The esophagus can be a difficult organ to image. Commonly used diagnostic imaging modalities, such as radiography, esophagography, ultrasonography, and endoscopy, have limitations for evaluation of the esophagus.1–3 In dogs, positive-contrast esophagography is frequently used to identify pathological esophageal lesions, but there is a risk patients may aspirate the contrast agent (eg, barium) or that the contrast agent may obscure visibility of small nodules.3 Ultrasonographic evaluation of the thoracic portion of the esophagus is hampered by the interposition of aerated lungs. Results of 1 study1 suggest that endoscopic ultrasonography may be useful for assessing esophageal wall integrity, but that modality is currently not readily available in most clinical settings. Endoscopy is a noninvasive atraumatic technique that allows visual inspection of the esophageal lining and biopsy of visible lesions, but it cannot be used to identify or evaluate periesophageal abnormalities, such as lesions located in the adjacent lymph nodes or lesions located in the muscular or submucosal layers of the esophagus, or to accurately describe the location and extent of lesions for surgical planning.2
Computed tomography is useful for evaluation of the location, severity, and extent of esophageal lesions as well as surrounding structures, and it is often used to glean information about lesion size and surrounding structures during tumor staging and surgical resection planning.4–9 The entire length of the esophagus can be evaluated with CT. In CT images, the esophagus appears as a round or ovoid shape with a soft-tissue density in the cervical and thoracic regions and has a more triangular shape at its most caudal aspect.10 However, the esophagus is a collapsible organ and is generally in a collapsed state. Collapse of the esophageal lumen or contraction of the muscular wall during the acquisition of standard unenhanced or contrast-enhanced CT images limits the information that can be obtained in regard to the extent and invasiveness of lesions. Several techniques can be applied to overcome that drawback, including use of oral effervescent granule contrast agents, an oxygen bag, or CO2 to achieve esophageal distention.5,7–9,11,12 In human medicine, the use of CO2 to distend the esophageal lumen is considered superior to other alternatives. That technique requires a continuous supply of CO2 to distend and maintain a pressure of 10 to 20 mm Hg within the esophageal lumen throughout the CT scan.5,9
The introduction of multidetector CT has made CT much more readily available in veterinary practice, and CT is frequently used to evaluate the gastrointestinal tracts of veterinary patients.13,14 A poorly distended gastrointestinal lumen can obscure luminal lesions such as nodules and masses, thereby decreasing the diagnostic efficacy of most imaging modalities including CT, and techniques such as CT-pneumocolonography and hydro-CT have been developed to circumvent that obstacle.15–17 In veterinary species, insufflation of gas through an orogastric or endotracheal tube is commonly used to distend the esophageal lumen and facilitate the delineation of intraluminal and mural masses during diagnostic imaging.3,4,10 However, detailed data regarding the optimal insufflation pressure to achieve continuous and consistent esophageal distention in dogs are currently unavailable.
The objective of the study reported here was to assess the feasibility of EICT for evaluation of the entire extent of the esophagus in dogs. We hypothesized that EICT, by which CO2 was insufflated into the esophagus during CT scanning, would be a feasible and noninvasive method for evaluation of the entire esophagus and its surrounding structures. A secondary objective was to determine whether images acquired by EICT could be used for MPR and virtual endoscopy.
Materials and Methods
Animals
All study procedures were reviewed and approved by the Seoul National University Institutional Animal Care and Use Committee. Seven university-owned adult Beagles (1 female and 6 males) with a mean ± SD weight of 14.27 ± 0.68 kg (range, 11.2 to 16.4 kg) were used for the study. All dogs were determined to be healthy on the basis of results of a physical examination, CBC, and serum biochemical profile. None of the dogs had evidence of an esophageal disorder. The dogs had a median body condition score of 6 (range, 4 to 8) on a scale of 1 to 9.
Anesthesia protocol
Dogs were anesthetized twice for CT scanning with at least 1 week between anesthesia sessions. The anesthesia protocol was standardized. Food but not water was withheld from each dog for at least 12 hours before anesthesia induction. Each dog was premedicated with acepromazine (0.01 mg/kg, IV), tramadol (2.0 mg/kg, IV), and atropine (0.01 mg/kg, IV). Anesthesia was induced with alfaxalone (2.0 mg/kg, IV) and maintained with isoflurane in oxygen, which was administered through an endotracheal tube by an anesthesia machine. Each dog was instrumented with a multiparameter patient monitor that included ECG, pulse oximetry, capnography, and direct blood pressure monitoring for the duration of each anesthesia session.
