound scans of the liver corresponding to the sagittal and transverse planes in Fig. 9.2, C and D. A, Plane 1, right kidney (RK) and right caudate liver lobe (L). Echogenicity of the liver parenchyma is normally equal to or slightly greater than that of the right kidney cortex. D, diaphragm. B and C, Plane 2 and plane 3. Depending on the exact scan plane, the gallbladder (GB), portal vein (PV), and caudal vena cava (CVC) may be imaged separately or together. The gallbladder appears as an anechoic structure with a thin, poorly visualized wall. The common hepatic duct (CHD) may be visualized ventral to the portal vein. Hepatic veins (HV) can be seen entering the caudal vena cava near the diaphragm (D). D, Plane 4, aorta. In this plane, the aorta (A), left hepatic vein (LHV), and left portal vein (LPV) are visualized. E, Plane 5, left hepatic lobes (L). The left portal vein (LPV) can be distinguished from the left hepatic vein (LHV) by the echogenic peripheral echoes surrounding all portal structures. D, Diaphragm. F, Plane 6, left lateral hepatic lobe (L). Terminal branches of the left portal vein (LPV) and left hepatic vein (LHV) can be visualized. The head of the spleen (S) is also visualized caudal to the liver. The spleen is normally more echogenic than the liver. D, Diaphragm. G, Plane 7, computed tomographic (CT) image (G1) and corresponding ultrasound image (G2) of the right lateral transverse view at the 11th intercostal space. This transducer position allows visualization of the aorta (A), caudal vena cava (VC or CVC), and portal vein (PV) by using the caudate and right liver lobes as an acous- tic window. The top of the ultrasound image represents the right side of the dog, and ventral is to the right because of the transducer’s orientation (see Fig. 9.1, C). Visualization of this region from the ventral abdomen is difficult because of interference from gas in the stomach. Continued 360 Small Animal Diagnostic Ultrasound GB LPV PV GB LHV RPV 6 H I GB LPV HV 6 BD PV HV D 8 CVC J CV 0 K .18 1 .18 1 2 m/s 3 BD L Fig. 9.3, cont’d H, Plane 8, right portal vein. The right portal vein (RPV) can be seen branching from the portal vein (PV) at this level. Cranial to this point, the portal vein continues ventrally and to the left as the left portal vein. GB, Gallbladder. I, Plane 9, left portal vein (LPV) and left hepatic vein (LHV). The left portal and left hepatic veins are now parallel to the scan plane. A portion of the gallbladder (GB) is also visible. J, Plane 10, left portal vein (LPV), caudal vena cava (CVC), and hepatic veins (HV). The left portal vein and the entrance of the hepatic veins into the caudal vena cava are clearly seen. The diaphragm (D) and a portion of the gallblad- der (GB) are also visible. K, Longitudinal image of normal cat liver with portal vein (PV), caudal vena cava (CV), and bile duct (BD, arrow). L, Longitudinal image of the bile duct (BD ) in cat. It is seen as an anechoic tube between the electronic cursors (0.27 cm). Color Doppler imaging shows lack of flow within the bile duct, which distinguishes it from the portal vein. The hepatic artery is seen just dorsal to the bile duct in the far field. CHAPTER 9 Liver 361 L LIVER R KIDNEY RK A B Fig. 9.4 Normal liver size (A) and hepatomegaly (B). A, On this sagittal image, the caudoventral margin of the medial sublobe of the left lateral lobe is seen to cover the cranial margin of the stomach. (L, liver; RK, right kidney). B, Sagittal image shows an enlarged liver that more completely covers the stomach and has enlarged rounded caudoventral margins, consistent with hepatomegaly. Liver enlargement should be suspected when there is (1) increased changes in echogenicity must be substantial to confidently conclude distance between the diaphragm and stomach, (2) increased extension that one or more of these organs is abnormal. of the liver ventral to the stomach or ventral to the right kidney, and (3) rounded liver margins (Fig. 9.4). Gallbladder and Biliary System Reduced liver size should be suspected when there is (1) poor visu- The gallbladder is seen as an anechoic, round to oval structure just to alization of the liver even without excessive stomach gas, (2) decreased the right of midline in most liver scans (see Fig. 9.3, B, H to J). The size distance between the diaphragm and stomach, and (3) decreased ex- is variable, depending on when the animal was fed, with as much as tension of the liver ventral to the stomach or poor visualization of the 53% of newly produced bile accumulating within the gallbladder dur- liver cranial to the right kidney. Abdominal radiographs should be ing prolonged withholding of food.22 Normal canine gallbladder vol- used to evaluate liver size in questionable cases, and the ultrasound ume is reported to be 1.0 mL/kg or less after food is withheld for interpretation should always be correlated with the clinical signs and 12 hours.23,24 Gallbladder motility can be determined by measuring the other diagnostic results. Some diseases associated with abnormalities gallbladder volume (length 3 width 3 height at largest dimension 3 0.53). in liver size are noted in Box 9.2. If the canine gallbladder volume is less than 1 mL/kg body weight, normal gallbladder contractility is assumed. If the volume is greater Hepatic Parenchyma than this, contractility can be determined by measuring the ejection The normal hepatic parenchyma has a uniform, medium level of fraction using gallbladder volumes before and at timed intervals after echogenicity (see Fig. 9.3). Only the hepatic and portal veins inter- a test meal. Normal percent contractility (in dogs) is greater than rupt this uniform echo pattern. The hepatic arteries and bile ducts 25%.12 In the cat, gallbladder volume is unrelated to body weight and run adjacent to the portal veins, as described previously, but usually is reported as 2.37 to 2.47 mL.25,26 Assessment of postprandial gallblad- cannot be seen in the normal animal. The parenchymal texture of der volume in cats has also been studied, with a mean gallbladder