Abstract
Objectives
The primary goals of this retrospective study were to describe a population of dogs with portal hypertension secondary to liver disease, and to assess whether prognosis could be inferred from historical, clinical, and clinicopathological data.
Animals and procedures
Dogs (N = 76) diagnosed with intrahepatic portal hypertension between 2011 and 2020 were included; dogs with known congenital hepatic anomalies were excluded. Effect on survival was assessed using univariable and multivariable Cox proportional hazards models for historical, clinical, and clinicopathological variables.
Results
Dogs survived for a median of 14 d (range: 0 to 2028 d), with 31.6% being euthanized within 2 d of diagnosis and 23.7% surviving longer than 2 mo. Presence of jaundice and duration of clinical signs, expressed in days, were significantly associated with outcome in the univariable analysis (HR = 1.846, 95% CI: 1.094 to 3.117, P = 0.02; HR = 0.995, 95% CI: 0.990 to 1.000, P = 0.033, respectively). However, only presence of jaundice was significantly associated with increased hazard of death in the multivariable analysis.
Conclusion
Results of this study show that portal hypertension is associated with a poor prognosis; however, some dogs can show prolonged survival.
Clinical relevance
Clinical data can guide decision-making for clinicians and owners.
Résumé
Indicateurs pronostiques chez les chiens atteints d’hypertension portale intra-hépatique
Objectifs
Les principaux objectifs de cette étude rétrospective étaient de décrire une population de chiens souffrant d’hypertension portale secondaire à une maladie hépatique et d’évaluer si le pronostic pouvait être déduit à partir de données historiques, cliniques et clinicopathologiques.
Animaux et procédures
Les chiens (N = 76) diagnostiqués avec une hypertension portale intrahépatique entre 2011 et 2020 ont été inclus; les chiens présentant des anomalies hépatiques congénitales connues ont été exclus. L’effet sur la survie a été évalué à l’aide de modèles de risques proportionnels de Cox univariés et multivariés pour les variables historiques, cliniques et clinicopathologiques.
Résultats
Les chiens ont survécu pendant une durée médiane de 14 jours (intervalle : 0 à 2028 jours), 31,6 % ayant été euthanasiés dans les 2 jours suivant le diagnostic et 23,7 % ayant survécu plus de 2 mois. La présence d’ictère et la durée des signes cliniques, exprimées en jours, étaient significativement associées au résultat de l’analyse univariée (HR = 1,846, IC à 95 % : 1,094 à 3,117, P = 0,02; HR = 0,995, IC à 95 % : 0,990 à 1,000, P = 0,033, respectivement). Cependant, seule la présence d’ictère était significativement associée à un risque accru de décès dans l’analyse multivariée.
Conclusion
Les résultats de cette étude montrent que l’hypertension portale est associée à un mauvais pronostic; cependant, certains chiens peuvent montrer une survie prolongée.
Pertinence clinique
Les données cliniques peuvent guider la prise de décision des cliniciens et des propriétaires.
(Traduit par Dr Serge Messier)
Introduction
Portal hypertension is a commonly recognized consequence of hepatic disease (1,2). There are multiple mechanisms and etiologies for the development of portal hypertension; these are classified as prehepatic (portal vein), intrahepatic (pre-sinusoidal, sinusoidal, post-sinusoidal), and post-hepatic. Intrahepatic sinusoidal portal hypertension is one of the most common forms of portal hypertension in dogs and is classically associated with hepatic fibrosis, in turn most frequently a sequel to chronic hepatitis (3–7). Although the pathophysiology is not fully understood, architectural hepatic changes (including fibrosis, capillarization of the sinusoids, presence of microthrombi in the intrahepatic vasculature, and regenerative nodule formation), functional processes (overproduction of inflammatory mediators, vasoconstriction, activation of hepatic stellate cells, splanchnic vasodilation, and neurohormonal activation), and angiogenesis are all suspected to contribute to the development and progression of intrahepatic sinusoidal portal hypertension (3,8).
