Ileal impaction in 245 horses: 1995–2007

Abstract

The objective of this study was to identify parameters that would assist in determining the probability of a successful outcome with medical management versus surgical intervention in horses with ileal impaction. Medical records of 245 horses admitted for ileal impaction were reviewed and placed into 2 groups: medical (med) and surgical (sx) treatment. Persistence of abdominal pain, gastric reflux, frequency of analgesic administration, and 1-year survival were evaluated. There were no differences in signalment, abdominal pain, or heart rate among groups; however, significantly more sx horses had peritoneal fluid abnormalities (51%) and produced gastric reflux (62%) than did med horses (38% and 15%, respectively). Eighty-nine percent of med horses required repeated analgesic administration for successful resolution. One-year survival was 91% for sx horses and 92% for med horses. Horses with ileal impaction responsive to analgesic therapy with minimal gastric reflux are likely to be managed successfully with medical treatment. Horses with persistent abdominal pain and gastric reflux are candidates for surgery.

Introduction

The most common site for intraluminal impaction of the small intestine in horses is the ileum. The incidence of ileal impaction varies with geographic location; the condition occurring more commonly in the southeastern United States. In retrospective studies of acute abdominal disease, ileal impaction has been reported to have a prevalence of 0.5% to 10.8% and accounts for 12% to 42% of all ileal obstructions (1). The exact cause is unknown, but an association with fine, high-roughage forage, Coastal Bermuda hay, and tapeworm infestation (Anoplocephala perfoliata) has been reported (1–3). Coastal Bermuda is typically a stemmy, fine hay that increases markedly in its lignin content and consequently, crude fiber content, as it matures. As the crude fiber content of the hay increases, the dry matter, protein, and energy digestibility decrease (4). Other factors (quality of Coastal Bermuda hay, decreased water intake, or weather change) are likely important factors associated with the disease. Dilution of Coastal Bermuda hay with another type of hay does not appear to eliminate the risk of ileal impaction and recent introduction of the hay and mistaken feeding of Coastal Bermuda hay appear to be contributing factors in many cases of ileal impaction (1–3). The impaction results from accumulation of ingesta in the ileum, resulting in complete intraluminal obstruction. Spasmodic contraction of the ileum and absorption of water from the ileum dehydrate the ingesta mass and exacerbate the impaction (1,5). In the western United States where legume hays are the primary source of roughage, ileal impaction rarely occurs. Less common causes of ileal impaction include ileal hypertrophy, mesenteric thrombotic disease, and ascarid impaction (1,5).

Clinical signs associated with ileal impaction are variable and depend on the duration of the impaction (5). Early in the disease, horses show intermittent signs of moderate to severe abdominal pain due to distention of the ileum and spasmodic contraction around the impaction. Gastrointestinal borborygmi are often present. Gastric reflux is often absent during the initial physical examination. During the initial transrectal examination, small intestinal distention may not be palpable, or a 5- to 8-cm diameter, firm, smooth-surfaced ileum may be detected originating at the cecal base and coursing from the right of midline obliquely downward, cranial, and to the left side (5). Affected horses usually have a transient response to analgesic therapy. Generally, 6 to 12 h following the onset of obstruction, rectal palpation reveals multiple loops of distended small intestine with variable signs of abdominal pain. Abdominal ultrasonography generally reveals distended, thin-walled, non-motile small intestine (5). As the severity of the small intestinal distention progresses, signs of abdominal pain become more persistent, borborygmi diminish, and abdominal pain becomes refractory to analgesic therapy.

Although ileal impactions in horses have been treated successfully by both medical and surgical treatments, the decision to continue with aggressive medical therapy or yield to surgical intervention may be difficult. The objective of this study was to identify clinical parameters that would assist the clinician in determining the probability of a successful outcome with medical management versus the need for surgical intervention. Our hypothesis was that horses with ileal impaction that are likely to respond successfully to medical therapy do not produce significant amounts of nasogastric reflux and are responsive to analgesic therapy, while horses with positive gastric reflux and persistent abdominal pain despite analgesic therapy will require exploratory celiotomy and surgical correction for a successful outcome.

