Oral Presentations
CORRELATION BETWEEN AGE AND ULTRASONOGRAPHIC B‐MODE APPEARANCE OF CECUM IN 29 HEALTHY CATS
S. Faverzani, G. Barella, M. Lodi, A. Bertolazzi From the Department of Health, Animal Science and Food Safety, State University of Milan, Milan, Italy (Faverzani, Barella, Lodi, Bertolazzi).
Background:
The sonographic appearance of the cecum has been described previously in young cats. We believe that there may be different aspects of the cecum in healthy subjects of different ages.
Purpose:
The present study aims to describe the B‐mode ultrasonographic appearance of cecum in healthy cats and its correlation with age.
Methods:
Twenty‐nine clinically healthy cats underwent ultrasonographic evaluation (two‐dimensional ultrasound with a 7.5–13 MHz linear transducer). The cecum was identified in transverse and longitudinal section. Two different patterns of ultrasonographic appearance were identified: hypoechoic homogenous and heterogeneous with a nonuniform hyperechoic band parallel to the lumen. Cecal content was recorded and wall measured (in longitudinal section) in the thin portion (S1, next to the colic junction) and in the thicker portion (S2, cul‐the‐sac). Medium values and standard deviations were recorded. Correlation between age and wall thickness (age/S1 and age/S2) was evaluated with Pearson correlation test. Friedman test was performed to examine the relationship between S1 and S2. Correlation between age and ultrasonographic appearance was evaluated with a logistic regression (using age as a continuous variable and binary output).
Findings:
Cats ranged from 1 to 14 years old (SD = 0.6). The cecal wall appeared hypoechoic homogenous in 19 cats while heterogeneous in 10. Logistic regression showed a correlation between age and ultrasonographic appearance (older cats tend to have an heterogeneous cecal wall, P = 0.016). Cecum content was gassy (15 cats) or mucous (14 cats). Medium thickness value was 0.20 cm (ranging 0.13–0.34 cm, SD = 0.04) in S1 and 0.27 cm (ranging 0.16–0.38 cm, SD = 0.05) in S2. Values of S1 and S2 were statistically different (P = 0). No significant statistical correlations were found between age/S1, age/S2, and age/luminal content (P > 0.05).
Conclusion:
Normal ultrasonographic appearances of feline cecum and their correlation with age are described. The cecal wall tends to be hypoechoic and homogenous in young cats, while heterogeneous in older individuals. We hypothesize that heterogeneous appearance of cecal wall could be explained by “cecal tonsil” theory: with aging, the immunostimulation of the lymphatic structures that are plenty in the cecal wall could lead to echostructural modifications. However, our hypothesis should be confirmed by histopathological examination even if all cats in our study were healthy and remained asymptomatic for 6 months after the initial evaluation.
EFFECT OF SEDATION ON CONTRAST‐ENHANCED ULTRASONOGRAPHY OF THE SPLEEN IN HEALTHY DOGS
C. Fina, F. Rossi, E. Stock, K. Vanderperren, L. Duchateau, J. H. Saunders From the Department of Medical Imaging, Faculty of Veterinary Medicine of Ghent, Ghent, Belgium (Fina, Stock, Vanderperren, Saunders), Veterinary Clinic dell'Orologio, Sasso Marconi, Italy (Rossi), and Department of Physiology and Biometrics, Faculty of Veterinary Medicine of Ghent, Ghent, Belgium (Duchateau).
Background:
Contrast‐enhanced ultrasonography of the spleen has been used to characterize focal and diffuse splenic lesions in veterinary medicine. This procedure is usually obtained in dogs using manual restraint, nevertheless uncooperative patients may need sedation. Secondary changes in splenic perfusion associated with sedation may occur and must be considered.
Purpose:
The purpose of this study was to investigate the effect of two commonly used sedative agents, butorphanol and dexmedetomidine, on splenic perfusion during contrast‐enhanced ultrasound of the spleen in healthy dogs.
Methods:
The study was designed as a prospective crossover study. Six adult healthy Beagle dogs were scanned three times each over 1 week by the same operator, with three different protocols assigned in a random sequence: no sedation, butorphanol (Dolorex®, Intervet Schering‐Plough, 0.2 mg/kg IM) or dexmedetomidine (Dexdomitor®, Pfizer Animal Health, 500 μg/m2 IM). Similar experiment was repeated 2 weeks later. All dogs received a bolus of second‐generation contrast medium (SonoVue®, Bracco, 0.03 ml/kg IV). When the dogs were sedated, contrast‐enhanced ultrasonography was obtained 20 min after sedation. Images were recorded and time‐intensity curves were generated using a commercial software program (QLAB, Philips, AG) for quantitative computerized analysis. Following parameters were calculated: mean baseline intensity, peak intensity (PI), arrival time (AT), time to peak (TTP), wash‐in, wash‐out, and area under the curve (AUC). Analysis was based on a linear mixed model with dog as random effect and sequence and treatment as fixed effects.
Findings:
The perfusion pattern observed in nonsedated dogs and dogs sedated with butorphanol was similar and consisted in rapid enhancement of the splenic arteries arising from the hilus, heterogeneous followed by more homogeneous enhancement of the parenchyma, and finally gradual, slow decrease of enhancement during the wash‐out phase. When dexmedetomidine was used, the whole organ was diffusely hypoechoic during the first 30 s in all dogs and enhancement was subjectively estimated to be weak during the entire procedure. There were no significant differences for any of the perfusion parameter values between the nonsedated dogs and the dogs sedated with butorphanol. Sedation with dexmedetomidine significantly increased AT and TTP (P < 0.001), and significantly decreased PI, wash‐in, and AUC (P = 0.018, P = 0.0014, and P = 0.0421).
Conclusion:
These results indicate that sedation with butorphanol causes no significant change in splenic perfusion and can be used in uncooperative dogs for contrast‐enhanced ultrasound of the spleen. In contrast, dexmedetomidine modifies significantly splenic enhancement. Clinicians should consider the hemodynamic consequences and evaluate splenic hypoperfusion changes with caution when dexmedetomidine is used.
ULTRASONOGRAPHIC CHARACTERIZATION OF CHRONIC GASTRIC DILATION VOLVULUS IN DOGS: A REPORT OF SIX CASES
C. Monteiro From the Ecovet Ecografia Veterinaria Movel and Departamento Clinico, Faculdade de Med Veterinaria, Lisboa, Portugal (Monteiro).
Background:
Radiography has always been the elective modality of diagnostic imaging in the diagnosis of gastric dilation volvulus (GVD). Nowadays ultrasonography (US) is often the first diagnostic imaging examination performed in animals with gastrointestinal (GI) signs, especially when cost is a concern; therefore recognizing the US signs of chronic GVD (CGDV) is valuable.
Purpose:
The aim of this study is to characterize the ultrasonographic findings of CGDV in dogs.
Methods:
From 2012 to 2014, in a retrospective study, images were obtained in six dogs with nonspecific GI signs and indication for an abdominal US (GE Vivid E with a microconvex transducer 5–9 MHz and linear transducer 9–13 MHz). There were five males and one female from four large breed dogs and one mixed breed. The relevant criteria evaluated were: gastric dilation, thickening and layering of the gastric wall, motility evaluation, gastric content, pyloric and duodenal position, enlargement of the spleen, location, and signs of infarction.
Findings:
All animals had moderate (n = 5) to severe (n = 1) gastric dilation with a mixed content of fluid, ingesta, and gas (n = 6). The gastric wall was thickened (n = 4) and there was no loss of wall layering (n = 6). There was deviation of the pylorus and duodenum dorsal and laterally to the left of the medial line (n = 6). No opening on the angular incisure fold was communicating the body and the pyloric antrum, indicating compartmentalization (n = 1). No motility was observed (n = 6). Splenomegaly was without signs of infarction (n = 6). All dogs were afterward subjected to a radiograph study (right and left lateral), where signs of GDV were identified. Five dogs were taken to surgery for exploratory coeliotomy, three had a 90° rotation of the stomach, one had a 180° rotation, in these four gastropexy was carried through. One dog presented only gastric dilation at surgery. In one of the dogs the owner declined surgery and after 1 year the dog was still alive with intermittent GI signs.
Conclusion:
Acute GV usually has an acute onset, nonetheless chronic forms are described, with recurrent subtle and varied signs, including bouts of weight loss, borborygmus, flatulence, retching, eructation, or vomiting. In these cases there is moderate gas dilation of the stomach and considerable fluid accumulation, which makes US a valid technique to access the stomach and duodenum. Left and right lateral radiographs are the recommended imaging modality to diagnose CGDV, but US findings, especially pyloric and duodenal displacement, are extremely valuable signs to suspect CGDV.
ULTRASOUND OF LYMPHATIC TISSUE IN THE SMALL INTESTINE OF YOUNG CATS
T. Hjorth, L. Lindstrom, M. Uhlhorn, K. Hansson From the Section of Diagnostic Imaging, University Animal Hospital, Swedish University of Agricultural Sciences (Hjorth, Uhlhorn), Department of Biomedical Sciences and Veterinary Public Health, Swedish University of Agricultural Sciences (Lindstrom), and Department of Clinical Sciences, Swedish University of Agricultural Sciences, Uppsala, Sweden (Hansson).
Background:
The normal ultrasonographic appearance of the feline small intestine has been described and shown to have good correlation with histological layers. We have noted a previously unreported, asymmetrically positioned hypoechoic layer in the submucosa of the distal jejunum and ileum.
Purpose:
The first goal was to histologically characterize the asymmetrically positioned hypoechoic layer. The second goal was to describe the prevalence and ultrasonographic features of the layer in a population of clinically healthy young cats.
Methods:
Three cats, 4–36 months old, were euthanized for disease not related to the gastrointestinal tract. In these cats an asymmetrically positioned hypoechoic layer in the distal part of the jejunum and ileum was found. Samples for histopathology were obtained from these areas. In the second part of the study, high frequency (8–18 MHz) ultrasonographic examination of the intestinal tract in 20 clinically healthy young cats was performed to document the presence of the layer. The length and thickness of the layer and intestinal wall thickness with and without the layer was measured. Measurements are expressed as mean or mean ± SD.
Findings:
Histopathology from the distal jejunum and ileum in three cats showed that the hypoechoic layer represents asymmetrically positioned normal lymphatic tissue in the submucosa. In the second part of the study, a hypoechoic layer in the submucosa was identified in the distal part of the jejunum and ileum in all 20 cats. In addition, a similar layer could also be seen in a more proximal segment of the jejunum in 10 (50%) of the cats. The layer could not be seen in the duodenum in any cat. The length of the layer in the jejunum measured 1.5 ± 0.7 cm and 4.8 ±1.6 cm in the distal jejunum–ileum. The thickness of the layer was 1.0 mm in both jejunum and ileum. The total thickness of the ileum wall including the hypoechoic layer was 2.9 mm, in comparison with the opposite part of the wall, without the hypoechoic layer, that measured 1.8 mm.
Conclusion:
Normal lymphatic tissue in the submucosa can be seen with ultrasound and is a common finding in young cats.
RADIOGRAPHIC AND ULTRASONOGRAPHIC APPEARANCE OF LESIONS ASSOCIATED WITH FELINE INFECTIOUS PERITONITIS: 44 CASES
C. Bonnissent, G. Benchekroun, E. Reyes Gomez, S. Le Poder, P. Pey From the Department of Internal Medicine (Bonnissent, Benchekroun), Department of Anatomical‐Pathology (Reyes Gomez), UMR 1161, INRA, ENVA, ANSES (Le Poder), and Department of Medical Imaging, Alfort Veterinary Teaching Hospital, Paris East University, National Veterinary School of Alfort, Maisons‐Alfort, France (Pey).
Background:
FIP is a viral disease that produces pyogranuloma and vasculitis in various organs. The antemortem diagnosis is often based on cumulative results rather than a single, definitive, simple test result. Radiographic and ultrasound findings can be helpful in this context.
Purpose:
The aim of this retrospective study was to characterize the lesions associated with FIP on thoracic radiographs and abdominal ultrasound. To our knowledge, the only investigating study on ultrasonography was inconclusive and no study describes thoracic radiographs (Lewis et al., 2010 JAAHA). The purpose was to find features/combination of imaging findings suggestive of FIP.
Methods:
Forty‐five cases were included from a review of medical records from the National Veterinary School of Alfort between 1997 and 2014. FIP diagnosis was based either on pathology (postmortem examination followed by histopathology) (n = 12) or reverse transcriptase polymerase chain reaction (PCR) detection of feline coronavirus virus from effusion, ultrasound‐guided fine needle aspiration, or cerebrospinal fluid (n = 33); seven cases had the diagnosis confirmed with both procedures. Thoracic radiographs or abdominal ultrasounds were performed in all cases. The radiologist and first author reviewed and correlated images with gross lesions together with pathologist and internist. The virologist performed PCR diagnosis.
Findings:
On thoracic radiographs (n = 25), the most common lesions were pleural effusion (n = 14), cardiomegaly (n = 10), and sternal lymphadenopathy (n = 8). Three cases with cardiomegaly had a postmortem examination or echocardiography (n = 2). An asymmetrically thickened interventricular septum or a left hypertrophic cardiomyopathy was observed on echocardiography and pyogranulomas had been seen on the interventricular septum during the postmortem examination. On abdominal ultrasound, an abdominal effusion was the most common finding (n = 24). Other common abdominal findings were isolated lymphadenopathy, mainly involving mesenteric lymph nodes (n = 17), bilateral nephromegaly (n = 16), splenomegaly (n = 15), hyperechoic renal cortex (n = 14), hypoechoic liver (n = 12), generalized lymphadenopathy (n = 10), generalized steatitis (n = 10). When present, abdominal effusion was frequently associated with hepatomegaly and hypoechoic liver (n = 12/24). In the absence of an effusion, hepatomegaly and liver hypoechogenicity were rarely observed. In cats without abdominal effusion (n = 13), on ultrasonography, the most frequent findings were nephromegaly (n = 8) and hyperechoic renal cortex (n = 6). On abdominal ultrasound, renal pyogranulomas created a hypoechoic notch along the subcapsular vascularization. On thoracic radiographs, pyogranulomas appeared as focal pleural thickening and irregularity or small ill‐defined perivascular/peribronchial nodules creating a micronodular interstitial lung pattern.
Conclusion:
This study makes an exhaustive list of FIP lesions on radiographs and ultrasound. While the lesions described are not pathognomonic, a combination of these findings is suggestive of FIP. When these lesions are detected, fine needle aspiration of abdominal organs for PCR should be considered. Cardiomegaly, sternal lymphadenopathy, or pleural effusions are other lesions that would be consistent with FIP. The description of these findings on thoracic radiographs or abdominal ultrasound may help the antemortem diagnosis of FIP.
COMPARISON OF ABDOMINAL ULTRASONOGRAPHY AND RADIOGRAPHY IN THE INVESTIGATION OF FELINE ABDOMINAL DISEASE: A RETROSPECTIVE STUDY OF 105 CASES AT THE UNIVERSITY OF GEORGIA
W. Won, A. Sharma, W. Wenbo From the College of Veterinary Medicine (Won, Sharma) and Statistical Consulting Center, University of Georgia, Athens, GA (Wenbo).
Background:
The investigation of feline abdominal disease commonly uses radiography and ultrasonography. Selection of one or both diagnostic tests should be based on the specific clinical question and confidence in test results. Additionally, the test results should provide information leading to a diagnosis and guide patient care. Performing a test with low diagnostic yield is inefficient and can result in increased cost to the client, longer inpatient hospital stay, and strains on hospital resources, including but not limited to personnel and time. At the University of Georgia, the current policy requires abdominal radiography prior to ultrasonography when performing diagnostic imaging for abdominal disease.
Purpose:
The purpose of this retrospective study was to compare the clinical diagnostic benefit between abdominal radiography and ultrasonography in the investigation of feline abdominal disease.
Methods:
One hundred and five feline cases were selected based on the criteria of both orthogonal abdominal radiography and ultrasonography examinations being completed during the same admission period. The cases were selected over a 2‐year time period. Inclusion criteria were based on feline patients presenting for signs consistent with abdominal disease and undergoing both radiography and ultrasonography for investigation of the presenting clinical signs. Cases for reevaluation of a previously diagnosed disease were excluded from the study.
Findings:
In this study, ultrasonography provided a tentative or definitive diagnosis in 62/105 cases (59.05%) compared to radiography (27/105, 25.7%). Of these, 25/89 (23.81%) were diagnosed both by ultrasonography and radiography. The cases that were diagnosed by ultrasonography alone were 37 (35.24%). The cases diagnosed only by radiography alone were 2 (1.90%). Forty‐one cases (39.05%) were neither diagnosed by ultrasonography nor radiography. These cases were diagnosed by other means such as serum chemistries, blood tests, MRI, necropsy, or other tests. Our statistics show that ultrasonography gave additional clinically relevant information in 80 cases (76.19%). Using a paired t‐test between the comparison of diagnoses made by ultrasonography vs. radiography, the study determined that the number of cases diagnosed by ultrasonography were significantly higher (P‐value < 0.0001).
Conclusion:
In the investigation of feline abdominal disease, ultrasonography provided more clinically relevant diagnostic information that was useful for guiding future diagnostic and therapeutic interventions in patient management. The results of this study suggest that ultrasonography may be used as the initial diagnostic imaging modality for evaluating feline abdominal disease.
DIAGNOSTIC ACCURACY OF RADIOLOGIC MEASUREMENTS OF DOGS AND CATS: A SYSTEMATIC REVIEW
C. R. Lamb, J. R. Nelson From the Department of Clinical Sciences and Services, The Royal Veterinary College, London, UK (Lamb, Nelson).
Background:
Studies describing measurement of structures in diagnostic images are published frequently. For example, of 52 original investigations published in Veterinary Radiology & Ultrasound in 2013, 13 (25%) reported measurements of normal animals and five (10%) described use of measurements as the basis for a diagnostic test.
Purpose:
The aim of the present study was to systematically review studies about diagnostic accuracy of radiologic measurements described for dogs and cats in order to reach generalizable conclusions about the value of making such measurements.
Methods:
Search was done using the ISI Web of Knowledge (Thomson Reuters) from inception to May 2013 in the subject category veterinary sciences. Search criteria were (1) within the title, terms veterinary or canine or feline or equine or dog or cat or horse; (2) also within the title, terms imaging or ultraso* or sono* or echo* or radiograph* or X‐ray or CT or MR or magnetic resonance or computed tomograph*; (3) within topic, terms measure* or size or thick* or quanti* or diagnos*. Studies were eligible for inclusion that employed length, angle, area, or volume measurements from radiographic, ultrasonographic, CT, or MR images as a diagnostic test; compared results of imaging with a reference standard; included at least 10 dogs or cats as subjects; and included sufficient data that a 2 × 2 table of results could be constructed. For each study, the total number of subjects, number of true positives, true negatives, false positives, and false negatives were recorded. Sensitivity, specificity, positive (PLR) and negative (NLR) likelihood ratios, and their respective binomial 95% confidence intervals were calculated. Methodological quality of studies was assessed using the QUADAS‐2 tool.
Findings:
The 4264 papers retrieved by initial search were reduced to 244 on the basis of title, to 43 on the basis of abstract, and to 26 by full review. These 26 studies described 40 tests that satisfied the inclusion criteria. Tests were radiographic in 22 (55%) instances and ultrasonographic in 18 (45%). Median (range) number of subjects was 64 (20–305), sensitivity was 77% (38–99%), specificity was 82% (50–99%), PLR was 4.1 (1–103), and NLR was 0.29 (0.01–1). Methodological quality of studies was generally low, with high risks of bias in patient selection in 95% studies and performance of the index test in 90% studies. Many authors overemphasized the clinical value of measurements by ignoring the accuracy of concurrent subjective image interpretation. The two studies that compared accuracy of radiologic measurements to subjective image interpretation alone found no difference.
