Criteria for Selection of Cases

Medical records of the CUHA from 1995 to 2005 were searched by use of a computer database²¹ with application of a series of 31 search strings for all dogs with disorders of extrahepatic biliary structures. Case records were examined for dogs with documented disease of the gallbladder. Twenty-five cases were derived from this search. An additional 13 case records of Shetland Sheepdogs were acquired from consulting veterinarians in North America (ranging from California to Pennsylvania), over the same 10-year interval (1995 to 2005). Criteria for inclusion were availability of routine clinicopathologic test results (hematologic, biochemical, and coagulation tests), availability of ultrasonographic images or reports completed by a board-certified veterinary radiologist, description of the gross appearance of the biliary tract at the time of surgical intervention or necropsy, and information regarding treatment and outcome.

Procedures

Data collection—Recorded information included signalment; clinical signs; results from routine clinicopathologic (including CBC, serum biochemical profile, and urinalysis) and coagulation tests; ultrasonographic findings; surgical interventions (eg, cholecystectomy, cholecystotomy, and cholecystoenterostomy); hepatic and gallbladder histologic findings; presence or absence (defined by surgical or necropsy reports and ultrasonographic images) of gallbladder rupture, bile peritonitis, pancreatitis, and obstruction of the extrahepatic biliary tract; and outcome (ie, survival [alive or dead] determined by status at 14 days after surgery, and cause of death [spontaneous or by euthanasia]) and whether death was related to the gallbladder disease. Ultrasonographic recordings performed at the CUHA were completed by the use of 2-dimensional real-time ultrasonography,^a after withholding food overnight; each study was either completed or reviewed by a single board-certified veterinary radiologist (AEY). Liver size was subjectively graded as normal, large, or small. Gallbladder volume was determined by use of the rotation ellipsoid method ([length × width × depth] × π/6 [π/6 = 0.523]), with dimensional measurements recorded in centimeters.^22–24 The thickness of the gallbladder wall was measured on at least 2 sites on opposing sides of the organ. The intralumenal surface of the gallbladder was evaluated for hypoechoic rim sign and other features (ie, finely striated, stellate central pattern) consistent with a gallbladder mucocele.² Images and ultrasonographic tapes were reviewed to define the presence or absence of features consistent with extrahepatic bile duct obstruction, gallbladder rupture, gravitationally dependent or nondependent echoic gallbladder sludge, and discrete choleliths casting an acoustic shadow. Hepatic parenchymal echogenicity was graded as normal, hypoechoic, or hyperechoic relative to the spleen. The presence of hepatic nodules or other abnormal hepatobiliary features was noted, and lesion size was measured. Dogs managed by primary care veterinarians had ultrasonography reports, images, or both prepared by board-certified veterinary radiologists. Unfortunately, tapes of each dog were not available for measurements as described for CUHA patients. Dynamic ultrasonographic studies^23,b were completed in 3 dogs after withholding food overnight. Gallbladder dimensions were determined before feeding and 15, 30, 45, 60, 90, and 120 minutes after feeding. The percentage reduction in gallbladder volume was calculated and compared with values determined in clinically healthy dogs (expected contraction resulting in at least 20% volume reduction).^23,b

Control hospital population—To determine whether Shetland Sheepdogs have a predisposition for disease involving the gallbladder, the number of ill dogs (dogs evaluated for clinical illnesses excluding dogs evaluated for routine health examinations or vaccinations and all dogs < 2 years of age) with and without gallbladder disease and the number of ill Shetland Sheepdogs with and without gallbladder disease evaluated at the CUHA during the time interval of this retrospective study were enumerated. Sex and reproductive status of ill dogs and ill Shetland Sheepdogs were determined.

Clinicopathologic variables—Hematologic variables included PCV, mean corpuscular volume, mean corpuscular hemoglobin concentration, and total and differential WBC and platelet counts. Blood counts were determined by use of automated procedures,^c and blood smears were individually reviewed by a board-certified veterinary clinical pathologist. The presence of gross lipemia was noted. Serum biochemical tests included concentrations of total protein, albumin, globulin, sodium, potassium, chloride, calcium, phosphate, magnesium, bicarbonate, BUN, creatinine, glucose, cholesterol, triglyceride, and total bilirubin and activities of alanine aminotransferase, aspartate aminotransferase, ALP, gamma glutamyltranspeptidase, amylase, lipase, and creatine kinase. Serum biochemical tests were completed by use of automated systems and reagents.^d Urinalyses included determination of specific gravity and routine dipstick and microscopic sediment evaluations. Coagulation tests included activated partial thromboplastin time and prothrombin time, concentrations of fibrinogen and D-dimers, and activity of antithrombin. The activated partial thromboplastin time, prothrombin time, and fibrinogen tests were performed promptly with citrated plasma and kit procedures with fibrometer end points.^e-i The concentration of D-dimers was determined by use of a latex agglutination test system.^j The antithrombin activity was determined by use of a chromogenic substrate with a spectrophotometric end point.^k

Gross examination of gallbladder and bile—One of the authors (SAC) grossly examined the excised gallbladder and bile from 10 dogs that underwent cholecystectomy. Gross inspection of serosal and luminal gallbladder surfaces was completed specifically noting the presence or absence of wall discoloration (darkened color suggesting necrosis) or wall perforation, abnormal wall thickness (2 mm = normal, < 2 mm = abnormally thin, and > 3 mm = abnormally thick), a tenacious viscoelastic mucin layer (> 3 mm) adherent to the luminal surface, inspissated biliary debris, gritty bile texture, overt choleliths, or hemobilia.