CT protocol
Once anesthetized and instrumented, each dog was positioned in sternal recumbency on the CT table. Prior to initiation of the CT scanning procedure, a 12F balloon-tipped Foley cathetera was inserted through the oral cavity and into the esophageal lumen such that the balloon was positioned just caudal to the cricopharyngeal muscles. The balloon was inflated with 5 mL of room air, and a survey CT scan of the cervical region was performed to confirm correct positioning of the catheter (Figure 1). Unenhanced and contrast-enhanced EICT scans from the level of C2 to the caudal aspect of the stomach were acquired during each anesthesia session. Immediately prior to each scan, a single breath–hold technique was implemented by which the dog was manually hyperventilated to induce transient apnea, thereby preventing motion artifact and decreasing lung volume during scanning. All scans were obtained by use of a 64-row multidetector CT scanner.b Scanner settings were as follows: 120 kV; 200 mA; slice thickness, 1 mm; tube rotation speed, 0.75 seconds; helical pitch, 27.0; pitch factor, 0.844; and interslice gap, 1 mm. Contiguous reconstructions were performed. The field of view was set to encompass the esophagus and stomach to its caudal aspect.
Representative sagittal survey CT image of the cervical region of a clinically normal adult Beagle that depicts correct positioning of a 12F balloon-tipped Foley catheter within the esophageal lumen prior to insufflation of CO2 for EICT. Notice the inflated balloon (dashed line) cranial to the catheter tip (arrow) was positioned just caudal to the cricopharyngeal muscles (asterisks).
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative sagittal survey CT image of the cervical region of a clinically normal adult Beagle that depicts correct positioning of a 12F balloon-tipped Foley catheter within the esophageal lumen prior to insufflation of CO2 for EICT. Notice the inflated balloon (dashed line) cranial to the catheter tip (arrow) was positioned just caudal to the cricopharyngeal muscles (asterisks).
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative sagittal survey CT image of the cervical region of a clinically normal adult Beagle that depicts correct positioning of a 12F balloon-tipped Foley catheter within the esophageal lumen prior to insufflation of CO2 for EICT. Notice the inflated balloon (dashed line) cranial to the catheter tip (arrow) was positioned just caudal to the cricopharyngeal muscles (asterisks).
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
During the first anesthesia session, an unenhanced EICT scan was performed before (insufflation pressure, 0 mm Hg) and after an electronic CO2 insufflatorc was used to deliver a continuous and sustained supply of CO2 through the Foley catheter into the esophageal lumen at a rate of 1 L/min until a pressure of 5 mm Hg was achieved. Following completion of the second unenhanced EICT scan (insufflation pressure, 5 mm Hg), the dog received iohexold (600 mg/kg, IV), a positive nonionic monomer contrast medium, and a contrast-enhanced EICT scan was performed 30 seconds later. After completion of the contrast-enhanced EICT scan, isoflurane administration was discontinued and the dog was allowed to recover from anesthesia in a routine manner. During the second anesthesia session, unenhanced and contrast-enhanced EICT scans were performed in the same manner as that described for the first anesthesia session except that the insufflation pressure achieved was 10 mm Hg. All dogs were monitored daily for signs of pain and difficulty swallowing for 1 week after each anesthesia session.
Image analyses
The CT images were reconstructed for analysis, and all measurements were performed on representative unenhanced and contrast-enhanced transverse images by use of image analysis software.e The esophagus was divided into 5 segments (cervical, thoracic inlet, heart base, caudal intrathoracic, and gastroesophageal). Measurements were obtained for each section at each of the 3 insufflation pressures (0, 5, and 10 mm Hg). The esophageal luminal cross-sectional area and wall thickness were measured at the level of C4 for the cervical segment, T1 for the thoracic-inlet segment, carina for the heart-base segment, and T7-8 intervertebral space for the caudal intrathoracic segment and 2 cm cranial to the gastroesophageal junction for the gastroesophageal segment. All images were reviewed in a soft tissue window with sufficient magnification and manual adjustments to maximize the conspicuity of the esophageal wall. Multiplanar reconstruction and curved MPR images were also obtained, and a dedicated image processing workstationf was used to generate virtual endoluminal images of the esophageal lumen.