Conclusion:
Evidence is weak that radiologic measurements are useful for diagnosis. Although measurements may have value in a description, they should not be emphasized as a basis for diagnosis either in teaching or clinical imaging reports.
DOPPLER WAVEFORM OF DUCTUS VENOSUS IN CANINE FETUSES: CORRELATION WITH NEONATAL MORTALITY
G. Barella, D. Groppetti, M. Faustini, M. Lodi, S. Faverzani, A. Pecile From the Department of Health, Animal Science and Food Safety (Barella) and Department of Veterinary Science and Public Health, State University of Milan, Milan, Italy (Groppetti, Faustini, Lodi, Faverzani, Pecile).
Background:
The Doppler ultrasonographic evaluation of the ductus venous (DV) blood flow has not been described in dogs yet. The evaluation of DV blood flow is considered in human medicine an important indicator of fetal wellbeing.
Purpose:
The present study aims to define if the Doppler ultrasonographic morphology of DV waveform can predict neonatal mortality in dogs.
Methods:
Forty‐six clinically healthy pregnant bitches underwent ultrasonographic evaluation (B‐mode and Doppler technique with a 7.5–13 MHz linear transducer). The bitch with the lowest gestational age was 36 days pregnant and the one with the highest was 66 days pregnant. Two fetuses for each pregnancy (except in case of singleton) were considered: the most cranially and the most caudally into the uterus. The wellbeing of the canine fetus was checked ultrasonographically evaluating the fetal movements and heartbeat. The DV appears as a short trumpet‐shaped vessel with a sparkling appearance, arising from the umbilical vein and entering the caudal vena cava. A sequence of at least three successive and symmetric waves was acquired using pulsed‐wave Doppler. Waveforms were classified as diphasic or threephasic. Odds ratios (OR) were calculated with 95% confidence intervals (CI), to compare the chance of neonatal mortality in litters with one or more DV threephasic waveforms with litters without threephasic waveforms.
Findings:
Eighty‐one of the 191 fetuses were evaluated. Neonatal mortality (death during the first 6 h after birth) recorded was 10%. Eighteen fetuses belonged to litters with neonatal mortality, in which threephasic waveform was observed in pregnancy. Six fetuses belonged to litters without neonatal mortality, in which threephasic waveform was observed in pregnancy. Forty‐nine fetuses belonged to litters without neonatal mortality, in which only diphasic waveform was observed in pregnancy. Eight fetuses belonged to litters with neonatal mortality, in which only diphasic waveform was observed in pregnancy. The correlation between the presence of a threephasic waveform and neonatal mortality was significant (OR, 18.38; 95% CI, 5.6–60.3, P < 0.0001). Considering only singleton or twin fetuses, 100% of fetuses were evaluated (27 fetuses from 19 bitches). Two of these 27 fetuses presented both threephasic waveform and neonatal mortality; the remaining 25 showed diphasic waveform without neonatal mortality. The correlation between the presence of a threephasic waveform and neonatal mortality was significant (OR, 255.00; 95% CI, 4.10–15,849.21, P = 0.0085).
Conclusion:
The evaluation of the DV during pregnancy to detect the presence of a threephasic waveform could be used to predict neonatal mortality in dogs. Our results appear to be very similar to those reported in human medicine, but further studies are needed.
QUANTITATIVE CONTRAST ULTRASOUND IMAGING OF BLADDER TUMORS IN DOGS: PRELIMINARY DATA
R. E. Pollard, K. Watson, D. Kruse, E. Ingham, K. Ferrara From the Department of Surgical and Radiological Sciences (Pollard) and Department of Biomedical Engineering, University of Davis, Davis, CA (Watson, Kruse, Ingham, Ferrara).
Background:
Lower urinary tract tumors are common in dogs. Surgical removal is often impossible making chemotherapy the standard treatment with animals surviving 6–12 months after diagnosis.
Purpose:
The purpose of this prospective study was to use contrast‐enhanced cadence‐contrast pulse sequence (CPS) ultrasound to noninvasively quantify microvessel density (MVD) in dogs with lower urinary tract tumors undergoing chemotherapy. We hypothesized that ultrasound estimates of MVD would remain stable with stable disease, increase with progressive disease, and decrease with remission.
Methods:
Dogs with a cellular diagnosis of transitional cell carcinoma and an ultrasonographically identifiable bladder or prostatic mass were eligible. Dogs were imaged prior to chemotherapeutic induction and at subsequent chemotherapy appointments. Ultrasound contrast material (Definity) was injected IV as a bolus (dogs & 8804; 20 kgs 0.1 mls; dogs > 20 kgs 0.2 mls). CPS video clips were obtained for 20 s. A region of interest was drawn around the tumor at peak enhancement. The percentage of enhanced pixels was calculated to estimate MVD.
Findings:
Six dogs with eight tumors were imaged on 51 occasions. Elevations in CPS estimates of MVD were associated with infectious cystitis (4/8 tumors). The mean (±standard deviation) percent change in estimated MVD for dogs with progressive disease was 14.7 ± 81.8, stable disease was 6.9 ± 67.3, and partial remission was 39.3 ± 67.7. Statistical significance was not achieved (P > 0.05).
Conclusion:
More data are necessary to determine if CPS estimates of MVD can predict chemotherapeutic response in dogs with transitional cell carcinoma of the lower urinary tract.
USE OF CONTRAST‐ENHANCED ULTRASONOGRAPHY IN NONCARDIAC THORACIC DISORDERS OF DOGS AND CATS
N. Linta, M. Baron Toaldo, M. Cipone, G. Bettini, A. Cordella, M. Quinci, V. Galli, A. Diana From the Department of Veterinary Medical Science, University of Bologna, Bologna, Italy (Linta, Baron Toaldo, Cipone, Bettini, Cordella, Quinci, Diana) and Borghesiana Private Veterinary Clinic, Rome, Italy (Galli).
Background:
Contrast enhancement ultrasonography (CEUS) has been employed for the diagnostic evaluation and guided biopsy of noncardiac thoracic disorders in humans. To the best of our knowledge, information regarding the use of CEUS for the evaluation of intrathoracic disorders has not been reported in the veterinary literature.
Purpose:
The first aim of this study was to evaluate the usefulness of CEUS in the diagnostic work‐up of thoracic noncardiac disorders of small animals and to describe the CEUS pattern of pulmonary, mediastinal, and thoracic wall lesions. The second aim was to assess the usefulness of CEUS as a direct guide for sample procedures.
Methods:
The clinical records of dogs and cats with thoracic disorders of noncardiac origin were retrospectively reviewed. All animals underwent thoracic radiography and B‐mode ultrasonography followed by CEUS, using a second‐generation contrast agent (Sonovue®). Cytological or histological confirmation was obtained. The distribution of contrast medium within the lesions was employed as a guide for cytological or biopsy sampling. All images and video clips were reviewed by the same experienced operator and the evaluation of contrast medium distribution was recorded for each lesion.
Findings:
Thirty‐four animals, 21 dogs, and 13 cats were included in the study. Thoracic disorders included 18 pulmonary lesions (primary carcinoma [10], lymphoma [1], sarcoma [1], abscess [1], and pneumonia [5]), 14 mediastinal lesions (lymphoma [6], thymoma [3], lipoma [1], mesothelioma [1], melanoma [1], metastatic carcinoma [1], and abscess [1]), and two thoracic wall lesions (osteosarcoma [2]). All neoplastic pulmonary lesions showed an inhomogeneous distribution of contrast medium, whereas four inflammatory lesions had homogenous distribution with typical pulmonary vessels ramification. All mediastinal lymphomas showed a homogeneous distribution pattern. The lung and mediastinal abscesses had peripheral enhancement of the wall with an avascular center. Finally, the thoracic wall lesions had an inhomogeneous distribution of contrast medium. All cytological and biopsy samples obtained after CEUS were diagnostic.
Conclusion:
Due to the lack of veterinary studies concerning CEUS of thoracic lesions, we compared our results to human literature. Contrast enhancement ultrasonography patterns of our pulmonary and mediastinal lesions were very similar to those previously reported in humans. In our cases, CEUS was poorly useful for the differentiation between benign and malignant thoracic tumors with the exception of mediastinal lymphoma showing a perfusion pattern similar to that observed in other organs. Tumors often contain necrotic areas, and biopsies from these regions will be nondiagnostic. Biopsy obtained after ultrasound contrast injection may help to define necrotic areas from viable tissue. On the basis of our preliminary results, the use of ultrasonographic contrast medium can be recommended for improving the diagnostic usefulness of cytology and biopsy sampling.
ULTRASOUND IMAGING IN CHIARI‐LIKE MALFORMATION SCREENING IN CKCSs
G. Spattini, C. Anselmi From the Clinica Veterinaria Castellarano, Reggio Emilia, Italy (Spattini, Anselmi).
Background:
Chiari‐like malformation (CM) is a neurological disease complex that implies foramen magnum dysplasia and various degrees of cerebellar vermis herniation. This could cause an alteration in the cerebrospinal fluid flow and common consequences are syringomyelia (SM) and/or hydrocephalus. The exact cause of CM has not been established, but there appears to be an association with reduced caudal cranial fossa size. In Cavalier King Charles Spaniels (CKCS), the condition has moderately high heritability. In literature 1, cervical radiology, ultrasound, computed tomography, and magnetic resonance are the imaging techniques involved in the CM study. MRI is the gold standard, allowing the visualization of cerebellar hernia and SM.
Purpose:
The purpose of this preliminary study is to evaluate which parameters, normally assessed by MRI, are measurable by ultrasound in the assessment of CKCSs potentially affected by CM.
Methods:
Two adult CKCSs with clinical signs consistent with CM were examined by ultrasound using the technique described in literature. A microconvex 8–11 MHz probe was used.
Findings:
In both patients, the cerebellum and the protrusion of the caudoventral portion through the foramen magnum were visualized. Furthermore, in both subjects a part of the spinal cord was examined and a central canal dilation of 2.4 mm and 2.8 mm, respectively, was observed. The spinal cord could be visualized up to the level of the C1‐C2 intervertebral disc space.
Conclusion:
In both patients, it was possible to evaluate some of the parameters usually assessed by MRI and used in CKCS breeding programs to reduce the incidence of CM/SM. Many breeders and most owners do not have the economic possibility of MRI examination to check for CM/SM. The changes proposed in literature 2 between different MRI scans during the natural progression of SM in time are the increase of the maximal syrinx width, height of the foramen magnum, length of cerebellar herniation, and caudal cranial fossa volume. Ultrasound examination could assess some of these measures in an economical and noninvasive way. However, this examination, when compared to the MRI, has significant limits. The most important are: limitations on the length of the cervical spine visualization and subjective assessment of the degree of cerebellum herniation. Further prospective and comparative studies are needed to assess if ultrasonography could be used as a first screening for CM in CKCS.
RENAL LESION DETECTION USING THREE‐DIMENSIONAL ULTRASONOGRAPHY IN CATS
M. Molazem, Y. Vali From the Department of Surgery and Radiology, Faculty of Veterinary Medicine, University of Tehran, Tehran, Iran (Molazem, Vali).
Background:
Three‐dimensional (3D) ultrasonography has become a realistic diagnostic imaging modality but still requires more investigation regarding its clinical applications.
Purpose:
The purpose of this study is to evaluate the potential of 3D ultrasonography in renal lesion visualization and diagnosis.
Methods:
A total of 10 cats with kidney disorders (hydronephrosis, polycystic, glomerulonephritis, acute tubular necrosis, and end stage renal disease) prospectively underwent two‐dimensional (2D) gray scale and power Doppler ultrasonography, 3D gray scale and power Doppler ultrasonography with use of the virtual organ computer‐aided analysis (VOCAL) imaging and vascularization histogram software program. The raw data were reconstructed and analyzed in the software blindly by the sonographer. The final subjective and objective results were compared with the cytological findings taken by fine needle aspiration.
Findings:
In detection of hydronephrosis and polycystic kidneys, 2D and 3D ultrasonography performed similarly. However, there was better visualization of the pelvis volume, and the number and volume of the cysts and their pressure on the cortical vascularization were better assessed by 3D sonography and with the VOCAL program. Glomerulonephritis and acute tubular necrosis had dramatically higher vascularization flow indices and flow index in comparison with normal kidneys. The degree of volume and perfusion of the end stage kidney were apparently better analyzed using quantitative VOCAL imaging and histogram program.
Conclusion:
Three‐dimensional ultrasonography and the VOCAL program seem to allow better anatomical visualization and internal tissue lesion detection in abnormal kidneys in cats.
PANCREATIC ECHOGENICITY IN DOGS WITH HYPERADRENOCORTICISM
A. Granger, L. Gaschen, T. Francis, M. Hilferty, M. Kearney From the Department of Veterinary Clinical Sciences, Louisiana State University, Baton Rouge, LA (Granger, Gaschen, Francis, Hilferty, Kearney).
Background:
Pancreatic echogenicity can greatly aid the clinician in diagnosing pancreatic disease. Pancreatic hyperechogenicity is anecdotally reported as a normal variant. In people, pancreatic hyperechogenicity has been associated with elevated cortisol levels. Elevation of cortisol levels occurs in with canine hyperadrenocorticism.
Purpose:
The purpose of this study is to determine the prevalence of pancreatic hyperechogenicity in dogs diagnosed with hyperadrenocorticism. It is hypothesized that an enlarged, heterogenous or hyperechoic pancreas will be identified in dogs with hyperadrenocorticism compared with a healthy control group.
Methods:
The LSU Diagnostic Imaging database and medical records were searched for dogs having a confirmed diagnosis of hyperadrenocorticism and an abdominal ultrasound. The pancreatic echogenicity was determined as compared to the surrounding mesentery. Each ultrasonographic image of the pancreas was subjectively graded on echogenicity and homogeneity. A final designation was determined by the consensus of three reviewers. The cases with unclear images were excluded. A final statistical analysis was performed with Fisher's exact test using JMP version 10.0, SAS Institute, Carey, NC to compare for differences in echogenicity categories between dogs with hyperadrenocorticism and healthy controls.
Findings:
A total of 93 dogs with a diagnosis of hyperadrenocorticism were compared a group of 74 healthy control dogs. Results showed a 39% prevalence of a hyperechoic pancreas as compared to 7% in the control group. A homogeneously hyperechoic pancreas was statically greater (P < 0.0001) in dogs with hyperadrenocorticism compared to healthy controls. An isoechoic heterogeneous (P < 0.0001) was statistically greater in healthy controls compared to dogs with hyperadrenocorticism.
Conclusion:
Dogs with hyperadrenocorticism have a statistically greater prevalence of a homogeneous hyperechoic pancreas compared to controls. The hyperechogenicity may result from mineralization, fatty replacement, or fibrosis due elevated blood cortisol levels. Additional concurrent diseases affecting the pancreas might have affected echogenicity. The retrospective nature of this study did not allow for tissue sampling to confirm the histological cause of the hyperechogenicity. Age, breed, and body condition score were not considered but could have had an influence.
GENERATION OF 3D PRINTED MODELS FROM ANIMAL COMPUTED TOMOGRAPHY SCANS FOR CLINICAL AND TEACHING PURPOSES
T. Schwarz, C. Bell, R. Collins, S. Rhind From the Royal (Dick) School of Veterinary Studies (Schwarz, Bell, Rhind) and Edinburgh College of Art, The University of Edinburgh, Edinburgh, UK (Collins).
Background:
Additive manufacturing or three‐dimensional (3D) printing is increasingly popular to generate models for clinical and teaching purposes. The 3D data sets and file formats necessary for 3D printing can be generated from CT scans providing high anatomic detail opening up many possibilities.
Purpose:
The aim of this project was to generate a series of 3D‐printed objects from CT‐scanned animal body parts for surgical planning and other procedures of veterinary patients and to improve the clinical skills of veterinary students enhancing spatial understanding.
Methods:
CT scans of body parts of a variety of species as well as postmortem specimen were used. To ensure high‐quality 3D prints, only helical CT scans with a slice width between 1 mm and 3 mm, overlapping image reconstruction interval were used. Medium and high frequency reconstruction algorithms were applied. DICOM images were loaded onto a professional DICOM software (Osirix) and a surface‐rendering model was generated, allowing export as a 3D file, which was further modified with 3D‐modeling software (rhino). 3D printing was performed with a professional 3D printer (CatalystEx) with a layer resolution of 0.17 mm using thermoplastic material (acrylonitrile butadiene styrene).
Findings:
A total of 30 prints were generated, ranging in size between 5 cm and 30 cm. All exported 3D files required some further digital cleanup with 3D‐modeling software. One millimeter slice width DICOM series reconstructed with a medium frequency algorithm resulted in the smoothest surfaces and highest anatomical detail. Minified and magnified prints were also generated. Printing time ranged between 4 and 30 h. Clinicians and students appreciated prints with complicated 3D details most, such as angular limb deformities, nasal neoplasia, skull anomalies, anatomical details of the canine external and middle ear, and angular deformities of the temporomandibular joints. Surgical planning prints were printed twice, allowing creative destruction of one of them. Real size prints were found most helpful for clinical purposes, whereas magnified prints were helpful for teaching purposes to explain small anatomic details. For clinicians 3D prints were also found very valuable for client communication, allowing demonstration of planned surgical procedures and to demonstrate the use of advanced technology to improve patient care.
Conclusion:
Three‐dimensional printing is an exciting new technology that is enthusiastically embraced by all involved. The technology fits in very well with the portfolio of the veterinary radiologist. It is affordable and relatively easy to use and can improve our communication with clinicians, students, and clients. Due to the often long printing times, 3D printing lends itself best for planned procedures. CT images are well suited for generation of 3D files, as they provide high detail of internal and external anatomical surfaces. Three‐dimensional prints offer the possibility of real‐size fitting of orthopedic and other implants.
OPTIMIZATION OF COMPUTED TOMOGRAPHY EXPOSURE SETTINGS TO AID IN RAPID PROTOTYPING (3D PRINTING) OF BONE MODELS
R. Freed, R. o'brien, D. Schaeffer, S. Joslyn From the College of Veterinary Medicine, University of Illinois, Urbana, IL (Freed, O'Brien, Schaeffer, Joslyn).
Background:
An area of study in rapid prototyping (3D printing) that is largely uninvestigated is the optimization of CT exposure settings for accurate isolation of desired tissue, such as bones, particularly at the level of close articulation.
Purpose:
The aim of this project is to investigate the optimal CT exposure setting that will allow for the most accurate articular margin definition aiding in bone isolation for rapid prototyping.
Methods:
This study involved CT scanning of a cadaveric elbow joint. CT scans were performed using a 16‐slice helical CT scanner. Four different reconstructive kernels (detail, bone, bone+, and edge) and the following exposure settings were used: 80–140kVp in increments of 20 kVp, and 80–200 mAs in increments of 40 mAs. Sixty‐four data sets were obtained and imported into a DICOM viewer and then further processed using the segmentation software. Segmentation smoothing factors were selected (0.1–0.9 in increments of 0.2). Each segmentation process was evaluated based on its ability to stay confined to the radius.
Findings:
A logistic regression model was used to investigate the statistical relationship between specific CT setting parameters and the probability of obtaining an “acceptable” CT image. The authors found that, using the comparison of odds ratios bone+ reconstructive kernel was associated with the highest probability of successful segmentation when compared to the other kernels. Through the use of Type 3 Analysis, the authors have also found that increasing kVp is more important than increasing mAs when trying to achieve optimal contrast resolution at the articular margins. No statistical significance was found between mAs settings and contrast resolution. The logistic regression of bone+ revealed that high kVp and a smoothing factor of 0.3 will have the greatest probability for segmentation success.
