SEARCH

SEARCH BY CITATION

Keywords:

  • Gastric neoplasia;
  • Gastroscopy;
  • Paraneoplastic syndrome;
  • Squamous cell carcinoma

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References

Background: Gastric neoplasia of horses is incompletely described.

Objective: Provide history, clinical signs, and clinicopathological and pathological findings associated with gastric neoplasia in horses.

Animals: Twenty-four horses with gastric neoplasia.

Methods: Retrospective study. History, clinical signs, and clinicopathological and pathologic findings in horses diagnosed histologically with gastric neoplasia were reviewed.

Results: Horses ranged in age from 9 to 25 years (median 18 years at presentation). There was no apparent breed or sex predisposition. The most common presenting complaints were inappetance (17/24), weight loss (14/24), lethargy (7/24), hypersalivation (7/24), colic (5/24), and fever (5/24). The most consistent clinical signs were tachypnea (10/19), decreased borborygmi (8/19), and low body weight (7/17). Useful diagnostic tests included rectal examination, routine blood analysis, gastroscopy, abdominocentesis, and transabdominal ultrasound examination. Anemia was the most common hematologic abnormality encountered (7/19), and hypercalcemia of malignancy was seen in 4/16 horses. Squamous cell carcinoma was the most common tumor identified (19/24), and was most often (14/19) found as a single ulcerated, necrotic mass in the nonglandular portion of the stomach. Other gastric neoplasms encountered were leiomyoma (n=2), mesothelioma (n=1), adenocarcinoma (n=1), and lymphoma (n=1). Metastatic neoplasia was found in 18/23 horses. The median time from onset of clinical signs to death was 4 weeks, and all horses died or were euthanized because of gastric neoplasia.

Conclusions: Squamous cell carcinoma is the most common primary gastric neoplasia in horses. The survival time after diagnosis of gastric neoplasia in horses is short.

Abbreviations:
SCC

squamous cell carcinoma

TNCC

total nucleated cell count

WBC

white blood cell

Gastric neoplasia is rare in horses, accounting for 1.5% of all equine neoplasms.1 The most common equine gastric tumor is primary squamous cell carcinoma (SCC),1 which represents 3–4% of all equine SCC.1–3 Although it has been reported that gastric SCC is the most common primary neoplasm of the equine gastrointestinal tract,2 1 study found an equivalent prevalence of primary intestinal lymphoma.4 Other gastric neoplasms reported in horses include adenocarcinoma,5 leiomyosarcoma,6 gastrointestinal stromal tumor,7 papilloma,1 and benign polyp.8 History and clinical signs associated with equine gastric neoplasia are often vague and seldom localize the disease process to the stomach, making antemortem diagnosis difficult. Therefore, the objective of the present retrospective study was to review the signalment, presenting complaint, physical examination findings, and clinicopathologic features of horses diagnosed histopathologically with gastric neoplasia.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References

A computer database was searched for horses examined at the Veterinary Medical Teaching Hospital, University of California at Davis, between 1990 and 2008. Horses were included in this study if a histopathologic diagnosis of gastric neoplasia was made via antemortem biopsy or at postmortem examination. Twenty-four horses diagnosed with gastric neoplasia were included in this study. Gastric neoplasms encountered were SCC (19 horses), leiomyoma (2), lymphoma (1), mesothelioma (1), and adenocarcinoma (1). The data derived from the medical records included signalment, history, clinical signs, laboratory results (CBC, biochemistry, peritoneal fluid cytology), gastroscopy results, abdominal ultrasound examination findings, treatment and outcome, as well as postmortem examination results. Data were summarized using descriptive statistics and are reported as median and range in this retrospective cohort study.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References

Twenty-four horses met the inclusion criteria. Nine (38%) were Quarter Horses, 5 (21%) were Arabians, 4 (16%) were Quarter Horse crosses, and 6 (25%) were other breeds, including 2 Shetland ponies, 1 National Show Horse, 1 Appaloosa, 1 Thoroughbred, and 1 Clydesdale. Fourteen (58%) were castrated males, 9 (38%) were females, and 1 (4%) was an intact male. The horses ranged in age from 9 to 25 years, with a median of 18 years.

