Introduction

Primary hypoplasia of the portal vein (PHPV) is a congenital disorder recognized primarily in dogs.¹ Severity of this condition is contingent upon the degree of hypoplasia of the portal vein, the presence of acquired portosystemic collateral circulation, and the extent of hepatocellular function lost. Mild cases of PHPV do not result in portal hypertension (PH). Portal hypertension, which is caused by the increased resistance of portal blood flow to the liver, eventually leads to the development of multiple acquired portosystemic shunts (MAPSS).^1,2

In cases of severe hypoplasia, dogs present with neurological dysfunction, polyuria, polydipsia, intermittent gastrointestinal signs, weight loss, small stature, and ascites.² The current veterinary literature shows that severely affected dogs have increased liver enzymes, serum bile acids (SBA), and plasma ammonia concentrations.^2,3 While various imaging modalities may document the presence of anomalous vasculature, histopathology of the liver is ultimately required, in part, to obtain a definitive diagnosis of PHPV.^1,3

The purpose of this case report is to describe severe PHPV in a young puppy that, despite exhibiting classical manifestations of this illness, had normal liver enzyme and paired SBA concentrations at the time of diagnosis.

Case Report

A 15 wk old male intact golden retriever puppy weighing 12.1 kg was presented for evaluation of abdominal distension of 3 to 4 days duration. Other notable historical changes included dull mentation, poor appetite, stargazing, and failure to grow.

Prereferral diagnostic evaluation included a complete blood count, serum chemistry panel with electrolytes, urinalysis, fecal analysis, fasting and 2 hr postprandial SBA concentrations, and abdominal effusion evaluation. The complete blood count was normal for a 15 wk old puppy.⁴ The only abnormality on the chemistry profile was panhypoproteinemia (3.5 g/dL; reference range, 4.5–7.3 g/dL), as characterized by mild hypoalbuminemia (2.2 g/dL; reference range, 2.6–3.7 g/dL) and hypoglobulinemia (1.3 g/dL; reference range, 2.2–3.5 g/dL). The urine was concentrated with a urine specific gravity of 1.052 and an inactive sediment. No ammonium biurate crystals were observed. Fecal flotation was negative for parasites. For measurement of SBA, the standard described technique was used.⁵ The patient was fasted for 12 hr, after which a sample of serum was obtained. He was fed eight tablespoons of a canine commercial diet^a, and a postprandial serum sample was obtained 2 hr later. The fasting SBA concentration was normal (11.1 μmol/L; reference range, <15.0 μmol/L) as was the 2 hr postprandial SBA concentration (12.2 μmol/L; reference range <20.0 μmol/L). The abdominal effusion was characterized as a pure transudate (total protein 2.0 g/dL). At the time of referral, the primary differential for ascites was congenital heart disease.

Upon presentation, physical examination showed a body condition score of 3/9 with diffuse muscle wasting, normal cardiopulmonary auscultation, marked abdominal distention with an obvious fluid wave, and normal rectal examination. There were no neurological deficits noted except dull mentation.

An abdominal ultrasound was done by a radiologist, which showed normal hepatic parenchyma, no visible arteriovenous fistula, and a moderate volume of ascites. Extensive interrogation of the portal vein, caudal vena cava, and aorta revealed no overt abnormalities, although mesenteric vasculature appeared prominent. No anomalous vessel(s) were identified. Measurement of portal vein velocity and portal vein-to-aortic ratio was not performed, as it is not considered standard at this institution. No turbulent blood flow, masses, or thrombi were detected in the portal vein. The gastrointestinal tract appeared normal with no evidence of edema. Thoracic radiographs were unremarkable, and an echocardiogram, performed by a cardiologist, was also normal. A fasting plasma ammonia^b concentration was elevated (113 μmol/L; reference range, 0–42 μmol/L). Prothrombin and partial thromboplastin times were normal.

Given the hyperammonemia, lack of cardiac disease, and lack of portal vein abnormalities, it was concluded that the transudative abdominal effusion was likely caused by PH. Computed tomography with angiography (CTA) or trans-splenic scintigraphy were recommended, but as they were unavailable on site, surgery was elected. An exploratory laparotomy was done to assess the portal vasculature and to obtain liver biopsies.

