In sexually intact female dogs with a caudal abdominal mass, determining the origin of lesions between the uterus and vagina is important for diagnosis and treatment plans.1–5 Hydrometra, mucometra, hemometra, and pyometra with or without cystic endometrial hyperplasia are generally suspected when a fluid-filled tubular or cystic structure that originates from the uterus is present.6,7 In the vagina, the tumors of muscle or fibrous tissue origin (leiomyoma, fibroma, and leiomyosarcoma) are most commonly found.3,6 However, if a fluid-filled tubular or cystic structure is found in the vagina, hydrocolpos secondary to congenital vaginal anomalies or cystic lesions, such as Gartner’s duct cyst, can be considered.8–10 A tumor lesion in the uterus can be resected using ovariohysterectomy.1 Depending on the extent of the tumor and its association with the urethra, pelvic osteotomy, episiotomy, and perineal urethrostomy may be required during vaginal tumor resection.2–5
Ultrasonography is the first step in the diagnosis of reproductive tract lesions. However, the visibility of the intrapelvic portion of the vagina is limited with ultrasonography, even when combining perineal and dorsal approaches.11 When the vaginal lesion is large enough to occupy the caudal abdomen and displace the uterus cranially, it can be difficult to distinguish the origin of the vaginal lesion from the uterus using ultrasonography or radiography.1,4,12
Due to the radiation exposure and anesthesia requirement, computed tomography (CT) is not routinely used to evaluate the reproductive tract in the dog. However, in a uterine or vaginal mass, CT evaluation of the location, size, extent, and characteristics of the mass may help determine the mass origin and make the surgical planning of the lesion.3,13,14 In human medicine, studies have assessed CT features of the uterus and vagina according to age, pregnancy status, and menstrual cycle.15,16 However, to the authors’ knowledge, CT has been applied to only a few cases of reproductive tract disease in veterinary medicine. These cases include tumors (leiomyoma, polypoid fibroleiomyoma, and hemangiosarcoma), congenital abnormalities (hydrocolpos with imperforate hymen and structural abnormalities with XX sex reversal), and a foreign body-associated urethrovaginal fistula.3,8,10,13,17,18
There has been a lack of research on normal CT features for the uterus and vagina and typical CT features that can distinguish between the uterine and vaginal masses in bitches. In addition, no studies have been reported on whether CT scans can distinguish between a large vaginal mass and a uterine mass when a large vaginal mass occupies the caudal abdomen. Therefore, we hypothesized that the CT appearance for diameter, attenuation, and contrast enhancement patterns are different between the uterus and vagina in dogs. In this study, CT features of the uterus and vagina were evaluated in clinically normal bitches according to estrus versus anestrus phases of the reproductive cycle and in bitches with large vaginal masses. The study aimed to present the typical CT features of the uterus and vagina during estrus versus anestrus and to assess whether the origin of large vaginal masses could be determined based on CT findings compared with surgical findings.
Materials and Methods
The study was approved by the Institutional Animal Care and Use Committee of Chonnam National University. The animals were cared for in accordance with the Chonnam National University Guidelines for Animal Experiments (CNU IACUC-YB-2018-64).
Prospective, experimental study in clinically normal bitches
Animals
Seven, sexually intact female, purpose-bred Beagles were used. The median age was 12 months (9 to 14 months), and the median weight was 8.5 kg (7.2 to 9.9 kg). The bitches were clinically healthy based on physical examination, complete blood count, serum biochemistry, electrolyte analysis, urinalysis, abdominal radiography, and ultrasonography results.
Determination of the phases of the estrus cycle
The phase of the estrus cycle was determined using physical examination, particularly the degree of vulvar edema and quality of discharge, and vaginal smears with Diff-Quick staining (Sysmex). The anestrus was determined with no evidence of vulvar edema and discharge on physical examination and predominant small parabasal cells in vaginal cytology. For determining the estrus, vaginal smears were performed at 2-day intervals at the onset of vulvar swelling and serosanguineous vaginal discharge, and estrus was determined with predominant large superficial cells and pyknotic or anuclear nuclei.19
CT study
In each dog, 6 CT scans were performed; in anestrus, pre- and postcontrast CT using intravenous contrast injection and CT vaginography, and in estrus, pre- and postcontrast CT using intravenous contrast injection and CT vaginography. After fasting for 6 hours, each dog received a 250-mL polyethylene glycol solution (Colyte; Taejoon Pharmaceutical Co) per os to empty the colon. Twelve hours later, general anesthesia was induced using intravenous injection of a combination of 0.75 mL/kg zolazepam hydrochloride-tiletamine hydrochloride (Zoletil; Virbac) and 0.03 mg/kg medetomidine hydrochloride (Domitor; Orion Corporation). Anesthesia was maintained using isoflurane (Terrell; Piramal) (1% to 2%) and oxygen (1 L/min). A 6-French balloon catheter (Silicone Foley Balloon Catheter; Sewoon Medical) was inserted into the urinary bladder to maintain emptying during the CT scan.