The median duration of clinical signs before examination was 3 weeks (range 1 day to 7.5 months). Historical signs included inappetance in 17 (71%) cases, weight loss in 14 (58%), lethargy in 7 (29%), hypersalivation in 7 (29%), colic in 5 (21%), pyrexia in 5 (21%), halitosis in 2 (8%), and choke and weakness associated with blood loss in 1 each (4%). Treatments before presentation included nonsteroidal anti-inflammatory drugs in 13 (54%) cases, antimicrobial drugs in 12 (50%), and gastroprotectant drugs in 3 (13%).

The following data are derived from horses with specifically recorded clinical signs at the time of initial examination. Thirteen of 16 (81%) horses had inappetance, and 10 of 19 (53%) horses had tachypnea (>20 breaths/min) with a median respiratory rate from all included horses of 24 breaths/min (range 12–56). Eight of 19 (42%) horses had decreased borborygmi, and 7 of 17 (41%) appeared lethargic. Seven of 17 (41%) horses were underweight; the median body condition score was 3.75/9 (range 2–5/9).9 Seven of 19 (37%) horses had tachycardia (>48 beats/min), with a median heart rate from all included horses of 44 beats/min (range 24–92). Seven of 22 (32%) horses demonstrated hypersalivation, and 4 of 18 (22%) horses had a fever (>101.5°F) with a median rectal temperature of 100.3°F (range 98–104.6). Two of 22 (9%) horses had halitosis, 1 demonstrated copious amounts of green discharge from the nose and mouth consistent with esophageal obstruction, and 1 had pale mucous membranes and tachycardia associated with acute blood loss. A rectal examination was recorded for 6 of 24 horses, 4 (67%) of which had abnormal findings. An abnormal mass was palpated in 2 (33%) of the 6 horses on which rectal examination was reported, both of which had gastric SCC with diffuse intra-abdominal metastases. Small intestinal distention was detected on rectal examination in 1 of 6 (17%) horses with gastric SCC and a concurrent strangulating lipoma. Finally, 1 of 6 (17%) horses had a cecal impaction on rectal examination and was subsequently diagnosed with gastric serosal mesothelioma.

CBC were available from 19 horses and revealed a median hematocrit of 35.6% (range 7.7–53). Seven (37%) horses were anemic (PCV < 30%). The peripheral white blood cell (WBC) count ranged from 6,490 to 28,840/μL, with a median of 12,060/μL. Five (26%) horses had leukocytosis (>11,600 WBCs/μL). The median neutrophil count was 9,032/μL (range 5,146–26,533), with 5 (26%) demonstrating mature neutrophilia (>6,800 neutrophils/μL). Three (16%) horses had increased immature (band) neutrophils in the peripheral blood. Eight (42%) horses were lymphopenic (<2,000 lymphocytes/μL), and the median lymphocyte count was 1,169/μL (range 578–3,120). Thirteen (68%) horses had a high plasma fibrinogen concentration (>400 mg/dL) with a median of 600 mg/dL (range 300–900).

Serum biochemical analyses were available for 16 horses. The median plasma total protein concentration was within the normal reference range (7 g/dL, range 4.2–8.9 g/dL), but 9 (56%) horses were hypoalbuminemic (<2.3 g/dL) with a median albumin concentration of 2.0 g/dL (range 1.6–3.1). In addition, 11 (58%) horses were hyperglobulinemic (>4.7 g/dL), and the median globulin concentration from all included horses was 5.0 g/dL (range 2.4–6.6). Of the 11 horses with an increased globulin, 10 were ultimately diagnosed with gastric SCC and 1 with gastric adenocarcinoma. Hepatic disease was suspected in 3 of 16 (19%) horses based on increased concentrations of both sorbitol dehydrogenase (>8 IU/L) and γ-glutamyl transferase (>22 IU/L), and neoplastic infiltrates were confirmed histologically in all 3 cases. Hypercalcemia (total calcium > 13.6 mg/dL) was observed in 4 (25%) horses, 3 of which were concurrently hypoalbuminemic.

