Making the Most of Abdominal Ultrasound
John S. Mattoon, DVM, DACVR
Small Animal Track
2012 ISVMA Annual Conference Proceedings
Abdominal ultrasound is an amazing diagnostic tool, yet it is often misunderstood.
Expectations are high, sometimes unrealistic, sometimes hindering rather than aiding
patient care. Understanding the type of information that can be gleaned from an
abdominal ultrasound is crucial, as is understanding limitations.
Ultrasound examination does not lead to a histological, cytological, or microbacterial
diagnosis. While this is stating the obvious, one of the most common and frequent
mistakes I see is just that; reaching diagnoses beyond the realm of the modality.
This can happen with inexperience, as lack of self-confidence may be overcompensated
by overzealous interpretation. Perhaps worse is the experienced sonographer who
believes that experience has afforded them a histological probe!
Next, it must be recognized that ultrasound can detect alterations in echogenicity and
echotexture before biochemical parameters are altered, or before clinical signs are
manifested. Conversely, we may be presented with very ill patients, with biochemical
abnormalities yet our ultrasound examination may be entirely normal. I know we have all
experienced these phenomena. It is frustrating, but it is our responsibility to recognize
these limitations as inherent in the modality. And it is our responsibility to move ahead
and utilize abdominal ultrasound to its fullest capabilities.
Why are we dong the abdominal ultrasound examination?
The answer to this question is important. Are we examining the patient because of
illness referable to the abdomen? And if so, what organ system has your highest
suspicion? Are we using ultrasound as a screening procedure in face of non-abdominal
disease, such as a patient with a suspected brain lesion? Is this a geriatric screening
procedure? Has the patient been traumatized? In a busy workday, you may not have the
luxury of performing a one-hour examination. Knowing the purpose of the examination is
vital in helping you focus your thoughts.
Knowing the type of information you can reliably gather from an ultrasound examination
is important. Knowing what you can add to that information by collecting a cytological
sample is also important. Knowing the limitations of cytology and the added benefits of
histological assessment through a tissue core sample is of the essence. Biopsies come
with some risk, even with ultrasound guidance. Understanding the complimentary role of
other imaging modalities is key, either as a potentially better “first test”, or “where do we
go next” if the ultrasound examination (and FNA and biopsy) fail to provide the correct
answer?
Knowing why we are doing the examination, and having a valid reason for doing so truly
benefit the patient and the client, and help us as sonographers focus our attention
appropriately.
The patient with signs referable to the gastrointestinal tract; is this patient best served by
doing an abdominal ultrasound examination, or is it better to begin with abdominal
radiographs?
In a traumatized patient, we are assessing for evidence of hemorrhage (abdominal fluid),
as well as urinary system integrity. What do we do when we find a splenic mass or a
“mottled” liver in this patient?
In a geriatric patient, you will undoubtedly find abnormalities on the ultrasound
examination. Do you ignore the findings, perform FNA sample collection?
How do we proceed?
First, we must develop reliable scanning methods, both organ by organ, but also to
optimize our scanning parameters to allow the best assessment possible for any given
patient. The first is easy and comes quickly with time. The second is more difficult and
probably will require some expert guidance along the way.
Heavy reliance on quantification of organs and structures (i.e., measuring things) is
unreliable at best and produces potentially harmful data at worst. Be careful with how
you use this data. A millimeter or 2 may not be abnormal at all! Ask yourself how critical
you are when measuring an adrenal gland. Is that 8.2 mm caudal pole dimension
translated to hyperadrenocorticism?
Comments on ultrasound guided sampling of organs
Ultrasound is routinely used to safely guide needle aspirates of organs, masses, and
effusions. There are considerations, however. For example, how do we know that the
specific area we aspirated truly reflects the pathology? How do we know that we have
sampled the pathology? Can we be certain that the deep liver nodule you skillfully
guided a needle into really produced a representative sample, or could the needle be
contaminated by near field “normal liver”? When aspirating “normal appearing organs”
with known biochemical alterations, are we certain that we are sampling the portion of
the organ responsible for the elevations in serum biochemical parameters? Of course,
the answer is no to all of the above.
Sampling error: Did I hit the target? How do I know what the target is, especially with
diffuse disease?
Organ Specific Comments
Liver
Ultrasound is very good in identifying parenchymal alterations such as nodular disease.
Some diffuse diseases can be confidently recognized, such as when the liver is
hyperechoic to the spleen, and/or has a finer or courser echotexture. In these cases,
many of us routinely perform fine needle aspirates (FNA) for a cytological examination
and hopefully a true diagnosis. Avoid ultrasound diagnoses such as “fatty liver” or
hepatic lipidosis as these are cytological/histological diagnoses. Differential diagnoses,
yes; ultrasound diagnoses, no!
