Abstract
OBJECTIVE To determine the ultrasonographic appearance of the major duodenal papilla (MDP) in dogs without evidence of hepatobiliary, pancreatic, or gastrointestinal tract disease.
ANIMALS 40 adult client-owned dogs examined because of conditions that did not include hepatobiliary, pancreatic, or gastrointestinal tract disease.
PROCEDURES Ultrasonographic examination of the MDP was performed. Each MDP was measured in 3 planes. Intraobserver reliability of measurements was determined, and associations between MDP dimensions and characteristics of the dogs were investigated. Histologic examination of longitudinal sections of the MDP was performed for 1 dog to compare the ultrasonographic and histologic appearance.
RESULTS The MDP appeared as a layered structure with a hyperechoic outer layer, hypoechoic middle layer, and hyperechoic inner layer that corresponded to the duodenal serosa, duodenal muscularis, and duodenal submucosa, respectively. Layers visible during ultrasonographic examinations were consistent with layers identified histologically. Intraobserver reliability was substantial for each plane of measurement. Mean ± SD length, width, and height of the MDP were 15.2 ± 3.5 mm, 6.3 ± 1.6 mm, and 4.3 ± 1.0 mm, respectively. An increase in body weight of dogs was significantly associated with increased values for all measurements.
CONCLUSIONS AND CLINICAL RELEVANCE The ultrasonographic appearance and approximate dimensions of the MDP of dogs without evidence of hepatobiliary, pancreatic, or gastrointestinal tract disease were determined. Additional studies are needed to evaluate possible ultrasonographic lesions of the MDP in dogs with hepatobiliary, pancreatic, or intestinal diseases and to investigate clinical implications of these lesions with regard to diagnosis and prognosis.
The MDP is the portion of duodenum through which the CBD and the pancreatic duct empty their respective secretions into the duodenal lumen. The principal pancreatic duct in dogs is the accessory pancreatic duct; the pancreatic duct may be absent in some dogs.1 The CBD enters the mesenteric wall of the duodenum 3 to 6 cm distal to the pylorus and has an intramural oblique portion of approximately 1.5 to 2 cm.2 The MDP is the opening of both the CBD and pancreatic duct (if present).1 However, the CBD and intramural portion of the CBD and pancreatic duct (if present) collectively are commonly referred to as the CBD in the veterinary literature.3,4 The minor duodenal papilla, where the accessory pancreatic duct opens into the duodenum, is approximately 3 cm aboral to the MDP. The muscular part of the CBD is composed of 2 layers of smooth muscle surrounding the intramural portion of the CBD and pancreatic duct. The tunica muscularis of the duodenum corresponds to the external layer. The internal layer is the musculus proprius of the bile duct, which is organized in bundles around the ampulla of the bile and pancreatic ducts and can be considered as the musculus sphincter ampullae hepatopancreaticae. However, there are interindividual variations in the amount of musculature of the duodenal papilla among dogs.1
In humans, tumors of the MDP account for 0.2% of all gastrointestinal tumors.5 Among these, 21% are adenomas and 79% are adenocarcinomas.6 Various imaging modalities are available to aid in the diagnosis of tumors of the MDP. Endoscopic ultrasonography is the most common and most accurate modality used.5 In human patients, endoscopic ultrasonography can be used to correctly identify a nodule in the MDP in 75% of cases, whereas transabdominal ultrasonography can be used successfully in only 15% of cases.7 In the veterinary literature, a ganglioneuroma of the MDP was reported in a dog, but abdominal ultrasonography was not performed, which possibly reflected a lack of availability of ultrasonography at the time of the report.8 In a series of 6 cats with diseases involving the MDP, 3 had sphincter dysfunction, 2 had obstruction by inspissated bile, and 1 had an infiltrating tumor.9
In humans and small animals, clinical signs associated with diseases involving the CBD are nonspecific and commonly include jaundice, signs of abdominal pain, weight loss, fever, and vomiting.7,8 In dogs, transabdominal ultrasonography is commonly used (because of its availability, rapidity for use, low cost, and lack of ionizing radiation) when a dog is examined because of these clinical signs. Investigators of 1 study3 found that the MDP was visible during abdominal ultrasonography in 42% of dogs.3 The poor performance of ultrasonography for detection of the MDP and the lack of specific clinical signs may partly explain the paucity of reports about diseases involving the MDP in the veterinary literature.8,9
The primary objective of the study reported here was to describe the ultrasonographic appearance of the MDP in dogs without evidence of hepatobiliary, pancreatic, or gastrointestinal tract disease; determine the approximate length, width, and height of the duodenal papilla in those dogs; and assess the repeatability of transabdominal ultrasonography for anatomic measurements of the MDP. A secondary objective of the study was to provide a preliminary comparison of the ultrasonographic and histologic appearance of the MDP.
