Transabdominal ultrasonography is increasingly performed in equine veterinary medicine to evaluate horses with various initial complaints, including acute and chronic colic, weight loss, fever of unknown origin, diarrhea, signs of nonspecific organ dysfunction, neoplasia, and internal abscess formation.1–4 In horses, complete ultrasonographic evaluation of the abdomen consists of assessment of parenchymal organs and gastrointestinal structures, including the stomach, descending duodenum, jejunum, cecum, and large colon.1,5 Ultrasonographic evaluation of the gastrointestinal tract primarily involves assessment of intestinal wall thickness, distention, and motility.1,2,5–7 Ultrasonographic imaging of the large intestine is usually limited to the near wall, because intestinal gas and ingesta reflect the ultrasound beam and limit assessment of the intestinal lumen and far wall. The ultrasonographic appearance of the gastrointestinal tract has been documented in clinically normal horses and ponies, with mean ± SD large intestinal wall thickness reportedly ranging from 1.8 ± 0.03 mm to 4.2 ± 0.3 mm.8–10
The ultrasonographic appearance of large intestinal abnormalities that require surgical treatment has been described. These include large colon displacement, colon torsion, and intussusception.6,7,11–16 Severe large intestinal thickening (ie, large intestinal wall thickness ≥ 9 mm) has been reported as an accurate preoperative indicator of colon torsion when identified in the ventral aspect of the abdomen in horses examined because of signs of surgical colic.13 Although signs of colic associated with colon torsion are often more pronounced than are signs associated with other causes of large intestinal thickening, it can be challenging to differentiate surgical from nonsurgical diseases in some horses. In such cases, it is possible that an ultrasonographic finding of severe large intestinal thickening (≥ 9 mm) could lead to a false-positive diagnosis of colon torsion. Such a scenario would be detrimental whether the owner opted for surgical intervention or euthanasia.13
The use of ultrasonography to aid in the diagnosis of right dorsal colitis has also been reported and is widely known among equine practitioners.17–19 Large intestinal abnormalities that induce ultrasonographically detectable changes in intestinal wall thickness or echogenicity, such as infiltrative intestinal disease or neoplasia, have been described but seem to be less well understood.4,20,21 Therefore, the purpose of the study reported here was to describe the clinical and ultrasonographic features and outcome of horses with severe large intestinal thickening (ie, large intestinal wall thickness ≥ 9 mm) identified by means of transabdominal ultrasonography.
Materials and Methods
Case selection criteria and medical records review
The medical records database of the William R. Pritchard Veterinary Medical Teaching Hospital at the University of California-Davis was reviewed to identify all horses that underwent ultrasonographic evaluation of the abdomen by the Large Animal Ultrasound Service between January 2003 and December 2010. Horses were included in the study if cecal or colonic wall thickness ≥ 9 mm was identified in any of 6 abdominal zones: left and right paralumbar fossa, left and right intercostal spaces (4th through 17th), and left and right ventrum (sternum to inguinal region, midline to costochondral articulations). Horses with a focal area of large intestinal wall thickening (defined as a single site ≥ 9 mm but < 2 cm in diameter) were excluded because the study objective was to identify patients with ultrasonographic evidence of severe regional or diffuse colonic thickening. Horses were also excluded if a complete ultrasonographic examination of the abdomen had not been performed.
Data extracted from medical records of horses included in the study consisted of signalment, history, initial complaint, results of laboratory testing, results of transabdominal ultrasonography, and outcome. For discharged horses that were returned for either recheck or unrelated examinations, additional outcome data were obtained from the medical record. For all discharged horses, telephone interviews with owners or referring veterinarians were performed by 1 author (EWB) to determine whether horses had had complete or incomplete resolution of clinical signs or were euthanized for related reasons.
Ultrasonographic examinations were performed with 1 of 2 machinesa,b equipped with a 2.5- to 5.0-MHz curvilinear transducer. For all examinations, acoustic gel was applied to the abdomen after clipping the hair with a No. 50 blade and washing the skin. Horses were placed in stocks and sedated with detomidinec (0.004 to 0.008 mg/kg [0.0018 to 0.0036 mg/lb], IV) as needed. Each abdominal zone was sequentially evaluated for abnormalities of the kidneys, liver, spleen, stomach, duodenum, small intestine, large colon, cecum, and cecal mesentery. Representative images of all abdominal structures within each zone were captured, and images were stored in the archival system of the ultrasound machine and printed on thermal paper. Video images were captured as deemed appropriate when they provided additional information. Measurements of intestinal wall thickness were acquired at the time of image acquisition with electronic calipers and software provided with the ultrasound machine. Maximal wall thickness measurements in each of the 6 zones were used for data analysis. Ultrasonographic examinations were performed or directly supervised by 1 of 2 clinicians (MBW and BV) with extensive experience in diagnostic ultrasonography. All ultrasonographic reports and images were reviewed by 3 authors (MBW, BV, and EWB), and ultrasonographic findings were extracted and recorded for data analysis.
