• Open Access

Comparison of Postprandial and Ceruletide Serum Bile Acid Stimulation in Dogs


  • This study was performed at the Small Animal Clinic, Justus-Liebig University Giessen, Germany. Data from this study were presented at the 17th ECVIM-CA Conference, Budapest, Hungary, 2007.

Corresponding author: Dr Nicole Bridger, Small Animal Clinic, Justus-Liebig University Giessen, Frankfurterstr 126, 35392 Giessen, Germany; e-mail: nicoleresk@gmx.de.


Background: Postprandial (PP) serum bile acid (SBA) stimulation is an important test for detecting hepatic dysfunction in dogs. However, this test is influenced by numerous variables, and a standardized approach using an injectable cholecystokinin analog (ceruletide) may be advantageous.

Hypothesis: Ceruletide SBA stimulation test is more sensitive than PP SBA stimulation in dogs.

Animals: Animals with portosystemic shunt (PSS) (n = 11) and dogs with upper respiratory disease (URD) (n = 9) were investigated. Healthy dogs (n = 13) and dogs with other diseases (n = 17) served as controls.

Methods: All dogs underwent SBA stimulation with food and ceruletide. Stimulation blood samples were drawn at 60/120 minutes and 20/30/40 minutes, respectively. Results were compared statistically, and the sensitivity and specificity were determined with receiver-operating characteristic curves.

Results: Stimulated SBA were significantly higher in both study groups than in controls. For dogs with PSS, the sensitivity and specificity (>35 μmol/L) were 100% postprandially (120 minutes) and 91 and 100%, respectively, postceruletide (30 minutes). The difference between these values was not statistically significant. For dogs with URD, the sensitivity and specificity (>22 μmol/L) were 44 and 88% postprandially (120 minutes) and 100 and 88% postceruletide (30 minutes).

Conclusions and Clinical Importance: Ceruletide SBA stimulation circumvents exogenous and endogenous influences associated with PP SBA stimulation. The results indicate that ceruletide SBA stimulation performs as well as PP SBA stimulation in dogs with PSS and is more sensitive for the detection of hepatic dysfunction in dogs with URD.

In companion animals, clinical signs associated with liver disease are often nonspecific (ie, gastrointestinal and/or neurologic signs), and therefore the laboratory assessment of liver function is an important diagnostic tool. In particular, serum bile acids (SBA) determined after a 12-hour fasting period and 2 hours after feeding have been shown to be very effective in the assessment of liver dysfunction and in particular the detection of portosystemic shunts (PSS).1–4

Bile acids such as cholic acid and chenodeoxycholic acid are formed by the conversion of cholesterol in the liver. In dogs, almost all bile acids are then conjugated to taurine.5 Following conjugation, the bile acids and nonbile acid organic anions, are actively excreted into the bile canaliculus, where they constitute the major component of bile.6 During fasting, the bile is stored in the gallbladder until the ingestion of food causes the release of cholecystokinin from endocrine cells of the gut into the circulation, which subsequently induces gallbladder contraction. This is a well-known mechanism that has been described in detail in dogs7 and humans.8 The released bile acids participate in the digestion of fat in the intestine and undergo intensive enterohepatic circulation. Both unconjugated and amidated bile acids are absorbed passively and actively from the ileum into the portal circulation.6 In a healthy liver, the bile acids are then re-extracted by hepatocytes and secreted back into the bile. Cholestatic conditions caused by the impairment of bile secretion and flow at the bile duct level or by a functional defect in bile formation at the level of the hepatocytes result in intrahepatic and systemic accumulation of bile acids. These conditions then limit hepatic bile acid uptake, which is considered a protective mechanism to reduce hepatocellular bile acid overload.6 In addition, disruption of the enterohepatic circulation induced by a shunting vessel that bypasses the liver and connects the portal venous system to the systemic circulation (ie, PSS) also results in increased SBA concentrations.