Statistical analysis—Histograms and box-and-whisker plots were used to evaluate data distributions. Most physical and clinicopathologic variables had non-Gaussian distributions and were analyzed by use of nonparametric statistical analyses. Descriptive statistics for these variables are expressed as median values and ranges. A χ² goodness-of-fit test was used to determine whether Shetland Sheepdogs were predisposed to gallbladder disease by comparing the proportion of Shetland Sheepdogs in the general ill-patient population to the proportion of Shetland Sheepdogs in the population of dogs with gallbladder disease. The odds ratio was calculated. A goodness-of-fit 1 × 4 χ² test was used to detect predisposition for sex and reproductive status in all ill Shetland Sheepdogs, compared with all ill dogs evaluated at the CUHA, whereas a sex predisposition in Shetland Sheepdogs with gallbladder disease could only be tested by use of a goodness-of-fit 1 × 2 χ² test. The Wilcoxon rank sum test was used to detect differences in age and clinicopathologic features of Shetland Sheepdogs that survived (14 days after surgery), compared with those that did not survive. A significant value of P ≤ 0.05 was modified with a Bonferroni correction for age, weight, and clinicopathologic variables such that P ≤ 0.002 was considered significant. A 2-tailed P value was used for nonparametric analyses. All statistical analyses were performed by use of a data analysis program.^l

Results

During the 10-year interval (January 1, 1995, to July 31, 2005) of this retrospective study, 29,707 ill dogs ≥ 2 years old, including 485 (1.6%) Shetland Sheepdogs, were evaluated at the CUHA. During that time, 245 dogs were admitted for primary gallbladder disease, including 25 (10.2%) Shetland Sheepdogs. A significantly (P < 0.001) lower percentage of Shetland Sheepdogs was found in the ill hospital population (1.6%), compared with the population admitted for gallbladder disease (10.2%). The odds ratio for Shetland Sheepdogs to develop disease of the gallbladder was 7.2. Therefore, in the hospital patient population, Shetland Sheepdogs appeared to have a predisposition for gallbladder disease.

Of all ill dogs ≥ 2 years of age, 4,059 (13.7%) were sexually intact males, 10,493 (35.3%) were neutered males, 2,795 (9.4%) were sexually intact females, and 12,360 (41.6%) were neutered females. Of the 485 ill Shetland Sheepdogs during this study period, 47 (9.7%) were sexually intact males, 178 (36.7%) were neutered males, 53 (10.9%) were sexually intact females, and 207 (42.7%) were neutered females. Of the 25 Shetland Sheepdogs with gallbladder disease, there were 4 (16%) sexually intact males, 6 (24%) neutered males, 0 sexually intact females, and 15 (60%) neutered females. There were no significant differences regarding sex or reproductive status in ill Shetland Sheepdogs versus Shetland Sheepdogs with gallbladder disease.

Including the case records of 13 Shetland Sheepdogs evaluated in private practices, there were 38 Shetland Sheepdogs with primary gallbladder disease. Affected dogs had a median age of 10.9 years (range, 3.2 to 14.7 years); there was no significant (P = 0.37) difference in median age between dogs that survived (n = 26) and dogs that died (12). Data regarding age, weight, and hematologic and serum biochemical findings were tabulated for Shetland Sheepdogs by survival category (Tables 1 and 2); 3 dogs evaluated because of acute abdominal signs and bile peritonitis underwent emergency surgery and died before completion of routine laboratory assessments. There was wide variability in WBC count and differential cell counts among dogs; a stress leukogram was suggested in many dogs by the presence of lymphopenia and eosinopenia. Nonsurvivors had significantly (P < 0.001) higher WBC and neutrophil counts than survivors. Common serum biochemical abnormalities included high serum activities of transaminases, ALP, and gamma glutamyltranspeptidase, as well as increased concentrations of cholesterol, triglycerides, and total bilirubin. Nonsurvivors also had significantly (P = 0.002) lower concentrations of potassium; 5 of 11 nonsurvivors were hypokalemic, compared with only 3 of 27 (11%) of survivors. Despite inappetence associated with illness, 26 of 37 (70%) dogs were hypercholesterolemic. Hypertriglyceridemia was detected in 8 of 16 dogs tested despite inappetence for > 48 hours. Four of these Shetland Sheepdogs already had been fed fat-restricted diets, yet they still had hypertriglyceridemia. In 1 dog, feeding a high-fat formula diet immediately preceded development of acute abdominal signs and jaundice associated with leakage of a previously subclinical mucocele.