With the dog in sternal recumbency, CT was performed using a 16-row multidetector CT scanner (Siemens Emotion 16) with the following settings: slice thickness, 1 mm; pitch, 0.8; rotation duration, 600 ms; tube voltage, 120 kV; and tube current, 120 mA. Then, a test bolus was performed at the aortic bifurcation after injection of 0.5 mL/kg iohexol (Omnipaque 300; GE Healthcare; 300 mg iodine/mL) at a rate of 3 mL/s using a power injector (LF OptiVantage; Liebel-Flarsheim Company LCC) through a 22-G cephalic vein catheter. Based on time-to-attenuation curve results, the arterial phase scan delay was determined as the time to 15% of peak enhancement of the aorta; the venous-phase scan delay was the time to peak enhancement of the caudal vena cava. Dual-phase contrast CT images were acquired after an intravenous injection of 2 mL/kg iohexol at 3 mL/s.
CT vaginography was performed immediately after the dual-phase contrast CT study, without changing the dog’s position. After a 12-French balloon catheter was placed into the vestibule and the tip of the catheter was inflated with saline, the vulva was clamped with atraumatic forceps to prevent contrast leakage.20,21 A 1:10 ratio of iohexol diluted with warm saline was used to avoid streaking artifacts and was slowly injected into the vagina at a dose of 1 mL/kg.20,21 CT vaginography images were then acquired. Complications associated with vaginography, such as a vaginal tear or rupture, were monitored for 1 week based on clinical signs (e.g., abdominal pain, vomiting, and anorexia).
Image analyses
Sagittal and dorsal plane reconstructions were performed in 1-mm slice thickness with 1-mm intervals. Images were processed in a soft tissue algorithm and viewed at a window width of 400 Hounsfield units (HU) and a window level of 40 HU. All CT images were evaluated at a workstation by 2 veterinarians (JWJ and EL) with 3 and 2 years of radiology experience, who were blinded to the reproductive phase of the subject. The CT images of each dog were presented to the veterinarians in random order.
The cervix was localized based on the site of insertion of the uterine artery in the uterus on the contrast CT images, confirming with CT vaginography. The vagina receives its blood supply from the vaginal artery, which arises from the internal pudendal artery.22 The main blood supply to the uterus comes from a branch of the vaginal artery and the uterine artery. The cervix was identified on the arterial phase CT images where the uterine artery embeds itself in the uterus, and the position of the cervix was confirmed on CT vaginography. On CT vaginography images, the cervix was located where the contrast filling terminated from the vaginal lumen based on the previous radiographic study using positive contrast vaginography.20,21 Subsequently, the vagina was localized approximately 2 cm caudal to the cervix; the uterine body was localized from the cervix to the uterine bifurcation; the uterine horns were localized from the uterine bifurcation to the ovary. CT images were assessed from 5 parts, including the vagina, uterine cervix, uterine body, and left and right uterine horns. After equally dividing the total length of each part, CT evaluation was performed at the 2 sites from each part; at one-third and two-third points, respectively. Quantitative and qualitative assessments of the CT images were performed (Table 1). The diameter was measured on the transverse image, and multiplanar reconstructions were performed to remain perpendicular to the axis of the structures measured to avoid volume averaging.
Quantitative and qualitative assessments of CT images.
