What Is the Evidence?

Scott L. Owens Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Purdue University, West Lafayette, IN 47907.

Search for other papers by Scott L. Owens in
Current site
Google Scholar
Nolie K. Parnell Small Animal Medical Clinic, Veterinary Teaching Hospital, Purdue University, West Lafayette, IN 47907.

Search for other papers by Nolie K. Parnell in
Current site
Google Scholar

Click on author name to view affiliation information


A 6-month-old sexually intact male Labrador Retriever was evaluated at the Purdue University Veterinary Teaching Hospital for a 3-month history of lethargy and failure to thrive and recent onset of vomiting and generalized neurologic deficits. Historical findings included exercise intolerance, occasional circling, polyuria, and polydipsia. At evaluation, the dog was considered of small stature for its breed, had a body condition of 2 on a 5-point scale (moderately underweight), and was lethargic and poorly responsive. Physical examination revealed a grade III/VI systolic heart murmur (point of maximum intensity, right cardiac apex) and ptyalism. A neurologic examination revealed delayed conscious pro-prioception in the hind limbs and decreased strength of hopping in all 4 limbs. Evaluation of cranial nerve function revealed an absent menace response bilaterally but presence of vision and tracking and a delayed response to nasal mucosa stimulation. These signs were consistent with a bilateral forebrain lesion.

Laboratory diagnostic testing included a serum biochemical analysis, CBC, urinalysis, and resting blood ammonia concentration. Serum abnormalities included a urea nitrogen concentration at the low end of the reference range (8 mg/dL; reference limits, 7 to 32 mg/dL), high alanine aminotransferase (118 U/L; reference limits, 3 to 69 U/L) and alkaline phosphatase (186 U/L; reference limits, 20 to 157 U/L) activities, low cholesterol concentration (120 mg/dL; reference limits, 125 to 301 mg/dL), and hyperammonemia (248.6 μmol/L; reference limits, 1 to 46 μmol/L). Urine abnormalities included dilute urine (urine specific gravity, 1.028; reference limit, ≥ 1.030) and moderate ammonium biurate crystalluria. The total serum bile acids concentration was abnormal (preprandial, 106.0 μmol/L [upper reference limit, 13 μmol/L]; postprandial, 130.9 μmol/L [upper reference limit, 25 μmol/L]). Thoracic radiography revealed nothing remarkable, and echocardiography revealed mild tricuspid valve regurgitation. Results of abdominal radiography indicated bilateral renomegaly and microhepatica. Ultrasonography of the abdominal cavity was performed; abnormalities included bilateral renomegaly with anatomically normal architecture and echogenic debris within the urinary bladder. An extrahepatic anomalous vessel was suspected; however, considerable colonic gas prevented complete examination and decreased the possibility of additional detail.

The combination of these findings was consistent with a diagnosis of portosystemic shunt (PSS). Although results of ultrasonography raised the suspicion of an extrahepatic shunt, Labrador Retrievers are more likely to have an intrahepatic shunt.1 To more accurately characterize the location of the shunt and determine whether interventional vascular correction was required, cranial mesenteric arteriography was performed with the dog anesthetized. One large intrahepatic anomalous vessel was identified, with 4 to 5 additional small vessels accompanying it. The vessel was located near the vena cava, and little vessel branching was seen within the hepatic parenchyma. Because of the size and location of the aberrant vessels, interventional closure techniques could not be performed. The dog recovered from anesthesia without complication.

The number and location of the vascular anomalies posed a substantial risk of surgical complications should complete occlusion have been attempted.2 Surgical intervention was discussed with and considered by the dog's owner but was ultimately declined. With an inability to further pursue vessel closure because of location and risk, the focus turned to long-term medical management and finding a means to optimize the dog's quality of life for as long as possible.

Formulation of the Clinical Question

Complete or partial occlusion of a PSS, when possible, is widely considered the ideal treatment because this allows slow neovascularization that will eventually deliver an increased volume of portal blood to the liver and improve hepatic function.2 When this is not possible because of financial constraints, owner unwillingness owing to potential serious complications, the number of aberrant vessels, or the location of the shunts, medical management is chosen instead of surgical intervention. Medical management of a dog with a PSS is an ongoing process that must be tailored as the dog ages. Many medications and dietary treatments can be used to maximize quality of life for affected dogs.

Clinical Question

What is the ideal medical management (including dietary and pharmaceutical) of an encephalopathic dog with PSS, and what prognostic information can be communicated to the owner?