Abdominal exploration, done by a board-certified surgeon, revealed a large amount of straw-colored effusion and multiple small vessels within the omentum, along the left abdominal dorsal wall, between the spleen and left kidney (Figure 1). Subjectively, the portal vein appeared to be of normal size. The liver appeared small but was homogenous. The remainder of the abdominal exploratory was unremarkable. A diagnosis of MAPSS was made. Two wedge liver biopsies (left medial lobe, left lateral lobe) and a duodenal biopsy were obtained. Additional small intestinal biopsies were not taken due to moderate hypotension (systolic blood pressure [SBP] of 65 mm Hg), which was unresponsive to synthetic colloid^c boluses, and concern for patient morbidity.

View Figure

• Download as Powerpoint
• Download Figure

During recovery, mild hypotension (SBP of 78 mm Hg) was noted. Laboratory abnormalities in the postoperative period included a mild anemia of 36% (reference range, 37–55%) and hypoalbuminemia (1.1 g/dL; reference range, 2.6–3.7 g/L), the latter of which was likely exacerbated by hemodilution and surgical blood loss. The patient was given 5 grams of canine-specific albumin^d (0.52 g/kg IV) over 4 hr. Canine-specific albumin was empirically selected to avoid approaching coagulopathic doses of synthetic colloid, for its colloidal support, and to increase albumin. To address persistent hypotension despite albumin, crystalloid^e, and synthetic colloid boluses, a continuous rate infusion (CRI) of norepinephrine^f (0.05–0.5 ug/kg/min) was initiated to maintain SBP over 90 mm Hg. Additional postoperative treatments included methadone^g (0.2 mg/kg IV q 6 hr), pantoprazole^h (1 mg/kg IV q 24 hr), metronidazoleⁱ (10 mg/kg IV q 12 hr), lactulose^j (0.25 mL/kg q 8 hr), spironolactone^k (1.3 mg/kg per os [PO] q 12 hr), a synthetic colloid solution (1 mL/kg/hr as a CRI), and a balanced crystalloid solution (3 mL/kg/hr as a CRI). The norepinephrine CRI was weaned after 14 hr. Oral spironolactone and lactulose were continued, and he was transitioned to other oral medications, including over-the-counter omeprazole (2 mg/kg PO q 24 hr), metronidazole^l (6.5 mg/kg PO q 12 hr), tramadol^m (2.6 mg/kg PO q 8 hr), and s-adenosyl-methionine with milk thistleⁿ (25 mg/kg PO q 24 hr on an empty stomach). Initially, the patient was started on a low-protein diet^o to decrease blood ammonia concentration and mitigate hepatic encephalopathy (HE). However, to address rapid reaccumulation of ascites, a lower-sodium diet^p was selected. The patient was discharged 72 hr postoperatively. The hypoalbuminemia had improved (1.4 g/dL; reference range, 2.6–3.7 g/dL) as had the hyperammonemia (49 μmol/L; reference range, 0–42 μmol/L).

The hepatic histopathologic findings included the following: increased arteriolar profiles, lobular atrophy, hypoplasia of the portal vein tributaries, and periportal sinusoidal dilation consistent with portal hypoperfusion. There was no bile duct proliferation, smooth muscle hypertrophy, regenerative nodules, or periportal fibrosis noted. Duodenal sections showed no significant abnormalities. Special stains were not requested.

Ten days after release, the patient represented for progressive abdominal distension. A chemistry profile showed low blood urea nitrogen (2.0 g/dL; reference range, 9.8–37.3 g/dL) and panhypoproteinemia characterized by hypoalbuminemia (1.6 g/dL; reference range 2.6–3.7 g/dL) and hypoglobulinemia (1.8 g/dL; reference range, 2.2–3.5 g/dL). Palliative abdominocentesis was done, during which 2.5 L of ascites were drained. The spironolactone dose was increased to 2.5 mg/kg PO q 12 hr.

