Reasons for performing study
Gastric impaction in the horse is poorly described in the veterinary literature.
Gastric impaction in the horse is poorly described in the veterinary literature.
To review the clinical and pathological features of gastric impaction.
The clinical details of horses presenting with colic over a 7-year period and cases in which gastric impaction was considered to determine the outcome were reviewed. Clinical and clinicopathological data were recorded.
Twelve cases of gastric impaction were recorded (1.4% of 857 horses hospitalised for colic). Diagnosis was achieved by ultrasonographic examination, gastroscopy, exploratory celiotomy and/or post mortem examination. Five out of 12 horses were successfully treated, 5/12 were subjected to euthanasia (3 at celiotomy and 2 due to recurrence of impaction) and 2/12 died. Three out of 12 horses had spontaneous gastric rupture despite attempted treatment (one was subjected to euthanasia at celiotomy and 2 died). Post mortem examination (7 horses) revealed gross muscular thickening of the stomach wall in 6/7 horses. Histological examination revealed focal fibrosis of the stomach wall in 4/6 and focal myositis in 1/6 horses.
Gastric impaction is a rare cause of colic and affected horses can present with acute, chronic or recurrent colic in the presence or absence of other gastrointestinal disease. Spontaneous gastric rupture may occur. A proportion of affected horses have gross thickening of the muscular layers of the stomach wall.
Gastric impaction, characterised by distension due to the accumulation of dehydrated ingesta in the stomach, is an unusual condition in the horse . Some studies have classified gastric impactions as primary or secondary in aetiology  but there is discrepancy and therefore confusion in the literature regarding the definition of primary and secondary gastric impactions. Suggested causes of gastric impaction include liver disease  (Senecio jacobaea [ragwort] toxicity [4, 5]), ingestion of feed stuffs that swell or form a mass in the stomach (e.g. persimmon seeds), dental disease, inadequate water supply, rapid food intake, abnormal gastric motility or secretion and intestinal obstructions [2, 6-12]. Although some cases of gastric impaction appear to arise secondary to intestinal obstructions , some also occur with another gastrointestinal disease . Alternatively, sporadic cases of gastric impaction with no concurrent disease process and no obvious cause have been reported [13, 14].
A range of presenting signs of horses with gastric impaction, from anorexia to severe abdominal discomfort, has been reported [2, 15]. Some horses present quickly after formation of the impaction (e.g. impactions due to persimmon phytobezoars ), whereas others may present after a prolonged period with apparent secondary dilatation and muscular thickening of the stomach wall [13, 14]. The reported success rates of treatment are variable for both surgical and medical therapies.
The present study aims to review the clinical and clinicopathological features of gastric impaction in which no predisposing hepatic, dental or ingesta-related causes could be identified and in which the gastric impaction was considered to determine the case outcome, including some horses that had concurrent intestinal disease.
The clinical records of horses presenting to Bell Equine Veterinary Clinic with acute, chronic or recurrent colic over a 7-year period were reviewed. Clinical records, clinicopathological data and post mortem findings were reviewed for horses presenting with gastric impaction in which no predisposing hepatic, dental or ingesta-related causes could be identified and in which the gastric impaction was considered to determine the case outcome. Horses with gastric impaction considered to be secondary to intestinal obstruction were excluded but cases where gastric impactions were considered to have occurred independently of other intestinal diseases were included. Diagnosis of gastric impaction was made either at exploratory celiotomy on identification of an abnormally enlarged stomach extending to or caudal to the last rib and containing a firm impaction of food material or via gastroscopy where the stomach remained full of food material (to the level of the cardia) following a minimum of 24 h starvation. In 8 cases, abdominal ultrasonography was performed as previously described . Post mortem examinations (n = 7) were performed within 30 min of death. The mass of the stomach contents and maximum thickness of the muscle layers of the distal oesophagus and stomach wall were recorded at the following sites: distal oesophagus (5 cm proximal to the cardia), cardiac sphincter, saccus caecus (10 cm distal to the cardiac sphincter) and pyloric sphincter. The stomachs of 6 control horses (all mature Thoroughbreds, Warmbloods or crosses), subjected to euthanasia for reasons unrelated to the gastrointestinal tract, were examined post mortem for comparative purposes. The stomach/oesophageal wall of affected horses was considered to have gross evidence of muscular thickening when its thickness was more than double the maximum muscle thickness of control horses. Selected tissue samples were fixed in 10% neutral buffered formalin, processed routinely to paraffin wax, sectioned at 4 μm and stained with haematoxylin and eosin (H&E).