Abdominocentesis was performed in 13 horses. Of these, 9 (69%) had an increased total nucleated cell count (TNCC > 2,000 cells/μL). The median TNCC in the abdominal fluid from all 13 horses sampled was 9,100/μL (range 100–180,000). Nine of 13 (69%) horses had an increased abdominal fluid protein concentration (>2.0 g/dL), 6 of which had a concurrent increase in TNCC. The median abdominal protein concentration from all horses sampled was 4.4 g/dL (range 1–7). The fluid was interpreted as exudate in 6 (46%), modified transudate in 4 (31%), transudate in 2 (15%) horses, and neoplasia in 1 (8%). The latter horse had abdominal fluid with large anaplastic nucleated cells with deeply basophilic cytoplasm, high nuclear to cytoplasmic ratio, and occasional mitotic figures suggestive of epithelial neoplasia. SCC was found in the stomach, spleen, liver, peritoneum, and hilar lymph nodes in this animal. Peritoneal fluid analysis from horses ultimately diagnosed with gastric SCC yielded a median WBC count of 8,550/μL and a total protein of 4.4 g/dL, while horses diagnosed with gastric neoplasia other than SCC had a median WBC count of 9,550/μL and a total protein of 3.75 g/dL.

Gastroscopy was performed in 13 horses. Of these, an abnormal proliferative process was visualized in 10 (77%). Eight of 10 (80%) horses had a single, multinodular mass in the squamous portion of the stomach, while 2 demonstrated diffuse, irregular proliferation of the squamous mucosa. One horse with a single gastric mass had an additional visible mass in the distal esophagus. Four of 13 (31%) horses demonstrated gastric ulceration, only 1 of which was in addition to a visible mass. One horse with a single intraluminal gastric mass had copious amounts of red-tinged gastric fluid.

Transabdominal ultrasound examinations were performed in 5 horses. Four of these horses had a complete examination of all abdominal structures, including both kidneys, both liver lobes, spleen, stomach, duodenum, small intestine, large colon, and cecum. The remaining horse had a partial ultrasound examination performed on the left side of the abdomen.

Four horses that underwent abdominal ultrasound evaluation were diagnosed with SCC and 1 with adenocarcinoma. Intra-abdominal masses were seen in 5 of 5 (100%) horses and included gastric, hepatic, splenic, omental, and peritoneal masses. Gastric masses were seen in 3 of the 4 (75%) horses with SCC. Two of 3 gastric masses were large, measuring from 16 to 22 cm in diameter. The 3rd mass was small, measuring 2.6 cm × 4.2 cm. Thickened omentum containing metastatic masses was seen between the spleen and stomach in the remaining horse with SCC. Multiple omental, hepatic, and peritoneal masses were seen in the horse with adenocarcinoma. All horses examined had metastatic disease. Severe peritoneal effusion was seen in 4 of 5 (80%) horses. Abnormalities of the spleen and right liver lobe were seen in 4 of 5 (80%) horses and included metastatic masses in 3 horses each. The remaining abnormalities included mild rounding of the right liver lobe and microsplenia, seen in 1 horse each. Other intra-abdominal findings included thickened small intestine and large colon seen in 3 and 2 horses, respectively. Bilateral renomegaly was seen in 1 horse with SCC that demonstrated multiple acute renal infarcts on postmortem examination.

Ultrasound-guided procedures were performed in 2 of 5 (40%) horses. Liver biopsy, performed in the horse with adenocarcinoma, yielded an erroneous diagnosis of poorly differentiated sarcoma. Ultrasound-guided biopsy of a splenic mass yielded a diagnosis of SCC in a horse with primary gastric SCC.

Of the 24 horses in the study, 22 were euthanized in the hospital, 1 died, and 1 was dead on arrival. The duration of onset of clinical signs to time of death ranged from the same day to 1 year, with the median elapsed time from onset of signs to death in horses with SCC being 4 weeks and in horses with non-SCC gastric neoplasia being 5 weeks.