Measurements of SBA in dogs with PSS resulted in a sensitivity of 64–100% in blood samples from fasted patients3,4,9 and 100% in samples taken postprandially.2–4 However, several factors can influence the postprandial (PP) values of SBA. Even when feeding a standardized test meal, PP SBA concentrations are subject to many variables such as individual differences in gallbladder emptying, the rate of gastric emptying, a delayed or increased passage of the meal in the bowels, or anorexia and vomiting in animals with hepatopathy.1 A useful alternative to PP SBA stimulation in dogs is an SBA stimulation test using the commercially available injectable synthetic cholecystokinin analog ceruletide.a,10 In 1997, Rufer and Grunbaum assessed 27 dogs suffering from a variety of primary and secondary hepatic diseases with the SBA stimulation test using ceruletide to detect hepatic dysfunction. In this study, all dogs with hepatic disease were found to have abnormal SBA stimulation results.10 Recently, dogs with tracheal collapse were demonstrated to have hepatic dysfunction using the ceruletide SBA stimulation test; amelioration thereof was demonstrated after resolution of tracheal collapse poststent implantation.11 However, both studies failed to compare the SBA stimulation test using ceruletide with the commonly used PP stimulation of SBA in dogs.

Therefore, the aim of this prospective study was to compare PP SBA stimulation with the SBA stimulation test using ceruletide in dogs with PSS and hypoxic upper respiratory disease (URD) (tracheal collapse and brachycephalic airway syndrome).

Materials and Methods


Thirteen healthy dogs (control group 1), 17 dogs with illnesses other than hepatopathy (control group 2), 11 dogs with PSS (study group 1), and 9 dogs with URD (study group 2) presented to the Small Animal Clinic, Internal medicine, Justus Liebig University Giessen, between January 2006 and July 2007 were included in the study. Control group 1 comprised dogs owned by veterinarians of the Small Animal Clinic, from whom informed consent was obtained. Dogs of a breed with a predisposition for primary hepatic disease were excluded. Several breeds were represented, including 4 medium-sized mixed breed dogs, 3 Beagles, 2 Golden Retrievers, and 1 each of 4 other breeds. Dogs were considered healthy based on history, physical examination, and normal CBC and serum biochemistry results. The median age was 4 years (range 1–10.5 years) and the median weight was 22 kg (range 7–61 kg). Control group 2 included animals that were ultimately diagnosed with illnesses other than hepatic disease. The median age was 2 years (range 8 months to 11 years) and the median weight was 22 kg (range 2–55 kg). There were 3 mixed breed dogs, 2 Yorkshire Terriers, 2 German Shepherd dogs, and 1 each of 10 miscellaneous breeds. Study group 1 was composed of animals with a congenital PSS. The median age was 1 year (range 4 months to 4.5 years) and the median weight was 7 kg (range 2.5–45 kg). Two Yorkshire Terriers and 9 miscellaneous breeds were represented in this group. Study group 2 included dogs with tracheal collapse and brachycephalic airway syndrome. The median age was 7 years (range 2–14 years) and the median weight was 6 kg (range 2.5–10 kg). Five dogs were Yorkshire Terriers, 2 were Pugs, and there was 1 Welsh Terrier and 1 Jack Russell Terrier.

The data collected for control group 2 and both study groups included history, physical examination, CBC, serum biochemical profile, and abdominal ultrasound. Histopathology results of the liver were available in 1 case. In dogs with URD (study group 2), thoracic radiographs and respiratory endoscopy instead of abdominal ultrasound were performed.

None of the dogs received steroids or antiepileptic drugs for at least 2 weeks before the study. The dogs received their last medications on the evening before the SBA test.

SBA Stimulation Test

All dogs underwent SBA stimulation with food (<5 kg body weight [BW] 2 teaspoons, >5 kg BW 2 tablespoons) or 0.3 μg/kg BW ceruletide IM, respectively, on consecutive days. A diet of moderate protein content and with an increased concentration of fiber was chosen to minimize metabolic complications such as hepatic encephalopathy (Royal Canin Hepatic Support). Before each test, the dogs were fasted for 12 hours. Blood samples were drawn at baseline, 60 and 120 minutes after feeding, and 20, 30, and 40 minutes postinjection, respectively. The blood samples were collected in plain tubes and left to clot; they were then centrifuged at 6,500 ×g for 1 minute, and the serum was used to measure SBA by a colorimetric test with endpoint determination in the ABX Pentra 400.b

Statistical Analysis

Distribution of normality was analyzed using the Shapiro-Wilk test. Statistical differences among the 4 groups were calculated using the Kruskal-Wallis test followed by the posthoc Mann-Whitney U-test. To control for type 1 errors, a Bonferroni Holm correction was used and P < .05 was considered to be statistically significant. In order to assess the ability of the ceruletide and PP SBA stimulation test to discriminate between the dogs in the study groups and those in control group 2, ie, to compare the sensitivity and specificity of both tests, receiver-operating characteristic (ROC) curves were drawn. For each ROC curve the area under the curve (AUC) was calculated; the greater the AUC, the better the performance of the test. The data were analyzed by the statistical software programs SPSSc and MedCalc.d