Twelve cases of gastric impaction fulfilling the inclusion criteria were identified, representing 1.4% of all colic cases presenting to the clinic during this time (n = 857).
The case details, relevant history, presenting signs, treatment and outcome of the cases are summarised in Table 1. A range of horse breeds (but no ponies) and ages (mean age 10.2 years; range 1–20 years) were represented with no apparent gender predisposition.
|Case No.||Breed||Sex||Age (years)||Pertinent long-term history||Signs initiating presentation||Method of diagnosis||Treatment||Outcome|
|1||Irish Draught x Thoroughbred||G||11||Recurrent colic||Colic 4 days||Gastroscopy||Medical||Discharged|
|2||Irish Draught x Thoroughbred||G||2||None|| |
Colic 3 days
Colic <24 h
|4||Welsh Cob||G||10||None|| |
Colic <24 h
Rocking horse stance
|5||Irish Draught||G||9||None|| |
Colic <24 h
|6||Thoroughbred cross||F||16||Recurrent colic 2 months exercise intolerance||Colic <24 h||Gastroscopy||Medical||Subjected to euthanasia 1 month post discharge|
|7||Irish Draught x Thoroughbred||F||9||None|| |
Colic <24 h
|Celiotomy||Medical then surgical||Subjected to euthanasia at surgery|
|8||Thoroughbred||F||16||None||Colic 2 weeks||Gastroscopy||Medical then surgical||Subjected to euthanasia at surgery|
|9||Thoroughbred||F||8||None||Colic <24 h||Celiotomy||Surgical||Discharged|
|10||Welsh Cob||G||20||None||Colic <24 h||Celiotomy||Surgical||Died following surgery|
|11||Friesian||F||8||None||Colic <24 h||Celiotomy||Surgical||Subjected to euthanasia 1 month post discharge|
|12||Warmblood||G||12||Recurrent colic||Colic <24 h||Celiotomy||Surgical||Subjected to euthanasia at surgery|
The mean heart rate on admission was 48 beats/min (range 36–60 beats/min), mean respiratory rate 18 breaths/min (range 12–40 breaths/min) and mean rectal temperature 37.9°C (range 36.3–40.0°C).
Routine haematological and serum biochemical profiles were performed in all horses at admission. Abdominal paracentesis was successful in 7/9 cases at admission. Relevant clinicopathological findings are summarised in Table 2.
|Case||Routine haematology and serum biochemistry||Plasma fibrinogen concentration||Abdominocentesis|
|Reference ranges|| |
WBC: 5.0–10.0 × 109/l
PMN: 1.2–6.8 × 109/l
TP: 53–76 g/l
Alb: 23–37 g/l
CK: 10–385 u/l
|<4.0 g/l||TNCC: <2.0 × 109/l|
|3||↑WBC (16.2 × 109/l) ↑PMN (13.9 × 109/l)||↑ 5.2 g/l||WNL|
|4||WNL||↑ 5.0 g/l||No fluid obtained|
|5||↓WBC (4.2 × 109/l)||↑ 4.9 g/l||Not attempted|
|7||↓WBC (2.0 × 109/l)||↑ 5.5 g/l||Yellow and turbid, TNCC = 73 × 109/l|
|8||↓Alb (22 g/l), ↓TP (46 g/l)||WNL||WNL|
TNCC = 0.7 × 109/l
|10||↑CK (1056 u/l)||WNL||Not attempted|
|11||↓PCV (21.6%), ↓Alb (15 g/l), ↑TP (80 g/l) ↑CK (733 u/l)||WNL||No fluid obtained|
The methods of diagnosis of gastric impaction are summarised in Table 1. Ultrasonographic findings provided supportive evidence of gastric distension based on published normal reference dimensions  in all 8 horses (Cases 1, 2, 3, 4, 6, 7, 8 and 10) in which it was performed. Results of rectal palpation were variable but caudal and medial displacement of the spleen was recorded in 6 horses (Cases 3, 5, 6, 8, 9 and 12). Distended small intestine was palpable per rectum and confirmed by ultrasonography in Case 10.
