Materials and Methods

Criteria for selection of cases—Medical records from the VMTH were reviewed to identify horses for which a diagnosis of rectal tear had been made from April 1985 through August 2006. Horses with third-degree perineal lacerations, those for which short-term survival was negatively affected by other conditions not related to the rectal tear, and those for which records had an extensive lack of information were excluded from the study.

Medical records review—Records were retrieved by use of the VMTH's medical and administrative computer system. Where necessary, the hard-copy medical records were also reviewed to obtain pertinent information. For the comparison population, the medical records of all horses evaluated at the VMTH during the same period were reviewed for information regarding age, breed, and sex. This general equine population was defined as total number of horses evaluated at the VMTH on an outpatient or inpatient basis regardless of the initial complaint or performed procedure.

Among the horses with rectal tears, data regarding age, sex, and breed were also collected for comparison with the VMTH's general equine population to assess risk factors for rectal injury in horses. On the basis of age, horses were categorized as follows: < 1 year old (group 1), ≥ 1 to < 5 years old (group 2), ≥ 5 to < 9 years old (group 3), and so forth in 4-year increments. Specific clinical or pathologic findings obtained from the records included etiology of the rectal injury and location, size, distance from the anus, and grade of the rectal tear. With regard to cause, tears were classified as a result of examination per rectum of the reproductive tract, examination per rectum of the gastrointestinal tract because of colic, examination per rectum as part of a general medical examination, dystocia (ie, mares with intraperitoneal and retroperitoneal rectal tears in which part of the perineal body and anal sphincter was intact), rectal prolapse, or unknown cause. The location of a tear within the rectum was classified as dorsal (10 to < 2 o'clock position), lateral (2 to < 4 o'clock position or 8 to < 10 o'clock position), ventral (4 to < 8 o'clock position), dorsolateral (when the tear extended over the dorsal and lateral location), and ventrolateral (when the tear extended over the ventral and lateral location). For each tear, size and distance from anus (in centimeters) were recorded and the grade was classified into 1 of 4 grades (ie, 1, 2, 3, or 4).^9–15

For each grade of rectal tear, treatments administered to the affected horses (eg, medical treatment, suture repair per rectum or via celiotomy, loop colostomy, placement of a temporary indwelling rectal liner, or euthanasia) were noted. Outcome (ie, short-term survival) was recorded as survival to discharge from the hospital versus nonsurvival (death or euthanasia). Medical treatments included administration of combinations of antimicrobials, anti-inflammatory drugs, and feces softeners. Depending on clinical signs, clinician preference, and severity of the tear, IV fluid therapy and manual evacuation of feces were performed.

Statistical analysis—By use of a Kruskal-Wallis test, data from horses with rectal tears in the various age categories were compared with data from horses of the corresponding age categories in the general VMTH population. The Kruskal-Wallis test was also used to assess the influence of rectal tear etiology on size of the rectal tear and its distance from the anus. By use of a χ² test, distributions of breed and sex among horses with rectal tears and among horses in the general VMTH population were compared. A χ² test was also used to evaluate the relationship between rectal tear etiology and location of the rectal injury and to assess the effects of the applied treatments (excluding euthanasia) on short-term survival rate. Differences were considered significant at a value of P < 0.05.

Results

From the medical records generated in the 21-year period of interest, 99 horses with rectal tears were eligible for inclusion in the study. These 99 horses represented 0.1% of the VMTH's general equine population (n = 74,053). The study population included 64 mares, 29 geldings, and 6 stallions. Among the horses with rectal tears, breeds included Arabian (n = 42), Quarter Horse (23), Thoroughbred (13), cross or mixed (5), American Paint Horse (4), American Miniature Horse (3), and other breeds (9). The median age of horses with rectal tears was 12 years (range, 6 months to 32 years). There was 1 horse < 1 year old (group 1), 11 horses ≥ 1 to < 5 years old (group 2), 15 horses ≥ 5 to < 9 years old (group 3), 20 horses ≥ 9 to < 13 years old (group 4), 18 horses ≥ 13 to < 17 years old (group 5), 12 horses ≥ 17 to < 21 years old (group 6), 10 horses ≥ 21 to < 25 years old (group 7), 3 horses ≥ 25 to < 29 years old (group 8), and 3 horses ≥ 29 to < 33 years old (group 9).

