In carnivores, the stomach plays an important role in the digestive process. It acts as a storage reservoir for ingesta, mechanically mixes the ingesta, propels food at an appropriate rate into the duodenum, initiates the digestion of protein and fat, and absorbs some vitamins and minerals.1,2
Gastric disease in dogs is poorly characterized. Gastritis is defined as inflammation of the stomach.1 Acute-onset vomiting occurs frequently in dogs3 and is often attributed to acute gastritis. However, histologic examination is required to definitively diagnose gastritis. Acute gastritis (in which there have been clinical signs for < 14 days) is often a self-limiting condition or responds to palliative treatment; thus, biopsy specimens are seldom obtained. Consequently, acute gastritis is rarely definitively diagnosed. It has been suggested in some studies that gastritis is a common histologic finding (in 35% of dogs investigated for chronic vomiting4 and 48% of dogs with no clinical signs5). It has also been stated6 that chronic gastritis (clinical signs have persisted for > 14 days) rarely develops without involvement of other parts of the gastrointestinal tract. Because of the anatomic and physiologic relationship between the stomach and small intestines, it is possible that pathologic changes of any kind at either site could affect the other. It is clinically relevant to know whether the stomach or intestines (or both) of a patient are diseased because the treatment selected may differ depending on the site affected.
Gastric biopsy specimens are often collected endoscopically in dogs, with duodenal biopsy specimens often being collected concurrently. Endoscopy allows observation of the mucosa of parts of the gastrointestinal tract and collection of biopsy specimens for subsequent histologic evaluation. However, use of this technique often results in collection of inadequate specimens, and the biopsy specimens collected may not represent all layers of the gastrointestinal wall. In addition, it can also be difficult to obtain biopsy specimens from dense fibrous lesions. Surgical collection of full-thickness biopsy specimens is more invasive than endoscopic collection of biopsy specimens, but surgical collection provides the advantages that biopsy specimens can be collected from dense fibrous lesions, the resulting biopsy specimens can contain large amounts of submucosa, and, in some cases, the biopsy specimens can be of greater diagnostic value.7 Additionally, the assessment of gastrointestinal histopathologic changes is somewhat controversial, as indicated by results of a study8 in which the interpretation of endoscopic biopsy specimens was subject to significant interobserver variation.
The first objective of the retrospective case series reported here was to test the hypothesis that intestinal pathologic changes are often concurrent with gastric pathologic changes in dogs. Our second objective was to characterize the historical, physical, clinicopathologic, imaging, and endoscopic findings in dogs with gastric histopathologic abnormalities.
Materials and Methods
Case selection—Medical records from dogs that had undergone gastrotomy, gastroduodenoscopy, or gastroscopy at the Veterinary Medical Teaching Hospital at Texas A&M University and in which biopsy specimens had been collected from 1 or more sections of the gastrointestinal tract were identified. The search was performed by use of the hospital's computerized database and spanned a 5-year period between September 2002 and September 2007. Records were then retrospectively reviewed. Dogs were eligible for inclusion when abnormal histopathologic findings had been identified during evaluation of gastric biopsy specimens. Dogs were excluded when a gastric biopsy specimen was not collected, results of gastric histologic examination were unremarkable, or the clinician's primary diagnosis was of a condition affecting an extragastrointestinal system. Dogs with a primary diagnosis of a disease of an extragastrointestinal system were excluded because their clinical findings and results of diagnostic tests were unlikely to reflect pathologic changes in the gastrointestinal system.
Medical records review—Information summarized from each dog's medical record included signalment, clinical signs at time of admission (ie, initial clinical signs), duration of initial clinical signs, physical examination findings, results of serum biochemical and hematologic analyses, results of diagnostic imaging, endoscopic findings, histopathologic findings, clinician's final diagnosis or diagnoses, treatment, and (when available) outcome and survival. Grading of the severity of histopathologic lesions was based on the assessment of the veterinary pathologist who reviewed each section because no standardized system of grading was in use during the study period.
Procedures—On the basis of the clinician's final diagnosis or the results of gastric and intestinal histologic examination (or both), dogs were classified into 2 groups. Group 1 comprised dogs with disease or pathologic changes primarily of the stomach. For dogs in group 1, the clinician's diagnosis indicated that the disease process primarily affected the stomach or, for dogs in which a definitive diagnosis had not been recorded, the veterinary pathologist's assessment was that there were no intestinal pathologic changes. Group 2 comprised dogs with more generalized disease or pathologic changes of the gastrointestinal system (ie, disease or pathologic changes of both the stomach and other parts of the gastrointestinal system, including the liver and exocrine portion of the pancreas). For dogs in group 2, the clinician's final diagnosis indicated that parts of the gastrointestinal system other than the stomach were affected or, for dogs in which a definitive diagnosis had not been recorded, there was histologic evidence of abnormalities of the intestines as well as the stomach. Data for each group were then analyzed and compared with results for the other group.
Statistical analysis—Numeric data were expressed as median and range. Categoric data were analyzed by use of a Fisher exact test or χ2 test, as appropriate. Significance was set at values of P ≤ 0.05 for all statistical analyses. Statistical analysis was performed by use of standard computer packages.a,b
Results
The medical records of 159 dogs that underwent gastroscopy, gastroduodenoscopy, or gastrotomy in which 1 or more biopsy specimens were collected from 1 or more parts of the gastrointestinal tract were reviewed. Eighty-seven dogs were excluded because a gastric biopsy specimen was not collected or gastric histologic findings were judged to be normal. Thus, 72 dogs had evidence of gastric pathologic changes. Five of these dogs were excluded because the clinician's primary diagnosis was of an extragastrointestinal condition. The remaining 67 dogs were included in the study. Of these, 25 (37.3%) dogs were classified as group 1 (dogs with disease or pathologic changes primarily of the stomach) and 42 (62.7%) were classified as group 2 (dogs with more generalized disease or pathologic changes of the gastrointestinal system).
