PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of cjvetresCVMACanadian Journal of Veterinary ResearchSee also Canadian Journal of Comparative MedicineJournal Web siteHow to Submit
 
Can J Vet Res. 2016 October; 80(4): 302–308.
PMCID: PMC5052882

Language: English | French

Evaluation of hyaluronic acid, procollagen type III N-terminal peptide, and tissue inhibitor of matrix metalloproteinase-1 as serum markers of canine hepatic fibrosis

Abstract

The only way to diagnose hepatic fibrosis in dogs is by histological assessment of a liver biopsy specimen. As this technique is invasive and susceptible to sampling variation, serum biomarkers are used to detect hepatic fibrosis in humans. The objective of this study was to assess the utility of hyaluronic acid (HA), procollagen type III N-terminal peptide (PIIINP), and tissue inhibitor of matrix metalloproteinase-1 (TIMP-1) as serum markers of canine hepatic fibrosis. Serum samples were collected from 47 dogs with histologically confirmed hepatobiliary disease and 24 healthy dogs in order to measure concentrations of HA, PIIINP, and TIMP-1. Hepatic fibrosis was staged using a 5-point scoring scheme. There was no correlation between serum concentrations of HA or PIIINP and the severity of hepatic fibrosis. There was a negative correlation between serum concentration of TIMP-1 and the severity of hepatic fibrosis (rs = −0.33; P = 0.036). It was not possible to use serum concentrations of HA, PIIINP, or TIMP-1 to discriminate between dogs with absent-to-moderate hepatic fibrosis and those with marked-to-very-marked fibrosis. The results of this study do not support the utility of measuring serum concentrations of HA, PIIINP, or TIMP-1 for diagnosing canine hepatic fibrosis. Further studies are needed to support this finding.

Résumé

Le seul moyen de diagnostiquer la fibrose hépatique chez les chiens est par évaluation histologique d’un échantillon de biopsie hépatique. Étant donné que cette technique est invasive et sujette à variation dans l’échantillonnage, chez l’humain des marqueurs sériques sont utilisés pour détecter la fibrose hépatique. L’objectif de la présente étude était d’évaluer l’utilité de l’acide hyaluronique (AH), du peptide N-terminal du pro-collagène de type III (PNPIII), et de l’inhibiteur tissulaire de la métalloprotéinase-1 matricielle (ITMP-1) comme marqueurs sériques de la fibrose hépatique canine. Des échantillons de sérums ont été récoltés de 47 chiens avec une pathologie hépatobiliaire confirmée histologiquement et 24 chiens en santé afin de mesurer les concentrations d’AH, PNPIII, et ITMP-1. La fibrose hépatique a été catégorisée en utilisant un schéma de pointage en cinq points. Il n’y avait pas de corrélation entre les concentrations sériques d’AH ou de PNPIII et la sévérité de fibrose hépatique. Il y avait une corrélation négative entre la concentration sérique d’ITMP-1 et la sévérité de la fibrose hépatique (rs = −0,33; P = 0,036). Il n’était pas possible d’utiliser les concentrations sériques d’AH, de PNPIII, ou d’ITMP-1 afin de différencier les chiens avec une fibrose hépatique absente à modérée de ceux avec une fibrose marquée à très marquée. Les résultats de la présente étude ne permettent pas de justifier l’utilité de mesurer les concentrations sériques d’AH, de PNPIII, ou d’ITMP-1 pour le diagnostic de la fibrose hépatique canine. Des études supplémentaires sont requises pour soutenir cette trouvaille.

(Traduit par Docteur Serge Messier)

Introduction

The development of hepatic fibrosis is an important event in the progression of liver disease and has been shown to be a negative prognostic factor in humans with chronic hepatitis (1). Currently, the only way to diagnose hepatic fibrosis in dogs is by histological assessment of a liver biopsy specimen. Liver biopsy is invasive, there is a risk of hemorrhage, and as only a small portion of the organ is evaluated (2); this technique is susceptible to sampling variation (3,4). Because of these disadvantages, serum markers of hepatic fibrosis have been developed for use in humans (5).