Assessment of the biliary system compliments ultrasound very well. We can confidently
ascertain the gallbladder, abnormal bile within the gallbladder, distended intrahepatic
and extrahepatic bile ducts, and localize calculi. However, biliary or intrahepatic gas is
probably easier to detect radiographically; differentiating between calculi and gas is
certainly easier.
Evaluation of the portal system is routinely preformed in patients suspected of having
portosystemic shunts. Many of these shunting lesions can be diagnosed with ultrasound,
but certainly not all. Judicious use of color Doppler is needed in many cases, as the
shunting vessels can be very small, beyond the resolving power of B-mode alone. If a
shunt is identified, the next question is to do your best to describe its location for the
surgeon, but this can be tricky and often we are not precisely correct (or worse!). If a
shunt cannot be identified, then the question is “did I miss it?” Or does the patient have
microvascular dysplasia? Secondary ultrasound findings that may indicate congenital
liver dysfunction include mineralization of the kidneys, usually in the form of a medullary
rim sign, though small discrete calculi may be seen. Calculi or “sand” may be present in
the urinary bladder. These are ammonium urate crystals.
Spleen
Ultrasound easily detects alterations in splenic parenchyma, particularly masses and
nodular disease. With higher resolution transducers, we are routinely seeing subtle
parenchymal alterations that we are unsure as to their significance, if any. Indeed, with
the newest high frequency transducers, it may be that we are actually seeing lymphoid
centers within the splenic parenchyma! Alternatively, we have experienced numerous
cases of completely normal spleens that when aspirated yield a diagnosis of mast cell
disease. Vascular abnormalities of the spleen are not uncommon, in particular splenic
vein thromboses can be reliably diagnosed, sometimes in clinically ill patients, in others
as an ‘incidental” finding.
Further investigation of other portal and systemic vessels is indicated, as is a search for
an underlying disease process (such as lymphoma, hyperadrenocorticism).
Urinary System
The urinary system is easily evaluated with ultrasound and has become the primary
imaging modality in many clinics. Ultrasound is more convenient and yields a more
detailed examination of the renal parenchyma and facilitates FNA and biopsy
procedures. Indeed, the advent of ultrasound has relegated the excretory urogram (EU)
in many hospitals to a test for ectopic ureters, urinary system integrity following trauma,
or as a crude estimation of comparative renal function.
As for other organs, diffuse renal disease is much more difficult to diagnose with
ultrasound than focal or multifocal disease. This is because not all diseases cause a
change in the sonographic appearance of an organ. There are many instances in which
renal failure is present, yet the ultrasound appearance of the kidneys is considered
normal. This is currently one of the greatest limitations of ultrasound. Thus, it is
important to remember that diffuse renal disease may be present without observed
changes on the ultrasound examination. Although sometimes frustrating, it is a fact that
ultrasound is much better at identifying focal or multifocal disease than diffuse pathology.
It must further be emphasized that when ultrasound abnormalities are identified, the
appearance is rarely specific for a particular disease process. Nonetheless, many forms
of diffuse renal pathology do show themselves on an ultrasound examination. Further,
the absence of observed ultrasound pathology in the face of renal failure can help the
veterinarian develop a list of reasonable differential diagnoses, by excluding certain
diseases, which do characteristically show ultrasound abnormalities.
Diffuse renal disease may cause increased cortical echogenicity with enhanced
corticomedullary distinction or result in decreased definition between the cortex and
medulla as a result of disease affecting both of these regions. Diseases diffusely
affecting the kidney include acute and chronic glomerulonephritis and interstitial
nephritis, bacterial infections (e.g., Leptospirosis), acute tubular necrosis from toxins
(e.g., ethylene glycol toxicosis), amyloidosis, endstage kidneys and nephrocalcinosis. In
the cat, lymphosarcoma, feline infectious peritonitis (FIP) and metastatic squamous cell
carcinoma have been reported to cause hyperechoic cortices, with maintained
corticomedullary definition. Reduced cortical echogenicity, or multifocal hypoechoic
nodules or masses have also been described with lymphosarcoma. The size of the
kidneys may be normal, enlarged, or small with diffuse renal disease. Acute nephritis,
FIP, lymphosarcoma generally cause renal enlargement. As previously mentioned,
determination of normal renal size for a particular patient is problematic. Therefore serial
examinations may be necessary to detect renal size changes in response to therapy.
Ethylene glycol toxicity (antifreeze) often will produce extremely hyperechoic cortices
and medullary tissue. Severe cases may produce complete acoustic shadowing. A rim
of hypoechogenicity has been described at the corticomedullary junction in some cases
of ethylene glycol toxicity. Concurrent peritoneal, retroperitoneal and subcapsular fluid
may be observed. Subtle to moderate renal enlargement is present.