The gold standard for diagnosis of portal hypertension is direct measurement of portal venous pressure. However, this is rarely performed in clinical settings. The presence of intrahepatic portal hypertension in dogs is typically inferred from the identification of its clinical consequences, including ascites, gastrointestinal hemorrhage, hepatic dysfunction/failure (e.g., hepatic encephalopathy, hyperammonemia, and other clinicopathological abnormalities), imaging findings (e.g., acquired portosystemic shunts, and reduced hepatopetal, or hepatofugal, portal vein blood flow), and exclusion of other causes (9,10). Definitive identification (and subsequently tailored treatment) of most underlying causes of intrahepatic portal hypertension requires histopathology but there are several reasons why clinicians and owners may elect not to pursue biopsy in these cases. Dogs with intrahepatic portal hypertension are typically considered to be poor anesthetic candidates, with evidence from human medicine supporting this perception (11,12). Affected dogs also commonly display multiple hemostatic abnormalities, with severe changes seen in some dogs with chronic hepatitis and cirrhosis (13–15). Percutaneous liver biopsy is generally not recommended in dogs with ascites and microhepatica due to the increased technical difficulty of the procedure and the effect of ascites on early detection of post-biopsy hemorrhage (16). These factors may also preclude or discourage surgical or laparoscopy-assisted biopsy. Analysis of abdominal effusion, when present, may also contribute little to diagnosis, as the protein content of the fluid may vary depending on the degree of capillarization of the hepatic sinusoids (3). These considerations prevent a definitive histopathological diagnosis being made in many cases. As a result, clinicians may be required to treat affected dogs without knowledge of the underlying cause.
Treatment options for portal hypertension are also limited. Many are extrapolated from human medicine, in whom detailed guidelines exist; however, several interventions have not been fully evaluated or are not available for use in dogs (3,17). This contrasts with canine chronic hepatitis, for which multiple studies and consensus guidelines have identified effective management, despite it often remaining idiopathic (9). Treatment of portal hypertension includes addressing the underlying etiology, if possible, and managing its complications. Current recommendations for the treatment of complications of portal hypertension include dietary sodium restriction, aldosterone antagonism/diuresis, paracentesis (if necessary for the management of discomfort secondary to ascites), dietary protein modification, lactulose, antibiotics, and gastroprotectant medications (3,9). Hepatoprotective medications, such as S-adenosylmethionine, vitamin E, and ursodeoxycholate, are often recommended for dogs with hepatic fibrosis (9). Expert opinion advises against the use of the anti-fibrotic drug colchicine, despite its theoretical benefits (5). Surgical attenuation of acquired portosystemic shunts has also failed to show a clinical benefit over medical management and is not recommended (5,18). Studies suggest a role for blockade of the renin-angiotensin-aldosterone system and endothelin-1 in dogs with portal hypertension and/or chronic hepatitis; β blockade, commonly used in human medicine, has so far been ineffective when used in dogs (3,19,20). There is little information on the frequency of use of these treatment options in dogs with portal hypertension.
Finally, there is also a lack of data on factors affecting the prognosis of dogs with intrahepatic portal hypertension. One study of 17 dogs with presinusoidal portal hypertension of varying etiology showed a poor prognosis, but did not identify abnormalities with prognostic significance (21). Ascites, a surrogate marker for the presence of cirrhosis and portal hypertension, has been shown to be a negative indicator of prognosis in dogs with chronic hepatitis; across studies, median survival times of 22.5 ± 15 d and 23 ± 23 d have been reported for dogs with ascites and cirrhosis, respectively, although some dogs display prolonged survival (2,9,22,23). Other markers shown to be of prognostic significance in dogs with chronic hepatitis include anorexia, jaundice, serum bilirubin and albumin concentrations, prothrombin and partial thromboplastin times, the degree of fibrosis, bridging fibrosis and necrosis on hepatic biopsy samples, and treatment with prednisolone or other immunomodulatory drugs, amongst others (9,22–25). All these studies focused primarily on dogs with chronic hepatitis, including only a small number of dogs inferred to have portal hypertension. Furthermore, the significance of clinical findings on the survival of dogs with portal hypertension has not been assessed. As highlighted by several studies, this hinders clinician and owner decision-making, resulting in many dogs being euthanized on diagnosis based on the perception of poor prognosis, rather than due to a lack of treatment options (2,21,22).
The aims of this study were to describe a large population of dogs with portal hypertension secondary to acquired hepatic disease, and to assess the ability of historical, clinical, and clinicopathological findings to predict survival in these dogs. It was hypothesized that dogs with portal hypertension survive longer than previously reported and that results of diagnostic investigations provide prognostic information.
Materials and methods
This was a retrospective, multicenter, observational, cohort study. The digital databases of the referral hospitals at Anderson Moores Veterinary Specialists, UK, North Downs Specialist Referrals, UK, University College Dublin, Ireland, University of Liverpool, UK, and Willows Veterinary Centre and Referral Service, UK, were searched to retrieve the records of all dogs diagnosed with intrahepatic portal hypertension between January 2011 and October 2020. Ethical approval for this study was granted by the Ethics Review Panel of the Royal College of Veterinary Surgeons (RCVS ERP 2019-03-Calleja).