Statistical analysis
Descriptive statistics were generated for esophageal luminal cross-sectional area and wall thickness. For each of those outcomes, paired Wilcoxon signed rank tests were used to assess whether the median difference was 0 between insufflation pressures of 0 and 5 mm Hg, 0 and 10 mm Hg, and 5 and 10 mm Hg at each of the 5 esophageal segments (cervical, thoracic inlet, heart base, caudal intrathoracic, and gastroesophageal). The Kruskal-Wallis test was used to compare the esophageal luminal cross-sectional area among the 5 segments at insufflation pressures of 5 and 10 mm Hg and esophageal wall thickness among the 5 segments at all 3 insufflation pressures evaluated. When necessary, pairwise comparisons were performed by means of Mann-Whitney U tests with the Bonferroni correction used to control for type I error inflation. Pearson correlation coefficients (r) were calculated to evaluate the relationship between esophageal luminal cross-sectional area and wall thickness at each of the 3 insufflation pressures. The Wilcoxon test was used to compare the esophageal wall thickness between unenhanced and contrast-enhanced EICT images at each of the 5 segments. Values of P < 0.05 were considered significant. All analyses were performed with standard statistical software.g
Results
EICT procedure
The esophagus was successfully distended with CO2 in all 7 dogs, and no adverse effects associated with anesthesia, esophageal insufflation, or CT scanning were observed in any of the dogs. All dogs recovered without complications from both anesthesia sessions. The mean ± SD CT scan duration was 10.76 ± 0.25 seconds (range, 9.8 to 10.9 seconds). The total procedure duration (including placement of the Foley catheter, esophageal insufflation, contrast medium injection, and CT scanning) was < 10 minutes for each dog. The total duration of esophageal insufflation (including CT scan acquisition) was < 30 seconds; thus, the insufflation pressure gauge was maintained at 5 mm Hg for only a short period of time during the first anesthesia session and was not consistently maintained at 10 mm Hg for any duration during the second anesthesia session.
Esophageal measurements
In all dogs, the esophageal lumen was not well visualized on CT images acquired before CO2 insufflation because the organ was collapsed. Distention of the esophagus but not the stomach was observed when the insufflation pressure was maintained at 5 mm Hg (Figure 2). When the insufflation pressure was maintained at 10 mm Hg, the esophagus was distended to a greater extent than when the insufflation pressure was maintained at 5 mm Hg, and the stomach became distended with gas as well. The mean ± SD esophageal luminal cross-sectional area and wall thickness at each of the 5 segments was summarized (Table 1). In general, the esophageal luminal cross-sectional area increased and the esophageal wall thickness decreased as the insufflation pressure increased, and the magnitudes of those changes were significantly > 0 for all pairwise comparisons except for the change in esophageal wall thickness between 5 and 10 mm Hg (P = 0.237). There was a significant (P < 0.001) negative correlation (r = −0.752) between esophageal luminal cross-sectional area and wall thickness. The mean luminal cross-sectional area of the esophagus at the thoracic-inlet, heart-base, and gastroesophageal segments was significantly (P < 0.01 for all comparisons) less than that at the cervical and caudal intrathoracic segments at insufflation pressures of 5 and 10 mm Hg. The mean esophageal wall thickness at the gastroesophageal segment was significantly (P < 0.01) greater than that at the other 4 segments prior to insufflation (insufflation pressure, 0 mm Hg) but did not differ significantly from that at the other 4 segments at insufflation pressures of 5 and 10 mm Hg. Following IV administration of iohexol, moderate uniform contrast enhancement of the esophageal wall was observed at insufflation pressures of 5 and 10 mm Hg. Subjectively, when the esophagus was surrounded by soft tissue such as fat and muscle, the esophageal wall conspicuity on contrast-enhanced EICT images was increased relative to that on unenhanced CT images (Figure 3). However, the esophageal wall thickness measurements did not differ significantly between unenhanced and contrast-enhanced EICT images.