The patient required frequent re-evaluation every 2–3 wk for palliative abdominocentesis over the subsequent 6 mo. Spironolactone was increased further to 4.4 mg/kg PO q 12 hr, while other medications remained the same. Furosemide^q (1.2 mg/kg PO q 12 hr) was added for its synergistic effects with spironolactone to control ascites.^3,5 The patient experienced signs of HE suspected to be caused by furosemide-induced potassium wasting. Furosemide was subsequently discontinued, after which the owners reported resolution of HE. Despite this, they contemplated euthanasia because of the frequency of palliative abdominocentesis (as often as weekly). The recommendation was made to add glucocorticoids for suspected protein-losing enteropathy (PLE) given persistent panhypoprotenemia. While the use of steroids may have been contraindicated in a muscle-wasted patient at high risk of gastrointestinal bleeding, the possible benefits of these drugs were thought to outweigh the risks. Dexamethasone^r was chosen instead of prednisone because of its lack of mineralocorticoid activity and therefore minimal effect on effective circulating volume. He was started on dexamethasone (0.11 mg/kg PO q 48 hr), after which the frequency of palliative abdominocentesis decreased to an average of every 3 to 4 wk. Dexamethasone dosing was elected every other day because of its long-acting effects and hepatic metabolism. At the time of this writing (18 mo after diagnosis), the patient still required abdominocentesis approximately every 3 to 4 wk, and his medical regimen included a restricted protein and sodium diet, amoxicillin, lactulose, spironolactone, s-adenosyl-methionine with milk thistle, famotidine, and dexamethasone. Repeat lab work at that time showed a normal complete blood count, normal chemistry panel, normal urinalysis, normal ammonia concentration, and elevated paired SBA concentrations (pre-SBA 67.9 μmol/L; post-SBA 42 μmol/L). His body condition had markedly improved, and the volume of ascites during palliative abdominocentesis had markedly diminished. His owners reported an excellent quality of life at home.

Discussion

Portosystemic shunts (PSS) are either congenital or acquired.² While congenital PSS are most commonly a single intrahepatic or extrahepatic macroscopic vessel, acquired PSS are usually extrahepatic and multiple. These anomalous vessels create a connection between the portal venous supply and systemic circulation, therefore bypassing the liver.^1,2 Acquired PSS develop as a consequence of sustained PH. The pathophysiology of PH is complex but is attributed in part to an increase in vascular resistance to and within the liver, an increase in overall blood flow to the liver, or both.^1,3 The classification of PH is based on the anatomic location of the lesion to which it is secondary (prehepatic, intrahepatic, or posthepatic).³

Figure 2 summarizes potential etiologies for PH. Prehepatic PH is caused by intraluminal obstruction or extraluminal compression of the portal vein. Intrahepatic PH occurs with increased resistance to portal vein flow in the portal vein tributaries, sinusoids, or small hepatic veins. Presinusoidal intrahepatic PH occurs as a consequence of increased resistance of portal blood through the portal vein tributaries. Possible etiologies of presinusoidal intrahepatic PH include PHPV, chronic cholangitis, hepatic arteriovenous fistulas, nodular hyperplasia, and ductal plate abnormalities.³ Noncirrhotic PH, microvascular dysplasia, hepatoportal scleral fibrosis, and idiopathic portal fibrosis (all previously considered separate etiologies of presinusoidal intrahepatic PH) have been reclassified as PHPV, according to the World Small Animal Veterinary Association guidelines on canine and feline liver diseases.^1,3 Primary hypoplasia of the portal vein has a wide range of clinical severity depending on the degree of hypoplasia of the portal vein, consequential development of MAPSS, and loss of hepatocellular function.¹ Sinusoidal intrahepatic PH is usually a result of fibrotic hepatopathies, while postsinusoidal intrahepatic PH is due to veno-occlusive diseases. Posthepatic PH may be due to right-sided heart failure, pericardial disease, or pulmonary hypertension.³

View Figure

• Download as Powerpoint
• Download Figure

Sequelae of sustained PH include ascites, development of MAPSS, HE, and gastric ulceration, among others.^2,3 Ascites occurs secondary to increased portal venous hydrostatic pressure when splanchnic vasculature dilates and lymphatics no longer accommodate excessive interstitial fluid production. Hypoalbuminemia, caused by hepatic synthetic failure and gastrointestinal losses secondary to PH-induced bowel congestion, may exacerbate the development of ascites. Fluid accumulation within the peritoneal cavity may cause decreased effective circulating volume, leading to a relative hypovolemia.³ As the portal–hepatic venous pressure gradient increases, MAPSS develop in cases of prehepatic and intrahepatic PH.^1,3 In dogs, most MAPSS are portal postcaval shunts, which develop in the region of the kidneys.⁶ As no pressure gradient exists between the portal and systemic circulation, ascites does not develop when a macroscopic congenital PSS and PHPV occur concurrently.^1,3 Not only is ascites considered a negative prognostic indicator in dogs with liver disease, but Adam and colleagues found its presence to be a strong predictor of MAPSS when comparing 54 dogs with PSS of various etiologies.^3,7