Cases were included if there was evidence of intrahepatic portal hypertension secondary to chronic liver disease (3,9). This was based on a combination of historical and clinical signs, compatible clinicopathological abnormalities (at least 1 of hypoglycemia, hypocholesterolemia, decreased urea concentration, hypoalbuminemia, hyperbilirubinemia, hyperammonemia, increased bile acid concentration (pre- or post-prandial values), or prolonged clotting times), and supportive diagnostic imaging or histopathological findings. Diagnostic imaging (computed tomography/ultrasonography) evidence of portal hypertension and chronic liver disease included either the presence of multiple acquired portosystemic shunts and a small or otherwise abnormally appearing liver, with or without ascites; or the presence of ascites in the absence of another identifiable cause with a small or abnormally appearing liver and a concurrent serum albumin concentration ≥ 15 g/L. Histopathological diagnosis of liver disease was not required for inclusion but was used to support inclusion if other inclusion criteria were not fulfilled. In all cases, diagnosis of intrahepatic portal hypertension was made by or under the guidance of a Board-certified internal medicine specialist.
Dogs with ascites and serum albumin concentration < 15 g/L without concurrent acquired portosystemic shunts, potential pre- or post-hepatic causes of portal hypertension or ascites (e.g., congestive heart failure, obstructive neoplasm/thrombus), and dogs with histologically confirmed primary hypoplasia of the portal vein or other congenital disease were excluded. Dogs with comorbidities not deemed relevant to their presentation and survival were included.
The following data were extracted from medical records, when available: breed, age, gender/neuter status, body weight and body condition score, and clinical signs (Table 1). This included onset and duration of signs prior to first presentation, presence or absence of ascites and fluid classification, serum concentrations/activities of urea, cholesterol, glucose, albumin, bile acids, bilirubin, ammonia, GGT, AST, ALP, and ALT, imaging findings including the presence or absence of acquired portosystemic shunts, cytological and/or histological diagnosis, final diagnosis, date of diagnosis, treatment given, date of death/euthanasia, and cause of death/euthanasia.
Table 1.
Clinical signs reported in 76 dogs with portal hypertension.
| Sign | n | % |
|---|---|---|
| Abdominal distension/ascitesa | 43 | 56.6 |
| Vomiting | 37 | 48.7 |
| Jaundicea | 35 | 46.1 |
| Lethargy/dullness | 27 | 35.5 |
| Anorexia/hyporexia | 24 | 31.6 |
| Melena/hematocheziaa | 14 | 18.4 |
| Polydipsia/polyuria | 14 | 18.4 |
| Mentation/behavior changea | 14 | 18.4 |
| Diarrhea | 8 | 10.5 |
| Weight loss | 7 | 9.2 |
| Ataxiaa | 5 | 6.6 |
| Hematemesisa | 3 | 3.9 |
| Digital dermatitis | 2 | 2.6 |
| Abdominal mass | 2 | 2.6 |
| Pyrexia | 2 | 2.6 |
| Seizurea | 2 | 2.6 |
| Constipation/straining to defecate | 1 | 1.3 |
| Dyspnea/panting | 1 | 1.3 |
| Regurgitation | 1 | 1.3 |
| Discomfort | 1 | 1.3 |
| Cough | 1 | 1.3 |
| Trembling | 1 | 1.3 |
| Hypersalivation | 1 | 1.3 |
| Lameness | 1 | 1.3 |
| Collapse | 1 | 1.3 |
Denotes signs more suggestive of liver disease or portal hypertension.
Data distributions were determined by the D’Agostino-Pearson method. Mean (± standard deviation) was used for parametric continuous data and median (range) for non-parametric continuous and ordinal data. Survival time was calculated from the date of diagnosis until the date of death or euthanasia, reported as median and range and displayed using the Kaplan-Meier method. Dogs were right censored if they were lost to follow-up or still alive at the time of data collection. The following explanatory variables were assessed using univariable Cox proportional hazards models: age, body weight, gender/neuter status, breed (Labrador retriever or other), duration of owner-reported clinical signs (days), body condition score, presence or absence of ascites, jaundice (supported by total bilirubin concentration > 35 μmol/L) (26) or multiple acquired shunts, urea, cholesterol, glucose, and albumin concentrations. Explanatory variables with a P-value ≤ 0.2 on univariable analysis were then examined using multivariable Cox regression. A backward stepwise linear regression was used to identify possible predictor variables, with sequential exclusion of variables between steps based on the highest P-value, and controlling for age at each step. The proportional hazards assumption was evaluated using log-minus-log plots for categorical data, and partial residual plots for both categorical and continuous data. Hazard ratio (HR) with 95% confidence interval (95% CI) were reported. P-values < 0.05 were considered significant. Statistical tests were performed using IBM SPSS (Statistics for Macintosh, Version 24.0, Armonk, New York, USA): IBM Corp and graphs constructed using GraphPad Prism version 8.0.0 for MacOS (GraphPad Software, San Diego, California, USA).