Thirty-three breeds were represented. Mixed-breed dogs accounted for the largest proportion (10/67 [14.9%] dogs). Other breeds commonly represented included Labrador Retrievers (n = 6 [9.0%]), Boston Terriers (5 [7.5%]), and Dachshunds (4 [6.0%]). Thirty-one (46.3%) dogs were male (12 sexually intact and 19 neutered), and 36 (53.7%) were female (5 sexually intact and 31 spayed). Median age of dogs included in the study was 6.0 years (range, 6 months to 15 years). Median duration of initial clinical signs was 1.8 months (range, 1 week to 24 months).
The most common initial clinical sign of the 67 dogs included in the study was vomiting (36 [53.7%] dogs). Other common initial clinical signs were weight loss (25 [37.3%] dogs), diarrhea (22 [32.8%] dogs), and a decrease in appetite or anorexia (17 [25.4%] dogs). Vomiting was recorded significantly (P = 0.001) more often in dogs that primarily had gastric disease (20/25 [80.0%] dogs of group 1), compared with the proportion of dogs in group 2 for which vomiting was recorded (16/42 [38.1%]). In contrast, weight loss was less common, but not significantly so (P = 0.118), for group 1 (6/25 [24.0%] dogs), compared with the results for group 2 (19/42 [45.2%] dogs). Similarly, diarrhea was recorded significantly (P < 0.001) less often as an initial clinical sign for dogs of group 1 (1/25 [4.0%] dogs), compared with results for dogs of group 2 (21/42 [50.0%] dogs).
A variety of physical examination findings were recorded for the 67 dogs. A decrease in body condition or poor body condition was the most frequent finding (20 [29.9%] dogs). Other commonly recorded physical examination findings were heart murmurs (13 [19.4%] dogs), changes in skin or coat (9 [13.4%] dogs), ascites (6 [9.0%] dogs), and abdominal discomfort (5 [7.5%] dogs). No significant abnormalities were recorded for 12 (17.9%) dogs.
The most commonly detected serum biochemical abnormalities were panhypoproteinemia (27/64 [42.2%] dogs), hypoalbuminemia (27/67 [40.3%] dogs), hypocalcemia (total calcium concentration; 23/64 [35.9%] dogs), hyperchloremia (21/66 [31.8%] dogs), hypocholesterolemia (20/63 [31.7%] dogs), decrease in serum cobalamin concentration (4/13 [30.8%] dogs), increase in alkaline phosphatase activity (18/64 [28.1%] dogs), and hypernatremia (18/67 [26.9%] dogs; Table 1). The most commonly detected hematologic abnormalities were lymphopenia (29/66 [43.9%] dogs), neutrophilia (22/66 [33.3%] dogs), eosinopenia (20/66 [30.3%] dogs), a decrease in RBC count (14/66 [21.2%] dogs), and leukocytosis (14/66 [21.2%] dogs; Table 2).
Results of serum biochemical analysis for 67 dogs with gastric histopathologic abnormalities included in the retrospectiv case series.
| Variable | No. of dogs | RR | No. (%) of dogs > RR | No. (%) of dogs < RR | No. (%) of dogs within RR | Median | Minimum | Maximum |
|---|---|---|---|---|---|---|---|---|
| ALP (U/L) | 64 | 24–147 | 18 (28.1) | 1 (1.6) | 45 (70.3) | 76 | 20 | 1,795 |
| ALT (U/L) | 66 | 10–130 | 13 (19.7) | 1 (1.5) | 52 (78.8) | 46.5 | 6 | 754 |
| GGT (U/L) | 62 | 0–25 | 6 (9.7) | NA | 56 (90.3) | 12 | 5 | 260 |
| Albumin (g/dL) | 67 | 2.4–3.6 | 5 (7.5) | 27 (40.3) | 35 (52.2) | 2.8 | 0.9 | 4.9 |
| Globulin (g/dL) | 64 | 1.7–3.8 | 12 (18.8) | 4 (6.3) | 48 (75.0) | 3.1 | 1.0 | 5.2 |
| Total protein (g/dL) | 64 | 5.7–7.8 | 5 (7.5) | 27 (42.2) | 32 (50.0) | 6.0 | 2.0 | 8.7 |
| Sodium (mmol/L) | 67 | 139–147 | 18 (26.9) | 5 (7.8) | 44 (65.7) | 145 | 135 | 154 |
| Potassium (mmol/L) | 67 | 3.3–4.6 | 5 (7.5) | 9 (13.4) | 53 (79.1) | 3.9 | 2.3 | 5.3 |
| Chloride (mmol/L) | 66 | 107–116 | 21 (31.8) | 9 (13.6) | 36 (54.5) | 115 | 87 | 125 |
| Calcium (mg/dL) | 64 | 9.3–11.8 | 1 (1.6) | 23 (35.9) | 40 (62.5) | 9.8 | 5.4 | 13.4 |