Conclusion:
When imaging the joints of medium to large breed dogs, we recommend using CT exposure setting ≥140 kVp, a bone+ reconstructive kernel and a segmentation smoothing factor of 0.3 to generate the highest quality of images for rapid prototyping.
GET THEM EARLY… EXPERIENCE WITH A DIAGNOSTIC IMAGING BASED WIDENING PARTICIPATION PROGRAMME FOR PRIMARY SCHOOL CHILDREN
R. Weller, G. Kimble, S. B. Channon From the Diagnostic Imaging, The Royal Veterinary College, London, UK (Weller, Kimble, Channon).
Background:
Inspiring children from a wide range of background to consider studying science is part of the work done by RVC Access and Widening Participation (promoted and legislated by the government), with the aim of helping young people have a fair chance of applying to higher education. It is important that children and teenagers have the opportunity to encounter motivating examples of biomedical sciences not only at a stage when they may start to be thinking about future study options, but before then, particularly if they have few role models who have attended university in their own circles of family and friends.
Purpose:
It is important that children and teenagers have the opportunity to encounter motivating examples of biomedical sciences not only at a stage when they may start to be thinking about future study options, but before then, particularly if they have few role models who have attended university in their own circles of family and friends. With this in mind, we have created a 60‐min session to engage primary school children in biomedical sciences, comprising five 10‐min stations.
Methods:
The Tesco quiz: radiographs of everyday objects (fruit, cereals, pastries, chocolate bars, and electronic gadgets) were shown and the children matched these to the supplied objects.
“Heads, shoulders, knees and toes”—children matched the radiographs to their own body and displayed the skeletons of dogs and horses.
“Spot the fracture”: children identified fractures on radiographs and skeletons.
“Spot the difference”: computed tomographic images with obvious left‐right asymmetry were displayed and the children asked to spot the difference.
“Watch your muscles move”: an ultrasound machine was used to show children how their arm and leg muscle moved when they changed position. Each station was looked after by a veterinary undergraduate student as a facilitator. Each primary school student was given a mini radiograph on a string as a reward and a black piece of paper and a bit of chalk and was asked to draw a whole skeleton or bone as homework.
Findings:
“Smiley face” feedback was excellent and negative verbal feedback focused on the fact that they were not allowed to eat the chocolate bars (school policy).
Conclusion:
The developed session based on mini station caters to the short attention span of the targeted age group. By linking the familiar with the unfamiliar first, the students lost their original “fear of the unknown” and the playful nature of the session made it age appropriate but still encouraged students to think outside their usual “box.”
ULTRASONOGRAPHY AND COMPUTED TOMOGRAPHY FEATURES OF LYMPH NODES IN HEALTHY CATS
M. Tobon Restrepo, R. Novellas, Y. Espada, A. Aguilar, X. Moll, S. A. E. B. Boroffka From the Departament de Medicina i Cirurgia Animals, Universitat Autonoma de Barcelona, Bellaterra, Barcelona, Spain (Tobon Restrepo, Novellas, Espada, Aguilar A, Moll), GIVET, Universitaria Lasallista, Medellin, Colombia (Tobon Restrepo), and Division of Diagnostic Imaging, Faculty of Veterinary Imaging, Utrecht University, Utrecht, the Netherlands (Boroffka).
Background:
Lymph node (LN) characterization is important for diagnosis and prognosis of neoplastic and infectious diseases. Information regarding imaging features of the normal anatomy of feline LNs is limited. Normal values for size, appearance, and ability to depict many of them using computed tomography (CT) and ultrasonography (US) are needed.
Purpose:
The aims of the study were (1) to assess the ability of US and CT to identify feline LNs in healthy cats and (2) to compare the imaging features with measurements from normal anatomic references.
Methods:
Anatomical study: five fresh feline cadavers were used to dissect all LNs using previously reported anatomic landmarks. Length, width, and thickness were measured using a manual caliper. Imaging studies: healthy cats were enrolled for the study. Complete body CT was performed using a 16‐slice CT scanner. Images were acquired with soft tissue algorithms pre‐ and postcontrast administration. After CT, an ultrasonographic study was performed in order to identify all the peripheral, thoracic, and abdominal LNs using a 4–13 MHz linear array transducer. Sagittal and transverse images of identified LNs were stored. Long and short axis measurements were performed with an electronic caliper.
Findings:
Fifteen cats were included. A total of 54 different LNs were found and described in the anatomic study. Thirty‐nine (72.2%) and 47 (87%) LNs were identified with CT in pre‐ and postcontrast images, and 35(64%) were identified with US. At least one LN per lymph center was identified with both CT and US. Ultrasonographically, peripheral LNs were fusiform and slender in shape, and slightly hypoechoic, with the exception of the axillary and popliteal LNs that were more frequently hyperechoic and with regular margins. Thoracic and abdominal LNs were more rounded and elongated in shape, isoechoic to surrounding fat and also with regular margins. In the CT studies, LNs most frequently had regular margins, rounded to elongated shape, and a homogeneous structure before and after the contrast administration. In obese cats, LNs were more difficult to identify by US and their identification was more time consuming. However, an increase in body fat made easier the visibility of LNs in CT images.
Conclusion:
Contrast‐enhanced CT allowed the visualization of more LNs than noncontrast‐enhanced CT and US. A possible explanation is the difficulty to evaluate the LNs of the thoracic cavity in US examinations. A previous study reported a higher visibility of abdominal LNs in CT images of dogs with more intra‐abdominal fat; the same situation was seen in the cats of this study. US and CT features of peripheral, abdominal, and thoracic LNs in healthy cats are similar to those previously reported for healthy dogs.
COMPARISON OF COMPUTED TOMOGRAPHY AND RADIOGRAPHY FOR THE EVALUATION OF SMALL INTESTINAL OBSTRUCTION IN DOGS
W. T. Drost, E. M. Green, L. J. Zekas, G. G. Habing From the Veterinary Clinical Sciences, Diagnostic Imaging & Radiation Oncology Service (Drost, Green, Zekas) and Veterinary Preventive Medicine, The Ohio State University, Columbus, OH (Habing).
Background:
Vomiting, often caused by intestinal obstruction, is common in dogs. Initial screening for intestinal obstruction includes radiography; equivocal radiographic signs often necessitate reevaluation or additional imaging procedures. In people, CT for intestinal obstruction is highly sensitive, specific, and accurate.
Purpose:
To determine the sensitivity and specificity of abdominal CT for detection of canine intestinal obstruction. Compare the sensitivity and specificity of CT to abdominal radiographs for detection of canine intestinal obstruction.
Methods:
Twenty dogs were enrolled. Fifteen dogs with clinical signs of gastrointestinal disease and five dogs without clinical signs of gastrointestinal disease and normal intestinal radiographs. All dogs had digital three‐view abdominal radiographs and abdominal CT (pre‐ and postcontrast). Seventeen dogs had abdominal surgery for intestinal evaluation including all five dogs without clinical signs. Three dogs had follow‐up to determine that the clinical signs resolved. Three experienced radiologists reviewed the imaging studies and were asked their confidence level for intestinal obstruction on a five‐point scale and if they would recommend surgery. Sensitivity and specificity was calculated.
Findings:
For the 20 dogs of various breeds, the average age was 6.1 years (range, 4 months and 12 years). Nine male (five castrated) and 11 female (10 spayed) dogs. Eight dogs had surgically confirmed intestinal obstruction and 12 dogs did not have obstruction. Seven dogs had intestinal foreign bodies and one dog had an intussusception. Of the dogs without intestinal obstruction, two had histopathologically confirmed enteritis, one had ischemic necrosis, one had pancreatitis, two had colonic foreign bodies, and four had nonintestinal neoplasia. All three radiologists agreed on 14/20 (70%) of the radiographic studies and 16/20 (80%) of the CT studies. Using a consensus opinion of the three radiologists, abdominal radiographs had sensitivity = 87.5% and specificity = 75%, while abdominal CT had sensitivity = 87.5% and specificity = 91.67%. Two radiologists were slightly more confident in their CT findings compared to radiographic findings.
Conclusion:
For experienced radiologists, abdominal radiographs and abdominal CT are sensitive and specific for the detection of intestinal obstruction in dogs. This reemphasizes the use of abdominal radiographs as an initial screening test for dogs with possible intestinal obstruction. Further investigation could focus on which cases would benefit from an abdominal CT study vs. follow‐up radiographs and if results would be similar for less‐experienced image interpreters.
TRIPLE PHASE COMPUTED TOMOGRAPHY PERFUSION CHARACTERISTICS OF NON‐NEOPLASTIC AND NEOPLASTIC SPIROCERCOSIS‐INDUCED OESOPHAGEAL NODULES
R. Kirberger, N. Cassel, N. Stander, E. Dvir From the Department of Companion Animal Clinical Studies, Faculty of Veterinary Science, University of Pretoria, Onderstepoort, Republic of South Africa (Kirberger, Cassel, Stander, Dvir).
Background:
Spirocercosis in dogs results in a caudal esophageal nodule that may undergo neoplastic transformation to a sarcoma over time.
Purpose:
The objective of this prospective study was to determine if computed tomography angiography (CTA), particularly nodule perfusion characteristics, could distinguish between non‐neoplastic and neoplastic nodules in a cohort of dogs with spirocercosis‐associated caudal esophageal nodules.
Methods:
Thirty‐eight dogs with spirocercosis were prospectively recruited. Dogs were classified into neoplastic (n = 15) or non‐neoplastic (n = 23) nodule groups based on tissue biopsy, postmortem histology, or response to treatment. Pre‐ and postcontrast (2 ml/kg of iohexol 300 mg I/ml at 3 ml/s using a pressure injector) early arterial, late arterial, and venous CTA images were evaluated at 6, 35, and 65 s postinjection trigger, respectively. Regions of interest (ROIs) were taken of the largest possible normal esophageal wall as well as of nodules (excluding the wall) at three different transverse slices and mean Hounsfield units (HU) as well as standard deviations recorded and averaged. The ROIs were placed in the same position on the same slice in each phase. Necropurulent areas with no perfusion and with HU < 25 as well as areas of mineralization were excluded from the ROI. Additionally, a subjective percentage of the nodule/neoplastic mass that was suspected to be necropurulent material or mineralized were recorded.
Findings:
Non‐neoplastic nodules were smooth and nonmineralized with a higher proportion of hypoattenuating necropurulent cavities compared to neoplastic nodules that had a more irregular surface with 93% having mineralized foci and rarely any hypoattenuating pockets. Non‐neoplastic nodules were significantly more perfused than neoplastic nodules with the difference being up to 23 HU. The difference was most marked in the early and late arterial phases (P 0.0005 and 0.00005, respectively). Perfusion ratios of the normal esophagus to the neoplastic and non‐neoplastic nodules did not differ significantly from each other. The perfusion findings were indicative of relative hypoperfusion of the esophageal sarcomas.
Conclusion:
Mineralization, lack of necropurulent material, and relative hypoprefusion were indicative of neoplastic transformation making CTA a valuable non‐invasive technique to determine malignant transformation of spirocercosis‐induced esophageal nodules. The relative hypoperfusion of sarcomas in this study correlated with other human and animal studies, which showed that sarcomas are less perfused than carcinomas.
COMPUTED TOMOGRAPHIC FEATURES OF PHARYNGEAL NEOPLASIA IN DOGS: A RETROSPECTIVE STUDY OF 25 CASES
G. Carozzi, A. Zotti, M. Alberti, F. Rossi From the Clinica Veterinaria dell'Orologio, Sasso Marconi, Italy (Carozzi, Rossi), Department of Animal Medicine, Production and Health, Faculty of Veterinary Science, University of Padova, Legnaro, Italy (Zotti), and Ospedale Veterinario I Portoni Rossi, Zola Predosa, Italy (Alberti).
Background:
Pharyngeal neoplasia is a challenge for the radiologist; one of the crucial aspects is to define exact origin and extension. Computed tomography (CT) is widely used to investigate head and neck cancer; however no exhaustive information about imaging of pharyngeal neoplasia is available in the veterinary literature.
Purpose:
The aim of this study was to describe CT features of pharyngeal neoplasia with further purpose to identify possible differences between included histotypes.
Methods:
CT findings of 25 dogs affected by pharyngeal neoplasia were reviewed. Location of lesions, size and shape, margins, relationship with surrounding structures and vessels, attenuation characteristics and enhancement pattern, regional lymph nodes changes, and presence of metastasis were recorded.
Findings:
The caseload included 15 carcinomas (seven tonsillar squamous cell carcinomas, two nontonsillar squamous cell carcinomas, three tonsillar carcinomas, and three adenocarcinomas); five sarcomas (two histiocytic, one undifferentiated, one extra‐skeletal osteosarcoma, and one fibrosarcoma); four melanomas, and one lymphoma. Oropharynx and laryngopharynx were more frequently involved. Lesions in the different groups were of similar size, with irregular shape, ill‐defined margins, and moderate to marked heterogeneous contrast enhancement. Lysis of hyoid bones was recorded in two carcinomas and infiltration of lingual artery in one case. Marked medial retropharyngeal lymphoadenomegaly was recorded in 10/13 carcinomas, in all sarcomas and in 2/4 melanoma. The single lymphoma case showed both medial retropharyngeal and mandibular lymphadenomegaly. Lung metastases were found in 2/5 sarcomas and 2/4 melanomas.
Conclusion:
In this study, most of CT findings were overlapping in the different groups of pharyngeal neoplasia. CT is precise to determine mass extension, lymph nodes involvement, and distant metastatic spread. However, it seems not to be able to suggest histotype and biopsy is always required for a final diagnosis.
COMPUTED TOMOGRAPHIC CHARACTERISTICS OF CANINE DONTOGENIC CYSTS
A. Destri, S. A. Janssens, L. F. H. Theyse, S. A. E. B. Boroffka From the Division of Diagnostic Imaging (Destri, Boroffka) and Division of Orthopaedic Surgery, Faculty of Veterinary Medicine, Utrecht, The Netherlands (Janssens, Theyse).
Background:
Odontogenic cysts (OC) are fluid‐filled cavities in the jaw bones originating from odontogenic epithelium. In veterinary literature there are few studies on canine OC and to the knowledge of the author no study describing CT features in canine OC exists.
Purpose:
The purpose of this retrospective study was to analyze CT features of OC for detailed presurgical evaluation and for differentiation from odontogenic neoplasms. In human medicine few reports show that both CT and cone beam CT images allows better understanding for localization, cortical destruction, and association with surrounding anatomic structures as compared to radiographs.
Methods:
In this study, 14 dogs with 18 OC lesions with a presurgical CT examination and a clinical diagnosed lesion compatible with OC during surgery and/or histopathology were included. Patient signalment and CT features were evaluated: mono‐ or bilateral, number of lesions, location, size on maximal diameter, changes concerning the teeth, features of bone lesions (lysis, cortical bone thickening or thinning), contents of the cystic lesions (fluid, soft tissue, mineral, or tooth material) by measuring the Hounsfield units in pre‐ and postcontrast.
Findings:
Nine of 14 cases were confirmed during surgery. In 5/14 cases, a histological confirmation was necessary. Four of five cases confirmed by histopathology were dentigerous cyst and 1/5 presented inflammatory characteristics. Most represented breed was the boxer (7/14 cases) and 9/14 were male. The medium age at diagnosis was 44.3 months. The distribution was nine mandibular, five maxillary, one premaxillary, one zygomatic and maxillary, and one had involved both premaxilla and rostral part of the mandibula. Bone lysis was present in 12 cases, cortical bone thinning in eight cases, cortical bone thickening in two cases, and no periosteal proliferation was observed. The OC contents included six cases fluid, two fluid and soft tissue, five soft tissue and mineral, two fluid and mineral, and three fluid, soft tissue, and mineral. There was ring contrast enhancement present in 7/8 OC and no general contrast enhancement was observed.
Conclusion:
Brachiocephalic dogs (boxer) were overrepresented in this study (50%) as described previously in the literature. Cyst‐like structures, associated with teeth and alveolar bone lysis are common features in both odontogenic tumor and OC. In a recent study 96% of odontogenic tumors showed general contrast enhancement, while in OC 88% ring contrast enhancement was observed and no contrast enhancement inside the cyst was seen in any case. In conclusion, with CT the precise extension of OC can be evaluated for surgical planning and contrast CT together with signalment may also help in differentiating an odontogenic tumor from an OC.
COMPUTED TOMOGRAPHY POSITIONING AND NASOPHARYNX ALLOMETRY IN DOGS AFFECTED BY BRACHYCEPHALIC AIRWAY SYNDROME
M. Longo, R. Stefano, M. Gobbetti, V. Bronzo, S. Modina, M. Di Giancamillo From the Department of Health, Animal Science and Food Safety (Longo, Stefano, Bronzo, Modina, Di Giancamillo) and Department of Veterinary Science and Public Health, Universita degli Studi di Milano, Milano, Italy (Gobbetti).
Background:
Brachycephalic dogs are often affected by the Brachycephalic Airway Syndrome (BAS). Literature is mainly concentrated on the size and shape of the soft palate compared to the clinical presentation, but CT features of the nasopharynx have been rarely discussed.
Purpose:
The aim of the work is to identify a standardized positioning of skull and neck in the CT evaluation of the upper airways and to compare skull measurements with functional and absolute volumetric measurements of the rostral nasopharynx, in brachycephalic dogs with different degrees of BAS.
Methods:
Twenty‐five brachycephalic dogs with BAS, 13 of them with nasal stertor on clinical presentation, were investigated. Each dog, intubated during general anesthesia, underwent upper airway CT in two different positions: with the skull elevated higher than the neck; with the skull and neck at the same level. Functional volume and absolute volume measurements of the rostral nasopharynx were obtained by a thresholding‐based segmentation algorithm. Volume of the mucosa was calculated as the difference between absolute and functional volume. To normalize differences in size, the volumes were statistically correlated (Shapiro–Wilk, Pearson tests) with the cranial index (CI), that is the ratio between cranial width and cranial length.
Findings:
Volumes of the nasopharynx appeared reduced when the skull was in an elevated position. The caudal nasopharynx showed an ellipsoid shape due to a reduced dorsoventral expansion, compared to a circular appearance when head and neck were placed at the same level. An inverse relationship between CI and functional volume was observed (P = 0.037) in all the patients regardless of their clinical signs, but no relationship between CI and absolute volume or volume of the mucosa were detected. Consequently, the reduction in the volume of air does not seem to be related to a redundant presence of soft tissues in this region, but primarily with a reduction of the space in which they lie. Interestingly, CI was the highest in pugs, suggesting that this value may represent a negative predictive value for the functional volume of the nasopharynx. Finally, unexpectedly, all the patients showed a prominent median septum localized at the level of the vomer, associated with an elliptic shape of the region.
Conclusion:
Due to the mobility of the soft palate, skull positioning seems to play a key role in the evaluation of the throat and positioning with the head and neck at the same level providing consistently higher volume measurements. Nasopharynx allometry suggests that the morphology of this region alone cannot be the only responsible factor for clinical presentation. The median nasal septum and the elliptical shape of rostral nasopharynx are likely to modify the column of air and create a more turbulent airflow, especially in dogs with the highest CI. Future perspectives will be oriented to the dynamic study of airflow in patients with BAS.
HYPOXIA DOSE PAINTING WITH HEAVY IONS
A. sØvik, E. Malinen From the Department of Companion Animal Clinical Medicine, University of Life Sciences (Søvik) and Department of Physics, University of Oslo, Oslo, Norway (Malinen).
Background:
Tumor hypoxia is a known cause of resistance to radiotherapy. Treatment strategies targeting hypoxia, for example hypoxia dose painting with megavoltage photons, may potentially improve treatment outcome. However, both physical and biological properties of the photons will limit the efficacy of such dose painting.
Purpose:
To investigate, through radiobiological modeling, whether hypoxia dose painting with heavy ions of intermediate to high linear energy transfer (LET) is expected to improve radiotherapy outcome compared to dose painting with photons of low LET.