Results
In total, 76 dogs fulfilled the inclusion criteria, 34 (44.7%) of which were male (21 castrated) and 42 (55.3%) of which were female (28 spayed). The mean age was 6.6 y (± 3.3 y), with a range from 0.9 to 12.0 y. Breeds included Labrador retriever (n = 16), cocker spaniel (n = 7), English springer spaniel (n = 4), golden retriever (n = 3), West Highland white terrier (n = 2), Cavalier King Charles spaniel (n = 2), greyhound (n = 2), German shepherd (n = 2), 24 other pure breeds (n = 1 each) and crossbreed (n = 14) dogs. Body weight was recorded in 73 dogs with a mean of 23.1 (± 12.6) kg. The body condition score was available for 60 dogs with a median of 4/9 (1–9).
The median duration of owner-reported clinical signs was 21 d (range: 1 to 360 d). Clinical signs were variable and included abdominal distension or ascites (the latter confirmed on diagnostic imaging), vomiting, icterus, lethargy, or decreased appetite in over 25% of cases (Table 1). Neurological signs (mentation/behavior change, ataxia, seizure) were reported in 21/76 dogs (27.6%), with mentation/behavior change the most frequently reported neurological sign.
Albumin, cholesterol, urea, and glucose concentrations were below the lower limit of the reference range in 48/75 (64.0%), 41/74 (55.4%), 40/76 (52.6%), and 1/71 (1.4%) dogs, respectively. Bilirubin concentration was above the reference interval in 42/75 (56.0%) dogs, and > 35 μmol/L in 23 (30.7%) cases. Hyperammonemia (> 50 μmol/L) was identified in 9 (81.8%) of the 11 dogs in which it was measured. Bile acid concentrations were increased in all 30 dogs in which it was measured (excluding dogs with concurrent hyperbilirubinemia). Ascites was present in 69 dogs and analysis was performed in 33 of these; it was characterized as a pure or modified transudate in 32/69 (46.4%) dogs and as a sterile neutrophilic exudate in 1 (1.4%) dog. In the latter dog, although a neutrophilic exudate is not typical for chronic liver disease, all other findings were consistent with liver disease and the dog responded well to treatment for chronic hepatitis with prednisolone, lactulose, prescription liver diet, spironolactone, short-term furosemide for refractory ascites, and omeprazole; this dog was still alive at the censor date (439 d after onset of clinical signs).
Ultrasonography was the most frequently used imaging modality (66/76, 86.8%). Computed tomography (CT) was performed in 16 (21.1%) dogs, 6 (7.9%) of which also underwent ultrasonography. Acquired portosystemic shunts were identified in 46/76 (60.5%) dogs, 33 (71.7%) of which were detected sonographically and 13 (28.3%) using computed tomography. Computed tomography and ultrasonography showed disagreement in 3 dogs, in which multiple acquired portosystemic shunts were identified using CT but not sonography. Computed tomography was not performed in any dog in which acquired portosystemic shunts were identified sonographically. Other hepatic abnormalities frequently identified on sonography and CT included microhepatica (n = 47, 61.8%), nodular change (n = 33, 43.4%), irregular margins (n = 27, 35.5%), and parenchymal heterogeneity (n = 24, 31.6%). The liver was described as sonographically normal in 3 (3.9%) dogs; of these, microhepatica and multiple acquired shunts were identified on subsequent computed tomography in 1 dog, whereas the other 2 dogs had histological diagnoses of chronic hepatitis with fibrosis; therefore, all 3 still fulfilled the inclusion criteria. Biopsies were performed in 28 (36.8%) dogs. Histopathological diagnosis was available for 27 dogs, 25 (92.6%) of which showed chronic hepatitis and fibrosis/cirrhosis; in 4 of these, hepatitis was copper-associated. Differentiation between primary and secondary copper accumulation could not confidently be made in all 4 cases; however, based on the available data for these cases, 1 case was primary, and 2 cases involved significant inflammation with associated copper. The final case was a Bedlington terrier with known COMMD1 mutation and severe hepatic changes. In the remaining 2 (7.4%) cases, the histological diagnoses of vacuolar hepatopathy were considered non-representative by the attending clinician.
The following concurrent conditions were identified in included dogs: 2 dogs were diagnosed with hypoadrenocorticism (one typical and one atypical, both well-controlled); 1 dog with lymphoplasmacytic enteritis; 1 dog was receiving metronomic chemotherapy (cyclophosphamide and meloxicam) for a previously diagnosed soft tissue sarcoma; an incidental lung mass was identified in 1 dog; hypothyroidism and pulmonary hypertension were diagnosed in 1 dog who subsequently died due to severe gastrointestinal hemorrhage secondary to portal hypertension and concurrent glucocorticoid administration; 1 dog was diabetic; acute and chronic pancreatitis were each suspected in 1 dog (based on sonography), without evidence of biliary obstruction; 1 dog, still alive at the censor date (1705 d after diagnosis of portal hypertension), was diagnosed with osteosarcoma; and 1 dog was incidentally diagnosed with a spindle cell tumor over the stifle. One dog with CT findings consistent with cirrhosis and portal hypertension (ascites and acquired portosystemic shunts) also had suspected emphysematous cholecystitis.