A definitive antemortem diagnosis was made by biopsy in 6 (25%) horses, 3 with SCC, 1 with mesothelioma, 1 with adenocarcinoma, and 1 with lymphoma. Of the 3 horses with SCC, a gastroscopy-mediated mass biopsy was performed in 2, and a transabdominal ultrasound-guided biopsy of a splenic mass was performed in 1. In 1 horse, a peritoneal mass collected during laparotomy yielded a diagnosis of adenocarcinoma but was interpreted as mesothelioma at necropsy. Transabdominal ultrasound-guided biopsy facilitated diagnosis of adenocarcinoma from a hepatic mass in 1 horse. Finally, lymphoma was diagnosed antemortem in 1 horse from a peripheral lymph node biopsy. The diagnosis of gastric neoplasia was confirmed at postmortem examination in the remaining 18 (75%) horses. Overall, 19 of 24 (79%) horses were diagnosed with gastric SCC, 2 (9%) with gastric leiomyoma, and 1 each (4%) of mesothelioma, adenocarcinoma, and lymphoma.

Of the 19 horses diagnosed with gastric SCC, a mucosal gastric mass or masses were present on postmortem examination in 16 (84%). A single mass was found in 14 of these horses, while 2 had multiple (>2) mucosal masses. Three of 19 (16%) horses diagnosed with gastric SCC had diffusely thickened gastric mucosa in the squamous portion of the stomach, 2 with multiple focally extensive ulcers of the nonglandular gastric mucosa. Of the 12 horses with SCC that had a gastroscopy performed, a single mass was visualized and then confirmed by necropsy in 7. Gastric ulcers were visualized by gastroscopy in 4 horses and were confirmed at necropsy, 1 of which had a single mass along the margo plicatus identified at necropsy that was not seen with gastroscopy. Gastroscopy identified 1 horse with multiple small proliferative lesions and gastric ulcers throughout the squamous portion of the stomach that were confirmed at necropsy. Metastasis was identified in 13 of 19 (68%) horses with gastric SCC. Of these, 5 demonstrated carcinomatosis, 1 hematogenous/lymphatic metastasis, 1 local invasion, and 7 had both carcinomatosis and hematogenous/lymphatic metastasis. Of the 5 horses that did not demonstrate metastatic neoplasia at necropsy, 1 presented with acute colic associated with strangulating venous infarction of the jejunum, and 1 with acute internal blood loss from a single mucosal gastric SCC.

Seventeen of 19 (89%) horses with gastric SCC demonstrated neoplastic transmural extension to the gastric serosa, while 2 (11%) horses had tumors that were contained within the gastric mucosa. These 2 horses did not have evidence of metastasis.

Gastric leiomyoma was found in 2 of 24 (8%) horses, both of which had a large solitary mass arising from the tunica muscularis of the squamous portion of the stomach. One was limited to the tunica muscularis and did not extend into the gastric mucosa, while the other demonstrated a rough, irregular surface with necrotic mucosa. Neither mass infiltrated the gastric serosa or metastasized. Gastroscopy was used to identify the rough, irregular mass in 1 horse with gastric leiomyoma.

One of 24 horses (4%) was each diagnosed with gastric mesothelioma, adenocarcinoma, and lymphoma. Multiple small nodules were present on the gastric serosa in the horse with mesothelioma, along with similar nodules throughout the parietal peritoneum, mesentery, and the serosa of the spleen and liver. In 1 horse, lymphoma was found in the submucosa and muscular layers of the stomach in addition to multiple lymph nodes. The horse diagnosed with gastric adenocarcinoma had multiple small masses present on the gastric serosa, parietal peritoneum, omentum, liver and spleen, and appeared to originate from the pancreas or biliary tree. The tumor did not originate from the glandular epithelium of the stomach. Overall, there was evidence of metastatic neoplasia in 18 of 23 (78%) horses.

Medications administered during hospitalization before definitive diagnosis of gastric neoplasia included gastroprotectant drugs in 6 of 19 (32%) horses, antimicrobial drugs in 8 (42%), and nonsteroidal anti-inflammatory drugs in 5 (26%). None of the horses were discharged from the hospital after a definitive histopathologic diagnosis of gastric neoplasia was made.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References