| Evaluation factors | CT images | CT plane | Evaluation methods |
|---|---|---|---|
| Quantitative assessment | |||
| Diameter | Postcontrast venous phase images | Transverse | Diameter was measured in the long and short axes and the mean value was calculated. |
| Enhancement of reproductive tract | Pre- and postcontrast images | Transverse | Attenuation of the wall was measured by tracing the regions of interest over the entire layer, careful not to include intra- or extraluminal structures. |
| Vascular attenuation | Postcontrast arterial phase images | Transverse | Attenuation of the uterine artery was measured using a circle region of interest over the entire lumen beside the uterine cervix. |
| Qualitative assessment | |||
| Contrast enhancement pattern of the wall | Postcontrast arterial and venous phase images | Transverse, dorsal, and sagittal | Contrast enhancement pattern of the wall was classified into 3 types: (1) central enhancement; (2) peripheral enhancement; and (3) transmural enhancement (enhancement of the entire wall). |
| Vascular change | Postcontrast arterial phase images | Transverse, dorsal, and sagittal | The uterine artery from the uterine cervix to the uterine horn was evaluated in terms of conspicuity and tortuosity. |
Statistical analyses
Interobserver agreement for the quantitative assessment was evaluated by calculating intraclass correlation coefficient (ICC) values. The agreement was defined as excellent if the ICC was > 0.8, good if it was 0.61 to 0.79, moderate if it was 0.41 to 0.6, and poor if it was < 0.4. The diameter and degree of contrast enhancement between each part were evaluated using a 1-way analysis of variance and multiple comparisons using the Kruskal-Wallis test and Duncan correction. The quantitative assessments between estrus and anestrus were conducted using the Wilcoxon signed-rank test. Data are presented as mean ± standard deviations. Statistical analysis was performed by 1 veterinarian (JWJ) under the supervision of 1 statistician (JK), using a standard software (SPSS version 23.0; IBM Corp). P < .05 was considered statistically significant.
Retrospective study of bitches with large vaginal lesions
Five sexually intact female dogs with large vaginal lesions were selected from the database at Chonnam National University Veterinary Teaching Hospital and Helix Animal Medical Center from July 2013 to December 2019. The inclusion criteria for the selection of the patients were (1) dogs undergoing abdominal radiography, ultrasonography, pre- and postcontrast CT, and surgery; (2) bitches with large masses in the caudal abdomen; (3) bitches with mass origin not determined using radiography and ultrasonography; (4) bitches with masses confirmed as vaginal origin based on surgery; and (5) bitches with masses confirmed through histopathologic examination.
The origin of the lesions was determined by 1 radiologist (JC) and 1 veterinarian (JWJ) based on vascularity, size, and contrast enhancement on CT images. The results were compared to surgical findings.
Results
Normal bitches
All CT images clearly revealed the uterine horn, body, cervix, and vagina in the normal dogs, and CT vaginography filled the vaginal lumen without any complications. Regardless of the phase of the estrus cycle, the uterine artery was clearly visible beside the cervix and the cervix could be localized based on the site of insertion of the uterine artery and CT vaginography in all bitches (Figure 1). CT vaginography revealed the cervix as a ventral crescent-shaped area filled with contrast agent.
Localization of the cervix in arterial phase CT study (A and B) and CT vaginography (C and D) in a normal dog. In the transverse plane of the arterial phase CT images (A), both sides of uterine arteries (white arrowheads) are visible beside the cervix (white asterisk). In a 10-mm slab maximum intensity projection (B), the uterine artery (white arrowhead) branches from the vaginal artery (black arrowhead), internal pudendal artery (short arrow), and internal iliac artery (long arrow), serially. On CT vaginography, the vaginal lumen (asterisk) is uniformly filled with a contrast agent to the level of the cervix (black arrow) dorsal to the urinary bladder (U) on the sagittal plane (C). There is a crescent-shaped, contrast agent filling (black arrow) at the cervix on the transverse plane (D). The left side of the image is the cranial part of the animal on the sagittal plane and the right side on the transverse plane, respectively.
Citation: American Journal of Veterinary Research 84, 1; 10.2460/ajvr.22.03.0052
The cervix (anestrus, 9.3 ± 1.0 mm; estrus, 15.8 ± 2.3 mm) was thicker than the uterine body (6.5 ± 1.5 mm; 9.7 ± 2.1 mm) and uterine horns (5.0 ± 1.14 mm; 7.0 ± 0.6 mm) (P < .05) and was similar to the vagina (8.4 ± 1.0 mm; 15.4 ± 2.2 mm), regardless of whether estrus or anestrus was present. The diameter of all parts of the uterus and vagina was markedly more increased during estrus than anestrus (P < .05).
Regardless of the scan phase or phase of the estrus cycle, the uterine horns, body, and cervix were enhanced more than the vagina (Table 2). In estrus, the degree of contrast enhancement was significantly increased compared to anestrus on the arterial phase in the uterine cervix (P = .046) and on both scan phases in the uterine horns and body (P < .05); however, there was no change in the vagina.