| ECO (mmol/L) | 65 | 21–28 | 5 (7.7) | 14 (21.5) | 46 (70.8) | 23 | 11 | 35 |
| Glucose (mg/dL) | 65 | 60–135 | 5 (7.7) | 0 (0.0) | 60 (92.3) | 101 | 61 | 311 |
| Cholesterol (mg/dL) | 63 | 120–247 | 12 (19.0) | 20 (31.7) | 31 (49.2) | 157 | 50 | 365 |
| SUN (mg/dL) | 66 | 5.0–29.0 | 2 (3.0) | 6 (9.1) | 58 (86.6) | 11.0 | 0.1 | 50.0 |
| Creatinine (mg/dL) | 67 | 0.3–2.0 | 0 (0) | 0 (0) | 67 (100) | 0.9 | 0.4 | 1.9 |
| Phosphate (mg/dL) | 67 | 2.9–6.2 | 1 (1.5) | 8 (11.9) | 58 (86.6) | 4.0 | 1.7 | 6.4 |
| Total bilirubin (mg/dL) | 62 | 0–0.8 | 1 (1.6) | NA | 61 (98.4) | 0.1 | 0.1 | 1.6 |
| cPLI (μg/L) | 8 | 0–200 | 0 (0) | NA | 8 (100) | 44 | 28 | 164 |
| cTLI (μg/L) | 11 | 5.7–45.2 | 0 (0) | 3 (27.3) | 8 (72.7) | 8.5 | 0.2 | 21.2 |
| Folate (μg/L) | 13 | 7.7–24.4 | 1 (7.7) | 2 (15.4) | 10 (76.9) | 12.7 | 5.2 | 25.0 |
| Cobalamine (ng/L) | 13 | 252–908 | 1 (7.7) | 4 (30.8) | 8 (61.5) | 313 | 99 | 974 |
ALP = Alkaline phosphatase. ALT = Alanine aminotransferase. GGT = γ-Glutamyltransferase. ECO2 = Enzymatically measured carbon dioxide. cPLI = Canine pancreatic lipase immunoreactivity. cTLI = Canine trypsin-like immunoreactivity. NA = Not applicable. RR = Reference range.
Results of hematologic analysis for 67 dogs with gastric histopathologic abnormalities included in the retrospective case series.
| Variable | No. of dogs | RR | No. (%) of dogs > RR | No. (%) of dogs < RR | No. (%) of dogs within RR | Median | Minimum | Maximum |
|---|---|---|---|---|---|---|---|---|
| PCV (%) | 66 | 31–56 | 3 (4.5) | 3 (4.5) | 60 (90.9) | 43.7 | 12.8 | 60.2 |
| RBCs (× 106/μL) | 66 | 5.5–8.5 | 2 (3.0) | 14 (21.2) | 50 (75.8) | 6.3 | 2.1 | 9.8 |
| Hemoglobin (g/dL) | 64 | 10.0–20.0 | 3 (4.7) | 3 (4.7) | 58 (90.6) | 14.85 | 2.9 | 22.4 |
| WBCs (cells/μL) | 66 | 6,000–17,000 | 14 (21.2) | 2 (3.0) | 50 (75.8) | 11,800 | 4,700 | 40,200 |
| Neutrophils (cells/μL) | 66 | 3,000–11,500 | 22 (33.3) | 1 (1.5) | 43 (65.2) | 9,334 | 2,875 | 34,170 |
| Lymphocytes (cells/μL) | 66 | 1,000–4,800 | 1 (1.5) | 29 (43.9) | 36 (54.5) | 1,146 | 31 | 5,959 |
| Monocytes (cells/μL) | 66 | 150–1,250 | 9 (13.6) | 4 (6.1) | 53 (80.3) | 581 | 88 | 3,245 |
| Eosinophils (cells/μL) | 66 | 100–1,250 | 1 (1.5) | 20 (30.3) | 45 (68.2) | 228 | 0 | 1,512 |
| Platelets (cells/μL) | 52 | 200,000–500,000 | 8 (15.4) | 6 (11.5) | 38 (73.1) | 344,000 | 50,000 | 881,000 |
See Table 1 for key.
Two of 25 (8.0%) dogs of group 1 had a serum albumin concentration less than the lower limit of the reference range, which was significantly (P < 0.001) fewer, compared with 25 of 42 (59.5%) dogs of group 2 that had a serum albumin concentration less than the lower limit of the reference range (Table 3). In dogs in which serum total protein concentration was measured, significantly (P < 0.001) fewer dogs of group 1 (3/23 [13.0%]) had panhypoproteinemia, compared with 24 of 41 (58.5%) dogs of group 2 with panhypoproteinemia. In dogs in which serum total calcium concentration was measured, significantly (P = 0.006) fewer dogs of group 1 (3/23 [13.0%]) had a decrease in serum total calcium concentration, compared with 20 of 41 (48.8%) dogs of group 2 with a decrease in serum total calcium concentration. In dogs in which serum cholesterol concentration was measured, significantly (P = 0.018) fewer dogs of group 1 (2/22 [9.1%]) had a serum cholesterol concentration less than the reference range, compared with 18 of 41 (43.9%) dogs of group 2 with a serum cholesterol concentration less than the reference range. Significantly (P = 0.044) more dogs of group 1 (11/25 [44.0%]) had an increase in serum sodium concentration, compared with 7 of 42 (16.7%) dogs from group 2 with an increase in serum sodium concentration.