SCC was the most common gastric neoplasm encountered in horses referred to our hospital. This tumor originates from squamous epithelial cells in the nonglandular portion of the equine stomach. Leiomyoma was diagnosed in 2 horses. Adenocarcinoma may originate from glandular epithelium in the stomach, but was found only involving the gastric serosa in 1 horse in the current study and represented metastasis. Lymphoma, which typically originates from the submucosal layer of the gastric wall, was diagnosed in 1 horse. Finally, mesothelioma originating from the peritoneal mesothelium was found in 1 horse. Unlike the horse, adenocarcinoma accounts for up to 90% of malignant gastric tumors in dogs.10–12 This is not surprising given the absence of gastric squamous epithelium in the canine stomach. However, the risk of equine gastric adenocarcinoma appears to be remarkably small despite the 1 : 1 ratio of glandular to squamous mucosa in the equine stomach. Similarly, gastric adenocarcinoma is exceedingly rare in cats.13,14 Gastric leiomyoma was much less common than gastric SCC in our study, which is consistent with previous literature.1,2 No specific etiology for the development of gastric neoplasia has been identified in domestic animals, making it difficult to explain the variation in tumor type frequency.

Mature horses were more likely to be diagnosed with gastric neoplasia. The mean age at presentation was 18 years, both for total cases and for those with gastric SCC. This is higher than in previous reports in which the mean age at admission was 12 years (range 7–15 years).6,15–22 No breed or sex predilection has been identified. The mean duration of clinical signs before presentation was 3 weeks in the current study, compared with 6 weeks from combined case reports of equine gastric SCC.21 Although insidious onset of clinical signs is consistent with a progressively expanding mass or masses, extreme variation from acute onset of signs to several years of mild signs has been reported and is consistent with our findings.20–22

Many clinical signs and clinicopathological findings in horses with gastric neoplasia are nonspecific. Inappetance was the most common presenting complaint in the current study, and was most likely because of 1 or more space-occupying lesion(s) in the gastric lumen, pain associated with the mass(es), gastric ulceration, peritonitis and/or esophagitis, esophageal obstruction, or lethargy. A reduced or absent appetite is the most common historical complaint in many reports of equine gastric neoplasia.6,17–21,23 Weight loss was also common, and was likely caused by anorexia and cachexia associated with neoplasia. Weight loss has been reported in gastric neoplasia and intestinal neoplasia of horses.4,6,17,18,20,24 Five of 24 (21%) horses in our study had a history of colic, including 2 with SCC, 1 with mesothelioma, and both horses with leiomyoma. Colic has been previously described in horses with gastric SCC,15,20,21,23 leiomyosarcoma,6 and adenocarcinoma,5 and may be because of pain associated with impaired gastric outflow and/or abdominal metastasis. Both horses in our study that were diagnosed with gastric leiomyoma had a history of recurrent colic, which is in agreement with previous reports of alimentary smooth-muscle tumors causing abdominal pain.4,25,26 Anxiety, abdominal pain, and anemia were the most likely causes of tachycardia in 37% of horses.

Hypersalivation was seen in 32% of horses in this report, all of which demonstrated a narrowing of the distal esophageal lumen into the cardia. This was because of distal esophageal hypertrophy in most horses and neoplastic infiltrates in others. The apparent increase in salivation may actually be due to salivary reflux secondary to decreased passage of saliva through the cardia or persistently increased intragastric pressure. Hypersalivation was previously reported in 2 horses with gastric SCC that had narrowed esophageal lumen diameter, consistent with our findings.15,24 Esophageal and/or gastric ulceration may have contributed to hypersalivation in some horses. Not surprisingly, esophageal obstruction occurs in horses with gastric neoplasia and narrowed cardia lumina,21,27 and was part of the history in 1 horse in the current study. Fever was described in the history or was found at presentation in 7 (29%) horses, 6 of which had gastric SCC. Fever was likely caused by inflammation, secondary to neoplasia, and may have been an immune response to necrotic mass surfaces or leakage of gastric ingesta into the peritoneal cavity secondary to SCC in some horses. Pyrexia occurs in approximately 50% of horses with gastric neoplasia.6,19–21 Tachypnea was a common clinical sign encountered at initial presentation in the current study. Increased respiratory rate is a nonspecific finding, and was attributed to pleural metastasis in only 2 (20%) of 10 horses in this study. In 1 study, respiratory distress caused by pleural effusion secondary to metastatic gastric SCC was the primary presenting complaint.22 In contrast, most horses with pleural metastasis present with a normal15,19,21 or slightly increased18 respiratory rate. Other potential causes of tachypnea in our study included systemic inflammatory response syndrome secondary to septic peritonitis, severe peritoneal effusion, colic, fever, and anemia. Halitosis was rarely reported in this study, in part because of the subjective nature of this finding. A single, ulcerated and necrotic SCC mass was found in the stomach of both horses that presented with halitosis. One horse presented with severe anemia secondary to a bleeding gastric SCC tumor, suggesting that gastric neoplasia should be considered as a differential diagnosis in horses that present with signs of acute internal hemorrhage. Although a rectal examination was performed in only 25% of horses in this study, most were abnormal. Rectal palpation was valuable in detecting an intra-abdominal mass in some horses with gastric neoplasia, as reported in previous literature,19–22 although normal rectal palpation does not exclude metastatic gastric neoplasia.17,19,20

