Causes of Gastrointestinal Colic at an Equine Referral Hospital in South Africa (1998–2007)

INTRODUCTION Equine veterinarians frequently treat horses with colic caused by visceral abdominal pain or non-gastrointestinal causes. Visceral abdominal pain is more commonly treated and may be caused by intestinal distension due to fluid, gas or ingesta, tension on the mesentery, ischaemia or infarction, ulceration of the mucosa, or peritoneal pain 36. Studies have examined the incidence of colic events 1,38,40 in general and in specific breeds 19,33,34. Some focused on specific causes such as caecal impaction 30 , perito-nitis

tems, Kaysville, Utah 2001; www.ncss.com).Descriptive statistics were calculated for each measured parameter and the data tested for normality.Parameter data that were not normally distributed were analysed as nonparametric data.Continuous data that were normally distributed are reported as mean ± standard deviation (SD) and continuous data that were not normally distributed are reported as median and range.

Study population
Of the 1201 medical records, 159 (13 %) were unavailable and 107 (9 %) were from horses admitted more than once.Therefore, 935 horses (78 %) were included in this study of which 72 % were admitted during normal hours and 28 % after hours.Medical and surgical intervention was performed on 411/238 horses respectively during normal hours and 147/93 horses respectively after hours.The pretreatment euthanasia/death at admission category included 26 horses admitted during normal hours and 20 horses after hours.The total admissions for January to December during the 10 years were calculated and a steady admission rate between 6.52 % and 11.55 % was obtained.The proportion of horses admitted during the 6 hottest months of the year, October to March, represented 51 % (n = 476) and those admitted during the remainder of the year represented 49 % (n = 459) of the study population.

Pre-admission data
The median duration of pre-admission colic symptoms recorded in 53 % of horses was 11 hours (range, 0.5-168 hours).Of these horses, 61 % were admitted within 12 hours after the onset of colic symptoms.The median duration of colic signs before admission for the surgical, medical and pre-treatment euthanasia/ death at admission category was 10, 12 and 14 hours, respectively.Pre-admission treatment was administered to 73 % of the horses and included anti-inflammatory and other analgesic drugs (78 %), oral and intravenous fluids (36 %), sedation (30 %) and antibiotics (9 %).
Mucous membrane colour was recorded in 885 horses (95 %), with cyanosis occurring more frequently in the pre-treatment euthanasia/death at admission category (23 %), while congestion was recorded more frequently in horses requiring surgery (24 %).Abdominal distension was recorded in 787 horses (84 %) and was present in 42 %.Borborygmi were recorded in 849 horses (91 %) of which 63 % had decreased borborygmi and 7 % had no borborygmi.
Pain character and severity were recorded in 798 horses (85 %).Twenty-two per cent of horses were alert and showed no pain, while depression, mild pain, marked pain and severe pain were recorded in 12, 44, 13 and 9 %, respectively.The pre-treatment euthanasia/death at admission category had the highest recorded percentage of marked and severe pain (48 %).

Treatment
Pre-treatment euthanasia/death at admission category.
Of the 935 horses reviewed in the study, 46 horses (5 %) received no treatment as 40 horses were euthanased due to a poor prognosis while 6 horses died following clinical examination.A diagnosis was made at necropsy on 31 horses but was not included in the study results as no treatment was performed (Fig. 2).
Impactions were reported in 75 horses (22 %) and affected the large colon (13 %), small colon (4 %), small intestine (4 %), caecum (1 %) and stomach (<1 %).Left dorsal displacement of the large intestine was recorded in 36 horses (11 %).Strangulating lipoma occurred in 2 horses but lesions were not corrected by intestinal resection.Strangulating intestinal lesions that were corrected by intestinal resection were recorded in 11 horses.Strangulation of the small intestines occurred in 8 horses and intramural obstruction in 3 horses.Of the 11 horses, 4 survived to discharge, 2 died and 5 were euthanased.

Gastrointestinal causes of colic
The ten most-common gastrointestinal causes of colic in horses admitted to the OVAH are depicted in Fig. 6
The median duration of hospitalisation for all horses was 4 days (1 to 42 days) while for horses treated medically or surgically it was 4 and 7 days, respectively.On average the median cost per day for the 935 cases included in this report was ZAR 716.41 (R0-13 026.86).The median cost per day for horses treated medically was ZAR 421.14 (R0-5597.82)and for those treated surgically was ZAR 1627.32 (R615-19 307.65).