volume the liver is coarser than that of the spleen. At the same scanning at 120 minutes after meal ingestion of 0.88 to 0.94 mL.25 Dependent depth and instrument gain settings, echogenicity of the liver is nor- echogenic, nonshadowing sediment (biliary sludge) may be present mally equal to that of the cortex of the right kidney or slightly more within the lumen in normal dogs.27 More recent reports indicate that or less echogenic (see Fig. 9.3, A).19-21 In the cat, a large amount of biliary sludge may indicate biliary stasis and decreased gallbladder renal cortical lipid deposition will result in a hyperechoic appear- emptying.28 The gallbladder wall is poorly seen or appears as a thin (1 ance relative to the hepatic parenchyma. The spleen has a somewhat to 2 mm) echogenic line in dogs.1,29,30 In cats, normal gallbladder wall higher echo intensity than the liver, but direct comparisons in the thickness is considered to be less than 1 mm.31 normal animal are difficult at the same depth except in the cranial The intrahepatic and extrahepatic bile ducts are not visualized in abdomen near the head of the spleen (see Fig. 9.3, F). There is usu- normal dogs and cats. The normal canine bile duct is not typically seen but ally a greater difference in echogenicity between the left kidney is occasionally identified ventral to the main portal vein (see Fig. 9.3, C). cortex and spleen than between the right kidney cortex and liver. The diameter of the common bile duct in normal dogs is reported to be This indirectly supports the observation that the spleen is slightly less than 3 mm.32,33 The common bile duct is consistently identified in more echogenic than the liver in normal animals. In practice, there normal cats, with a reported diameter of 4 mm or less.34 (see Fig. 9.3, K, L). seems to be enough individual variability to make it difficult to rec- The bile duct can be identified immediately ventral to the portal vein, ognize subtle changes in the diffuse echogenicity of the liver, spleen, surrounded by hyperechoic tissue, and distinguished from adjacent or kidney that would consistently assist with diagnosis. Therefore vessels in the porta hepatis by use of Doppler ultrasonography. 362 Small Animal Diagnostic Ultrasound BOX 9.2 Clinical Disease Processes Associated With Abnormal Liver Size Hepatomegaly b. Restrictive pericarditis 1. Hepatitis (infectious or noninfectious) c. Pericardial effusion a. Infectious (viruses; bacteria; fungi [histoplasmosis]; parasites [schisto- d. Thrombi or masses in caudal vena cava somiasis]) e. Budd-Chiari syndromes b. Immune-mediated hepatitis (acute, chronic hepatitis) 5. Obstruction of biliary system c. Medication-associated hepatotoxicity a. Biliary atresia d. Disseminated intravascular coagulopathy b. Choledochal cysts 2. Storage abnormalities c. Cholelithiasis a. Glycogen storage diseases Microhepatica b. Disorders of lipid storage (hepatic lipidosis, diabetes mellitus, hyperad- 1. Congenital portosystemic shunts renocorticism) 2. Ductal plate malformations 3. Infiltrative disease 3. Hepatic fibrosis a. Primary hepatic tumors (e.g., hepatocellular carcinoma) a. Chronic hepatitis b. Infiltration of lymphoid cells (leukemia, lymphoma cells), mast cells, b. End-stage liver disease (cirrhosis) histiocytic cells 4. Hypotension c. Metastatic neoplasia (carcinoma, sarcoma) 5. Hypovolemia 4. Hepatic congestion 6. Hypoadrenocorticism a. Congestive heart failure (right sided) DOPPLER EVALUATION OF THE NORMAL LIVER Doppler ultrasonography adds important information about blood 0 flow hemodynamics within the liver, such as with portosystemic shunts, portal hypertension, and thrombosis. A description of general Doppler techniques can be found in Chapter 1. .06 Doppler evaluation of the normal liver is presented here, and the ab- normal findings associated with specific liver diseases are presented in their appropriate sections. Doppler ultrasonography of the liver may be used to evaluate flow velocity, flow direction, and spectral patterns of the hepatic arteries, caudal vena cava, and the hepatic and portal .06 veins (Fig. 9.5). Flow velocity is usually determined by the uniform m/s 5 insonation method in which a large Doppler sample volume is used to overlap the walls of the vessel.35,36 The maximal velocity method can also be used; in this method, maximal velocity measured in the center of the vessel lumen is multiplied by 0.57 to determine mean flow veloc- ity.37 This method may be preferable if there is considerable artifact as LIVER LT TRANS a result of wall motion with the uniform insonation method. Incident angles of less than 60 degrees should be used with both methods to Fig. 9.5 Color doppler image of a normal dog liver. Transverse image obtain reliable velocity measurements. These two Doppler techniques of the left side of the liver showing the hepatic veins (in blue, blood flow for determining flow velocity, if used properly, show no significant dif- coursing away from transducer toward the caudal vena cava) and portal ference in reliability.35 veins (in red, blood flow coursing toward the transducer, away from the porta hepatis into the periphery of the liver). Caudal Vena Cava and Hepatic Veins The Doppler spectral patterns of the caudal vena cava and hepatic veins are complex and depend on changes in cardiac activity, respiration, and intra-abdominal pressure.38,39 Therefore, the caudal vena cava and After right atrial contraction, there is a rapid flow of blood from the hepatic vein waveforms should be obtained at normal end-expiration hepatic veins and caudal vena cava toward the right atrium during in a quiet animal whenever possible. The vena cava is visualized by use rapid atrial filling (S wave). Flow slows as the right atrium becomes of either a ventral (subcostal) or right intercostal approach, ventral and filled, and velocity starts to return toward baseline (V wave). When the to the right of the aorta and slightly dorsal and to the right of the portal tricuspid valve opens and the right ventricle fills, there