The following treatments were prescribed in various combinations: antibiotics (n = 30, 39.5%), spironolactone (n = 29, 38.2%), omeprazole (n = 29, 38.2%), lactulose (n = 27, 35.5%), prednisolone (n = 24, 31.6%), diet formulated for dogs with liver disease (n = 18, 23.7%), furosemide (n = 11, 14.5%), and D-penicillamine (n = 1, 1.3%). Penicillins were the most prescribed antibiotic class, with 21/30 (70%) dogs receiving clavulanate-potentiated amoxicillin, amoxicillin, or ampicillin.
Median survival time was 14 d (range: 0 to 2028 d), with 16/76 (21.1%) dogs euthanized on the day of presentation and an additional 8/76 (10.5%) dogs within the first two days (Figure 1). One dog was euthanized 16 d after initial diagnosis due to acute complications following percutaneous liver biopsy; another dog was euthanized 183 d after diagnosis after developing an intestinal intussusception. All other non-survivors died or were euthanized as a direct result of their liver disease. Eighteen (23.7%) dogs survived longer than 60 d after diagnosis, 8 (44.4%) of which survived longer than 365 d. Ten (13.1%) of the 76 dogs included in this study were still alive after the censor date (survival range: 20 to 2028 d).
Figure 1.
Kaplan-Meier survival curve of dogs with portal hypertension.
Based on univariable Cox regression analysis (Table 2), duration of clinical signs in days (HR = 0.995, 95% CI: 0.99 to 1.0, P = 0.033) was significantly associated with decreased hazard of death (mortality). The presence of clinically appreciable jaundice, supported by a serum bilirubin concentration > 35 μmol/L (HR = 1.846, 95% CI: 1.094 to 3.117, P = 0.022) (Figure 2) was significantly associated with increased hazard of death. Within the multivariable analysis, only jaundice was associated with mortality in the models tested (HR = 1.848, 95% CI: 1.093 to 3.122, P = 0.022) when controlling for age.
Table 2.
Univariate analysis of explanatory variables examined.
| Variable | Hazard ratio | P-value |
|---|---|---|
| Age | 1.000 95% CI (0.994 to 1.005) | 0.952 |
| Body weight | 1.008 95% CI (0.987 to 1.030) | 0.444 |
| Gender | 0.906 95% CI (0.558 to 1.473) | 0.691 |
| Neuter status | 1.034 95% CI (0.628 to 1.702) | 0.895 |
| Labrador/Labrador cross | 1.081 95% CI (0.613 to 1.907) | 0.788 |
| Duration of clinical signsb | 0.995 95% CI (0.990 to 1.000) | 0.033 |
| Body condition score at presentation | 0.985 95% CI (0.814 to 1.193) | 0.880 |
| Jaundicea,b | 1.846 95% CI (1.094 to 3.117) | 0.022 |
| Ascites | 1.871 95% CI (0.743 to 4.712) | 0.177 |
| Presence of acquired shunts | 0.81 95% CI (0.496 to 1.324) | 0.401 |
| Urea | 1.027 95% CI (0.936 to 1.128) | 0.571 |
| Cholesterol | 0.983 95% CI (0.876 to 1.103) | 0.771 |
| Glucose | 1.001 95% CI (0.895 to 1.119) | 0.984 |
| Albumin | 0.961 95% CI (0.914 to 1.010) | 0.115 |
Jaundice defined as total bilirubin concentration > 35 μmol/L.
Statistically significant variables.
Figure 2.
Kaplan-Meier survival curve (solid line) (95% CI, dashed line) of dogs with portal hypertension stratified by the presence or absence of jaundice with total bilirubin concentration > 35 μmol/L.
Twenty-six dogs had clinical signs of more than 30 d duration, 10 (38.5%) of which survived longer than 60 d after diagnosis. Nine dogs (19.6%) of a total of 46 with duration of clinical signs less than 30 d survived longer than 60 d after diagnosis. Of 52 dogs with bilirubin concentration < 35 μmol/L, 17 (32.7%) dogs survived longer than 60 d after diagnosis, whereas of 23 dogs with bilirubin concentration > 35 μmol/L, 3 (13%) dogs survived longer than 60 d after diagnosis.