Comparison of results for serum biochemical analysis of dogs in group 1 and group 2.
| Variable | No. of dogs | RR | No. (%) of dogs in group 1 > RR | No. (%) of dogs in group 2 > RR | No. (%) of dogs in group 1 < RR | No. (%) of dogs in group 2 < RR | No. (%) of dogs in group 1 within RR | No. (%) of dogs in group 2 within RR | P value* |
|---|---|---|---|---|---|---|---|---|---|
| ALP (U/L) | 64 | 24–147 | 7 (30.4) | 11 (26.8) | 0 (0) | 1 (2.4) | 16 (69.6) | 29 (70.7) | 0.779† |
| ALT (U/L) | 66 | 10–130 | 4 (16.7) | 9 (21.4) | 0 (0) | 1 (2.4) | 20 (83.3) | 32 (76.2) | 0.755† |
| GGT (U/L) | 62 | 0–25 | 1 (4.5) | 5 (12.5) | NA | NA | 21 (95.5) | 35 (87.5) | 0.409† |
| Albumin (g/dL) | 67 | 2.4–3.6 | 3 (12.0) | 2 (4.8) | 2 (8.0) | 25 (59.5) | 20 (80.0) | 15 (35.7) | < 0.001‡ |
| Globulin (g/dL) | 64 | 1.7–3.8 | 7 (30.4) | 5 (12.2) | 1 (4.3) | 3 (7.3) | 15 (65.2) | 33 (80.5) | 0.194‡ |
| Total protein (g/dL) | 64 | 5.7–7.8 | 4 (17.4) | 1 (2.4) | 3 (13.0) | 24 (58.5) | 16 (69.6) | 16 (39.0) | < 0.001‡ |
| Sodium (mmol/L) | 67 | 139–147 | 11 (44.0) | 7 (16.7) | 2 (8.0) | 3 (7.1) | 12 (48.0) | 32 (76.2) | 0.044‡ |
| Potassium (mmol/L) | 67 | 3.3–4.6 | 2 (8.0) | 3 (7.1) | 5 (20.0) | 4 (9.5) | 18 (72.0) | 35 (83.3) | 0.461‡ |
| Chloride (mmol/L) | 66 | 107–116 | 6 (24.0) | 15 (36.6) | 3 (12.0) | 6 (14.6) | 16 (64.0) | 20 (48.8) | 0.470‡ |
| Calcium (mg/dL) | 64 | 9.3–11.8 | 0 (0) | 1 (2.4) | 3 (13.0) | 20 (48.8) | 20 (87.0) | 20 (48.8) | 0.006† |
| ECO (mmol/L) | 65 | 21–28 | 2 (8.3) | 3 (7.3) | 6 (25.0) | 8 (19.5) | 16 (66.7) | 30 (73.3) | 0.851‡ |
| Glucose (mg/dL) | 65 | 60–135 | 3 (12.5) | 2 (4.9) | 0 (0) | 0 (0) | 21 (87.5) | 39 (95.1) | 0.350† |
| Cholesterol (mg/dL) | 63 | 120–247 | 6 (27.3) | 6 (14.6) | 2 (9.1) | 18 (43.9) | 14 (63.6) | 17 (41.5) | 0.018‡ |
| SUN (mg/dL) | 66 | 5.0–29.0 | 1 (4.2) | 1 (2.4) | 0 (0) | 6 (14.3) | 23 (95.8) | 35 (83.3) | 0.406† |
| Creatinine (mg/dL) | 67 | 0.3–2.0 | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 25 (100) | 42 (100) | — |
| Phosphate (mg/dL) | 67 | 2.9–6.2 | 0 (0) | 1 (2.4) | 1 (4.0) | 7 (16.7) | 24 (96.0) | 34 (81.0) | 0.242† |
| Total bilirubin (mg/dL) | 62 | 0–0.8 | 1 (4.5) | 0 (0) | NA | NA | 21 (95.5) | 40 (100) | 0.355† |
| cPLI (μg/L) | 8 | 0–200 | 0 (0) | 0 (0) | NA | NA | 5 (100) | 3 (100) | ND |
| cTLI (μg/L) | 11 | 5.7–45.2 | 0 (0) | 0 (0) | 0 (0) | 3 (30.0) | 1 (100) | 7 (70.0) | ND |
| Folate (μg/L) | 13 | 7.7–24.4 | 0 (0) | 1 (10.0) | 0 (0) | 2 (20.0) | 3 (100) | 7 (70.0) | ND |
| Cobalamine (ng/L) | 13 | 252–908 | 0 (0) | 1 (10.0) | 0 (0) | 4 (40.0) | 3 (100) | 5 (50.0) | ND |
Group 1 comprised 25 dogs with disease or pathologic changes primarily of the stomach, and group 2 comprised 42 dogs with more generalized disease or pathologic changes of the gastrointestinal system.
Results were considered significant at values of P < 0.05.
Fisher exact test.
χ2 test.
ND = Not determined because sample size was too small. — = Not determined because results for all dogs in both groups were within the reference range.
See Table 1 for remainder of key.
Four of 24 (16.7%) dogs from group 1 had a monocyte count below the lower limit of the reference range, which was significantly (P = 0.007) lower than the proportion of dogs from group 2, in which no dogs had a monocyte count below the lower limit of the reference range (Table 4). Two of 24 (8.3%) dogs from group 1 had an RBC count below the lower limit of the reference range, compared with 12 of 42 (28.6%) dogs from group 2. However, this difference in proportions was not significant (P = 0.066).