is a second wave vein (see Fig. 9.3, G). Doppler measurements of the caudal vena cava in of rapid flow out of the hepatic veins toward the right atrium (D the liver region are obtained with the vena cava in a long-axis view and wave); this wave is slightly smaller than the first. The right atrium the transducer angled cranially to obtain an incident angle of less than contracts again, and the cycle is repeated. Therefore the pulsatile flow 60 degrees (see Fig. 9.3, C). within the hepatic veins and caudal vena cava is referred to as triphasic The normal caudal vena cava and hepatic venous pulsed Doppler because there are two antegrade pulses (S and D waves) and one retro- waveform demonstrates marked variation in the direction and velocity grade pulse (A wave). The A wave is so named because it occurs during of flow because of right atrial activity (Fig. 9.6).38,39 Hepatic venous atrial systole, whereas the S and D waves are named because they occur flow is retrograde into liver during right atrial contraction (A wave). during ventricular systole and diastole, respectively. CHAPTER 9 Liver 363 11:05:11 AM describes the use of the right portal vein branch rather than the main portal vein for Doppler evaluation.41 Doppler ultrasonography has been used to assess the appearance and closure time of the ductus venosus in neonatal Irish wolfhounds.42 In this study, the ductus venosus was closed at 9 days after birth. This is slightly longer than reported using more invasive techniques.43 As- sessment of closure of the ductus venosus in other breeds has not been reported but is expected to be similar. Until further studies are avail- able, a patent ductus venosus detected after the first few weeks of life 0.18 should be considered abnormal. Confirmation by other means, such as a laboratory test results, scintigraphy, or computed tomography (CT) v angiography, is essential. m/s Normal portal blood flow velocity (PBFV) in the main portal vein is relatively slow and nearly uniform with small undulations in the Doppler spectral pattern (Fig. 9.7). The portal vein has intestinal capil- laries at one end and hepatic sinusoids at the other, so it is insulated D from the variable pressures and flow that normally exist in arteries and S 0.18 systemic veins. The small undulations in velocity are thought to be the Fig. 9.6 Pulsed wave doppler image of a normal hepatic vein. Here result of respiratory motion of the diaphragm, with velocity increasing is an example of a duplex image showing the brightness (B)-mode im- during expiration and decreasing during inspiration.35,37 age, electrocardiogram (green line), and spectral tracing. The B-mode Mean PBFV in the main portal vein in normal unsedated dogs has image shows the gallbladder, right medial and quadrate liver lobes, been reported to be 18 6 7.6 cm/sec,37 14.7 6 2.5 cm/sec,35 and hepatic veins, and the sample gate with angle correction, placed in the 19.2 6 0.5 cm/sec.44 In another study, mean PBFV was determined common tributary of these two veins. The waveform consists of two based on weight: 16.95 6 5.79 cm/sec in dogs weighing up to 10 kg, retrograde waves (a and v), which represent blood flow away from the 16.98 6 3.04 cm/sec in dogs weighing between 10.1 and 20.0 kg, and heart, and two antegrade waves (S and D), which represent blood flow 17.39 6 4.77 cm/sec in dogs weighing 20.1 kg or more.41 Thus the toward the heart. The various wave components are a result of pressure normal range of PBFV is approximately 10 to 25 cm/sec (see Fig. 9.7, A). differences within the right atrium and the hepatic veins. The A wave (a in the figure) occurs after the P wave on the ECG. It results from The mean PBFV in healthy, unsedated cats has been reported to be 10 to elevated right atrial pressure caused by right atrial contraction that cre- 12 cm/sec45 or 17.1 cm/sec (range, 9.7 to 18.1 cm/sec) (see Fig. 9.7, B).46 ates a burst of reversed flow into the hepatic veins. The S wave occurs Mean portal blood flow (PBF) in dogs has been reported to be 31 6 after the QRS complex. It results from a negative pressure in the right 9 mL/min/kg,37,47 40.9 6 13 mL/min/kg,35 and 33.8 6 1.2 mL/min/kg.44 atrium with corresponding flow from the hepatic veins to the heart. This When determined by weight, it was 51.7 6 20.55 mL/min/kg in dogs negative pressure is created by right atrial relaxation and movement of weighing up to 10 kg, 38.28 6 8.15 mL/min/kg in dogs weighing be- the tricuspid annulus toward the cardiac apex. The V wave (v in the fig- tween 10.1 and 20.0 kg, and 32.19 6 13.23 mL/min/kg in dogs weigh- ure) occurs toward the end of the T wave on the ECG tracing. It results ing 20.1 kg or more41 by four different investigators. Mean PBF (mL/ from increased right atrial pressure secondary to overfilling against a min/kg) is calculated by multiplying the cross-sectional area (cm2) of closed tricuspid valve. The D wave is a diastolic wave that results from the portal vein by the mean PBFV (cm/sec) and dividing by body negative pressure created in the right atrium secondary to blood flow- ing into the right ventricle (away from the hepatic veins) through an weight (kg). Eating is known to increase normal portal flow, whereas open tricuspid valve. exercise and upright posture may decrease normal portal flow. These considerations should be taken into account in interpreting PBFV and PBF, but the magnitude and variability of these effects as Normal respiratory activity affects flow velocity in the hepatic veins they apply to clinical cases are still largely unknown. Therefore, it and caudal vena cava. During inspiration, there is increased flow is important to perform PBFV and PBF Doppler evaluations in a toward the heart because of decreased intrathoracic pressure and consistent manner (e.g., unsedated animals, before feeding, in lateral increased intra-abdominal pressure. These effects are reversed with or dorsal recumbency) to reduce variability and provide meaningful expiration. Contraction of the abdominal muscles against a closed results. glottis