Gastric Erosion And Ulceration
Gastric erosions and ulcers are associated with a number of primary gastric and non-gastric disorders (Table Association of Gastric Ulceration and Erosion with Specific Diseases). Clinical signs range in duration and severity, from acute to chronic and mild to life threatening. The pathomechanisms underlying gastric damage can be broadly attributed to impairment of the gastric mucosal barrier (defined above) through direct injury, interference with gastroprotective prostaglandins (PGE2), mucous or bicarbonate, decreased blood flow, and hypersecretion or gastric acid.
Association of Gastric Ulceration and Erosion with Specific Diseases
|Gastric Problem||Related Diseases|
|Metabolic / Endocrine||Hypoadrenocorticism, uremia, liver disease, mastocytosis, d. i. c.|
|Hypergastrinemia and other APUDomas|
|Neoplastic||Leiomyoma, adenocarcinoma, lymphosarcoma|
|Drug-induced||Nonsteroidal and steroidal anti-inflammatories|
|Idiopathic||Stress, spinal surgery, exercise induced (sled dogs)|
Perhaps the most predictable recipe for gastric erosion is the combination of a non-steroidal anti-inflammatory and a glucocorticoid, either alone, or in combination with interver-tebral disk disease.
Nonsteroidal anti-inflammatory drugs cause direct mucosal damage and interfere with prostaglandin synthesis. Flunixin mcglumine, aspirin, and ibuprofen have all been associated with erosions in healthy dogs. To circumvent toxicity caused by the inhibition of “friendly prostaglandins” (PGE2), drugs that preferentially block “inducible” cyclooxygenase (COX-2) have been developed. These COX-2 selective agents, such as carprofen, meloxicam, derccoxib, and potentially etodolac, are less ulcerogenic in normal dogs. However, even COX-2 selective drugs such as meloxicam are ulcerogenic in combination with dexamethasone, and their safety in sick animals remains to be determined.
High doses of glucocorticoids alone, such as dexamethasone and methylprednisolone, have also been associated with gastric erosions but the mechanisms by which they induce damage are not clear. Unlike NSAIDs, their effects are not ameliorated by PGE2 analogs.
Hypersecretion of gastric acid in response to histamine release from mast cell tumors, and gastrin from gastrinomas has also been clearly implicated as a cause of gastroduodenal ulceration and esophagitis in dogs and cats.
Renal failure, hepatic failure, hypoadrenocorticism, and hypotension are frequently proposed as risk factors for gastric erosion or ulceration, although few details have been published on the pathogenesis, frequency, or severity of gastric damage in these conditions. In a recent study of dogs with renal failure, ulceration was present in only 1 of 28 dogs. The predominant findings in these dogs were mucosal edema, vasculopathy, and mineralization that correlated to the degree of azotemia and calcium phosphorous product.
Sled dogs in the Iditarod are prone to develop gastric erosions and / or ulcers. This finding is similar to exercising humans and horses in whom the pathogensis is not understood hut is responsive to acid suppression.
Erosions and ulcers are aJso a sequela of gastric cancer and gastritis and are discussed elsewhere in this chapter.
Vomiting, hematemesis, and melena may be present in patients with gastric erosions or ulcers. Pale mucous membranes, abdominal pain, weakness, inappetance, hypcrsalivation (potentially associated with esophagitis as a consequence of gastric acid hypersecretion), and evidence of circulatory compromise are more variably present. Access to toxins and drugs, particularly NSAlDs, should be determined.
Clinicopathologic testing is directed at identifying diseases associated with gastric erosions and ulcers (see Table Association of Gastric Ulceration and Erosion with Specific Diseases) and the consequences of erosion / ulceration. The complete blood count may reveal anemia that is initially regenerative but can progress to become microcytic, hypochromic, and minimally regenerative. When accompanied by thromhocytosis and decreased iron saturation or low serum ferritin, these findings are characteristic of chronic bleeding and iron deficiency. Lack of a stress leukogram and eosiniphilia in dogs is supportive of hypoadrenocorticism. Eosinophilia could also be consistent with dietary allergy, eosinophilic gastroenteritis, mastocytosis, or a hyperseosinophilic syndrome. A neutrophilic leukocytosis and a left shift may indicate inflammation or possible gastric perforation. Examination of a buffy coat smear may help to detect mastocytosis.
Biochemistry and urinalysis may reveal findings consistent with dehydration (azotemia and hypersthcnuria), renal failure (e. g., azotemia and isosthenuria), hepatic disease (e. g., increased liver enzymes or bilirubin; decreased cholesterol, albumin, BUN], or hypoadrenocorticism (i. e., Na+:K+ ratio <27:1). It will also identify electrolyte and acid base abnormalities associated with vomiting and gastrointestinal ulceration. The presence of a metabolic alkalosis, hypochloremia, hypokalemia, and acid urine is consistent with upper gastrointestinal obstruction (physical or functional) or a hypersecretory state. Testing should be performed to detect abnormalities in primary and secondary hemostasis that may be associated with gastrointestinal bleeding. Serum gastrin and potentially histamine concentrations can be evaluated where acid hypersecretion is suspected as a cause of ulceration.