Statistical analysis
Continuous data were summarized as mean ± SD and median; categorical data were summarized as percentages. For the purposes of data analysis, horses were classified as survivors if they were still alive at the time of follow-up and had complete (ie, fully recovered) or partial (ie, partially recovered) resolution of their initial clinical signs; horses were classified as nonsurvivors if they died or were euthanized. Horses lost to follow-up were excluded from analyses. The Wilcoxon rank sum test was used to compare age, duration of clinical signs prior to initial evaluation, results of abdominocentesis, and large intestinal wall thickness measurements between survivors and nonsurvivors. The Fisher exact test was used to test whether outcome (ie, survivor vs nonsurvivor) was significantly associated with sex, the presence or absence of specific clinical signs, or the presence or absence of specific ultrasonographic findings. For categorical data, odds ratios and their 95% CIs were calculated if a significant association was identified. All statistical analyses were performed with commercially available statistical software.d Values of P < 0.05 were considered significant.
Results
Patient characteristics
Twenty-five horses met the inclusion criteria, including 8 Quarter Horses or Quarter Horse crossbreds, 3 Thoroughbreds, 2 Appaloosas, 2 Morgans, 2 ponies or pony crossbreds, 2 Arabians, and 1 each of the following: warmblood, Peruvian Paso, miniature horse, Tennessee Walker, Missouri Fox Trotter, and draft crossbred. Nine of the horses were geldings, 13 were females, and 3 were stallions. Ages ranged from 3 to 28 years (median, 13 years). Horses were used for a variety of purposes and included 11 pleasure horses, 2 show horses, 1 pet horse, 1 reining horse, 1 rodeo horse, and 1 western pleasure horse. Three horses were retired. There were 5 broodmares (3 Thoroughbreds, 1 Arabian, and 1 warmblood; age range, 10 to 13 years). Three broodmares were evaluated 4, 10, and 100 days after foaling, and 1 broodmare was evaluated at 120 days of gestation. The remaining broodmare had foaled earlier in the year, but the date was not recorded.
Clinical signs
Duration of clinical signs prior to initial examination at our hospital ranged from 1 to 30 days in 24 horses and was 5 months in the remaining horse (overall median, 2 days). Eleven of the 25 horses were evaluated within 24 hours after the onset of clinical signs, and 17 horses underwent ultrasonographic examination within 24 hours after initial evaluation. Initial complaints included colic (n = 13), diarrhea (10), reduced appetite (10), weight loss (9), lethargy (4), fever (2), and hematuria (1). Of the 13 horses examined because of colic, 9 had signs of colic at the time of initial examination. Severity of colic was reported in the medical record as mild (n = 5), moderate (3), or severe (1). Two of the 13 horses with colic had a history of chronic colic, and all 13 had had an episode of colic within 30 days prior to the initial evaluation. Horses with diarrhea initially had soft (ie, cow-pie consistency; n = 8) feces or intermittent (1) or profuse (1) watery diarrhea. Two horses with diarrhea had a recent history of antimicrobial administration. Three horses had undergone abdominal surgery 6 to 26 days prior to ultrasonographic examination, including 1 horse that had undergone surgery (but not intestinal resection) for treatment of colon torsion 6 days earlier, 1 horse that had undergone cecal biopsy during exploratory celiotomy 12 days earlier, and 1 horse with large colon infarction that underwent resection of 60% of the large colon 26 days earlier.
Ultrasonographic findings
Severe large intestinal thickening (ie, large intestinal wall thickness ≥ 9 mm) was the predominant ultrasonographic finding in all horses, with thickness of the affected portions of intestinal wall ranging from 9 to 46.6 mm (mean ± SD, 18.8 ± 6.8 mm; median, 18). Severe large intestinal thickening was identified in the right intercostal abdominal zone in 18 horses (Figure 1), right ventrum in 18, left ventrum in 16, right paralumbar fossa zone in 10, left intercostal zone in 7, and left paralumbar fossa zone in 4. Twenty of the 25 (80%) horses had severe large intestinal thickening identified in the ventral aspect of the abdomen (left ventrum, right ventrum, or both). Severe large intestinal thickening was identified in 5 zones in 5 (20%) horses, 4 zones in 3 (12%) horses, 3 zones in 6 (24%) horses, 2 zones in 8 (32%) horses, and 1 zone in 3 (12%) horses. No horses had severe large intestinal thickening (≥ 9 mm) in all 6 abdominal zones. Mild to moderate large intestinal thickening was recorded in adjacent zones in 8 of 25 (32%) horses.
Anatomic location of zones evaluated during abdominal ultrasonography of 25 horses suspected to have abdominal abnormalities. The right and left paralumbar fossa zones (black) were evaluated from the level of the tuber coxae to the inguinal region. The right and left 4th to 17th intercostal zones (dark gray) were evaluated from the ventral lung margins to the costochondral articulations. The right and left ventrum zones (light gray) were evaluated from the sternum to the inguinal region and from the ventral midline to the costochondral articulations. Numbers shown in each zone corresponded to the number of horses with severe large intestinal thickening (ie, large intestinal wall thickness ≥ 9 mm) identified during ultrasonographic examination.