Representative unenhanced transverse CT images of a clinically normal adult Beagle obtained at the level of C4 (cervical segment; A, B, and C), T1 (thoracic-inlet segment; D, E, and F), carina (heart-base segment; G, H, and I), and T7-8 intervertebral space (caudal intrathoracic segment; J, K, and L) and 2 cm cranial to the gastroesophageal junction (gastroesophageal segment; M, N, and O) before (insufflation pressure, 0 mm Hg; left column) and after CO2 was insufflated into the esophageal lumen to achieve a pressure of 5 (middle column) or 10 (right column) mm Hg. Within each segment, notice that the diameter of the esophagus (white arrow) increased as the insufflation pressure increased. The trachea (arrowhead) or tracheal bifurcation (outlined arrow) is indicated when present.
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative unenhanced transverse CT images of a clinically normal adult Beagle obtained at the level of C4 (cervical segment; A, B, and C), T1 (thoracic-inlet segment; D, E, and F), carina (heart-base segment; G, H, and I), and T7-8 intervertebral space (caudal intrathoracic segment; J, K, and L) and 2 cm cranial to the gastroesophageal junction (gastroesophageal segment; M, N, and O) before (insufflation pressure, 0 mm Hg; left column) and after CO2 was insufflated into the esophageal lumen to achieve a pressure of 5 (middle column) or 10 (right column) mm Hg. Within each segment, notice that the diameter of the esophagus (white arrow) increased as the insufflation pressure increased. The trachea (arrowhead) or tracheal bifurcation (outlined arrow) is indicated when present.
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative unenhanced transverse CT images of a clinically normal adult Beagle obtained at the level of C4 (cervical segment; A, B, and C), T1 (thoracic-inlet segment; D, E, and F), carina (heart-base segment; G, H, and I), and T7-8 intervertebral space (caudal intrathoracic segment; J, K, and L) and 2 cm cranial to the gastroesophageal junction (gastroesophageal segment; M, N, and O) before (insufflation pressure, 0 mm Hg; left column) and after CO2 was insufflated into the esophageal lumen to achieve a pressure of 5 (middle column) or 10 (right column) mm Hg. Within each segment, notice that the diameter of the esophagus (white arrow) increased as the insufflation pressure increased. The trachea (arrowhead) or tracheal bifurcation (outlined arrow) is indicated when present.
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative contrast-enhanced transverse CT images that depict the cervical (A and B), thoracic-inlet (C and D), heart-base (E and F), caudal intrathoracic (G and H), and gastroesophageal (I and J) segments of the esophagus of a clinically normal adult Beagle obtained after CO2 was insufflated into the esophageal lumen to achieve a pressure of 5 (left column) or 10 (right column) mm Hg and 30 seconds after administration of iohexol (600 mg/kg, IV). Notice that the esophageal wall was more conspicuous in the contrast-enhanced images than the unenhanced images of Figure 2, especially when the esophagus was surrounded by soft tissues such as fat and muscles (A, B, C, and D). See Figure 2 for remainder of key.
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative contrast-enhanced transverse CT images that depict the cervical (A and B), thoracic-inlet (C and D), heart-base (E and F), caudal intrathoracic (G and H), and gastroesophageal (I and J) segments of the esophagus of a clinically normal adult Beagle obtained after CO2 was insufflated into the esophageal lumen to achieve a pressure of 5 (left column) or 10 (right column) mm Hg and 30 seconds after administration of iohexol (600 mg/kg, IV). Notice that the esophageal wall was more conspicuous in the contrast-enhanced images than the unenhanced images of Figure 2, especially when the esophagus was surrounded by soft tissues such as fat and muscles (A, B, C, and D). See Figure 2 for remainder of key.
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative contrast-enhanced transverse CT images that depict the cervical (A and B), thoracic-inlet (C and D), heart-base (E and F), caudal intrathoracic (G and H), and gastroesophageal (I and J) segments of the esophagus of a clinically normal adult Beagle obtained after CO2 was insufflated into the esophageal lumen to achieve a pressure of 5 (left column) or 10 (right column) mm Hg and 30 seconds after administration of iohexol (600 mg/kg, IV). Notice that the esophageal wall was more conspicuous in the contrast-enhanced images than the unenhanced images of Figure 2, especially when the esophagus was surrounded by soft tissues such as fat and muscles (A, B, C, and D). See Figure 2 for remainder of key.
