Cintia R. Oliveira VetsChoice Radiology, Madison, WI, USA In small animals, the spleen is a mobile organ, with only its proximal extremity or head being fixed to the stomach by the gastrosplenic ligament, in the left craniodorsal aspect of the abdomen. The body and distal extremity, or tail of the spleen, are not fixed and therefore variable in position. On lateral radiographic views in dogs, the tail of the spleen is seen as a triangular soft tissue opacity caudal and slightly ventral to the liver and pylorus (Figure 24.1). Occasionally, the head of the spleen may be seen in dogs or cats on lateral views in the dorsal and cranial abdomen just caudal to the stomach and dorsal to the right kidney margins (Figure 24.1). On ventrodorsal views, the body of the spleen is seen as a triangular soft tissue opacity in the cranial and left aspects of the abdomen, caudolateral to the stomach and craniolateral to the left kidney in both dogs and cats (Figures 24.1 and 24.2). If the spleen is positioned alongside the left body wall, it may not be visible on lateral radiographs but may be visible entirely on the ventrodorsal view. The feline spleen differs from the canine spleen in that it is smaller and thinner, and less variable in size and position (Figure 24.2). The body and tail of the feline spleen are not usually seen on lateral views in normal cats. The head may be visible on lateral views as a small triangular soft tissue opacity caudal and dorsal to the stomach. Like the dog, on the ventrodorsal view the head of the spleen can be seen in the left cranial abdomen caudolateral to the stomach and craniolateral to the left kidney. Frequently, the entire feline spleen is visible on ventrodorsal views since its tail extends caudally along the left lateral abdominal wall (Figure 24.2). The normal thickness of the feline spleen is 1 cm as measured on radiographs and ultrasound [1]. On ultrasound (US), the canine spleen is identified on the left side caudal and lateral to the stomach. Since its body and distal extremity are movable, they often extend into the caudal abdomen and across midline to the right side and the spleen often appears folded on itself (Figures 24.3 and 24.4) [1]. The feline spleen may be difficult to find ultrasonographically in some cats because of its small size (Figure 24.5). Intercostal windows may be necessary with the transducer placed in a more dorsal and cranial position than expected [1]. As opposed to dogs, the feline spleen rarely folds on itself unless enlarged. Sonographically, when enlarged, the spleen may extend more caudally, being visible next to the urinary bladder, and the splenic margins become rounded or blunted, as opposed to the normal sharp appearance [1, 2]. Ectopic spleen has been reported in dogs and cats. It has been described ultrasonographically as round to triangular homogeneous structures isoechoic to spleen in the perisplenic region and with computed tomography (CT) as multiple heterogeneous intraabdominal and intrahepatic masses resembling hepatic tumor with abdominal metastasis [3, 4]. Splenic thickness greater than 1 cm in cats is suggestive of splenomegaly. A recent study found a weak correlation between radiographic and ultrasonographic measurement of splenic size in cats. In that study, the feline spleen measured ultrasonographically had a mean of 8.0 ± 1.6 mm, similar to previous reports [5]. In dogs, there is no absolute normal size for the spleen. Suspect splenomegaly on ultrasound can be confirmed with radiographs or palpation, but radiographic assessment of splenic size is subjective because the spleen varies in size in normal dogs and cats. Radiography and ultrasound are still the primary imaging methods used to evaluate the spleen, in part due to their noninvasive nature and the ease with which evaluation and sampling can be carried out with US. Contrast‐enhanced US can be used for further characterization of focal and diffuse splenic lesions although the accuracy of contrast‐enhanced US in differentiating between benign and malignant splenic lesions is conflicting [36–10]. A study evaluating the assessment of feeding vessels with contrast‐enhanced US found that echogenicity and persistent hypoperfusion cannot be used alone to distinguish between malignant and benign focal splenic lesions but the presence of tortuous and persistent feeding vessels may be helpful in distinguishing malignancy since no benign lesions in the study had such a characteristic [11]. Contrast enhancement parameters for normal spleen have been published using different contrast agents [8]. The use of CT to evaluate the