DISCUSSION
The aim of assessing a horse with colic is to determine the severity of the colic and to select the appropriate type of medical or surgical intervention.The knowledge of common causes and plausible outcomes of colic events may aid veterinarians when considering appropriate diagnostic methods, therapeutic options and in determining the prognosis for the individual horse.Although studies conducted in other countries dealt with the causes of gastrointestinal colic, this information may not be relevant to horses with colic in South Africa due to geographical differences.To the authors' knowledge this is the 1st study in South Africa to establish causes and outcomes of horses with gastrointestinal colic at a referral hospital.
In the present study the incidence of all colic cases showed no seasonal trends which is in contrast with other stud-  ies 3,12,18 .In a study involving Swedish horses admitted for gastrointestinal colic, a larger proportion of all colic cases were seen in the winter months 12 .Studies performed at a United Kingdom-based referral hospital 3 and at a Thoroughbred training premises in the British Isles 18 indicated an apparent peak in horses with colic during spring and autumn.The sex distribution of horses in the present study was not evenly divided between females (43 %) and males (57 %) and agrees with similar studies 1,27 , while the median age distribution was in contrast to that reported in other studies 1,12 .
At the OVAH equine hospital, equine veterinarians and veterinary students are encouraged to employ the history, physical examination results and clinicopathological data when assessing a horse with colic.The usefulness of various clinical signs and clinical pathology to determine of the appropriate management of horses with colic has been previously reported 13,21,26,31,37,41 .Reports have indicated that heart rate 21,31,41 , mucous membrane colour 21 , capillary refill time 37 , packed cell volume 21,31,37 and level of pain 31,37,41 are important predictors of outcome in horses that are hospitalised for colic.However, another study reported that white cell count, packed cell volume, blood pH and colour of the mucous membranes did not show any prognostic significance 41 .Furthermore, abdominal fluid analysis cannot be used alone to accurately predict lesion type or outcome for horses with colic 13 , but it may contribute to the decision to proceed to surgery 26 .
In the present study clinical examination data, packed cell volume and total serum protein concentration were recorded in > 75 % of admissions and frequently used to assess horses.Diagnostic procedures performed such as rectal examination (93 %), nasogastric intubation (84 %) and abdominocentesis (53 %) were commonly used to compliment a physical examination.
While the causes of colic in the present study were numerous and in some cases life-threatening, 55 % of all admissions responded to medical intervention and were subsequently discharged.In this study, 32 % of horses referred for medical intervention had large colon impactions while 7 % were suspected to have tympany, which was consistent with results reported in other studies 1,29 .A diagnosis could not be made in 25 % of horses that had medical intervention.It is not uncommon not to determine the cause of colic, as seen in a Canadian study where a diagnosis could not be made in 13 % of horses receiving medical intervention 1 and 10 % of horses in a  study in Georgia, USA 27 .
From the results it is apparent that most horses that had surgical intervention suffered a displacement of the large colon (29 %), while intestinal impactions (22 %), small intestinal strangulation (18 %) and large intestinal torsion (11 %) occurred less often.Similar findings were reported in a Canadian study where surgical intervention was performed on horses with displacements (24.5 %) and torsion (14.3 %) of the large colon 1 .In the present study strangulating lipoma occurred in <1 % of all admissions, which is in contrast to studies reporting an incidence of 12.5 % 1 and 10 % 15 .Moreover, several studies have reported that a strangulating lipoma be an important consideration when examining an aged horse with suspected small intestinal disease 1,5,11,15 .The finding in this study that 2 % of all the colic admissions had caecal impactions is in contrast to a reported incidence of 4.1 % in another study 30 .Horses with enterolithiasis accounted for <1 % in the present study, which is similar to a study that reported an incidence of 1.7 % 8 , but is in contrast to another study that reported an incidence of 15.1 % 16 .These studies further concluded that breed and diet influence the risk of enterolithiasis, which explains the discrepancies in study results 8,16 .
The common causes of gastrointestinal colic in horses in this study were consistent with results found in several studies in other countries 1,12,27,28 .A Canadian study reported that the 9 most common causes of colic were large colon impactions, large colon displacements, spasmodic colic, large colon volvulus, lipoma, strangulating small intestinal lesions, enteritis, peritonitis and verminous arteritis 1 .A study of 3100 Swedish horses with colic reported that impaction occurs most frequently, followed by torsion/volvulus, enteritis, sand impaction and incarceration of the small intestines 12 .In 2 studies at the University of Georgia, USA, involving 229 and 449 horses, the most common colic causes were large colon displacement and volvulus, large colon impaction and small intestinal obstruction 27,28 .
Compared with this study, a similar pattern of short-term survival was reported in other studies 1,10,21,24,41 .In a report of the outcome of horses with colic admitted to the Liverpool University Equine Hospital, UK, over 12 months, 76 % to 88 % survived the colic surgery 10 , while the survival for horses undergoing surgical intervention for colic at the Bell Equine Veterinary Clinic had a short-term survival rate to discharge of 70.3 % for all horses and 83.15 % for those recovering from anaesthesia 24 .In a Canadian report, a sur-vival rate of 93.6 % for medically treated horses and 73.5 % for surgically treated horses was reported 1 , while in a report of horses with colic in Oslo, Norway, the total survival rate was 63 % with a survival rate of 78 % and 48 % for the medically and surgically treated horses, respectively 21 .In the present study, 10 horses (3 %) needed a 2nd exploratory laparotomy of which 50 % was discharged, which is consistent with results from a similar study 25 .
This study had several limitations.Although the equine hospital at the OVAH serves a wide geographical region, the results of this study may be biased towards causes of colic that occur more frequently within the OVAH region.Horses admitted may be more likely to have suffered more severe causes of gastrointestinal colic.Owing to the retrospective nature of this study, unavailable records and missing data made accurate interpretation difficult.
In conclusion, most horses in this study were Thoroughbreds (54 %), male (57 %) and with a mean age of 8.2 years.Clinical examination, transrectal palpation, nasogastric intubation, packed cell volume and total serum protein determination was most commonly performed following admission.The most common causes of horses receiving medical intervention were impactions (39 %), tympany (7 %) and displacement of the large colon (6 %) while displacement (29 %), impaction (22 %) and small intestinal strangulating lesions (18 %) occurred most often in horses receiving surgical intervention.Overall, impaction and displacement of the large colon was the leading cause of colic in the population of horses.Fifty-five per cent of all the admissions responded favourably to medical intervention, while the recovery rate for horses receiving both medical and surgical intervention was comparable with those of other studies.

Fig. 4 :
Fig. 4: Specific causes of colic affecting the large intestine in horses that received surgical intervention at the Equine Clinic of the OVAH (1998-2007) (percentage of n = 331).

Fig. 5 :
Fig. 5: Specific causes of colic affecting the stomach, small intestine and peritoneum in horses that received surgical intervention at the Equine Clinic of the OVAH (1998-2007) (percentage of n = 331).SI = small intestine.