Citation: Journal of the American Veterinary Medical Association 253, 1; 10.2460/javma.253.1.108
The ultrasonographic appearance of the severely thickened intestinal walls varied within and between horses (Figure 2). Diffusely anechoic to hypoechoic wall thickening was the predominant feature in 21 of 25 (84%) horses. Affected intestinal walls appeared edematous in these horses and often had striations that did not correspond to intestinal layers. Eight of these horses also had scattered areas of echogenic or heterogeneous wall thickening. Intestinal layering was detectable in some horses or in other locations within the same horse. Visualization of intestinal layering or striations was impeded in horses with thick subcutaneous or retroperitoneal fat layers. In the remaining 4 horses, intestinal walls had either a predominantly echogenic appearance or a mixed, somewhat heterogeneous appearance. In horses with substantial luminal fluid, the far wall of the colon or cecum was intermittently visible.
Ultrasonographic appearance of severe large intestinal thickening in 3 horses; all 3 horses recovered with medical management. A—Severe large intestinal thickening and an edematous, homogeneously hypoechoic appearance of the intestinal wall in the right ventrum abdominal zone in a 12-year-old Thoroughbred broodmare. B—Severe large intestinal thickening with an edematous striated appearance in the left ventrum zone in a 7-year-old German Riding Pony with a maximal wall thickness of 34.0 mm (between arrows). Visible striations do not appear confined to a single intestinal layer. C—Severe large intestinal thickening (large intestinal wall thickness between arrows, 12.5 mm) with visible layering in the right ventrum zone in an 11-year-old Welsh Pony with eosinophilic typhlitis-colitis. D—Severe large intestinal thickening (large intestinal wall thickness between arrows, 10.2 mm) in the right intercostal spaces zone in the same horse as in panel C. All images were obtained with a 2.5- to 5.0-MHz curvilinear transducer.
Citation: Journal of the American Veterinary Medical Association 253, 1; 10.2460/javma.253.1.108
Multiple enlarged lymph nodes (Figure 3) were visualized within the cecal mesentery in 9 of the 25 (36%) horses. This included 5 horses with confirmed or presumed neoplasia (3 with lymphoma, 1 with adenocarcinoma, and 1 with an 8-cm-diameter cecal mural-luminal mass and small left renal masses of unknown origin). Additional ultrasonographic abnormalities were evident in 7 horses, including the horse with renal masses. Four horses had hepatic abnormalities, including 3 horses with hepatomegaly and diffuse hepatic disease (one of which also had splenic nodules and another of which also had severe small intestinal thickening) and 1 horse with a focal hepatic mass (Figure 4). Three horses with hepatic abnormalities underwent ultrasound-guided biopsy, which yielded a diagnosis of lymphoma in the horse with splenic nodules, adenocarcinoma in the horse with a focal hepatic mass, and only nonspecific findings in one of the horses with hepatomegaly. Severe gastric thickening was identified in 2 horses (gastric wall thickness, 26 and 32 mm). Both underwent gastroscopy, which revealed moderate ulceration in one horse and minor ulceration adjacent to the margo plicatus in the other.
Ultrasonographic images obtained from a 23-year-old Arabian mare with hepatosplenic lymphoma, severe large intestinal thickening, and cecal lymphadenopathy. A—Enlarged lymph nodes (LN) adjacent to the lateral cecal vein (V) and artery (A) were detected within the cecal mesentery during examination from the right paralumbar fossa to the right ventrum zones. B—Severe edematous thickening of the wall of the large colon was detected (arrows). Luminal fluid allowed visualization of the far wall of the colon. C—One of multiple splenic nodules (arrow) that were visualized in the horse. Results of histologic examination of ultrasound-guided biopsy samples of this nodule and the right liver lobe were diagnostic for lymphoma.
Citation: Journal of the American Veterinary Medical Association 253, 1; 10.2460/javma.253.1.108
Ultrasonographic images obtained from a 28-year-old Quarter Horse mare initially evaluated because of chronic weight loss, decreased appetite, and intermittent diarrhea. A—Severe colonic wall thickening (colonic wall thickness indicated by the bracket, 46.4 mm) was present throughout the right and left ventrum zones and intercostal spaces zones. B—A focal hepatic mass (arrowheads) within the right liver lobe was visualized from the right 11th intercostal space. Histologic examination of an ultrasound-guided biopsy sample of this mass yielded a diagnosis of adenocarcinoma. Severe large intestinal thickening (bracket) was also evident. This horse died at home 1 week later.
Citation: Journal of the American Veterinary Medical Association 253, 1; 10.2460/javma.253.1.108
Excess peritoneal fluid was detected ultrasonographically in 12 horses and was subjectively graded as mild (n = 3), moderate (8), or severe (1). Abdominocentesis yielded peritoneal fluid in 20 of 21 horses. Eleven samples were interpreted as transudates, 2 as modified transudates, and 2 as exudates; the remaining 5 samples were not classified. Peritoneal fluid samples from 2 horses were submitted for bacterial growth, and neither yielded any growth. Fluid analysis revealed total protein concentrations ranging from 0.4 to 4.6 g/dL (mean ± SD, 1.7 ± 1.0; median, 1.4) and total nucleated cell counts ranging from 180 to 70,410 cells/dL (mean ± SD, 6,130 ± 16,000 cells/dL; median, 1,200 cells/dL). All but 2 samples had a total nucleated cell count < 7,500 cells/dL. Cytologic examination of 16 samples revealed that the neutrophil fraction ranged from 12% to 96% (mean ± SD, 56% ± 30%; median, 65%). No neoplastic cells, eosinophils, or intra- or extracellular bacteria were found in any sample.