after inspiration, known as a Valsalva maneuver, raises intra The portal vein congestion index (CI, cm 3 sec) is a measure of thoracic pressure and causes flow to diminish or stop within the caudal vascular resistance and may increase with hepatic cirrhosis and other vena cava. This can occur in animals during straining or vocalization. liver diseases.37 The CI is determined by dividing the portal vein cross- sectional area (cm2) by the average blood flow velocity (cm/sec). The Portal Vein CI for normal, unsedated dogs is approximately 0.04 6 0.015 cm 3 sec.37 The portal vein is visualized by either a ventral or right intercostal ap- A second report lists CI as 0.022 6 0.01 cm 3 sec for dogs weighing proach, whereby it can be seen entering the liver at the porta hepatis. up to 10 kg, 0.039 6 0.009 cm 3 sec for dogs weighing between It is ventral and slightly to the left of the caudal vena cava. The rela- 10.1 and 20.0 kg, and 0.043 6 0.009 cm 3 sec for dogs weighing at tionship of the aorta, caudal vena cava, and portal vein is easiest to least 20.1 kg.41 recognize in the transverse view (see Fig. 9.3, G). Nevertheless, Doppler measurements of the main portal vein at the porta hepatis are obtained Hepatic Arteries with the portal vein in a long-axis view and the transducer angled Normal hepatic arterial blood flow has a typical low resistance arterial cranially to obtain an incident angle of less than 60 degrees (see waveform with a distinct systolic and diastolic component. The resis- Fig. 9.3, G). A scanning protocol for systematic ultrasound evaluation tive index (RI) is a unitless measure of the resistance of blood flow of the portal system in dogs has been proposed.40 An additional report through an organ. It is calculated by subtracting the end-diastolic 364 Small Animal Diagnostic Ultrasound 0 0 1 1 2 .12 .06 2 3 4 3 5 .12 .06 4 m/s 6 m/s 39.36 27.90 cm/s cm/s PV 1 PV A 5.35 48.10 B 3.5s 34.98 Fig. 9.7 Doppler evaluation of normal portal veins in a dog (A) and cat (B). A, This triplex image shows a sagittal brightness (B)-mode image of a dog portal vein (PV ) with color Doppler and a pulse-wave (PW) spec- tral image tracing. The color Doppler image depicts portal blood flow as blue (away from transducer, toward the liver to the left). The PW spectral image shows a fairly uniform portal vein blood flow at 223.0 cm/sec (at the point of the electronic cursor, 1). The blood flow is below the baseline (away from transducer). The small positive waveforms depict adjacent arterial blood flow. B, Cat portal vein (PV ) triplex image shows a portal vein velocity of approximately 210 cm/sec. velocity from the peak systolic velocity and then dividing the result by have mixed echogenicity.49-54 Mixed patterns result when there are dilated the peak systolic velocity. The mean peak systolic velocity and RI of the venous sinusoids, central necrosis, or hemorrhage. One study reported hepatic artery in normal, fasted dogs was reported to be 1.5 6 0.4 m/sec that hematomas may occur in approximately 35% of the cases.55 Target and 0.68 6 0.04, respectively.36 lesions have a hypoechoic rim with an isoechoic or hyperechoic center In humans, the RI is thought to be a reliable indicator of hepatic (see Fig. 9.9, A) and are occasionally seen with nodular hyperplasia but vascular resistance. In dogs, the importance of hepatic arterial RI mea- are more often associated with neoplasia.54 Clinical experience indicates surements in clinical cases remains to be determined. Two dogs with that nodular hyperplasia may appear similar to some forms of focal vacu- congenital arterioportal fistulas had a lower mean RI than normal, but olar hepatopathy and extramedullary hematopoiesis. When leisons are in three dogs with portal vein thrombosis and two dogs with acquired large, nodular hyperplasia may resemble hematomas, abscesses, necrosis, hepatic insufficiency, the RI was normal.36 Hepatic arterial RI determi- and primary or metastatic liver neoplasia. A liver biopsy (or fine-needle nations may potentially be useful for diagnosing congenital portosys- aspiration for cytology) is indicated, but the histologic or cytologic temic shunts or for serially evaluating changes in hepatic arterial appearance may be nondiagnostic or difficult to distinguish from hepa- flow after surgical intervention; however, differences in hepatic blood tocellular adenoma or well-differentiated hepatocellular carcinoma.56 flow in dogs of different ages or breeds may limit the usefulness of the Metastatic neoplasia can usually be ruled out on the basis of the liver procedure. Thus additional studies are indicated.36 sampling results. Therefore, a liver biopsy or aspirate primarily serves to rule out neoplasia or other diseases, not to confirm nodular hyperplasia. Contrast harmonic ultrasound may help in differentiating benign FOCAL OR MULTIFOCAL HEPATIC from malignant hepatic nodules.57 All malignant nodules were hypoechoic to liver parenchyma at peak contrast enhancement, PARENCHYMAL DISEASE whereas benign nodules were isoechoic to the surrounding normal Focal lesions in the liver as small as 5 mm or less can be visualized on contrast-enhanced liver. ultrasound examination, especially when the image is optimized for spatial resolution by using high frequencies and careful setting of the Cysts beam focus. Nevertheless, the ultrasound appearance of focal disease is Hepatic cysts are usually detected incidentally. Incidentally, clinical nonspecific, with a long list of potential etiologies. signs may be present if widespread polycystic disease replaces the liver parenchyma, mechanical compression of a vital structure oc- Nodular Hyperplasia curs, or the cyst becomes secondarily infected. Typical hepatic cysts Nodular hyperplasia has a variable appearance and cannot be differenti- are characterized by thin and well-defined walls, absence of internal ated from other nodular diseases on the basis of ultrasonography alone echoes, sharp distal borders, peripheral reflective and refractive (Figs. 9.8 and 9.9). Nodular