Materials and Methods
Animals
Forty adult client-owned dogs were included in the study. None of the dogs had evidence of gastrointestinal tract, hepatobiliary, or pancreatic disease as determined on the basis of the medical history and results of physical examination and ultrasonography. Because this study included only abdominal ultrasonographic examinations performed during routine clinical evaluations, consent of the owners for participation of their dogs in the study was not specifically requested. Approval was granted by the Committee on Research Ethics at the School of Veterinary Science of the University of Liverpool.
Dogs with a mildly heterogeneous hepatic parenchyma during ultrasonographic examination were included only if cytologic examination of the liver revealed no substantial disease other than benign hyperplasia. Dogs with a history of vomiting, diarrhea, anorexia, constipation, signs of nausea, or signs of abdominal pain during the 3 months immediately preceding the ultrasonographic examination were excluded. Dogs with abnormal plasma activities of alanine aminotransferase, alkaline phosphatase, aspartate aminotransferase, or canine-specific pancreatic lipase or concentrations of bile acids during the week of the ultrasonographic examination were also excluded. Several dogs were excluded because of gas in the stomach or duodenum (or both) or because of a deep-chested conformation that prevented identification of the MDP. The number of dogs excluded for failure to identify the MDP was not recorded.
Ultrasonographic examination
All ultrasonographic examinations were performed in B-mode with the same ultrasound machinea and curvilinear microconvex probe (8 to 11 MHz). All dogs were sedated or anesthetized and then positioned in right or left lateral recumbency for the procedure. The MDP was identified a few centimeters aboral to the pylorus as a small protrusion on the mesenteric border of the duodenum or by following the CBD to the duodenum. In several dogs, the minor duodenal papilla could be identified a few centimeters distal to the MPD; however, it was substantially smaller in all these dogs, which prevented confusing the MPD for the minor duodenal papilla.
Length of the MDP was measured on the longitudinal view, whereas height and width of the MDP were measured with the duodenum in cross section (transverse view). Length of the MDP was measured at the level of the duodenal serosa and parallel to the long axis of the duodenum (Figure 1). Height was measured on the axis of the duodenal diameter; it was measured from the duodenal mucosa to the most external part of the MDP at the location where the MDP was the highest. Width was measured perpendicular to the axis of the duodenal diameter at the level of the duodenal serosa at the point where the MDP was the widest. All measurements were obtained 3 times; the ultrasound transducer was repositioned between successive measurements.
Longitudinal (A) and transverse (B) ultrasonographic views used for measurements of the MDP of a dog. Notice that the length is 2.122 cm on the longitudinal view, and the height is 3.766 mm on the transverse view. The scale on the right side is in centimeters.
Citation: American Journal of Veterinary Research 77, 6; 10.2460/ajvr.77.6.597
Images were obtained by 2 experienced ultrasonographers (JRM and GMW). Static images were recorded for retrospective review. Ultrasonographic images were exported to a software program.b The MDP measurements were obtained retrospectively as a consensus between a resident in a veterinary radiology training program (JRM) and a board-certified veterinary radiologist (SML).
Postmortem examination
Postmortem examination was performed on the cadaver of a recently euthanized dog donated for research purposes. It was confirmed that the dog was free of gastrointestinal tract, hepatobiliary, or pancreatic disease. Abdominal ultrasonography was not performed on this dog before it was euthanized.
Portions of the pancreas, duodenum, and liver were harvested and fixed in neutral-buffered 10% formalin. The proximal part of the descending duodenum was transected longitudinally at the mid duodenal papilla for macroscopic evaluation. Longitudinal histologic sections of the MDP were obtained, processed routinely, and embedded in paraffin; sections were cut at a thickness of 4 μm and stained with H&E stain.