The final diagnoses for control group 2 included 5 dogs with neurologic disease and 4 dogs with gastrointestinal disease. In these dogs, ammonia tolerance testing was unremarkable. In three of the patients with gastrointestinal disease, the final diagnosis was food hypersensitivity, and in the 4th dog, the diarrhea was acute and suspected to be secondary to dietary indiscretion. The remaining dogs in control group 2 were affected by the following disorders: a combination of gastrointestinal and cardiac disease, cardiac disease, protein-losing nephropathy, urinary tract infection, allergic skin disease, pneumonia, hypoglycemia in a toy breed, and a combination of endocarditis and pneumonia. In the latter case, histopathology of the liver was unremarkable. In study group 1, a congenital PSS was diagnosed in 9 dogs, and there were 2 cases of hepatic dysfunction with a residual shunt postclosure of a congenital shunt. Study group 2 included 7 dogs with tracheal collapse and 2 dogs with brachycephalic airway syndrome.

None of the samples tested were hemolyzed or lipemic.

Baseline and Stimulated SBA

Baseline SBA concentrations were significantly higher in study group 1 than in both control groups (Fig 1). There were no differences in baseline SBA concentrations between control groups 1 and 2. However, baseline SBA concentrations were significantly higher in study group 1 compared with study group 2. No significant difference was found depending on the order in which the test was performed.

Figure 1.

 Box plots for postprandial serum bile acid (SBA) stimulation at baseline, 60 and 120 minutes postfeeding (left) and ceruletide SBA stimulation at baseline, 20, 30, and 40 minutes postinjection (right). Both study groups and both control groups are shown. Outliers and extremes are marked with a circle (○) or an asterisk (*), respectively. Significant differences between study group 1 and the other groups are shown by the unbroken bars. Significant differences between study group 2 and the other groups are shown by the broken bars. The levels of significance are marked by the number of # signs (###P≤ .001; ##P≤ .01; #P < .05).

SBA concentrations in both study groups compared with both control groups were significantly higher for each stimulation time point postceruletide injection (Fig 1). In addition, dogs in study group 1 had significantly higher stimulation values at each time point than dogs in study group 2. Dogs in study group 1 had significantly higher SBA than dogs in both control groups 60 and 120 minutes postfeeding. In study group 2, SBA were significantly higher than in both control groups 120 minutes postfeeding. At 60 and 120 minutes, study group 1 was found to have significantly higher SBA than study group 2.

ROC Curves

ROC curves were drawn for study groups 1 and 2, each time point, and both tests. For study group 1 and the ceruletide SBA stimulation test, the AUC was greatest at 30 minutes (0.989). In the PP SBA stimulation test, the AUC was greatest at 120 minutes (1.000), though this was not significantly different from the 30-minute value for ceruletide (P= .637). For study group 2 and the ceruletide SBA stimulation test, the AUC was greatest at 30 minutes (0.966), whereas in the PP test the AUC was greatest at 120 minutes (AUC 0.791); the difference was not significant (P= .067). For study group 2, the differences between the AUCs were significant for ceruletide stimulation at 20, 30, and 40 minutes in comparison with 60 minutes postprandially. For study group 1 and both methods, there were no significant differences between the AUCs at any time point.

Sensitivity and Specificity

The sensitivities and specificities at 120 minutes postprandially and 30 minutes postceruletide administration are shown in Table 1.

Table 1.   Sensitivity (sens.) and specificity (spec.) for ceruletide (30 minutes postinjection) and postprandial serum bile acid stimulation in dogs with a congenital portosystemic shunt and upper respiratory disease at different cut-off values.
Portosystemic Shunt (n = 11)
Cut-Off (μmol/L)Ceruletide 30 Minutes Postinjection (AUC = 0.989)120 Minutes Postfeeding (AUC = 1.000)
Sens. (%)95% CISpec. (%)95% CISens. (%)95% CISpec. (%)95% CI
Upper Respiratory Disease (n = 9)
Cut-Off (μmol/L)Ceruletide 30 Minutes Postinjection (AUC = 0.966)120 Minutes Postfeeding (AUC = 0.791)
Sens. (%)95% CISpec. (%)95% CISens. (%)95% CISpec. (%)95% CI
  1. The area under the curve (AUC) for each time point is given as well as the 95% confidence intervals (CI) for each cut-off. Optimal cut-off values as determined by the receiver-operating characteristic curves are shaded and marked with an asterisk (*).