hyperplasia may occur in up to 70% of older zones, and strong distal acoustic enhancement (Fig. 9.10, A).58 Rever- dogs48 but is not detected this frequently during hepatic ultrasound ex- beration artifacts may be present at the proximal portion of the cyst aminations because the lesions are often isoechoic to surrounding liver and should not be mistaken for an irregular wall or internal debris. parenchyma. The earliest age that nodules were found was between 6 and Cysts may be congenital or acquired and may be solitary (Fig. 9.11; 8 years in a series of 50 consecutive postmortem examinations of dogs, see also Fig. 9.10, B, C) or multiple (see Fig. 9.10, A). They may affect whereas all dogs older than 14 years had nodules.49 These benign lesions the parenchyma or biliary tract. Acquired cysts (bilomas or biliary may appear isoechoic, hypoechoic, moderately hyperechoic, cavitary, or pseudocysts) that form outside the biliary tract may result from CHAPTER 9 Liver 365 A B LIVER C D F E Fig. 9.8 Hepatic nodular hyperplasia, spectra of patterns in the dog. A, Lesions (arrows) associated with focal nodular hyperplasia may be isoechoic and not visualized ultrasonographically unless they appear as a nodule near the liver margin. They sometimes appear more echogenic (B and C), less echogenic (D), or with mixed echo- genicity compared with surrounding hepatic parenchyma (E). F, Large hyperplastic hepatic nodules seen in cross section of a specimen of a dog liver. Note mild convexity of the liver surface in association with the nodules. G, Gross cut section of a dog liver showing diffuse, multiple, small, benign hyperplastic nodules. L, Liver. (F, Courtesy Dr. R. Fairley, Gribbles Veterinary Pathology. In Zachary JF, McGavin MD: Pathologic basis of veterinary disease 5th edition, Elsevier, St. Louis, 2012; G, courtesy Washington State University College of Veterinary Medicine, Washington Animal Disease Diagnostic Labora- tory, Pullman, WA.) G 366 Small Animal Diagnostic Ultrasound 0 0 1 1 1 1 2 2 3 3 4 4 5 6 5 A B 7 6 0 1 1 Fig. 9.9 Benign hepatic hyperplasia in a geriatric greyhound presented for ascites; a spectra of patterns in same patient. 2 A, Isoechoic to hypoechoic hyperplastic nodule (1.32 cm, between electronic cursors), slightly protruding from the liver surface. Truly isoechoic nodule protrudes from the caudal liver margin (arrows). 3 B, Hypoechoic hyperplastic nodule (1.20 cm, between electronic cursors) in the cranioventral liver. C, Mixed echogenicity hyper- 4 plastic nodule (1.39 cm, between electronic cursors). 5 C 6 trauma or inflammatory disease, but this diagnosis requires a com- history, physical findings, and laboratory results are used to narrow the patible history.59 Cyst-like structures may originate from the biliary differential diagnosis. Ultrasound-guided aspiration of the cyst and tract in Cairn terriers,60 West Highland white terriers,61 and other cytologic evaluation with bacteriologic culture are advisable. Serial breeds.62-64 The kidneys should always be evaluated because polycys- examinations also aid diagnosis because there is little or no change tic renal disease may accompany hepatic cysts in both the dog and in the cyst on serial examinations compared with cyst-like masses cat.60,61,63-67 Careful scanning sometimes suggests communication of produced by trauma, inflammation, necrosis, or neoplasia. the cyst-like structure with the biliary tract if localized ductal ectasia is present. Segmental dilations of the bile duct, consistent with cho- Hematoma ledochal cysts and unassociated with biliary obstruction, have been The internal appearance of a hematoma changes as it ages (Fig. 9.13; reported in cats.68 Cysts smaller than 1 cm may be recognized be- see also Fig. 9.12). Acute parenchymal hemorrhage less than 24 hours cause there is a marked change of acoustic impedance at the liver- old is echogenic.70 Within the first week, a hematoma becomes more cyst interface. In our experience, a large number of very small cysts hypoechoic and better defined, with a mixture of solid and fluid may also produce a hyperechoic lesion because of the presence of components. Over the next several weeks, the hematoma becomes multiple reflecting interfaces. increasingly less distinct as fluid is resorbed and spaces are filled with On occasion, cysts do not meet all of the aforementioned criteria granulation tissue. (Fig. 9.12; see also Fig. 9.10, C, D). They may possess irregular walls, The variable appearance of a hematoma over time is similar to that septations, internal debris, or solid elements. In such cases, the differ- of necrosis, abscess, or tumor and cannot be differentiated by ultraso- ential diagnosis must be widened to include traumatic, toxic, inflam- nographic appearance alone. Hematomas have also been reported to matory, or neoplastic disease. Complex cystic lesions include hematoma, occur in approximately 35% of cases of nodular hyperplasia, which abscess, cystic metastasis, hemorrhage, or necrotic tumors. Biliary may give them a complex appearance.55 Other important factors, such cystadenomas, seen primarily in older cats, are benign hepatic tumors as history of trauma or coagulation disorder, age of the animal, pres- that have a characteristic cystic appearance.69 Vascular lesions such as ence of fever and leukocytosis with a left shift, serum biochemical portacaval shunts and arteriovenous fistulas may also appear cyst-like, abnormalities, and change in appearance on serial examinations, but these can be readily identified using Doppler evaluation. The must be considered. Without a history of trauma, bleeding disorder, CHAPTER 9 Liver 367 0 1 LIVER 2 3 4 LT LIVER A B 5 D 0 5 LIV LT SAG C Fig. 9.10 Hepatic cysts. A, A typical hepatic cyst (black arrows) has a thin and well-defined far wall, no in- ternal echoes or septations, and strong distal acoustic enhancement (white arrow). B, This small hepatic cyst (approximately 5 mm, arrow) has anechoic fluid and far wall enhancement but little distal acoustic enhance- ment because it is