Gastrointestinal obstruction is one of the primary differential diagnoses for vomiting dogs.1,2 Plain radiography is a readily available screening test for the presence of mechanical obstruction. Radiographic features suggestive of mechanical obstruction include intestinal dilation, a visible foreign body, a so-called gravel sign, intestinal stacking, and 2 distinct groupings of bowel that differ in diameter.1
Radiographic determination of intestinal dilation can be based on objective measurement of intestinal diameter. This most commonly involves calculation of the ratio of the intestinal diameter to the height of the midbody of L5 (SI:L5 ratio).1,3,4 Published ratios consistent with mechanical obstruction vary from 1.6 to 2.4, suggesting some degree of inaccuracy associated with radiographic identification of mechanical obstruction.1–4 Indeed, some authors have suggested that use of the SI:L5 ratio does not significantly increase diagnostic accuracy beyond that achieved by subjective assessment alone.3 To better identify 2 distinct groupings of bowel, calculation of the ratio of smallest to largest intestinal diameter measurements has been suggested as a more accurate test for mechanical obstruction.4
The limited accuracy of radiography in identification of mechanical obstruction coupled with the greater availability of ultrasonography to veterinary practitioners has resulted in increased use of abdominal ultrasonography for the identification of gastrointestinal obstruction.2 Early on, ultrasonography was recognized as an important adjunct to radiography for definitive diagnosis of intestinal foreign bodies.5 In children, abdominal ultrasonography is the imaging modality of choice for patients with suspected intussusception and is considered highly sensitive, specific, and cost-effective.6 Whereas abdominal ultrasonography has some limitations, including variability in operator experience, interference of intestinal gas, and availability, it also has a reported diagnostic accuracy of 97% in the diagnosis of intestinal obstruction.2 However, other research7 has revealed some limitations of ultrasonography versus exploratory laparotomy for correct assessment of the gastrointestinal tract.
The usefulness of CT for the identification of gastrointestinal obstruction in children has been extensively described, and CT is considered the primary imaging modality for children for which the leading differential diagnoses do not include intussusception or appendicitis.6 In veterinary patients, the usefulness of CT for the evaluation of dogs with signs of acute abdominal pain has also been reported, and with the increased availability of MDCT, examinations can be rapidly performed with patients sedated.8,9 To the authors' knowledge, specific investigation of the feasibility and effectiveness of CT in the diagnosis of mechanical obstruction in dogs has not been reported.
The purpose of the study reported here was to compare noncontrast abdominal CT with abdominal ultrasonography for the identification and localization of mechanical gastrointestinal obstruction in dogs, with exploratory laparotomy as the reference standard. A secondary goal was to report and compare intestinal diameter ratios between dogs with and without obstruction. We hypothesized that abdominal CT could be performed more quickly and would be more accurate in the diagnosis of mechanical gastrointestinal obstruction than abdominal ultrasonography. We further hypothesized that dogs with mechanical gastrointestinal obstruction would have larger intestinal diameter ratios than dogs without such obstruction.
Materials and Methods
Animals
Client-owned dogs examined for suspected obstructive gastrointestinal disease at the Small Animal Hospital of the University of Florida were prospectively enrolled in the study with owner consent between September 15, 2013, and December 15, 2014. Prior to enrollment, a clinical history was collected, and physical examination, CBC, serum biochemical analysis, and abdominal radiography were performed. Dogs that were suspected of having a complete or partial gastrointestinal obstruction without clinically important comorbidities (eg, hypovolemic shock, sepsis, or aspiration pneumonia) were considered eligible for inclusion. Dogs were deemed ineligible when they had a clinically important comorbidity, history of previous gastrointestinal surgery or septic peritonitis, known intestinal adhesions, or radiographic evidence of intestinal plication.
All dog owners were given the option for standard treatment independent of the study and were allowed to withdraw their dogs from the study at any point. The study protocol was approved by the University of Florida Institutional Animal Care and Use Committee (protocol No. 201307975).