Methods:
Dynamic contrast‐enhanced MR images of a spontaneous canine nasal osteosarcoma were used to create tentative images of tumor hypoxia. The images from this particular patient were chosen for the subsequent modeling as the tumor contained a relatively high hypoxic fraction with a complex spatial distribution, factors likely to make successful radiotherapy challenging. The tumor hypoxia images were used as input in a tumor control probability (TCP) model, incorporating the dependence of cellular radiation sensitivity on partial pressure of oxygen and LET. The output of this model was the optimal dose prescription map that would maximize TCP for a given radiation quality. The dependence of the dose maps on LET was investigated, and the resulting TCP was compared to that from conventional therapy. Replanning, that is modification of the treatment plan during therapy in response to changes in the spatial hypoxia distribution, was also considered, as hypoxia images taken before each treatment fraction with conventional radiotherapy were available for this patient.
Findings:
Heavy ions gave significantly higher TCP than photons for a given tumor dose, regardless of whether hypoxia dose painting or conventional therapy was employed. Hypoxia dose painting increased the TCP compared to conventional therapy, irrespective of LET. The optimal dose distribution became more homogeneous and the magnitude of the increase in TCP from dose painting was reduced with increasing LET. However, this reduction was only modest when LET was increased from low (0.1–5 keV/μm) to intermediate (10–30 keV/μm, typical of therapeutic carbon ions) values. The importance of replanning diminished with increased LET.
Conclusion:
Dose painting with particles such as carbon ions has the potential to increase TCP compared with photons. The resulting dose distributions for heavy ions beams are less heterogeneous than for photons, which is expected to increase ease of delivery as well as to make the treatment plans more robust against changes in tumor oxygenation during treatment.
RADIOTHERAPY OF THYMOMA IN RABBITS: OPTIMIZATION OF VMAT SETUP
M. Dolera, G. Mazza, G. Urso, L. Malfassi From the La Cittadina Fondazione Studi e Ricerche Veterinarie, Romanengo, CR, Italy (Dolera, Mazza, Urso), 2Radioterapia Oncologica, Ospedale Lodi‐Casalpusterlengo, Italy (Malfassi).
Background:
Radiotherapy is a treatment option for rabbit thymoma but toxicity is frequent.
Purpose:
The aim of this study was to define an optimized volumetric modulated arc therapy (VMAT) treatment plan. The hypothesis was that image‐guided radiation therapy (IGRT) with VMAT technique would allow to deliver higher doses to the target with better sparing of organs at risk (OARs) in comparison to 3D conformal radiotherapy.
Methods:
Fifteen client‐owned rabbits naturally suffering from thymoma were treated with LINAC‐based hypofractionated VMAT technique. Dose prescription was 40 Gy in six fractions. Treatment plans were computed with Monte Carlo statistic algorithm and Monaco 3.0 planning system, verified with Dosimetry Check system and Scansidose Delta4 system. Positioning was verified with cone beam CT. For all patients, a treatment consisting in a single 360° arc optimized over continuous dose rate variation was planned. For the prescription target volume (PTV) coverage, we considered the V95% and the V107% levels (the PTV volume receiving less than 95% and more than 107% of the dose prescription, respectively). The equivalent uniform dose (EUD) was evaluated and dose constraints for OARs were derived from literature. Throughout treatment, in according to the volumetric reduction of the tumor, patients underwent two further CT simulations and TPS were immediately renewed.
Findings:
All the plans fulfilled the PTV, OAR, and EUD constraints and no one was rejected. The mean follow‐up was 24 months. In all patients tumor disappeared (complete response) within 40 days after the treatment and at 1 and 2 years neither relapses nor major complications were observed. No morbidity or mortality events were reported in treated animals. Treatment schedule of 40 Gy in six fractions with appropriate plan setup is effective in disease control and showed better results in comparison with published data of 3D conformal radiotherapy or surgical therapy.
Conclusion:
IGRT with VMAT technique is suitable in rabbits suffering from thymoma. Considering the high anesthesiological and surgical risk connected with rabbit thymoma, VMAT radiotherapy technique could be an option also in resectable patients.
CANINE MENINGIOMAS: COMPARISON OF PALLIATIVE THERAPY, SURGERY, AND VMAT STEREOTACTIC FRAMELESS RADIOSURGERY
M. Dolera, L. Malfassi, S. Marcarini, M. Sala, G. Urso From the La Cittadina Fondazione Studi e Ricerche Veterinarie, Romanengo, CR, Italy (Dolera, Malfassi, Marcarini, Sala) and Radioterapia Oncologica, Ospedale Lodi‐Casalpusterlengo, Italy (Urso).
Background:
Meningiomas represent 50% of primary intracranial tumors in dogs. There are limited comparative studies regarding the various treatment modalities.
Purpose:
Aim of this study was to compare palliative therapy, surgery, and volumetric modulated arc therapy (VMAT) frameless stereotactic radiosurgery in canine meningiomas.
Methods:
Data were collected retrospectively from 198 dogs referred to one institution over a 5‐year period with histopathologically confirmed meningiomas or where MRI findings were consistent with meningioma. Dogs were grouped by anatomical site (supratentorial—E, infratentorial—T, spinal—S) and by therapeutic option (palliation—P, surgery—S, radiosurgery—R). Surgery goal was total tumor resection. LINAC‐based VMAT radiosurgery was performed in one to five fractions. Serial clinical and MRI examinations were conducted. Signalment, clinical signs, neuroanatomic tumor location, relapse specifics, adverse events, best response according to RECIST criteria, and overall survival (OS) time were evaluated. MRI findings of irradiated meningiomas were investigated, with particular relevance of volume, signal intensity, contrast enhancement, mass effect, brain edema. Overall survival estimates were calculated using Kaplan–Meier method and the differences between groups were compared using logrank analysis. Multivariate analysis was performed using Cox regression.
Findings:
Ninety‐one dogs (51 E, 33 T, 7 S) had been palliated, 69 dogs (33 E, 31 T, 5 S) had been treated with stereotactic radiosurgery, 38 dogs (32 E, 1 T, 5 S) with surgery. Overall survival in PE was 190 days, in PT 38 days, in PS 89 days, in RE 781 days, in RT 654 days, in RS 813 days, in SE 567 days, in ST 3 days, in SS 210 days. Variables predictive of OS were localization and therapy option. Two months after the radiotherapy, according to RECIST criteria in 89% of cases there was stable disease, in 7% partial response, in 4% complete response; 24 months after the radiotherapy, in 63% there was a partial response, in 28% stable disease, in 9% complete response. Follow‐up MRI examinations showed a signal increase of the tumor in TSE T2‐weighted pulse sequence, a reduction of contrast enhancement in 72%, a reduction of brain edema in 94% of irradiated dogs.
Conclusion:
Dogs suffering from meningiomas undergoing stereotactic radiosurgery had superior outcome compared with those treated by conventional surgery or palliation in the same cohort. Comparing our results with previous studies, OS of irradiated dogs by VMAT radiosurgery is better than that of dogs definitively irradiated by 3D conformal radiotherapy (5–12 months) or treated by surgery and adjuvant radiotherapy (15 months) or traditional LINAC radiosurgery performed with conical collimators (18 months). Volumetric modulated arc therapy frameless radiosurgery could represent the best treatment option for canine meningiomas, particularly in presence of comorbidity, high anesthesiological risk, or other medical contraindications.
DIAGNOSTIC VALUE OF LATEROLATERAL RADIOGRAPHS IN EQUINE CERVICAL ARTICULAR PROCESS JOINT OSTEOARTHRITIS
R. Wood, J. Mullard, M. Tambaschi, R. Weller From the Department of Veterinary Clinical Sciences, The Royal Veterinary College, North Mymms, Hertfordshire, UK (Wood, Mullard, Tambaschi, Weller).
Background:
Arthropathy of the articular process joints of the cervical region has been associated with a wide variety of clinical presentations. Laterolateral radiographs are the most frequently employed diagnostic test for articular process joint osteoarthritis (OA). However, the diagnostic value of laterolateral radiographs in the diagnosis of cervical articular process joint OA has not yet been established.
Purpose:
This study aimed to determine the value of laterolateral radiographs as a diagnostic test for articular process joint OA by comparing them against computed tomography (CT) and postmortem gross examination.
Methods:
Plain laterolateral radiographs and CT scans were taken for cervical articular process joints from 27 cadaver necks. Radiographs were interpreted by two independent observers. Interobserver agreement was tested for radiographic interpretation using the kappa statistic. Results from both observers were compared to results found on CT and postmortem including frequency and distribution of lesions and overall agreement.
Findings:
The results showed that the majority of lesions were found to be at C3‐C4 on postmortem and CT but C5‐C7 on radiographs. Interobserver agreement for presence/absence of articular process joint OA on radiographs was found to be fair (k = 0.34). Sensitivity was found to be low for both observers against both CT and postmortem. Although distribution was found to be similar, prevalence of OA differed between the two gold standards (with 78 individual joints identified as positive for OA on CT compared to 33 joints on PM). Specificity was found to be low for both observers against both CT and postmortem.
Conclusion:
The distribution of lesions contradicted the wealth of anecdotal evidence suggesting that arthropathy of the caudal articular process joints is a common finding in nonataxic horses. Interobserver agreement was found to be lower than demonstrated in previous studies suggesting that interpretation of laterolateral radiographs is less precise than previously thought. The sensitivity of radiographs was low when compared to both gold standards, however the gold standards showed different results for prevalence for OA in the population. This changed the positive predictive values between the data sets. The observers were asked to identify presence or absence of OA without comment on the different features of OA assessed. This suggests that different features of OA are assessed on laterolateral radiographs, CT, and postmortem and are differently weighted by clinicians in reaching a diagnosis. Further studies are warranted to investigate the clinical significance of different radiological features and gross pathological lesions as diagnostic and prognostic indicators of cervical articular process joint OA.
EXTERNAL ROOT RESORPTION IN THE HORSE
J. Arnbjerg From the Diagnostic Imaging, Copenhagen University, København, Denmark (Arnbjerg).
Background:
Resorption is a physiological process in deciduous teeth. In man resorption may result from periodontal membrane damage, dental fracture, tooth correction, or duel to infections but seldom connected with hypercementosis. Staszyk (2008) suggests that hypercementosis is a precursor to resorption.
Purpose:
The aim of this study was to follow horses with abnormal root resorption in the incisors but without hypercementosis. Two horses with dental resorption without hypercementosis (I: with infection and the other, II: with trauma as etiology) were followed clinically and radiographically for 2 and 5 years, respectively.
Methods:
Infectious gingivitis was observed in a 22‐year‐old Icelandic horse with a history of loss of dental pieces from mandibular incisors. The intraoral radiographs revealed fragmentation and resorption in all mandibular incisors and 103 and 203 with infectious reaction in the alveolar bone. After extraction of severely affected teeth, the gingivitis resolved. A 6‐year‐old horse had a ventral dislocated fracture of the mandible involving 301, 302, and 303. The radiographs also revealed a fracture of 303 about 1 cm from apex and dislocation of all incisors. This fracture was repositioned and stabilized using interdental wires.
Findings:
The horse was reevaluated approximately 3 and 19 months after initial presentation. Progressive extensive resorption of maxillary incisors was deteriorated by resorption of the remaining maxillary incisors and again the gingivitis around the persistent incisors. After euthanizing, the head was CT scanned showing internal and external resorption but still no hypercementosis. II: The horse was followed up yearly. The apical 10 mm of 301 and 302 gradually disappeared and the pulpa of all three incisors were still open. Five years after the accident, the separately apex of 303 was anchylosed with totally dentine‐filled pulpa. Sykora et al (2013) has shown infectious agents in correlation with resorption. As long as the infection is not involving the gingiva, the patient seems clinical sound. The damaged periodontal membrane resulted in resorption, but when the membrane is intact isolated root pieces did survive, but anchylosed to the alveolar bone.
Conclusion:
The gingivitis healed spontaneously and the horse did clinical well for 17 months, when gingivitis and increased resorption and CT showed internal and external resorption but no hypercementosis. II: Showed resorption, where the periodontal membranes were damaged, and the pulpa seem died, but the old pulpa on the occlusal surface became black. Where the periodontal membrane was intact and separated from 403, the piece stayed alive and gradually anchylosed. Taken together, the two cases show resorption, damaged periodontal membranes pulpa death tooth fracture anchylosis, gingivitis but no hypercementosis. Resorption can occur in clinical sound equine patients without hypercementosis.
MORPHOLOGICAL SONOGRAPHIC PARAMETERS OF COMMON CAROTID ARTERY IN FOALS
A. C. S. Aguiar, G. F. Viana, V. M. D. V. Machado From the Department of Animal Reproduction and Veterinary Radiology, Veterinary Medicine and Animal Husbandry School, Botucatu, SP, Brasil (Aguiar, Viana, Machado).
Background:
Common carotid artery (CCA) is the main source of arterial supply for the brain and reduced supply caused, for example by stenosis, can lead to brain lesions, affecting white matter as already described in humans and rats. Physiologic data regarding sonographic evaluation of the CCA in horses are lacking.
Purpose:
The aim of this study was to obtain normal values for CCA intima‐media thickness (IMT) and diameter in American Quarter Horse foals and check for differences between sexes.
Methods:
Ten American Quarter Horses aged between 13 months and 19 months old were submitted to bilateral ultrasonographic scans of common carotid artery. Five females aged between 17 and 19 months old, an average of 18. 5 males aged between 13 and 18, average of 16.3 months old. The equipment used was My LAB30‐ESAOTE and a linear transducer of 18 MHz. Three to five measurements of IMT and lumen diameter (D) were made using at least three different images for each artery. Common carotid artery was accessed on the lower neck, around C5 and C6.
Findings:
Cardiac rate was in average 54 bpm among males and 55 among females. Mean value for right IMT was 0.96 ± 0.2 mm for males and 0.95 ± 0.12 mm for females. Mean left IMT was 0.90 ± 0.15 mm for males and 0.95 ± 0.14 mm for females. Mean right CCA luminal diameter was 9.62 ± 0.88 mm for males and 9.41 ± 0.45 mm for females. Mean left CCAD was 9.28 ± 0.80 mm for males and 9.08 ± 0.21 mm for females. There was no difference between IMT between sexes. Right diameter was larger than the left one. Males presented larger diameters than females, even though they were younger.
Conclusion:
There are very few studies on equine carotid artery, especially regarding normal values. The higher values in diameter for males, even though they were younger than females, suggest that there is a difference between sexes. This finding is in accordance with Cipone et al. (1997) that noticed diameter was greater in males than females and it was independent of weight. The authors have no knowledge of literature regarding the difference between left and right side diameters.
ULTASONOGRAPHIC ARTIFACTS APPEARANCE IN THE COLLATERAL LIGAMENTS OF THE DISTAL INTERPHALANGEAL JOINT WITH THE LIGAMENT UNDER DIFFERENT TENSION LEVELS
S. J. A. De Blaauw, I. Verschoor, C. Crijns From the DAP Bodegraven BV, Bodegraven, The Netherlands (De Blaauw, Verschoor, Crijns) and Department of Veterinary Medical Imaging and Small Animal Orthopaedics, Ghent University, Merelbeke, Belgium (Crijns).
Background:
Desmopathy and enthesiopathy of the collateral ligaments of the distal interphalangeal (DIP) joint are commonly causing foot‐related lameness in horses. The appearance of relaxation artifacts during ultrasonographic evaluations of these ligaments has previously been described in the literature.
Purpose:
The aim of this work is to evaluate the possibility of influencing the appearance of relaxation artifacts during ultrasonographic evaluation of the collateral ligaments of the DIP joint, if the ligament is scanned under different tension levels.
Methods:
Ultrasonographic examination of the lateral collateral ligament of the left front DIP joint was performed. The ligament was imaged three times with the DIP joint each time in three different positions: lateral extension, neutral, and lateral flexion. A 12° wedge under the medial or lateral quarter and heel was used to place the DIP joint in lateral extension and flexion. Just distal to the proximal insertion of the ligament, the area, circumference, width, and thickness of the ligament were measured. For the different positions, the mean values of the parameters measured in the three examinations were calculated. The differences between the measurements with the joint in different positions were analyzed using analysis of variance followed by Bonferroni's post hoc test. Retrospectively, one observer scored blindly the shape and echogenicity of the ultrasonographic images in four different categories: (1) flat‐oval and homogenous, (2) oval and homogenous, (3) oval and heterogeneous, and (4) round‐oval and heterogeneous.
Findings:
Twenty‐eight adult Dutch Warmblood horses without foot‐related lameness were included in the study. The data showed a normal distribution. ANOVA showed significant differences between the measurements in lateral extension and lateral flexion position (P: area < 0.0001; circumference 0.0001; thickness < 0.0001) and between the measurements in lateral extension and neutral position (P: area < 0.0001; circumference 0.025; thickness < 0.0001), for all parameters except for the width, which differed not significantly between any groups. There was also no significant difference between the measurements in neutral and lateral flexion position. The echogenicity of the ligament was scored in category 1 or 2 when the ligament was examined with the DIP joint in lateral extension, in category 2, 3, or 4 when the ligament was examined with the DIP joint in a neutral position and in almost all cases in category 3 or 4 when the ligament was examined with the DIP joint in lateral flexion.
Conclusion:
Changing the position of the joint can influence the ultrasonographic appearance of the collateral ligaments of the DIP joint. The appearance of relaxation artifacts can be reduced if the ligament is tensed and can be evoked if the ligament is more relaxed during ultrasonographic examination.
USEFULNESS OF ULTRASONOGRAPHY OF DEEP CERVICAL LYMPH NODES IN DIAGNOSIS OF NEOPLASTIC DISEASE IN HORSES
L. Evrard, R. Fonseca, G. Bolen, V. Busoni From the Diagnostic Imaging Section, Clinical Department of Small Animals and Equids, University of Liege, Liege, Belgium (Evrard, Fonseca, Bolen, Busoni).
Background:
Horses with neoplastic disease often present nonspecific clinical signs including dyspnea, weight loss, or chronic/recurrent colic. Diagnostic imaging is routinely performed on those patients.
Purpose:
This study aims (1) to describe ultrasonographic (US) technique and anatomy of deep cervical lymph nodes in horses, (2) to present imaging findings in a series of horses with deep cervical lymphadenomegaly and cytologically or histologically confirmed neoplasia.
Methods:
Deep cervical lymph nodes were explored using a 5–7.5 MHz curvilinear transducer. Ultrasonographic examination of the base of the neck and the thoracic inlet was performed in five normal horses to establish normalcy. Medical records were reviewed to select horses with deep cervical lymphadenomegaly at ultrasonography and a final diagnosis of neoplastic disease.
Findings:
In normal horses deep cervical lymph nodes appeared as well‐defined hypoechoic ovoid structures around 7–8 mm thick, in the area of bifurcation of common carotid arteries. Seven horses responded to the selection criteria. All of them had undergone US examination of abdomen and thorax and a radiographic examination of the thorax. A transrectal ultrasonography was done in one horse. Abdominal abnormalities were detected in all horses and included abdominal lymphadenopathy (four horses), peritoneal effusion (three horses), hepatic (three horses), and splenic (two horses) abnormalities. Intrathoracic abnormalities were detected in all horses and included mediastinal masses (seven horses), lung parenchymal abnormalities (five horses), and pleural effusion (four horses). Of the seven horses, four were diagnosed to have lymphoma, one had an adenocarcinoma, one had an undifferentiated round cell tumor, one had macroscopic neoplastic findings at necropsy but final classification is pending.
Conclusion:
Lymphadenomegaly of deep cervical lymph nodes can be detected in horses with diffuse neoplastic disease. Deep cervical lymph nodes are well visualized when enlarged and accessible for fine needle aspiration. Their US assessment can be easily added to routine imaging assessment of thorax and abdomen in case of suspected neoplasia in horses.