Materials and methods

Medical records of 245 horses admitted to the University of Georgia Veterinary Medical Teaching Hospital for suspected or confirmed ileal impaction between 1995 and 2007 were reviewed. Horses with a suspected diagnosis of ileal impaction, which were included in the study, had a history of recurrent abdominal pain, a history of being fed Coastal Bermuda hay, and clinical signs consistent with non-strangulating small intestinal obstruction determined by evidence of small intestinal distention based on transrectal palpation and/or abdominal ultrasound examination. Horses with confirmed ileal impaction were diagnosed by palpation of the impaction at admission, or confirmed at surgery or at postmortem examination. Horses with evidence of strangulating intestinal obstruction (serosanguineous peritoneal fluid or lesion identified at exploratory celiotomy) were excluded. Horses were placed into one of 2 groups: horses that were successfully treated with medical therapy alone (med) and those that required exploratory celiotomy for resolution of the impaction (sx).

For horses in both med and sx groups, details of age, breed, sex, deworming history, type of hay fed (timothy, alfalfa, Coastal Bermuda grass, fescue, orchard grass, or other), deworming history, and duration of clinical signs prior to admission were obtained from the medical record. Clinical parameters obtained from the record at admission included: time from initial signs of abdominal pain to referral; degree of abdominal pain (intermittent, mild, moderate, or severe); heart rate; mucous membrane color and capillary refill time; borborygmi (present or absent); presence of gastric reflux; transrectal examination findings (ileal impaction palpable, presence or absence of distended small intestine); and cytology of peritoneal fluid [color, total protein (g/L), nucleated cell count (cells/μL)].

Data retrieved from the medical record for both groups during hospitalization included: persistence of abdominal pain, type and frequency of visceral analgesic medication administration, persistence of gastric reflux, and 1-year survival. For horses requiring exploratory celiotomy, data retrieved also included: method of correction (manual reduction, enterotomy, or jejunocecostomy), duration of postoperative gastric reflux, and the need for additional celiotomies.

Statistical analysis

All analyses were performed using SAS V 9.2 (SAS Institute, Cary, North Carolina, USA). A chi-squared test was used to compare abnormal and normal proportions (temperature and abdominocentesis results) and yes and no proportions (palpation of the ileum, gastric reflux, survival and re-impaction of the ileum) between surgical and medical groups. A Wilcoxon rank-sum test was used to compare scored variables, time to referral (1 = < 5 h, 2 = 6 to 12 h, 3 = 13 to 24 h, 4 = > 24 h); heart rate [0 = < 51 beats/min (bpm), 1 = 51 to 59 bpm, 2 = 60 to 69 bpm, 3 = > 69 bpm]; duration of reflux (0 = none, 1 = < 12 h, 2 = 12 to 24 h, 3 = 25 to 36 h, 4 = > 36 h); and number of administrations of an analgesic (0 = none, 1×, 2×, 3×, or 4×) between surgical and medical groups. All tests were 2-sided and the significance level was α = 0.05. The chi-square was implemented using PROC FREQ and the Wilcoxon rank-sum test in PROC NPAR1WAY in SAS.

