Click on a section title for more information:

Gastric dilatation-volvulus (GDV) is a range of conditions of varying clinical severity.

  • Dilation can occur without volvulus in many breeds. It occurs frequently in young puppies after eating with no apparent ill effects.

  • Dilation without volvulus in adult dogs can cause shock as severe as GDV.

  • Volvulus occasionally occurs as a chronic condition in dogs and results in mild gastrointestinal signs without hemodynamic compromise.  When volvulus occurs, the pylorus moves from its normal position in the right cranial abdominal quadrant. It passes between the stomach and the body wall to the left side of the abdomen (180° volvulus) and may keep rotating. At the same time the fundus of the stomach rotates to the right. This type of volvulus occurs in more than 95% of GDV dogs.  {Click here for diagram.}

Many potential causes of GDV have been proposed:

  • Anatomical studies have shown that Irish setters with an increased thoracic depth/width ratio are at increased risk of GDV.  Others suggest that abnormal thoracoabdominal dimensions might influence the physiology of the stomach, esophagus, gastroesophageal junction and the diaphragm, perhaps inhibiting the animals’ ability to vomit or eructate.

  • Overeating, post prandial exercise, and food type have all been incriminated in causing GDV, but no hard clinical or experimental evidence supports these assumptions.  In a study of 600 dogs at a military center, changes in feeding patterns, post-prandial exercise and diet had no effect on the incidence of GDV.  Contradictory studies have been published on the effect of diet (soy bean meal) on the incidence of GDV.

  • Several studies have suggested that delayed gastric emptying (in some cases associated with increased plasma gastrin levels) may cause GDV.  These studies in which gastric emptying was delayed on barium swallows were performed after GDV had occurred, making it difficult to establish cause versus effect.

  • The gas in the stomach of GDV dogs was originally thought to be from bacterial fermentation because of its high CO2 levels.  However, subsequent measurements have shown no increase in hydrogen and methane.  The gas is now thought to be swallowed air.  The higher CO2 content is due to reaction of gastric acid and bicarbonate.  Anecdotally, owners often report that affected dogs are greedy eaters, possibly swallowing a large amount of air with their food.

An understanding of the causes of shock seen in animals with GDV is vital to planning appropriate treatment.  Severe gastric distension results in compression of the abdominal caudal vena cava.  Angiographic studies of the caudal vena cava show shunting of blood from the obstructed vena cava into intervertebral and azygos veins, and reflux filling of the iliac, deep circumflex iliac, and renal veins.  The collateral circulation is unable to handle the normal venous return.  Gastric volvulus and distension result in obstruction of the portal vein supplying the liver.  Lack of venous return to the heart results in a decrease in cardiac outputBlood pressure falls.  The carotid baroreceptors respond to the falling blood pressure and trigger reflexes that: i) decrease vagal tone; ii) cause release of norepinephrine from the sympathetic nerve terminals; and iii) cause release of epinephrine from the adrenal medulla.  These changes result in peripheral vasoconstriction and an increase in heart rate, as the body tries to maintain perfusion to the brain and heart.

Experimental studies which dilated the stomach to 30 mm Hg (gastric pressure in spontaneous GDV = 9-62 mm Hg; mean = 22.9) showed a decrease in cardiac output of 64% and a decrease in mean arterial pressure of 48%.  Other experiments have shown an increase in portal venous pressure, decreased coronary blood flow, and increased total systemic peripheral resistance.  In total, the substantial decrease in cardiac output results in poor perfusion and oxygen delivery to many tissues.

Cardiac arrhythmias are common in GDV.  {Click here for an ECG example.}  They probably have many causes, the most important being myocardial ischemia.  Arrhythmias similar to those found in dogs with GDV were created in coronary artery ligation studies.  These arrhythmias developed 6-12 hours after coronary artery ligation.  This is similar to arrhythmias seen in clinical GDV.  Coronary blood flow in experimental GDV is decreased 50%.  Histologic lesions compatible with ischemia are seen in both experimental and spontaneous GDV.  Circulating cardiostimulatory substances such as epinephrine, and cardioinhibitory substances such as myocardial depressant factor have also been implicated in the generation of arrhythmias.

Organs other than the heart are significantly affected by the decrease in cardiac output.  The stomach is the organ most severely affected.  In experimentally induced volvulus without dilation, gastric perfusion was only 22% of that of control dogs.  Decreased gastric perfusion results in serosal hemorrhage and edema of stomach wall, which begins in fundus, and spreads to body.  Severe compromise to the gastric wall results in necrosis and perforation, with resultant peritonitis.

Infarction of splenic arteries and thrombosis of splenic veins result in splenic necrosis.  In the author's experience, splenic torsion is rare in dogs with GDV.  Although a large spleen can often be palpated on physical examination, it is most often simply congested because of portal venous obstruction.