In human patients with hepatic fibrosis, an increased rate of extracellular matrix turnover results in matrix components being released into the bloodstream (5). Hyaluronic acid (HA) is a highly evolutionarily conserved glycosaminoglycan component of the extracellular matrix (6) that is used as a serum biomarker of hepatic fibrosis (5). In a study of humans with chronic hepatitis-C, the sensitivity and specificity of HA for distinguishing between patients with extensive fibrosis and those with milder fibrosis were 86% and 70%, respectively (7). It has previously been shown that HA increased in dogs with hepatic disease, including cirrhosis (8,9) and congenital portosystemic shunts (CPSS) (10).

In a study of human patients with chronic hepatitis-C, measurement of serum concentrations of procollagen type III N-terminal peptide (PIIINP) had a sensitivity of 92% and a specificity of 76% for differentiating between patients with extensive fibrosis and those with milder fibrosis (7). Serum concentrations of PIIINP were increased in growing dogs, but not in dogs with chronic bronchopulmonary disease (11). Serum concentrations of PIIINP have not previously been assessed in dogs with hepatobiliary disease.

Tissue inhibitor of matrix metalloproteinase-1 (TIMP-1) is a protein that inhibits the action of matrix metalloproteinase-1, thus slowing the degradation of extracellular matrix during fibrosis (12). In a study of human patients with chronic hepatitis-C, measurement of serum concentration of TIMP-1 had a sensitivity of 75% and a specificity of 70% for differentiating between patients with extensive fibrosis and those with milder fibrosis (7). To the authors’ knowledge, serum concentrations of TIMP-1 have not previously been assessed in dogs with hepatobiliary disease.

We hypothesized that serum concentrations of HA, PIIINP, and TIMP-1 would be positively correlated with the histological stage of hepatic fibrosis and would allow the discrimination between dogs with absent-to-moderate hepatic fibrosis and those with marked-to-very-marked fibrosis. The main objectives of this initial exploratory study were to determine if there is a correlation between serum concentrations of these analytes and the histological severity of hepatic fibrosis and to determine if these markers allow the discrimination between dogs with absent-to-moderate hepatic fibrosis and those with marked-to-very-marked fibrosis. The secondary objectives of the study were to compare serum concentrations of HA, PIIINP, and TIMP-1 among dogs with different types of hepatobiliary disease and healthy dogs, as well as to partially validate a nonspecies-specific serum HA assay for use with canine serum.

Materials and methods

Animals

Dogs over 1 y of age (over 2 y for large and giant breeds) with histologically confirmed hepatobiliary disease diagnosed at Gulf Coast Veterinary Specialists or Texas A&M University Veterinary Medical Teaching Hospital from March 1, 2011 to February 28, 2013 were enrolled in this prospective observational study. This age cutoff was used to avoid enrolling growing dogs. Diagnosis of hepatobiliary disease was based on a combination of clinical signs, laboratory testing, diagnostic imaging findings, histological evaluation of a liver biopsy specimen, and in some cases, findings after surgical exploration of the abdominal cavity.

These dogs were divided into 4 groups: i) chronic hepatitis; ii) hepatic neoplasia, which could be primary or secondary; iii) congenital portosystemic shunt (CPSS); and iv) other hepatobiliary diseases, including vacuolar hepatopathy, nodular regeneration, and gallbladder mucocele. When possible, an extra liver biopsy specimen was collected from the dogs with hepatobiliary disease for evaluating the severity of fibrosis.

Healthy staff-owned dogs over 1 y of age (over 2 y of age for large and giant breeds) were enrolled at the Texas A&M University Veterinary Medical Teaching Hospital. The health of these dogs was assessed by an owner questionnaire, physical examination, complete blood (cell) count (CBC), serum biochemistry profile, and measurement of serum concentration of pancreas-specific lipase (Spec cPL Test; IDEXX Laboratories, Westbrook, Maine, USA). Dogs with clinically important abnormalities were excluded from the study.