small. The liver is enlarged, based on a thickened, rounded caudoventral margin. C, Atypical hepatic cysts do not meet all of the criteria of a typical cyst and may have septations, walls that are irregular or thickened, or echogenic internal contents even though there is distal acoustic enhancement (arrows). D, Multiple, small hepatic cysts as seen here (arrows) may be difficult to differentiate from hepatic neoplasia. Distal acoustic enhancement is poor. Benign hepatic tumors, such as hepatic cystadenomas, and some malignant primary or metastatic hepatic tumors may have a similar appearance. An ultrasound-guided aspiration is indicated in questionable cases. L, Liver. or laboratory findings suggestive of other diseases, the lesion should many were hypothermic. In addition, affected cats were less likely to be percutaneously aspirated for culture and cytology to obtain a de- have elevated liver enzymes than dogs, making the diagnosis in cats finitive diagnosis. more difficult.76,77 A liver abscess may produce lesions that run the spectrum of an- Abscess echoic, hypoechoic, hyperechoic, and mixed patterns, depending on Hepatic abscesses are relatively uncommon in dogs and cats, with the age of the abscess and the appearance of the centralized necrosis predisposing factors including septicemia, diabetes mellitus, pancre- (Figs. 9.14 to 9.16).72,74 In some cases, the contents are isoechoic, which atitis, hepatic neoplasia, previous liver biopsy, long-term steroid use, makes the differentiation from normal or a solid mass extremely dif- and concurrent hepatobiliary disease.71-78 Hepatic abscesses may be ficult. If gas is present within the abscess because of gas-producing associated with infection of other organs or organ systems, with bacteria, a hyperechoic pattern is found (see Figs. 9.14, C, and 9.15, positive bacterial cultures in the majority of cases. More than one A, B).72,74 Distal acoustic enhancement is variable, depending on the organism may be involved; Escherichia coli appears to be the most contents of the abscess.72,74 Diagnosis depends on correlation with the common.71 Hepatic abscesses have been reported in dogs with pene- history and ultrasound-guided aspiration of the lesion for culture trating foreign objects, liver lobe torsion, and hepatic neoplasia.74,75 and cytology. Once a hepatic abscess is diagnosed, serial ultrasound Compared with dogs, fewer cats with hepatic abscess were febrile and examinations are ideal for monitoring the response to therapy. 368 Small Animal Diagnostic Ultrasound 0 0 1 1 1 2 2 2 3 3 4 4 5 5 RT LIVER RT LIVER A B 6 6 0 Fig. 9.11 Large simple, solitary hepatic cyst in a 14-year old 1 neutered male diabetic shiba inu presenting for vomiting. Liver enzymes were elevated and a mucocele (not shown) was found. This cyst was felt to be incidental to the presenting signs. 2 A, Sagittal image of a large anechoic cyst arising from the right caudal margin of the liver showing acoustic enhancement of the far field tissues. B, Transverse image of the cyst (3.84 3 4.07 cm, 3 between electronic cursors) shows strong acoustic enhance- ment. Note the reverberation artifact in the near field of the cyst, RK 4 mimicking internal debris. C, Sagittal image shows the cyst’s rela- tionship to the right kidney (RK ) and adjacent liver blood vessels. Near-field artifact is seen, as in B. 5 C 6 Liver abscesses in the dog usually have a centralized anechoic to of liver abscess using ultrasound-guided ethanol infusion in combina- hypoechoic region with an irregular, poorly defined hyperechoic mar- tion with antibiotics has also been described.80 gin (see Fig. 9.14, B).71,72,74 There may be small echogenic clumps Chemical, viral, toxic, or immune-mediated insults may cause within the abscess. Distal acoustic enhancement may be seen if lique- hepatic necrosis, as may liver lobe torsion (see later section). The pat- faction necrosis is present. Abscesses containing gas may be diffusely tern may be multifocal or diffuse and is similar to that of hepatitis, hyperechoic or have multifocal areas of increased echogenicity; they multifocal abscess formation, or neoplasia (Figs. 9.17 and 9.18). may also have acoustic shadowing.72,74 Doppler evaluation may be helpful to confirm its avascular nature, Liver abscesses may be safely aspirated for diagnosis and culture but in practice necrosis cannot be reliably differentiated from absces- with a thin needle (such as a spinal needle), especially if they are deep sation. The history may suggest the agent responsible, but sampling is within the parenchyma or surrounded by a thick rim that prevents required to eliminate a diagnosis of necrosis associated with other leakage.72,79,80 The sonographer should choose the shortest possible conditions. route that avoids vital structures. To prevent contamination of the pleural cavity, an intercostal route should not be chosen. Multifocal or Trauma multilocular abscesses or those not easily accessible percutaneously Laceration of the liver can occur with blunt abdominal trauma. The must be drained surgically, and a sample for culture can be obtained at laceration can be small or large and can be accompanied by a hema- that time. Percutaneous needle drainage in combination with appro- toma. Subcapsular hematomas or capsular disruption can also occur priate antibiotic administration for treatment of hepatic abscess (Fig. 9.19). The need for surgery depends on the size of the laceration, has been reported in four dogs without complications, and one surviv- the amount of blood in the peritoneal cavity, and the patient’s clinical ing dog in this study had nothing but antimicrobial treatment.72 In status. CT is considered the gold standard for evaluating blunt another report, 1 of 13 dogs with hepatic abscess had a successful out- abdominal trauma in humans but ultrasound still has an important come when treated with antibiotics.71 One of the authors of this chapter role in the initial triage of patients that are not hemodynamically successfully treated a dog with multiple