Plain radiographs are not usually helpful in diagnosing gastric erosions or ulcers but may help to rule out other causes of vomiting, such as foreign bodies, peritonitis, and gastric perforation. Contrast radiographs may reveal filling defects but do not allow detailed mucosal evaluation or sampling.
Ultrasonography can be performed to evaluate the gastric wall for thickening associated with ulcers and masses and also helps to rule out non-gastric causes of vomiting. The information provided by radiography and ultrasound is complementary to endoscopic evaluation, which is the diagnostic test of choice.
Endoscopy allows the direct evaluation of gastric damage and mucosal sampling. NSAID-associated ulcers tend to be found in the antxum and are not usually associated with marked mucosal thickening or irregular edges. This contrasts with ulcerated tumors that frequently have thickened edges and surrounding mucosa. Ulcers should be biopsied at the periphery to avoid perforation. Endoscopic biopsies are not ideal for diagnosing infiltrative gastric tumors and several biopsies from the same site are usually taken to enable sampling of deeper tissue. Endoscopic guided fine-needle aspirates, with use of a needle and tubing in the biopsy channel, can also be used to sample deep lesions. Even with this approach the diagnosis may be missed, and surgical biopsy required for a definitive diagnosis.
The combination of mucosal erosion or ulceration, antral mucosal hypertrophy, copious gastric juice, and esophagitis is highly suggestive of a gastric hypersecretory state. It is prudent to measure gastric pH and serum gastrin in patients with gastric erosion / ulceration that is not associated with drugs or gastric tumors. Dogs with mast cell tumors and hyperhistaminemia-induced acid hypersecretion have low serum gastrin concentrations. Finding a combination of gastric pH less than 3 and a high serum gastrin concentration prompts further investigation of gastrinoma by secretin stimulation test, ultrasonography (liver and pancreas), and pentertreotide scintigraphy.
Treatment of gastric erosions and ulcers is directed at the underlying cause, which ensures adequate hydration and perfusion, including blood transfusion if needed, and restoring electrolyte and acid base disturbances. Additional support is directed at shoring up the gastric mucosal barrier by enhancing mucosal protection and cytoprotection, and decreasing gastric acid secretion. Where vomiting is persistent, antiemetics may help to reduce fluid loss, discomfort, and the risk of esophagitis.
The rate of fluid administration depends on the presence or absence of shock, the degree of dehydration, and the presence of diseases (e. g., cardiac or renal), which predispose to volume overload. Patients with a history of vomiting who are mildly dehydrated are usually responsive to crystalloids (e. g., LRS or 0.9% NaCl) at a rate that will provide maintenance and replace both deficits and ongoing losses over a 24-hour period. Potassium depletion is often a consequence of prolonged vomiting or anorexia, and most polyionic replacement fluids contain only small amounts of potassium. Therefore KC1 is added to parenteral fluids on the basis of serum levels.
Patients with signs of shock require more aggressive support. The volume deficit can be replaced with crystalloids at an initial rate of 60 to 90 mLAg / h, then tailored to maintain tissue perfusion and hydration. Colloid solutions can also be used to treat animals in shock to reduce the amount of crystalloid required (e. g., Hetastarch, hemaccel at 10 to 20 mL / kg IV over 4 to 6 hours). Plasma, colloids, packed cells, or whole blood is occasionally required to treat severe hypoproteinemia or anemia, which can develop in vomiting animals with severe ulceration or HGE.
Central venous pressure monitoring and evaluation of urine output are necessary in patients with severe gastrointestinal disease, particularly those complicated by third space losses of fluid into the gut or peritoneum.
The effect of vomiting on acid-base balance is hard to predict and therapeutic intenention to correct acid-base imbalances should be based on blood gas determination. Where severe metabolic acidosis is present (pH <7.1, HCO3- <10 mmol / L), sodium bicarbonate (l mmolAg) can be given under careful supervision for the development of worsening hypokalemia, and hypocalcemia, and CSF acidosis. Further bicarbonate supplementation is based on repeated blood gas analysis. Metabolic alkalosis usually responds to replacing volume deficit, chloride, and potassium with IV 0.9% NaCl + KG. Diagnostic investigations should initially center on ruling out upper gastrointestinal obstruction. The administration of antisecretory drugs such as H2 antagonists may help to limit Cl–efflux into gastric juice.
Reducing Acid Secretion and Providing Mucosal Protection
Pharmacologic inhibition of acid secretion can be effected by blocking H2 (cimeridine, ranitidine, famotidine), gastrin (proglumide), and acetylcholine (atropine, pirenzipine) receptors, and by inhibiting adenyl cyclase (PGE analogs) and H+/K+ ATPase (e. g., omeprazole). H Long-acting somatostatin analogs such as octreotide directly decrease the secretion of gastrin and gastric acid.