Portal hypertension–induced bowel congestion may lead to a PLE, which is characterized by panhypoproteinemia and histopathologically by bowel edema, lacteal dilation, and lymphangiectasia.⁵ While the histopathological findings in this case did not show such lesions, it is possible that a single duodenal biopsy may not have been representative of the true distribution and severity of the intestinal disease. Ideally, biopsies of all small intestinal segments would have been done as well as a malabsorption panel, which includes trypsin-like immunoreactivity, cobalamine, and folate. While the young age of the patient and hepatic immaturity may have contributed to the hypoglobuminemia seen in this case, a PLE is a more plausible explanation given the lack of blood loss, overt immunodeficiency, and the presence of concurrent hypoalbuminemia.

Hepatic encephalopathy is an incompletely understood and multifactorial syndrome that manifests as a wide range of neurological clinical signs, ranging from behavioral changes to seizures or even comatose states. The pathogenesis is presumed to be associated with the accumulation of various compounds in circulation, including ammonia, glutamine, manganese, bile acids, fatty acids, and false neurotransmitters, amongst others.^2,3,8,9 Ammonia is implicated as the major compound responsible for HE. Causes of hyperammonemia include shunting of portal blood, hepatic failure with loss greater than 70% of function, urea cycle enzyme deficiencies, methylmalonic acidemia, and urethral obstruction–induced hyperammonemia.⁵ A breakdown product of protein ingested by colonic bacteria, ammonia enters the systemic circulation and bypasses the liver through the PSS.^2,3 It crosses the blood–brain barrier and works synergistically with an inflammatory response to cause astrocyte swelling and cerebral edema leading to the clinical manifestations of HE. Metabolic derangements including dehydration, hypoglycemia, hypokalemia, hyponatremia, azotemia, and alkalemia exacerbate HE. High-protein diets and gastrointestinal ulcers do the same.⁸ The patient described here had clinical signs of HE and hyperammonemia without any other metabolic derangements. Dogs with hyperammonemia may have urine that is hyposthenuric or isosthenuric (>50%) or that contains urate crystals (26–57%).² Despite these reported incidences, the patient in this report had a urine concentration of 1.052 and lacked urate crystalluria.

The most common tests used to infer the presence of PH are liver-function tests such as blood ammonia or SBA.^2,3,10 Bile acids, which aid in the digestion of fats, are produced and conjugated in the liver, excreted into bile, and—in healthy individuals—95% are absorbed in the distal small intestine (ileum). Many factors influence the enterohepatic circulation of bile acids, including completeness of gallbladder contraction, rate of gastric emptying, intestinal transit time, efficiency of ileal bile acid reabsorption, and the frequency of enterohepatic cycling.⁵ While primary veterinarians perform SBA routinely, the logistical complications associated with blood ammonia concentrations often preclude this test from being done regularly. As evaluation of the blood sample must occur within 30 min of collection, freezing the ammonia sample is required prior to submission unless a bedside ammonia test is available.^10,11

Evidence in the literature suggests that postprandial SBA concentration may be the most sensitive blood test for detection of PSS but that blood ammonia concentration is more specific for this condition.^3,10,12,13 Overall, it is considered very unusual that an animal with both normal fasting and postprandial SBA concentrations would have a PSS of any cause.¹¹ In reviewing 338 cases in the literature of PSS of varying etiologies, only 20 cases (5.9%) presented with normal SBA. Sixteen (4.7%) of these cases were diagnosed with a single macroscopic congenital PSS. The remaining four cases (1.1%) had acquired shunting; however, only fasting SBA were available for review. It is not clear if any of these patients had an elevated ammonia or were encephalopathic.^7,10,11,13 Based on the aforementioned studies, it can be concluded that it is exceptionally rare for SBA to be normal in cases of MAPSS. In this particular case, the patient had normal pre- and post-SBA concentrations on presentation, despite the severity of his symptoms. While we can attribute the HE to hyperammonemia, and while the histopathologic diagnosis of portal hypoperfusion with concurrent PH may explain the ascites, we would have expected the SBA to be abnormal.