Attenuation (Hounsfield unit) of each reproductive tract according to estrus or anestrus and scan phases.
| Arterial phase | Venous phase | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Estrus cycle | Vagina | Cervix | Body | Left horn | Right horn | Vagina | Cervix | Body | Left horn | Right horn |
| Anestrus | 69.9 | 82.6 | 78.6 | 70.3 | 71.6 | 74.7 | 91.0 | 92.6 | 94.5 | 92.5 |
| Estrus | 77.2 | 119.0 | 134.7 | 108.1 | 111.2 | 84.2 | 111.0 | 137.5 | 125.8 | 127.9 |
The uterus and vagina had diverse contrast enhancement patterns, based on estrus versus anestrus (Figure 2). During anestrus, the transmural enhancement pattern was found in all parts; in 100% of the vagina, 85% of the uterine body, and 64% of the uterine horns. The enhancement pattern was more varied during estrus (transmural [57%] and central [42%] patterns in the vagina; central [57%)] and peripheral patterns [35%] in the uterine body; and peripheral pattern [89%] in the uterine horns). In the cervix, the central enhancement pattern was dominant during both anestrus (64%) and estrus (92%). During estrus, strongly enhanced layers were present in the uterine wall. The layers surrounded the circumference of the caudal part of the cervix and were subsequently continuous to the uterine body and horns in 6 of the 7 dogs. The cervical opening appeared to be folded into the vaginal lumen with intraluminal fluid in 4 of the 7 dogs. During estrus, the uterine artery was tortuous and dilated and had higher attenuation (290 ± 66 HU) than during anestrus (165.7 ± 39.6 HU). It was easily identified beside the cervix and followed to the uterine horns in all bitches.
Contrast-enhanced patterns of the vagina (A and B), cervix (C and D), uterine body (E and F), and uterine horn (G and H) during anestrus (A, C, E, and G) and estrus (B, D, F, and H) on the transverse plane of the venous-phase CT images. The vagina (white arrow) shows a transmural enhancement pattern with slight enhancement during anestrus (A) and estrus (B). The cervix (white arrowhead) shows a central pattern during anestrus (C) and estrus (D); the pattern is more distinct during estrus than anestrus. The uterine body (white arrow) and uterine horn (white arrowhead) show transmural patterns during anestrus (E and G); central and peripheral patterns were present during estrus (F and H). The left side of the image is the right side of the animal.
Citation: American Journal of Veterinary Research 84, 1; 10.2460/ajvr.22.03.0052
The mean ICC values for diameter and attenuation of the uterine artery between the 2 reviewers were 0.92 (0.61 to 0.99) and 0.99, respectively, indicating good to excellent interobserver agreement.
Retrospective study
Of 5 bitches evaluated in the retrospective study, 3 had large vaginal tumors (leiomyoma, leiomyosarcoma, and adenocarcinoma), and 4 had cystic lesions (hematocolpos with vaginal obstruction and Gartner’s duct cyst). All were small breed dogs (3 Yorkshire Terriers and 2 Maltese Terriers). The median body weight was 3.6 kg (2.3to 6 kg), and the median age was 9.8 years (8 to 11 years). The dogs presented with abdominal distension (n = 4), dysuria (3), and dyschezia and tenesmus (1). A large mass was palpated in the caudal abdomen in all dogs.
Lateral radiographs revealed a soft tissue opacity mass from the caudal abdomen between the descending colon and urinary bladder. The mass was continuous to the pelvic cavity in 3 bitches (dogs 1, 2, and 5). In 2 bitches (dogs 3 and 4), the mass was suspected to originate from the uterus rather than the vagina, because the caudal margin of the mass was cranial to the pelvic cavity.
Ultrasonography revealed that the lesions were heterogeneous masses with (dog 3) or without (dogs 1 and 2) blood flow as determined by Doppler interrogation or a cystic structure surrounded by a thin membrane containing echogenic (dog 4) or anechoic (dog 5) fluid. Both the ovaries and uterine horns were separate from the lesions in all bitches. However, the anatomical relationships between the lesion and the uterine body, cervix, and vagina could not be determined due to the large size of the mass extending into the pelvic cavity. In dog 4, ultrasound-guided aspiration of the cyst was attempted to obtain samples and for the relief of dysuria. Because of its high viscosity, only a small amount of dark red fluid was aspirated.