For study group 1, a sensitivity and specificity of 100% (95% confidence interval [CI] 71–100% and 80–100%, respectively) was reached at 120 minutes (cut-off 35 μmol/L), which was superior to the performance at 60 minutes postprandially. For the ceruletide SBA stimulation test, the 30- and 40-minute values postadministration were superior to the 20-minute value. At 30 and 40 minutes, the sensitivity was 91% (10/11 dogs) (CI 59–99%) and the specificity was 100% (CI 80–100%) at a cut-off of 32 and 48 μmol/L, respectively. In study group 1, 2 of 11 (18%) dogs had baseline values within the normal range, though ceruletide and/or PP SBA stimulation revealed abnormal liver function (1 dog with a congenital PSS and 1 dog with a residual shunt postclosure).

In study group 2, the 120-minute value in the PP SBA stimulation test had excellent sensitivity (100%, CI 66–100%), but specificity was poor (53%, CI 28–77%) (cut-off 13 μmol/L). At 60 minutes, sensitivity was very poor (44%, CI 14–79%), though the specificity was good (94%, CI 71–99%) (cut-off 23 μmol/L). In contrast, the ceruletide test had optimal sensitivity (100%, CI 59–100%) and good specificity (88%, CI 64–98%) at 30 minutes (cut-off 22 μmol/L). The 20- and 40-minute value did not perform as well. The baseline SBA acid concentration was within normal limits in all patients, though ceruletide and/or PP SBA stimulation revealed abnormal liver function.

At a cut-off of 22 μmol/L, the sensitivity (100%, CI study group 1 71–100%, CI study group 2 59–100%) and specificity (88%, CI study groups 1 and 2 64–98%) for the ceruletide SBA stimulation test at 30 minutes was identical in both study groups.


This is the 1st study to compare PP and ceruletide SBA stimulation in dogs with PSS and dogs with upper airway problems.

To date, measurements of SBA postprandially have been shown to be a good marker for the detection of PSS in dogs.2–4 However, differences in gastric emptying, gallbladder contraction, and variations in type or amount of food used in the PP SBA stimulation test can cause inconsistent SBA results.1 Only recently, ceruletide-stimulated SBA have been shown to indicate hepatic dysfunction in dogs with primary hepatic disease10 and tracheal collapse.11

In the past, the induction of gallbladder contraction with a cholecystokinin analog such as ceruletide has been used in both human medicine12 and dogs.10,13 The results of this study suggest SBA stimulation of ceruletide to be equally sensitive for the detection of PSS in comparison with the commonly used 120-minute value in the PP test. The performance of the 60-minute PP value was found to be poor, probably because the time interval was too short for the bile acids to reach the systemic circulation. In humans, the comparison of SBA after application of ceruletide to PP SBA stimulation revealed that the test meal alone was superior in differentiating controls from the patients with liver disease. However, the addition of ceruletide caused a greater percentage increase of SBA.14 In healthy dogs, Rothuizen et al13 showed that the gallbladder was more effectively emptied with an infusion of cholecystokinin octapeptide than with a test meal. This is consistent with our results, which also revealed a consistently higher median of SBA concentrations after application of ceruletide compared with PP SBA stimulation. Though the ceruletide SBA stimulation test showed similar results in comparison with PP SBA stimulation in diagnosing dogs with congenital PSS, the practicability and reliability of ceruletide SBA stimulation are obvious. The test can be performed in a shorter time period (30 versus 120 minutes) and is independent of food intake and the presence of vomiting in the ill patient.