Gallbladder and bile—Gross inspection of the gallbladder removed from 10 dogs disclosed a segmentally darkened wall in the area of rupture in 3 dogs with bile peritonitis. This area corresponded with histologic findings of mural necrosis or infarction. In these areas, the gallbladder wall was subjectively thin. In dogs with gallbladder mucoceles not associated with bile peritonitis, the gallbladder wall was subjectively normal or slightly thick (n = 7 dogs). A tenacious viscoelastic mucin layer was strongly adhered to the interior wall of the gallbladder in 8 dogs with a gallbladder mucocele. One excised gallbladder (with ligated stoma) was examined ultrasonographically before gross examination. The hypoechoic rim sign corresponded with the observed mucin layer. Black choleliths (presumably containing bilirubinate) were confirmed in 3 dogs. Eight of 10 dogs with a gallbladder mucocele had black-green bile with a gritty texture located in the center of the gallbladder lumen. The biochemical composition of biliary material was investigated in one of these dogs, which confirmed the presence of a bilirubinate precipitate, mucin, and calcium carbonate. Hemobilia was identified only in 1 dog that had a necrotic gallbladder wall. In 2 dogs, a firm rubbery matrix filled the gallbladder lumen.

Ultrasonographic imaging—All 25 CUHA dogs were evaluated by use of abdominal ultrasonography. In 44% (11/25) of dogs, gallbladder abnormalities were discovered during abdominal ultrasonography for other primary disease processes, including recurrent pancreatitis (n = 6 dogs), protein-losing nephropathy (2), pituitary-dependent hyperadrenocorticism (2), hypothyroidism (1), pancytopenia (1), and diabetes mellitus associated with superficial necrolytic dermatitis (hepatocutaneous syndrome; 1); 2 dogs had multiple disorders. Sixteen of 25 (64%) dogs had ultrasonographic features consistent with a gallbladder mucocele (striated nongravitationally dependent gallbladder contents, hypoechoic rim sign, and a large gallbladder in the absence of an enlarged extrahepatic bile duct or evidence of intrahepatic bile duct distension).² Five of these 16 dogs had an abnormally thick gallbladder wall ranging from 3 to 5 mm, and 6 had a hypoechoic layer (hypoechoic rim sign) ranging in thickness from 1 to 10 mm adjacent to the inner surface of the gallbladder wall. Hyperechoic nondependent material containing hypoechoic oval foci was imaged within the gallbladder in 9 of 16 dogs. This complex echogenic pattern was interpreted to represent pockets of liquefied biliary secretions or mucin entrapped within tenacious gallbladder bile in the mucocele nidus. In 2 dogs, formation of the classic “kiwi-fruit” gallbladder mucocele pattern was detected on sequential ultrasonographic evaluations completed at 3- and 6-month intervals, respectively. Of the 9 CUHA dogs that lacked ultrasonographic evidence of a definitive gallbladder mucocele, 4 had nondependent echoic gallbladder sludge and 6 had discrete cholecystoliths. Biliary tract rupture was suspected in 20% (5/25) of dogs²; 4 of these had a well-defined gallbladder mucocele on ultrasonographic imaging. Hepatomegaly was suspected in 44% (11/25) of dogs, and hepatic parenchyma was judged to be hyperechoic in 28% (7/25) of dogs; 6 of 11 dogs with hepatomegaly had hyperechoic hepatic parenchyma. One dog had a small but hyperechoic liver; this dog had severe vacuolar hepatopathy and clinical features of liver failure. Sixteen percent (4/25) of dogs had ultrasonographic features consistent with pancreatitis (ie, hypoechoic and irregular pancreas, hyperechoic peripancreatic fat, and focal peripancreatic fluid), whereas 12% (3/25) had discrete hepatic masses ranging in diameter from 0.5 to 3.0 cm, 16% (4/25) had an irregular liver marginal contour (nodular), 24% (6/25) had ultrasonographic features of major bile duct obstruction, and 12% (3/25) had numerous small (2-mm-diameter) hypoechoic foci throughout the hepatic parenchyma (confirmed on histologic evaluation as severe vacuolar hepatopathy consistent with hepatocutaneous syndrome).

Dynamic gallbladder studies were completed in 3 dogs with nongravitationally dependent gallbladder debris but lacking gallbladder distension and the hypoechoic rim sign commonly associated with a mature gallbladder mucocele. Dynamic studies detected failure of the gallbladder to contract in 2 dogs and only minimal contraction (15%) in the third dog.