Disseminated Intravascular Coagulation (DIC) is frequently seen in dogs with GDV.  Possible contributing factors include pooling of blood in caudal vena cava, portal vein, splanchnic circulation, tissue hypoxia, acidosis, endotoxemia, and sepsis.

Dogs are often large, deep chested breeds.  They frequently have a history including recent ingestion of large meal and large amount of water.  Dogs are usually restless, uncomfortable and anxious.  The majority of affected dogs salivate, and may retch or attempt to vomit.  As shock worsens, dogs become progressively weaker and collapse.

Dogs often present in shock, with pale mucous membranes, prolonged capillary refill time, and rapid, weak, thready pulses.  The abdomen can vary from unremarkable on palpation, through distended and firm, to tympanic.  An enlarged spleen is often palpable; this is due to splenic congestion and rarely indicates torsion.  If presentation has been delayed, dogs can be collapsed and comatose.  This usually represents severe shock with possible gastric rupture.

Acid/base and electrolyte imbalances are not consistently seen in every dog with GDV.  In 1982, Wingfield, et al found no statistically significant difference between mean blood gas, pH and hemoglobin saturation values in normal dogs and dogs with GDV.  However, plasma lactate levels have been shown to be prognostic in dogs with GDV {dePapp, Drobatz, Hughes}.  Ninety-nine percent of dogs with blood lactate levels less than 6 mmol/L survived GDV, compared with 58% survival in dogs with blood lactate concentrations greater than 6 mmol/L.  A high plasma lactate level was also predictive of gastric necrosis at surgery.  Lactate is produced during anaerobic metabolism by tissues, and has been shown to be a prognostic indicator for critically ill humans. 

Radiographs are made to determine if the stomach is just dilated or in an abnormal position.  A right lateral radiograph is the most helpful view.  In right lateral recumbency, the pylorus will often lie dorsally if a GDV is present.  The pylorus will thus appear as a separate, dorsal gas filled structure on this view.

One must differentiate between GDV and gastric dilation without torsion (done by radiography).  Other diseases causing acute abdominal distension include intestinal volvulus, splenic torsion, abdominal effusion or hemorrhage.

The aim of postoperative management is to maintain perfusion to the tissues.  This is achieved with intravenous, balanced electrolyte solutions.  GDV dogs lose substantial amounts of fluid into the peritoneal cavity and gastrointestinal tract, so reasonably high fluid rates are often required for the first few days.  The dog's mucous membrane color, capillary refill time, PCV/TS, urine output, and blood pressure should be monitored to assess the adequacy of perfusion.  The dog's abdomen should be monitored for gastric distension and the stomach tube vented every 2 hours if necessary.  Dogs recovering well from surgery can be offered water and a small amount of food on the second day after surgery.  These dogs can be gradually weaned off their intravenous fluids over 2 days.

Cardiac arrhythmias often begin 12-24 hours after surgery.  Continuous EKG monitoring is ideal.  Contributing factors include poor myocardial perfusion, electrolyte disturbances, acidosis, myocardial depressant factor, and DIC.  In many cases these arrhythmias are not life threatening.  As we have discussed, they are thought to be caused by poor myocardial perfusion, so the clinician's first thought when arrhythmias occur should be: Does this animal have adequate perfusion?  Does this animal need a bolus of intravenous fluids?  Antiarrhythmic drugs should only be considered if the animal is adequately volume replaced and has an arrhythmia which is life threatening or causing poor perfusion.  For example, R waves occurring directly on top of T waves (R on T phenomenon) is known to precede ventricular fibrillation.  Ventricular tachycardia occurring with a heart rate of greater than 200-220 probably impairs ventricular filling and, therefore, cardiac output.

Disseminated intravascular coagulation can persist after surgery.  Contributing factors include pooling of blood in portal circulation and caudal vena cava, sepsis, vascular thrombosis, endotoxemia, acidosis, tissue hypoxia, and splenic congestion.  The diagnosis can be confirmed by a prolonged activated clotting time, or abnormalities in platelet, FSP and PTT values in a coagulation screen.  Treatment should be aimed at the underlying cause.  The clinician should attempt to prevent microvascular thrombosis by maintaining adequate perfusion to the tissues with intravenous fluid therapy.

Sepsis can occur in the postoperative period.  Animals become lethargic, febrile and shocky.  Although the first instinct is to suspect gastric leakage, in many cases a silent aspiration pneumonia is the culprit.  Thoracic radiographs and a peritoneal lavage should be performed to identify the source of sepsis.  Broad spectrum bacteriocidal antibiotic coverage is instituted.  Cardiovascular support is required.

The most important goal of treatment is correction of circulatory collapse.  Aggressive cardiovascular stabilization before surgery is far more important than the surgical procedure itself.  Prior to 1983 at VHUP, the success rate with GDV was approximately 50%.  Since then, a protocol of aggressive preoperative correction of cardiovascular collapse has been instituted and the success rate has risen to 85%.  After correcting the circulatory collapse, treatment goals include decompression of the stomach, differentiation of dilation vs GDV, re-positioning and pexying the stomach if volvulus exists, and early diagnosis and treatment of complications.