The discrimination between dogs with upper airway disease and control dogs was possible using ceruletide. This finding supports previous work demonstrating hepatic dysfunction in dogs with tracheal collapse.11 The PP test, however, was unable to consistently detect hepatic dysfunction in study group 2. There are several possible explanations for this finding. First, ceruletide directly induces gallbladder contraction and is considerably more potent than cholecystokinin-induced contraction secondary to the consumption of food.13,15 Second, individual and illness-influenced differences in gastric emptying after feeding play a role that can result in peak SBA concentrations occurring anywhere from 1 to 8 hours postprandially1 and may be missed with a single measurement. It must be kept in mind, however, that to date both the frequency of hepatic pathology and the etiology of hepatic dysfunction in dogs with upper airway disease are largely unknown, and further studies are necessary to elucidate the true prevalence and to characterize and confirm the hepatic lesions in affected dogs by tissue samples. Nonetheless, the results in study group 2 suggest that the ceruletide SBA stimulation test may be advantageous for the noninvasive detection of mild hepatic dysfunction otherwise not found on routine screening. Though nonspecific, an abnormal SBA stimulation test result could alert the clinician to the presence of occult liver disease and give an indication for a liver biopsy. This test could become useful when other routine test results are still within normal limits, such as in Doberman Pinschers and Labrador Retrievers with chronic hepatitis.16,17 Thus, further studies investigating the ceruletide SBA stimulation test as an early marker of primary and secondary liver disease characterized by histopathology of hepatic biopsies are warranted.

Because of its structural similarity to cholecystokinin,18 adverse effects of ceruletide might include those associated with the physiologic effects of cholecystokinin. Apart from gall bladder contraction, cholecystokinin stimulates pancreatic enzyme secretion and growth, increases intestinal motility, relaxes the stomach, constricts the pylorus, and inhibits gastric emptying when food is in the stomach, thereby inducing gastric distension.19 Adverse effects that have been reported in humans are usually minimal and self-limiting, including mild abdominal pain and transient diarrhea related to the increase in intestinal motility.20 As yet, these adverse effects have not yet been seen in dogs. Because of delayed gastric emptying, there is a potential risk for the development of gastric dilatation and volvulus, though this has not yet been reported.10 The development of pancreatitis is also a potential side effect; however, this should happen only if a significant overdose of ceruletide is given inadvertently. Simpson et al21 demonstrated that in dogs, the administration of ceruletide at a dose of 10 μg/kg/h induced edematous pancreatitis within 2 hours. Other reported adverse effects in humans are pain at the injection site and nausea.20 Rufer and Grunbaum10 reported vomiting in 2 of 39 dogs after injection of ceruletide. Some dogs in our study showed a pain reaction to the injection with ceruletide; however, as this was not a consistent finding, it is difficult to interpret whether this reaction was related to the injection per se or the application of the drug. Contraindications for the administration of ceruletide, therefore, might include the presence of acute pancreatitis and mechanical obstruction of the bile duct.

The sensitivity of pre- and postprandial SBA to detect hepatic dysfunction is discussed controversially; some authors have shown little difference between both parameters for most disorders,22 whereas others have demonstrated PP SBA to be superior.2,23,24 A recent study found baseline SBA concentrations to have a sensitivity of 92.2% for the detection of a congenital PSS and 81% for acquired shunting.9 In our study, none of the dogs with tracheal collapse had increased baseline SBA and 2 of 11 dogs with a PSS were missed with only a fasting SBA measurement. These findings clearly indicate that baseline SBA concentrations are not consistently elevated in dogs with hepatic dysfunction and that the determination of stimulated SBA can give additional information when liver disease is suspected. However, it must be kept in mind that an abnormal SBA stimulation test result is not a diagnosis per se and is not an appropriate test to discriminate between the various diseases. It is rather an indicator to consider whether further investigations, eg, a liver biopsy, are indicated to establish a diagnosis. Furthermore, not every primary hepatic disease—especially localized inflammatory or neoplastic disease—is associated with abnormalities in hepatic function. Therefore, an SBA stimulation test can only aid and not replace additional diagnostic modalities in the recognition of liver disease.

In conclusion, the ceruletide SBA stimulation test is a useful alternative to the PP SBA stimulation test and we recommend using the 30-minute value for diagnostic purposes. This test is not subject to the many variables that affect the PP SBA stimulation test and has been shown to perform as well as the PP SBA stimulation test in the detection of congenital PSS. Furthermore, the ceruletide SBA stimulation test is more sensitive than the PP SBA stimulation test for the detection of mild hepatic dysfunction in dogs with upper airway disease.


aTakus, Pharmacia GmbH, Karlsruhe, Germany

bHoriba ABX, Montpellier, France

cSPSS 5 for Windows, SPSS Inc, Chicago, IL

dMedCalc for Windows, Version, MedCalc Software, Mariakerke, Belgium