Bile acids may be lower than expected for various reasons. These reasons include disturbances in bile acid synthesis, biotransformation, conjugation, transport, or reabsorption.⁵ While evidence of a PLE could not be corroborated based on a single duodenal biopsy in this case, the authors presume that the lower-than-expected SBA concentration was related to prolonged gastrointestinal transit time and decreased ileal reabsorption related to this patient’s PLE. The authors also cannot rule out small intestinal bacterial overgrowth, which may have led to bacterial conjugation and a subsequent decrease in ileal reabsorption of SBA.⁵

Serum bile acids were not repeated prior to surgery, as there was sufficient information to support PH given the presence of a pure transudate, lack of cardiac changes, and hyperammonemia. Also, the results were deemed reliable, as the standard protocol for the measurement of SBA had been followed.⁵ Serum bile acids were done, however, at the time of this writing. Interestingly, the SBA returned elevated (preprandial was higher than postprandial, likely due to premature gallbladder contraction). Glucocorticoid-induced cholestasis is one plausible explanation for this increase; however, serendipitous treatment of a PLE could also explain this change. While it could be argued that any increase in SBA weakens the premise of this case report, the authors feel that this result actually strengthens our hypothesis for why the SBA returned lower than was originally expected. The patient had been receiving medical management including the use of glucocorticoids for the prior 18 mo. Resolution of the panhypoproteinemia and diminishment of ascites supports the improvement of the patient’s presumed PLE. With said improvement, one would expect re-establishment of appropriate bile acid reabsorption in the ileum, leading to an increased SBA, as is typically seen in patients with PH and MAPSS. This case suggests that despite a reported sensitivity of up to 100% in some studies for the detection of PSS, SBA cannot be used alone to rule out the presence of MAPSS.

In conjunction with biochemical testing, diagnostic imaging is helpful in ascertaining underlying etiologies of PH. Ultrasound is often the initial imaging modality of choice because of its limited invasiveness and relatively low cost.² It is used to confirm the presence and allow sampling of abdominal effusion, and according to the World Small Animal Veterinary Association guidelines on liver disease, ultrasound is sufficient for the identification of occlusive portal vein diseases and arteriovenous fistulas.¹ While not as sensitive or specific as CTA, ultrasound is also used to detect the presence of MAPSS. Ultrasound can also assess portal vein dynamics, which may support PH. These include an enlarged portal vein, decreased portal vein velocity (<10 cm/sec), and portal-vein-to-aorta ratio of <0.65.³ The ultrasonographic appearance of the liver is variable among causes of PH and, in many cases, may be normal. Other imaging techniques used to identify possible causes of PH include angiography, transcolonic or trans-splenic scintigraphy, contrast-enhanced helical computed tomography, and contrast-enhanced magnetic resonance angiography.^2,3,14 A recent retrospective study found CTA to be more sensitive (96% versus 68%) and specific (89% versus 84%) when compared to abdominal ultrasound for the detection of PSS of any cause.¹⁵ Helical CTA may be the ideal imaging modality for the diagnosis of PSS because of its fast scan time, good contrast and resolution, and ability to make three-dimensional images for surgical planning. The presence of MAPSS, however, can be difficult to confirm.^6,14,15 As seen in this case, MAPSS may be missed with ultrasound alone, even in the hands of an experienced radiologist. In this case, exploratory laparotomy was needed to confirm the presence of MAPSS.

The diagnosis of PHPV is made when no macroscopic congenital PSS, arteriovenous fistula, or portal vein obstruction is identified and there is histopathological evidence of hepatic hypoperfusion in the absence of inflammation and nodular regeneration.^1,3,16 With portal hypoperfusion, the portal vein profiles are typically diminished to absent. Consequently, the hepatic arterioles have an increased blood flow leading to increased numbers of arteriolar profiles within the portal tracts.¹ Varied morphology may be present depending on the degree of hypoplasia of the portal vein. Portal fibrosis, hepatocellular atrophy, vacuolar degeneration, bile ductule proliferation, lipogranulomas, hypoplasia of intrahepatic portal tributaries, smooth muscle hypertrophy, and sinusoidal dilation may also occur.^1,6 In this patient, the histopathology showed moderately increased arteriolar profiles, mild vacuolar degeneration, lobular atrophy, hypoplasia of the portal tributaries, and periportal sinusoidal dilation. No bile duct proliferation, lipogranulomas, inflammation, or fibrosis was documented. The authors find it interesting that given the severity of clinical signs in the case presented, the histopathological changes were not more severe. Even though the liver appeared grossly homogenous during surgery, as only two wedge liver biopsies were obtained from the left liver lobes, it is possible that nonuniform vascular changes were not represented. Ideally, biopsies from both the left and right liver lobes would have been obtained, as portal blood flow may be variable between the lobes of the liver.