Comparison of results for hematologic analysis of dogs in group 1 and group 2.
| Variable | No. of dogs | RR | No. (%) of dogs in group 1 > RR | No. (%) of dogs in group 2 > RR | No. (%) of dogs in group 1 < RR | No. (%) of dogs in group 2 < RR | No. (%) of dogs in group 1 within RR | No. (%) of dogs in group 2 within RR | P value* |
|---|---|---|---|---|---|---|---|---|---|
| PCV (%) | 66 | 31–56 | 0 (0) | 3 (7.1) | 2 (8.3) | 1 (2.4) | 22 (91.7) | 38 (90.5) | 0.234‡ |
| RBCs (× 106/μL) | 66 | 5.5–8.5 | 1 (4.2) | 1 (2.4) | 2 (8.3) | 12 (28.6) | 21 (87.5) | 29 (69.0) | 0.066† |
| Hemoglobin (g/dL) | 64 | 10.0–20.0 | 0 (0) | 3 (7.1) | 2 (9.1) | 1 (2.4) | 20 (90.9) | 38 (90.5) | 0.228‡ |
| WBCs (cells/μL) | 66 | 6,000–17,000 | 6 (25.0) | 8 (19.0) | 1 (4.2) | 1 (2.4) | 17 (70.8) | 33 (78.6) | 0.755† |
| Neutrophils (cells/μL) | 66 | 3,000–11,500 | 6 (25.0) | 16 (38.1) | 0 (0) | 1 (2.4) | 18 (75.0) | 25 (59.5) | 0.416† |
| Lymphocytes (cells/μL) | 66 | 1,000–4,800 | 1 (4.2) | 0 (0) | 11 (45.8) | 18 (42.9) | 12 (50.0) | 24 (57.1) | 1.000† |
| Monocytes (cells/μL) | 66 | 150–1,250 | 5 (20.8) | 4 (9.5) | 4 (16.7) | 0 (0) | 15 (62.5) | 38 (90.5) | 0.007‡ |
| Eosinophils (cells/μL) | 66 | 100–1,250 | 0 (0) | 1 (2.4) | 7 (29.2) | 13 (13.0) | 17 (70.8) | 28 (66.7) | 1.000† |
| Platelets (cells/μL) | 52 | 200,000–500,000 | 1 (5.3) | 7 (21.2) | 3 (15.8) | 3 (9.1) | 15 (78.9) | 23 (69.7) | 0.272‡ |
Radiographs were obtained for 34 of 67 (50.7%) dogs. Findings for the gastrointestinal system or abdomen included gastric foreign bodies (n = 3 [8.8%]), esophageal dilation (2 [5.9%]), and abdominal effusion (2 [5.9%]). No important radiographic findings for the gastrointestinal system were recorded for 18 of 34 (52.9%) dogs.
Abdominal ultrasonography was performed in 62 of 67 (92.5%) dogs. Abdominal effusion was recorded for 13 of 62 (21.0%) dogs. Other common findings were thickening of the gastric wall (n = 13 [21.0%] dogs), enlargement of abdominal lymph nodes (8 [12.9%] dogs), loss of layering of the gastric wall (4 [6.5%] dogs), pancreatic changes (4 [6.5%] dogs), and thickening of the intestinal walls (3 [4.8%] dogs). No significant findings were recorded during abdominal ultrasonography for 19 of 62 (30.6%) dogs. Abdominal effusion was recorded significantly (P = 0.003) more frequently for group 2 (13/41 [31.7%] dogs), compared with results for group 1 (0/21 [0%] dogs). Thickening of the gastric wall was found significantly (P = 0.002) more commonly in dogs of group 1 (9/21 [42.9%] dogs), compared with only 3 of 41 (7.3%) dogs of group 2 that had thickening of the gastric wall.
Gastroduodenoscopy (including collection of biopsy specimens) was performed in 62 of 67 (92.5%) dogs, and biopsy during gastrotomy was performed in the remaining 5 (7.5%) dogs. The most commonly recorded observation during endoscopy for the 62 dogs was of no clinically important findings (17 [27.4%] dogs). Other frequently recorded findings included gastric ulcers, erosions, or mucosal hemorrhage (13 [21.0%] dogs), dilated lacteals (7 [11.3%] dogs), gastric masses (5 [8.1%] dogs), and Physaloptera spp infection (4 [6.5%] dogs).
Lymphoplasmacytic gastritis was the most common gastric histopathologic finding recorded (34/67 [50.7%] dogs). Of the 34 dogs with lymphoplasmacytic gastritis, the severity grade assigned by the veterinary pathologist who examined the sections was minimal (5 [14.7%] dogs), mild (19 [55.9%] dogs), moderate (8 [23.5%] dogs), and marked (1 [2.9%] dogs). A severity grade was not assigned for 1 (2.9%) dog.
Other common findings included various degrees of gastric fibrosis (13/67 [19.4%] dogs), spiral-shaped bacteria (8 [11.9%] dogs), carcinoma or adenocarcinoma (5 [7.5%] dogs), ulcers or erosions (4 [6.0%] dogs), mixed cellular infiltrates (4 [6.0%] dogs), lymphoma (2 [3.0%] dogs), nonspecific gastritis (2 [3.0%] dogs), atrophy (2 [3.0%] dogs), and lymphocytic gastritis (2 [3.0%] dogs). In addition, eosinophilic gastritis, suppurative gastritis, pyogranulomatous gastritis, plasmacytic gastritis, epithelial hyperplasia, lymphoid hyperplasia, foveal hyperplasia, histiocytic gastritis, and other miscellaneous findings were each recorded for 1 (1.5%) dog.