MAGNETIC RESONANCE IMAGING OF THE DEEP DIGITAL FLEXOR TENDON IN THE EQUINE DIGIT OF FIT TO COMPETE HIGH LEVEL SHOW JUMPERS
T. Schulze, J. Olive, C. J. Lischer From the Klinik fur Pferde, Allgemeine Chirurgie und Radiologie, Berlin, Germany (Schulze, Lischer) and Centre Hospitalier Veterinaire Saint‐Martin, Saint‐Martin‐Bellevue, France (Olive).
Background:
The potential of MRI to identify lesions that are not detectable with other imaging modalities has led to a growing demand for MRI as a part of prepurchase examinations. Veterinarians are asked to speculate how these findings will affect future performance. So far there is limited information about the clinical significance and MRI findings in competing high‐level sport horses. Tendinopathy of the distal aspect of the deep digital flexor tendon (DDFT) is often considered as a primary cause of lameness. The present study describes MR abnormalities of the DDFT in thoracic feet of show jumpers actively competing in high classes.
Purpose:
The hypothesis was that high‐level show jumpers show some signal alterations in the distal aspect of the DDFT, which do not exclude them from competing.
Methods:
MRI studies of the foot were obtained on 95 forelimbs of 48 warmblood show jumping horses participating in at least two competitions (minimum 1.40 m classes) within a maximum of 3 weeks both before and after imaging. In addition to the regular vet check during the competition, every horse underwent a standardized clinical lameness examination before MRI. A questionnaire, filled out by the owners, provided further information about previous history of lameness or treatments. Standardized MRI was performed on a standing 0.27T low‐field system (Hallmarq). MRI examinations were reviewed separately by two veterinarians blinded to the identity of the horses. One veterinarian repeated the blinded retrospective interpretations 3 months after the first assessment in order to determine intraobserver reproducibility. Location and length of DDFT lesions were assessed according to a suggestion from Vanel et al. (2012).
Findings:
Signal abnormalities in the DDFT could be found in 37 of 95 limbs in T1‐GRE sequences, but could only be seen in two limbs on T2‐FSE images. Multifocal signal abnormalities were seen in seven limbs on T1‐GRE sequences but not on T2‐FSE images. Sixty‐two percent of the signal abnormalities in the DDFT on T1‐GRW images were located suprasesamoidean, 30% sesamoidean, and 8% infrasesamoidean. A magic‐angle effect was frequently seen and is discussed as possible cause for false‐negative and false‐positive findings as well as volume averaging artifacts. Sequences essentially affected by patient movement were excluded from this evaluation. However, mild movement artifacts still need to be considered in interpretations of MR examinations of standing horses.
Conclusion:
In conclusion, high‐level show jumpers frequently show DDFT lesions on T1‐weighted images only and this should not preclude continuation or return to athletic activity. Conversely, very few competing horses had only small DDFT lesions visible on T2‐FSE images and these are more likely detrimental to such activity.
IMAGING AND PATHOLOGIC FINDINGS IN INTERVERTEBRAL DISC DISEASE IN THE EQUINE CERVICAL SPINE
S. Veraa, W. Bergmann, I. D. Wijnberg, W. Back, A. J. M. Van Den Belt From the Division of Diagnostic Imaging (Veraa, van den Belt), Department of Pathobiology (Bergmann), Department of Equine Sciences, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands (Wijnberg, Back), and Department of Anaesthesia and Surgery of Domestic Animals, Faculty of Veterinary Medicine, Ghent University, Ghent, Belgium (Back).
Background:
Intervertebral disc disease is a well‐recognized entity in small animal and human medicine. In horses only case reports are available describing protrusion and in a single case extrusion of the intervertebral disc. Clinical diagnostic imaging of intervertebral disc disease in horses has so far been limited to radiographic studies.
Purpose:
The purpose is to describe the imaging findings, including in vivo radiography and ex vivo computed tomography (CT) and magnetic resonance imaging (MRI), and the macroscopic pathological findings of the cervical intervertebral discs in four adult horses.
Methods:
The cervical vertebral column of four adult horses (one without clinical complaints related to the cervical spinal column or intervertebral disc space collapse on radiography, three with clinical complaints related to the spinal column of which one with and two without intervertebral disc space collapse on radiographic examination) was dissected at the level of Th2‐3 with placement of a ligature to preserve the cerebrospinal fluid. Ex vivo CT and MRI scans (T1‐, T2‐weighted sequences without and with fat suppression) were obtained and evaluated by a board‐certified radiologist, after which the vertebral column was macroscopically evaluated by a board‐certified pathologist.
Findings:
Intervertebral disc protrusions of varying degree without apparent spinal cord compression were present in all adult horses and could be detected both in CT and MRI images. Intervertebral disc extrusion in combination with a hyperintense focus in the spinal cord in T2‐weighted images was present in one horse at C6‐7. The intervertebral disc was slightly hyperattenuating compared to the spinal cord in CT images. The intervertebral discs displayed mainly a patchy isointens signal on T1‐weighted images and patchy hypointense signal on T2‐weighted images compared to the spinal cord. The nucleus pulposus could only be seen as a very small and vague hyperintense slit with a hypointense rim on T2‐weighted images. Gross macroscopic findings differed from the CT and MRI findings in such that in three horses a disrupted yellowish intervertebral disc was found, which could not be appreciated as a severely degenerated disc in imaging findings. No abnormal intervertebral disc was found in the horse without clinical complaints related to the cervical spinal column.
Conclusion:
Intervertebral disc protrusion and extrusion can be appreciated with CT and MRI; however the severity of degeneration was not appreciated. Additional MRI sequences should be introduced into the standard imaging protocol of intervertebral disc disease to fully evaluate equine intervertebral disc degeneration.
EVALUATION OF COMPUTED TOMOGRAPHY VS. POSTMORTEM FINDINGS FOR DIAGNOSIS OF OSTEOARTHRITIS IN THE EQUINE CERVICAL SPINE
J. Mullard, R. Wood, M. Tambaschi, R. Weller From the Department of Veterinary Clinical Sciences, The Royal Veterinary College, North Mymms, Hertfordshire, UK (Mullard, Wood, Tambaschi, Weller).
Background:
Advanced imaging of the equine cervical vertebrae is an area of interest due to the many pathologies that can occur in this area. Lateral and oblique radiographs are routinely used to assess bony lesions, however computed tomography (CT) is gaining popularity and currently there is only limited information about its diagnostic value for the assessment of cervical lesions in the horse.
Purpose:
The aim was to determine the sensitivity and specificity of CT for the diagnosis of osteoarthritic lesions of the articular process joints in the equine cervical spine.
Methods:
CT scans of 25 Thoroughbred‐type cadaver necks were obtained and reviewed independently by three observers for the presence of osteoarthritis of the articular process joints. Osteroarthritis was deemed present if ostophytes were seen, it was recorded which joint was affected and if it was uni‐ or bilateral. CT findings were compared with gross postmortem examination of new bone proliferation of the cervical skeleton and sensitivity and specificity calculated.
Findings:
New bone proliferation was defined as the presence of asymmetric enlargement, bony buttresses, lipping, or medial facet formation. As only the osseous skeleton remained, changes in joint space were not assessed. CT was found to be highly sensitive for observer A and C (0.78% and 0.91%), while observer B had lower sensitivity (0.20%), but specificity was high for observer B (0.74%) and low for observer A and C (0.48%, 0.09%) compared to gross postmortem examination. Sixty‐eight percent (n = 17) of cervical spines were found to have new bone proliferation present on gross postmortem examination, with 27% (n = 30) of facet joints affected, with the majority (70%) occurring at C3‐C4.
Conclusion:
Differences in sensitivity and specificity between observers for CT are thought to be explained by the differing experience in reading CT images of the cervical spine. CT is highly sensitive and specific for bony lesions of the equine cervical neck, depending on reviewer experience. This finding will help us to better understand the pathology in a research context and in the future aid diagnosis, however this hinges on the advent of a CT scanner that can image further than the cranial spine in the live horse.
COMPUTED TOMOGRAPHIC PITUITARY GLAND HEIGHT‐TO‐BRAIN AREA RATIO MEASUREMENTS IN HEALTHY EQUINES
C. P. Crijns, H. J. J. van Bree, G. van Loon, S. Schauvliege, A. Martens, I. M. V. L. Gielen From the Department of Veterinary Medical Imaging and Small Animal Orthopaedics (Crijns, van Bree, Gielen), Department of Internal Medicine and Clinical Biology of Large Animals (van Loon), and Department of Surgery and Anaesthesiology of Domestic Animals, Ghent University, Merelbeke, Belgium (Schauvliege, Martens).
Background:
Pituitary pars intermedia dysfunction (PPID) is a common endocrinologic disorder in horses. Contrast‐enhanced computed tomography (CT) has been described for diagnosing an enlarged pituitary gland in dogs and horses. The reference values described in the literature however do not take into account the factors that influence the size of the pituitary gland between individuals. The pituitary gland size in horses can be influenced by age, size, gender, and by seasonal variation.
Purpose:
The aim of this work is to determine the pituitary gland height‐to‐brain area ratio (P:B ratio) reference values, as previously described in dogs, which can be used for all equines, independent of their size.
Methods:
Contrast‐enhanced CT studies of PPID free horses, divided in three different age groups—young (<5 years), adult (5–15 years), and aged (>15 years); three different size groups—small (<250 kg), average (250–550 kg), and large (>550 kg); and two gender groups—male, female, were reviewed. The pituitary gland height was measured on the CT slice at punctum maximum. On the same CT slice, the brain area was calculated using a segmentation tool of the DICOM viewer. Using these measurements, the P:B ratio could be calculated for each horse. Cut‐off reference values for the different age and gender groups were determined.
Findings:
Twenty‐one horses and ponies were included in the study: 10 young (<5 years), five adults (5–15 years), and six aged (>15 years)/three small (<250 kg), 11 average (250–550 kg), and seven large (>550 kg)/eight male and 13 female. Using the P:B ratio instead of the pituitary height, less variation in the results was detected within the different age groups and different size groups. The P:B ratio increased with the age of the horses and was higher for mares compared to stallions and geldings. Calculated reference values for young horses are: female ≤0.12 and male ≤0.11; for adult horses: female ≤0.14 and male ≤0.11; for aged horses: female ≤0.17 and male ≤0.13.
Conclusion:
This study provides P:B ratio reference values for assessment of pituitary gland size in different sized individuals, from miniature horses to large warmbloods. Because age and gender influence the P:B ratio, age‐ and gender‐specific reference values should be used to evaluate the equine pituitary gland.
MAGNETIC RESONANCE DACRYOCYSTOGRAPHY IN THE HORSE
G. Manso‐diaz, M. Garcia‐lopez Jose, R. Herron Vilella, M. I. Garcia‐real, O. Taeymans From the Department of Animal Medicine and Surgery, School of Veterinary Medicine, Universidad Complutense de Madrid, Madrid, Spain (Manso‐Diaz, Herron Vilella, Garcia‐Real), Department of Clinical Sciences, Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA (Garcia‐Lopez Jose), and Dick White Referrals, Six Mile Bottom, Cambridgeshire, UK (Taeymans).
Background:
Diseases affecting the nasolacrimal duct (ND) are uncommon in the horse and when present usually result in obstruction of its lumen. Detailed evaluation of the ND can be performed by outlining it with contrast media; a method known as dacryocystography. In the horse, this technique has been described using radiography and CT.
Purpose:
The purpose of this study was to provide a comprehensive anatomical description of the normal equine ND using magnetic resonance imaging (MRI), and to develop a practical method for performing magnetic resonance dacryocystography (MRD) in horses.
Methods:
The dislocated heads of seven fresh equine cadavers were used. Transverse 3D T1‐weighted spoiled gradient echo (T1W‐FFE3D) images were acquired with a 0.23T magnet. Approximately 5 ml of contrast was injected through each nasolacrimal ostia (until overflow was noted at the puncta). The study consisted of two parts. In the first part, pharmacological olive oil and gadolinium were compared as contrast medium, using three horses. Two doses of gadolinium were used: 50% diluted and pure gadolinium 0.5 mmol/ml. In the second part, influence of different head recumbencies using olive oil as the sole contrast medium was studied using the remaining four horses. The ND was identified on transverse images and multiplanar reconstructions (MPR). Degree of distention and completeness of contrast filling along the course of the duct were assessed for each horse.
Findings:
Injection with olive oil resulted in better distention, and resulted in complete filling of the duct in all horses. Injection of 50% gadolinium resulted in incomplete filling of the ND, while injection of pure gadolinium resulted in slightly better filling of the duct. The nondependent duct showed better distention compared to the dependent duct when scanning in lateral recumbency. No differences were observed between ducts in dorsal recumbency. The degree of distention in this position was slightly smaller than the degree of distension observed in the nondependent ND in lateral recumbency.
Conclusion:
MRI is a valid imaging modality for assessing ND in horses. Acquisition of 3D sequences allows performing MPR of the ND. As opposed to filling the ND with gadolinium, injection with olive oil resulted in increased distention and complete filling of the duct. This technique can equally be performed in lateral or dorsal recumbencies, but the latter might be preferred, as degree of distention of the duct is not influenced by recumbency in this position.
BEAM HARDENING EFFECTS OF LEAD SHIELDING DURING BONE SCINTIGRAPHY IN THE HORSE
U. C. M. Kafka, A. Carstens, J. P. Slabbert From the Department of Companion Animal Clinical Studies, Section of Diagnostic Imaging, University of Pretoria, Pretoria, Republic of South Africa (Kafka, Carstens) and iThemba LABS, Stellenbosch, Western Cape, Republic of South Africa (Slabbert).
Background:
During 99mTc bone scintigraphy in the horse, personnel may be subjected to at least 1–2 h of radiation exposure emitted from the horse, as a result of extensive gamma ray interaction within the patient.
Purpose:
This study aimed to assess whether lead shielding would reduce exposure during scintigraphic examinations. Interaction of lead with emitted radiation may produce radiation, which is more harmful than scatter radiation emitted from the horse. Removal of lower energies within a spectrum with resultant higher energy peaks is called beam hardening.
Methods:
Five horses were injected with 150mCi 99mTc, scanned initially without lead shielding, followed by a number of combinations of lead coats of different lead thicknesses (0.25, 0.35, and 0.5 mm of lead equivalent) placed on the horse at varying locations (in front of the head, side of the head, shoulder, bladder, and pelvic region), and on coat hangers mimicking coated personnel. All resultant energy spectra and the number of counts/second were measured by a nuclear physicist with a spectrometer and recorded.
Findings:
Energy spectra emitted from the horses injected with 99mTc‐MDP contained a large component of lower energies as a result of patient physical matter interaction. Two energy peaks appeared behind lead shielding, one at 86 keV and another at the typical gamma peak of 140.5 keV for 99mTc. As the thickness of the lead varied, so did the relative heights of the two peaks. With increasing lead thickness, there was a progressive reduction in number of counts. The difference in reduction of radiation exposure measured behind the 0.35 mm and 0.5 mm lead coats on hangers was small. Although a higher energy 140.5 keV peak is theoretically biologically safer than 86 keV, the energy intensity behind lead shielding reduced substantially. There was no evidence of the expected beam hardening effect. Instead, lead shielding resulted in softening of the 99mTc gamma spectrum. Radiation exposure to personnel was lowest in front of the head and highest at the shoulder and bladder region.
Conclusion:
Despite the prominence of a lower 86 keV peak, the wearing of lead shielding during bone scintigraphy is strongly recommended. It markedly reduces the intensity of radiation and removes harmful scatter energies emitted from the patient during scintigraphic examination of horses. The horse should be draped with a 0.5 mm lead coat and bystanding personnel should wear a 0.35 mm lead coat. The handler should stand directly in front of the horse's head.
MAGNETIC RESONANCE IMAGING IN SMALL RUMINANTS WITH CONFIRMED BRAIN DISEASE
K. Ertelt, A. Oevermann, C. Stahl, J. Lauper, D. Henke, D. Gorgas From the Division of Clinical Radiology, Department of Clinical Veterinary Medicine (Ertelt, Stahl, Gorgas), Division of Experimental Clinical Research, Department of Clinical Research and Veterinary Public Health (Oevermann), Clinic for Ruminants, Department of Clinical Veterinary Medicine (Lauper), and Division of Neurology, Department of Clinical Veterinary Medicine, Vetsuisse Faculty Bern, Bern, Switzerland (Henke).
Background:
The prevalence of brain disease in small ruminants dying of natural causes is high (8.1%). Antemortem diagnosis is difficult, because when noticed, animals are presented in advanced stages of disease with unspecific neurologic signs. Magnetic resonance imaging (MRI) is the imaging method of choice, but MRI findings in small ruminants are not well described.
Purpose:
Purpose of the study is the description of MRI findings in small ruminants with histopathological confirmed neurological disorders localized to the brain.
Methods:
MRI examination was performed in nine small ruminants (six sheep, three goats) presented between December 2011 and April 2014 because of neurologic disorders. In two patients, the MRI examination was performed postmortem. Examination protocol included T2‐weighted images in transverse and sagittal plane and fluid attenuation inversion recovery and T1‐weighted images in transverse and dorsal plane. Postcontrast T1‐weighted images were performed in all living animals. Evaluation of MRI included presence and locations of lesions, lesion distribution, border definition, signal intensity, and contrast enhancement. All animals underwent a pathological examination.
Findings:
Neuropathological examination revealed clostridial encephalomalacia (2, therefrom 1 postmortem), listeric encephalitis (2), polioencephalomalacia (1), abscess within the basioccipital bone (1 postmortem), brain abscess (1), viral cerebellitis (1), and migration tracts caused by parasitic infection (1). In the latter two, no lesions were visible on MRI. In the remaining seven animals focal (2), multifocal (2), or diffuse lesions (3) were visible with mass effect being present in five animals. All toxic metabolic diseases (costridium enterotoxemia and polioencephalomalacia) showed a diffuse, bilateral symmetric distribution. The majority of lesions were hyperintense on T2‐weighted images and hypointense on T1w images. Lesion signal intensity differed in both cases of abscess (one postmortem) and in one case of clostridium enterotoxemia, which showed focal hyperintensity on T1‐weighted images with hypointense signal on T2‐weighted images. Contrast enhancement was present in two of five animals, with a ring‐like pattern in the brain abscess and mild enhancement in one case of listeric encephalitis.
Conclusion:
MRI identified brain lesions in seven of nine animals. In these animals, MRI features were typical for metabolic‐toxic disease, listeric encephalitis, and brain abscess, respectively. Despite not all lesions were visible, identification of metabolic/toxic disease and the differentiation from infectious disorders such as listeric encephalitis seems to be possible. The awareness of the variable neurological disorders occurring in small ruminants and knowledge of their MRI features is important for diagnosis of neurologic disease in small ruminants.
COMPUTED TOMOGRAPHY AS A SCREENING TOOL FOR ARTICULAR OSTEOCHONDROSIS IN PIGS
K. Olstad, J. Kongsro, E. Grindflek, N. I. Dolvik From the Department of Companion Animal Clinical Sciences, Faculty of Veterinary Medicine and Biosciences, Norwegian University of Life Sciences, Equine Section, Oslo, Norway (Olstad, Dolvik) and Norsvin, Hamar, Norway (Kongsro, Grindflek).
Background:
In some countries, boar performance testing includes assessment of body composition through CT scanning. CT can therefore potentially also be used to screen and select against heritable articular osteochondrosis, but the accuracy of CT screening for osteochondrosis is currently unknown.
Purpose:
The aims of the current study were to calculate the positive predictive value of CT for the diagnosis of articular osteochondrosis using histological examination as the reference standard, and to monitor development of osteochondrosis because initiation and resolution of defects will affect accuracy if only a single screening is feasible.