The study was approved by the Texas A&M University Institutional Animal Use Committee. Informed owner consent was given before the dogs were enrolled in the study.

Serum samples

At the time of liver biopsy or immediately before euthanasia and necropsy, 3 to 5 mL of blood were collected from the dogs by jugular venipuncture and placed into sterile anticoagulant-free tubes. Once a firm blood clot had formed, the blood was centrifuged at 1300 × g for 15 min to separate the serum from the red blood cells. The serum was stored at −80°C until analysis.

Assays

The assays were conducted in batches containing samples from healthy and diseased dogs. Samples were run in duplicate using materials, including standards and quality controls, provided by the respective manufacturers according to their instructions. Serum concentrations of HA were measured with a commercially available enzyme-linked immunosorbent assay (ELISA) marketed for use in humans and other animals (Hyaluronan ELISA Kit; Echelon Biosciences, Salt Lake City, Utah, USA). This assay has previously been used for measuring HA in canine serum (1316). Assay precision was assessed by calculating the intra-assay coefficients of variation (%CV) for 3 samples (low, medium, and high concentration) run 6 times on the same ELISA plate. Repeatability was assessed by calculating the inter-assay %CV for 3 samples (low, medium, and high concentration) run 7 times on different days. We assessed analytical accuracy by calculating observed-to-expected ratios (%) when equal volumes of 2 of 4 canine serum samples were mixed in every possible combination. Assay linearity was assessed by calculating observed-to-expected ratios (%) when 5 samples were serially diluted, 1:2, 1:4, 1:8, and 1:16, with sample buffer. Serum concentrations of PIIINP were measured with a human radioimmunoassay (UniQ PIIINP RIA; Orion Diagnostica, Espoo, Finland) that has previously been validated for use with canine serum (11). Serum concentrations of TIMP-1 were measured using a commercially available canine ELISA (TIMP-1 ELISA; USCN Life Science, China) that the manufacturer has validated for use with canine serum.

Histological assessment

Liver biopsies were fixed in neutral buffered formalin, processed for routine histology, and embedded in paraffin. Serial sections of liver were stained with hematoxylin and eosin and picrosirius red (Picrosirius Red Stain Kit; Polysciences, Warrington, Pennsylvania, USA). Hepatic fibrosis was staged by a board-certified veterinary anatomic pathologist (ARH) using a 5-point scoring scheme, i.e., 0 — absent, 1 — mild, 2 — moderate, 3 — marked, and 4 — very marked. This scheme is an adaptation of the Ishak scheme and was devised by Drs. van den Ingh, Grinwis, and Rothuizen at the University of Utrecht and has previously been used to assess hepatic fibrosis in dogs (17). Information about the clinical history or serum concentrations of extracellular matrix components was not provided to the pathologist during the scoring process.

Statistical methods

The distribution of continuous data was assessed using the Kolmogorov-Smirnov test and by visual inspection of frequency histograms. Non-parametric data were expressed as the median (minimum–maximum). Serum concentrations of HA, PIIINP, and HA were compared among the groups of dogs using the Kruskal-Wallis test, followed by post-testing with Dunn’s test as needed. Comparisons of continuous variables between 2 groups of dogs were made using Mann-Whitney U-tests. The relationships between serum HA, PIIINP, and TIMP-1 concentrations and the hepatic fibrosis stage were assessed using Spearman’s rank correlation (rs). A statistical software package was used for these calculations (GraphPad Prism 5; GraphPad Software, LaJolla, California, USA). Statistical significance was set as P < 0.05.