Evidentiary Search Strategy

The preferred method to answer the clinical question is the evidence-based medicine approach. Three separate search databases were used: CAB Abstract and the PubMed database to search for relevant peer-reviewed journal articles and the Veterinary Information Network to search for recent abstracts presented at major companion animal internal medicine conferences during the past 3 years. Search terms included the following: dogs, portosystemic shunt, medical, and congenital. These 3 brief searches yielded abundant clinical data and research regarding surgical correction of PSSs as well as literature reviews on the disease, yet only a few reports specifically dealing with medical and dietary treatment were identified. Relevant literature included comprehensive narrative review articles; reports of retrospective studies addressing intrahepatic shunt survival rates, prognostic indicators, medical versus surgical treatment, and medical management alone; reports of prospective studies (clinical trials and case series); and a non-peer-reviewed abstract. When a lack of prospective studies on the efficacy of various medications for treating this condition was discovered, the search was expanded to include relevant human literature.

Review of the Evidence

Many studies involving comparison of surgical techniques and outcomes for dogs with a PSS have been reported, but little information of higher evidentiary value (eg, findings of randomized, controlled clinical trials, experiments, or cohort studies vs those of case reports and case series) was identified on medical management of this condition. Medication recommendations are often extrapolated from human medicine, in which considerably more research of medical management of PSSs has been performed and reported. Anecdotal information and expert opinion are often the only evidence available for constructing a therapeutic plan. Although it is impossible to derive every facet of a treatment plan from experimental studies of PSS management, reports of 2 such studies were available for formulating an appropriate diet. In 1 prospective clinical trial in dogs with experimentally induced portosystemic shunts,3 the minimum daily recommended protein requirement in adult dogs with PSS was evaluated, with the results indicating 2.1 g of crude protein/kg/d (0.95 g of crude protein/lb/d) is the minimum amount that will preserve total body protein stores. In the other study,4 which was a double-blinded crossover study, findings suggested the importance of the source of the dietary protein, in addition to protein restriction, in helping to manage hyperammonemia in dogs with a PSS. The conclusion was that a soy-based protein is less ammoniagenic than a meat-based protein.

Veterinary-specific evidence in the form of clinical trials or prospective cohort studies supporting use of medications in dogs with an inoperable PSS was not identified during the literature searches. The largest retrospective case series5 in which medical management alone was evaluated revealed that dogs with a congenital PSS could be expected to live between 9 months and 4 years or greater if blood urea nitrogen concentration was within reference limits, the condition was not diagnosed until later than 2 years of age, and the dogs received appropriate medical management. Most dogs in that study were managed with multimodal therapy including antimicrobials, lactulose, and a diet with moderate protein restriction. However, the dogs in that study had a single anomalous vessel, whereas the dog in this report had multiple aberrant vessels.

The lack of veterinary evidence prompted consideration of findings in humans medically treated for a PSS. Commonly used medications for PSSs in humans include nonabsorbable disaccharides such as lactulose and antimicrobials such as metronidazole or neomycin. A randomized controlled clinical trial6 involving humans with a PSS revealed that lactulose and neomycin were equally effective at resolving clinical signs associated with hepatic encephalopathy, whereas another randomized prospective clinical trial7 revealed that metronidazole is equally effective as neomycin in similar situations. These 2 studies, while not involved with hepatic encephalopathy, still provided scientific evidence worthy of consideration in devising a treatment plan for the dog in this report. In addition, a literature review8 was identified regarding use of lactulose, antimicrobials, and other strategies in the treatment of humans with hepatic encephalopathy. Findings of that review suggested lactulose be used as a first-line treatment and that antimicrobial administration be reserved for patients that remain symptomatic or cannot tolerate the lactulose.

Given the aforementioned evidence, what decision would you make?

Clinical Decision and Outcome

Prior to instituting a detailed medical plan, it was important to review all prognostic indicators that would help to set reasonable expectations for the owner. Younger age at diagnosis, low BUN concentration, hypoproteinemia, and low PCV are all negatively associated with long-term survival rate in dogs with congenital PSS.2,5 Other commonly identified abnormalities, including hyperammonemia, high total serum bile acids concentration, and histopathologic changes in the liver, appear not to be associated with the probability of survival.5,9 The dog in this report was < 2 years of age at the time of diagnosis and had a mild normocytic normochromic nonregenerative anemia, and both young age and anemia are reportedly negative prognostic indicators. The dog's BUN and serum total protein concentrations, however, were within reference limits. This prognostic information must be interpreted cautiously, however, as it relates to single anomalous vessels, whereas this case represents a dog with multiple aberrant vessels.

A long-term medical plan was formulated to best address all foreseeable complications of the disorder. One of the most challenging aspects of long-term medical management of PSSs in dogs is protein balance. Protein restriction must be implemented to prevent ammonia production, which results in hepatic encephalopathy secondary to an ineffective urea cycle. However, iatrogenic hypoproteinemia arising from protein restriction and secondary malnutrition increases the risk for systemic protein breakdown, hepatic degeneration, decreased muscle mass, and decreased oncotic pressure leading to third-space loss of fluids.3,4 On the basis of the findings of the 2 aforementioned dietary studies, a commercial therapeutic diet for hepatic healtha containing protein from egg and soybean was chosen for the dog in this report. After calculation of protein intake (minimum of 2.1 g of crude protein/kg/d), the dog's required daily intake of the diet was estimated to be 3 1/3 cups. This quantity was deemed sufficient to meet daily nutrient and energy requirements.