Eleven of 13 dogs managed by primary care veterinarians were evaluated by use of abdominal ultrasonography. Classic features of a gallbladder mucocele were identified in 9 of 11 dogs,² whereas the remaining 2 dogs had abdominal effusion and thickened biliary structures consistent with bile peritonitis and cholecystitis. The 2 dogs not imaged were surgically explored because of acute abdominal signs thought to be associated with extrahepatic bile duct obstruction and pancreatitis. At surgery, necrotizing cholecystitis, bile peritonitis, and gallbladder mucocele were grossly confirmed in each dog.

Definitive diagnoses and treatments—Of 38 Shetland Sheepdogs with gallbladder disease, 40% (15/38) were evaluated for acute onset of vomiting, signs of abdominal pain, or diarrhea. Eleven percent (4/38) of dogs had unexplained persistent but modest increases in serum ALP activity that instigated use of abdominal ultrasonography. Primary health problems preceding gallbladder mucocele identification were identified in 66% (25/38) of dogs and included pancreatitis, hyperlipidemia, exposure to steroid hormones, hypothyroidism, protein-losing nephropathy, diabetes mellitus, cholelithiasis, and gallbladder dysmotility; some dogs had several disorders. Seven of 38 (18%) dogs had extrahepatic bile duct obstruction secondary to pancreatitis, whereas 10 of 38 (26%) dogs were thought to have chronic relapsing pancreatitis. Ten of 38 (26%) dogs had a well-documented history of hyperlipidemia; 3 of these became acutely ill after being fed a high-fat, protein-restricted diet. Another known hyperlipidemic dog became acutely ill after administration of immunosuppressive doses of glucocorticoids as treatment for an immune-mediated pancytopenia. Six of 38 (16%) dogs had abnormally high adrenal steroid hormone concentrations; 3 had abnormally high progesterone concentrations associated with atypical adrenal hyperactivity, and 3 had high glucocorticoid concentrations associated with classic hyperadrenocorticism. Five of 38 (13%) dogs had hypothyroidism that was recently diagnosed or had illness that interrupted hormone replacement. Two dogs had protein-losing nephropathy, and 1 dog had diabetes mellitus. Six of 38 (16%) dogs had choleliths confirmed at surgery. Three of 20 (15%) CUHA patients had severe vacuolar hepatopathy consistent with hepatocutaneous syndrome. Three dogs evaluated at CUHA had biliary dyskinesia (dysmotility and amotility) confirmed by use of dynamic ultrasonographic gallbladder volume studies.

Untreated dogs—Eight untreated dogs were either euthanized or died within 2 weeks of diagnosis of a gallbladder disorder as a result of primary illnesses (ie, protein-losing nephropathy, transitional cell carcinoma of the urinary bladder, hepatocutaneous syndrome, extrahepatic bile duct obstruction, cardiac failure associated with hyperadrenocorticism, splenic hemangiosarcoma, and hepatic neoplasia). One dog was euthanized for acute pancreatitis, gallbladder mucocele, and bile peritonitis, without treatment.

Medical treatment—Criteria used to select for medical rather than surgical management included a lack of signs of illness related to the gallbladder abnormality, client refusal to pursue surgery, or client inability to afford surgery. Dogs treated medically without surgical intervention (n = 7) were prescribed a severely fat-restricted diet if hypertriglyceridemic, ursodeoxycholic acid (10 to 15 mg/kg [4.5 to 6.8 mg/lb], PO, divided into 2 doses/d), and s-adenosylmethionine (20 mg/kg [9.1 mg/lb], PO, on an empty stomach), and owners were advised to return the dog every month or bimonthly for sequential abdominal ultrasonographic examinations to monitor their gallbladder status. Two of these dogs died within 2 weeks of diagnosis of a gallbladder abnormality (1 died of gallbladder rupture and 1 died from pulmonary thromboembolism related to protein-losing nephropathy), 2 dogs were lost to followup, and 3 dogs survived and were treated as advised. Three surviving dogs were monitored as recommended. In 1 dog, the gallbladder mucocele resolved, and in the other dogs, the gallbladder abnormality remained static during a 6-month interval.

Surgical treatment—Surgery was performed on 23 of 38 (60%) dogs, including 10 dogs at CUHA and 13 dogs at private practices. One dog received surgery twice at a 6-month interval. Cholecystectomy was performed in 18 of 38 (47%) dogs, cholecystoenterostomy was performed in 4 of 38 (11%) dogs, and 1 dog underwent a combined cholecystotomy and choledochotomy for removal of biliary debris. Cholecystectomy was completed when gallbladder wall viability appeared grossly abnormal, suggesting necrosis, and was accompanied by removal of inspissated or mucoid biliary debris from the common and cystic bile ducts and verification of duct patency. Cholecystotomy was performed to remove inspissated biliary material in dogs with an apparently viable gallbladder wall. Choledochotomy was performed to remove inspissated biliary material that could not be otherwise removed from the common bile duct. Cholecystoenterostomy was performed in dogs with a viable gallbladder wall or cystic duct remnant that required decompression of the common bile duct because of inability to clear occluding debris.