Results

Forty-seven dogs with hepatobiliary disease were enrolled in the study. The median age of the dogs was 10 y (range: 1 to 17 y). Twenty were neutered males (43%), 3 were intact males (6%), 22 were spayed females (47%), and 2 were intact females (4%). The following breeds were represented: 3 Labrador retrievers (6%), 3 miniature schnauzers (6%), 3 Yorkshire terriers (7%), 2 Chihuahuas (4%), and 16 mixed-breed dogs (32%). Twenty dogs (43%) had chronic hepatitis, 17 (36%) had hepatobiliary neoplasia (13 with hepatocellular carcinoma/adenoma, 2 with hemangiosarcoma, 1 with cholangiocarcinoma, and 1 with lymphoma), 4 dogs (9%) had CPSS, and 6 (13%) had other hepatobiliary disease (2 dogs with gall bladder mucoceles, 2 with nodular hyperplasia, and 1 with portal vein hypoplasia and portal hypertension, and 1 with acute hepatitis). Twenty-four healthy dogs were enrolled in the study. The median age of these dogs was 3 y (range: 1 to 13 y). Eight were neutered males (33%), 11 were spayed females (46%), and 5 were intact females (21%). The following breeds were represented: 3 miniature schnauzers (13%), 2 German shepherds (8%), 2 hounds (8%), 2 Australian shepherds (8%), and 3 mixed-breed dogs (13%).

All the hepatic samples were deemed to be adequate for evaluation. Hepatic fibrosis stage scores were assigned to the dogs as follows: 6 dogs (13%) had no fibrosis, 10 dogs (21%) had mild, 14 dogs (30%) had moderate, 8 dogs (17%) had marked, and 9 dogs (19%) had very marked fibrosis.

The intra-assay %CV for the HA ELISA was 6.2%, 1.9%, and 12.2% for low, medium, and high concentration samples, respectively. The inter-assay variability %CV for the assay was 15.1%, 13.4%, and 15.3% for low, medium, and high concentration samples, respectively. The mean (± standard deviation; minimum–maximum) observed-to-expected ratio for spiking recovery experiments was 97.2% (± 7.2%; range: 89.5% to 110.9%). The mean observed-to-expected ratio for the dilutional parallelism experiments was 96.0% (± 16.9%; range: 75.4% to 129.3%).

Serum concentrations of HA were measured in 45 dogs with hepatobiliary disease (96%) and in 24 healthy dogs (100%). There was no correlation between serum concentration of HA and the severity of hepatic fibrosis (rs = 0.24; P = 0.114; Figure 1). There was no significant difference in serum concentrations of HA between dogs with absent-to-moderate hepatic fibrosis and those with marked-to-very-marked fibrosis (P = 0.598; Table I). Serum concentrations of HA were higher in healthy dogs (median: 201 ng/mL; range: 84 to 1464 mg/mL) than in dogs with neoplasia (median: 126 ng/mL; range: 82 to 1532 ng/mL; P < 0.05; Table I). There were no other significant differences in serum concentrations of HA among the groups of dogs (Table II).

Figure 1
There was no correlation between serum concentrations of hyaluronic acid (HA) and the stages of hepatic fibrosis in dogs (rs = 0.24; P = 0.114).
Table I
Serum concentrations of extracellular matrix components in dogs at different stages of hepatic fibrosis
Table II
Serum concentrations of extracellular matrix components in healthy dogs and dogs with various types of hepatobiliary disease

Serum concentrations of PIIINP were measured in 39 dogs with hepatobiliary disease (83%) and in 24 healthy dogs (100%). There was no correlation between serum concentration of PIIINP and the severity of hepatic fibrosis (rs = 0.12; P = 0.472; Figure 2). There was no significant difference in serum concentrations of PIIINP between dogs with absent-to-moderate hepatic fibrosis and those with marked-to-very-marked fibrosis (Table II; P = 0.707). There were no significant differences in serum concentrations of PIIINP among the groups of dogs (P = 0.440; Table I).

Figure 2
There was no correlation between serum concentrations of procollagen type III N-terminal peptide (PIIINP) and the stages of hepatic fibrosis in dogs (rs = 0.12; P = 0.472).