Intestinal biopsy specimens were collected for 60 of 67 (89.6%) dogs. Histologic evidence of intestinal pathologic changes was recorded for 43 of 60 (71.7%) dogs. No clinically relevant findings were recorded for the remaining 17 (28.3%) dogs. Commonly recorded findings were lacteal dilatation (12 [20.0%] dogs), lymphoplasmacytic enteritis (11 [18.3%] dogs), dilatation or abscesses of the intestinal crypts (8 [13.3%] dogs), eosinophilic enteritis (5 [8.3%] dogs), plasmacytic enteritis (5 [8.3%] dogs), and villous blunting (4 [6.7%] dogs). Other findings were recorded less frequently than the aforementioned lesions. Of the 11 dogs in which lymphoplasmacytic enteritis was identified, 2 were graded by the veterinary pathologist who examined the section as minimal disease, 5 as mild disease, 1 as moderate disease, and 1 as severe disease. Two dogs with lymphoplasmacytic enteritis were not classified.
A final diagnosis was recorded for 61 of 67 (91.0%) dogs, whereas no definitive diagnosis was recorded for the remaining 6 (9.0%) dogs. Lymphangiectasia (confirmed or suspected) was recorded as the diagnosis for 12 of the 67 (17.9%) dogs. Other common diagnoses included chronic gastritis of unknown cause (10 [14.9%] dogs), gastric carcinoma or adenocarcinoma (6 [8.9%] dogs, with suspected gastric carcinoma in 1 of these dogs), Physaloptera spp infection (5 [7.5%] dogs), exocrine pancreatic insufficiency (4 [6.0%] dogs), inflammatory bowel disease (4 [6.0%] dogs), and lymphosarcoma (3 [4.5%] dogs).
Discussion
The objectives of the retrospective evaluation reported here were to test the hypothesis that intestinal pathologic changes are often concurrent with gastric pathologic changes in dogs and to characterize the clinical findings and results of diagnostic tests in dogs with gastric histopathologic abnormalities. Of the dogs in which both gastric and intestinal biopsy specimens were collected, 43 of 60 (71.7%) had concurrent intestinal pathologic changes. Thus, it is reasonable to conclude that in the study population of dogs with gastric histopathologic abnormalities, there were concurrent intestinal pathologic changes in most of the patients. However, it is important to mention that a proportion of dogs (17/60 [28.3%]) did not have evidence of concurrent intestinal disease and that the clinical relevance of some of these histopathologic abnormalities is uncertain. These findings support the practice of collecting multiple gastric and duodenal biopsy specimens every time gastroduodenoscopy is performed because failing to obtain a biopsy specimen from either of these sites could result in histologic changes being overlooked. Compared with the dogs in this study, 50% of human patients with antral gastritis or pangastric patterns of lymphocytic gastritis have duodenal villous atrophy, which indicates that concurrent chronic gastritis and intestinal disease are common in human patients.9 Our study did not evaluate the effect that the area of the stomach involved would have on the prevalence of concurrent intestinal disease. Additionally, cases were not limited to dogs with chronic gastritis.
Lymphoplasmacytic gastritis was the most common histopathologic finding recorded in 50.7% of the dogs. This supports results of other studies10–13 in which lymphoplasmacytic gastritis was also found to be the most common histologic change detected during evaluation of gastric biopsy specimens. Most of the lymphoplasmacytic gastritis was of minimal or mild severity (14.7% and 55.9%, respectively), as judged by the veterinary pathologists who examined the histologic sections. However, the importance of such minimal or mild changes is not known. Indeed, investigators in another study5 found evidence that gastritis (diagnosed on the basis of histologic examination) was a common finding, which was reported in 48% of dogs without any clinical signs. The prevalence of histopathologically diagnosed gastritis in dogs without clinical signs of disease in that study5 raises questions about the relationship between histologically determined infiltration of the gastric mucosa with inflammatory cells and clinically relevant gastric disease. Results of the study reported here indicated that lymphoplasmacytic gastritis was a common histopathologic finding in dogs in which gastric biopsy specimens were collected. They also suggested that lymphoplasmacytic gastritis is not specific to dogs with isolated gastric disease because it also was recorded in dogs with more generalized disease or pathologic changes of the gastrointestinal system, including dogs with intestinal diseases such as lymphangiectasia.
Lymphangiectasia and suspected lymphangiectasia were the most common final diagnoses in the study reported here; they were recorded in 17.9% of dogs. This finding concurs with results of a study14 in which investigators evaluated full-thickness intestinal biopsy specimens and determined that lymphangiectasia was the most common intestinal histopathologic finding (reported in 59% of dogs with gastrointestinal signs and 20% of control dogs). The reported prevalence of lymphangiectasia in control dogs without gastrointestinal signs in that study14 raises questions regarding the clinical importance of histopathologically mild forms of lymphangiectasia. The term lymphangiectasia or suspected lymphangiectasia has been used to reflect the controversy in diagnosing this condition without results for examination of a full-thickness biopsy specimen.15 Some clinicians require histologic confirmation based on evaluation of a full-thickness intestinal biopsy specimen to diagnose this condition. Other clinicians make the diagnosis on the basis of a combination of evidence of a protein-losing enteropathy, evaluation of endoscopic biopsy specimens, and response to treatment. When collecting data for the study reported here, the clinician's final diagnosis was recorded for each dog directly from the medical record, which was preferable to the authors attempting to make a diagnosis on the basis of retrospective evaluation of clinical data.