Discussion
This study shows that, as previously reported, survival of dogs diagnosed with portal hypertension secondary to primary acquired liver disease is poor, with a median survival time of 14 d. Comparatively, prognosis for humans with ascites secondary to cirrhosis is also relatively poor, with 20% of those with ascites surviving less than 1 y (17). However, in this study, almost 1/3 of dogs were euthanized at or within 48 h of diagnosis. This group may have included some dogs euthanized because of perceived poor prognosis without attempting treatment, thereby negatively biasing the results. Almost 1/4 of dogs survived more than 2 mo from diagnosis, and over 10% were alive more than 1 y after diagnosis.
This study identified clinicopathological information which could help guide both clinician and owner decision-making. Dogs with clinical evidence of jaundice experienced shorter survival. This is consistent with studies in both dogs and humans which identified jaundice/hyperbilirubinemia as a negative prognostic marker in patients with liver disease (22,23,27,28). Jaundice may reflect more severe hepatic dysfunction, accounting for decreased survival; severe hyperbilirubinemia can also result in kernicterus, although this was not reported in this population. A total bilirubin concentration > 35 μmol/L was chosen for survival analysis, rather than hyperbilirubinemia per se, as this would provide a clinical parameter for both veterinarians and owners to help guide decision-making. A bilirubin concentration > 35 μmol/L would also be expected to cause clinical evidence of jaundice, although there is significant inter-observer variation (29). Univariable analysis also showed an association between duration of clinical signs and mortality. This was not significant on multivariable analysis, suggesting, at most, a weak association. However, the study may have been underpowered.
The findings of this study suggest that treatment should be considered for all dogs with intrahepatic portal hypertension, rather than euthanasia on diagnosis. Discussions with owners about prognosis should highlight that the current poor survival data may, in part, be due to the large numbers of dogs in which treatment is never attempted, and that a subpopulation of dogs can experience prolonged survival. Particularly, treatment should be considered in those cases in which there is no evidence of jaundice or serum bilirubin concentration is < 35 μmol/L.
In contrast to previous studies (2,9,22,23), ascites was not determined to be prognostic in this study. This could suggest that the primary determinants of survival are hepatic disease severity, response to treatment (when attempted), and portal hypertension, rather than the presence or absence of ascites, which may simply be a marker of portal hypertension. Therefore, when previous studies identified ascites as a negative prognostic marker compared with non-ascitic cases of liver disease, the dogs with ascites could have included cases of portal hypertension, which in turn could have resulted in a worse prognosis. It should be noted that most (60.5%) dogs in this study had multiple acquired portosystemic shunts identified on imaging. These acquired shunting vessels, which form as a direct result of portal hypertension, can act as a pressure release system which could, in combination with lymphatic vessels, minimize the formation of ascites (26). These acquired shunts may be another reason why ascites was seen in most cases but did not appear to be a negative prognostic indicator. Other potential causes for this dissimilarity include the inclusion of only dogs with advanced liver disease (all with an expected shorter survival time) and the potential heterogeneity of the underlying hepatic disease processes in the current study.
Dogs included in this study received various treatments, often in combination. The lack of treatment standardization (e.g., dose, frequency, duration, varying reasons for prescription, likely bias towards using some drugs in dogs experiencing longer survival) precluded examination of treatment effect on survival. Further studies regarding the use of glucocorticoids, spironolactone, and other treatments in similar populations are needed, ideally underpinned by a better understanding of the pathophysiology of portal hypertension.
This study has several limitations. Although this is the largest collection of dogs with portal hypertension to date, the study remains relatively small. Analysis of a larger data set would have increased statistical power and possibly identified other/stronger relationships. The retrospective nature of the study, its focus on the referral population, and lack of standardization of investigations and treatments are further limitations. Computed tomography, which has a higher sensitivity for portosystemic shunts and thrombosis than ultrasound, was performed in a small number of dogs, possibly because of the need for heavy sedation or anesthesia. Follow-up was also lost in a small number (7) of dogs. Finally, although all included dogs were presented and, for non-survivors, euthanized/died because of chronic liver disease and portal hypertension, the presence of comorbidities may have contributed somewhat to the decision not to pursue treatment. However, only a small number of dogs had comorbidities in this population.
The underlying etiology of liver disease and portal hypertension was not confirmed in all cases. The criteria used, particularly the requirement for imaging, excluded dogs with pre- and post-hepatic portal hypertension. Based on the available histological results, the exclusion of dogs with known congenital hepatopathies, and a previous study of liver disease in dogs (7), most dogs included herein were likely to have suffered from an acquired intrahepatic cause of portal hypertension. As previously discussed, chronic hepatitis is reported to be the most common cause of intrahepatic portal hypertension in the UK, whereas developmental anomalies are more common in comparatively younger dogs (3,5–7,18,21,30,31). However, the definitive underlying cause of portal hypertension cannot be ascertained without biopsy. In the current study, liver biopsy was not performed in most dogs. There are several reasons why clinicians may choose not to pursue biopsy including the perception that these dogs are poor candidates for anesthesia and/or biopsy, or cost. Proponents of biopsy argue that a histopathological diagnosis is necessary to optimize treatment, and that this may outweigh the risk in at least some cases. The risks associated with biopsy could not be determined in the current study because the method and timing of biopsy were not always recorded, and the decision to perform biopsy was likely influenced by the perceived prognosis. A prospective, randomized, controlled study would be required to assess the associated risk. Inclusion of dogs without a histopathological diagnosis could also be considered a limitation; however, excluding these dogs would have made the sample population less representative of dogs managed in practice.