Serum concentrations of TIMP-1 were measured in 41 dogs with hepatobiliary disease (87%) and in 24 healthy dogs (100%). There was a negative correlation between serum concentration of TIMP-1 and the severity of hepatic fibrosis (rs = −0.33; P = 0.036; Figure 3). There was no significant difference in serum concentrations of TIMP-1 between dogs with absent-to-moderate hepatic fibrosis and those with marked-to-very-marked fibrosis (P = 0.124; Table I). There were no significant differences in serum concentrations of TIMP-1 among the groups of dogs (P = 0.139; Table II; Figure 4). Post-hoc analysis revealed that there was no significant difference in serum concentration of TIMP-1 between dogs with hepatic neoplasia (median: 45 ng/mL; range: 6 to 100 ng/mL) and dogs with non-neoplastic hepatobiliary disease (median: 22 ng/mL; range: 5 to 156 ng/mL; P < 0.1 but > 0.05). Post-hoc analysis showed that dogs with primary hepatobiliary neoplasia (hepatocellular adenomas, hepatocellular carcinomas, and cholangiocarcinomas) had higher serum concentrations of TIMP-1 than those with tumors and those with neoplasia secondarily affecting the liver (lymphoma and hemangiosarcoma), with median concentrations of 47 ng/mL (range: 25 to 100 ng/mL) and 6 ng/mL (range: 6 to 12 mg/mL), respectively (P = 0.0115).

Figure 3
There was a weak negative correlation between serum concentrations of tissue inhibitor of matrix metalloproteinase-1 (TIMP-1) and the stages of hepatic fibrosis in dogs (rs = −0.33; P = 0.036).
Figure 4
Serum concentrations of tissue inhibitor of matrix metalloproteinase-1 (TIMP-1) in healthy dogs and dogs with hepatobiliary disease. The solid horizontal bars represent the median serum concentration of TIMP-1 for that group. For the group of dogs with ...

Discussion

The HA ELISA that we used in this study had acceptable precision, accuracy, and linearity for measuring canine serum concentrations of HA. The inter-assay %CVs of 15.1%, 13.4%, and 15.3% were slightly higher than desirable, which suggests suboptimal repeatability. We do not think this is likely to have affected our conclusions, however, as there was no trend for serum concentrations of HA to be related to the severity of hepatic fibrosis. The relevance of the finding that dogs with hepatic neoplasia had lower serum concentrations of HA than healthy dogs is unknown and neither is the reason that this occurred. There were only small differences in the median concentration of HA between the groups, however, and it may therefore not be clinically important. We are not aware of a similar relationship in humans with hepatic neoplasia. There was no relationship between the severity of hepatic fibrosis and serum concentrations of HA in the dogs enrolled in this study. This is in contrast to studies in humans in whom HA has been shown to be a useful marker of hepatic fibrosis (5,7) and a previous study in dogs in which those with hepatic cirrhosis had higher serum concentrations of HA than those with non-cirrhotic hepatic disease, those with non-hepatic disease, or healthy dogs (8).

Another study also showed that serum concentrations of HA were higher in dogs with advanced hepatic fibrosis than in those with mild fibrosis (9). The aforementioned studies in dogs used a different assay than the one we used. Previous studies using the ELISA that we used in our study, however, found that Chinese Shar pei dogs with cutaneous mucinosis have higher serum concentrations of HA than healthy dogs or Chinese Shar peis without cutaneous mucinosis (1316), which suggests that this assay is capable of detecting elevated serum concentrations of HA in dogs. The median serum concentrations of HA in 5 healthy dogs reported in 1 of these studies was 244 ng/mL (range: 166 to 302 ng/mL), which is similar to that of the 23 healthy dogs from our study with 201 ng/mL (range: 84 to 1464 ng/mL) (15). Another possible reason for the discrepancy in results between our study and these previous studies on dogs with hepatic disease (8,9) is that our study contained relatively few dogs with marked or very marked hepatic fibrosis and potentially serum concentrations of HA only increase in dogs with very marked fibrosis (n = 9).