CT examination was performed to determine the exact origin of the masses and to plan surgeries (Supplement Table 1). In all bitches, the blood vessels around the uterus and vagina were severely distorted or displaced by the mass, so it was difficult to localize the cervix based on the uterine artery. The cervix was localized based on CT features, a thick, moderately enhancing region (105 to 124 HU) with a central enhancement pattern. The lesions caudal to the cervix were determined as vaginal origin in 4 bitches (dogs 1, 2, 4, and 5). In dog 1 and dog 2, the CT images revealed the lesions were eccentric mural vaginal tumors. Based on surgery (ovariohysterectomy (dog 1) and partial vaginectomy (dog 2) and histology results, they were confirmed as intramural vaginal leiomyoma originating from the left-sided vaginal wall and vaginal leiomyosarcoma, respectively (Figure 3). In dog 3, the intrapelvic mass appeared to have broad contact with the vagina (7.5 mm). However, based on the CT images, it was unclear whether the mass originated from the uterus or the cranial part of the vagina, because the cervix, uterine body, and right uterine horn were effaced by the mass due to its large size. Based on an exploratory laparotomy, the mass was confirmed as a vaginal-origin adenocarcinoma adhered to the urinary bladder, rectum, and ureter.
Postcontrast CT images of dog 1 with vaginal leiomyoma and dog 2 with vaginal leiomyosarcoma. In dog 1 (A–C), the cervix (arrowheads) with central enhancement is on the right, cranial to the mass (asterisk) and the pelvic part of a mass is extensively attached to the left wall of the vagina (arrows), without a clear margin between the mass and vaginal wall. In dog 2 (D and F), the mass (asterisk) is enhanced heterogeneously (118 HU) with a low-attenuation center (38 HU) and has peripheral feeding vessels. The cervix (arrowheads) is cranial to the mass. The mass was diagnosed as originating from the vagina. A distended right uterine horn (F, white arrow) with intraluminal fluid was present. The left side of the image is the right side on the transverse (A and D) and dorsal (B and E) planes, and the cranial part of the animal on the sagittal plane (C and F), respectively.
Citation: American Journal of Veterinary Research 84, 1; 10.2460/ajvr.22.03.0052
In dog 4 and dog 5, the significantly large cystic structures caudal to the cervix were observed on CT images (Figure 4). In dog 4, the structure was suspected to be a dilated vagina because it was continuous to the pelvic cavity, without separation from the vaginal lumen. The ratio of the vestibulovaginal junction width to the maximal vaginal width was 0.2, indicating that vestibulovaginal stenosis was present (vestibulovaginal ratio: < 0.20, severe; 0.20 to 0.25, moderate; 0.26 to 0.35, mild).5,10 Hematocolpos secondary to vestibulovaginal stenosis was diagnosed based on CT and fluid sample findings. A digital vaginal examination revealed a narrowed lumen of the vestibulovaginal junction (cingulum septum), with complete obstruction between the vagina and vestibule. A partial vaginectomy and ovariohysterectomy were performed, and hematocolpos secondary to vestibulovaginal stenosis was found during surgery. In dog 5, a large cyst caudal to the cervix occupied the pelvic cavity. A fluid-filled vagina was found separately on the right side of the cyst. CT vaginography was performed to assess the communication between the cyst and vaginal lumen. The vaginography images did not reveal an opening between the cyst and the vaginal lumen. However, the HU of the cystic fluid was increased from 12 HU to 47 HU on CT vaginography, which indicated there was communication between the cyst and the vaginal lumen. A Gartner’s duct cyst within the paravaginal space was tentatively diagnosed and was confirmed using laparotomy.
Postcontrast CT images of dog 4 with a hematocolpos with vaginal stenosis (A, B) and dog 5 with Gartner’s duct cyst (C, D). In dog 4, a large cystic structure (asterisk) is present from the caudal abdomen to the pelvic cavity. This cyst continues cranially with the cervix (arrowhead, A, B) and uterine body (arrow, A). In the pelvic cavity, the vagina is not found separate from the cystic structure. The cyst was diagnosed as a distended vagina, based on CT images. In dog 5, transverse (C) and dorsal (D) plane images indicate a thin-walled cyst containing a large amount of fluid is located left and caudal to the urinary bladder (U). A normal cervix (arrowhead, D) and left uterine horn (arrow, D) are cranial to the cyst. Sagittal (E) and dorsal (F) plane CT vaginography images indicate the vaginal lumen (arrowheads, E) is homogeneously filled with a contrast agent, is separate from the cyst, and continues to the level of the cervix (long arrow, F). The left side of the image is the cranial part of the animal on the sagittal plane and the right side on the transverse and dorsal planes, respectively.
Citation: American Journal of Veterinary Research 84, 1; 10.2460/ajvr.22.03.0052
Discussion
The uterus and vagina were investigated in terms of diameter, attenuation, and contrast enhancement pattern to identify characteristic CT features to differentiate between the vagina and uterus in normal bitches and bitches with large vaginal masses. The cervix was identified from other uterine parts based on diameter and enhancement pattern, and the origin of mass was determined. Regardless of estrus or anestrus, the vagina was thicker than the uterine body and horns and had a central enhancement pattern. The CT features of the cervix could be used to determine the origin of the vaginal lesions in 4 dogs.