Chronic gastritis was the second most common final diagnosis and was recorded in 14.9% of dogs. An underlying cause of the chronic gastritis was not recorded in any of these dogs. This finding concurs with the notion that the cause of chronic gastritis often remains elusive.1,4,6 Gastric carcinoma, adenocarcinoma, and Physaloptera spp infection were also recorded frequently and should be ruled out in dogs with vomiting that is not self-limiting, does not respond to palliative treatment, or is chronic or recurrent in nature.
The biochemical abnormalities most commonly detected indicated gastrointestinal protein loss (ie, loss of albumin leading to a decrease in serum total protein concentration) with a subsequent decrease in serum total calcium concentration. Hypocholesterolemia was also a commonly recorded finding that may have been attributable to malabsorption. The group of dogs with disease or pathologic changes of both the extragastric portions of the gastrointestinal system and the stomach had hypoalbuminemia, panhypoproteinemia, hypocalcemia, and hypercholesterolemia more frequently than did dogs with disease or pathologic changes primarily of the stomach. Therefore, these abnormalities may act as markers of more diffuse disease of the gastrointestinal system. Many of the dogs of group 2 had protein-losing enteropathies caused by conditions such as lymphangiectasia or inflammatory bowel disease. It is likely that the protein loss in the dogs of our study was from the gastrointestinal tract and that it was attributable in most dogs to concurrent intestinal disease. Evaluation of the fecal α-1 proteinase inhibitor concentration in combination with serum total protein concentration could have been used to provide further evidence of protein-losing enteropathy16 in the hypoproteinemic dogs of our study.
Diarrhea was a less common finding and vomiting a more common finding in the group of dogs with disease or pathologic changes primarily of the stomach, compared with findings for dogs with more generalized disease or pathologic changes of the gastrointestinal system. Therefore, a clinician examining a dog with vomiting but no diarrhea and no evidence of gastrointestinal protein loss should be more suspicious of a disease process limited to the stomach. In contrast, a clinician examining a dog with clinical signs of gastrointestinal disease and evidence of gastrointestinal protein loss should be more suspicious of a more generalized disease process of the gastrointestinal system, which may also affect the choice of initial treatment.
Serum biochemical and hematologic examination are generally performed in animals examined because of clinical signs suggestive of gastric disease to rule out secondary causes of gastrointestinal disease or to identify metabolic sequelae of gastrointestinal disease, rather than for use in making a diagnosis. Noninvasive tests specific for gastric disease would be highly desirable. In humans, serum concentrations of pepsinogen, gastrin-17, and ghrelin are promising as markers of gastritis.17,18 Measurement of serum pepsinogen concentration has not yet been reported to be useful in the investigation of gastric disease in dogs. In addition, serum pepsinogen concentrations increase substantially during the post-prandial period.19 However, evaluation of gene expression and proteomes of normal and diseased gastric tissue may lead to the discovery of markers that could aid in the diagnosis of gastric disease in dogs.
Endoscopy often did not reveal any macroscopic changes, even though there were histopathologic changes. Gastric ulcers and erosions were the most common endoscopic findings. Endoscopic detection of gastric worms allowed a diagnosis of Physaloptera spp infection to be made in 4 dogs. Additionally, gastric mucosal changes (masses or ulcers) were apparent on gastros-copy of all dogs with gastric carcinoma or adenocarcinoma. Similarly, another retrospective study20 revealed that large, deep ulcers with thickened, irregular rims and walls were common findings in dogs with gastric carcinoma.20 This supports the value of gastroscopic examination of the gastric mucosa as a screening tool for neoplastic lesions and allows for biopsy specimens to be collected from abnormal areas.
The study reported here provided support for many of the generally accepted concepts about gastric disease. In addition, our results suggested that intestinal histopathologic changes are often concurrent in dogs with gastric histopathologic changes and that the most common histopathologic change of the stomach is lymphoplasmacytic gastritis. Assigning the dogs to 2 groups on the basis of the site of disease was helpful in distinguishing dogs with isolated gastric disease or pathologic changes from dogs in which other parts of the gastrointestinal system were concurrently affected. However, by its nature, this process was subjective and possibly introduced bias. Another potential source of bias in this study was the retrospective design. The decision to collect intestinal biopsy specimens was at the discretion of each clinician. Therefore, this study may have overestimated the prevalence of concurrent intestinal pathologic changes because intestinal biopsy specimens were more likely to have been collected from dogs in which the clinician was suspicious of intestinal involvement. However, clinicians at this institution routinely collect both gastric and intestinal biopsy specimens when performing gastroduodenoscopy, thus reducing the effect of this bias. This was supported by the fact that 89.6% of the dogs in the study had biopsy specimens collected from both sites. Interpretation of the histopathologic findings was complicated because > 1 veterinary pathologist reviewed the histologic sections and the grading of histologic sections was not based on a standardized grading criterion.