Methods:
Material consisted of 19 male Landrace piglets that were purpose‐bred for the study and sired by the boar with the national highest phenotypic estimated breeding value for osteochondrosis. All four limbs of each piglet were scanned in a 32‐slice helical CT at one to eight examination intervals from piglet age 70–180 days (weight: 20–128 kg). The majority of piglets were anesthetized, scanned, and recovered; however, at each interval from one to two piglets were euthanized and the left distal femur processed for histological examination. All piglets that remained at the eighth and final interval were euthanized and the left distal femur examined histologically.
Findings:
Areas of ischemic chondronecrosis, that is osteochondrosis were identified in histological sections from each of the 36 focal radiolucent defects in the ossification front suspected of representing osteochondrosis in CT scans. This corresponds to a positive predictive value of 100% (95% CI: 90–100). The positive predictive value of macroscopic diagnosis of osteochondrosis was calculated for comparison and found to be 79% (95% CI: 67–84). Within the hind and fore limbs, the incidence of defects was highest in the stifle (n = 321) and elbow joint (n = 112), respectively. Ninety‐eight percent of the total number of defects that arose in the stifle and elbow up to the maximum age of 180 days were detectable by the seventh examination interval when piglets were a mean of 159 days old (mean weight: 98 kg). The proportion of defects that resolved was lowest in the stifle at 51% and highest in the elbow at 69%.
Conclusion:
Ninety‐eight percent of the defects that arose in the stifle and elbow by slaughter weight (130 kg) were detectable by a single screening at 98 kg body weight. Sixty‐nine percent of defects could have resolved before screening at a single time in the elbow. This category of false‐negative diagnosis may not adversely affect prevalence in adult pigs following selection as it will result in inclusion of pigs that are disposed for healing. Combined with the excellent positive predictive value, these results support the conclusion that optimally timed CT examination is a highly accurate screening tool for osteochondrosis.
COMPUTED TOMOGRAPHY DIAGNOSES IN AVIANS SUFFERING FROM UPPER RESPIRATORY TRACT DISEASE
M. Gumpenberger, A. Scope From the Diagnostic Imaging (Gumpenberger) and Department for Companion Animals and Horses, University Clinic for Small Animals, University of Veterinary Medicine, Avian and Reptile Medicine, Veterinarplatz 1, 1210 Vienna, Austria (Scope).
Background:
Computed tomography (CT) has become a routine procedure for avians suffering from respiratory disease at our clinic. One main advantage of a quick multislice helical CT is the possibility to scan avian patients when awake. Moderate patient movement artifacts usually do not hamper meaningful interpretation of lung or air sac disorders. But in the last 24 months the number of avians presented with respiratory distress due to disorders of the upper respiratory tract has been increasing at our clinic. Due to the mostly severe dyspnea in these birds, we tried to perform most of the examinations unsedated.
Purpose:
Radiographs usually do not provide sufficient information for conclusive diagnoses of the upper respiratory tract or are not performed due to the need of anesthesia. The purpose of this study was to demonstrate the advantage and diagnostic value of CT examinations in (mostly conscious) birds suffering from upper respiratory tract disease.
Methods:
Most birds were scanned consciously in an upright position in a cardboard box (optionally equipped with a perch), whereas some were sedated and positioned in sternal recumbency. Quick and effective examination was achieved by using a multislice helical CT (Siemens Somatom Emotion, Siemens AG). All scans were performed with 130 kV, 80 mAs, and 0.6 mm thin slices. Quick animal handling by “bird‐trained” staff allowed further reduction of examination time. In general, CT scans (including preliminary procedures) of birds ranging from budgerigar to African gray size were performed within less than 5 min. Only the examination of larger species like geese or swans was slightly more time consuming.
Findings:
Between January 2010 and May 2014, 34 avian patients with a history of respiratory tract disease received a CT examination. Seven of 34 patients suffered from upper respiratory disease: 4/7 presented with tracheal lumen obstruction by aspergillous granuloma or inflammatory products, one with scar tissue induced trachea constriction after a biting wound, and two with (partial) tracheal collapse. Definite diagnoses were established in 5/7 birds with cytology and histopathology. Two birds are still alive. Patient movement hindered the final diagnosis in the bird with the biting wound, therefore the scan was repeated under general anesthesia.
Conclusion:
According to our experience, CT was crucial for final diagnosis and therefore prognosis and therapy. Movement artifacts of nonsedated patients are always to be taken into account as possible limitations, though. In tiny birds even minimal patient movements interfere with correct diagnoses of small suspected tracheal lesions.
AORTIC ANEURYSMS IN ASIATIC BLACK BEARS (URSUS THIBETANUS) RESCUED FROM BILE FARMS IN CHINA
M. Bando, N. Webster, L. Nelson, C. Hartley, D. Donaldson, J. Dwyer From the Animals Asia (Bando, Dwyer), Adelaide Veterinary Specialist and Referral Centre (Webster), Veterinary Teaching Hospital, Pullman, WA(Nelson), and Centre for Small Animal Studies, Animal Health Trust, Newmarket, Suffolk, UK (Hartley, Donaldson).
Background:
Animals Asia has rescued 276 Asiatic black bears (Ursus thibetanus) from bile farms in China, where bears experience serious welfare issues. Aortic aneurysms in animals, while not commonly reported, are associated with conditions including heredity, parasites, nutrition, and hypertension.
Purpose:
The purpose of this study was to investigate aortic aneurysm in the captive Asiatic black bear population at the Animal Asia, China Bear Rescue Centre, Chengdu.
Methods:
Aortic aneurysms have been identified in nine (3.2%) bears at the China Bear Rescue Centre (Longqiao, Chengdu), from radiographs, computed tomography, and necropsies. Eight of the nine bears died. Four died of dissecting aortic aneurysms. One of the eight deceased was found dead with hemopericardium without evidence of aortic dissection. Necropsies confirmed aortic aneurysms in three bears euthanized due to spinal paresis, a nasal tumor, and unresponsive head pressing.
Findings:
Clinically, one of the deceased bears exhibited signs described as coughing attributed to severe aortic dilation. The surviving bear remains asymptomatic. Preliminary histopathology of necropsied bears revealed no primary aortic abnormalities. High prevalence of retinal and renal pathology in farmed bears may be attributed to hypertension and suggests an association with aortic aneurysms in this population. Seven (4.6%) of 152 bears at CBRC receive oral enalapril (0.5 mg/kg BID, Enalapril, Yabao Pharmaceutical Group, Shanxi, China) and amlodipine (0.05–0.1 mg/kg SID‐BID, Norvasc, Beijing Yimin Pharmaceutical, Beijing, China) as treatment for suspect hypertensive retinopathy following fundic examination.
Conclusion:
In humans, aortic aneurysms and associated hypertension is well documented. To the authors’ knowledge, this is the first report of aortic aneurysms in ursids. Understanding the etiology of aortic aneurysms in this population is of importance considering the severe health implications when rehabilitating farmed bears.
SYNCHRONIZED COMPUTED TOMOGRAPHY‐DIGITAL RADIOGRAPHY IMAGING IN HOOFED MAMMALS’ DISTAL LIMB
G. Galateanu, C. A. Szentiks, Z. Mezo, G. Straus, M. F. Bertelsen, T. B. Hildebrandt From the Leibniz Institute for Zoo and Wildlife Research, Berlin, Germany (Galateanu, Szentiks, Mezo, Hildebrandt), Tierpark Berlin, Berlin, Germany (Straus), and Copenhagen Zoo, Copenhagen, Denmark (Bertelsen).
Background:
Chronic foot disease represents a tremendous clinical challenge and is a major health concern for herbivores in captivity, often being a reason for euthanasia. Currently, radiography is the only imaging technique used to diagnose bone pathology in animals under “field conditions.” Yet, unlike in elephants for which numerous studies exist, keystone data, such as normal foot radiographic appearance is very scarce for other hoofed mammals. The current assumption is that the anatomy of wild herbivores can be extrapolated from their domestic relatives.
Purpose:
This study was designed to identify and depict normal radiographic appearance of distal foot in a wide variety of captive hoofed mammals.
Methods:
Using state‐of‐the‐art tools of 128‐slice computed tomography (CT) and a dedicated workstation, the authors initiated a pioneering approach termed “synchronized CT—digital radiography (DR), (Synch DR)” based on X‐ray projections derived from tridimensional (3D) CT reconstructed images. For each foot studied, eight 3D CT images (45° apart), equivalent to eight standard radiographic views, together with additional projections were obtained. Each such 3D CT image was transformed into a Synch DR image and the following autopodial elements were analyzed: podial elements, metapodials, and phalanges. Distal limb excerpts were collected after the unrelated death of the animals, in the context of mandatory postmortem examination and disease diagnosis. Seventy‐nine captive wild odd‐toed ungulates (Perissodactyla) and even‐toed ungulates (Artiodactyla) from a wide variety of species and subspecies (n) were encompassed in our study. The order Perissodactyla included all extant families: Equidae (n = 3), Tapiridae (n = 2), and Rhinocerotidae (n = 2). Artiodactyla's order included: Suidae (n = 2), Tayassuidae (n = 1), Hippopotamidae (n = 2), Tragulidae (n = 1), Camelidae (n = 4), Giraffidae (n = 3), Cervidae (n = 6), and Bovidae (n = 15).
Findings:
High‐resolution CT‐DR synchronization (1) depicted normal anatomic features and their radiographic appearance for each foot in eight radiographic views; (2) showed that species‐dependent anatomical variations must be taken into account, emphasizing the unique traits found in each taxa; (3) endowed us with key radiographic knowledge from many endangered species, with a limited number of captive individuals; (4) enabled the identification of species‐dependent radiographic views that allow a better visualization of autopodial elements than the traditional orthogonal and oblique projections. Distinct anatomical differences were found between the domestic and wild species both in Perissodactyla and Artiodactyla orders (e.g., Equidae, Hippopotamidae, Bovidae). Furthermore, noticeable autopodial dissimilarities were found between two wild species belonging to the same taxonomic family (e.g., Southern white and Indian rhinoceroses).
Conclusion:
This study provides important reference data on radiographic appearance of normal anatomy; taxa‐specific, optimal radiographic views for better autopodial visualization; and comparative radiographic anatomy results in hoofed mammals’ distal limb. Such major advances in diagnostic imaging facilitate straightforward implementation and easy interpretation of ungulate foot radiographic images taken in the field worldwide.
RADIOGRAPHIC AFFECTIONS OF THE TARSUS OF THE DROMEDARY CAMEL (CAMELUS DROMEDARIUS)
R. O. Ramadan From the College of Veterinary Medicine and Animal Resources, King Faisal University, Al‐Ihssa, Saudi Arabia (Ramadan).
Background:
The anatomy of the tarsus of camel has been investigated in isolated specimen using MRI, CT, radiography, and ultrasound (Hagag et al., 2012). However radiographic affections of the tarsus have rarely been reported (Ramadan 1994). The aim of the study was to record the radiographic affections of the tarsus of the dromedary camel.
Materials and findings:
Camels were presented at the Veterinary Teaching Hospital over the last 30 years suffering from lameness seated at the tarsal region. Sex, age, and breed of the camels were recorded during clinical examination. Radiographs were obtained in the lateromedial, dorsoplantar, and occasionally in oblique directions using a mobile X‐ray machine.
Results:
The tarsus of camel was subject to a variety of affections. Its congenital anomalies were upright tarsus conformation, wedge‐shaped tarsal bone, and hypoplastic fibula tarsal bone. Traumatic injuries included physis‐related fractures of Salter Harris (Type I in the proximal fibula tarsal bone and type I or type II in the distal tibia). These injuries were noted in both sexes up to 4 years of age. Also transverse or comminuted fractures of the fibula tarsal bone, dislocation of central tarsal bone, and degenerative joint disease were of frequent occurrences.
Discussion/Conclusion:
The tarsus of the dromedary camel is the major pivoting center during mating. Therefore any restriction in its movement might terminate the reproductive life of the camel. Fractures of the fibula tarsal bone or fractures and dislocation of the central tarsal bone of young camel up to 1 year old could be treated with plates and screws as well as tension band wiring system. Degenerative joint disease could be both traumatic and septic in nature, which is not far from its counterpart in cattle. Longstanding degenerative joint disease might progress to ankylosing arthritis. Radiography is a cheap and effective imaging modality for diagnosing bone lesions in camel.
COMPUTED TOMOGRAPHY, MAGNETIC RESONANCE IMAGING, AND GROSS ANATOMICAL STUDIES ON THE HEAD OF THE OSTRICH (STRUTHIO CAMELUS)
H. Mohyeddin, A. Veshkini From the Department of Clinical Sciences, College of Veterinary Medicine, Garmsar Branch, Islamic Azad University, Garmsar, Iran (Mohyeddin) and Department of Diagnostic Imaging, College of Veterinary Medicine, Tehran Sciences and Research Branch, Islamic Azad University, Tehran, Iran (Veshkini).
Background:
The present study was performed to provide a detailed anatomic, computed tomography (CT), and magnetic resonance imaging (MRI) description of the structures of the head in the ostrich. The bony structures of the skull such as beak, hyoid apparatus, sinuses, and orbital structures were also evaluated by 3D reconstruction volume rendering in CT. Soft tissue structures of the head such as eye, brain, pharynx, and larynx were better evaluated by MR imaging in three saggital, coronal, and dorsal plans. This study provides detailed anatomic information of the ostrich head as seen with CT and MRI.
Purpose:
The current study aimed not only to provide a reference for MRI, CT, and anatomic description of the ostrich's head but also create a basic study for comparison of sinuses, airways, and eye and orbit of the animals living in the desert and evaluate the differences of these structures with other animals.
Methods and findings:
Seven clinically normal adult ostrich heads (four males and three females) were collected from a slaughterhouse. The heads were scanned with a 64‐slice CT scanner in 1 mm section immediately in bone and soft tissue window and then MR images were taken by 0.5 Tesla scanner in three plains and T1 and T2 and proton density sequences. Then, the heads were frozen in −20°C. After 3 days the heads were sectioned in 3 mm and each section in CT and MRI were labeled according to anatomical section. Three‐dimensional volumetric rendering reconstruction was performed in bony structure of the head. Sinuses, eye, orbit, brain, pharynx, and laryngeal structures were identified and compared with anatomical part of the same structures.
Conclusion:
Our study contributes to a better anatomical knowledge of the head of the ostrich by means of CT and MRI. These data are useful initial reference for clinical study and compare airway, sinuses, eye, and other structures of the head of the desert habitat with other animals. The eyes of this animal‐like ostrich and camel are large and they have identifiable compartment like big eyelid and eyelash and lacrimal and harderian glands to prevent environmental pathogens. Also narrow upper airways and big sinuses prevent hot air and dust to enter in lower respiratory system.
CANINE CENTRAL TARSAL BONE AS REVEALED BY HIGH‐RESOLUTION COMPUTED TOMOGRAPHY AND FINITE‐ELEMENT ANALYSIS
G. Galateanu, I. I. Feibis, F. gΦritz, A. D. Gruber, T. B. Hildebrandt, R. Shahar From the Reproduction Management Department, Leibniz Institute for Zoo and Wildlife Research, Berlin, Germany (Galateanu, Gφritz, Hildebrandt), Koret School of Veterinary Medicine, The Hebrew University of Jerusalem, Rehovot, Israel (Feibis, Shahar), and Institute of Veterinary Pathology, Freie University of Berlin, Berlin, Germany (Gruber).
Background:
Central tarsal bone (CTB) fractures in racing Greyhounds have been the subject of considerable investigation, yet, the mechanism of these fractures is not fully understood and the actual extrapolation (talar compression) from human medicine may not be accurate. The main question that persists is why 96% of CTB fractures occur on the right leg (situated in adduction) when turning on bends in counterclockwise races?
Purpose:
This study was designed to shed light on the role played by talar compression on CTB's body and calcaneal impingement on the plantar process of the CTB (PPCTB) in the pathogenesis of CTB fractures.
Methods:
Using state‐of‐the‐art tools of 128‐slice computed tomography (CT), we investigated the calcaneal and talar tridimensional (3D) position relative to CTB and PPCTB in 221 tarsal joints from 101 dogs of 45 different breeds and 19 mixed‐breed dogs. Additionally, multiple CT scans were acquired postmortem from eight dogs’ distal hind limbs excerpts, with the motion range varying between abduction, adduction, hyperflexion, and hyperextension. In order to evaluate the 3D distribution of stresses and strains in the CTB under different loading conditions, we obtained a high‐resolution CT scan of the bone of a German Shepherd cross dog. All slices of the scan were segmented, and image‐processing software was used to create a 3D model, which was meshed into 53,574 tetrahedral elements to create a finite‐element model. The model included several sets of boundary conditions (applied forces and distal restraint) to allow parametric analysis. Finite‐element analysis was then performed using Patran software.
Findings:
Finite‐element analysis provided evidence for differences in the location of maximal stresses for different loading scenarios, and in particular higher stresses were located in areas reported to experience higher frequency of clinical fractures when loads were applied to the plantar process of the CTB. High‐resolution CT images revealed that the talus and calcaneus perform a complex movement of circumduction. Consequently, the calcaneal contact surface with the PPCTB is greater in hyperextension than in hyperflexion. This calcaneo‐PPCTB contact surface is also greater in adduction than in abduction. In brief, the complex pattern of the movement of CTB, talus, and calcaneus provides a maximal contact area between the calcaneus and PPCTB in adduction and hyperextension and minimal contact area in abduction and hyperflexion.
Conclusion:
Such insights are important not only because they enhance the basic understanding of tarsal kinematics and stress and strain distribution, but also because they could lead to the design of new diagnostic strategies in racing Greyhounds.
AGE‐RELATED CHANGES IN SUBCHONDRAL BONE DENSITY IN THE TALUS OF HEALTHY LABRADOR RETRIEVERS
W. Dingemanse, M. Muller‐gerbl, I. Jonkers, J. Vander Sloten, H. Van Bree, I. M. V. L. Gielen From the Department of Veterinary Medical Imaging and Small Animal Orthopaedics, Faculty of Veterinary Medicine, Ghent University, Ghent, Belgium (Dingemanse, van Bree, Gielen), Institute of Anatomy, Basel University, Basel, Switzerland (Muller‐Gerbl), Faculty of Kinesiology and Rehabilitation Sciences, Human Movement Biomechanics Research Group (Jonkers), and Faculty of Engineering Science, Biomechanics Section, KU Leuven, Leuven, Belgium (Vander Sloten).
Background:
The articular cartilage and the underlying subchondral bone, supporting the cartilage, form a functional unit. Subchondral bone density is influenced by joint biomechanics and limb function and reflects the loading history of the joint. Subchondral bone density can be evaluated using computer tomographic osteoabsorptiometry (CTOAM).
Purpose:
The evaluation of subchondral bone density patterns provides more information on long‐term joint loading. Joint conformation, age, and (orthopedic) pathology all contribute to a specific density pattern. The goal of this study was to evaluate the subchondral bone density in the talus of healthy Labrador Retrievers of different ages. The hypothesis was an increase in subchondral bone density in the older dogs, comparable with findings in humans.
Methods:
Computer tomographic (CT) images (four‐slice helical CT scanner Lightspeed Qx/i, General Electric Medical Systems, Milwaukee, WI) were acquired. The CT parameters were 120 kVp and 300 mAs. Contiguous, 1.25 mm collimated, transverse images were obtained using a soft tissue reconstruction algorithm. The CTOAM workflow was completed using commercially available computer software (Analyze, Mayo Clinics). Computer tomographic data were obtained from eight Labrador Retrievers at ages of 10 and 18 months. Overall density and the location of density maxima were compared between left and right joints and between different ages.