Vascular anatomy landmarks are commonly used for localization of the organ and to determine lesion origin.23,24 The uterine artery was readily identified on the postcontrast CT images of normal bitches. The artery was then used to localize the cervix and distinguish the uterus from the vagina. However, the uterine artery could not be used for localization of the cervix when distortion by the vaginal masses was present.
CT vaginography was also used for localization of the cervix. The termination of the contrast-filled vaginal lumen had a crescent shape because the contrast agent entered the vaginal fornix, a cranioventral recess of the cervix. These typical features were found in normal dogs and dog 5 with Gartner’s duct cyst.
The uterus and vagina have different histological structures. The uterus comprises 3 layers: the endometrium, myometrium, and perimetrium. The vagina comprises 4 layers: the mucosa, propria submucosa, muscularis, and adventitia or serosa.25 These histologic differences between the 2 organs could be related to differences in CT features in terms of diameter, attenuation, and contrast enhancement patterns.
The vagina and cervix were significantly thicker than the uterine body and horns, regardless of estrus versus anestrus. This result was consistent with the results of a previous study using ultrasound.6 During estrus, all parts of the uterus and vagina were significantly thicker than during anestrus. During proestrus, estradiol stimulates histomorphologic changes in the walls of the uterus and vagina.22,25–27 The endometrium and myometrium of the uterus begin to thicken from edematous change, congestion, and muscle cell hypertrophy as the progesterone level increases during early estrus; maximum thickness occurs during estrus and early diestrus.22,25,27,28 Subsequently, the changes in the uterus and vagina regress to complete atrophy during anestrus.
CT appearances of the normal uterus and cervix were assessed in humans. Various contrast enhancement patterns were found depending on age, menstrual status, and imaging phase (delay after injection of the contrast material bolus).16 Similarly, in our study, the increase in contrast enhancement according to estrus versus anestrus was marked in the uterus. The uterus receives dual arterial supplies via the ovarian and uterine arteries.22 A study28 using ultrasound Doppler examination revealed that the uterine artery had a large blood flow volume during proestrus and estrus, but the pulsed-wave Doppler pattern indicated the presence of high resistance during anestrus and diestrus. This greater blood supply of the uterine artery due to the estrogen effect may increase the uterine enhancement during estrus compared with anestrus.
Meanwhile, in this study, contrast enhancement of the vagina revealed only mild estrus cycle-related changes. The contrast enhancement of the vagina found in this study may be related to the histopathologic change of the vagina in which the loose connective tissue in the vaginal epithelium changes into the collagen-rich stroma and the muscle bundle thickens according to the estrus cycles.2 There were differences in contrast enhancement in the uterus versus the vagina in normal dogs. The vagina and all parts of the uterus except the cervix had a transmural pattern during anestrus. During estrus, the vagina had a transmural enhancement pattern, but central and peripheral enhancement patterns were commonly found in the uterine body and uterine horns, respectively. During estrus, stratum vasculare, collections of blood vessels between the inner and outer layers of the myometrium in the uterus, develop as the estradiol levels increase. Glandular hypertrophy within the endometrium and fluid of the uterine lumen is apparent histologically.25,26 These histologic changes may contribute to the peripheral enhancement of the uterine body and horns and unenhanced intraluminal fluid on CT images. The cervix showed central enhancement regardless of estrus versus anestrus. Circumferential enhancement of the cervix into the vaginal lumen was also present during estrus. In the cervix, the arteries and veins are distributed inside the lamina propria, below the superficial epithelium.27 This distribution may account for the central enhancement of the cervix seen in the CT images. During estrus, the caudal part of the cervix, which appears in the vaginal lumen, may contribute to the circumferential enhancement of the cervix in the CT images. This result is comparable to the bull’s eye or fusiform appearance with multiple layers found on ultrasound.29
When a space-occupying lesion is located between the descending colon and urinary bladder and urethra, a uterine body or cervix origin is prioritized.1 However, in our study, when the vaginal lesion was large enough to extend from the caudal abdomen to the pelvic canal, radiography and ultrasonography had limited ability to determine the origin of the lesion. These results suggested that uterine-origin and vaginal-origin lesions should be included in the differential list for lesions located between the descending colon and urinary bladder/urethra.