Data from the study population and from the general equine population were compared to determine risk factors for rectal tears in horses. Rectal tears occurred more often (P < 0.001) in older horses, particularly in those ≥ 9 years old (Figure 1). Although the VMTH's general equine population was comprised primarily of mares, mares were overrepresented in the study population. There were significantly (P < 0.001) more mares with rectal tears than affected geldings or stallions (Figure 2). Of the various breeds, Arabian and American Miniature Horses were predisposed (P < 0.001) to rectal tears, compared with other breeds represented in the general equine population (Figure 3).

View Full Size

Distribution (%) of horses with rectal tears (black bars; n = 93) by age group, compared with data from the general equine population (gray bars; 74,053) of a VMTH, from April 1985 through August 2006. Age groups were defined as follows: group 1, < 1 year old; group 2, ≥ 1 to < 5 years old; group 3, ≥ 5 to < 9 years old; group 4, ≥ 9 to < 13 years old; group 5, ≥ 13 to < 17 years old; group 6, ≥ 17 to < 21 years old; group 7, ≥ 21 to < 25 years old; group 8, ≥ 25 to < 29 years old; and group 9, ≥ 29 to < 33 years old. *Within an age group, value in horses with rectal tears was significantly (P < 0.001) greater than the value in horses of the general VMTH population.

Citation: Journal of the American Veterinary Medical Association 233, 10; 10.2460/javma.233.10.1605

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Distribution (%) of horses with rectal tears (black bars; n = 99) by sex category, compared with data from the general equine population (gray bars; 74,053) of a VMTH, from April 1985 through August 2006. *Within a sex category, value in horses with rectal tears was significantly (P < 0.001) greater than the value in horses of the general VMTH population. †Percentage of mares with rectal tears was significantly (P < 0.05) greater than the percentages of stallions or geldings in the study population.

Citation: Journal of the American Veterinary Medical Association 233, 10; 10.2460/javma.233.10.1605

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Distribution (%) of horses with rectal tears (black bars; n = 99) by breed category, compared with data from the general equine population (gray bars; 74,053) of a VMTH, from April 1985 through August 2006. *Within a breed category, value in horses with rectal tears was significantly (P < 0.001) greater than the value in horses of the general VMTH population. AR = Arabian. AMH = American Miniature Horse. QH = Quarter Horse. TB = Thoroughbred. CB = Crossbreed. OTH = Other breeds. APH = American Paint Horse.

Citation: Journal of the American Veterinary Medical Association 233, 10; 10.2460/javma.233.10.1605

• Download Figure
• Download figure as PowerPoint slide

The 2 most important causes of rectal tears were colic-related examination per rectum of the gastrointestinal tract (n = 44) or examination per rectum of the reproductive tract (35); examination per rectum as part of general medical examinations, dystocia, and rectal prolapse were the causes of rectal tears in 4, 4, and 3 horses, respectively. In 9 horses, the proximate cause of rectal tears was unknown. In 1 horse in which the rectal tear resulted from a colic-related examination per rectum of the gastrointestinal tract, histologic examination of rectal tissues revealed that the rectum was predisposed to rupture; the rectal tear occurred through a devitalized region of necrosis and inflammation in the muscularis, which developed in association with severe small colon impaction.

The location of the rectal tear was specified in records of 73 of the 99 horses. In 37 of the 73 (51%) horses, the rectal tear was located dorsally. Ventral and lateral tears were present in 14 (19%) and 12 (16%) horses, respectively. Rectal tears were located dorsolaterally in 6 (8%) and ventrolaterally in 4 (6%) horses. Although rectal tears developed more often in a dorsal location, there was no significant (P > 0.05) influence of cause on the location of the tear. The size of the rectal tears ranged from 1 to 27 cm. The median size of tears that were a result of dystocia was 25 cm. The median size of tears that were a result of rectal prolapse, colic-related examination per rectum of the gastrointestinal tract, examination per rectum of the reproductive tract, and examination per rectum as part of a general medical examination was 11.5, 8, 6, and 4 cm, respectively. Of the various causes, only dystocia had a significant (P < 0.02) influence on the size of the tear. The distance of the rectal tear from the anus ranged from 4 to 60 cm. The median distance from the anus for tears that were a result of rectal prolapse was 41 cm. The median distance from the anus for tears that were a result of examination per rectum of the reproductive tract, colic-related examination per rectum of the gastrointestinal tract, dystocia, and examination per rectum as part of a general medical examination was 35, 30, 25, and 23.5 cm, respectively. There was no significant (P > 0.05) association between the etiology and the distance of the rectal tear from the anus.