Results

A total of 245 cases of suspected or confirmed ileal impaction were retrieved from the medical record. One hundred and thirty-three horses (54%) were managed with medical therapy alone (med) and 112 horses (46%) required ventral midline celiotomy to resolve the impaction (sx). There were no significant differences in the distribution of age, breed, or sex, between groups. Signs of abdominal pain were described as mild 57/245 (23%), moderate 137/245 (56%), and severe 51/245 (21%). Heart rate at initial examination was within normal limits (< 48 bpm) in 105/245 (43%) horses and elevated (> 48 bpm) in 140/245 (57%). Seventy-seven of 133 (58%) medically treated horses and 63/112 (56%) of surgically treated horses were presented with an elevated heart rate (> 48 bpm). There were no significant differences in signs of abdominal pain or heart rate at admission between groups. At admission, the ileum was palpable upon transrectal examination in 67/245 (27%) of all horses. The ileum was palpable in 39/133 (29%) of med horses, which was not significantly different than that of sx horses, 28/112 (25%). In all horses, small intestinal distention was evident either upon transrectal examination or ultrasound evaluation. Cytologic evaluation of peritoneal fluid revealed abnormalities (total protein concentration > 20 g/L or nucleated cell count > 2000 cells/μL) (6) in 100/245 (41%) horses. Significantly more sx horses had abnormalities detected on cytological evaluation of peritoneal fluid (57/112; 51%) than did med horses (43/133; 32%). Of horses with suspected ileal impaction that were managed with medical treatment, only 20/133 (15%) produced gastric reflux during treatment. Significantly more of the sx treated horses 69/112 (62%) produced gastric reflux compared with medically treated horses (Table 1).

Table 1.

Comparison of clinical parameters, necessity for repeated administration of analgesic medication, and survival among horses treated with medical versus surgical management for successful treatment of ileal impaction in horses

	Medical management (n = 133)	Surgical management (n = 112)
Elevated heart rate	77 (58%)	63 (56%)
Positive gastric reflux	20 (15%)	69 (62%)a
Palpable ileal impaction	39 (29%)	28 (25%)
Abnormal peritoneal fluid	43 (32%)	57 (51%)a
Required 1 repeat of analgesic administration for successful treatment	74 (55%)	NA
Required 2 to 4 repeats of analgesic administrations for successful resolution	22 (17%)	NA
1-year survival	123 (92%)	102 (91%)
Re-impaction of ileum	2 (1.5%)	2 (1.8%)

^aIndicates significant difference between groups (P < 0.05).

All horses were treated with an indwelling nasogastric tube, intravenous balanced isotonic fluids, and continuous monitoring. Ninety-six (72%) of med horses had signs of persistent abdominal pain that required repeated administration of analgesic medications: flunixin meglumine (Banamine; Schering Plough, Kenilworth, New Jersey, USA), xylazine (Rompum; Bayer Co., Shawnee Mission, Kansas, USA), detomidine (Dormosedan; Pfizer Animal Health, Westchester, Pennsylvania, USA), butorphanol tartrate (Torbugesic; Fort Dodge Animal Health, Division of Wyeth, Overland Park, Kansas, USA), or N-butyl-scopolammonium bromide (Buscopan; Boehringer Ingelheim Vetmedica, St. Joseph, Missouri, USA) for successful resolution of clinical signs. Seventy-four horses (55%) required a single administration of analgesic medication and 22 (17%) required 2 to 4 administrations of analgesic medications. Medically treated horses received significantly more administrations of analgesic medications than did surgically treated horses. One hundred and twenty-three/133 (92%) medically managed horses survived at least 1 y. Eleven/133 (8%) of med horses were euthanized because of economic reasons. Ileal impaction was confirmed at necropsy in all 11 of these horses.

In all sx horses, exploratory celiotomy was performed after failure of medical therapy as evidenced by persistent signs of abdominal pain despite analgesic therapy, worsening small intestinal distention, increasing production of gastric reflux and deteriorating cardiovascular status. The estimated length of ileal impactions ranged from 0.30 to 1.5 m, but an accurate mean length could not be determined from review of the surgical reports. The impaction was reduced by infusion of the impaction with a mixture of 1% sodium carboxymethylcellulose (Aqualon Co., Wilmington, Delaware, USA), balanced isotonic fluids, and 2% lidocaine HCl (Lidocaine HCL 2%; Hospira, Lake Forest, Illinois, USA) and extra-luminal massage in 107/112 (96%) horses; distal jejunal enterotomy and evacuation with warm water lavage in 5/112 (4%) horses. One-to-two liters of 1% sodium carboxymethylcellulose (SCMC) solution were used in all surgically treated horses to facilitate intestinal manipulation and minimize serosal trauma during extra-luminal message of the impaction. Jejunocecostomy was not required in any of the horses.