Statistical analysis
Statistical analyses were performed with computerized statistical analysis software.c Primary outcome variables were ultrasonographically measured length, width, and height of the MDP. Intraobserver reliability of the repeated ultrasonographic measurements was assessed through calculation of the intraclass correlation coefficient (2-way single measure for absolute agreement) with corresponding 95% CIs. For all subsequent analyses, the mean of the repeated measurements was used. Independent variables were derived from information obtained from the signalment data, clinical records, imaging reports, and patient follow-up monitoring. Variables assessed included those related to each dog (breed, body weight, sex, and age), reason for examination (ie, primary complaint), ultrasonographic findings, and final diagnosis.
Descriptive statistics were calculated for variables as appropriate; continuous data were summarized as mean ± SD if normally distributed or median with interquartile ranges if nonnormally distributed. Categorical data were collected into appropriate groups and expressed as frequencies with 95% CIs when applicable. Normality of distribution for all continuous variables was assessed with graphical analysis and a Kolmogorov-Smirnov test. Logarithmic transformation was applied to variables with a nonnormal distribution. Once measurements were found to have a normal distribution, the mean, SD, 5% quantile, and 95% quantile were calculated for each ultrasonographic measurement.
Associations between independent variables and mean measurements of MDP dimensions were assessed by use of linear regression analysis. Variables that had an association with papilla size on initial univariable analysis (P < 0.25) were considered for incorporation into a final multivariable model for that outcome. For any pair of independent variables with a correlation coefficient ≥ 0.70, only the variable with the smallest P value was considered for subsequent inclusion in the multivariable analysis. Multivariable models were constructed by use of a manual backward stepwise procedure, with variables with values of P < 0.05 retained in the model. Potential confounding factors were assessed by examining parameter estimates for substantial changes after removal of those factors.
Results
Dogs
Forty dogs met the inclusion criteria. There were 16 neutered females, 9 sexually intact males, and 15 neutered males that ranged from 15 months to 13.2 years of age (mean, 22.2 months). Body weight ranged from 3.8 to 49 kg (mean, 9.4 kg). There were 31 purebred dogs comprising 5 Labrador Retrievers, 4 Staffordshire Bull Terriers, 3 German Shepherd Dogs, 2 Bichon Frises, 2 Border Collies, 2 Boxers, 2 Cavalier King Charles Spaniels, 2 Cocker Spaniels, 1 Beagle, 1 Bearded Collie, 1 Chihuahua, 1 Chinese Crested Dog, 1 Golden Retriever, 1 Irish Setter, 1 Lhasa Apso, 1 Yorkshire Terrier, and 1 Vizsla. There were 9 crossbred dogs. The most common reasons for ultrasonographic evaluation were tumor staging (n = 26 dogs), urinary tract disease (4), paraparesis (2), proteinuria (2), polyuria-polydipsia (1), seizures (1), polycythemia (1), neutropenia (1), signs of abdominal discomfort (1), and tenesmus (1).
Ultrasonographic appearance
The MDP had a similar ultrasonographic appearance in all dogs examined (Figure 2). The MDP appeared as a layered structure on both the longitudinal and transverse views. Progressing from outer to inner, there was a thin and even hyperechoic rim, a slightly thicker hypoechoic rim, and an irregularly shaped hyperechoic center. This layering was not always conspicuous and appeared faint or interrupted in some dogs. The hyperechoic outer layer, hypoechoic layer, and hyperechoic inner layer corresponded to the duodenal serosa, duodenal muscularis, and duodenal submucosa, respectively. On the transverse view, the hyperechoic center was fan shaped with a semicircular external part protruding from the duodenal wall and a tubular internal part extending through the duodenal wall toward the lumen. In most images, pancreatic parenchyma was visible surrounding the MDP.
Longitudinal section of the duodenum, bile duct, and MDP (A) and their corresponding appearance on longitudinal (B) and transverse (C) ultrasonographic views. In panel A, notice the MDP (arrow). Structures visible macroscopically and ultrasonographically include the duodenal layers (mucosa [1], submucosa [2], muscularis [3], and serosa [4]). Layers visible at the MDP include the intramural portion of the CBD (asterisk), duodenal muscularis mucosa and ductus choledochi muscle sphincter (circle), and duodenal muscularis and ductus choledochi muscle sphincter (arrowhead). In panels B and C, the scale on the right side is in centimeters.