Two dogs died during general anesthesia during cholecystectomy, and 1 dog did not recover from general anesthesia after cholecystectomy; each of these dogs had bile peritonitis before surgery. Survival following cholecystectomy was reported for 11 of 18 dogs; 12 of the 18 dogs treated with cholecystectomy had gross evidence of necrotizing cholecystitis or a ruptured biliary tract and limited surgical options. Survival following cholecystoenterostomy was reported for 4 of 4 dogs, and survival following cholecystotomy-choledochotomy was reported for the single dog that received this surgery. However, the dog returned within 6 months because of recurrent clinical signs of gallbladder mucocele with a devitalized gallbladder wall and died from bile peritonitis after cholecystectomy. No other dog that survived the acute postoperative interval (2 weeks) developed postoperative extrahepatic biliary tract disease (the longest follow-up was 7 years). Bile peritonitis subsequent to gallbladder rupture was confirmed in 12 of 38 (32%) dogs (7 dogs at CUHA and 5 dogs at other hospitals). Of 12 dogs with bile peritonitis, 4 died, 4 were euthanized, and 4 recovered. Most cholecystectomized dogs that survived > 2 weeks (9/11) were managed after surgery by use of a fat-restricted diet and long-term administration of ursodeoxycholic acid as a choleretic agent.

Six of 17 (35%) bacteriologic cultures of liver or bile yielded positive results for aerobic bacteria (A-hemolytic Streptococcus spp, β-hemolytic Streptococcus spp, Enterococcus spp, Staphylococcus intermedius, β-hemolytic Escherichia coli, unencapsulated E coli [n = 2], Enterococcus spp, and Pseudomonas spp [3]); 2 cultures yielded > 1 isolate. The only anaerobic bacterium isolated was Proprionibacterium spp from liver tissue. Although each of these dogs had necrotizing cholecystitis and 3 had a ruptured gallbladder that caused bile peritonitis, each dog survived. Dogs with positive results of bacterial cultures received extended antimicrobial administration (usually for 4 weeks) with an appropriate antimicrobial chosen on the basis of culture and susceptibility testing.

Histologic findings—Histologic specimens of liver and gallbladder were submitted for 20 of 23 (87%) dogs. Cholecystitis was confirmed histologically in 16 of 20 (80%) dogs; the inflammatory infiltrate in the gallbladder wall was suppurative in 2 of 16 dogs or mixed (suppurative and nonsuppurative) in 14 of 16 dogs. Gallbladder mucosal cystic hyperplasia was confirmed in a gallbladder biopsy specimen of all 20 dogs. Hepatic histologic changes included a periportal neutrophilic or mixed inflammatory infiltrate in 9 of 20 (45%) dogs and a moderate to severe vacuolar hepatopathy in 7 of 20 (35%) dogs.

Survival—Twelve of 38 (32%) Shetland Sheepdogs included in this study either died or were euthanized within 2 weeks of diagnosis and treatment except for 1 dog that died from bile peritonitis associated with recrudescent gallbladder mucocele formation 6 months after combined cholecystotomy and choledochotomy. One dog died from acute renal failure associated with septic bile peritonitis, and 3 dogs were euthanized because of complications associated with biliary disease (hepatic abscesses [n = 1], severe pancreatitis and gallbladder necrosis [1], and postoperative bile peritonitis [1]). Two dogs were euthanized because of the grave prognosis warranted for hepatic failure attributed to a severe vacuolar hepatopathy consistent with the histologic features of the hepatocutaneous syndrome. One dog died at home because of severe pancreatitis and necrotizing cholecystitis. Two dogs died during anesthesia during cholecystectomy, and 1 dog did not recover from anesthesia after cholecystectomy. One dog died because of a nonhepatic primary disease process (amyloidosis that caused systemic thromboembolism). Two dogs were lost to follow-up; 1 dog had myelogenous leukemia, and the other had severe cardiac failure and hyperadrenocorticism.