Anemia was the most common hematological abnormality found in this study and was most often described as normocytic and normochromic, implying anemia of chronic inflammation. Internal hemorrhage was found in 2 horses with gastric SCC, 1 with a bleeding intragastric tumor and 1 with a ruptured metastatic liver mass. Anemia has been previously reported in 7 of 16 (44%) horses with gastric SCC, which is consistent with our findings.21 Six horses with gastric SCC had neutrophilia, hyperfibrinogenemia, and hyperglobulinemia consistent with chronic inflammation. Hypoalbuminemia was the most common biochemical abnormality and was associated with hyperglobulinemia in 78% of cases. Other causes of hypoalbuminemia include liver dysfunction and protein losing enteropathy associated with intestinal metastasis. Hyperglobulinemia was likely because of chronic inflammation or septic peritonitis associated with neoplasia. Hypoalbuminemia and hyperglobulinemia have been reported in previous cases of equine gastric SCC.18,20,22 Although increased liver enzyme activity in serum was present in only 19% of horses, all were associated with liver metastases from gastric SCC. Consistent with findings from other studies,4,28 hepatic neoplasia should thus be considered in horses with increased serum concentration of liver enzymes. Moreover, horses with primary gastric adenocarcinoma might present with signs of hepatic encephalopathy including lethargy, transient mania, blindness, and head pressing that are associated with metastatic tumors in the liver.5,28. Hypercalcemia was found in 25% of horses tested, all of which had gastric SCC. These increases were profound, even in the face of hypoalbuminemia. The most common paraneoplastic syndrome described in the horse is because of the release of parathyroid hormone-related peptide, which causes calcium resorption from the bone and subsequent hypercalcemia. The production of this hormone-like substance has been associated with SCC, multiple myeloma, and lymphoma in the horse.3,19,27,29–31 Hypercalcemia of malignancy is a common finding in small animals with carcinoma,29,32,33 although normal blood work does not rule out gastric SCC or metastasis.15,17,21

Peritoneal fluid was analyzed in 13 horses, 11 of which were abnormal. The fluid was determined to be exudate in 6 cases, 5 of which had gastric SCC, and 1 had an adenocarcinoma. All horses with exudative fluid had neoplastic infiltrates involving multiple intra-abdominal organs. Two of these horses had septic peritonitis secondary to SCC-induced gastric leakage. Nonseptic exudate was the most common interpretation of peritoneal fluid in several horses with gastric SCC or leiomyosarcoma.6,17,18,20 In 1 study, peritoneal fluid analysis confirmed the presence of intra-abdominal neoplasia in 44% of equine patients examined.34 A 2nd study reported neoplastic cells in peritoneal fluid from 1 of 6 horses with gastric SCC, all 6 of which demonstrated metastasis to the liver.21 A 3rd study reported neoplastic epithelial cells in peritoneal fluid from 3 of 5 horses with gastric SCC.20 In addition, neoplastic cells may be identified in pleural fluid from horses with metastasis to the thoracic cavity.18,22 In the current study, neoplastic cells were obtained from abdominocentesis in 1 horse that demonstrated liver rupture and subsequent hemoabdomen secondary to metastatic gastric SCC. Importantly, the absence of inflammatory or exfoliative neoplastic cells does not exclude a diagnosis of gastric SCC.15,21,34

