Portosystemic shunts (PSS) are abnormal vessels allowing communication between the splanchnic and systemic circulations. They are either congenital or acquired, single or multiple, extrahepatic or intrahepatic.
Clinical signs associated with HE can be broadly separated in neurological, gastrointestinal and urinary. Neurological signs include depression, listlessness, ataxia, pacing, circling, head pressing, cortical blindness, seizures and coma. They have been classically described as correlated with meals, but it has been shown to only be the case in 25% to 50% of patients. Gastrointestinal signs occur in about 30% of canine cases and include anorexia, vomiting and gastrointestinal bleeding. In cats, ptyalism is a very common clinical sign, present in 75% of cases. Urinary signs include dysuria, stranguria, pollakiuria and haematuria and are associated with ammonium urate crystalluria.
Clinical pathology changes most commonly encountered in animals with PSS include mild to moderate microcytic normochromic nonregenerative anaemia, elevated alkaline phosphatase (ALP) and alanine aminotransferase (ALT) activities, hypoalbuminaemia, decreased blood urea nitrogen (BUN), hypocholesterolaemia and hypoglycaemia. Coagulation times are almost consistently prolonged in affected dogs, these abnormalities are infrequently clinically significant. Urinalysis abnormalities include hyposthenuria, ammonium biurate crystalluria and signs of secondary urinary tract infection. The most common laboratory tests used for diagnosis of liver dysfunction include bile acids stimulation test (BAST) and plasma ammonia levels.
Survey radiographs are of limited value in the diagnostic of PSS. Portovenography, either jejunal or transplenic, can be performed pre- and/or intraoperatively and is very sensitive and specific for PSS detection. Ultrasound is still widely used for assessment of PSS, with 74% to 95% sensitivity and 67% to 100% specificity. Scintigraphy is quite sensitive for detection of PSS, but cannot discriminate between EHPSS and IHPSS, nor can it give any information of the number of shunts. Computed tomographic angiography has become the gold-standard for characterisation of PSS,2 as it is non-invasive, more accurate than portographies3 and ultrasound,4 and allows precise characterisation of the shunt morphology5.
A medical treatment can be used to decrease the clinical signs and optimise patients prior to surgical correction of operable CPSS, or for long-term management of multiple acquired or inoperable CPSS.
It aims at decreasing the clinical signs associated with HE by reducing the amount of neurotoxic by-products of protein metabolism. It is therefore based on the administration of a low-protein diet, antibiotics to decrease the number of ammonia-producing bacteria in the digestive tract, lactulose to decrease the production and absorption of ammonia, gastroprotectants and antacids to prevent or treat gastrointestinal ulcers and bleeding. Seizure control is sought through administration of phenobarbital, potassium bromide, gabapentin, or propofol. Levetiracetam is another anticonvulsant commonly used to prevent or control seizures in PSS patients,6 which the author routinely prescribes to patients for a few days before surgery at 10–20 mg/kg TID.
Identification of the Shunt
Most PSS are found to terminate on the caudal vena cava at the level of the epiploic foramen, which should therefore be inspected first. Any blood vessel found to enter the caudal vena cava between the renal and phrenico-abdominal veins and the liver should be considered as potentially being a PSS. Blood flow turbulences are often visible in the caudal vena cava or shunt vessel, which further supports the shunt identification. If no abnormal vessel is found through the epiploic foramen, the omental bursa must be opened by tearing the superficial leaf of the greater omentum. This allows inspection of the tributaries of the portal vein. Most often, shunting vessels originate from the gastrosplenic vein in dogs and left gastric vein in cats. Portoazygos shunts are found as abnormal vessels penetrating the diaphragmatic crura or aortic hiatus. Occasionally, shunts cross through the diaphragm at the oesophageal hiatus. If no abnormal vessel is found, the presence of an IHPSS should be evaluated. This is done by identifying the left hepatic veins cranial to the liver for the left lobe, or branches of the portal vein caudal to the liver for the right and centre lobes. Temporary occlusion of any of these veins communicating with the shunt results in immediate signs of portal hypertension. Acute or gradual occlusion of this vessel is then carried out. In addition, any dilation or abnormally soft, fluid-filled consistency of a liver lobe should raise the suspicion of the presence of an IHPSS passing through it.
Two types of occlusions are used for PSS attenuation: acute and gradual.
Acute surgical attenuation consists of completely or partially ligating the abnormal vessel. In many cases, complete ligation is not possible without life-threatening portal hypertension. Partial occlusion may result in persistent shunting and necessity for a second surgery to ligate the shunt further. Acute attenuation always carries a risk of postoperative portal hypertension, which may cause death or necessitate reintervention to remove the ligature.