Postoperatively, 56/112 (50%) horses produced reflux, with 7/112 (6%) refluxing for < 24 h; 26/112 (23%) refluxing for 24 to 36 h; 11/112 (10%) refluxing for 27 to 48 h; and 12/112 (11%) refluxing for more than 48 h. Eight of 112 (7%) sx horses required a second celiotomy because of persistent ileus, intestinal displacement, or adhesions. One hundred and two of 112 (91%) surgically treated horses survived at least 1 y. Ten of 112 (9%) horses were euthanized before discharge for persistent pain, ileus, and/or economic considerations. One-year mean survival rates between medically managed and those requiring surgery was not significantly different.

Re-impaction of the ileum occurred in 4/245 (1.6%) horses. Two surgically treated horses were re-admitted 6 mo and 4 y after surgery, and 2 medically managed horses were re-admitted 4 mo and 1 y after the initial admission. In all 4 horses, the diagnosis of ileal impaction was confirmed by transrectal palpation of the impaction at re-admission.

Discussion

The decision to manage ileal impaction in horses with aggressive medical therapy or yield to surgical intervention may be difficult. In this study, clinical and physical examination parameters evaluated at admission such as the severity of abdominal pain, heart rate, presence or absence of borborygmi, transrectal examination findings, and abdominal ultrasound examination findings were not helpful in determining whether or not medical management or exploratory celiotomy would lead to successful resolution of the impaction.

In 27% of the horses, the ileal impaction was identified on transrectal palpation, thereby, confirming the diagnosis. This is similar to data from 2 previous smaller studies in which the ileum was palpable in 25% and 39% of horses, respectively (7,8), but higher than in 1 study, in which only 10% of the impactions were palpated (2). The reason for these differences may be the stage of the disease when the horses were evaluated. Early in the disease, the ileum may be more easily palpable in the right caudal abdomen, but with increased duration, the development of multiple loops of distended small intestine may preclude the examiner from palpating the impaction. Interestingly, there were no differences in the ability to palpate the ileum in horses that resolved with medical management versus those that required surgery.

Horses that responded to medical management produced significantly less gastric reflux both at admission and during treatment. Only 15% of horses that responded favorably to medical management produced gastric reflux compared with 62% of horses that required surgery. A lower percentage of medically treated horses refluxed in this study compared with a previous study of 10 horses with ileal impaction that were managed medically, in which 40% of horses produced reflux (9). Our results in surgically treated horses are in agreement with 2 previous studies of surgical treatment of ileal impactions in which, 56% and 54% of horses produced reflux (7,8). Therefore, it appears that the absence of gastric reflux may be a valuable indicator in determining which horses will respond favorably to medical management. While 50% of the surgically treated horses refluxed after surgery, only 11% refluxed for more than 48 h after surgery. Small intestinal distention, intestinal manipulation, and anesthesia are all reported contributing factors for postoperative ileus (10). While some degree of postoperative ileus should be expected in horses that are treated surgically, ileus and gastric reflux most often resolve within 24 to 36 h after surgery.

Seventy-two percent of the medically managed horses had persistent signs of abdominal pain that required repeated administrations of analgesics for successful resolution of the impaction. The majority (55%) of horses required only 1 additional administration of analgesia for resolution; however, some horses required up to 4 administrations of analgesic treatment before resolution of the clinical signs. It is imperative to continuously monitor clinical parameters during medical management of horses with suspected ileal impaction to assure adequate preservation of the patient’s cardiovascular status. It is not uncommon, however, for horses to require at least 1, and sometimes 2 or more, administrations of a visceral analgesic during the course of successful medical management. It is important to note, however, that referral to a hospital facility for close monitoring and possible surgical treatment should always be offered to owners when horses are refractory to initial medical therapy and require multiple administrations of analgesic medication. While horses that were successfully managed medically received a significantly greater number of repeat administrations of analgesics compared with surgically managed horses, this likely reflects a more rapid progression of the clinical signs and an earlier decision for surgery in the surgically managed group.