This study had some limitations. First, normal CT features were only assessed in young Beagles, so further study is required to investigate differences in CT features according to age, breed, and the number of pregnancies and parturition. Second, normal CT enhancement patterns were only assessed during anestrus and estrus. Further study is required to investigate the enhancement patterns which could differ during proestrus and diestrus. Third, the capability of CT to distinguish the origin of a vaginal lesion was investigated in only a small number of dogs because we only included bitches with large vaginal masses and a workup that included sequential radiography, ultrasonography, CT, and surgery. Fourth, volume averaging may increase the measured diameter and affect the CT attenuation values, particularly when the tube of interest is not perpendicular to the scan plane.
This study described CT features of the uterine horns, body, cervix, and vagina in terms of diameter, attenuation, and enhancement pattern according to estrus versus anestrus. CT features that could be used to differentiate the uterus and vagina were also assessed. The uterine cervix could be localized by contrast enhancement pattern, anatomical location of the uterine artery, and CT vaginography. However, in bitches with large vaginal lesions, vascular distortion and displacement by these lesions limited the use of the uterine artery as a landmark for determination of lesion origin. The CT images revealed the cervix as a focal thick region with central enhancement, regardless of estrus versus anestrus in the normal bitches and most of the bitches with vaginal lesions. CT vaginography facilitates the localization of the cervix and enables the identification of communications between cystic vaginal lesions and the vaginal lumen.
The results of this study indicated that when a bitch has a large lesion in the caudal abdomen, not only the uterus but also the vagina should be suspected as the possible organs of origin. Using CT, the cervix can be localized based on a typical enhancement pattern and focal thickening and the vagina can be distinguished from the uterus.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
This research was supported by the Animal Medical Institute of Seoul National University and Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Science, ICT, and Future Planning (NRF-2021R1A2C200573011).
The authors declare that there were no conflicts of interest.
References
- 1.↑
Patsikas M, Papazoglou LG, Jakovljevic S, et al. Radiographic and ultrasonographic findings of uterine neoplasms in nine dogs. J Am Anim Hosp Assoc. 2014;50(5):330–337. doi:10.5326/JAAHA-MS-6130
- 2.↑
Nelissen P, White RA. Subtotal vaginectomy for management of extensive vaginal disease in 11 dogs. Vet Surg. 2012;41(4):495–500. doi:10.1111/j.1532-950X.2011.00948.x
- 3.↑
Weissman A, Jiménez D, Torres B, Cornell K, Holmes SP. Canine vaginal leiomyoma diagnosed by CT vaginourethrography. J Am Anim Hosp Assoc. 2013;49(6):394–397. doi:10.5326/JAAHA-MS-5922
- 4.↑
Manothaiudom K, Johnston SD. Approach to vaginal/vestibular masses in the bitch. Vet Clin North Am Small Anim Pract. 1991;21(3):509–521. doi:10.1016/S0195-5616(91)50057-7
- 5.↑
Kieves NR, Novo RE, Martin RB. Vaginal resection and anastomosis for treatment of vestibulovaginal stenosis in 4 dogs with recurrent urinary tract infections. J Am Vet Med Assoc. 2011;239(7):972–980. doi:10.2460/javma.239.7.972
- 6.↑
Davidson AP, Baker TW. Reproductive ultrasound of the bitch and queen. Top Companion Anim. Med 2009;24(2):55–63. doi:10.1053/j.tcam.2008.11.002
- 7.↑
Pretzer SD. Clinical presentation of canine pyometra and mucometra: a review. Theriogenology. 2008;70(3):359–363. doi:10.1016/j.theriogenology.2008.04.028
- 8.↑
Yoon H, Han SH, Kim J, Kim K, Eom K. Urogenital anomalies and urinary incontinence in an English Cocker Spaniel dog with XX sex reversal. J Vet Intern Med. 2018;32(3):1166–1171. doi:10.1111/jvim.15119
- 9.
Kim HJ, Kim JK, Choi JH, et al. A Gartner duct cyst of the vagina causing dysuria and dyschezia in a Yorkshire Terrier. J Vet Sci. 2007;8(4):427–429. doi:10.4142/jvs.2007.8.4.427
- 10.↑
Blackford Winders CL, Tobias KM. Perineal protrusion secondary to imperforate hymen and hydrocolpos in an 8-year-old spayed female dog. Case Rep Vet Med. 2016;2016:8067967. doi:10.1155/2016/8067967
- 11.↑
Kim H, Lee K, An J, Lee S, Jeong H, Lee T, Han J, Cho Y. Feasibility of non-transabdominal ultrasonographic approach for canine rectum and anal canal imaging. Poster presented at: the 2019 European Veterinary Diagnostic Imaging (EVDI) Congress; August 21–24, 2019; Basel, Switzerland. Assessed October 15, 2019. https://onlinelibrary.wiley.com/doi/epdf/10.1111/vru.12818
- 12.↑
Thacher C, Bradley R. Vulvar and vaginal tumors in the dog: a retrospective study. J Am Anim Hosp Assoc. 1983;183(6):690–692.