Gastroscopy was used to aid in diagnosis of gastric neoplasia in over half of the cases in this study. Based on postmortem findings, it may have been useful in visualizing an abnormal proliferative process in the majority of cases; however, the gastric mucosa was normal or only ulcerated in some horses with gastric neoplasia. Additionally, neoplastic infiltrates affecting the distal esophagus and/or distal esophageal hypertrophy may prevent passage of the gastroscope into the stomach. Gastroscopy may appear normal in horses with gastric leiomyoma, leiomyosarcoma, or serosal tumors, because the mucosa is typically unaffected in these cases. However, a mural mass may create a convex gastric bulge that can be detected by gastroscopy. Gastroscopy-guided biopsies may also be attempted, although 5 of 7 attempts in the current study yielded nondiagnostic samples or identified gastritis. The majority of previous case reports were published before gastroscopes were readily available, although gastroscopy did not reveal abnormalities in 1 horse with gastric leiomyosarcoma.6 Taken together, gastroscopy should be attempted in horses that present with signs and clinicopathologic data that suggest gastrointestinal disease, but failure to identify abnormalities on gastroscopy do not rule out gastric neoplasia.

Intra-abdominal masses were seen in 100% of horses that underwent transabdominal ultrasound examination. Gastric ultrasound has been utilized to identify gastritis, gastric impaction, gastric distension, and gastric neoplasia.35–38 The ultrasonographic findings in horses with gastric SCC include visualization of a heterogeneous mural gastric mass in the left side of the abdomen associated with the greater curvature of the stomach,37,38 similar to 2 of the 3 gastric masses reported here. Metastatic lesions are commonly described in cases of SCC37–39 and were seen affecting the liver, spleen, omentum, and/or peritoneum in all horses that underwent transabdominal ultrasound examination. Although transabdominal ultrasound examination was utilized in only 5 of 24 horses in this study, ultrasound findings correlated closely with postmortem findings, demonstrating its value as a diagnostic procedure in horses with clinicopathological features suggestive of gastric neoplasia. Ultrasound-guided procedures such as aspiration, biopsy or both were underutilized in this study and should be considered when accessible primary or metastatic lesions are visualized.

A primary gastric tumor was found in 21 of 24 (88%) horses in the current study, with 2 cases (adenocarcinoma, mesothelioma) involving metastatic gastric lesions and 1 (multicentric lymphoma) that is undetermined. Consistent with several reports, all cases of gastric SCC were primary neoplasms, with most demonstrating a single gastric mass in the nonglandular portion of the stomach.15,17–22,24

Metastatic neoplasia was found on postmortem examination in nearly 80% of cases in this study, which emphasizes the advanced stage of disease by the time clinical signs are apparent and diagnostic tests are performed. Most horses with gastric SCC were diagnosed with both carcinomatosis and hematogenous/lymphatic metastasis to multiple organs. Carcinomatosis refers to implantation of neoplastic cells onto the serosal surface of the body cavity or organs, while hematogenous or lymphatic metastasis refers to spread of tumor cells to distant sites through vessels. Only 1 horse in this study demonstrated hematogenous/lymphatic metastasis alone, and 1 had a tumor that extended into the diaphragm and mesenteric root. As expected given that leiomyoma is a benign tumor, neither horse with gastric leiomyoma had evidence of metastatic neoplasia. The horse with gastric lymphoma had multicentric disease affecting multiple lymph nodes. In the horse with adenocarcinoma of the gastric serosa, the tumor was not contiguous with the gastric mucosa and likely represented metastasis from a primary pancreatic or biliary tumor. Finally, the horse with mesothelioma involving the serosa of the stomach had diffuse infiltration and disseminated tumors throughout the abdomen. Similar to the overall metastatic rate found in our study, metastasis is found in 58 to 74% of small animals at the time of gastric neoplasia diagnosis.11–13

Treatment of most abdominal neoplasia, including gastric neoplasia, is restricted in horses because of the poor accessibility of much of the abdominal viscera, the advanced stage of disease by the time a diagnosis is reached, and financial constraints involved with extensive surgical procedures and chemotherapeutics. No horses in this study were discharged from the hospital after a diagnosis of gastric neoplasia was made. Future advances in medical and surgical treatment modalities may increase the survival time in horses diagnosed with gastric neoplasia if the primary gastric tumor is detected before metastasis.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References