Ninety-two percent of medically treated horses survived at least 1 y. These results are consistent with 1 report of 10 cases of medically managed ileal impactions in which all 10 horses survived to discharge (9), but in marked contrast to a report of 288 cases of ileal impaction, in which medical treatment was attempted in 13 cases, with only a 10% to 20% survival rate (11). Treatment failures with medical management early in the course of the latter study may have led to a surgeon’s preference to pursue early surgical intervention. This would explain the high percentage (95%) of surgically treated horses and low success rates for medical management. In the present study, 8% of medically treated horses were euthanized for economic reasons. In all 11 of these horses, an ileal impaction was confirmed at necropsy, supporting the initial presumptive diagnosis.

In contrast to previous studies (2,7,8), jejunocecostomy was not required in any of the horses that underwent exploratory celiotomy and surgical correction. This may be related to the surgeons’ preference for infusion of SCMC, balanced isotonic fluids, lidocaine mixture, or enterotomy to facilitate reduction of the impaction (5). Treatment of ileal impaction with jejunocecostomy has been associated with decreased survival rates (7), and should not be performed unless additional pathology such as ileal hypertrophy, stricture, or dysfunction is evident (2). A distal jejunal enterotomy and evacuation and reduction of the impaction with warm water lavage were necessary in 5/112 surgically treated horses. It is the authors’ opinion that jenunal enterotomy and warm-water lavage are associated with less serosal trauma of the ileum and distal jejunum and less postoperative morbidity than multiple and prolonged attempts at reduction by extra-luminal massage.

Ninety-one percent of surgically treated horses survived at least 1 y, which was, markedly improved over earlier reports (7,11,12), but in agreement with more recent reports (2,8). However, 7% of these horses required a second celiotomy and 5.7% were euthanized secondary to complications of persistent ileus, intestinal displacement, small intestinal adhesions, indicating that although the overall prognosis for ileal impaction is good, serious complications may result in less favorable outcomes. Although survival rates for horses with ileal impaction managed medically or surgically in the present study were not significantly different, medical treatment is associated with less patient morbidity and decreased expense to the owner, and when feasible, should be the first choice of treatment.

Re-impaction of the ileum has been reported previously (2,7), and occurred in 1.6% of horses in this study. One impaction occurred as late as 4 y after surgery. In all of the horses of this study, re-impaction of the ileum was associated with reintroduction of the horse to Coastal Bermuda hay. Therefore, feeding Coastal Bermuda hay to horses, which have had a previous history consistent with an ileal impaction, should be avoided whenever possible.

One inherent weakness in this study is the presumptive diagnosis of ileal impaction in horses that were treated medically in which the ileal impaction was not confirmed by palpation. In the southeastern United States where these horses were treated, ileal impactions are a common occurrence. The combination of a history of consuming Coastal Bermuda hay, clinical signs and clinicopathologic laboratory findings consistent with simple, non-strangulating, small intestinal obstruction are highly suggestive of ileal impaction. In addition, in all suspect horses in which medical therapy was attempted, but failed, and surgery was not an option, the diagnosis of ileal impaction was confirmed at postmortem examination. However, it is possible that some of the cases in this report in which an ileal impaction was diagnosed, but not confirmed by surgery, may have had a lesion other than and ileal impaction.

In summary, diagnosis of ileal impaction should be considered in horses evaluated for clinical signs consistent with non-strangulating, small intestinal obstructive disease and a history of being fed a poor quality roughage, Coastal Bermuda hay, or being dewormed with an anthelmintic with poor efficacy against tapeworms. Horses with an ileal impaction that are likely to be managed successfully with medical treatment are diagnosed early in the disease, have abdominal pain that is responsive to analgesic therapy, and do not develop gastric reflux during treatment. Horses with persistent signs of abdominal pain that is not responsive to analgesic therapy, persistent gastric reflux, progressive small intestinal distention, or deteriorating cardiovascular status are candidates for exploratory celiotomy and surgical correction. The long-term prognosis for horses treated with either medical management or surgical intervention is good.