Sedation and anesthesia
Dogs were sedated or anesthetized prior to ultrasonography and CT. Because of the stable nature of the included dogs, sedation was preferred in most instances. Choice of pharmacological agents and doses administered was at the discretion of the attending anesthesiologist and surgeon. In general, premedications or sedatives included opioids such as methadone and α2-adrenoceptor agonists such as dexmedetomidine hydrochloride. Dexmedetomidine was typically administered at a dose of 5 to 10 μg/kg (2.3 to 4.5 μg/lb), I V. In some dogs, butorphanol tartrate (0.2 mg/kg [0.1 mg/lb], IV) was combined with dexmedetomidine for enhanced sedation with analgesia. For dogs imaged while anesthetized, anesthetic induction was performed with propofol, and anesthesia was maintained with isoflurane in 100% oxygen.
Abdominal ultrasonography
Abdominal ultrasonography was performed by a board-certified radiologist or radiology resident under the supervision of a board-certified radiologist. Dogs were evaluated in dorsal recumbency by use of a microconvex 5- to 8-MHz or 4- to 9-MHz broadband transducer.a,b When applicable, a linear 5- to 17.5-MHz or 8- to 18.5-MHz broadband transducer was also used. The time required to complete each ultrasonographic examination was calculated on the basis of recorded acquisition times of the first and last images. Images were reviewed in real time, and findings were discussed with the attending surgeons.
Abdominal CT
An MDCT unitc–e was used to perform all abdominal CT examinations. Dogs were positioned in dorsal recumbency, and routine abdominal-volume acquisition protocols were used. Transverse images were reconstructed with a 2- to 3-mm slice thickness by use of bone and soft tissue algorithms. Data volumes were reformatted in sagittal and dorsal planes with a 2- to 3-mm slice thickness by use of a soft tissue algorithm. The time required for each examination was determined on the basis of the acquisition times listed on the first scout image and the last transverse reconstruction.
Image analysis
Image analysis and evaluation was performed by a board-certified radiologist. For each dog, images were assessed for the presence and location of obstruction and for diagnosis (eg, foreign body, adhesions, or intussusception). The ultrasonographic diagnosis of obstruction was made on the basis of subjective assessment of intestinal dilation, recognition of 2 distinct groupings of bowel differing in diameter, or identification of intestinal plication, as previously described.2,8 The presence of a foreign body alone was considered a positive test result, indicating complete mechanical obstruction only when intestinal dilation or 2 distinct bowel groupings were also present. For CT, similar criteria were used to subjectively assess the presence of intestinal dilation and 2 distinct groupings of bowel differing in diameter, with or without the identification of a foreign body suggestive of mechanical obstruction.
Bowel diameter and ratios were measured at the time of data analysis and compared between dogs with and without obstruction. For each dog, a board-certified radiologist (MDW) identified and measured the maximal and minimal small intestinal diameters, from serosa to serosa, in dorsal and transverse planes. Each measurement was made 3 times, and the mean of the largest and smallest of the measurements was used to calculate intestinal diameter ratios. Intestinal diameter ratios were calculated in dorsal and transverse CT planes, then compared between dogs with and without obstruction.
Surgery
All dogs underwent laparoscopy, exploratory laparotomy, and surgical treatment of their primary gastrointestinal disease as part of a parallel investigation.10 Dogs underwent surgery if the constellation of findings of physical examination, laboratory tests, and abdominal CT and ultrasonography was consistent with mechanical obstruction or, as for 2 dogs, if surgery was indicated for diagnosis and treatment. A complete obstruction was deemed to exist when there was an obvious, focal transition from distended to nondistended intestine, palpation of a foreign body or mass that obliterated most of the intestinal lumen, or both. In contrast, a partial obstruction was deemed to exist when there was an intraluminal foreign body or extramural lesion causing incomplete attenuation of the intestinal lumen.
Statistical analysis
Values of continuous variables are reported as median (IQR). Image acquisition times for abdominal CT and ultrasonography were compared with the Wilcoxon signed rank test. No adjustment was made for multiple comparisons. Intestinal diameter ratios measured via CT were compared between dogs with and without mechanical gastrointestinal obstruction by use of the Wilcoxon rank sum test. Sensitivity, specificity, PPV, and NPV of abdominal CT and ultrasonography for the diagnosis of gastrointestinal obstruction were calculated, with findings of exploratory surgery considered the reference standard. Statistical softwaref was used for all analyses; values of P < 0.05 were considered significant.