spleen is usually reserved for specific disease processes such as trauma, vascular disorders, staging of neoplasia or in acute abdomen cases (Figure 24.3). However, CT has been used with increasing frequency to evaluate splenic disease and its use is paramount for complete detection and evaluation of vascular disorders. A recent study described the presence of collateral circulation using CT angiography in a dog with splenic vein obstruction and presumed splenic vein hypertension [12]. Magnetic resonance imaging (MRI) is not used routinely to evaluate the spleen since US and CT are faster and allow for immediate sampling of lesions. Studies describing the normal MRI appearance of the canine and feline spleen are available as well as studies evaluating splenic disease (Figure 24.3). Elastography (ultrasound technique for determining stiffness of the spleen) of the normal spleen has also been reported but currently its use is still reserved for research. The imaging features of the normal spleen in four imaging modalities are listed in Table 24.1. A recent study evaluated the US appearance of the spleen in nine normal puppies from 4 to 60 weeks using a high‐frequency linear transducer and found that the splenic parenchymal patterns changed with age from a granular appearance at 4 weeks of age to a reticulonodular pattern characterized by well‐defined hypoechoic nodules which was most marked at 28–36 weeks of age, which then gradually changed into a more homogeneous granular pattern at 60 weeks [13]. Care must be taken during ultrasound examination of very young dogs since some of these patterns resemble disease, including neoplasia. Splenic neoplasia has a high prevalence in dogs, with a high frequency of malignant tumors, and in the presence of hemoabdomen, a splenic mass is more likely to be neoplasia [14, 15]. Hemangiosarcoma is the most prevalent malignant neoplasia of the spleen in dogs (Figures 24.6–24.9) [15]. Other malignant tumors of the spleen in dogs include sarcoma, lymphoma, carcinoma, mast cell tumor, and metastasis (Figures 24.10–24.18) [16–19]. In a study of 455 cats with splenic disease, primary and metastatic neoplasia accounted for 37% of cases [20]. Most cats had mastocytoma followed by lymphoma, myeloproliferative disease, and hemangiosarcoma. A much smaller percentage of benign lesions was found, including accessory spleen (4%), hyperplastic nodules and/or hematomas (4%), splenitis (2%), and thromboembolism with infarction (1%) [20]. Splenic nodules are a common nonspecific finding in dogs for which differentials include nodular hyperplasia, extramedullary hematopoiesis (EMH), hematoma, or neoplastic or infectious lesions [21]. These nodules can be of variable size and echogenicity and no ultrasound criteria can be confidently applied to differentiate benign from malignant nodules. Nodular hyperplasia and EMH are a common finding in older dogs and they should not be regarded as metastatic lesions in the presence of a primary neoplasia elsewhere without sampling [20, 22]. Splenic masses are usually vascular, and bleeding often occurs if the masses rupture. Bleeding of splenic hemangiosarcoma is one of the most common causes of canine hemoabdomen [23]. In some dogs with hemoabdomen, a “sentinel clot sign” may be detected in proximity to the bleeding source. The sentinel clot sign is the highest attenuation hematoma adjacent to a bleeding organ or mass [24]. Active bleeding can be detected on postcontrast CT images by the presence of a serpiginous or amorphous, highly attenuating area originating from an organ parenchyma which would be consistent with extravasation of contrast medium into the peritoneal space [24]. With radiography, neoplasia can be seen as a distinct mass or changes to the splenic contour/margins (Figure 24.8). A variety of changes can be seen with US and CT and there are no specific findings to differentiate between benign or malignant disease or between specific types of neoplasia [17, 21, 25]. TABLE 24.1 Imaging characteristics of normal canine and feline spleen. CT, computed tomography; MRI, magnetic resonance imaging; STIR, short tau inversion recovery; T1W, T1 weighted; T2W, T2 weighted; US, ultrasound. On US, neoplasia commonly manifests as poorly marginated nodules or masses that are anechoic or hypoechoic, with or without target‐like lesions, or are complex in appearance and may distort the splenic contour [17, 21,26–28]. With CT, a multitude of characteristics have been reported with overlap between benign and malignant masses [25, 29, 30]. Malignant masses may be precontrast hypoattenuating and enhance