There was no difference in serum concentrations of PIIINP among healthy dogs, those with chronic hepatitis, CPSS, hepatobiliary neoplasia, or other hepatobiliary disease. Additionally, no relationship was observed between the severity of hepatic fibrosis and serum concentrations of PIIINP. In humans, serum concentrations of PIIINP have been shown to be useful in distinguishing between patients with no or mild fibrosis and those with moderate or severe fibrosis (5,7). It is not known why serum concentrations of PIIINP were not increased in these dogs with hepatic fibrosis. One possible explanation is that this protein is not leaked into the bloodstream of dogs with hepatic fibrosis. Type III collagen was increased in the livers of dogs with chronic hepatitis (18), which suggests that this form of collagen is important in canine hepatic fibrosis. It is interesting that a previous study did not find that serum concentrations of PIIINP were increased in dogs with chronic bronchopulmonary disease, but that concentrations in bronchoalveolar lavage fluid were increased (11). In humans, PIIINP has also been shown to have a moderately strong positive correlation with the severity grade of hepatic necroinflammatory activity (19). We therefore cannot rule out that this was a confounding factor that obscured the presence of a true relationship between PIIINP and the severity of hepatic fibrosis in our study.

Upon visual inspection of the data, the median serum concentration of TIMP-1 in dogs with hepatic neoplasia (median: 45 ng/mL; range: 6 to 100 ng/mL) was higher than that of healthy dogs (median: 20; range: 5 to 100 ng/mL) or a combined group of dogs with non-neoplastic hepatobiliary disease formed post-hoc (median: 22; range: 5 to 156 ng/mL), but the differences were not statistically significant. This was possibly due to type-II error, as the sample size in each group was relatively small. Additionally, dogs with primary liver neoplasia (hepatocellular adenoma/carcinoma and cholangiocarcinoma) had significantly higher serum concentrations of TIMP-1 than those with lymphoma and hemangiosarcoma.

Humans with a variety of tumors, including hepatic metastases (20) and pulmonary neoplasms (21), have been shown to have increased serum concentrations of TIMP-1. In 1 study of humans with hepatic metastases, higher serum concentrations of TIMP-1 were shown to be a poor prognostic factor (20). Dogs with spontaneously occurring mammary tumors have been shown to have a relatively low tissue TIMP-1 activity compared to rats with induced mammary tumors (22), but to our knowledge, serum concentrations of TIMP-1 have not previously been reported in dogs with hepatobiliary neoplasia. It is important to emphasize, however, that it was not hypothesized a priori that serum concentrations of TIMP-1 would be higher in dogs with primary hepatic neoplasia than in metastatic tumors. This means that we cannot draw any definitive conclusions from this finding. Further studies in a larger group of dogs with hepatic neoplasia are needed to confirm or refute our findings and to assess serum concentrations of TIMP-1 in dogs with other types of neoplasia.

There was a weak negative correlation between serum concentration of TIMP-1 and the severity of fibrosis. As there was no significant difference in the fibrosis stages assigned to dogs with hepatobiliary neoplasia and those with other hepatobiliary diseases, it is unlikely that the non-significant trend for increased serum concentrations of TIMP-1 observed in dogs with neoplasia acted as a confounding factor, which explains this unexpected negative correlation. There was no difference in serum concentrations of TIMP-1 between dogs with absent-to-moderate fibrosis and those with marked-to-very-marked fibrosis. These findings contradict those in studies of humans in which TIMP-1 is a useful marker of hepatic fibrosis and is positively correlated with its severity (5,7). The reason for this difference between the 2 species is not known.