Results
Animals
Sixteen dogs with suspected mechanical gastrointestinal obstruction were enrolled in the study. Signalment and clinical characteristics of the dogs are summarized elsewhere.10 For 14 dogs, abdominal ultrasonography and CT were performed with the dog sedated, whereas for 2 dogs, general anesthesia was used for the CT examination. General anesthesia was necessary in a brachycephalic dog to protect an airway, whereas it was necessary in another dog with a mesenteric mass to provide better support during a planned injection of contrast medium.
Diagnosis
Median time required for acquisition of CT images was 2.5 minutes (IQR, 2.0 to 3.8 minutes) and for ultrasonographic images was 26.0 minutes (IQR, 22.0 to 35.8 minutes; P < 0.001). The presence and location of mechanical gastrointestinal obstruction and the causative diagnosis based on imaging and surgical findings were summarized (Table 1). Results of CT and exploratory surgery (laparoscopy and exploratory laparotomy) indicated that 10 dogs had a complete obstruction, 3 had a partial obstruction, and 3 had no obstruction. The jejunum was the most commonly affected segment of intestine in obstructed dogs (11/12 dogs).
Findings of exploratory surgery (laparoscopy and exploratory laparotomy), ultrasonography, and CT for each of 16 dogs with suspected mechanical gastrointestinal obstruction.
| Obstruction status | Obstruction location | Diagnosis | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Dog No. | Surgery | Ultrasonography | CT | Surgery | Ultrasonography | CT | Surgery | Ultrasonography | CT |
| 1 | Nonobstructive | Nonobstructive | Nonobstructive | Diffuse SI and LI | Diffuse SI and LI | Diffuse SI and LI | Adhesions | Mesenteric mass* | Adhesions |
| 2 | Partially obstructed | Partially obstructed | Partially obstructed | Jejunum | Jejunum | Jejunum | Foreign body | Foreign body | Foreign body |
| 3 | Partially obstructed | Partially obstructed | Partially obstructed | Gastric, ileum | Gastric* | Gastric, ileum | Foreign body | Foreign body | Foreign body |
| 4 | Obstructed | Obstructed | Obstructed | Jejunocolic | Jejunocolic | Jejunocolic | Intussusception | Intussusception | Intussusception |
| 5 | Nonobstructive | Nonobstructive | Nonobstructive | Liver | Liver | Liver | Hepatopathy | Hepatopathy | Hepatopathy |
| 6 | Obstructed | Obstructed | Obstructed | Jejunum | Jejunum | Jejunum | Adhesions | Cause not seen* | Adhesions |
| 7 | Obstructed | Obstructed | Obstructed | Jejunum | Jejunum | Jejunum | Foreign body | Foreign body | Foreign body |
| 8 | Partially obstructed | Partially obstructed | Partially obstructed | Pylorus to colon | Colon, pylorus* | Colon, ileum, pylorus* | Linear foreign body | Foreign body* | Foreign body* |
| 9 | Obstructed | Obstructed | Obstructed | Jejunum | Jejunum | Jejunum | Foreign body | Foreign body | Foreign body |
| 10 | Obstructed | Obstructed | Obstructed | Jejunum, gastric | Jejunum, gastric | Jejunum, gastric | Foreign body | Foreign body | Foreign body |
| 11 | Obstructed | Obstructed | Obstructed | Duodenum, jejunum | Duodenum, jejunum | Duodenum, jejunum | Foreign body | Foreign body | Foreign body |
| 12 | Obstructed | Obstructed | Obstructed | Jejunum | Jejunum | Jejunum | Foreign body | Foreign body | Foreign body |
| 13 | Nonobstructive | Partially obstructed* | Nonobstructive | Diffuse SI | Diffuse SI | Diffuse SI | Functional ileus | Mechanical obstruction* | Functional ileus |
| 14 | Obstructed | Obstructed | Obstructed | Jejunum | Jejunum | Jejunum | Foreign body | Foreign body | Foreign body |
| 15 | Obstructed | Obstructed | Obstructed | Gastrojejunal | Gastrojejunal | Gastrojejunal | Linear foreign body | Linear foreign body | Linear foreign body |
| 16 | Obstructed | Obstructed | Obstructed | Gastrojejunal | Gastrojejunal | Gastrojejunal | Linear foreign body | Linear foreign body | Linear foreign body |
Finding differs from that of exploratory surgery (reference standard).