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Summary statistics for esophageal luminal cross-sectional area and wall thickness at each of 5 segments as determined by measurements obtained from unenhanced transverse EICT images acquired at insufflation pressures of 0, 5, and 10 mm Hg.
| Insufflation pressure (mm Hg) | P value* | ||||||
|---|---|---|---|---|---|---|---|
| Esophageal segment | Variable | 0 | 5 | 10 | 0 vs 5 mm Hg | 0 vs 10 mm Hg | 5 vs 10 mm hg |
| Cervical | Luminal cross-sectional area (cm2) | 1.44 ± 0.31 | 4.11 ± 0.29 | 4.99 ± 0.17 | 0.018 | 0.018 | 0.018 |
| Wall thickness (cm) | 0.15 ± 0.02 | 0.11 ± 0.01 | 0.09 ± 0.01 | 0.043 | 0.018 | 0.237 | |
| Thoracic inlet | Luminal cross-sectional area (cm2) | 0.49 ± 0.20 | 2.47 ± 0.24 | 3.54 ± 0.28 | 0.018 | 0.018 | 0.018 |
| Wall thickness (cm) | 0.18 ± 0.02 | 0.12 ± 0.01 | 0.09 ± 0.01 | 0.028 | 0.018 | 0.042 | |
| Heart base | Luminal cross-sectional area (cm2) | 0.18 ± 0.09 | 1.98 ± 0.33 | 3.42 ± 0.43 | 0.018 | 0.018 | 0.018 |
| Wall thickness (cm) | 0.19 ± 0.02 | 0.13 ± 0.01 | 0.09 ± 0.01 | 0.018 | 0.018 | 0.028 | |
| Caudal intrathoracic | Luminal cross-sectional area (cm2) | 0.91 ± 0.36 | 4.26 ± 0.32 | 5.26 ± 0.30 | 0.018 | 0.018 | 0.018 |
| Wall thickness (cm) | 0.22 ± 0.02 | 0.11 ± 0.01 | 0.09 ± 0.01 | 0.018 | 0.018 | 0.075 | |
| Gastroesophageal | Luminal cross-sectional area (cm2) | 0.22 ± 0.11 | 2.42 ± 0.34 | 3.37 ± 0.37 | 0.018 | 0.018 | 0.018 |
| Wall thickness (cm) | 0.29 ± 0.02 | 0.15 ± 0.01 | 0.13 ± 0.01 | 0.018 | 0.018 | 0.028 | |
Each variable was measured at the level of C4 for the cervical segment, T1 for the thoracic-inlet segment, carina for the heart-base segment, and T7-8 intervertebral space for the caudal intrathoracic segment and 2 cm cranial to the gastroesophageal junction for the gastroesophageal segment.
For paired Wilcoxon signed rank tests that were used to assess whether the median difference was 0 between the indicated insufflation pressures; values of P < 0.05 were considered significant.
Image reconstruction
Multiplanar reconstruction and curved MPR were used to characterize the distended esophagus. Results revealed that the entire esophageal lumen could be viewed in 1 cross-sectional dorsal or sagittal curved MPR image (Figure 4). Additionally, virtual endoscopy was performed without any problems because distention of the esophagus allowed the lumen to be viewed from all angles.
Representative unenhanced dorsal (A and B) and sagittal (C and D) MPR (A and C) and curved MPR (B and D) images of the cervical and thoracic regions of a clinically normal adult Beagle following insufflation of CO2 into the esophageal lumen to achieve a pressure of 5 mm Hg. Notice that curved MPR allowed the entire esophagus (arrows) and esophageal lumen to be viewed in 1 cross-sectional image in both the dorsal and sagittal planes. In panel D, notice that the esophageal luminal diameter was smaller at the heart base segment than that at the other segments, likely owing to the presence of the tracheal bifurcation (arrowhead).
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative unenhanced dorsal (A and B) and sagittal (C and D) MPR (A and C) and curved MPR (B and D) images of the cervical and thoracic regions of a clinically normal adult Beagle following insufflation of CO2 into the esophageal lumen to achieve a pressure of 5 mm Hg. Notice that curved MPR allowed the entire esophagus (arrows) and esophageal lumen to be viewed in 1 cross-sectional image in both the dorsal and sagittal planes. In panel D, notice that the esophageal luminal diameter was smaller at the heart base segment than that at the other segments, likely owing to the presence of the tracheal bifurcation (arrowhead).