Findings:
The subchondral bone density distribution was very similar between left and right tali and between different dogs. The medial and lateral trochlear ridges showed a distinctly different density pattern. In general, the lateral trochlear ridge showed a higher overall density in comparison to the medial trochlear ridge. Two distinct density maxima were seen in all joints, one proximally on the medial talar ridge and one distally on the lateral talar ridge. The overall density was significantly increased at the ages from 10 to 18 months, however the density distribution was similar.
Conclusion:
Using CTOAM, regional density variation in the subchondral bone can be visualized in the canine talus. In Labrador Retrievers, a consistent distribution, with two separate density maxima, was found. In younger dogs, the location of the density maxima is the same, but the overall density is lower. The strength of CTOAM in biomechanical research is its ability to show the morphological effects of forces acting on the joint. Actual joint loading is very difficult to determine in vivo. Because CT is a noninvasive technique, this method can be applied in a clinical setting to study both healthy joints and joint pathology. This way, CTOAM can provide new insights in physiological joint loading, and alterations in pathological conditions.
POSTMORTEM COMPUTED TOMOGRAPHY FINDINGS AND CORE NEEDLE BIOPSY IN COMPARISON TO AUTOPSY IN EIGHT BERNESE MOUNTAIN DOGS WITH HISTIOCYTIC SARCOMA
F. C. Hostettler, D. J. Wiener, M. M. Welle, H. Posthaus, U. Geissbuhler From the Clinical Radiology, Department of Clinical Veterinary Medicine (Hostettler, Geissbuhler) and Department of Veterinary Pathology, Vetsuisse Faculty, Bern, Switzerland (Wiener, Welle, Posthaus).
Background:
With the rising esteem of companion animals in our society, the owner compliance for autopsy, which is the gold standard for postmortem examination, is diminishing. Minimally invasive techniques might increase the acceptance of investigations after death.
Purpose:
The aim of this study was to compare the findings of postmortem computed tomography (pm‐CT) combined with core needle biopsy to conventional autopsy in Bernese Mountain Dogs with confirmed histiocytic sarcoma.
Methods:
The current study is part of a project performed in contribution with the Swiss Club for Bernese Mountain Dogs, a Swiss animal crematory, the Department for Pathology and the Clinical Radiology of the Vetsuisse Faculty of the University of Bern. Dog owners and breeders were informed via the club and the crematory about our interest in spontaneously dead or euthanized dogs for the study. A postmortem whole body CT was performed at the Clinical Radiology. CT‐guided core needle tissue samples were taken immediately afterwards. Then a conventional autopsy was performed at the Department of Veterinary Pathology.
Findings:
CT revealed in all eight dogs space‐occupying focal or multicentric lesions, with characteristics suspicious for histiocytic sarcoma. Such lesions were consistently detected in the lung (5/5), cranial mediastinum/thoracic lymph nodes (6/6), spleen (4/4), and kidney (1/1). Hepatic lesions, however, were missed in pm‐CT in three of four cases. Three lesions were suspicious in pm‐CT but not confirmed in autopsy (mesentery, right medial iliac lymph node, lung (2)). Core needle biopsy revealed a definite diagnosis in six of 10 bioptied organs. A definite diagnosis of histiocytic sarcoma was put in five of eight biopsied dogs. In one dog, no neoplastic cells were identified in none of the core tissue samples. In two dogs, histiocytic sarcoma was part of the differentials but a definite tumor classification was not possible. In one dog, histiocytic sarcoma was confirmed by core tissue biopsy but just in the list of differential diagnoses in autopsy.
Conclusion:
Postmortem computed tomography turned out to be a fast, dignified, and effective method to develop lesions suspicious for histiocytic sarcoma in pulmonary, cranial mediastinal/thoracic lymphatic, splenic, and renal tissue of Bernese Mountain dogs. Possible reasons for missing a definite diagnosis in core tissue biopsy include autolysis, small tissue sample size, central tumor necrosis, and skills of the histologist. Investigations have to be focused on these factors to improve the accuracy of the method.
ADIPOSE‐DERIVED CANINE STEM CELLS: BEHAVIOR AFTER ENDOREM® LABELING FOR MAGNETIC RESONANCE IMAGING
K. Kolecka, K. H. Amort, M. J. Schmid, C. Reich, M. Kramer, S. Arnhold From the Department of Veterinary Clinical Sciences, Small Animal Clinic (Kolecka, Amort, Schmid, Kramer) and Institute of Veterinary Anatomy, Histology and Embryology, Justus Liebig University Giessen, Giessen, Germany (Reich, Arnhold).
Background:
Clinical tracking of canine mesenchymal stem cells has been limited to a small number of reports in the past. A detailed long‐term follow‐up of behavior and signal intensity is necessary before an application of stem cells in vivo.
Purpose:
Purpose of the study was to assess the influence of Endorem® labeling on the proliferation ability, cytoskeleton, and multipotency of the stem cells. Agar phantoms with labeled cells were analyzed in T2TSE and T2FFE to show the duration of Endorem® labeling using a 1 Tesla MRI tomograph.
Methods:
Seven samples with canine adipose‐derived stem cells were incubated with the Endorem® concentration of 319.2 μg/ml Fe for 24 h. The efficiency of the labeling was investigated with the aid of the Prussian blue staining and the MACS method. Cytoskeleton and proliferation were assessed. Agar phantoms with labeled cells were analyzed in T2TSE and T2FFE MRI sequences 1, 2, and 3 weeks after Endorem® labeling. Histograms and unlabeled negative controls were used to determine the signal intensity of the samples.
Findings:
Iron particles were detectable within the cells 3 weeks after labeling with Endorem®. The percentage of the labeled canine stem cells decreased within 3 weeks. Phalloidin staining showed that the present concentration of iron particles had no negative influence on the cytoskeleton. Labeled cells underwent osteogenic and adipogenic differentiation. The chondrogenesis was negatively affected. The labeling had no adverse effects on the proliferation of the stem cells. It was determined in both sequences (T2TSE and T2FFE) that the hypointensity caused by Endorem® lasted at least for 3 weeks.
Conclusion:
The present study shows that Endorem® is a suitable MRI contrast agent for clinical tracking studies with canine adipose‐derived stem cells at 1 Tesla.
BONE MINERAL DENSITY EVALUATION BY MEANS OF QUANTITATIVE COMPUTED TOMOGRAPHY UTILIZING A CALIBRATION PHANTOM
G. Galateanu, J. Saragusty, C. A. Szentiks, Z. Mezo, G. Straus, J. Mews, T. B. Hildebrandt From the Leibniz Institute for Zoo and Wildlife Research, Berlin, Germany (Galateanu, Saragusty, Szentiks, Mezo, Hildebrandt), Tierpark Berlin, Berlin, Germany (Straus), and Toshiba Medical Systems Europe, Clinical Application Research Center, Zoetermeer, The Netherlands (Mews).
Background:
The exact modus operandi for performing quantitative computed tomography (qCT) using a dedicated bone mineral density (BMD) calibration phantom for veterinary clinician is a conundrum, in spite of scientists’ and clinicians’ increasing interest in this subject matter.
Purpose:
Whereas qCT scanning protocols in human medicine are highly standardized, these strict conditions cannot be met by nonhuman subjects.
Methods:
For establishment of reliable protocols for BMD evaluation in autopodial bones (podial, metapodial, and phalangeal elements), we conducted a qCT study employing a high‐resolution, 128‐slice CT scanner (Aquilion CX, Toshiba, Tochigi, Japan), a BMD calibration phantom (B‐MAS 200, Kyoto, Japan) with five rods equivalent to five different bone densities, and a dedicated multisoftware workstation (ViTREAR 2 version 4.0 medical diagnostic software; Vital Images Inc., Minnetonka, MN). One hundred fifty‐six limbs from 39 wild hoofed animals, obtained postmortem, underwent helical CT scans. In order to avoid any inaccuracy produced by a narrow bone sample, the following conditions were produced: (1) studied specimens included a wide variety of species and subspecies (n = 27); (2) a large spectrum of ages was encompassed, ranging from fetus, through newborn, subadult, adult to geriatric; (3) various autopodial bones were measured; (4) different bone compartments (cortical, trabecular, mixed) and the whole bone were assessed. Two protocols were used for each specimen: “Protocol A” with 120 kVp, variable mA, rotation time 0.6 s; and “Protocol B” with 100 kVp, 300 mA, rotation time 1.0 s. Some excerpts underwent additional CT scans maintaining identical exposure factors as in protocols A and B, but varying CT couch height, object‐BMD phantom distance and positioning, foot recumbency, slice thickness, fresh vs. frozen excerpts.
Findings:
Significant statistical differences were found between qCT studies performed with dissimilar CT protocols: (1) when different scanning settings were applied (fixed vs. variable exposure factors); (2) the center of scanning field of view had changing locations (object‐centered/BMD phantom centered/ noncentered); (3) at inconstant object‐BMD phantom distances; (4) slice thickness higher than 3 mm; (5) dissimilar object relative to phantom positioning (orthogonal vs. oblique). The analyzed groups did not differ statistically: (1) at slice thicknesses of 0.5 mm, 1 mm, and 2 mm; (2) between fresh and frozen (at –20°C) excerpts; (3) between dorsal and palmar/plantar recumbencies.
Conclusion:
Recognizing the scarcity of precise methodological data on BMD evaluation in veterinary medicine, this study provides important reference data, a modus operandi, and basis for accurate qCT analysis performed with a dedicated BMD calibration phantom.
ASSOCIATION BETWEEN CLINICAL IMPROVEMENT AND RESTORATION OF THE BRAIN IN DOGS WITH HYD ROCEPHALUS AFTER CSF SHUNTING
M. J. Schmidt, K. H. Amort, M. Kramer, N. Ondreka From the Department of Veterinary Clinical Sciences, Clinic for Small Animals, Justus‐Liebig University, Giessen, Germany (Schmidt, Amort, Kramer, Ondreka).
Background:
Congenital hydrocephalus is one of the most common malformations of the central nervous system in dogs characterized by abnormal dilatation of the cerebral ventricular system. The benefit of ventriculoperitoneal shunt (VPS) implantation and its potential for long‐term survival of affected animals has been demonstrated recently. One challenge remaining in recognition of the beneficial effects of shunting is understanding the association between decrease of ventricular volume, reconstitution of the cerebral mantle, and clinical response in dogs. The efficacy of VPS procedures in reversing the pathophysiological consequences of increased intraventricular pressure and its effect on the neurological outcome have been reported to vary. Some clinical signs seem to be reversible after cerebrospinal fluid (CSF) drainage whereas others are somewhat persistent. It is unclear whether the absence of therapeutic benefit is a consequence of insufficient or arresting ventricular drainage or a result of permanent tissue destruction persisting after reduction of CSF pressure.
Purpose:
The purpose of this prospective study was therefore twofold: first, to examine whether restoration of the cerebral mantle occurs in dogs being treated for hydrocephalus during a follow‐up period of at least 3 months, and second, to assess whether a volume reduction may co‐occur with a positive clinical response to shunting.
Methods:
Twenty‐three dogs underwent magnetic resonance imaging (1.0T, Philips Intera) immediately before, immediately after, and 3 months after surgical VPS implantation. The volumes of the ventricular system and cerebral hemispheres were determined at both time points and compared. A two‐tailed Fisher's exact test was used to observe the association between volume reduction of the cerebral mantle and clinical signs of the dogs.
Findings:
Reduction of the ventricular volume and restoration of the cerebral tissue could be documented in 16 of 23 dogs. Improvement of clinical signs was noted in 15 of 23 dogs, the remaining seven dogs did not show improvement. Fourteen of 15 dogs with clinical improvement and two of seven dogs without clinical improvement showed reconstitution of the cerebral mantle. The association between reconstitution and improvement of clinical sings was statistically significant (P = 0.001).
Conclusion:
Based on the results of this study comparative volume measurements of pre‐ and postsurgical ventricular size and cerebral parenchyma could turn out to be important prognostic factors in prediction of long‐term outcome of dogs with hydrocephalus after VPS.
MAGNETIC RESONANCE DIFFUSION TENSOR IMAGING IN DOGS WITH COMPRESSIVE AND NONCOMPRESSIVE SPINAL CORD INJURY
G. Middleton, L. Gaschen, K. Saile, J. Wignall, M. Kearney From the Department of Veterinari Clinical Sciences, Louisiana State University, Baton Rouge, LA (Middleton, Gaschen, Saile, Wignall, Kearney).
Background:
Diffusion tensor imaging (DTI) is an MRI sequence that can be used to characterize the directional properties of the diffusion of water molecules in biological tissues. Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) represent axonal injury when abnormal.
Purpose:
The purpose of this study is to identify the utility of DTI for assessing neuronal integrity in dogs with compressive and noncompressive spinal cord lesions and comparing FA and ADC with the degree of compression, spinal cord signal change, chronicity of clinical signs, anti‐inflammatory administration, and patient outcome.
Methods:
The MRI examination of dogs that presented for clinical signs localized to the spinal cord over a 3‐year period was included in the study. Fractional anisotropy and ADC values were obtained. Dogs were divided into six categories based on presenting and discharge neuroscores. MRI lesion type was recorded and patients were divided into compressive and noncompressive groups. Patients within the compressive group were further divided into degree of compression as mild (<25%), moderate (25‐50%), or severe (>50%). Values for FA, ADC, and DWI were obtained from the affected cord region.
Findings:
One hundred and two total canine patients included in the study with 88 categorized within the compressive group and 14 within the noncompressive group. Forty‐six total patients had intramedullary hyperintense signal including 34 patients in the compressive group and 12 patients in the noncompressive group. Presumptive diagnosis of noncompressive patients included granulomatous eningoencephalitis (n = 7), Hansen's type 3 intervertebral degenerative disc disease/fibrocartilaginous embolism (n = 2), myelomalacia (n = 2), degenerative myelopathy (n = 2), and peripheral nerve sheath tumor (n = 1). There was a statistically significant decrease in FA, increase in ADC, and increase in DWI in patients within the noncompressive group compared to compressive. No significant changes in DTI parameters between degrees of compression. Statistically significant increase in DWI in patients with intramedullary hyperintensity though FA and ADC values did not vary statistically between patients with and without spinal cord signal change. Neither DTI parameters nor length of spinal cord signal change differed statistically with patient outcome/neuroscore. Neither chronicity of clinical signs nor anti‐inflammatory administration prior to presentation correlated with DTI parameters.
Conclusion:
These results indicate that there is a difference in diffusion between compressive and noncompressive lesions; however, DTI does not correlate with degree of compression or presence of hyperintensity. Neither DTI nor length of intramedullary hyperintensity correlated with patient outcome and therefore should not be used as a predictor of patient prognosis in the population examined. These findings are in agreement with a previous study performed by Pease et al. It remains unclear what duration or degree of compression is required for neuronal degeneration to occur and it is possible that such pathology was not even present in our study population.
FUNCTIONAL MAGNETIC RESONANCE IMAGING TO ESTIMATE NORMAL DIFFUSION IN THE CANINE BRAIN
A. Hartmann, C. SOffler, K. Failing, A. Schaubmar, M. Kramer, M. J. Schmidt From the Clinic for Small Animals (Surgery/ Radiology) (Hartmann, Soffler), Unit for Biomathematics and Data Processing (Failing, Schaubmar), Clinic for Small Animals (Surgery) (Kramer), and Clinic for Small Animals (Surgery/ Neurology), Justus‐Liebig‐University, Giessen, Germany (Schmidt).
Background:
In human medicine diffusion‐weighted magnetic resonance imaging (DWI) is used in the diagnosis of cerebral ischemia, epilepsy, brain neoplasms, and encephalitis. In veterinary medicine DWI is still in its infancy.
Purpose:
The purpose of this study was to evaluate the normal distribution of diffusion in selected brain areas of healthy dogs to increase the knowledge of the principle function of canine brains and to provide baseline information for further studies.
Methods:
Ten healthy Beagle dogs with an average age of 2.4 ± 0.6 years were examined. MR examination was performed using a 1.0 Tesla MR system and a sensitivity encoding coil. Diffusion‐weighted images using a single shot echo planar imaging sequence with a b‐value of b = 0 s/mm tot de macht 2 and b = 800 s/mm tot de macht 2 were acquired in a dorsal and transverse plane with diffusion gradients in all three planes. On the apparent diffusion coefficient (ADC) maps regions of interest were drawn around the caudate nucleus, thalamus, piriform lobe, hippocampus, semioval center, and cerebral cortex and the ADC values were recorded.
Findings:
There are significant differences between the hippocampus and the semioval center compared to all other regions and differences between thalamus, caudate nucleus, and piriform lobe. The hippocampus shows the highest ADC value (1044.29 ± 165.21 × 10−6 mm2/s [mean ± standard deviation]). The semioval center shows the lowest ADC value (721.39 ± 126.28 × 10−6 mm2/s). Thalamus (807.82 ± 128.41 × 10−6 mm2/s), cerebral cortex (843.15 ± 98.24 × 10−6 mm2/s), caudate nucleus (902.85 ± 193.88 × 10−6 mm2/s), and piriform lobe (915.44 ± 159.84 × 10−6 mm2/s) have similar ADC values. We also discovered a significant difference between the average ADC value of the right and left cerebral hemispheres. The left hemisphere, except for the hippocampus, shows higher ADC values. Significant differences exist for the caudate nucleus, thalamus, and piriform lobe.
Conclusion:
Differences between brain regions can be attributed to their affiliation to white and gray matter and therefore to differences in myelination and neuronal density. The semioval center (white matter structure) shows the lowest ADC value due to myelination that limits diffusion. The remaining regions (gray matter structures) show higher ADC values, may be due to neuronal density. Variations between the right and left hemisphere could be the expression of differences in anisotropy, for example caused by pawedness. These results give a first estimate of normal distribution of diffusion in canine brains. A variation from reported ADC values could be a sign for pathologies.
SHAPE OF THE FELINE CEREBELLUM AND OCCIPITAL BONE RELATED TO BREED ON MAGNETIC RESONANCE IMAGING OF 200 CATS
X. Huizing, A. Sparkes, R. Dennis From the Animal Health Trust, Lanwades Park, Kentford, Newmarket, Suffolk, UK (Huizing, Dennis) and International Cat Care, High Street, Tisbury, Wilts, UK (Sparkes).
Background:
The relationship between skull morphology and cerebellar herniation has been investigated in various species (dogs, lions, and more recently alpacas) but studies in cats are less abundant. In our experience cats display a wide diversity in shape of the cerebellum and occipital bone on magnetic resonance imaging (MRI) with the vermis sometimes appearing to be impacted into, or herniated through, the foramen magnum. Several anatomical studies have been conducted to characterize the shape and morphology of the feline skull as well as quantifying various skull parameters but the MRI features of the feline cerebellum have not previously been described in the literature.
Purpose:
The aim of this study was to investigate the variation in cerebellar and occipital bone shape on MRI in neurologically normal cats and to assess whether there was any correlation with breed. Our hypothesis was that, in absence of clinical neurological signs, feline cerebellar herniation is more prevalent in brachycephalic cats.
Methods:
Imaging records at the Animal Health Thrust were searched retrospectively between 2000 and 2013 to include adult cats that underwent MRI investigation that included the brain. Inclusion criteria included a normal neurological examination and an absence of evident intracranial disease on MRI. MRI was performed for a multitude of reasons including nasal disease, head tilt, and respiratory problems. Since the MRI was done as part of a clinical workup, no ethical approval was necessary; owner consent is obtained before every scan as a routine procedure in our hospital. Two hundred cats were included and midline sagittal and transverse T2W images were used to document the shape of the cerebellum and occipital bone. From the MR images, the morphology of the cerebellum was matched to one of the following three categories: convex caudal surface with rounded caudoventral margin, flattened angular caudoventral margin, and herniation of the cerebellar vermis through the foramen magnum. The width and length of the cranial cavity were also measured and an intracranial cavity length to width ratio (ICCR) was calculated and used as an indication of brachycephalia.