Discussion

Findings of the present study indicated that Shetland Sheepdogs are predisposed to gallbladder disorders, specifically gallbladder mucocele. Furthermore, findings suggested that metabolic disturbances or gallbladder dysmotility may be associated with development of gallbladder biliary sludge in these dogs. In humans, gallbladder sludge is described as microlithiasis and is composed of a complex mixture of cholesterol crystals, precipitated bile pigments, mucin, and bile salts.^{5,10-12,25,26} As such, gallbladder sludge is recognized as an early and reversible phase of cholelith formation. Use of research models confirms that the gallbladder epithelium plays a dominant role in biliary sludge formation and that reduced gallbladder motility, other causes of bile stasis, or enhanced gallbladder water absorption augment biliary sludge formation.^{11–13,25,26} However, microlithiasis in canine gallbladder bile differs from the syndrome described in humans and the classic prairie dog model of cholelithiasis. In dogs, gallbladder microlithiasis involves bilirubinate pigments complexed with calcium carbonate (termed pigment choleliths) and has been induced in healthy dogs fed a protein-restricted, methionine-deficient, high-carbohydrate diet.^27–29 Examination of gallbladder contents from 10 dogs after cholecystectomy confirmed that bile from the gallbladder mucoceles grossly resembled that in dogs with experimentally induced microlithiasis. Furthermore, the gallbladder debris investigated in 1 dog reported here consisted of a similar complex of precipitated bilirubinate, mucin, and calcium carbonate as that found in dogs with nutritionally induced microlithiasis. In most dogs, gritty black-green bile was found in the gallbladder and a viscoelastic mucin layer was tightly adhered to the inner surface of the gallbladder. However, a firm, pale rubbery matrix filled the gallbladder lumen in 2 dogs, suggesting that there may be diverse mechanisms promoting gallbladder mucocele formation. The absence of a black-green color in the gallbladder contents of these dogs could be explained by a dominant contribution of gallbladder mucin and impaired flow of hepatic bile through the cystic duct. Unfortunately, the chemical composition of the gallbladder contents in these dogs was not investigated. The ultrasonographic images and clinicopathologic features of these dogs were not different from dogs with the more common black-green gritty gallbladder bile. Each of the 2 dogs had necrotizing cholecystitis that was seemingly related to the tense gallbladder wall distension caused by its firm gelatinous contents.

The gallbladder normally functions to store, concentrate (up to 10-fold concentration, compared with hepatic bile), and modify (adding mucin and immunoglobulins) bile.²⁵ In the enteric canal, bile facilitates digestive processes and delivers biliary IgA that assists in protecting enteric surfaces from infectious agents. The flow of bile into the alimentary canal imparts a regulatory effect on the microbial population and an illdefined positive effect regarding health of enterocytes. Impaired enteric bile flow has been associated with increased risk for transmural bacterial translocation that promotes hepatobiliary and systemic infections. Six of 17 dogs in this case series had positive results of bacteriologic culture of bile or liver tissue.

Mucin, a high–molecular-weight glycoprotein, is directly secreted into gallbladder bile by mucosal glands. Although gallbladder mucin is essential for protecting the gallbladder mucosa from concentrated biliary constituents, it also plays an important role in diseases of the gallbladder.^30,31 The physicochemical characteristics of mucin provide viscoelastic properties to bile. Viscosity of bile increases when it becomes supersaturated (dehydrated), and this change compromises the mechanical cleansing of the gallbladder that normally occurs with gallbladder contraction.^30,31 Inflammation of the gallbladder wall associated with septic or nonseptic cholecystitis, as well as other disorders enhancing gallbladder prostaglandin production, increases gallbladder mucin production.^25,31 In experimental and human cholelithiasis, mucin serves as a pronucleating agent that binds lipids and bilirubinate into a dense conglomerate.^25,31 Results of experimental cholelithiasis support the notion that mucin hypersecretion precedes cholecystolith formation and accompanies gallbladder inflammation.^31,32 Gallbladder epithelium and mucin hypersecretion also importantly contribute to the formation of bile sludge or microliths in species predisposed to pigmented choleliths (bilirubinate-calcium carbonate stones) as occurs in dogs.^33,34 Finding a thick layer of mucin adherent to the inner gallbladder surface and black-green gritty bile debris in dogs with a gallbladder mucocele is fully consistent with changes described in dogs with experimentally induced pigment microlithiasis.³⁴ However, the thickness of the mucin layer in the dogs reported here exceeded that reported in dogs with experimental cholelithiasis, perhaps reflecting the associated cystic mucosal hyperplasia or inflammation.

Although cystic hyperplasia of the mucus glands in the wall of the gallbladder is a consistent pathologic change associated with gallbladder mucocele, its etiopathogenic role remains unproven.^3,10 Terminology of the syndrome is confusing, and it has been referred to as mucinous hyperplasia, cystic hyperplasia of the gallbladder, mucinous cyst of the gallbladder, mucosal cysts of the gallbladder, cystic mucinous hypertrophy of the gallbladder mucosa, mucinous cholecystitis, and cystic glandular cholecystitis.¹⁰ In the initial report¹⁰ of cystic mucinous hypertrophy of the canine gallbladder, an association with gallbladder mucocele formation was described. Results of experimental studies^34–36 (with species other than canine and with canine gallbladder epithelium) indicate that enhanced exposure to bile acids increases the rate of gallbladder epithelial turnover, development of mucosal hyperplasia, and epithelial mucus secretion associated with microlithiasis. Increased concentrations of hydrophobic bile acids (specifically deoxycholate, a secondary bile acid derived from cholic acid) are most detrimental.³⁶ Thus, gallbladder cystic mucosal hyperplasia may represent a response to gallbladder bile stasis and supersaturation and may develop secondary to gallbladder dysmotility. Although the findings of the present study supported an association between gallbladder mucosal cystic hyperplasia and mucocele formation, whether cystic hyperplasia of the gallbladder wall plays an etiopathogenic role, is secondary to gallbladder dysfunction or dysmotility, or reflects a normal aging change remains unclarified.