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Representative unenhanced dorsal (A and B) and sagittal (C and D) MPR (A and C) and curved MPR (B and D) images of the cervical and thoracic regions of a clinically normal adult Beagle following insufflation of CO2 into the esophageal lumen to achieve a pressure of 5 mm Hg. Notice that curved MPR allowed the entire esophagus (arrows) and esophageal lumen to be viewed in 1 cross-sectional image in both the dorsal and sagittal planes. In panel D, notice that the esophageal luminal diameter was smaller at the heart base segment than that at the other segments, likely owing to the presence of the tracheal bifurcation (arrowhead).
Citation: American Journal of Veterinary Research 80, 1; 10.2460/ajvr.80.1.61
Discussion
In the present study, EICT performed at insufflation pressures of 5 and 10 mm Hg appeared to be a safe and effective method to visualize the entire esophagus of clinically normal medium-sized adult dogs. The mean esophageal luminal cross-sectional area increased as the insufflation pressure increased, and there was a significant negative correlation between esophageal luminal cross-sectional area and wall thickness. Although the esophagus was distended to a greater extent at an insufflation pressure of 10 mm Hg relative to that at 5 mm Hg, the esophagus was adequately distended for CT imaging at an insufflation pressure of 5 mm Hg. Moreover, the stomach became distended with gas at an insufflation pressure of 10 mm Hg but not at 5 mm Hg.
In veterinary medicine, contrast esophagography and endoscopy are the most frequently used modalities to assess the esophagus for abnormalities, although both modalities have limitations for esophageal evaluation.2,3 Conventional CT of hollow organs such as the esophagus also has limitations in regard to evaluation of the lumen and mucosal surface of those organs. However, for the dogs of the present study, EICT was successfully used to visualize the esophageal lumen and wall thickness and made it easy to perform curved MPR and virtual endoscopic imaging of the esophageal lining. Adequate gaseous distention of the esophagus during CT is critical for evaluating esophageal wall thickness and performing virtual endoscopy.7
For the clinically normal dogs of the present study, the luminal cross-sectional area was smaller at the thoracic-inlet, heart-base, and gastroesophageal segments of the esophagus than at the cervical and caudal intrathoracic segments. There are extraesophageal structures adjacent to those 3 segments that might have restricted esophageal distention, as evidenced by the fact that foreign bodies within the esophagus are most commonly identified at those sites.18 Results of the present study also indicated that the thickness of the wall of the esophagus was not uniform throughout its length. The esophageal wall was thickest at the gastroesophageal segment, which is an area subject to high pressure owing to the caudal esophageal sphincter at the gastroesophageal junction.1,18,19 However, although the mean esophageal wall thickness at the gastroesophageal segment was significantly greater than that at each of the other 4 segments prior to insufflation of CO2, it did not differ significantly among the 5 segments at insufflation pressures of 5 and 10 mm Hg. That finding most likely reflected thinning of the esophageal wall as the lumen was distended with gas.
In anesthetized dogs of another study,20 the mean ± SD resting caudal esophageal sphincter pressure was 21 ± 10 cm H2O (15.45 ± 7.36 mm Hg). The resting pressure reflects the pressure when the sphincter is closed. Investigators of that study20 postulated that the caudal esophageal sphincter opens when the intraesophageal pressure slightly exceeds the resting sphincter pressure. For all dogs of the present study, the caudal esophageal sphincter opened and the stomach became distended with gas at an insufflation pressure of 10 mm Hg but not 5 mm Hg. Therefore, we believed that infusion of sufficient CO2 to maintain an insufflation pressure of 10 mm Hg exceeded the resting pressure of the caudal esophageal sphincter causing it to open and allow the gas to pass into the stomach. Given that an insufflation pressure of 5 mm Hg resulted in adequate distention of the esophagus for evaluation of the wall at all segments of interest except perhaps at high pressure areas such as the gastroesophageal junction, we believe that an insufflation pressure of 5 mm Hg will generally be sufficient for EICT in dogs.