Overall, this study shows that, although portal hypertension carries a poor prognosis in most cases, prolonged survival is possible, and this should be considered before euthanasia is recommended. Use of clinical and clinicopathological data, specifically the presence of jaundice, can assist decision-making for clinicians and owners. CVJ
Footnotes
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (hbroughton@cvma-acmv.org) for additional copies or permission to use this material elsewhere.
References
- 1.Sevelius E. Diagnosis and prognosis of chronic hepatitis and cirrhosis in dogs. J Small Anim Pract. 1995;36:521–528. doi: 10.1111/j.1748-5827.1995.tb02801.x. [DOI] [PubMed] [Google Scholar]
- 2.Raffan E, McCallum A, Scase TJ, Watson PJ. Ascites is a negative prognostic indicator in chronic hepatitis in dogs. J Vet Intern Med. 2009;23:63–66. doi: 10.1111/j.1939-1676.2008.0230.x. [DOI] [PubMed] [Google Scholar]
- 3.Buob S, Johnston AN, Webster CR. Portal hypertension: Pathophysiology, diagnosis, and treatment. J Vet Intern Med. 2011;25:169–186. doi: 10.1111/j.1939-1676.2011.00691.x. [DOI] [PubMed] [Google Scholar]
- 4.Routhuizen J World Small Animal Veterinary Association Liver Standardization Group. WSAVA standards for clinical and histological diagnosis of canine and feline liver disease. Edinburgh, UK: Saunders/Elsevier; 2006. p. xi.p. 130. [Google Scholar]
- 5.Eulenberg VM, Lidbury JA. Hepatic fibrosis in dogs. J Vet Intern Med. 2018;32:26–41. doi: 10.1111/jvim.14891. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rothuizen J. Important clinical syndromes associated with liver disease. Vet Clin North Am Small Anim Pract. 2009;39:419–437. doi: 10.1016/j.cvsm.2009.02.007. [DOI] [PubMed] [Google Scholar]
- 7.Watson PJ, Roulois AJ, Scase TJ, Irvine R, Herrtage ME. Prevalence of hepatic lesions at post-mortem examination in dogs and association with pancreatitis. J Small Anim Pract. 2010;51:566–572. doi: 10.1111/j.1748-5827.2010.00996.x. [DOI] [PubMed] [Google Scholar]
- 8.Center SA. Chronic liver disease: Current concepts of disease mechanisms. J Small Anim Pract. 1999;40:106–114. doi: 10.1111/j.1748-5827.1999.tb03050.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Webster CRL, Center SA, Cullen JM, et al. ACVIM consensus statement on the diagnosis and treatment of chronic hepatitis in dogs. J Vet Intern Med. 2019;33:1173–1200. doi: 10.1111/jvim.15467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Toulza O, Center SA, Brooks MB, Erb HN, Warner KL, Deal W. Evaluation of plasma protein C activity for detection of hepatobiliary disease and portosystemic shunting in dogs. J Am Vet Med Assoc. 2006;229:1761–1771. doi: 10.2460/javma.229.11.1761. [DOI] [PubMed] [Google Scholar]
- 11.Csikesz NG, Nguyen LN, Tseng JF, Shah SA. Nationwide volume and mortality after elective surgery in cirrhotic patients. J Am Coll Surg. 2009;208:96–103. doi: 10.1016/j.jamcollsurg.2008.09.006. [DOI] [PubMed] [Google Scholar]
- 12.Kiamanesh D, Rumley J, Moitra VK. Monitoring and managing hepatic disease in anaesthesia. Br J Anaesth. 2013;111(Suppl 1):i50–61. doi: 10.1093/bja/aet378. [DOI] [PubMed] [Google Scholar]
- 13.Webster CR. Hemostatic Disorders Associated with Hepatobiliary Disease. Vet Clin North Am Small Anim Pract. 2017;47:601–615. doi: 10.1016/j.cvsm.2016.11.009. [DOI] [PubMed] [Google Scholar]
- 14.Prins M, Schellens CJ, van Leeuwen MW, Rothuizen J, Teske E. Coagulation disorders in dogs with hepatic disease. Vet J. 2010;185:163–168. doi: 10.1016/j.tvjl.2009.05.009. [DOI] [PubMed] [Google Scholar]