It is important to discuss several limitations of this work. Firstly, as there were relatively few dogs (9 out of 48; 19%) that were assigned a score of very marked hepatic fibrosis, it is possible that this was why we failed to find a relationship between any of the extracellular matrix components and the severity of hepatic fibrosis because they are only increased in dogs with very marked fibrosis. Even if this was the case, however, the utility of these markers would be limited if they can only distinguish between dogs with mild fibrosis and those with very marked fibrosis. The sample sizes in some of the groups were small and this may have led to type-II error when evaluating differences in the analytes between these groups. Additionally, we cannot completely exclude the possibility that some of the dogs that were enrolled in this study had subclinical disease that was causing fibrosis of another organ, therefore increasing their serum concentration of extracellular matrix components and acting as a confounding factor. The group of dogs with hepatobiliary disease in this study, however, would be similar to the population of dogs in which these markers would be used if they were shown to be valuable. It is therefore important to recognize their lack of apparent diagnostic utility in this population.

In conclusion, serum concentrations of HA, PIIINP, and TIMP-1 were not able to discriminate between dogs with and without hepatic fibrosis. Therefore, the results of this study do not support the utility of measuring serum concentrations of HA, PIIINP, or TIMP-1 for diagnosing canine hepatic fibrosis. Further studies are needed to confirm this finding.

Footnotes

Preliminary results of this study were presented as abstracts at the 2013 European College of Veterinary Internal Medicine Conference in Liverpool, September 4 to 6, 2013 and at the 2014 American College of Veterinary Internal Medicine Forum in Nashville, June 5 to 7, 2014.

Some of the dogs enrolled in this study were part of a separate study of novel inflammatory markers of hepatic disease, the manuscript of which is currently under review with another journal. Craig SM, Fry JK, Rodrigues Hoffmann A, Manino P, Heilmann RM, Suchodolski JS, Steiner JM, Hottinger HA, Hunter SL, Lidbury JA. Serum C-reactive protein and S100A12 concentrations in dogs with hepatic disease. J Small Anim Pract 2016 Jun 7. doi: 10.1111/jsap.12504. [Epub ahead of print] PubMed PMID: 27271454.\This study forms part of Dr. Lidbury’s PhD thesis at the College of Veterinary Medicine & Biomedical Sciences, Texas A&M University, 4474 TAMU, College Station, Texas.

Conflict of interest

None of the authors has any financial or personal relationships that could inappropriately influence or bias the content of this paper.