Clinical course and outcome
Twenty-two of the 25 horses were hospitalized, with hospitalization time ranging from 2 to 26 days. Two horses were evaluated only as outpatients, including the horse in which adenocarcinoma was diagnosed by means of ultrasound-guided hepatic biopsy and a horse with gastric ulcers that was discharged with instructions for the owner to administer omeprazole (this horse did not have ultrasonographic evidence of gastric wall thickening). One horse was euthanized on the day of initial evaluation, and necropsy yielded a diagnosis of epitheliotropic T-cell lymphoma affecting the cecum, right ventral colon, and small intestine.
One horse was found to have a 360° colon torsion on exploratory celiotomy. This horse was evaluated 4 days after foaling because of mild signs of colic, and the decision to perform surgery was made on the basis of ultrasonographic findings of severe large intestinal thickening and a markedly distended colonic vessel in the right ninth intercostal space. One of the other hospitalized horses underwent exploratory celiotomy and was found to have pelvic flexure retroflexion 8 days after ultrasonographic examination. This horse had been examined because of diarrhea, and ultrasonographic findings consisted of cecal lymphadenopathy and severe large intestinal thickening in 5 of 6 abdominal zones. There was no clinical or ultrasonographic evidence of displacement until 7 days later, when the horse developed signs of colic while still hospitalized. Colonic biopsy at the time of surgery yielded a diagnosis of lymphoplasmacytic colitis that was determined to be a contributing factor for the pelvic flexure retroflexion.
Rectal biopsy was performed in 4 of 25 horses and was nondiagnostic in 2 horses, yielded a diagnosis of T-cell lymphoma in 1 horse, and yielded a diagnosis of eosinophilic-lymphoplasmacytic proctitis in 1 horse. Cecal biopsy was performed during exploratory laparotomy in the latter horse, and results confirmed eosinophilic typhlitis as the cause of cecal thickening. The horses in which rectal biopsy samples were negative for clinically important abnormalities included the horse confirmed to have hepatic adenocarcinoma and the horse with lymphoplasmacytic colitis.
In addition to the 10 horses with diarrhea or loose feces on initial evaluation, 4 horses developed diarrhea while hospitalized. Bacterial culture of fecal samples yielded growth of Salmonella spp in 2 of 19 horses, 1 with soft (ie, cow-pie consistency) feces at initial evaluation and 1 that developed diarrhea while hospitalized. Clostridium difficile was cultured in small numbers from feces of 3 of 11 horses, including 2 horses for which the organism was recovered only from enrichment broth. All 3 horses were negative for C difficile antigen and negative for C difficile toxin at the time of ultrasonographic examination; however, 1 horse had been positive for C difficile toxin and antigen at the time of surgery 1 month earlier. Two of these horses had had soft (ie, cow-pie consistency) feces at admission or while hospitalized. None had classic signs of infectious colitis.
Recheck ultrasonography was performed in 3 horses and revealed complete resolution of or substantial improvement in colonic thickening. This included a 5-day postoperative recheck examination of the horse with confirmed colon torsion (Figure 5), a 9-day recheck examination of a horse treated for colitis, and a 7-week recheck examination of the horse that later underwent surgery for pelvic flexure retroflexion.
Preoperative and 5-day postoperative ultrasonographic images obtained from a 13-year-old Warmblood gelding with colon torsion. A—Severe large intestinal thickening (large intestinal wall thickness, 22.9 mm) was seen in the right paralumbar fossa zone. Similar large intestinal thickening was identified in 5 of 6 abdominal zones. B—Recheck ultrasonography performed 5 days after surgery revealed normal intestinal wall thickness throughout the abdomen. Wall thickness in the right 15th intercostal space was 2.8 mm.
Citation: Journal of the American Veterinary Medical Association 253, 1; 10.2460/javma.253.1.108
Considering the 22 horses for which outcome information was available, the survival rate was 55% (12/22). Eleven horses were reported to have fully recovered (ie, complete resolution of clinical signs) on the basis of medical record review or communication with the owner or referring veterinarian between 4 months and 3 years (median, 450 days) after discharge. The horse with a surgical diagnosis of lymphoplasmacytic colitis was considered to have partially recovered because of ongoing chronic intermittent colic. Survivors included 2 horses for which results of fecal bacterial culture were positive for C difficile but results of a fecal immunoassay for C difficile toxin and antigen were negative, the horse with eosinophilic typhlitis-proctitis, and the broodmare that required surgery for a 360° colon torsion. The remaining 4 broodmares fully recovered without surgical intervention. Despite having only mild to moderate signs of colic at the time of initial evaluation, these 4 mares were suspected on the basis of signalment and recent parturition to have had colon torsion that resolved spontaneously. Also, 2 of these broodmares underwent surgery 1 and 4 years later for 360° colon torsion 5 weeks and 5 days after foaling, respectively. Only 1 surviving horse had an additional ultrasonographic finding (severe gastric wall thickening) that was considered clinically important. Subsequent gastroscopy revealed moderate ulceration for which this horse was treated with omeprazole. The 2 remaining horses that survived were a 23-year-old Appaloosa gelding discharged after 4 days of hospitalization and a 7-year-old German Riding Pony Stallion that had the second largest wall thickness measurement (34 mm) in the study. This horse was hospitalized for 24 days after developing inappetence and soft (ie, cow-pie consistency) feces following treatment with trimethoprim-sulfamethoxazole because of recent dental surgery. Repeated ultrasonographic examination 9 days after the initial ultrasonographic examination revealed marked improvement and only a focal persistent area of severe large intestinal thickening in the right cranioventral aspect of the abdomen. None of the surviving horses had clinicopathologic or ultrasonographic findings suggestive of neoplasia. The 2 surviving horses with histologic confirmation of lymphoplasmacytic colitis and eosinophilic typhlitis also had cecal lymphadenopathy.