- 15.Reece J, Pavlick M, Penninck DG, Webster CRL. Hemorrhage and complications associated with percutaneous ultrasound guided liver biopsy in dogs. J Vet Intern Med. 2020;34:2398–23404. doi: 10.1111/jvim.15942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Rawlings CA, Howerth EW. Obtaining quality biopsies of the liver and kidney. J Am Anim Hosp Assoc. 2004;40:352–358. doi: 10.5326/0400352. [DOI] [PubMed] [Google Scholar]
- 17.Aithal GP, Palaniyappan N, China L, et al. Guidelines on the management of ascites in cirrhosis. Gut. 2021;70:9–29. doi: 10.1136/gutjnl-2020-321790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Boothe HW, Howe LM, Edwards JF, Slater MR. Multiple extrahepatic portosystemic shunts in dogs: 30 cases (1981–1993) J Am Vet Med Assoc. 1996;208:1849–1854. [PubMed] [Google Scholar]
- 19.Sakamoto Y, Sakai M, Sato K, Watari T. Plasma renin activity and aldosterone concentration in dogs with acquired portosystemic collaterals. J Vet Intern Med. 2020;34:139–144. doi: 10.1111/jvim.15661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sakamoto Y, Sakai M, Watari T. Hepatic and plasma endothelin-1 in dogs with chronic hepatitis. J Vet Intern Med. 2017;31:764–769. doi: 10.1111/jvim.14687. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.James FE, Knowles GW, Mansfield CS, Robertson ID. Ascites due to pre-sinusoidal portal hypertension in dogs: A retrospective analysis of 17 cases. Aust Vet J. 2008;86:180–186. doi: 10.1111/j.1751-0813.2008.00284.x. [DOI] [PubMed] [Google Scholar]
- 22.Gómez Selgas A, Bexfield N, Scase TJ, Holmes MA, Watson P. Total serum bilirubin as a negative prognostic factor in idiopathic canine chronic hepatitis. J Vet Diagn Invest. 2014;26:246–251. doi: 10.1177/1040638713520602. [DOI] [PubMed] [Google Scholar]
- 23.Poldervaart JH, Favier RP, Penning LC, van den Ingh TS, Rothuizen J. Primary hepatitis in dogs: A retrospective review (2002–2006) J Vet Intern Med. 2009;23:72–80. doi: 10.1111/j.1939-1676.2008.0215.x. [DOI] [PubMed] [Google Scholar]
- 24.Shih JL, Keating JH, Freeman LM, Webster CR. Chronic hepatitis in Labrador Retrievers: Clinical presentation and prognostic factors. J Vet Intern Med. 2007;21:33–39. doi: 10.1892/0891-6640(2007)21[33:chilrc]2.0.co;2. [DOI] [PubMed] [Google Scholar]
- 25.Strombeck DR, Miller LM, Harrold D. Effects of corticosteroid treatment on survival time in dogs with chronic hepatitis: 151 cases (1977–1985) J Am Vet Med Assoc. 1988;193:1109–1113. [PubMed] [Google Scholar]
- 26.Ettinger SJ, Feldman EC, Côté E. Textbook of Veterinary Internal Medicine: Diseases of the Dog and the Cat. 8th ed. St. Louis, Missouri: Elsevier; 2017. p. 2181. [Google Scholar]
- 27.Kilpatrick S, Dreistadt M, Frowde P, et al. Presence of systemic inflammatory response syndrome predicts a poor clinical outcome in dogs with a primary hepatitis. PLoS One. 2016;11:e0146560. doi: 10.1371/journal.pone.0146560. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kamath PS, Wiesner RH, Malinchoc M, et al. A model to predict survival in patients with end-stage liver disease. Hepatology. 2001;33:464–470. doi: 10.1053/jhep.2001.22172. [DOI] [PubMed] [Google Scholar]
- 29.Ruiz MA, Saab S, Rickman LS. The clinical detection of scleral icterus: Observations of multiple examiners. Mil Med. 1997;162:560–563. [PubMed] [Google Scholar]
- 30.Bunch SE, Johnson SE, Cullen JM. Idiopathic noncirrhotic portal hypertension in dogs: 33 cases (1982–1998) J Am Vet Med Assoc. 2001;218:392–399. doi: 10.2460/javma.2001.218.392. [DOI] [PubMed] [Google Scholar]
- 31.Van den Ingh TS, Rothuizen J, Meyer HP. Portal hypertension associated with primary hypoplasia of the hepatic portal vein in dogs. Vet Rec. 1995;137:424–427. doi: 10.1136/vr.137.17.424. [DOI] [PubMed] [Google Scholar]