Citation: American Journal of Veterinary Research 77, 6; 10.2460/ajvr.77.6.597
Reliability of measurements
The calculated intraclass correlation coefficients for repeated ultrasonographic measurements of MDP length, width, and height were all > 0.85, which indicated substantial intraobserver reliability. The intraclass correlation coefficient was 0.93 (95% CI, 0.88 to 0.96) for length, 0.91 (95% CI, 0.85 to 0.95) for width, and 0.86 (95% CI, 0.78 to 0.92) for height. Mean values of all MDP measurements were not normally distributed but had a normal distribution after logarithmic transformation. The independent continuous variables of age and body weight were normally distributed. Geometric mean ± SD, 5% quantile, and 95% quantile of all MDP measurements were calculated (Table 1).
Ultrasonographic measurements of length, width, and height of the MDP in 40 dogs without evidence of hepatobiliary, pancreatic, or gastrointestinal tract disease.
| Variable | Geometric mean ± SD* | 5% quantile | 95% quantile |
|---|---|---|---|
| Length (mm) | 15.2 ± 3.5 | 10.7 | 23.2 |
| Width (mm) | 6.3 ± 1.6 | 4.0 | 8.7 |
| Height (mm) | 4.3 ± 1.0 | 3.0 | 6.0 |
Geometric mean is the back-transformed mean of the logarithmically transformed data.
Association of MDP dimensions with independent variables
Linear regression analysis revealed an association of body weight and sex with most outcome measurements of MDP dimensions (Table 2). The variables age, reason for examination (ie, primary complaint), ultrasonographic findings, and final diagnosis were not associated with MDP dimensions. When controlling for body weight in the final multivariable models, sex no longer had a significant effect, which indicated that body weight was the only variable that remained in the final linear regression models. Increasing body weight was significantly associated with increased values for all measurements (Figure 3).
Box-and-whisker plots of ultrasonographic measurements for the length (A), width (B), and height (C) of the MPD in 40 dogs of various body weights. Each box represents the interquartile range, the horizontal line within each box represents the median, the whiskers represent the range, and circles represent outliers.
Citation: American Journal of Veterinary Research 77, 6; 10.2460/ajvr.77.6.597
Coefficients and P values for univariable regression analysis of independent variables that were associated with ultrasonographically measured length, width, and height of the MDP in 40 dogs without evidence of hepatobiliary, pancreatic, or gastrointestinal tract disease.
| Length | Width | |||||
|---|---|---|---|---|---|---|
| Variable | Coefficient | P value | Coefficient | P value | ||
| Sex | ||||||
| MN | — | — | — | — | ||
| MS | −0.26 | 0.10 | −0.45 | 0.30 | ||
| FN | −0.67 | 0.03 | −0.60 | 0.10 | ||
| Body weight (kg) | 0.005 | < 0.001 | 0.004 | 0.004 | ||
Variables with values of P < 0.25 were considered for incorporation into a final multivariable model.
— = Not applicable (referent). FN = Female neutered. MN = Male neutered. MS = Male sexually intact.
Postmortem examination
No abnormalities were detected during macroscopic and histologic examination of the liver, pancreas, duodenum, and MDP. Macroscopically, the MDP was identified in the duodenal mucosa as a small elevation located at the end of a short rim corresponding to the last intramural portion of the CBD. The CBD exited the liver and was located along the serosal surface of the duodenum (segment referred to as the free portion) until it entered obliquely into the intestinal wall. The free portion segment was followed by an intramural portion of approximately 1.5 to 2 cm that finally opened into the MDP (Figure 2).
Histologically, the CBD was a single layer of columnar epithelial cells. The intramural portion was surrounded by 2 layers of smooth muscle. The external layer was the tunica muscularis of the duodenum, and the inner layer corresponded to the musculus proprius of the CBD, which wrapped around the end portion of the CBD from the infundibular portion to the duodenal orifice and comprised the ductus choledochi muscle sphincter. The ductus choledochi muscle sphincter formed a funnel-shaped layer characterized by bands of smooth myofibers in the mucosal portion. At the level of the submucosa, the musculus proprius also formed a ring of muscle (ie, ampullae hepatopancreaticae muscle sphincter) that surrounded the termination of the bile and pancreatic ducts (Figure 4).