Ten of 22 (45%) horses for which outcome information was available died or were euthanized within 2 weeks after initial evaluation or discharge, including both horses with Salmonella colitis, the horse that had previously undergone colon resection and was positive for C difficile on fecal bacterial culture and had previously been positive for C difficile antigen on fecal immunoassay, both horses with gastrointestinal lymphoma, and the horse with hepatic adenocarcinoma and presumed large colon involvement. The remaining 4 horses did not undergo necropsy, and the etiopathogenesis of severe large intestinal thickening was unknown. These horses consisted of 1 horse that died with signs of multiple organ dysfunction syndrome, 1 horse with hepatic enlargement and nonspecific biopsy findings, 1 horse with severe gastric wall thickening, and 1 horse with severe large intestinal thickening as the only notable ultrasonographic finding. Three horses were lost to follow-up, including the horse with a cecal mass, the horse with histologically confirmed hepatosplenic lymphoma, and a 26-year-old gelding with severely thickened colon visible in 5 of 6 abdominal zones.
Statistical analysis
Severe large intestinal thickening (ie, large intestinal wall thickness ≥ 9 mm) was significantly more likely to be detected in ventral abdominal zones than in upper zones (ie, paralumbar fossa and intercostal space zones combined; P = 0.001; OR, 3.3; 95% CI, 1.6 to 6.8). Severe large intestinal thickening was also significantly more common on the right side (46/75 right-sided zones) than on the left side (27/75 left-sided zones) of the abdomen (P = 0.003; OR, 2.8; 95% CI, 1.5 to 5.5). There were no significant differences between survivors and nonsurvivors for continuous (Table 1) and categorical (Table 2) variables, including sex, history of colic, fever or inappetence, whether abdominocentesis was performed, and the presence of large intestinal thickening ≥ 9 mm in any individual zone.
Median (IQR) values for continuous variables recorded for 22 horses with severe large intestinal thickening (ie, large intestinal wall thickness ≥ 9 mm) identified by means of abdominal ultrasonography that were alive at the time of follow-up (survivors; n = 12) or that died or were euthanized (nonsurvivors; 10).
| Variable | Survivors | Nonsurvivors | P value |
|---|---|---|---|
| Age (y) | 11.5 (10–15) | 18.5 (8–24) | 0.23 |
| Duration of clinical signs (d) | 1 (1–3.5) | 5 (2–14) | 0.06 |
| Time between initial examination and ultrasonography (d) | 1 (1–1) | 1 (1–6) | 0.40 |
| Peritoneal total protein (g/dL) | 1.3 (1.15–1.7) | 1.6 (1.1–2.4) | 0.60 |
| Peritoneal nucleated cell count (cells/μL) | 530 (233–4,100) | 1,030 (710–3,030) | 0.25 |
| Peritoneal neutrophil fraction (%) | 56 (15–78) | 61 (13–87) | 0.94 |
| Peritoneal lactate (mmol/L) | 2.4 (2.1–4.8) | 2.0 (1.6–2.7) | 0.37 |
| Maximal large intestinal wall thickness (mm) | |||
| Right paralumbar fossa | 17.7 (16.2–21.3) | 19 (13.1–22.2) | 0.88 |
| Right intercostal spaces | 15.7 (11.6–19.6) | 15.0 (13.1–27.0) | 0.91 |
| Right ventrum | 22.2 (17.7–27.8) | 18.0 (16.4–22.2) | 0.25 |
| Left paralumbar fossa | 10 (9.5–12.8) | NA | NA |
| Left intercostal spaces | 17.1 (12.8–18.1) | 23.6 (22.4–24.7) | 0.06 |
| Left ventrum | 20.3 (18–24.1) | 14.7 (13.2–24.8) | 0.40 |
| No. of abdominal zones with severe thickening | |||
| Right-sided zones | 1.5 (1–3) | 2 (1–3) | 0.63 |
| Left-sided zones | 1 (0.5–2) | 1 (1–1) | 0.57 |
| All zones | 2 (2–4.5) | 3 (2–4) | 0.76 |
| Total large intestinal wall thickness (mm) | |||
| Right-sided zones | 35.6 (20.3–59.4) | 32.5 (24.2–62.2) | 0.79 |
| Left-sided zones | 27.1 (19–36) | 14.7 (12.9–43.3) | 0.62 |
| All zones | 44.6 (15.9–77.9) | 46.2 (28.5–70.1) | 0.67 |
| Mean large intestinal wall thickness | |||
| Right-sided zones | 18.7 (12.7–22.8) | 16.3 (13.5–25.8) | 0.87 |
| Left-sided zones | 16.8 (13.9–20.9) | 14.7 (13.2–25.1) | 0.96 |
| All zones | 17.8 (12.2–19.5) | 15.4 (12.9–24.2) | 0.90 |
Total thickness measurements represented the sum of maximal wall thickness measurements from each included zone. P values were calculated by use of the Wilcoxon rank sum test.