In the present study, EICT was performed with dogs positioned in sternal recumbency and during periods of transient apnea to minimize lung volume and prevent motion artifact during scan acquisition. Lung volume and other mediastinal structures affect esophageal pressure. When human subjects are in a supine position, gravity causes the heart and great vessels to compress the esophagus and increases esophageal pressure.21 Moreover, esophageal pressure is inversely related to lung volume.21 The use of manual hyperventilation to induce transient apnea decreases lung volume, thereby decreasing pressure on the structures of the mediastinum including the esophagus.21 Thus, sternal positioning and low end-tidal lung volume should per mit maximal distention of the esophagus.21 However, that may not be advisable in patients with cardiorespiratory disease. Although esophageal insufflation and EICT scan duration was < 30 seconds for the dogs of the present study, all the dogs were clinically normal. Potential adverse effects associated with anesthetizing and inducing transient apnea in dogs with cardiorespiratory deficits should be considered before EICT is performed.
No major adverse effects specific to CO2 insufflation for EICT were described in a study9 of human patients with esophageal cancer. Carbon dioxide is the most commonly used gas to induce pneumoperitoneum during laparoscopic procedures. It is an extremely soluble gas that readily diffuses into the blood and tissues, potentially causing hypercapnia and acidosis, and is subsequently exhaled by the lungs.22 No adverse effects associated with esophageal insufflation of CO2 were observed in the dogs of the present study, and all dogs had uneventful recoveries from both anesthesia sessions. Also, for the dogs of this study, the amount of CO2 insufflated was small and duration of the EICT procedure was short, compared with corresponding values for laparoscopic procedures.22 In human medicine, CO2 insufflation during diagnostic or therapeutic endoscopic procedures is considered safe and effective, particularly when compared with more invasive and painful alternatives.23 Although EICT appeared to be a simple and quick method for esophageal evaluation in dogs, the heart rate, blood pressure, and end-tidal CO2 of patients should be closely monitored throughout the procedure.
The present study had some limitations. Esophageal insufflation CT requires equipment (eg, a mechanical insufflator) that is not needed for conventional CT, but that equipment was readily acquired and easily used. Only a small number of clinically normal dogs were evaluated. The EICT protocol might need to be modified for dogs with clinical disease, especially those with partial esophageal obstructions or lesions that affect esophageal wall strength or compliance. Potential modifications in clinical patients might include higher insufflation pressures than those evaluated in this study or altered positioning (eg, lateral recumbency). Additional studies are necessary to determine the optimal protocol and diagnostic efficacy of EICT for detection of dogs with esophageal lesions. Also, further research is warranted to evaluate the insufflation pressure necessary to cause distention of the stomach with CO2 as well as the potential advantages and disadvantages associated with gaseous distention of the stomach during EICT.
Results of the present study indicated that EICT was a safe and feasible technique for evaluation of the esophagus in clinically normal dogs. An insufflation pressure of 5 mm Hg appeared to be sufficient for adequate visualization of the esophageal lumen and measurement of the esophageal luminal cross-sectional area and wall thickness. Esophageal insufflation CT may be a useful and noninvasive imaging method for evaluation of esophageal and extraesophageal lesions in dogs with clinical disease, but further research is necessary to determine the optimal protocol for the procedure and validate its diagnostic efficacy in diseased patients.
Acknowledgments
Supported by the Research Institute for Veterinary Science, Seoul National University.
ABBREVIATIONS
| EICT | Esophageal insufflation CT |
| MPR | Multiplanar reconstruction |
Footnotes
Yushin Medical Co Ltd, Bucheon, South Korea.
Aquilion 64, Toshiba Medical Systems Corp, Ωtawara, Japan.
Endoflator, Karl Stortz, Tuttlingen, Germany.
Omnipaque 300, GE Healthcare, Cork, Ireland.
INFINITT, Infinitt Healthcare Co Ltd, Seoul, South Korea.
Vitrea 2, Vital Images Inc, Minnetonka, Minn.
SPSS Statistics for Windows, version 23.0, IBM Corp, Armonk, NY.
References
1. Baloi PA, Kircher PR, Kook PH. Endoscopic ultrasonographic evaluation of the esophagus in healthy dogs. Am J Vet Res 2013;74:1005–1009.