Decreasing gastric acid secretion with an H2 receptor antagonist has been shown to promote mucosal healing in dogs with a variety ot experimentally induced ulcers and erosions. Famotidine is an attractive choice as it does not inhibit P450 enzymes and can be given once daily. The additional prokinetic activity of ranitidine or nizatidine (mediated by anticholinesterase activity) may make them good choices in the face of delayed gastric emptying associated with defective propulsion. In patients with severe or persistent gastric ulceration that is refractory to H2 antagonists, more complete inhibition of gastric acid secretion can be achieved with a H+/K+-ATPase inhibitor such as omeprazole (0.2 to 0.7 mg / kg SID PO—dogs). Omeprazole is the initial drug of choice in patients with acid hypersecretion secondary to r. ast cell tumors and gastrinoma (Zollinger-Ellison syndrome). Omeprazole has been shown to have few long-term side effects in dogs, but it should be used with caution in patients with liver disease and reviewed for interactions with drugs such as cisapride.
In sled dogs with exercise-associated gastric hemorrhage treatment with omeprazole significantly reduced mean gastric severity score compared to placebo but also was associated with increased frequency of diarrhea (omeprazole 54%, placebo 21%). The authors recommended further investigation of diarrhea associated with omeprazole treatment hefore omeprazole can be recommended for routine prophylactic treatment in these athletes.
The combination of omeprazole and the long acting somatostatin analog Octreotide effectively reduced vomiting in a dog with gastrinoma (Octreotide 2 to 20 μg / kg SC TID). Octreotide can also be employed to rapidly decrease gastric acid secretion in patients discovered to have large ulcers at endoscopy and may also be useful for controlling gastric bleeding (see human studies).
The PGE2 analog, misoprostol, protects against NSA1D-induced erosions in dogs at doses that do not inhibit acid secretion (3 to 5 μg / kg PO TID in dogs) and may be given to dogs receiving chronic NSAIDs for arthritis. The main side effect of misoprostol is diarrhea and it should not be given to pregnant animals.
The mucosal protectant polyaluminum sucrose sulfate (sucralfate) binds to areas denuded of mucosal epithelium regardless of the underlying cause and is useful lor treating gastric erosions and ulcers and esophagitis. Sucralfate can be given to patients receiving injectable antacids, but it may compromise absorption of other oral medications and is probably best separated from these by 2 hours or so.
In contrast to the efficacy of misoprostol and H2 antagonists in preventing NSAID-induced erosions, the prophylactic administration of various combinations of misoprostol, cimetidine, and omeprazole has not been shown to prevent gastric erosions in dogs with or without intervertebral disk disease receiving high-dose glucocorticoids. However, these drugs may speed healing of gastric lesions in these patients. Sucralfate is probably the drug of choice for treating gastrointestinal ulceration in patients receiving high doses of corticosteroids because it is not dependent on the premise that acid is causing or delaying healing.
Mast cell tumors are also worth considering separately as gastric ulceration is a frequent and severe complication. Mast cell tumors are thought to cause vomiting via the central effects of histamine on the CRTZ and the peripheral effects of histamine on gastric acid secretion (with resultant hyperacidity and ulceration). Treatment of mastocyosis with H1 and H2 histamine antagonists (e. g., diphenhydramine and famotidine) should reduce the central and peripheral effects of histamine. Corticosteroids are used to decrease tumor burden. Where acid hypersecretion is present, or is suspected, it is likely best managed with proton pump inhibitors (e. g., omeprazole 0.2 to 0.7 mg / kg SID). Somatostatin analogs may also be useful for controlling refractory gastric acid hypersecretion (Octreotide 2 to 20 μg / kg SC TID).
Antiemetics can be used where vomiting is severe or compromising fluid and electrolyte balance, or causing discomfort. The initial agent used in dogs is usually metoclopramide, which antagonizes D2-dopaminergic and 5HT3-serotonergic receptors and has cholinergic effects on smooth muscle (1 mg / kg / 24h CRI IV). Phenothiazine derivatives such as chlorpromazine and prochlorperazine are antagons of alpha1 and alpha2-adrenergic, H1 -and H2-histaminergic, and D2-dopaminergic receptors in the vomiting center and CRTZ and are used if metoclopramide is ineffective and the patient is normotensive. Nonselective cholinergic receptor antagonists (other than the M1 specific antagonist- pirenzipine) such as atropine, scopolamine, aminopentamide, and isopropamide are generally avoided as they may cause ileus, delayed gastric emptying, and dry mouth.
Antibiotics and Analgesia
Prophylactic antibiotic cover (e. g., cephalosporins, ampicillin) may be warranted in animals with shock and major gastrointestinal barrier dysfunction. Leukopenia, neutrophilia, fever, and bloody stools are additional indications for prophylactic antibiotics in animals with vomiting or diarrhea. Initial choices in these situations include ampicillin or a cephalosporin (effective against gram-positive and some gram-negative and anaerobic bacteria), which can be combined with an aminoglycoside (effective against gram-negative aerobes) when sepsis is present and hydration status is adequate, Enrofloxacin is a suitable alternative to an aminoglycoside in skeletally mature patients at risk of nephrotoxicity from an aminoglycoside.
Analgesia can be provided using opioids like buprenor-phine (0.0075 to 0.01 mg / kg IM).
Surgery may be required when the cause of ulceration is unclear or to resect large non-healing ulcers or those about to perforate.