The grade of the rectal tear was specified in records of 89 of the 99 horses. In 19 horses, the rectal tear involved only the mucosae and submucosae (grade 1). All such grade 1 rectal tears were successfully treated either medically (n = 16) or via blind suture repair (3). Conservative medical treatment was also applied to grade 2 rectal tears in 2 horses, which was successful in both instances. Medical or surgical treatment was performed in 14 of 26 horses that had a grade 3 rectal tear. All horses that underwent loop colostomy (n = 3) or suture repair via celiotomy (2) survived to discharge from the hospital. Blind suture repair was performed in 5 horses, of which 2 were eventually discharged from the hospital. Of the 4 horses with grade 3 rectal tears that were treated medically, 2 survived. Among the 42 horses that had grade 4 rectal tears, 32 horses were euthanatized at admission. Of the 9 horses with grade 4 rectal tears that underwent attempts at treatment or repair, only 1 horse survived to discharge from the hospital; that horse underwent suture repair via celiotomy. In the other 8 horses with grade 4 rectal tears, blind suture repair (n = 5), placement of a temporary indwelling rectal liner (1), and medical treatment (2) were unsuccessful. Irrespective of the method of treatment, the rate of short-term survival (ie, survival to discharge from the VMTH) among horses with grade 1, 2, 3, or 4 rectal tears was 100%, 100%, 38%, and 2%, respectively. No significant (P > 0.05) association between any treatment and shortterm survival rate was identified.

Discussion

To our knowledge, this is the first reported study in which comparisons between horses with rectal tears and horses in a defined general population were undertaken to assess risk factors associated with the development of rectal tears. On the basis of our findings, the identified risk factors included age (ie, ≥ 9 years old), sex (ie, female), and breed (ie, Arabian and American Miniature Horse). The greater risk of rectal tears in older horses may be attributable to a more pendulous reproductive tract,²⁷ the increased frequency of rectal examinations of subfertile older broodmares, and an increased incidence of medical issues necessitating examination per rectum. In addition to increasing age, decreased elasticity,¹⁷ degenerative changes, denervation, and previous injury⁶ of the rectal wall can predispose the rectal wall to injury.

There is some controversy regarding sex predilection of rectal tears in horses. Results of previous studies^2,6 suggest that stallions and geldings are more predisposed to rectal tears; however, in our study population, more mares had rectal tears than did geldings or stallions. The increased frequency of rectal injuries in mares may be attributed to the fact that examinations per rectum of the reproductive tract are performed frequently in mares, particularly during the breeding season; however, it is still important to keep in mind that stallions and geldings are less accustomed to palpation per rectum.^2,6 Factors such as restless behavior, small rectum, and rectal straining most likely contribute to the greater occurrence of rectal tears in Arabians and American Miniature Horses.^2,6,9,17 Awareness of important risk factors can be translated into appropriate measures to reduce the risk of rectal tear development, such as provision of adequate restraint (use of a twitch or sedation), application of large amounts of lubricants, and administration of a lidocaine enema or spasmolytic drugs (eg, N-butylscopolammonium bromide and propantheline).

In the present study, examination per rectum of the gastrointestinal tract because of colic or examination per rectum of the reproductive tract represented the most frequent types of procedure associated with rectal injury. A similar finding in 2 retrospective studies^12,13 has been reported. The frequency of rectal tear development in horses that underwent colic-related examination per rectum of the gastrointestinal tract may not be surprising given that the rectal mucosa may be friable and dry as a result of dehydration, that multiple examinations per rectum are likely undertaken, and that adequate restraint under certain circumstances may be difficult to achieve, all of which are potential predisposing factors for rectal tears.⁹ Therefore, veterinarians conducting this type of examination per rectum should be particularly cognizant of this association.