LI = Large intestine. SI = Small intestine.
Sensitivity and specificity of abdominal ultrasonography for the diagnosis of mechanical gastrointestinal obstruction (with findings of exploratory surgery as the reference standard) were 100% and 67%, respectively. The PPV for ultrasonography was 93%, and the NPV was 100%. For CT, sensitivity and specificity for the diagnosis of mechanical gastrointestinal obstruction were both 100%, and the PPV and NPV were also both 100%.
Disagreement among diagnostic modalities was detected for 1 dog (dog 13) only (Figure 1). In this dog, a false-positive diagnosis of obstruction was made on ultrasonography. Ultrasonographic findings in that dog included generalized small intestinal dilation with a suspected region of mural narrowing in the distal portion of either the jejunum or ileum. However, CT and surgical findings did not support this finding. Biopsy specimens were obtained during surgery, and the ultimate diagnosis was mild gastrointestinal inflammatory enteropathy.
Dorsal (A) and sagittal (B) CT images and representative ultrasonographic image (C) of the abdomen of an 11-year-old spayed female English Bulldog (dog 13) with suspected mechanical gastrointestinal obstruction. Possible obstruction of the distal portion of the jejunum or ileum was diagnosed via ultrasonography, but laparoscopy and exploratory laparotomy revealed no obstructive lesion. The lumen of the small intestinal segments (L) was uniformly and diffusely dilated but tapered abruptly at the level of the ileum. Functional ileus was considered as a differential diagnosis, but mechanical obstruction remained the primary differential diagnosis, and exploratory surgery was performed. At surgery, no obstructive lesion was identified, and biopsy specimens were obtained. The ultimate diagnosis was gastrointestinal inflammatory enteropathy. ST = Stomach.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Dorsal (A) and sagittal (B) CT images and representative ultrasonographic image (C) of the abdomen of an 11-year-old spayed female English Bulldog (dog 13) with suspected mechanical gastrointestinal obstruction. Possible obstruction of the distal portion of the jejunum or ileum was diagnosed via ultrasonography, but laparoscopy and exploratory laparotomy revealed no obstructive lesion. The lumen of the small intestinal segments (L) was uniformly and diffusely dilated but tapered abruptly at the level of the ileum. Functional ileus was considered as a differential diagnosis, but mechanical obstruction remained the primary differential diagnosis, and exploratory surgery was performed. At surgery, no obstructive lesion was identified, and biopsy specimens were obtained. The ultimate diagnosis was gastrointestinal inflammatory enteropathy. ST = Stomach.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Dorsal (A) and sagittal (B) CT images and representative ultrasonographic image (C) of the abdomen of an 11-year-old spayed female English Bulldog (dog 13) with suspected mechanical gastrointestinal obstruction. Possible obstruction of the distal portion of the jejunum or ileum was diagnosed via ultrasonography, but laparoscopy and exploratory laparotomy revealed no obstructive lesion. The lumen of the small intestinal segments (L) was uniformly and diffusely dilated but tapered abruptly at the level of the ileum. Functional ileus was considered as a differential diagnosis, but mechanical obstruction remained the primary differential diagnosis, and exploratory surgery was performed. At surgery, no obstructive lesion was identified, and biopsy specimens were obtained. The ultimate diagnosis was gastrointestinal inflammatory enteropathy. ST = Stomach.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
In 1 other dog (dog 6), ultrasonography allowed the correct identification of small intestinal obstruction, but the cause of the obstruction was unknown, and a foreign body was considered a differential diagnosis. In this same dog, CT clearly revealed a focal area of transition between distended and collapsed intestine, absence of an intraluminal foreign body, and disruption of the properitoneal fat line coupled with close association of the small intestine (Figure 2). In another dog (dog 1), obstruction was identified, and a mural lesion was characterized on both CT and ultrasonography (Figure 3). However, exploratory surgery revealed that the obstructive component was the result of sheet-like connections of nonvascular fibrous tissue between jejunal segments.