IQR = Interquartile (25th to 75th percentile) range. NA = Not applicable.
Values for categorical variables recorded for the horses in Table 1.
| Variable | Survivors | Nonsurvivors | P value | ||
|---|---|---|---|---|---|
| Sex | 0.22 | ||||
| Mare | 6 | 6 | |||
| Gelding | 3 | 4 | |||
| Stallion | 3 | 0 | |||
| Clinical signs at initial examination | |||||
| Colic | 6 | 3 | 0.22 | ||
| Fever | 0 | 2 | 0.19 | ||
| Inappetence | 3 | 5 | 0.17 | ||
| Weight loss | 2 | 5 | 0.10 | ||
| Signs of depression | 0 | 3 | 0.08 | ||
| Diarrhea | 3 | 6 | 0.09 | ||
| Abdominocentesis performed | 9 | 8 | 0.38 | ||
| Cecal lymphadenopathy | 2 | 5 | 0.10 | ||
| Severe large intestinal thickening identified | |||||
| Right paralumbar fossa | 5 | 3 | 0.30 | ||
| Right intercostal spaces | 7 | 8 | 0.21 | ||
| Right ventrum | 8 | 8 | 0.30 | ||
| Left paralumbar fossa | 3 | 0 | 0.14 | ||
| Left intercostal spaces | 4 | 2 | 0.30 | ||
| Left ventrum | 6 | 8 | 0.13 | ||
Data represent number of horses. P values were calculated by use of the Fisher exact test.
Discussion
Results of the present study suggested that severe large intestinal thickening (ie, large intestinal wall thickness ≥ 9 mm) can be present in horses with multiple disease processes. A previous study13 reported that such a finding could be used as an accurate and reproducible indicator for colon torsion in horses with colic determined to require surgical intervention. At our hospital, ultrasonographic evaluation of horses with surgical colic is typically performed at the point of care by attending surgeons or surgery residents. These cases were not included in the present study because of variations in the ultrasonographic imaging technique, inconsistent documentation of ultrasonographic findings for the purposes of retrospective analysis, and the large number of personnel involved in ultrasonographic examinations during the study period. Whereas patients with colon torsion and other causes of surgical colic were underrepresented in this study, 1 of the 25 horses that were included was found to have a 360° large colon torsion despite having only mild colic signs. Additionally, the remaining 4 broodmares were initially referred for evaluation of suspected colon torsion on the basis of signalment, recent parturition, and signs of pain prior to referral, but these horses had only mild to moderate colic signs. All 4 mares were discharged 3 to 5 days later after receiving basic supportive care. Although a definitive diagnosis was not made in these horses, spontaneous resolution of large colon torsion or displacement remained a possibility. It was notable that 2 of these broodmares underwent surgery for treatment of colon torsion in subsequent breeding seasons.
It is well-known that the distinction between medical and surgical colic is not always straightforward. In the study by Pease et al,13 the finding that 12 horses with colon torsion were amenable to ultrasonographic examination and that those examinations required 15 minutes each suggests a lesser degree of pain in that study population than is typically associated with colon torsion. Although the extent of pain was not reported in that study, horses were excluded if intractable pain precluded ultrasonographic examination.13 It is important to note that 52% (13/25) of the horses in the present study had signs of colic on initial evaluation. Whereas many had only mild colic signs during the initial examination, several had a history of more severe pain prior to referral. Evidence from the present and a previous13 study seem to suggest that clinical decision-making for and management of patients with large intestinal wall thickness ≥ 9 mm should include careful consideration of the patient's clinical appearance in conjunction with ultrasonographic findings, especially for horses that do not have classic surgical colic signs.
The value of a thorough and systematic ultrasonographic examination of the abdomen should also be emphasized, as additional ultrasonographic findings helped achieve a definitive diagnosis in several patients of the present study. Identification of abdominal structures has been found to be superior in horses in which the hair has been clipped, compared with results of focused ultrasonographic techniques performed without clipping.22 However, detailed examinations may not always be possible or practical in certain situations. The recently described FLASH protocol (fast localized abdominal sonography of horses) is a focused technique that has gained traction among equine veterinarians when evaluating patients with colic.23 Although not as extensive as the systematic examination performed in the present study, the protocol described23 should help emergency clinicians assess most abdominal structures in horses with colic, thus assisting with the clinical decision-making process in an expeditious manner.