less than the splenic parenchyma on postcontrast images, or they can have mild or strong heterogeneous enhancement [25, 30]. Less commonly, neoplastic lesions may be hyperechoic or hyperattenuating on US and CT respectively [25, 30]. Hemangiosarcomas can appear as a mass of variable size, may be solid or cavitated, isoechoic, hypo‐ or hyperechoic compared to the splenic parenchyma, or have mixed echogenicity [31]. Ultrasonographically, they are indistinguishable from hematomas, abscesses, EMH or benign neoplasia such as hemangioma [26,31–33]. With CT, they have been found to have marked generalized enhancement in early‐phase images that persisted in delayed images, or homogeneous poor enhancement on all phases. In one study, 77% of the hemangiosarcomas and all hematomas had contrast accumulation compatible with active hemorrhage [34, 35]. Other malignant neoplasia that have similar US appearance include other sarcomas such as histiocytic sarcoma, lymphoma, and metastatic disease [18, 28]. In a CT study comparing malignant and nonmalignant splenic masses in dogs, malignant masses had significant lower attenuation values (Hounsfield units, HU) on both pre‐ and postcontrast images and the authors recommend a cutoff value of less than 55 HU as being suggestive of a malignant mass [34]. A study comparing CT with histopathologic diagnosis of dogs with hepatic or splenic masses found a large overlap between pre‐ and postcontrast CT features of malignant and nonmalignant masses and concluded that dual‐phase CT provides limited specific diagnostic information in these patients [35]. A more recent study found that precontrast lesion attenuation was significantly different between splenic malignant and benign tumors, with malignant tumors having a mean and standard deviation lesion attenuation of 40.3 ± 5.9 HU, and benign tumors 52.8 ± 6.8 HU [36]. Target‐like lesions seen on US, nodules with a hyperechoic center surrounded by a hypoechoic rim resembling a bull’s eye target, have been associated with malignant lesions in both humans and dogs, although these lesions can also be seen with a variety of benign diseases (Figures 24.17, 24.18, and 24.32) [27]. Sampling of any splenic lesion is necessary for a final diagnosis. Nodular hyperplasia and hematoma are among the most common splenic lesions in dogs [1]. These lesions are a particularly common finding in ultrasound examinations of older dogs and should not be regarded as malignant lesions without sampling. In a study evaluating 1480 dogs with splenic disease, nodular hyperplasia and hematoma made up the majority of the cases [20]. In a study evaluating 105 dogs presented with incidentally found nonruptured splenic masses, 70.5% had a benign lesion, of which 50% had nodular hyperplasia. In a similar study in cats, however, these lesions comprised only 4% of the cases [20]. This suggests that splenic nodules or masses in cats are less likely to be benign compared to dogs and therefore more aggressive sampling may be indicated. With US, these benign lesions have a variable appearance and therefore cannot be distinguished from malignant nodules (Figures 24.19–24.22). With CT, these lesions tend to be hyperattenuating on precontrast images and remain so on postcontrast images or be isoattenuating on precontrast images and hyperattenuating on postcontrast images (Figures 24.23 and 24.24) [25, 34]. Another differential for such lesions is EMH.
CHAPTER 24
Spleen
Introduction
Focal Splenic Changes
Primary and Secondary Neoplasia
Radiographs
US
CT
MRI
Dog
Soft tissue opaque
Finely textured, hyperechoic compared to liver and renal cortex
Splenomegaly is common since patients usually under general anesthesia
Precontrast: homogeneous soft tissue attenuation
Postcontrast:
Arterial: heterogeneous enhancement
Venous: homogeneous enhancement
Delayed: homogeneous enhancement
T2W: hyperintense compared to liver
T1W precontrast: hypointense compared to liver
T1W postcontrast: homogeneously contrast enhancing, slightly hypointense compared to liver
STIR: hypointense compared to liver
Cat
Soft tissue opaque
Finely textured, hyperechoic compared to renal cortex
Splenomegaly is common since patients usually under general anesthesia
Postcontrast:
Arterial: patchy heterogeneous enhancement
Venous: patchy heterogeneous or homogeneous enhancement
Delayed: homogeneous enhancement
T2W: hyperintense compared to liver
T1W precontrast: hypointense compared to liver
T1W postcontrast: markedly enhancing, first 60 seconds heterogeneous pattern of high and low signal intensity, after 60 seconds homogeneous enhancement
Nodular Hyperplasia