Following evaluation, 8/12 horses were treated medically (for ≥24 h); 2 of these subsequently underwent an exploratory celiotomy due to clinical deterioration or recurrence of an impaction; both were subjected to euthanasia. Four horses underwent exploratory celiotomy within 24 h of presentation due to their degree of discomfort. Seven of the 12 horses were discharged from the clinic, 5 following medical treatment and 2 following celiotomy.
Six horses received only medical treatment. One of these (Case 5) was treated medically but died shortly after admission. This horse was tachycardic (60 beats/min) and pyrexic (40°C) and had hyperfibrinogenaemia (5 g/l). The horse was treated with antibiotics (penicillin and gentamicin), flunixin meglumine and i.v. fluids but had increasing signs of endotoxic shock and died 12 h following admission. Post mortem examination revealed gastric rupture and peritonitis. The remaining 5 horses treated medically were diagnosed by gastroscopy and survived to discharge from the clinic (Cases 1, 2, 3, 4 and 6). On presentation, these 5 surviving horses had mild colic; 3 also had inappetence and 2 spontaneous nasogastric reflux. Two of these 5 horses had shown abnormal signs for less than 24 h, while the other 3 showed abnormalities for 3 to 4 days. The medical treatment in these 5 horses involved starvation and repeated gastric lavage. Gastric lavage was performed with water via nasogastric tube 3–4 times daily and, once the impactions began to clear, oral water, electrolytes ± carbonated drink (Diet Coca Cola) were administered by nasogastric tube twice daily . In one horse the impaction was infiltrated with water via the gastroscope. In these 5 horses, impaction resolution, confirmed by gastroscopy, occurred between 2 days and 2 weeks. In cases in which the impaction took longer than 2 days to resolve, sloppy feed and short periods of grazing were introduced as the impaction started to resolve. Two of these horses had mild colic during the first 24 h after admission and one horse had a mild episode of colic after lavage of the impaction. At hospital discharge, instructions were given to feed mashes and sloppy feeds, small quantities of soaked hay and, if possible, grass.
In addition to gastric impaction, one of the 5 horses diagnosed and treated medically (Case 3) had megaoesophagus and oesophageal impaction (diagnosed by oesophagoscopy). This horse had mild oesophageal dilation and delayed oesophageal emptying after resolution of the oesophageal and gastric impactions (demonstrated by barium contrast oesophageal radiography), but had no clinical complications following discharge. Gasterophilus intestinalis larvae were identified on the gastric squamous mucosa in one horse (Case 3) and in another (Case 1) the pylorus was swollen with erythematous mucosa.
Three of these 5 horses (Cases 1, 2 and 6) returned to the clinic for re-examination (including routine gastroscopy after 12 h of starvation) 4 weeks following discharge. Signs of colic were not reported by the owners of any of these horses during these 4 weeks. In Case 1, food was still present in the stomach, and the horse was starved for a further 24 h. Repeat gastroscopy (after 36 h of starvation) showed that food material was still present in the pyloric region, and there was generalised erythema around the margo plicatus. The horse was discharged on oral sucralfate therapy (20 mg/kg bwt per os q. 6 h) with instructions to feed a low residue diet. In Case 2, gastroscopy showed hyperkeratosis of the squamous mucosa adjacent to the margo plicatus. In Case 6, food was still present in the stomach after 48 h of starvation confirming recurrence of the gastric impaction; the horse was subjected to euthanasia at the owner's request.
Six horses underwent an exploratory celiotomy for diagnosis and/or treatment of concurrent diseases. Of these, 4 horses (Cases 9, 10, 11 and 12) underwent an exploratory celiotomy within 24 h of admission; none had a history of previous colic and all had clinical signs for <24 h. All of these horses were found to have other gastrointestinal conditions in addition to gastric impaction (impacted transverse colon [Case 9], epiploic foramen entrapment of the ileum [Case 10] and right dorsal displacement of the large colon [Cases 11 and 12]). One horse (Case 12) was subjected to euthanasia at surgery at the owner's request. Following standard treatment of the other intestinal disorders in the remaining 3 horses, the gastric impactions were left in 2 (Cases 10 and 11) and partially broken down by infusion of sterile saline into the impaction through the stomach wall and manual manipulation in the third horse (Case 9). All 3 horses that recovered from anaesthesia were treated with gastric lavage and dietary management, as specified for medically managed cases, post operatively. One horse (Case 10) died 48 h following surgery due to gastric rupture while the remaining 2 horses (Cases 9 and 11) recovered and were discharged. Case 11 was re-examined 2 weeks following discharge; the owner reported no signs of colic but abnormal gastric emptying was suspected following re-examination based on the gastroscopic identification of food within the stomach after 24 h of starvation. Four weeks following discharge the horse presented with colic and was found to have a gastric impaction. The horse was subjected to euthanasia at the owner's request.