Photomicrograph of a tissue section of the intramural portion of the MDP of a dog. The mucosa of the CBD is composed of simple cuboidal epithelium. The submucosa contains multiple mucous glands (circle). Notice the ductus choledochi muscle sphincter (arrows), submucosa (black arrowhead), tunica muscularis of the duodenum (asterisk), muscularis mucosa (star), and mucosa (white arrowhead). The vertical bar on the left side indicates the intramural portion of the CBD. H&E stain; bar = 250 μm.
Citation: American Journal of Veterinary Research 77, 6; 10.2460/ajvr.77.6.597
Discussion
Ultrasonographic scanning techniques and ultrasonographic appearance for the MDP of cats have been published.10 However, to the authors' knowledge, detailed anatomic and ultrasonographic descriptions of the MDP in dogs are not available. The study reported here provided a description of the ultrasonographic appearance of the MDP in dogs without ultrasonographic lesions or clinical signs involving hepatobiliary, pancreatic, and gastrointestinal tract diseases. The dogs in this study were not free of disease, and most underwent abdominal ultrasonography for staging of neoplastic disease. Presence of chronic or subclinical hepatic, pancreatic, or duodenal disease could not be ruled out. However, on the basis of a lack of relevant clinical and laboratory findings, the authors considered it unlikely that important lesions of the MDP were present in these dogs.
The MDP was visualized in only 42% of the dogs in a previous study.3 The frequency of visualization of the MDP was not specifically evaluated in the study reported here. However, the authors believed that, although it was challenging in some dogs, the MDP could be viewed in most dogs that met the inclusion criteria. This may have reflected the fact that the dogs examined in the present study were smaller, with a mean body weight of 9.4 kg, compared with a body weight of 21.5 kg for that previous study.3 Additionally, most dogs evaluated in the present study were anesthetized, whereas dogs evaluated in the previous study3 were conscious or only sedated. The amount of pressure applied to the abdomen to enable viewing of the MDP was sometimes substantial, and this amount of pressure may not be tolerated well by conscious dogs. Similarly, the main reason for not being able to view the MDP in the study reported here was the presence of contents within the stomach and size and conformation of the dog. Viewing of the MDP of large and deep-chested dogs was much more challenging than it was in small dogs.
Ultrasonographic examination revealed that the MDP appeared layered, at least partially, in all dogs. These ultrasonographic layers were consistent with the layers observed during histologic examination, which consisted of the 4 layers of the duodenum (mucosa, submucosa, muscularis, and serosa) and the specialized muscular structures of the CBD (Figure 2). The outer hyperechoic layer corresponded to the duodenal serosa. The CBD was tubular, oblique within the duodenal wall, and hyperechoic and extended to the duodenal lumen. The intramural portion of the CBD was surrounded by a hypoechoic layer. The luminal part of this layer represented a combination of the duodenal lamina muscularis mucosae (which is part of the tunica mucosa) and the ductus choledochi muscle sphincter, whereas its serosal part was a summation of the duodenal muscularis and the ductus choledochi muscle sphincter.
The relatively small number of dogs included in the present study did not allow the establishment of reference intervals. However, the objective of the study was not to determine a valid reference interval but, rather, to describe the ultrasonographic appearance of the MDP of dogs and relate it to results of a histologic examination. This will be necessary for accurate detection of lesions affecting the MDP. We determined an approximate mean ± SD for the length (15.2 ± 3.5 mm), width (6.3 ± 1.6 mm), and height (4.3 ± 1.0 mm) of the MDP in dogs. As expected, the length, width, and height of the duodenal papilla were not significantly associated with age, breed, or sex. However, all 3 measurements were positively associated with body weight. This finding, although not entirely expected, appeared logical because duodenal wall thickness and pancreatic thickness are also positively correlated with body weight.11,12 Different approximate values of the dimensions could have been calculated for dogs on the basis of body weight groups. However, differences in MDP dimensions on the basis of body weight did not appear sufficiently large enough to warrant this evaluation. Additionally, there were insufficient numbers of dogs to allow body weight–specific MDP dimensions to be calculated. However, it should be considered that larger dogs will typically have values toward the upper limit of these dimensions, and smaller dogs will have values toward the lower limit.