hepatic abscesses with antibi- stable and cannot undergo CT. Contrast-enhanced ultrasonography otics alone. Thus it may be reasonable to treat with antimicrobials was found to improve the initial detection of liver lesions in patients alone if other interventions are not possible. Successful management with blunt abdominal trauma and also help with follow-up healing CHAPTER 9 Liver 369 0 HEP CYST MASS 5 STOMACH LIVER A B 0 0 1 1 2 5 3 4 10 RT HEP CYST TRANS C 5 D 97% 0 1 5 2 10 RT HEP CYST SAG E F Fig. 9.12 Complex, large liver cysts and hematomas in a 2-year-old spayed female serval (large cat) with an enlarged abdomen after an injury from a fall. A, Sagittal ultrasound shows an enormous, complex hepatic cyst. The small gallbladder is identified in the upper left of the image. B, Moving caudally, a hepatic cyst (HEP CYST) is seen in addition to a very large, homogeneous echogenic structure (MASS). C, Power color Doppler image of the mass in B showing its avascular nature. Careful observation showed slight motion and layering of its contents. The appearance along with the history allowed a presumptive diagnosis of hemorrhage within a cyst. D, Transverse image shows complex cystic structures with septa- tions and varying degrees of echogenic content. E, An extended field-of-view image was necessary to accurately assess the full size of the numerous and complex cysts. F, Surgery photo show two huge, thin-walled hemorrhagic cysts arising from the liver; a smaller G cyst is also seen. G, Surgical photograph showing the hemorrhagic and pedunculated nature of the cyst. 370 Small Animal Diagnostic Ultrasound assessment.81-86 Liver lesions may appear hyperechoic, mixed echogenic, or hypoechoic with non–contrast-enhanced ultrasound. These patterns change after administration of contrast material. The normal liver parenchyma becomes hyperechoic after contrast administration, but the area of injury appears anechoic or hy- poechoic, with well-defined margins. Active bleeding or extravasation can also be visualized.87 The use of contrast agents enhances visualiza- tion of liver injuries relative to what can be achieved without their use. Contrast-enhanced sonography may also eventually prove useful in animals for evaluating solid organ injuries from blunt abdominal trauma. Liver Lobe Torsion Although rare, liver lobe torsion may lead to tissue infarction and necrosis, resulting in formation of a liver mass that is hypoechoic or mixed echogenicity. Cystic or cavitated areas may be present Fig. 9.13 Hematoma. A sagittal view through the left liver lobes in this (see Figs. 9.17 and 9.18).75,88-90 Echogenic foci with reverberation cat revealed a lesion with mixed anechoic and echogenic contents artifact or shadowing may be seen within the mass if gas- (arrows). This represents a hematoma from a liver biopsy performed producing organisms are present. Absent or reduced blood flow 2 weeks previously. Liver Right GB GB A B 0 LIVER 1 5 LIVER LT C D Fig. 9.14 Liver abscesses appearing in a spectrum of patterns. A, This liver abscess (arrow) has mixed an- echoic and echogenic contents produced by necrotic debris and no obvious rim. It would be difficult to differenti- ate this lesion from an organizing hematoma or focal neoplastic process without a compatible history and clinical signs. An abscess was confirmed at necropsy and was attributed to adjacent pancreatitis. (GB, gallbladder) B, Thick, rimmed, cavitated mass with anechoic contents (between calipers). (GB, gallbladder) C, A gas containing abscess is present within the liver (arrow). The large amount of gas is creating a “dirty” acoustic shadow. D, A large echogenic interface represents the boundary of a large liver abscess (between electronic cursors). CHAPTER 9 Liver 371 24.0 24.0 cm/s A LIVER B LIVER Fig. 9.15 Liver abscess in a 10-year old neutered male golden retriever with severe lethargy. A, A heteroge- neous, poorly demarcated mass is present within the liver, characterized by hypoechoic areas mixed with smaller, very echogenic foci, representing gas pockets. B, Color Doppler image shows good peripheral blood flow, but no flow within the abscess. C, Photograph of the hepatic abscess taken during surgery. The abscess bulges out from the surface of the liver. Severe suppurative hepatitis and abscessation were found on histology. C 0 0 GB 5 5 LIVER GB LT LIVER A B 0 Fig. 9.16 Liver abscess, characterized by cavitation, hy- perechoic liver parenchyma, and generally poor hepatic detail. A, Image of the liver showing the gallbladder (GB). B, Heterogeneous liver parenchyma with a lobulated hypoechoic lesion in the near field and poor overall detail. C, Hypoechoic 5 area surrounded by a thick hyperechoic wall, representing the liquefactive portion of the liver abscess. LIVER C 372 Small Animal Diagnostic Ultrasound 0 0 5 5 10 10 LIVER MID LIVER CD MID A B 0 .18 Fig. 9.17 Hepatic necrosis from liver lobe torsion. A, Sagittal image of the central portion of the liver shows a heterogeneous, complex mass. B, Sagittal image of the caudal margin of the liver shows a hypoechoic mass and a swollen, rounded liver margin. C, Color Doppler image of the mass shows no blood flow. The final .18 diagnoses were massive necrotizing bacterial hepatitis of the left 5 m/s lateral and caudate liver lobes, a result of localized ischemia because of lobar torsion. LIVER MASS C A B Fig. 9.18 Left lateral liver lobe torsion. A, Sagittal image of the left liver shows a maintained shape of liver lobe with mixed echogenicity (irregular hyperechoic nodules). Echogenic fluid is noted adjacent to the liver lobe. B, Intraoperative image of torsed left lateral liver lobe. CHAPTER 9 Liver 373 0 0 2 .02 1 5 5 .02 m/s R LIVER R LIVER A B Fig. 9.19 Liver laceration in a dog that fell from a truck. A, Heterogeneous area within the liver that pro- trudes from the liver capsule (5.20 3 2.83 cm, between electronic cursors). B, Color Doppler image of the area shows peripheral blood flow only. This patient had a hemoabdomen (not shown) and was later deter- mined to