Two horses were treated medically (>24 h) before undergoing an exploratory celiotomy and being subjected to euthanasia. One horse (Case 7) presented with peritonitis and leucopenia. It deteriorated despite intensive medical treatment. The horse was therefore taken to surgery 36 h following admission where a gastric rupture was identified necessitating euthanasia under general anaesthesia. The second horse (Case 8) showed signs of lethargy for 2 weeks prior to admission. Gastroscopy identified a gastric impaction, which was successfully cleared after 5 days of medical treatment. Despite slow re-feeding the horse re-impacted prompting further diagnostic evaluation. Celiotomy revealed a thickened, firm stomach wall and the horse was subjected to euthanasia at the owner's request in light of the guarded prognosis.
All 7 of the horses that either died (n = 2) or were subjected to euthanasia (n = 5) underwent a routine post mortem examination.
In 3 affected horses, the mass of ingesta in the stomach was recorded and ranged from 15 to 35 kg (mean 25 kg). In the 6 control horses, the mass of ingesta in the stomach ranged from 0.26 to 0.72 kg (mean 0.36 kg). In addition to gastric impaction, 6 horses had evidence of gross thickening of the muscle layers of the wall of the saccus caecus (thickness of the muscle layer ranging from 10 to 35 mm [range of normal horses 1–5 mm]) (Table 3) (Fig 1); the stomach wall felt abnormally thick and ‘leathery’ in these cases. Two horses also had gross thickening of the distal oesophageal muscle and one had thickening of the pylorus. In the control horses, the muscle layer tapered from the cardiac sphincter (median thickness 11.5 mm) to the saccus caecus (median thickness 3.5 mm). In horses with muscular thickening, the muscle of the cardiac sphincter was moderately thickened, but the muscle layer of the saccus caecus was severely thickened (up to 35 mm thick) and then tapered down to normal thickness in the pyloric region of the stomach (Table 3). Tissue samples from 6/7 horses were examined histopathologicallly. Four had evidence of focal fibrosis (Fig 2) within the stomach wall and in one of these there was also focal myositis (Fig 3).
|Control horses (median and [range])||Case 5||Case 6||Case 7||Case 8||Case 10||Case 11||Case 12|
|Distal oesophagus||6 [4–7]||10||NR||15||NR||NR||7||NR|
|Cardiac sphincter||11.5 [9–15]||35||18||25||18||NR||12||15|
|Saccus caecus||3.5 [1–5]||35||25||25||20||4||15||10|
|Pyloric sphincter||11 [9–14]||20||14||12||NR||NR||10||13|
Gastric impaction was present in a small percentage of horses with colic examined at the hospital over the study period. Horses with this condition presented with a wide range of clinical signs and concurrent gastrointestinal abnormalities. While some horses showed relatively mild signs of colic and were successfully treated medically, others presented with severe colic and systemic compromise. In these latter cases, other intestinal abnormalities were often present.
The aetiology of gastric impactions is poorly understood. Although some studies classify gastric impactions as primary or secondary in aetiology  others distinguish between acute and chronic disease . By definition, primary disease is a disease arising spontaneously and not associated with or caused by a previous disease or injury . Definitively ruling out previous or associated disease or injury in clinical cases is not always practical. It has been stated that diagnosis of a primary gastric impaction requires that all other diseases that delay gastrointestinal transit be excluded through exploratory celiotomy or necropsy . In clinical practice, cases have been classified as primary impactions without such procedures . Equally, abnormal gastrointestinal motility patterns may not be identified during exploratory celiotomy or necropsy . It follows that categorisation of these cases into primary or secondary gastric impaction is controversial and was considered to be beyond the scope of this small case series.
Categorisation of horses into those with acute or chronic gastric impaction is also controversial. Two horses (Cases 5 and 7) presented with spontaneous gastric rupture, both of which had shown abnormal clinical signs for less than 48 h. This suggests rapid progression of the impactions from initial stages through to rupture and, therefore, is consistent with an acute disease. However, both horses presenting with gastric rupture had increased plasma fibrinogen concentrations on presentation suggesting the existence of an inflammatory process for several days. Both horses also had post mortem evidence of gross muscular thickening, likely due to hypertrophy or hyperplasia, and histological evidence of fibrosis of the stomach wall consistent with a more chronic pathogenesis. It is, therefore, more likely that these impactions remained sub-clinical until they became so large as to compromise the integrity of the stomach wall.