Sagittal (A) and transverse (B) CT images of the abdomen of a 7-year-old sexually intact male Chihuahua (dog 6) with interruption of the properitoneal fat line that is suggestive of enteroparietal adhesions in humans.11 Note the multifocal areas in which the intestinal segments are adjacent to the body wall but between the dilated, gas-filled segment and the ventral body wall, soft tissue–attenuating material (arrowheads) is visible, representing adhesions later identified via laparoscopy. This dog had a history of prior laparotomy.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Sagittal (A) and transverse (B) CT images of the abdomen of a 7-year-old sexually intact male Chihuahua (dog 6) with interruption of the properitoneal fat line that is suggestive of enteroparietal adhesions in humans.11 Note the multifocal areas in which the intestinal segments are adjacent to the body wall but between the dilated, gas-filled segment and the ventral body wall, soft tissue–attenuating material (arrowheads) is visible, representing adhesions later identified via laparoscopy. This dog had a history of prior laparotomy.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Sagittal (A) and transverse (B) CT images of the abdomen of a 7-year-old sexually intact male Chihuahua (dog 6) with interruption of the properitoneal fat line that is suggestive of enteroparietal adhesions in humans.11 Note the multifocal areas in which the intestinal segments are adjacent to the body wall but between the dilated, gas-filled segment and the ventral body wall, soft tissue–attenuating material (arrowheads) is visible, representing adhesions later identified via laparoscopy. This dog had a history of prior laparotomy.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Dorsal CT (A) and ultrasonographic (B) images of the abdomen of a 2-year-old neutered male Australian Shepherd (dog 1) with pythium that resulted in mural thickening and loss of wall layering (asterisks) on the ultrasonographic image and layered intestinal segments connected by soft tissue–attenuating material (arrowheads) on CT images.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Dorsal CT (A) and ultrasonographic (B) images of the abdomen of a 2-year-old neutered male Australian Shepherd (dog 1) with pythium that resulted in mural thickening and loss of wall layering (asterisks) on the ultrasonographic image and layered intestinal segments connected by soft tissue–attenuating material (arrowheads) on CT images.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Dorsal CT (A) and ultrasonographic (B) images of the abdomen of a 2-year-old neutered male Australian Shepherd (dog 1) with pythium that resulted in mural thickening and loss of wall layering (asterisks) on the ultrasonographic image and layered intestinal segments connected by soft tissue–attenuating material (arrowheads) on CT images.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Intestinal diameter ratios
In dogs with mechanical gastrointestinal obstruction (n = 13), the median ratio of largest to smallest intestinal segments as measured on CT scans was 3.9 (IQR, 2.4 to 4.3) in the dorsal plane and 4.0 (IQR, 2.5 to 4.3) in the transverse plane. For dogs without mechanical gastrointestinal obstruction (n = 3), median ratios were 2.0 (IQR, 1.5 to 2.4) and 1.6 (IQR, 1.3 to 2.1), respectively. Dogs with gastrointestinal obstruction had significantly larger intestinal diameters on CT scans than did dogs without obstruction in both the dorsal (P = 0.04) and transverse (P = 0.01) planes. The intestinal ratios for 4 dogs with gastrointestinal obstruction were within the IQR of ratios for dogs without obstruction. However, because of the presence of foreign material within the lumen of the stomach or jejunum (Figure 4) or findings indicative of a linear foreign body (Figure 5), a diagnosis of mechanical obstruction was still made.