Our findings suggest an association between dyslipidemias (hypertriglyceridemia and hypercholesterolemia) and gallbladder mucocele formation. Consistent with this observation is that previous case reports^2,3 of canine gallbladder mucoceles also have included dogs with disorders known to augment dyslipidemia (ie, dogs receiving glucocorticoids or dogs with hyperadrenocorticism). A number of Shetland Sheepdogs included in the present case series were historically classified as hyperlipidemic while consuming a variety of adult-maintenance dog foods before developing gallbladder disease. At the time of diagnosis of biliary tract disease, 70% of Shetland Sheepdogs remained hypercholesterolemic and 8 of 16 dogs tested were hypertriglyceridemic despite anorexia for a minimum of 48 hours. Hyperlipidemia and hypercholesterolemia are not unique to Shetland Sheepdogs with biliary disease, because this also has been recognized by the authors and others in Shetland Sheepdogs that lacked clinicopathologic features or ultrasonographic evidence of gallbladder abnormalities. This dyslipidemia in Shetland Sheepdogs has been proposed to represent a breed-related genetic abnormality.⁹ Three of 10 dogs with a history of hyperlipidemia developed acute clinical signs of gallbladder mucocele after consuming high-fat, protein-restricted diets. This phenomenon may reflect the strong stimulation for gallbladder contraction induced by an ingested fatty meal on a devitalized gallbladder wall segment, which may lead to rupture of the gallbladder. A subset of dogs with gallbladder mucocele in the present study had additional medical conditions known to aggravate dyslipidemia (ie, 6 dogs had hyperadrenocorticism or sex-hormone–related adrenal hyperplasia, 6 dogs had unexplained severe vacuolar hepatopathy consistent with a corticosteroid hormone disturbance, 1 historically hyperlipidemic dog developed clinical signs of gallbladder mucocele shortly after initiation of immunosuppressive glucocorticoid administration, 5 dogs had recently diagnosed hypothyroidism or had illnesses that interrupted hormone replacement, and 1 dog had diabetes mellitus). A link between corticosteroid hormones and bile lithogenicity is established for progestational compounds in several species, including dogs.^5,11-13 Furthermore, an association between progestational hormones, vacuolar hepatopathy, and gallbladder mucosal hyperplasia in dogs has been experimentally confirmed.^37–39

Results of the present study indicated that gallbladder mucoceles can be subclinical yet can transform acutely into a clinical condition. Clinicopathologic findings in Shetland Sheepdogs with clinical signs were not different from those described in other reports of canine gallbladder mucoceles and reflected biliary tract obstruction, gallbladder inflammation or ischemia, and bile peritonitis.^1–4 Because Shetland Sheepdogs with gallbladder disease were mostly middle-aged to older adult dogs, a variety of chronic disease conditions contributed to hematologic and biochemical abnormalities. Unfortunately, few clinicopathologic variables provide prognostic information unique from other criteria of critical illness, metabolic failure, hypoperfusion, or sepsis. Neutrophilic leukocytosis, with or without left shift and monocytosis, has been the predominant finding in previous reports of canine gallbladder mucoceles.^2,3 High WBC counts were associated with gallbladder rupture and bile peritonitis.² Similarly, nonsurviving dogs in the present study had higher WBC and neutrophil counts than surviving dogs, likely reflecting disease severity or tissue injury, bile peritonitis, and perhaps endogenous cortisol release. However, contrary to a previous report,² thrombocytopenia was not found. Nonsurviving dogs in the present study had lower concentrations of potassium than surviving dogs. Hypokalemia likely reflects the more serious nature of a dog's illness and develops subsequent to chronic vomiting, third-space fluid accumulation (eg, as occurs in bile peritonitis), and administration of fluids to offset hypotension. Clinical signs in dogs with gallbladder mucocele may develop secondary to pathologic distension of the gallbladder or major bile duct occlusion secondary to duct occlusion by inspissated biliary debris or cholecystoliths. However, we identified 11 dogs lacking clinical signs of gallbladder mucocele, and these dogs had no serum biochemical abnormalities or merely had unexplained high ALP activity. The dogs without clinical signs had some owner complaint, physical finding, or laboratory test (such as an unexplained high ALP activity) that was judged to warrant an abdominal ultrasonographic examination.

The ultrasonographic features of gallbladder mucocele include gallbladder distension with immobile echogenic bile producing a stellate or finely striated pattern resembling a kiwi fruit on cross section.² In prior reports,² dilation of the cystic and common bile ducts and hepatomegaly were described in 62% and 71% of cases, respectively. In the present study, we identified common bile duct obstruction in only 22% and hepatomegaly in only 44% of Shetland Sheepdogs with gallbladder disease. The difference between our findings and prior reports of canine gallbladder mucoceles reflects the fact that not all dogs in the present study had a mature mucocele that caused clinical signs.