Typically, dogs with severe PHPV are young (<2 yr) with ascites, HE, and gastrointestinal signs. Dogs with PHPV often have moderate-to-severe increases in liver enzymes, mild-to-moderate hyperbilirubinemia, hyperammonemia, increased SBA concentrations, hypoalbuminemia, and transudative abdominal effusions.^{3,7,13,16–18} The reader should not assume that any dog with hyperammonemia and a transudative abdominal effusion has PHPV with secondary PH. Causes for hyperammonemia have already been discussed. The cause of the hyperammonemia in this case can be explained by shunting of portal blood. Etiologies of transudative abdominal effusions other than PH include the following: decreased oncotic pressure secondary to hypoalbuminemia (PLE, protein-losing nephropathy, or liver failure), increased hydrostatic pressure (right heart failure), and (rarely) increased vascular permeability.⁵ This case of PHPV is unique because of the absence of liver enzyme elevations and normal SBA concentrations. Also, to the authors’ knowledge, this is one of the youngest dogs diagnosed with this condition and the only case of PHPV with both normal paired SBA and liver enzyme concentrations at the time of diagnosis in the veterinary literature.^7,11,16–18

Treatment of PH primarily involves correction of the underlying disease. Unfortunately, there is no viable surgical treatment for MAPSS, as collateral vessels develop in response to PH and will recur when ligated.^2,3 Medical management includes dietary sodium restriction, diuretic therapy, palliative abdominocentesis, reduction of gastric acid production, provision of liver protectant/antioxidant medications, and amelioration of hyperammonemia and HE. Control of HE is accomplished with a low-protein diet, antibiotics, and lactulose.² In this case, given the presence of panhypoproteinemia from suspected PLE, a fat-restricted, hydrolyzed protein diet was considered in addition to glucocorticoid administration.

Experimental therapies to decrease PH such as the creation of a transjugular intrahepatic PSS has been described in human medicine but has never been documented in veterinary medicine.³ Caudal vena caval banding has been reported in the veterinary literature; however, this technique is not recommended, as it does not affect survival time.¹⁹ In human medicine, a variety of medications including beta blockers, somatostatin, and vasopressin analogues are used to induce splanchnic vasoconstriction. Statins, angiogenesis inhibitors, and renin-angiotensin system blocking drugs are used to induce sinusoidal vasodilation in order to lower portal venous pressure.^3,20 These drugs have not been proven safe nor effective in veterinary medicine and therefore cannot be recommended.³

Conclusion

Primary hypoplasia of the portal vein is a common congenital disorder in dogs. However, PHPV with secondary PH and the development of MAPSS is uncommon based on current veterinary literature. The clinical manifestations of this disease are highly variable depending on the degree of portal vein hypoplasia, the presence of MAPSS, and the extent of hepatocellular dysfunction. Diagnosis of this condition not only requires documentation of the typical findings of portal hypoperfusion histopathologically but also necessitates the exclusion of an arteriovenous fistula, portal vein thrombosis, and congenital macroscopic PSS. Other diagnostic criteria typically include abnormal liver enzymes and the documentation of poor hepatocellular function with increased SBA and fasting ammonia concentrations.

It is apparent from this case that PHPV cannot be ruled out based solely on normal liver enzyme and SBA concentrations. Serum bile acid concentrations were lower than expected at the time of presentation, presumably related to the patient’s PLE. The subsequent increase in SBA concentration can be explained by successful treatment of this patient’s presumed PLE. The authors recognize that this case report would have been strengthened had we obtained intestinal biopsies from all intestinal segments, done a CTA, a malabsorption panel, and repeat SBA concentrations prior to the institution of long-term medical management. Diagnosis ultimately required exploratory laparotomy to visually confirm the presence of MAPSS and to histopathologically diagnose portal hypoperfusion. Overwhelmingly, while the literature does not describe the management of severely affected dogs over the long term, this patient has demonstrated that with supportive palliative care, quality of life can be maintained.