A general increase in renal echogenicity (cortical and medullary), with loss of the
corticomedullary junction is noted in cases of acute and chronic inflammatory disease,
amyloidosis, some types of toxicity and endstage kidneys in dogs and cats. Endstage
kidneys are small, distorted, irregular, and may not resemble a kidney at all. Endstage
renal disease may be seen in older patients, or in young or even juvenile patients, a
result of congenital renal dysplasia. The appearance of the kidneys may be
asymmetrical. This is seen regularly in cats with renal failure. The opposite kidney may
appear normal and in fact hypertrophy in an effort to compensate for diminished renal
function.
The renal medullary rim sign has been described in a number of disease processes,
including hypercalcemic nephropathy (lymphosarcoma), ethylene glycol ingestion,
pyogranulomatous vasculitis (feline infectious peritonitis), acute tubular necrosis of
undetermined etiologies and chronic interstitial nephritis. It is often seen in dogs with
portosystemic shunts. It is recognized as a very echogenic rim parallel to the
corticomedullary junction and usually results from mineral deposits within the outer
medullary tubular lumens or tubular basement membranes. In the case of FIP,
mineralization is not seen histologically. It should also be noted that the medullary rim
sign has been described in normal cats caused by a band of mineral within the lumens
of the renal tubules. The medullary rim sign thus provides an ultrasonographic finding
indicating primary renal disease in some, but not all patients. This rim sign is also
frequently seen in dogs and cats without clinical or biochemical signs of renal disease.
Thus, interpretation of this sonographic finding must be correlated with other pertinent
data.
It has also been shown that the degree of cortical echogenicity is positively correlated to
the amount of fat vacuoles in the cortical tubular epithelium of cat kidneys. Kidneys with
a plentiful amount of fat vacuoles demonstrated a great difference between the
hyperechoic cortical tissue and the hypoechoic medulla. The cortical echogenicity
becomes similar to the highly echogenic renal sinus. Cats without a large number of
cortical fat vacuoles had less echogenic cortices. Thus, the definition between the
cortex and the medulla is less apparent, as the two regions of the kidney are more
similar in echogenicity.
Ultrasound is at its best when there is focal disease; focal renal lesions are readily
imaged using ultrasound. Still, while a few types of focal disease can be definitively
diagnosed using ultrasound, it must be remembered that even most focal renal disease
will require cytology or histology for a final diagnosis. Renal cysts are the exception
Pancreas
Diagnosing pancreatic disease can be challenging in many instances because blood
work may not be specific. Clearly, lipase and amylase have essentially no diagnostic
value. PLI seems to be the most sensitive blood test available. Radiography may help
define the presence of pancreatitis, especially in dogs, but it is not a sensitive or specific
test. Ultrasound is a non-invasive method that can often detect abnormal pancreatic
tissue, or in many cases show regional mesenteric pathology. Progression and healing
can be followed as well. Feline pancreatic disease is more problematic, as the changes
seen are often subtle, and in many cases ultrasound cannot reliably detect pathology
seen with histology.
Gastrointestinal Tract
Ultrasound evaluation of the GIT can be very rewarding, and can offer information
unavailable with radiography and CT, such as bowel wall layering. However, there have
been numerous instances of readily apparent GIT disease on radiographs that were
completely overlooked or misinterpreted during an ultrasound examination. Foreign body
obstructions, in particular partial obstruction are often overlooked, as is free abdominal
air from bowel rupture. As a rule of thumb, I suggest abdominal radiographs be made for
any patient suspected of having gastrointestinal disease. Ultrasound alone is often nondiagnostic and misleading, sometimes to the serious detriment of the patient.
Lymph nodes
Ultrasound is invaluable in assessing abdominal lymph nodes. FNA of enlarged lymph
nodes is commonly performed. When assessing lymph nodes, it is important to
distinguish mesenteric lymph nodes (which drain the GIT) from the systemic lymph
nodes located along the lumbar spine (such as the medial iliac lymph nodes). The latter
do not drain the GIT so enlargement should prompt investigation of diseases of the
urogenital system and soft tissues of the perineal region.
Fluid
Fluid is easily detected using ultrasound and can be collected for analysis. It must be
emphasized that the character of sonographically detected fluid cannot reliably predict
clinical pathological analysis!
References
1. Adams WH, Toal RL, Breider MA: Ultrasonographic findings in dogs and cats with
oxalate nephrosis
2. Biller DS, Bradley GA, Partington BP: Renal medullary rim sign: Ultrasonographic
evidence of renal disease. Vet Radiol Ultrasound 1992;33:286-290.
3. Forrest LJ, O’Brien RT, Tremelling MS, et al: Sonographic renal findings in 20 dogs
with leptospirosis. Vet Radiol Ultrasound 1998;39:337-340.
4. Yeager AE, Anderson WI: Study of association between histologic features and
echogenicity of architecturally normal cat kidneys. Am J Vet Res 1989;50:860-863.
5. Wood AK, McCarthy PH: Ultrasonographic-anatomic correlation and an imaging
protocol of the normal canine kidney. Am J Vet Res 1990;51:103-108.