2. Guiford WG. Upper gastrointestinal endoscopy. In: Constantinescu GM, McCarthy TC, eds. Veterinary endoscopy for the small animal practitioner. St Louis: Elsevier Saunders, 2005;279–321.
3. Kirberger RM, van der Merwe LL, Dvir E. Pneumoesophagography and the appearance of masses in the caudal portion of the esophagus in dogs with spirocercosis. J Am Vet Med Assoc 2012;240:420–426.
4. Kirberger RM, Cassel N, Stander N, et al. Triple phase dynamic computed tomographic perfusion characteristics of spirocercosis induced esophageal nodules in non-neoplasitic versus neoplastic canine cases. Vet Radiol Ultrasound 2015;56:257–263.
5. Ulla M, Gentile EMJ, Cavadas D, et al. Esophageal cancer characterization with pneumo-64-MDCT. Abdom Imaging 2012;37:501–511.
6. Brissot HN, Burton CA, Doyle RS, et al. Caudal mediastinal paraesophageal abscesses in 7 dogs. Vet Surg 2012;41:286–291.
7. Onbaş O, Eroglu A, Kantarci M, et al. Preoperative staging of esophageal carcinoma with multidetector CT and virtual endoscopy. Eur J Radiol 2006;57:90–95.
8. Mazzeo S, Caramella D, Gennai A, et al. Multidetector CT and virtual endoscopy in the evaluation of the esophagus. Abdom Imaging 2004;29:2–8.
9. Ulla M, Gentile E, Yeyati EL, et al. Pneumo-CT assessing response to neoadjuvant therapy in esophageal cancer: imaging-pathological correlation. World J Gastrointest Oncol 2013;5:222–229.
10. Petite A, Kirbergr R. Mediastinum. In: Schwarz T, Saunders J, eds. Veterinary computed tomography. Chichester, England: John Wiley and Sons, 2011;249–260.
11. Liu BR, Liu BL, Wang XH, et al. Esophageal insufflation computed tomography for the diagnosis and management of esophageal submucosal tumors. Surg Endosc 2017;31:2350–2355.
12. Ulla M, Cavadas D, Muñoz I, et al. Esophageal cancer: pneumo-64-MDCT. Abdom Imaging 2010;35:383–389.
13. Hoey S, Drees R, Hetzel S. Evaluation of the gastrointestinal tract in dogs using computed tomography. Vet Radiol Ultrasound 2013;54:25–30.
14. Vignoli M, Saunders J. Gastrointetinal tract. In: Schwarz T, Saunders J, eds. Veterinary computed tomography. Chichester, England: John Wiley and Sons, 2011;325–330.
15. Steffey MA, Daniel L, Taylor SL, et al. Computed tomographic pneumocolonography in normal dogs. Vet Radiol Ultrasound 2015;56:278–285.
16. Terragni R, Vignoli M, Rossi F, et al. Stomach wall evaluation using helical hydro-computed tomography. Vet Radiol Ultrasound 2012;53:402–405.
17. Yamada K, Morimoto M, Kishimoto M, et al. Virtual endoscopy of dogs using multi-detector row CT. Vet Radiol Ultrasound 2007;48:318–322.
18. Kyles AE. Esophagus. In: Tobias KM, Johnston SA, eds. Veterinary surgery: small animal. St Louis: Elsevier Saunders, 2012;1461–1483.
19. Evans HE, De Lahunta A. The digestive apparatus and abdomen. In: Miller's anatomy of the dog. 4th ed. St Louis: Elsevier Saunders, 2013;304–307.
20. Pettersson GB, Bombeck CT, Nyhus LM. The lower esophageal sphincter: mechanisms of opening and closure. Surgery 1980;88:307–314.
21. Ferris BG Jr, Mead J, Frank NR. Effect of body position on esophageal pressure and measurement of pulmonary compliance. J Appl Physiol 1959;14:521–524.
22. Fukushima FB, Malm C, Andrade ME, et al. Cardiorespiratory and blood gas alterations during laparoscopic surgery for intrauterine artificial insemination in dogs. Can Vet J 2011;52:77–79.
23. Maeda Y, Hirasawa D, Fujita N, et al. Carbon dioxide insufflation in esophageal endoscopic submucosal dissection reduces mediastinal emphysema: a randomized, double-blind, controlled trial. World J Gastroenterol 2016;22:7373–7382.