Three horses included in the present study were evaluated ultrasonographically 6, 13, and 26 days after colon-related surgeries. It is arguable that colon wall thickness in these patients may have been affected by surgical manipulation, previous strangulation, colon resection, or some combination of these factors. One of these horses underwent surgical cecal biopsy 12 days prior to ultrasonographic examination. We suggest it was unlikely that surgical manipulation alone could have accounted for the severe diffuse changes observed in this horse. This is supported by the results of a previous study24 in which the authors reported no change in intestinal wall thickness in healthy ponies following celiotomy. Two other horses in the present study underwent exploratory celiotomy for treatment of large colon torsion 6 and 26 days prior to ultrasonographic examination. Involution of the colonic wall after surgery for colon torsion without resection has been found to be significantly prolonged in horses that subsequently develop multiple organ dysfunction syndrome.25 This may explain the persistent severe thickening evident 6 days after surgery in the horse that later died secondary to multiple organ dysfunction syndrome. In the horse that underwent ultrasonographic examination 26 days after surgery, it would seem unlikely that thickening persisted solely as the result of surgery.
Of the 22 horses in the present study for which outcome information was available, 11 (50%) recovered fully, 1 (5%) recovered partially, and 10 (45%) died or were euthanized because of problems related to the large intestine. Of the 3 horses lost to follow-up, 2 were suspected to have neoplastic infiltration of the colon on the basis of histologic diagnosis of hepatosplenic lymphoma in one and ultrasonographic evidence of cecal and renal masses in the other. It was unknown whether the severe large intestinal thickening detected in these horses was a primary or secondary pathological process, as neither underwent necropsy. Taking these cases into consideration, the prognosis for horses with severe large intestinal thickening in the study population was considered guarded to poor. It was interesting that the extent of wall thickening did not differ between survivors and nonsurvivors. In fact, 6 of 12 survivors and 4 of 10 nonsurvivors had maximal wall-thickness measurements > 20 mm. The number of zones affected was also not significantly different between groups. Although definitive diagnoses were not made in all horses that recovered, clinical diagnoses included inflammatory colitis and spontaneous resolution of suspected colon displacement or torsion.
In the present study, large intestinal thickening was most often found in the ventral aspect of the abdomen (20/25 [80%] horses). We suggest that this can be explained by the anatomic location of a large portion of the large intestine along the ventral abdominal wall. However, it was also likely affected by the diffuse nature of the disease processes. Whereas the left side of the abdomen was less frequently affected, compared with the right side for patients of this study, 19 of 25 (76%) horses had severe large intestinal thickening in at least 1 zone of the 3 left-sided zones.
Severe colonic thickening was identified in the region of the right dorsal colon (right intercostal spaces) in 18 of 25 (72%) horses in the present study; however, no horse had abnormal colon located exclusively in this region. This underscored the importance of evaluating beyond the right intercostal spaces in horses with clinical signs suggestive of right dorsal colitis, a diagnosis with clinical signs similar to those reported for many horses in the present study. In our clinical experience, it is not uncommon for veterinarians to request evaluation of the right intercostal spaces to rule out right dorsal colitis in horses with suggestive clinical signs. We would note that 7 of 25 (28%) horses had no large intestinal thickening in the right intercostal spaces. If ultrasonographic examination had been limited to the region of the right dorsal colon in these horses, a delay in diagnosis or underdiagnosis could have compromised treatment and outcome.
In the present study, ultrasonographic evidence of severely thickened large intestine in the right paralumbar fossa region was interpreted to represent the cecum. This was further confirmed by visualization of the lateral cecal artery and vein adjacent to the body wall. Evaluation of the cecal mesentery also allowed identification of cecal lymphadenopathy in 9 of 25 horses.26 It is possible that laparoscopic biopsy of enlarged cecal lymph nodes for histologic examination may have yielded a definitive diagnosis in some horses, but this was not performed in any patients of the present study. Laparoscopic biopsy may have provided a minimally invasive option for patients in which rectal biopsy or biopsy of abnormal abdominal organs failed to reveal the cause of large intestinal thickening.
Limitations of the present study included the lack of a definitive diagnosis in some patients. As a retrospective study of horses with severe disease, the intensive nature of treatment and owner financial constraints limited diagnostic testing in some horses. Although a documented cause for severe large intestinal thickening was not identified in 7 of 12 survivors, inclusion of these cases was considered important to represent those horses that were apparently able to recover despite severe and extensive large intestinal thickening.
In summary, severe thickening of the large intestine can occur in horses with multiple disease processes and similar clinical signs. Ultrasonographic evaluation of the entire abdomen was important to document the severity and extent of large intestinal thickening, which infrequently involved only the right dorsal colon. Also, ultrasonographic examination allowed the identification of additional abnormalities that helped establish a definitive diagnosis in many cases, assisting clinicians in determining a prognosis and guiding treatment planning. Consideration of all clinical and diagnostic imaging findings is important to determine the best clinical course of action for horses with severe large intestinal thickening.
Acknowledgments
Presented in abstract form at the 57th Annual Convention of the American Association of Equine Practitioners, San Antonio, Tex, November 2011.
ABBREVIATIONS
| CI | Confidence interval |
Footnotes
Vingmed System 5, GE Healthcare, Wauwatosa, Wis.