In addition to nutritional management, a decision was made to treat the dog with lactulose (0.5 mL/kg [0.23 mL/lb], PO, q 8 h) for the duration of its life to manage the clinical signs of hepatic encephalopathy. The dose was subject to adjustment depending on weight gain, serum ammonia concentration, and fecal consistency. The decision to use lactulose was made on the basis of human and veterinary literature reviews1,8 and the supposition that lactulose would acidify the dog's colonic contents and prevent bacteria-produced ammonia from being absorbed into the systemic circulation. Antimicrobial administration was not initiated, given the findings of the literature review.8 Because gastrointestinal bleeding and ulcerations are another commonly encountered complication in dogs with a PSS, administration of a proton pump inhibitor (omeprazole, 1.0 mg/kg [0.45 mg/lb], PO, q 24 h) to reduce gastric acid production was also initiated.

Nine months after initiating medical treatment, the dog was no longer hyperammonemic (serum ammonia concentration, 43 μmol/L; reference limits, 1 to 46 μmol/L) and was clinically normal. Its body condition score improved to 3.5/5 (ideal). The serum albumin concentration improved to within reference limits (2.6 g/dL), and the mild microcytic normochromic anemia remained constant.


Decision making in veterinary internal medicine should be based on available evidence. However, as was the situation with the dog with an inoperable PSS, there are deficiencies in the veterinary literature that must be filled by clinical experience and research findings from other species. Because no peer-reviewed prospective clinical trials were identified in which surgical correction was compared with medical management in dogs with a naturally occurring PSS, information needed to be carefully extrapolated from studies of lower evidentiary value, narrative reviews, and clinical experience. A further complication was that there are multiple forms and locations of vascular anomalies; therefore, it can be difficult for veterinarians to tailor a treatment approach specific for an individual patient on the basis of the literature that is available. The number or location of aberrant vessels is also important information in devising a treatment plan because the end result of restricted hepatic circulation, hepatic encephalopathy, must be addressed in affected animals. Furthermore, location of aberrant vessels influences the probability of postsurgical complications, with perioperative complication rates for intrahepatic PSSs reportedly as high as 77% and the likelihood of postsurgical death as high as 28%.2

Given the limited amount of evidence found and the specific nature of the PSS in the dog in this report, we based our treatment plan on the best evidence at hand. In this situation, the outcome was successful. At this point in veterinary research, it is unlikely that veterinarians will be able to identify clinical trials that involve animals with the exact conditions as in the animals they treat. Research resources such as funding and manpower simply are not comparable with those in human medicine. This does not negate the worth of clinical experience or other information considered of low evidentiary value in the decision-making process, nor does it negate the potential usefulness of research findings in other species, as the outcome here suggests.


Prescription Diet l/d Canine, Hill's Pet Nutrition Inc, Topeka, Kan.


  • 1.

    Berent AC, Tobias KM. Portosystemic vascular anomalies. Vet Clin North Am Small Anim Pract 2009; 39: 513541.

  • 2.

    Papazolgou LG, Monnet E, Seim HB III. Survival and prognostic indicators for dogs with intrahepatic portosystemic shunts: 32 cases (1990–2000). Vet Surg 2002; 31: 561570.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3.

    Laflamme DP, Allen SW, Huber TL. Apparent dietary protein requirement of dogs with portosystemic shunt. Am J Vet Res 1993; 54: 719723.

  • 4.

    Proot S, Biourge V, Teske E, et al. Soy protein isolate versus meat-based low-protein diet for dogs with congenital portosystemic shunts. J Vet Intern Med 2009; 23: 794800.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 5.

    Watson PJ, Herrtage ME. Medical management of congenital portosystemic shunts in 27 dogs—a retrospective study. J Small Anim Pract 1998; 39: 6268.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6.

    Conn HO, Leevy CM, Vlahcevic ZR, et al. Comparison of lactulose and neomycin in the treatment of chronic portal-systemic encephalopathy. Gastroenterology 1977; 72: 573583.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7.

    Morgan MH, Read AE, Speller DC. Treatment of hepatic encephalopathy with metronidazole. Gut 1982; 23: 17.

  • 8.

    Al Sibae MR, McGuire BM. Current trends in the treatment of hepatic encephalopathy. Ther Clin Risk Manag 2009; 5: 617626.

  • 9.

    Parker JS, Monnet E, Powers BE, et al. Histologic examination of hepatic biopsy samples as a prognostic indicator in dogs undergoing surgical correction of congenital portosystemic shunts: 64 cases (1997–2005). J Am Vet Med Assoc 2008; 232: 15111514.

    • Crossref
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 112 0 0
Full Text Views 607 549 253
PDF Downloads 156 100 14