Obstruction of the extrahepatic biliary tract is uncommon in dogs but can involve the MDP. It can be secondary to acute pancreatitis, cholelithiasis, pancreatic and duodenal tumors, or abscesses.13,14 It is important to have information about the anatomy and size of the MDP to allow evaluation for its involvement in these conditions, which might change the treatment. Further studies are needed to establish reference intervals for dimensions of the MDP, determine the possible ultrasonographic changes of the MDP in clinical conditions, and evaluate the implications of these changes with regard to diagnosis, treatment, and prognosis.
Acknowledgments
All procedures described within this study were performed at the University of Liverpool School of Veterinary Science, Neston, United Kingdom.
The authors declare that there were no conflicts of interest or extrainstitutional funding or support.
Presented in part as an abstract at the 17th Congress of the International Veterinary Radiology Association, Freemantle, WA, Australia, August 2015.
ABBREVIATIONS
| CBD | Common bile duct |
| CI | Confidence interval |
| MDP | Major duodenal papilla |
Footnotes
Logiq 7, General Electric Medical System, Milwaukee, Wis.
OsiriX, V.4.1.1, 64-bit, Pixmeo, Geneva, Switzerland.
SPSS for Windows, version 21.0, SPSS Inc, Chicago, Ill.
References
1. Evans H, DeLahunta A. The digestive apparatus and abdomen. In: Evans H, DeLahunta A, eds. Miller's anatomy of the dog. 3rd ed. Philadelphia: WB Saunders, 1993;281–337.
2. Dyce KM, Sack WO, Wensing CJG. The abdomen of the dog and cat. In: Dyce K, Sack WO, Wensing CJG, eds. Textbook of veterinary anatomy. 4th ed. Philadelphia: WB Saunders, 2009;434–453.
3. Barberet V, Schreurs E, Rademacher N, et al. Quantification of the effect of various patient and image factors on ultrasonographic detection of select canine abdominal organs. Vet Radiol Ultrasound 2008; 49: 273–276.
4. Gaillot HA, Penninck DG, Webster CR, et al. Ultrasonographic features of extrahepatic biliary obstruction in 30 cats. Vet Radiol Ultrasound 2007; 48: 439–447.
5. Zbar AP, Maor Y, Czerniak A. Imaging tumours of the ampulla of Vater. Surg Oncol 2012; 21: 293–298.
6. Carter JT, Grenert JP, Rubenstein L, et al. Tumors of the ampulla of Vater: histopathologic classification and predictors of survival. J Am Coll Surg 2008; 207: 210–218.
7. Skordilis P, Mouzas IA, Dimoulios PD, et al. Is endosonography an effective method for detection and local staging of the ampullary carcinoma? A prospective study. BMC Surg 2002; 2: 1.
8. Van Den Ingh TS, Rothuizen J. Ganglioneuroma of Vater's papilla and extrahepatic cholestasis in a dog. Vet Pathol 1984; 21: 254–256.
9. Furneaux RW. A series of six cases of sphincter of Oddi pathology in the cat (2008–2009). J Feline Med Surg 2010; 12: 794–801.
10. Etue SM, Penninck DG, Labato MA, et al. Ultrasonography of the normal feline pancreas and associated anatomic landmarks: a prospective study of 20 cats. Vet Radiol Ultrasound 2001; 42: 330–336.
11. Penninck DG, Zeyen U, Taeymans ON, et al. Ultrasonographic measurement of the pancreas and pancreatic duct in clinically normal dogs. Am J Vet Res 2013; 74: 433–437.
12. Delaney F, O'Brien RT, Waller K. Ultrasound evaluation of small bowel thickness compared to weight in normal dogs. Vet Radiol Ultrasound 2003; 44: 577–580.
13. Herman BA, Brawer RS, Murtaugh RJ, et al. Therapeutic percutaneous ultrasound-guided cholecystocentesis in three dogs with extrahepatic biliary obstruction and pancreatitis. J Am Vet Med Assoc 2005; 227: 1782–1786.
14. Mayhew PD, Richardson RW, Mehler SJ, et al. Choledochal tube stenting for decompression of the extrahepatic portion of the biliary tract in dogs: 13 cases (2002–2005). J Am Vet Med Assoc 2006; 228: 1209–1214.
