Portosystemic shunts: Ultrasonography vs CT (Proceedings)

The normal vascular flow to the liver is dual with a larger portion coming portal vein (80%) than the hepatic arteries (remaining 20%). The efferent flow follows the hepatic veins to the caudal vena cava. Within the liver the hepatic portal veins are seen as anechoic, branching, tapering tubular structures. Hepatic arteries and hepatic ducts cannot be seen in normal patients. Color Doppler facilitates the identification of the hepatic veins and helps us verify which vessel we are examining.

Portosystemic shunts represent amongst the most common vascular anomalies. These represent an abnormal communication between the portal vasculature and either the caudal vena caval or azygos veins. Intra-hepatic shunts are more common in large breed dogs and extrahepatic shunts in small breed dogs. Microvascular dysplasia is most common in Yorkshire terriers. Multiple acquired shunts are only seen in acquired cirrhosis or developmental hepatic arterial dysplasia patients. The methods for evaluation portosystemic shunt include ultrasound, ultrasound combined with radiography or computed tomography. This lecture will discuss all three modalities with an emphasis towards more modern forms of imaging, concluding with CT angiography by multi-detector modern systems

Ultrasound is the mainstay of the characterization portosystemic shunts in general practice. However, it is amongst the most rigorous ultrasound examinations due to the wide variety of manifestations of these anomalies. While many tools are used to simplify the ultrasound search for portosystemic shunts, most will not cover shunts under all circumstances. For the evaluation of portosystemic shunts the measurements of the portal vein, immediately before entering the liver, compared to the diameter of the aorta at the level where the aortic crosses the diaphragm is useful. These measurements are typically made in transverse section and the normal portal vein is more than one half the diameter of the aorta. However, this measurement scheme does not include intrahepatic shunts, acquired hepatic shunts, or microvascular dysplasia. So, for the evaluation of solitary extra-hepatic shunts a strong level of suspicion the patient has a shunt would be reached if the portal vein was less than one half the diameter of the aorta. This is a reasonably easy measurement to obtain in most patients by utilizing a right-sided intercostal approach. 

Additional schemes for evaluating portosystemic shunts include evaluation of the caudal vena cava. The normal caudal vena cava has no turbulent flow. Any evidence of turbulent flow, especially between the level of the right renal vein and the diaphragm should provide strong suspicion of an abnormal vessel representing a portosystemic shunt. Aberrant vascular drainage into the vena cava results in turbulent flow. Often we can find the exact site of the beginning of the turbulent flow and follow the vessel draining in at that level. This may often be traced to the origin of the shunt, usually the portal vein, gastric vein or splenic vein branches. Alternatively, you can look for sites where flow is leaving the portal vein. Flow departing from the portal vein is always abnormal since all portal vein branches enter the portal vein. Additional features of portosystemic shunts, while not specific they are supportive, include small liver, hypovascular liver, large kidneys and bladder stones. These bladder stones are lucent on radiographs because they are of urate composition. Approximately 10% of all portosystemic shunt patients present with urinary tract signs associated with the urate stones.

While many studies indicate the sensitivity and specificity of ultrasound for the diagnosis and portosystemic shunts, most radiologists confirm the challenge of both the detection of shunts as well as verification that the dog is normal in a large percent of patients. Therefore, additional modality should be considered to improve the accuracy of portosystemic shunt detection and better characterization. This includes the combination of ultrasound and radiography. Historically angiography has been used as an intraoperative tool for the characterization of portosystemic shunts. However, this requires a surgical exploratory and catheterization of a jejunal vein. A newer test has been proposed whereby the splenic vein can be injected with iodinated radiographic contrast for subsequent radiographic angiogram. Typically these patients should be heavily sedated or anesthetized, placed in right lateral recumbency with the spleen in the upper most part of the abdomen available for imaging with ultrasound. Using ultrasound guidance, a 25gauge needle can be directed into a large splenic vein, subsequent aspiration on the attached syringe should indicate strong blood flow indicating localization of the needle tip within the large vein lumen and thereafter 3 to 5 ml (depending on patient weight) of contrast is injected. The radiograph is taken after initiation of the injection but before the end of the injection. This test has been proven to be very sensitive and specific for shunts at the level of cranial to the entrance of the splenic vein into the portal vein. This represents approximately 95% of all portosystemic shunts. The additional advantage is that it also is helpful for acquired and intrahepatic shunts. There is no imaging test for the diagnosis of microvascular dysplasia.

The final diagnostic test, and that preferred by the lecturer, is CT angiography. Historically CT angiography has been a rigorous CT procedure including at least three separate data collection points including survey CT, a test injection to calculate the timing of portal vein enhancement of the liver and the final CT angiogram. This procedure was established during the era of single detector CT and is no longer necessary with multi-detector CT systems. With newer systems (greater than 4 detectors), a single CT angiographic study is all that is necessary. This study is composed of three consecutive, rapid, limited abdominal scans after a hand injection of intravenous contrast in a peripheral vein. This study is referred to as a triple phase CT angiogram for portosystemic shunts. Usually the shunt is seen best in the second phase of this triple phase study. The rigor of this study is no longer the technical expertise of the imager, in this case a technician running the CT equipment, but rather the radiologists interpretation of the angiogram on the subsequent imaging set. A review using this technology demonstrated 100% accuracy for portosystemic shunts in patients in a clinical setting. These studies typically require approximately 10 minutes to perform, required no advanced radiologic expertise available during the scan, and open the opportunity for this advance modality in places where there is no boarded radiologist available. These techniques are easily taught to CT technicians and can readily be performed in an accurate and timely fashion in any practice setting.

In summary, the imaging of patients with suspected portosystemic shunt runs the gamut of ultrasound, radiology and computed tomography. Improved ultrasound equipment provides increased accuracy for the detection of shunts with ultrasound. The use of Doppler sonography greatly improves the characterization of the aberrant flow and possible turbulence associated with the shunt vessel. Measurement of the aorta compared to the portal vein often give strong suggestion of a portosystemic shunt, but has a large number of exceptions that limits utility of this tool. Ultrasound guidance of the injection of an iodinated contrast into a splenic vein with subsequent radiographic angiography is a simple, but technically rigorous test with a high degree of accuracy and better anatomic characterization of the portosystemic shunts for the surgeon. Finally, new advances in multi-detector CT demonstrate that this tool is the current superior test for the evaluation of portosystemic shunts. CT angiography is now a very simple, rapid and highly accurate test without the requirement of boarded radiologist available during the scan. This CT protocol is easily taught to anyone with a working knowledge of CT and images can thereafter be sent for evaluation of the CT angiogram.

References

Laitinen, MR, Matheson, JS, O'Brien, RT. Novel technique of multislice CT angiography protocol for diagnosis of portosystemic shunts in sedated dogs. Open Journal of Veterinary Medicine 3:115-120. 2013. DOI 10.4236/ojvm.2013.32019

M. A. D'Anjou, D. Pennick, L. Cornejo and P. Pibarot, Ultrasonographic Diagnosis of Portosystemic Shunting in Dogs and Cats, Veterinary Radiology & Ultrasound, Vol. 45, No. 5, 2004, pp. 424-437. doi:10.1111/j.1740-8261.2004.04076.x