Biosound Technos, Biosound Esaote Inc, Indianapolis, Ind.
Dormosedan, Orion Corp, Espoo, Finland.
SAS, version 9.3, SAS Institute Inc, Cary, NC.
References
1. Reef VB. Equine diagnostic ultrasound. Philadelphia: WB Saunders Co, 1998;273–363.
2. Klohnen A, Vachon AM, Fischer AT Jr. Use of diagnostic ultrasonography in horses with signs of acute abdominal pain. J Am Vet Med Assoc 1996;209:1597–1601.
3. Vaughan B, Whitcomb MB, Pratt SM, et al. Ultrasonographic appearance of abdominal organs in 14 horses with systemic Cor y nebacter ium pseudotuberc ul osis infection, in Proceedings. 50th Ann Conv Am Assoc Equine Pract 2004;50:63–69
4. Taylor SD, Pusterla N, Vaughan B, et al. Intestinal neoplasia in horses (1990–2005). J Vet Intern Med 2006;20:1429–1436.
5. Freeman SL. Diagnostic ultrasonography of the mature equine abdomen. Equine Vet Educ 2003;15:319–330.
6. Fontaine GL, Rodgerson DH, Hanson RR, et al. Ultrasound evaluation of equine gastrointestinal disorders. Compend Contin Educ Pract Vet 1999;21:253–262.
7. Freeman SL. Ultrasonography of the equine abdomen: findings in the colic patient. In Pract 2002;24:262–273.
8. Hendrickson EHS, Malone ED, Sage AM. Identification of normal parameters for ultrasonographic examination of the equine large colon and cecum. Can Vet J 2007;48:289–291.
9. Bithell S, Habershon-Butcher JL, Bowen M, et al. Repeatability and reproducibility of transabdominal ultrasonographic intestinal wall thickness measurements in Thoroughbred horses. Vet Radiol Ultrasound 2010;51:647–651.
10. Epstein K, Short D, Parente E, et al. Gastrointestinal ultrasonography in normal adult ponies. Vet Radiol Ultrasound 2008;49:282–286.
11. Santschi EM, Slone DE Jr, Frank WM II. Use of ultrasound in horses for diagnosis of left dorsal displacement of the large colon and monitoring its nonsurgical correction. Vet Surg 1993;22:281–284.
12. Martin BB Jr, Freeman DE, Ross MW, et al. Cecocolic and cecocecal intussusception in horses: 30 cases (1976–1996). J Am Vet Med Assoc 1999;214:80–84.
13. Pease AP, Scrivani PV, Erb HN, et al. Accuracy of increased large-intestine wall thickness during ultrasonography for diagnosing large-colon torsion in 42 horses. Vet Radiol Ultrasound 2004;45:220–224.
14. Whitcomb MB. Advanced abdominal ultrasound for chronic colic, in Proceedings. Am Coll Vet Surg Symp 2005;39–43.
15. Grenager NS, Durham MG. Ultrasonographic evidence of colonic mesenteric vessels as an indicator of right dorsal displacement of the large colon in 13 horses. Equine Vet J Suppl 2011;39:153–155.
16. Ness SL, Bain FT, Zantingh AJ, et al. Ultrasonographic visualization of colonic mesenteric vasculature as an indicator of large colon right dorsal displacement or 180 degrees volvulus (or both) in horses. Can Vet J 2012;53:378–382.
17. Cohen ND, Mealey RH, Chaffin MK, et al. The recognition and medical management of right dorsal colitis in horses. Vet Med 1995;90:687–692.
18. Cohen ND. Right dorsal colitis. Equine Vet Educ 2002;14:212–219.
19. Jones SL, Davis J, Rowlingson K. Ultrasonographic findings in horses with right dorsal colitis: five cases (2000–2001). J Am Vet Med Assoc 2003;222:1248–1251.
20. Barr BS. Infiltrative intestinal disease. Vet Clin North Am Equine Pract 2006;22:e1–e7.
21. Sheats MK, van Wetter AJNJ, Snyder LA, et al. Disseminated large granular lymphoma in a horse. Equine Vet Educ 2008;20:459–463.
22. Williams S, Cooper J, Freeman S. Evaluation of normal findings using a detailed and focused technique for transcutaneous abdominal ultrasonography in the horse. BMC Vet Res 2014;10(suppl 1):S5.
23. Busoni V, De Busscher V, Lopez D, et al. Evaluation of a protocol for fast localised abdominal sonography of horses (FLASH) admitted for colic. Vet J 2011;188:77–82.
24. Epstein K, Short D, Parente E, et al. Serial gastrointestinal ultrasonography following exploratory celiotomy in normal adult ponies. Vet Radiol Ultrasound 2008;49:584–588.
25. Sheats MK, Cook VL, Jones SL, et al. Use of ultrasound to evaluate outcome following colic surgery for equine large colon volvulus. Equine Vet J 2010;42:47–52.
26. Vaughan B, Whitcomb MB, Pusterla N. Ultrasonographic findings in 42 horses with cecal lymphadenopathy, in Proceedings. 59th Ann Conv Am Assoc Equine Pract 2013;59:238.