Sagittal (A) and dorsal (B) CT images of the abdomen of a 6-year-old neutered male Golden Retriever (dog 3) with minimal to no distention but with mineral foreign material in the stomach (ST), ileum (arrows), and cecum (arrowheads). This dog was not believed to have an obstruction, but the presence of gastrointestinal foreign material coupled with vomiting and inappetence resulted in the decision to pursue surgical intervention.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Sagittal (A) and dorsal (B) CT images of the abdomen of a 6-year-old neutered male Golden Retriever (dog 3) with minimal to no distention but with mineral foreign material in the stomach (ST), ileum (arrows), and cecum (arrowheads). This dog was not believed to have an obstruction, but the presence of gastrointestinal foreign material coupled with vomiting and inappetence resulted in the decision to pursue surgical intervention.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Sagittal (A) and dorsal (B) CT images of the abdomen of a 6-year-old neutered male Golden Retriever (dog 3) with minimal to no distention but with mineral foreign material in the stomach (ST), ileum (arrows), and cecum (arrowheads). This dog was not believed to have an obstruction, but the presence of gastrointestinal foreign material coupled with vomiting and inappetence resulted in the decision to pursue surgical intervention.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Sagittal (A) and dorsal (B) CT and ultrasonographic (C) images of the abdomen of a 3-year-old neutered male Springer Spaniel (dog 16) with a linear foreign body. Plication is evident (arrowheads) via both modalities, with a linear foreign body directly visible on the ultrasonographic image (arrows). There is no evidence of distention, and heterogeneous soft tissue–and fluid-attenuating foreign material is present in several intestinal segments on the CT images.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Sagittal (A) and dorsal (B) CT and ultrasonographic (C) images of the abdomen of a 3-year-old neutered male Springer Spaniel (dog 16) with a linear foreign body. Plication is evident (arrowheads) via both modalities, with a linear foreign body directly visible on the ultrasonographic image (arrows). There is no evidence of distention, and heterogeneous soft tissue–and fluid-attenuating foreign material is present in several intestinal segments on the CT images.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Sagittal (A) and dorsal (B) CT and ultrasonographic (C) images of the abdomen of a 3-year-old neutered male Springer Spaniel (dog 16) with a linear foreign body. Plication is evident (arrowheads) via both modalities, with a linear foreign body directly visible on the ultrasonographic image (arrows). There is no evidence of distention, and heterogeneous soft tissue–and fluid-attenuating foreign material is present in several intestinal segments on the CT images.
Citation: Journal of the American Veterinary Medical Association 251, 3; 10.2460/javma.251.3.315
Discussion
Findings of the present study suggested that noncontrast abdominal CT was feasible, rapid, and accurate for determining the presence and cause of mechanical gastrointestinal obstruction in dogs in which such obstruction is suspected. Therefore, the use of CT in place of or in addition to abdominal ultrasonography may provide useful information and prevent some dogs without obstruction from having to undergo an emergent surgery. Although the PPV for CT was greater than that of ultrasonography (100% vs 93%), only 3 dogs without a mechanical obstruction were included, precluding formal statistical comparison of accuracy between the 2 imaging modalities.
Gastrointestinal obstruction is a common and serious disease in dogs and humans, and diagnostic imaging is important in the assessment of affected patients. In the study reported here, abdominal ultrasonography and CT were highly effective and accurate in the diagnosis of mechanical gastrointestinal obstruction in dogs, with disagreement between the 2 modalities noted for only 1 dog (Figure 1). In this dog, a false-positive diagnosis of obstruction was made via ultrasonography, reducing the PPV to 93%. In contrast to ultrasonography, results of CT and exploratory surgery (laparoscopy and exploratory laparotomy; used as the reference standard) had 100% agreement, yielding a PPV for CT of 100%. Nonetheless, this finding must be interpreted in light of the low prevalence of nonobstructive, functional gastrointestinal disease in the study sample. The authors consider it likely that the proportions of included dogs with and without obstruction were reflective of the clinical scenario in an academic referral hospital where suspected gastrointestinal obstruction in most dogs has a mechanical cause. Evaluation of diagnostic images by board-certified radiologists (as was done in the present study) generally minimizes the incidence of false-positive diagnoses of gastrointestinal obstruction.