The most common location of rectal tears in the present study was dorsal (10 to < 2 o'clock position), which is in agreement with findings of previous studies.^4,6,11,17 There is some suggestion that tearing on the dorsal aspect of the rectum might be attributable to particular anatomic features. The dual arterial blood supply from the mesocolon (the mesentery joining the rectum to the dorsal portion of the abdominal wall) penetrates the rectum dorsally, which may make the rectum susceptible to trauma in this area.^4,6,17 Second, histologic examination of the sites where the arteries enter the muscular layer reveals evidence of a decrease in thickness of the circular muscular layer, compared with other locations along the rectum.¹⁷ Furthermore, there is a ventral deviation of the rectum at the pelvic inlet, which coincides with a decrease in luminal size of the rectum.^6,17

In the present study, no association between the type of examination per rectum and the distance from the anus to the rectal tear was identified. In addition to penetration of the rectal wall by the fingertips, rectal tears may develop as a result of contractions of the rectum around the clinician's hand; such contractions may occur at any location within the rectum.⁹ With regard to the specific type of examination (ie, examination per rectum of the reproductive tract, examination per rectum of the gastrointestinal tract because of colic, or examination per rectum as part of a general medical examination), the location of the abdominal organs in relation to the distance from the anus varies among the different sizes of horses.

Not surprisingly, dystocia was associated with large rectal tears. Delivery of a malpositioned fetus⁴ and forceful abdominal contractions that often accompany dystocia may contribute to increased pressure on the rectum against the pelvis, which may lead to tearing of the rectum.

Conservative medical treatments have been applied mainly to grade 1 and 2 rectal tears in horses¹¹; however, medical management of grade 3 rectal tears can be successful, as determined in the present study and in other investigations.^{3,13,15,28,29} Of the surgical treatments for horses with grade 3 rectal tears in our study, loop colostomy and suture repair via celiotomy appeared to be superior (5/5 horses survived to discharge from the hospital), compared with blind suture repair via rectum (2/5 horses survived short-term) or medical treatment (2/4 horses survived short-term). However, it is difficult to make conclusions about the effectiveness of each treatment because of the limited number of horses that underwent those procedures. Each technique has certain advantages and disadvantages, and some techniques are not applicable in certain circumstances. The accessibility of the rectal tear can be a major issue in suture repair via celiotomy.²⁰ Viability of the rectal wall in the region of the tear, contamination of the tear, and distance from the anus to the tear are important features that need be evaluated carefully before the decision is made to suture the tear via the rectum.⁹ Tears that are > 25% of the rectal diameter or located > 50 cm inside the rectum are indications for the use of colostomy.¹² Besides the fact that colostomy is a 2-step technique, there are several complications associated with that procedure.^22,24 It has been hypothesized that elimination of feces accumulation around a rectal tear is responsible for the successful outcome associated with repeated manual evacuation of the rectum. That conservative treatment technique is much easier and less expensive than surgery; however, it is not recommended for rectal tears > 10 cm in diameter, and it is fairly labor intensive during the first few days of treatment.²⁸

For horses with full-thickness (grade 4) rectal tears in the present study, suture repair via celiotomy was the only approach that was successful. There are limited reports of other successful techniques, such as suturing the tear per rectum by use of a Deschamps needle,¹⁶ closure of a tear per rectum by use of a stapling device,^5,18 combination of colostomy and direct suturing, combination of placement of a temporary indwelling rectal liner and direct suturing,¹³ and repair via celiotomy.²⁰ Many of these reports include a limited number of animals, and comparison of outcomes is therefore difficult to interpret.

The overall short-term survival rate among horses with grade 1, 2, and 4 rectal tears in the present study was fairly similar to values determined in other large retrospective studies.^12,13 However, with regard to horses with grade 3 rectal tears, the survival rate in the present study (38%) appeared to be lower than survival rates in 2 other studies^12,13 (approx 60% to 70%). This lower survival rate for horses with grade 3 rectal tears in our study can be most likely attributed to the large number of horses euthanatized at admission because of economic constraints or fecal contamination of the abdominal cavity. Because of a lack of information in the database, we were not able assess whether initial management performed immediately after the occurrence of rectal tears (prior to referral to the VMTH) was adequate; it is known that an appropriate initial management increases the survival rate.^12–14,30

Because all horses with grade 1 and 2 rectal tears in the present study survived to discharge from the hospital, it was not possible to compare the efficacy of treatments for grade 1 and 2 rectal tears. Although some treatments for grade 3 and 4 rectal tears were promising, an insufficient number of treated horses in these groups precluded detection of a significant association between the different treatments and short-term survival. Therefore, no conclusions regarding an ideal treatment for grade 3 and 4 rectal tears in horses could be made. The treatment most likely to correct a rectal tear should be selected on the basis of grade of the rectal tear, viability of the disrupted tissue, surgeon's experience, associated costs,¹³ distance of the tear from the anus, and localization of the tear.²⁸