References

1. Everhart JE, Wright EC, Goodman ZD, et al. Prognostic value of Ishak fibrosis stage: Findings from the hepatitis C antiviral long-term treatment against cirrhosis trial. Hepatology. 2010;51:585–594. [PMC free article] [PubMed]
2. Petre SL, McClaran JK, Bergman PJ, Monette S. Safety and efficacy of laparoscopic hepatic biopsy in dogs: 80 cases (2004–2009) J Am Vet Med Assoc. 2012;240:181–185. [PubMed]
3. Bedossa P, Dargere D, Paradis V. Sampling variability of liver fibrosis in chronic hepatitis C. Hepatology. 2003;38:1449–1457. [PubMed]
4. Kemp SD, Zimmerman KL, Panciera DL, Monroe WE, Leib MS. Histopathologic variation between liver lobes in dogs. J Vet Intern Med. 2015;29:58–62. [PMC free article] [PubMed]
5. Baranova A, Lal P, Birerdinc A, Younossi ZM. Non-invasive markers for hepatic fibrosis. BMC Gastroenterol. 2011;11:91. [PMC free article] [PubMed]
6. Chen WY, Abatangelo G. Functions of hyaluronan in wound repair. Wound Repair Regen. 1999;7:79–89. [PubMed]
7. Leroy V, Monier F, Bottari S, et al. Circulating matrix metalloproteinases 1, 2, 9 and their inhibitors TIMP-1 and TIMP-2 as serum markers of liver fibrosis in patients with chronic hepatitis C: Comparison with PIIINP and hyaluronic acid. Am J Gastroenterol. 2004;99:271–279. [PubMed]
8. Kanemoto H, Ohno K, Sakai M, et al. Blood hyaluronic acid as a marker for canine cirrhosis. J Vet Med Sci. 2009;71:1251–1254. [PubMed]
9. Glińska-Suchocka K, Orlowska A, Spużak J, Jankowski M, Kubiak K. Suitability of using serum hialuronic acid concentrations in the diagnosis of canine liver fibrosis. Pol J Vet Sci. 2015;18:873–878. [PubMed]
10. Seki M, Asano K, Sakai M, et al. Serum hyaluronic acid in dogs with congenital portosystemic shunts. J Small Anim Pract. 2010;51:260–263. [PubMed]
11. Schuller S, Valentin S, Remy B, et al. Analytical, physiologic, and clinical validation of a radioimmunoassay for measurement of procollagen type III amino terminal propeptide in serum and bronchoalveolar lavage fluid obtained from dogs. Am J Vet Res. 2006;67:749–755. [PubMed]
12. Arthur MJ, Mann DA, Iredale JP. Tissue inhibitors of metalloproteinases, hepatic stellate cells and liver fibrosis. J Gastroenterol Hepatol. 1998;13( Suppl):S33–38. [PubMed]
13. Grützner N, Heilmann RM, Cranford SM, Holzenburg A, Suchodolski JS, Steiner JM. Inflammatory, immunological, and intestinal disease biomarkers in Chinese Shar-Pei dogs with marked hypocobalaminemia. J Vet Diagn Invest. 2015;27:31–40. [PubMed]
14. Olsson M, Meadows JR, Truvé K, et al. A novel unstable duplication upstream of HAS2 predisposes to a breed-defining skin phenotype and a periodic fever syndrome in Chinese Shar-Pei dogs. PLoS Genet. 2011;7:e1001332. [PMC free article] [PubMed]
15. Zanna G, Fondevila D, Bardagi M, Docampo MJ, Bassois A, Ferrer L. Cutaneous mucinosis in shar-pei dogs is due to hyaluronic acid deposition and is associated with high levels of hyaluronic acid in serum. Vet Dermatol. 2008;19:314–318. [PubMed]
16. Docampo MJ, Zanna G, Fondevila D, et al. Increased HAS2-driven hyaluronic acid synthesis in shar-pei dogs with hereditary cutaneous hyaluronosis (mucinosis) Vet Dermatol. 2011;22:535–545. [PubMed]
17. Fieten H, Biourge VC, Watson AL, Leegwater PA, van den Ingh TS, Rothuizen J. Nutritional management of inherited copper-associated hepatitis in the Labrador retriever. Vet J. 2014;199:429–433. [PubMed]
18. Spee B, Arends B, van den Ingh TS, et al. Transforming growth factor beta-1 signalling in canine hepatic diseases: New models for human fibrotic liver pathologies. Liver Int. 2006;26:716–725. [PubMed]
19. Holoman J, Glasa J, Galbavy S, et al. Serum markers of liver fibrogenesis, and liver histology findings in patients with chronic liver diseases. Bratisl Lek Listy. 2002;103:70–75. [PubMed]
20. Bunatova K, Pesta M, Kulda V, et al. Plasma TIMP1 level is a prognostic factor in patients with liver metastases. Anticancer Res. 2012;32:4601–4606. [PubMed]
21. Ylisirnio S, Hoyhtya M, Turpeenniemi-Hujanen T. Serum matrix metalloproteinases -2, -9 and tissue inhibitors of metalloproteinases -1, -2 in lung cancer-TIMP-1 as a prognostic marker. Anticancer Res. 2000;20:1311–1316. [PubMed]
22. Kawai K, Uetsuka K, Doi K, Nakayama H. The activity of matrix metalloproteinases (MMPS) and tissue inhibitors of metalloproteinases (TIMPs) in mammary tumors of dogs and rats. J Vet Med Sci. 2006;68:105–111. [PubMed]

Articles from Canadian Journal of Veterinary Research are provided here courtesy of Canadian Veterinary Medical Association