There are multiple factors that will influence sensitivity and specificity for the diagnosis of intrahepatic and extrahepatic anomalous vessels in dogs. The 3 most common factors are the availability of equipment, the operator and the patient/owner.

Radiography is one of the most commonly used imaging modalities for the screening of animals with suspected portosystemic shunt. The sensitivity and specificity are low. We may see microhepatia and enlargement of the kidneys. Occasionally, mildly mineralized calculi in the urinary system is visualized. Contrast radiography such as intraoperative cranial mesenteric venography will be able to outline the anomalous vessels, but this is an invasive procedure. Cranial mesenteric angiography which needs a fluoroscopy machine is another method of choice. Again, this is also an invasive procedure. The advantage of these two procedures is the ability to visualize small intrahepatic portal veins which otherwise may not be seen on computed tomography (CT).

Patient preparation for ultrasound is very important in the search for both intrahepatic and extrahepatic anomalous vessels. Most of the patients with anomalous vessels have microhepatia, thus it is not easy to visualize the liver if the patient has excessive movement and panting. Sedating or putting the animal under GA will help to increase the sensitivity of detecting any anomalous vessel. If a patient has a postprandial stomach, this may reduce the acoustic window and interfering with detection of the normal and abnormal blood vessels. There is variation in the type of the shunt vessels, which includes different sizes and lengths of the anomalous vessel. It is easier to detect a large diameter anomalous vessel, thus the sensitivity and specificity are higher. It is also easier to trace a short anomalous vessel, and to be able to recognize the origin and insertion of the anomalous vessel. A small diameter, long and tortuous anomalous vessel could be detected, but most of the time it is difficult to trace the entire vessel. The operator needs to have a very strong knowledge of the normal anatomy of the blood vessels, and also be familiar with the various type of anomalous vessels. The last factor is the ultrasound machine used. As a general rule we need a good ultrasound machine with better resolution and good colour Doppler sensitivity to detect any anomalous vessels. This is especially true when the blood vessel(s) is small. In the last decade, the availability of better resolution and colour Doppler ultrasound machines and literatures of various anomalous vessels has promoted the use of ultrasound in the investigation of intrahepatic and extrahepatic anomalous vessels.

The sensitivity and specificity of ultrasound detection of intrahepatic shunts has been reported to be 100% and 100% while the sensitivity and specificity of detection of extrahepatic shunts has been reported to be 90% and 97% respectively.

There are 2 steps in ultrasonographic examination of the anomalous vessel, the screening and the confirmation. It has been reported that there is usually a change in size of the blood vessels (portal vein and caudal vena cava) with occurrence of extrahepatic shunt. Thus, a comparison of portal vein and caudal vena cava (PV/CVC), and also portal vein and aorta (PV/Ao) ratio has been published. A PV/Ao ratio of 0.7 to 1.25 is considered normal. Thus, dogs and cats with PV/Ao ratio of ≤0.65 were found to have either an extrahepatic portocaval shunt or idiopathic, noncirrhotic portal hypertension. Dogs and cats with a PV/Ao of ≥0.80 may have other types of shunts or be normal. It is good to use the PV/Ao ratio as a screening procedure before looking for an anomalous vessel. When a patient with a PV/Ao ratio of <0.65 and has clinical suspicion of an anomalous vessel, the portal vein and caudal vena should be examined carefully. If no anomalous vessel is identified, then abdominal contrast CT is recommended to rule out any small anomalous vessel. If an anomalous vessel is identified ultrasonographically, contrast CT may not be needed.

The anomalous vessel should be classified as congenital or acquired, intrahepatic or extrahepatic, and singular or multiple. Most anomalous vessels are congenital, and singular. Large breed dogs tend to have intrahepatic while smaller breed dogs to have extrahepatic anomalous vessels. In patients with acquired anomalous vessels, this normally occurs secondary to severe liver disease, hepatic AV malformations, portal vein thrombosis or hepatic vein outflow obstruction with portal hypertension.

There are three types of anomalous vessels in intrahepatic shunts: the left divisional, central divisional and right divisional shunt. The central divisional anomalous vessel is the most difficult to trace due to its short length. Both left and right divisional anomalous vessels are wide and tortuous.

There are many types of extrahepatic anomalous shunt vessels. The most commonly seen are splenoazygus, splenocaval and splenophrenic. Less commonly seen anomalous shunts are the splenocaval, right gastric-caval, double right gastric-caval and double right gastric-azygus. The presence of extra blood vessel(s) adjacent to the aorta makes splenoazygus shunts readily recognized with both intercostal and subcostal approaches. However, due to the long and tortuous nature of the anomalous vessel, it is not easy to trace the entire vessel back to the splenic vein. An anomalous splenocaval shunt vessel is normally short and could be traced at the region of the portal hepatis. Due to the cranial position of the anomalous splenophrenic shunt vessel adjacent to the diaphragm, this type of shunt vessel is more challenging to visualize due to movement of the diaphragm secondary to respiration.

An acquired shunt is normally secondary to portal hypertension, and multiple shunt vessels or varices are present. Most of the time, these varices are short, small in diameter and tortuous. They are commonly located cranial to the left kidney or at the region of the splenic vein. Esophageal varices have been reported. Whenever there is suspicion or confirmation of an acquired shunt, measure of portal blood flow velocity should be performed. Normal portal blood flow velocity is between 10–25 cm/sec. Any patients with portal blood flow velocity of < 10 cm/sec has a higher probability of developing acquired extrahepatic shunts.

Computed tomography has become the imaging modality of choice for most clinicians nowadays. This is because it is not invasive and the detail anatomy of the anomalous blood vessels can be outlined with post-procession 3D reconstruction. The information of diameter, length and the actual insertion of the shunt vessel could be acquired from the CT study. Computed tomography normally is fast (less than 10 minutes), and with 64 slice machines, occasionally this study could be performed under heavy sedation. The only disadvantage of this is the intrahepatic portal veins may not be identified.

The ultimate choice of modalities/technique used in the investigation of anomalous vessels is again depending on the patient preparation, availability of equipment and also operator experience and preference. Recently, the clinician/surgeon preference has become a deciding factor on the choice of the techniques. Some surgeons want to know the distance between the portal hepatis and the anomalous vessel, the size and length of the anomalous vessel and also the exact connection point of the anomalous vessel for surgical planning. A 3D reconstruction of the anomalous vessel is always requested by surgeons prior to surgical correction.