Although the incidence of a false-positive diagnosis of mechanical gastrointestinal obstruction appeared low in the present study, this incidence was likely clinically important in some situations. For example, in the dog with a false-positive ultrasonographic result, exploratory surgery was subsequently performed. This dog had a history of progressive vomiting and had been unresponsive to medical management for more than a week. Because of the severe condition of the dog and the unknown accuracy of abdominal CT for the diagnosis of gastrointestinal obstruction, the decision was made to perform exploratory gastrointestinal surgery and obtain biopsy specimens for definitive diagnosis. There was no apparent benefit of the emergent surgical procedure performed in this dog, and a prospective study to compare CT findings in a large number of dogs with and without obstruction is indicated to better evaluate the accuracy of abdominal CT for this purpose in dogs.
In 1 other dog, ultrasonography allowed the correct identification of small intestinal obstruction, but CT was useful to exclude the presence of an intraluminal foreign body and to reveal disruption of the properitoneal fat line coupled with close association of the small intestine (Figure 2), These characteristics have been identified in humans with enteroparietal adhesions to the body wall related to prior laparotomy.11 In another dog in which a mural lesion was detected via CT and ultrasonography, exploratory surgery was needed to determine that the obstructive component was the result of sheet-like connections of nonvascular fibrous tissue between jejunal segments that are often indicative of enteroenteral adhesions in humans.11 The CT findings associated with abdominal adhesions have not been well described in the veterinary literature but may be similar to those reported for humans and the dogs of the present study. A presurgical diagnosis of adhesions may have particular importance when considering a laparoscopic-assisted surgical approach in dogs with gastrointestinal obstruction, given that adhesions are a contraindication and reason for conversion to laparotomy in dogs.10 Therefore, performance of abdominal CT before minimally invasive GI surgery may be helpful to further reduce the rate of conversion to laparotomy in dogs with suspected gastrointestinal obstruction.10
Although dogs with mechanical gastrointestinal obstruction had significantly greater intestinal diameter ratios than dogs without obstruction in the present study, these values should not be used independently to determine whether obstruction exists given that linear foreign bodies and partial obstructions may not result in marked intestinal distention. Indeed, although ratios for 4 dogs with gastrointestinal obstruction were within the IQR of ratios for dogs without obstruction, a diagnosis of mechanical obstruction was still made. Intestinal diameter ratios measured on CT images with no evidence of intestinal plication may help distinguish dogs that can be managed medically from those that require surgical intervention. Alternatively, it may be that, because of the superior contrast resolution of CT, calculation of intestinal diameter ratios may be unnecessary. Additional research involving construction of receiver operating characteristic curves and larger numbers of dogs with and without obstruction is required to better assess and define cutpoints and the associated sensitivity and specificity of intestinal diameter ratios for the diagnosis of mechanical gastrointestinal obstruction in dogs.
The amount of time required for ultrasonographic image acquisition was dramatically more than that for CT image acquisition, which was performed approximately 10 times as fast as ultrasonography. It is important to consider, however, that the time required to perform an ultrasonographic examination often includes interpretation, which typically occurs concurrent with the examination. Acquisition of the CT images happens rapidly, and with the availability of MDCT and multiplane reformatting of data, thousands of images may be available for review. Interpretation time for abdominal CT scans was not recorded in the present study, but if included, the difference between CT and ultrasonography in acquisition times may have been less dramatic.
The nature of the image acquisition by ultrasonography is vastly different from that by CT. An ultrasonographer typically acquires representative images or video clips, which are stored or archived and can be later reviewed. The entirety of the gastrointestinal tract may be completely imaged ultrasonographically, but recorded images typically represent only samples of the total examination. In contrast, CT data acquisition can record the entire abdomen, including the musculoskeletal structures. Further, the data are available in their entirety after the CT examination is complete, and they can be reviewed and manipulated by the surgeon or other clinician at any time after completion.
The availability of ultrasonography has been steadily increasing, as has the availability of CT. Results of abdominal ultrasonography can be operator dependent, and an experienced ultrasonographer may not always be available. With the increasing availability of MDCT and use of teleradiology, a certified radiologist can be easily accessed for interpretation of a CT examination. The rapid acquisition of CT images while animals are sedated is a clear benefit over ultrasonography, which may require an ill or vomiting patient to remain in dorsal recumbency for an extended period.