The clinical signs associated with gallbladder mucocele are similar to those typically ascribed to pancreatitis. A clinical diagnosis of pancreatitis in a dog with a gallbladder mucocele may have been erroneous if based on clinical signs, clinicopathologic abnormalities, and ultrasonographic features. The gold standard of pancreatic biopsy for confirming a diagnosis of pancreatitis was not pursued in any dog in this retrospective study. Ultrasonographic pancreatic abnormalities in some dogs may have been caused by transmural leakage of bile from a diseased gallbladder, focal bile peritonitis, or extension of inflammation from cholecystitis because the pancreas and gallbladder are located in the right cranial abdominal quadrant. A causal role of hyperlipidemia or the influence of corticosteroid hormones on gallbladder mucocele formation and pancreatitis also would explain the apparent dual appearance of these disorders in some dogs. Alternatively, pancreatitis causing stasis of bile flow in the extrahepatic biliary system may have augmented gallbladder mucocele formation in some dogs. It is also possible that a clinical diagnosis of gallbladder disease in a dog with pancreatitis may be erroneous.

Dynamic ultrasonographic evaluation of gallbladder contractility confirmed that gallbladder dysmotility preceded formation of an organized mucocele in 3 dogs. We hypothesize that gallbladder dysmotility impedes the cleansing expulsion of mucin and bile from the gallbladder, thereby promoting mucocele formation. Because gallbladder motility depends on complex processes involving the parasympathic (autonomic) pathway and enteric neurohormones (eg, vagal innervation, motilin, and cholecystokinin), there are multiple mitigating factors.^15,40,41 Unfortunately, the role of these factors in gallbladder dysmotility and mucocele formation has not been investigated in dogs. However, in humans with cholecystoliths, the role of gallbladder motility in generation of biliary pain has been explored.¹⁶ Individuals with normal motility have severe pain (biliary colic), whereas those with gallbladder dysmotility have clinically silent disease.¹⁶ Thus, we speculate that gallbladder dysmotility in dogs may contribute to the insidious nature of a developing mucocele and the lack of clinical signs during this process.

Surgical management is usually advocated as the primary treatment for dogs with gallbladder mucocele. Cholecystectomy provides direct resolution of the problem but must be combined with surgical removal of congealed biliary secretions from the extrahepatic biliary structures and administration of broad-spectrum antimicrobials. The poor survival of dogs treated with cholecystectomy in this case series was similar to previous reports^2,3 and reflects the compromised status of the gallbladder at the time of surgery. Twelve of the 18 dogs in the present case series treated by cholecystectomy had 1 or more of the following features: necrotizing cholecystitis, ruptured gallbladder, bile peritonitis, or sepsis. Mucocele recrudescence in 1 dog surgically managed only by manual removal of biliary debris suggests the need for gallbladder removal or cholecystoenterostomy. The latter procedure should only be completed if the viability of the gallbladder wall is judged to be good. Four dogs with a viable gallbladder were successfully treated with cholecystoenterostomy and surgical removal of congealed biliary secretions from their extrahepatic biliary structures. In 1 dog without clinical signs, a biliary mucocele resolved following administration of ursodeoxycholic acid and feeding a severely fat-restricted diet.

Antimicrobial treatment is recommended before, during, and for several weeks after cholecystectomy in clinically affected dogs. A broad-spectrum antimicrobial should be considered because gram-positive and enteric gram-negative organisms were identified in 35% of the samples in the present study. Chronic administration of ursodeoxycholic acid and s-adenosylmethionine also may assist with the mechanical cleansing of microbes from the biliary system (choleresis and reduced bile viscosity), as well as provide anti-inflammatory and antioxidant benefits.^42–48

The present case series indicated that Shetland Sheepdogs have an inherent predisposition for gallbladder disease leading to gallbladder mucocele formation. On the basis of these findings, we recommend that ultrasonographic gallbladder evaluation be used for routine health care surveillance in middle-aged to older Shetland Sheepdogs, especially those with unexplained high serum ALP activity and those known to have hyperlipidemia (hypertriglyceridemia), hypercholesterolemia, or disease processes that affect cholesterol or triglyceride metabolism. If an unusual gallbladder appearance suggests early mucocele formation, a dynamic gallbladder study may reveal whether dysmotility is a concurrent and potentially contributing factor. The best treatment for dogs with gallbladder dysmotility remains to be proven. Although resolution of a subclinical gallbladder mucocele occurred in a single dog with medical therapy (ursodeoxycholic acid and feeding an ultra–low-fat diet), the efficacy of medical management remains unproven in the face of complicating health problems and gallbladder dysmotility. If medical management is undertaken, ultrasonographic evaluations at monthly or bimonthly intervals and periodic evaluation of serum biochemical profiles are recommended. Considering the poor response to cholecystectomy in clinically affected dogs, earlier elective surgical intervention (before gallbladder wall infarction or necrosis or development of septic cholecystitis or bile peritonitis) should be evaluated as a potentially better treatment option.