On presentation, two large bore catheters are placed in the cephalic veins.  If the cephalic veins are not available, the jugular vein is used.  Fluid resuscitation through the saphenous veins is unlikely to be successful because of the caudal vena caval obstruction.  Blood for a minimum or extended data base is drawn and resuscitation is begun.  Either isotonic crystalloids (90 ml/kg in the first hour) or hypertonic saline (7% NaCl in 6% Dextran: 5 ml/kg given over 5 minutes) followed by crystalloid are administered.  {Click here for a picture of crystalloid fluids.}  The administration of corticosteroids remains controversial.  Corticosteroids have many theoretical benefits, but have not been unequivocally demonstrated to improve survival in GDV.  Prophylactic antibiotics are also somewhat controversial, but rational arguments are made for their use.  GDV dogs do have increased levels of circulating endotoxin, perhaps indicating increased GI mucosal permeability.  Poor perfusion to the liver could inhibit reticuloendothelial function.

Improving tissue perfusion by fluid resuscitation and subsequent gastric decompression and de-rotation can potentially result in the production of damaging, highly reactive oxygen free radicals.  These radicals can cause significant tissue injury (“Re-perfusion Injury”) that may be as damaging as the initial hypoperfusion episode.  It is possible that treatment to prevent free radical generation may be beneficial in dogs with GDV.  Of the drugs trialed in experimental models, desferoxamine, an iron chelator, shows the most promise for clinical application. 

During resuscitation, perfusion should be monitored using clinical and laboratory parameters.  A continuous electrocardiogram is connected.  Once the animal has been stabilized, gastric decompression is attempted using a silicone or rubber tube.  The tube is pre-measured to the level of the stomach and marked.  A 2" roll of tape is placed in the dog’s mouth and the tube passed through the tape and slowly into the esophagus and stomach.  If resistance is encountered at the level of the cranial esophageal sphincter, the tube must not be forced, as this could cause rupture of the caudal esophagus.  In some fractious animals, sedation and intubation is necessary for gastric decompression.  Balanced anesthesia is used (Valium: 0.125-0.25 mg/kg IV, Oxymorphone: 0.025-0.05mg/kg IV).  If orogastric intubation is unsuccessful, the stomach is decompressed by trocharization. The abdomen is carefully palpated, and the enlarged spleen is avoided.  A large gauge catheter (10-12F) is placed into the stomach percutaneously.

Surgery is indicated for dogs with GDV or dogs with radiographic dilation only that cannot be adequately decompressed with a stomach tube or by trocharization.  The animal is anesthetized once it is stable, and an exploratory laparotomy is performed.  Anesthesia should be performed using an oxymorphone/valium combination, titrating the minimal amount of drug required for intubation.  A large ventral midline incision is made.  The stomach is initially decompressed and then rotated back into its normal position.  To find the pylorus, trace the duodenum (identifiable by the attached pancreas) cranially from the duodenocolic ligament.  Gently de-rotate the stomach, assess the stomach for viability, then examine the spleen.  Many procedures have been described to "pexy" the pyloric antral region of the stomach to the right body wall.  These include the tube gastropexy, the incisional gastropexy, the muscular flap gastropexy, the circumcostal gastropexy, the belt loop gastropexy, and various modifications of the above.  The aim is to create a permanent adhesion between the antral region of the stomach and the right body wall.  All of these techniques will work; at VHUP we perform tube gastropexies because it allows gastric decompression or feeding postoperatively.

Ancillary procedures include splenectomy and gastric resection.  The spleen should only be removed when it has been damaged by the gastric volvulus; it is rarely twisted itself.  Partial gastrectomy is required when gastric necrosis has occurred, usually along the greater curvature.  Although gastric necrosis is associated with poorer survival rates, in a recent study of cases at VHUP, 70% of dogs with gastric resections survived to discharge (Brockman et al, 1993).

Gastric outflow enhancing procedures have been performed in an attempt to decrease re-occurrence of gastric dilation or GDV.  Pyloromyotomy and pyloroplasty have been described.  However, as we have already discussed, gastric outflow obstruction is no longer thought to play a major role in the pathogenesis of GDV.  These procedures have not been shown to decrease re-occurrence, and have in fact been associated with a higher immediate postoperative complication rates.

Key indicators for a poor prognosis are detection of gas within the gastric wall (indicating gastric wall compromise), detection of free gas in the abdominal cavity (indicating gastric rupture), and plasma lactate levels above 6 mmol/L at presentation (predicting 58% survival of GDV in dogs).

Copyright © 2002 University of Pennsylvania School of Veterinary Medicine
VHUP Department of Clinical Studies
Student Web Designers: Srinivas Lingareddy, V'03 (2001) and Karen O'Connor, V'04 (2002)