Limitations of the present study include the small sample size as well as the nature of the included dogs. Specifically, all enrolled dogs were suspected of having mechanical gastrointestinal obstruction on the basis of their clinical signs and results of physical examination and abdominal radiography. However, we believe that the study design and results were reflective of the typical clinical setting in which dogs routinely undergo examination and abdominal radiography prior to exploratory surgery. Further, we did not statistically compare the accuracy of ultrasonography and CT, so conclusions about superiority cannot be made.
The speed with which abdominal CT was performed relative to abdominal ultrasonography in the present study and the increasing access to MDCT and expert CT image interpretation may render CT examination of the abdomen in dogs with suspected mechanical gastrointestinal obstruction the preferable diagnostic technique. Intestinal diameter ratios may provide additional information regarding the need for surgical intervention versus medical management in dogs with intestinal foreign material, but further research is required to investigate this possibility.
Acknowledgments
Supported by a seed grant from the American Kennel Club Canine Health Foundation
Presented in part as an abstract at the American College of Veterinary Radiology Annual Scientific Meeting, Minneapolis, October 2015.
ABBREVIATIONS
| IQR | Interquartile range |
| MDCT | Multidetector-row CT |
| NPV | Negative predictive value |
| PPV | Positive predictive value |
Footnotes
Philips iU22, Philips Healthcare, Andover, Mass.
Hitachi Preirus, Hitachi Medical Systems, Zug, Switzerland.
Toshiba Acquilion 8, Toshiba America Medical Systems, Tustin, Calif.
Toshiba Acquilion 16, Toshiba America Medical Systems, Tustin, Calif.
Toshiba Prime 160, Toshiba America Medical Systems, Tustin, Calif.
JMP, version 9.0.2, SAS Institute Inc, Raleigh, NC.
References
1. Graham JP, Lord PF, Harrison JM. Quantitative estimation of intestinal dilation as a predictor of obstruction in the dog. J Small Anim Pract 1998; 39: 521–524.
2. Sharma A, Thompson MS, Scrivani PV, et al. Comparison of radiography and ultrasonography for diagnosing small-intestinal mechanical obstruction in vomiting dogs. Vet Radiol Ultrasound 2011; 52: 248–255.
3. Ciasca TC, David FH, Lamb CR. Does measurement of small intestinal diameter increase diagnostic accuracy of radiography in dogs with suspected intestinal obstruction? Vet Radiol Ultrasound 2013; 54: 207–211.
4. Finck C, D'Anjou M-A, Alexander K, et al. Radiographic diagnosis of mechanical obstruction in dogs based on relative small intestinal external diameters. Vet Radiol Ultrasound 2014; 55: 472–479.
5. Tidwell AS, Penninck DG. Ultrasonography of gastrointestinal foreign bodies. Vet Radiol Ultrasound 1992; 33: 160–169.
6. Hryhorczuk AL, Lee EY. Imaging evaluation of bowel obstruction in children: updates in imaging techniques and review of imaging findings. Semin Roentgenol 2012; 47: 159–170.
7. Pastore GE, Lamb CR, Lipscomb V. Comparison of the results of abdominal ultrasonography and exploratory laparotomy in the dog and cat. J Am Anim Hosp Assoc 2007; 43: 264–269.
8. Shanaman MM, Hartman SK, O'Brien RT. Feasibility for using dual-phase contrast-enhanced multi-detector helical computed tomography to evaluate awake and sedated dogs with acute abdominal signs. Vet Radiol Ultrasound 2012; 53: 605–612.
9. Shanaman MM, Schwarz T, Gal A, et al. Comparison between survey radiography, B-mode ultrasonography, contrast-enhanced ultrasonography and contrast-enhanced multi-detector computed tomography findings in dogs with acute abdominal signs. Vet Radiol Ultrasound 2013; 54: 591–604.
10. Barry KS, Case JB, Winter MD, et al. Diagnostic usefulness of laparoscopy versus exploratory laparotomy for dogs with suspected gastrointestinal obstruction. J Am Vet Med Assoc 2017; 251: 307–314.
11. Ghonge NP, Ghonge SD. Computed tomography and magnetic resonance imaging in the evaluation of pelvic peritoneal adhesions: what radiologists need to know? Indian J Radiol Imaging 2014; 24: 149–155.