Of the 6 horses treated medically, 5 were discharged with initial successful resolution of the impaction, 3 of which were re-examined 4 weeks later; one had re-impacted, one had possible delayed gastric emptying and one had relatively normal gastroscopy findings. The success rate for the horses that underwent exploratory celiotomy was not as high as medical treatment. However, comparing treatment outcomes in retrospective series is difficult particularly when the presenting signs and disease severity clearly vary between individual animals. Of the 3 horses that recovered from surgery, one horse subsequently suffered gastric rupture and one re-impacted following discharge. In the horse that developed a gastric rupture (Case 10), the impaction was not broken down at surgery as it was the surgeon's expectation that the impaction would resolve once the epiploic foramen entrapment was corrected. It is uncertain whether evacuation, infusion of fluid into the impaction and/or manual breakdown of the impaction at surgery would have prevented the subsequent rupture.
Of the 7 horses that underwent post mortem examination, 6 had thickening of the muscle layers of the stomach wall. Of the 6 horses that underwent histopathological examination, 4 had focal fibrosis of the muscle layers and one had focal myositis. Thickening of the gastric muscle layers has been described previously in horses with gastric impaction [13, 14]; the cause was not identified. In one study, muscular thickening in combination with dilatation and mucosal ulceration was considered to represent evidence of long-standing irreversible disease .
Smooth muscle thickening has been identified in other regions of the alimentary tract in horses, including the oesophagus [20, 21], small intestine (most commonly the ileum)  and caecum . Two horses in the current study (Cases 5 and 7) had concurrent muscular thickening of the distal oesophagus, but neither of these horses had any clinical signs indicative of oesophageal dysfunction. The muscular coat (tunica muscularis) of the stomach consists of 3 incomplete layers; external longitudinal, middle circular and internal oblique fibres (the oblique fibres being arranged in 2 layers) [24, 25]. In the horses in this study that had gross muscular thickening of the stomach wall, the distribution of the thickened muscle suggested that the oblique muscle fibres of the cardia and saccus caecus were affected predominantly; however, this was difficult to confirm by gross or histolopathological examination.
There was a range of histolopathological lesions in the muscle layers of the stomach, including vacuolation of smooth muscle cells, fibrosis and focal myositis. The gross thickening of the gastric muscle layers in 6 horses was a striking post mortem observation but the precise cause of this thickening (i.e. hyperplasia or hypertrophy of smooth muscle cells) was not specifically determined. In future studies, histopathological evaluation of control tissue from horses subjected to euthanasia without gastrointestinal disease as well as those subjected to euthanasia due to gastrointestinal disease but without gastric impaction may help validate the abnormal histopathological findings in horses with gastric impactions. Whether the gross and histological abnormalities are the result of gastric impaction or related to the cause is unclear. Muscular hypertrophy and reduced neuron density have been reported in horses with chronic recurrent caecal impactions and it has been hypothesised that these impactions may be secondary to uncoordinated contractions of the caecum ; a similar pathogenesis could be responsible for gastric impaction, supporting the need for further histopathological studies including evaluation of neuron density in these cases. Rupture of the oesophagus associated with muscular hypertrophy [20, 21] and rupture of the ileum associated with ileal muscular hypertrophy  have been previously reported and are probably the result of decreased elasticity, secondary to the muscular thickening. A similar mechanism may have been responsible for the spontaneous gastric rupture that occurred in 2 horses with gastric muscular thickening in this series.
In conclusion, gastric impaction is an uncommon cause of colic in horses. Presentation with other gastrointestinal disease necessitating surgical intervention is not uncommon. Gross muscular thickening, focal myositis and focal fibrosis of the stomach wall have been identified in cases of gastric impaction but the cause and effect relationship remains undetermined.
No competing interests have been declared.
The authors thank colleagues at Bell Equine Veterinary Clinic who contributed to the care of these cases; and D. Fews, M.J. Day and technical staff in the Comparative Pathology Department, University of Bristol.
All authors were equally involved with the study design, data collection, data analysis and interpretation, and preparation of the manuscript.