have immune-mediated anemia and thrombocytopenia. within the mass can be demonstrated using Doppler ultrasonogra- involvement of the liver is possible with either primary or metastatic phy. Abdominal fluid may accumulate secondary to necrosis, tumors, and they cannot be differentiated solely on the ultrasound venous congestion, or bacterial peritonitis. Torsion may be more appearance.49,50,52,53,58,96,101-105 The most common primary hepatic likely to occur in large breed dogs, although medium and small neoplasia in the cat is biliary cystadenoma (bile duct adenoma). breed dogs have also been affected.75,91 The left liver lobe (left me- Cholangiocarcinoma is the most common malignant hepatic tumor, dial, left lateral sublobes) are most commonly affected, although followed by hepatocellular carcinoma.98-100 As in the dog, lym- torsion of all lobes has been reported.88-94 Liver lobe torsions usu- phoma, MCT, and histiocytic sarcoma also occur. The ultrasound ally occur in dogs, but cases have been described in cats.92,95 In the appearance of canine and feline hepatic neoplasia is variable, and in largest reported series90 of liver lobe torsion in dogs, prognosis some cases, diffuse forms of the disease may show no ultrasound with surgical resection of the involved liver lobe was very good abnormalities. The same tumor may have a variety of appearances, with 11 out of 12 dogs surviving.90 even within the same animal (Figs. 9.24 and 9.25). Furthermore, different histologic types of primary or metastatic tumors may ap- Neoplasia pear remarkably similar ultrasonographically. Tumors may abscess Detection of parenchymal lesions compatible with hepatic neoplasia in or hemorrhage, altering their appearance because of necrosis, bacte- the dog and cat is one of the most important diagnostic and prognos- rial gas formation, or cavitation.77 Ultrasound alone cannot be used tic uses of ultrasonography, but CT provides a more detailed and to distinguish benign from malignant focal disease. Nevertheless, thorough evaluation of the liver. Nevertheless, ultrasonography is larger focal hepatic masses in the presence of peritoneal effusion more readily available and rarely requires sedation or anesthesia. In were associated with malignant liver disease in dogs.106,107 Nodules dogs, metastatic liver tumors, typically arising from neoplasia of the and masses can be hypoechoic or hyperechoic, or have mixed echo- pancreas, spleen and gastrointestinal tract, are 2.5 times more com- genicity. Target lesions (nodules with a hypoechoic rim and echo- mon than primary hepatic tumors.96-98 Primary hepatobiliary tumors genic or isoechoic center) are commonly, though not exclusively, are more common than metastatic tumors in cats.96,99,100 associated with neoplasia. In one report, one or more target lesions Primary liver tumors may appear as solitary masses confined to in the liver or spleen had a positive predictive value of 74% for ma- one lobe (Fig. 9.20), multifocal nodules involving several lobes, or lignancy, whereas finding multiple target lesions in multiple organs diffuse, coalescing nodules involving all lobes.96 The most common had a positive predictive value of 81% for malignancy.54 Multifocal primary hepatic tumor in dogs is hepatocellular carcinoma, which is target lesions are commonly encountered with carcinomas but are most often massive in classification. Other primary tumors include not limited to them.54 Hemangiosarcoma, lymphoma, and benign cholangiocarcinoma (second most common primary hepatic tumor conditions such as nodular hyperplasia, pyogranulomatous hepati- in dogs), neuroendocrine tumor (carcinoid), hemangiosarcoma, tis, chronic active hepatitis, and cirrhosis may also produce target leiomyosarcoma, and fibrosarcoma. Benign tumors include hepato- lesions.54 cellular adenoma and bile duct adenoma (Fig. 9.21).96-99,100-106 The Although the ultrasonographic appearance of hepatic neoplasia canine liver is frequently involved with disseminated histiocytic is extremely variable, the appearance of some common hepatic sarcoma, lymphosarcoma, and mast cell tumor (MCT).96 Metastatic tumors have been described. Hepatic lymphoma has a variety of tumors are commonly sarcomas of the spleen (hemangiosarcoma, ultrasound appearances.108-112 Commonly there is no abnormal fibrosarcoma, leiomyosarcoma; Fig. 9.22) or gastrointestinal tract, appearance; however, hepatomegaly, coarse parenchyma, ill-defined or carcinomas originating from the stomach, bowel, pancreas, thy- hypoechoic areas, hypoechoic nodules, target lesions, diffuse roid, anal gland, and mammary gland (Fig. 9.23). Focal or diffuse hypoechogenicity, and diffuse hyperechogenicity have been noted 374 Small Animal Diagnostic Ultrasound 0 0 5 5 10 LIVER LT 10 LIVER LT A B C Fig. 9.20 Hepatocellular Carcinoma in a Dog. A and B, Sagittal images of the left side of the liver show a very abnormal, nodular appearance of the liver parenchyma. C, Specimen photograph of the liver tumor within the left lateral lobe. (C, Courtesy Washington State University College of Veterinary Medicine, Washington Animal Disease Diagnostic Laboratory, Pullman, WA.) CHAPTER 9 Liver 375 LIVER A B LIVER LIVER C D LIVER E F Fig. 9.21 Primary hepatic neoplasia in a variety of appearances in the dog and cat. A to F: Hepatocellular carcinoma. A, Target lesion (hypoechoic rim and hyperechoic center) associated with hepatic carcinoma. Target lesions commonly indicate malignancy but have also been associated with benign disease. B, Focal hypoechoic mass. C, Diffuse hepatic involvement, with both hyperechoic and hypoechoic areas. D, Focal, irregular hyperechoic mass. E, Focal mass with hypoechoic and anechoic areas. F, Large hyperechoic mass with focal hypoechoic nodular areas. Continued 376 Small Animal Diagnostic Ultrasound LIVER G H LIVER I1 I2 LIVER J K Fig. 9.21, cont’d G, Histiocytic sarcoma. Focal hypoechoic masses. H, Lymphosarcoma. Multiple focal hypoechoic nodules. I, Ultrasound image (I1) and gross specimen (I2) of cholangiocarcinoma in a cat. Focal,