Gradual attenuation of a PSS aims at giving time for the intrahepatic portal vasculature to develop as the PSS is progressively attenuated, thereby decreasing the risk of portal hypertension, and can be achieved with ameroid constrictors, cellophane bands or hydraulic occluders.
The author currently prefers cellophane banding over other techniques. Cellophane bands placed around shunt vessel induce inflammation around the vessel, leading to its occlusion. The rate of occlusion is therefore dependant on the amount of inflammation induced. They are reported to induce complete occlusion of the PSS within 6 months. Cellophane can be used for shunt vessels of any size, as well as for EHPSS and IHPSS. In early reports, it was advocated to attenuate shunts intraoperatively to less than 3 mm of diameter for complete occlusion to occur, but more recent evidence suggests that no attenuation is necessary to achieve long-term complete occlusion, regardless of the shunt size,7 which avoids the risk of inducing portal hypertension. The cellophane band is therefore just applied around the shunt vessel and secured with 3 or 4 vascular clips.
If any attenuation has been induced surgically, the patient is monitored for signs of portal hypertension, which may warrant reintervention to relieve the shunt occlusion.
Postoperative hypothermia and hypoglycaemia are frequent, and must be actively monitored and promptly addressed. Postoperative seizures occur in 3% to 7% of dogs and 8% to 22% of cats after PSS attenuation, typically up to 4 days after surgery. Treatments include midazolam, propofol, barbiturate or levetiracetam administration.
Once discharged from the hospital, animals are maintained on a low-protein diet, antibiotics, lactulose and gastrointestinal protectants. Bile acids levels are monitored 1, 3 and 6 months after surgery. In the absence of clinical signs of HE, antibiotics are stopped 1 month after surgery and lactulose administration is discontinued a few weeks later. Approximately 3 months after surgery, if tolerated, the diet is progressively returned to normal.
Both short- and long-term, dogs with CPSS treated surgically have better survival rates and less persistent clinical signs than those treated medically.8,9 In dogs, reported mortality rates range from 2% to 32% after EHPSS surgical attenuation and from 0% to 27% after IHPSS surgical attenuation. Good to excellent outcomes in surviving dogs are reported in 84% to 94% for EHPSS and 50% to 100% for IHPSS.
In cats, perioperative mortality ranges from 0% to 23%. Good to excellent long-term outcome is reported in 33% to 80% of surviving cats.
1. Gerritzen-Bruning MJ, van den lngh TS, Rothuizen J. Diagnostic value of fasting plasma ammonia and bile acid concentrations in the identification of portosystemic shunting in dogs. Journal of Veterinary Internal Medicine. 2006;20:13–19.
2. Zwingenberger AL, Schwarz T, Saunders HM. Helical computed tomographic angiography of canine portosystemic shunts. Veterinary Radiology & Ultrasound. 2005;46:27–32.
3. Parry AT, White RN. Comparison of computed tomographic angiography and intraoperative mesenteric portovenography for extrahepatic portosystemic shunts. Journal of Small Animal Practice. 2017;58:49–55.
4. Kim SE, Giglio RF. Reese DJ, Reese SL, Bacon NJ, Ellison GW Comparison of computed tomographic angiography and ultrasonography for the detection and characterization of portosystemic shunts in dogs. Veterinary Radiology & Ultrasound. 2013;54:569–574.
5. Nelson NC, Nelson LL. Anatomy of extrahepatic portosystemic shunts in dogs as determined by computed tomography angiography. Veterinary Radiology & Ultrasound. 2011;52:498–506.
6. Fryer KJ, Levine JM, Peycke LE, Thompson JA, Cohen ND. Incidence of postoperative seizures with and without levetiracetam pretreatment in dogs undergoing portosystemic shunt attenuation. Journal of Veterinary Internal Medicine. 2011;25:1379–1384.
7. Frankel D, Seim H, MacPhail C, Monnet E. Evaluation of cellophane banding with and without intraoperative attenuation for treatment of congenital extrahepatic portosystemic shunts in dogs. Journal of the American Veterinary Medical Association. 2006;228:1355–1360.
8. Greenhalgh SN, Reeve JA, Johnstone T, Goodfellow MR, Dunning MD, O’Neill EJ, et al. Long-term survival and quality of life in dogs with clinical signs associated with a congenital portosystemic shunt after surgical or medical treatment. Journal of the American Veterinary Medical Association. 2014;245:527–533.
9. Greenhalgh SN, Dunning MD, McKinley TJ, Goodfellow MR, Kelman KR, Freitag T, et al. Comparison of survival after surgical or medical treatment in dogs with a congenital portosystemic shunt. Journal of the American Veterinary Medical Association. 2010;236:1215–1220.