Inflammation was the most common type of pathologic process of the stomach of dogs recorded in the study reported here. Results of histologic examination of normal gastric tissue need to be better characterized to enable clinicians and researchers to understand the importance of common changes, such as minimal or mild lymphoplasmacytic gastritis. The cause of gastritis often remains unknown, but the breakdown of mucosal immune tolerance as a result of intraluminal bacterial or dietary antigens is a possible factor in the development of chronic gastric inflammation. The use of more subjective grading criteria for the histopathologic diagnosis of gastric inflammation (such as the Updated Sydney System21 used for humans or the system developed by a World Small Animal Veterinary Association study group22) will also aid in our ability to understand those changes in small animal patients that are important and reduce interobserver variation in interpretation. Similarly, relatively little is known about the etiopathogenesis of gastric neoplasia in dogs. Additional studies of gastric microflora, mucosal cytokine expression, novel biomarkers for gastric inflammation, and tumorgenesis of gastric neoplasia in dogs will hopefully enable the development of greater insight into the pathogenesis of gastric disease in dogs.
References
- 1.↑
Simpson KW. Diseases of the stomach. In: Hall E, Simpson J, Williams DA, eds. BSAVA manual of small animal gastroenterology. 2nd ed. Telford, England: British Small Animal Veterinary Association, 2004;151–175.
- 2.
Argenzio RA. Gastrointestinal motility. In: Reece WO, ed. Dukes' physiology of domestic animals. 12th ed. Ithaca, NY: Cornell University Press, 2004;391–404.
- 3.↑
Hubbard K, Skelly BJ, McKelvie J, et al. Risk of vomiting and diarrhea in dogs. Vet Rec 2007;161:755–757.
- 4.↑
Simpson KW. Gastric disease. In: Ettinger SJ, Feldman EC, eds. Textbook of veterinary internal medicine. 6th ed. Philadelphia: WB Saunders, 2004;1310–1331.
- 5.↑
Happonen I, Linden J, Saari S, et al. Detection and effects of Helicobacters in healthy dogs and dogs with signs of gastritis. J Am Vet Med Assoc 1998;213:1767–1774.
- 7.↑
Mansell J, Willard MD. Biopsy of the gastrointestinal tract. Vet Clin North Am Small Anim Pract 2003;33:1099–1116.
- 8.↑
Willard MD, Jergens AE, Duncan RB, et al. Interobserver variation among histopathologic evaluations of intestinal tissues from dogs and cats. J Am Vet Med Assoc 2002;220:1177–1182.
- 9.↑
Hayat M, Arora DS, Wyatt JI, et al. The pattern of involvement of the gastric mucosa in lymphocytic gastritis is predictive of the presence of duodenal pathology. J Clin Pathol 1999;52:815–819.
- 10.
Wiinberg B, Spohr A, Dietz DD, et al. Quantitative analysis of inflammatory and immune responses in dogs with gastritis and their relationship to Helicobacter species infection. J Vet Intern Med 2005;19:4–14.
- 11.
Happonen I, Saari S, Castren L, et al. Occurrence and topographical mapping of gastric Helicobacter-like organisms and their association with histological changes in apparently healthy dogs and cats. Zentralbl Veterinarmed [A] 1996;43:305–315.
- 12.
Geyer C, Colbatzky F, Lechner J, et al. Occurrence of spiralshaped bacteria in gastric biopsies of dogs and cats. Vet Rec 1993;133:18–19.
- 13.
Eaton KA, Dewhirst FE, Paster BJ, et al. Prevalence and varieties of Helicobacter species in dogs from random sources and pet dogs: animal and public health implications. J Clin Microbiol 1996;34:3165–3170.
- 14.↑
Kleinschmidt S, Meneses F, Nolte I, et al. Retrospective study on the diagnostic value of full-thickness biopsies from the stomach and intestines of dogs with chronic gastrointestinal disease symptoms. Vet Pathol 2006;43:1000–1003.
- 15.↑
Peterson PB, Willard MD. Protein-losing enteropathies. Vet Clin North Am Small Anim Pract 2003;33:1061–1082.
- 16.↑
Murphy KF, German AJ, Ruaux CG, et al. Fecal alpha1-proteinase inhibitor concentration in dogs with chronic gastrointestinal disease. Vet Clin Pathol 2003;32:67–72.
- 17.
Plebani M, Basso D. Non-invasive assessment of chronic liver and gastric diseases. Clin Chim Acta 2007;381:39–49.
- 18.
Checchi S, Montanaro A, Pasqui L, et al. Serum ghrelin as a marker of atrophic body gastritis in patients with parietal cell antibodies. J Clin Endocrinol Metab 2007;92:4346–4351.
- 19.↑
Suchodolski JS, Steiner JM, Ruaux CG, et al. Serum concentrations of pepsinogen A in healthy dogs after food deprivation and after feeding. Am J Vet Res 2003;64:1146–1150.
- 20.↑
Sullivan M, Lee R, Fisher EW, et al. A study of 31 cases of gastric carcinoma in dogs. Vet Rec 1987;120:79–83.
- 21.↑
Dixon MF, Genta RM, Yardley JH, et al. Classification and grading of gastritis. The updated Sydney System. International Workshop on the Histopathology of Gastritis, Houston 1994. Am J Pathol 1996;20:1161–1181.
- 22.↑
Day MJ, Bilzer T, Mansell J, et al. Histopathological standards for the diagnosis of gastrointestinal inflammation in endoscopic biopsy samples from the dog and cat: a report from the World Small Animal Veterinary Association Gastrointestinal Standardization Group. J Comp Pathol 2008;138(suppl 1):S1–S43.