- 13.↑
Wenzlow N, Tivers M, Selmic L, Scurrell EJ, Baines SJ, Smith KC. Haemangiosarcoma in the uterine remnant of a spayed female dog. J Small Anim Pract. 2009;50(9):488–491. doi:10.1111/j.1748-5827.2009.00769.x
- 14.↑
Russo M, England GCW, Catone G, Marino G. Imaging of canine neoplastic reproductive disorders. Animals. 2021;11(5):1213. doi:10.3390/ani11051213
- 15.↑
Yitta S, Hecht EM, Slywotzky CM, Bennett GL. Added value of multiplanar reformation in the multidetector CT evaluation of the female pelvis: a pictorial review. Radiographics. 2009;29(7):1987–2003. doi:10.1148/rg.297095710
- 16.↑
Yitta S, Hecht EM, Mautner EV, Bennett GL. Normal or abnormal? Demystifying uterine and cervical contrast enhancement at multidetector CT. Radiographics. 2011;31(3):647–661. doi:10.1148/rg.313105057
- 17.↑
Agut A, Carrillo JD, Anson A, Belda E, Soler M. Imaging diagnosis—urethrovaginal fistula caused by a migrating grass awn in the vagina. Vet Radiol Ultrasound. 2016;57(3):E30–E33. doi:10.1111/vru.12311
- 18.↑
Barozzi MB, Saba CF, Gendron KP. CT characteristics of uterine and vaginal mesenchymal tumours in dogs. J Small Anim Pract. 2021;62(4):293–299. doi:10.1111/jsap.13293
- 20.↑
Kyles AE, Vaden S, Hardie EM, Stone EA. Vestibulovaginal stenosis in dogs: 18 cases (1987–1995). J Am Vet Med Assoc. 1996;209(3):1889–1893.
- 21.↑
Root MV, Johnston SD, Johnston GR. Vaginal septa in dogs: 15 cases (1983–1992). J Am Vet Med Assoc. 1995;206(1):56–58.
- 22.↑
Evans HE, De Lahunta A. The urogenital system. In: Evans HE, De Lahunta A, eds. Miller’s Anatomy of the Dog. 4th ed. Elsevier; 2013:361–405.
- 23.↑
Upile T, Jerjes W, Nouraei SAR, Singh SU, Kafas P, Sandison A, Sudhoff H, Hopper C. The stylomastoid artery as an anatomical landmark to the facial nerve during parotid surgery: a clinico-anatomic study. World J Surg Oncol. 2009;28:71. doi:10.1186/1477-7819-7-71.
- 24.↑
Pagani E, Tursi M, Lorenzi C, Tarducci A, Bruno B, Mondino ECB, Zanatta R. Ultrasonographic features of adrenal gland lesions in dogs can aid in diagnosis. BMC Vet Res. 2016;12(1):267. doi: 10.1186/s12917-016-0895-1.
- 25.↑
Rehm S, Stanislaus DJ, Williams AM. Estrous cycle-dependent histology and review of sex steroid receptor expression in dog reproductive tissues and mammary gland and associated hormone levels. Birth Defects Res B Dev Reprod Toxicol. 2007;80(3):233–245. doi:10.1002/bdrb.20121
- 26.↑
Eurell JA, Frappier BL. Female reproductive system. In: Dellmann’s Textbook of Veterinary Histology. 6th ed. John Wiley & Sons; 2013:256–278.
- 27.↑
Goericke-Pesch S, Schmidt B, Failing K, Wehrend A. Changes in the histomorphology of the canine cervix through the oestrous cycle. Theriogenology. 2010;74(6):1075–1081. doi:10.1016/j.theriogenology.2010.05.004
- 28.↑
Nogueira I, Almeida L, Angrimani D, Brito MM, Abreu RA, Vannucchi CI. Uterine haemodynamic, vascularization and blood pressure changes along the oestrous cycle in bitches. Reprod Domest Anim. 2017;52:52–57. doi:10.1111/rda.12859
- 29.↑
Mattoon JS, Nyland TG. Small Animal Diagnostic Ultrasound. 3th ed. Elsevier Health Sciences; 2014:634–653.