Findings:
Our results showed that cerebellar morphology varied strongly and significantly (P = 0.0099) between breeds with a significant higher incidence of herniation in Persian cats compared with other breeds. Variations in occipital bone shape were not characteristic of specific breeds but morphology did differ significantly (P = 0.006) between male and female cats. Our results also showed a significant difference (P < 0.0001) in median values of the ICCR between breeds, with Persian cats having a higher ICCR compared to every other breed. Importantly, high ICCR was significantly (rs = 0.1986, P = 0.006) correlated to cerebellar herniation.
Conclusion:
In conclusion, our results showed that breeds with a higher ICCR, that is brachycephalic breeds, and Persians in particular, exhibit a significantly higher incidence of cerebellar shape consistent with herniation of the vermis. It is important to recognize this morphological variation in order to avoid false‐positive diagnoses of raised intracranial pressure and pathological herniation on MRI.
SUSPECTED HYDRATED NUCLEUS PULPOSUS EXTRUSIONS ASSOCIATED WITH THE DORSAL LONGITUDINAL LIGAMENT IN DOGS
M. Dolera, L. Malfassi, G. Mazza, S. Finesso, M. Sala From the La Cittadina Fondazione Studi e Ricerche Veterinarie, Romanengo, CR, Italy (Dolera, Malfassi, Mazza, Finesso, Sala).
Background:
Disc‐associated intraspinal cysts and cervical hydrated nucleus pulposus extrusions are lesions of suspected disc origin recently described in veterinary literature.
Purpose:
To describe MRI, surgical and cytological findings in patients with acute hydrated cervical disc extrusions were associated with the dorsal longitudinal ligament (DLL).
Methods:
Prospective study of dogs with evidence of cervical spinal cord compression by material associated with the DLL. Inclusion criteria were: availability of high‐field MRI examination, microsurgical decompression, and cytological examination of the removed material. The following MRI features were assessed: grading of spinal cord compression, intervertebral spaces involved, signal intensity and distribution of the material, T2‐W spinal cord signal, thickness, and signal intensity of the discs. Spinal cord compression was evaluated in cross‐sectional T2‐W images. Surgical findings considered were: macroscopic appearance of the material, its localization, and relationship with the DLL.
Findings:
Thirty‐six patients satisfied inclusion criteria. The anatomical site was C4‐C5 in 88%, C3‐C4 in 12% of cases. Although on MRI the lesions were found to be extradural and ventrally located under the DDL, ligament incision had to be performed in all patients to permit lesion removal, indicating an intraligamentous localization. On MRI, lesions had signal intensity similar to the cerebrospinal fluid in both T1‐W and T2‐W sequences. The distribution in transverse images was central in all cases. The thickness of the corresponding discs was slightly reduced in 75% of cases. Signal intensity of the nucleus pulposus was normal in 82% of cases, and slightly reduced in T2‐W sequences in 18%. Spinal cord compression ranged from 13.5% to 57.2% (mean 36.5%). At surgery the material was found to be liquid with a gelatinous component in 42% of cases, plain liquid in 33%, liquid with frustules in 25%. Although a direct cytological diagnosis of disc material is not possible, cytological examination allowed to exclude the presence of inflammation, bacteria, mycosis, neoplasia, or foreign material.
Conclusion:
In this study, 36 cases of suspected hydrated cervical disc extrusions are described. Findings were consistent with an intraligamentous localization of extruded disc material, and cytology excluded other underlying disease processes. A similar appearance has been previously described with disc‐associated intraspinal cysts and further studies are needed to clarify the pathophysiology of the diseases in dogs.
NECK PAIN IN DOGS AND CATS
M. Dolera, L. Malfassi From the La Cittadina Fondazione Studi e Ricerche Veterinarie, Romanengo, CR, Italy (Dolera, Malfassi).
Background:
Epidemiological studies on neck pain have not been conducted. Imaging findings of canine and feline diseases that cause neck pain are not well described.
Purpose:
The aim of this research was to define the causes of neck pain in dogs and cats and describe typical MRI patterns of pathologies that cause neck pain in order to establish the sensitivity and specificity of the method.
Methods:
A retrospective clinical study considered 178 animals (nine cats and 169 dogs) with neck pain examined with MRI. The inclusion criteria were neck pain identified at clinical examination, an MRI imaging of head, neck and chest, a confirmed diagnosis based on cytology–histopathology or surgical findings or lab work, and a complete follow‐up. Data considered were: localization of lesions determining the symptomatology, type of pathology responsible for neck pain (malformation, traumatic, vascular, inflammation, degenerative, neoplastic), primary or indirect involvement of nervous system, any neurological or systemic signs, and imaging findings.
Findings:
Ninety‐five percent of patients suffered from neurological pathology whereas 5% were non‐neurological. Lesions were in the neck in 72.5% of cases, in 17.4% both head and neck were involved. In 9% of cases the lesions were only in the head, in 1.1% in the chest. 52.5% of dogs suffered from intervertebral degenerative disc disease (IVDD), 19.5% were neoplastic, 14.5% inflammatory, 8.5% malformation, 2.5% traumatic, 2.5% vascular. A total of 66.6% of cats were neoplastic, 22.2% traumatic, 11.2% inflammatory. Seventy‐eight percent of cases showing a cervical spinal lesion had IVDD. Seventy‐seven percent of cases with brain and spinal lesions had an inflammatory disease, whereas in all cases of neck pain with a lesion located in the head, the pathology is neoplastic. Sensitivity and specificity of MRI were, respectively, 100/100 for malformation, 100/75 for vascular diseases, 95/90 for inflammations, 96/94 for degenerative diseases, 94/89 for neoplastic diseases.
Conclusion:
The findings of this study indicate that lesions responsible for neck pain can localize in the neck, head, or chest. Pain and lesion localization does not always correlate. MRI has a high diagnostic sensitivity and specificity for diseases that cause spinal pain. Based on the results of this study, MRI can be the method of choice to investigate neck pain.
CANINE CARDIAC MAGNETIC RESONANCE IMAGING: COMPARISON OF ASSESSMENT OF STROKE VOLUME BASED ON MORPHOLOGY AND PHASE CONTRAST ANGIOGRAPHY
M. Dennler, H. Richter, J. Novo Matos, M. Baron Taldo, T. Glaus, P. R. Kircher From the Section of Diagnostic Imaging (Dennler, Richter, Kircher) and Division of Cardiology, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland (Novo Matos, Baron Taldo, Glaus).
Background:
The heart and the pulmonary and systemic circulations form a closed system. Average stroke volume is equivalent in the left and right ventricle in the normal heart, independent of measuring technique and location.
Purpose:
Objectives of the study were to compare stroke volume as a functional parameter based on morphological image information and time compensated phase contrast angiography acquired by cardiac magnetic resonance imaging (cMRI) in healthy Beagle dogs.
Methods:
Ten healthy Beagle dogs, four male and six female with a mean age of 3.8 ± 0.9 years and a mean body weight of 13.3 ± 2.3 kg underwent cMRI. Cine gradient echo sequences (sBTFE, white blood sequence) in short axis (SAX) of the left and right ventricle served for quantification of stroke volume based on morphology by drawing endo‐ and epicontours of the myocardium in each plane and phase. From the long axis views stroke volume was acquired from left ventricular two (2Ch) and four chamber (4Ch) planes based on the two phases in end diastole and end systole. Two‐dimensional, time‐resolved phase contrast angiography was performed over the aorta (AV), pulmonary artery (PV), and both atrioventricular (MV and TV) valves. These sequences allowed quantification of stroke volume based on flow information in each location. Due to small sample size, only descriptive statistics were performed.
Findings:
Mean standard deviation of the results of flow quantifications (AV: 16.9 ± 4.9 ml, PV: 18.4 ± 6.2 ml, MV: 16.9 ± 5 ml, and TV: 17.6 ± 6.5 ml) and morphologic quantification of the left ventricular stroke volume in long axis (4Ch: 15 ± 4.1 ml; 2Ch: 20.2 ± 4.4 ml) varied minimally. Morphologic quantification in short axis gave somewhat small left (13.6 ± 3.2 ml) and right (9.4 ± 2.8 ml) ventricular stroke volumes.
Conclusion:
Flow quantification using cardiac magnetic resonance imaging is a feasible, simple method for the assessment of cardiac left and right ventricular function in dogs. Assessment of stroke volume based on morphologic image information resulted in more variable measurements, especially on the right side, probably due to the complex anatomy of the right ventricle.
COMPUTED TOMOGRAPHY, RADIOLOGY, AND ECHOCARDIOGRAPHY IN CATS NATURALLY INFECTED WITH AELUROSTRONGYLUS ABSTRUSUS
G. Lacava, E. Zini, O. Domenech, F. Romano, S. Manzocchi, E. Auriemma From the Department of Radiology, Istituto Veterinario di Novara, Granozzo con Monticello, NO, Italy (Lacava, Zini, Domenech, Romano, Manzocchi, Auriemma).
Background:
Feline aelurostrongylosis represents a common parasitic pulmonary infection in cats. Production of eggs and larval migration induce a massive inflammatory response characterized by severe interstitial pneumonia with alveolar, bronchial, and arterial lesions. So far, few cases described the imaging findings of cats with naturally occurring aelurostrongylosis.
Purpose:
The aim of the study was to describe radiographic, computed tomographic (CT), and echocardiographic findings in cats naturally infected with Aelurostrongylus abstrusus.
Methods:
Inclusion criteria were positive results to fecal parasitology (Baermann test) and being treatment naive. The age of cats ranged from 8 to 120 months (mean 33 months). There were seven males (50%) and seven females (50%) and all cats were domestic shorthair cats.
Findings:
The most common presenting clinical abnormalities were coughing (10/14, 71%) and dyspnea (5/14, 36%). Radiographic findings were generalized unstructured interstitial pulmonary pattern (8/14, 57%), mixed structured (nodular) broncho‐interstitio‐alveolar infiltrate (3/14, 21%) and broncho‐interstitial pattern with bronchial walls thickening (3/14, 21%). Sternal lymphadenopathy was detected on thoracic radiographs in six cats (42%). On CT, imaging features were mixed broncho‐interstitio‐alveolar infiltrate with ground‐glass appearance in six (42%), interstitio‐alveolar with multiple pulmonary nodules in five (36%), interstitial ground glass in three (21%), regional lymph node enlargement in 11 cats (10 sternal, three cranial mediastinal, and three tracheobronchial lymph nodes), and subpleural thickening in four (28%). None of the thoracic radiographs revealed subpleural thickening. Lymphadenopathy documented with thoracic radiography was also detected by CT scan, whereas four cases documented with the latter were not using the former. In one case suspected lymphadenopathy based on radiography was not confirmed with CT. In all cases pulmonary vessels were within normal limits for size, shape, and attenuation. No echocardiographic evidence of pulmonary hypertension or cardiac abnormality was found in any cat.
Conclusion:
In conclusion, CT yielded more through characterization of pulmonary and mediastinal lesions when compared with thoracic radiographs in cats naturally infected with A. abstrusus. Although feline aelurostrongylosis has been previously associated with severe histopathological lesions of lung arteries, in the present study no evidence of pulmonary hypertension was detected according to echocardiography.
ASSESSMENT OF PULMONARY HYPERTENSION IN DOGS WITH CHRONIC PULMONARY DISEASES BY THORACIC COMPUTED TOMOGRAPHY ANGIOGRAPHY
T. Couvreur, E. Roels, A. C. Merveille, C. Clercx, K. Mcentee, G. Bolen From the Department of Medical Imaging, Centre Hospitalier Chretien, Liege, Belgium (Couvreur), Internal Medicine Section, Department of Small Animals and Equidae Clinic, Faculty of Veterinary Medicine, University of Liege, Liege, Belgium (Roels, Merveille, Clercx, McEntee), Laboratory of Physiology and Physiopathology, Faculty of Medicine, Universite Libre de Bruxelles, Brussels, Belgium (McEntee), and Diagnostic Imaging Section, Department of Small Animals and Equidae Clinic, Faculty of Veterinary Medicine, University of Liege, Liege, Belgium (Bolen).
Background:
Pulmonary hypertension (PH) is a complex syndrome, which negatively influences quality of life and survival. Chronic alveolar hypoxia and lung parenchymal infiltration observed in cases of canine idiopathic pulmonary fibrosis (CIPF) and chronic lower airway diseases (CLAD) may lead to PH. In dogs, noninvasive diagnosis of PH and its repercussions on the right heart is still challenging.
Purpose:
In humans, thoracic CT imaging may assess PH. This study was intended to investigate, in normal dogs and dogs with CIPF or CLAD, the reproducibility of PH CT angiography (CTA) parameters and their correlation with similar echocardiographic variables.
Methods:
CTA images obtained from six dogs with CIPF (Group A), seven dogs with CLAD (Group B), and nine healthy controls (Group C) were retrospectively reviewed. The diameter of the right (RV) and left ventricles (LV) was obtained from transverse images (T) and both from manually reformatted four‐chambers (4C) and short axis (SA) cardiac views. Diameters of pulmonary trunk (PT) and aorta (Ao) and of the pulmonary right vein (PV) and its adjacent pulmonary artery (PA) were measured in T. Measurements were performed twice blindly by two radiologists. Measurements and ratios (PV/PACT, PT/AoCT and RV/ LVCT in the three planes) were compared between groups and correlated with same ratios measured by echocardiography (ultrasonography).
Findings:
Statistical analyses were performed using XLStatR software. Values are given as mean ± SD. Intra‐ and interobserver agreements were good, small bias were observed for all parameters except for RV in T and SA views. Pulmonary right vein /PACT was significantly lower in group A (0.547 ± 0.151) compared to group C (1.034 ± 0.196, P = 0.001). Values of group B were intermediate (0.859 ± 0.275). Pulmonary right vein /PACT correlated with PV/PAUS measured in both bidimensional (r = 0.777, P < 0.0001) and M‐modes (r = 0.638, P = 0.002). Pulmonary trunk /AoCT was significantly higher in groups A (1.178 ± 0.097) and B (1.235 ± 0.178) compared to group C (0.964 ± 0.128, P = 0.002) but no correlation was found with PT/AoUS. Right ventricle/LVCT in T, SA, and 4C views were significantly increased in group A (T: 0.873 ± 0.101; SA: 0.866 ± 0.065; 4C: 0.881 ± 0.065) compared to group C (T: 0.657 ± 0.114, P < 0.05; SA: 0.669 ± 0.068, P < 0.05; 4C: 0.710 ± 0.075, P < 0.05). Correlations between RV/LVCT, in T, SA, and 4C views, and RV/LVUS were found (T: r = 0.633, P = 0.002; SA: r = 0.647, P = 0.001; 4C: r = 0.608, P = 0.003).
Conclusion:
In conclusion, in dogs, PV/PA, PT/Ao, and RV/LV ratios measured on thoracic CTA images are reproducible and may serve in the assessment of PH and its repercussions on the right heart size. These thoracic CTA ratios correlate with similar echocardiographic ratios except for PT/Ao.
LEFT‐SIDED CAUDAL VENA CAVA (CVC), LEFT RETROCAVAL URETER, AND INGUINALRING STENOSIS ASSOCIATED WITH A TRONCULAR VASCULAR ANOMALY IN A DOG
S. Borgonovo, P. Pisani, F. Bolzoni, A. Bettinelli From the Clinica Veterinaria Crema, Crema, CR, Italy (Borgonovo, Bolzoni, Bettinelli) and Free Practitioner, Milano, MI, Italy (Pisani).
Background:
In human medicine, the diagnosis of congenital inferior vena cava anomalies and vascular malformations is more common in asymptomatic patients, and congenital retrocaval ureter occurs in approximately 1–5% of total births. In veterinary medicine, however, there are few cases describing an association between these diseases.
Purpose:
A 12‐month‐old neutered female Pit Bull Terrier was evaluated for mammary, vulvar, and left hind limb edema. No other physical abnormalities were detected. The hematology results did not reveal any significant anomalies. The patient underwent urogenital ultrasound, echocardiographic exam, and CT abdominal angiography.
Methods:
Urogenital ultrasound and echocardiographic exam were performed with a Mindray M5 Scan, equipped with a convex probe 35C50 EA (2.5–5 MHz) and a linear probe 75L38 EA (5–10 MHz). CT abdominal angiography was performed under general anesthesia with a 16‐slice Siemens Somatom CT Scanner (thick/index 1.5/1 mm, scan time 0.5 s), precontrast, 45 s after 800 mg iodine/kg iohexol (Omniscan 300) IV injection, and in the late phase after 60 s.
Findings:
Echocardiographic exam and urogenital ultrasound examination did not reveal any abnormalities. CT angiography did, however, show some abdominal anomalies. First of all, it revealed a malpositioned caudal vena cava (CVC): cranially, near the aortic trifurcation, the CVC crossed the midline of the abdomen from right to left and then continued ventrolaterally toward the left side of the abdominal aorta, level with the cranial renal artery ostia, and extended as far as the common iliac vein bifurcation. The left proximal ureter deviated medially and dorsally and it appeared to be compressed between the dorsal wall of the CVC and the ventral surface of the psoas muscles (left retrocaval ureter). The remaining portion of the left ureter was barely visible distal to the occlusion. No signs of left pyelectasia or hydroureter were found. Finally, the left external inguinal ring showed marked signs of stenosis in addition to cranial displacement of the pudendoepigastric trunk, which was localized within the inguinal canal (troncular vascular anomaly).
Conclusion:
To the authors’ knowledge, this is the first report documenting multiple vascular diseases characterized by troncular vascular anomaly of the pudendoepigastric trunk, left‐sided CVC (LsCVC), ipsilateral retrocaval ureter, and inguinal ring stenosis. In human medicine, troncular vascular anomalies are widely described and management may include medical (vascular sclerosing therapy) and surgical interventions. There are few papers about the association of LsCVC and retrocaval ureter in veterinary literature, but no one has described a troncular vascular anomaly and particularly the simultaneous presence of all three.
ABERRANT SUBCLAVIAN ARTERIES: COMPUTED TOMOGRAPHY FINDINGS IN 13 DOGS
D. Della Santa, A. Arcangeli, T. Mannucci From the Tac Veterinaria Fiorentina, San Piero a Sieve, Italy (Della Santa), Clinica Veterinaria Apuana, Carrara, MS, Italy (Arcangeli), and Department of Veterinary Sciences, University of Pisa, Pisa, PI, Italy (Mannucci).
Background:
Computed tomography (CT) is the diagnostic method of choice in patients with suspected vascular ring anomaly.
Purpose:
The aim of the present paper is to describe the CT findings detected in dogs with aberrant subclavian arteries.
Methods:
In this retrospective study, the CT archives of Tac Veterinaria Fiorentina, Clinica Veterinaria Apuana, and Dipartimento di Scienze Veterinarie of Pisa University were searched for cases of aberrant subclavian arteries in dogs. The clinical, computed tomographic, and, when available, surgical reports were retrieved. Evaluated clinical parameters include: presence of clinical signs consistent with a vascular ring anomaly and reason of the CT study. Evaluated CT parameters are: type and morphology of the aberrant subclavian arteries, presence of concurrent vascular abnormalities, reason, and degree of esophageal constriction.
Findings:
Thirteen dogs were included in the study (four boxers). Aberrant left subclavian artery (LSA) was diagnosed in 4/13 cases (all of them scanned because of a suspected vascular ring anomaly) concurrently with persistent right aortic arch (PRAA) with left ligamentum arteriosum that was considered the only cause of esophageal constriction in 3/4 cases. In 1/4 cases, the aberrant LSA (associated with PRAA and right‐sided ligamentum arteriosum) caused the esophageal constriction. Aberrant right subclavian artery (RSA) was diagnosed in 9/13 cases. Six of nine dogs were scanned for unrelated reasons while 3/9 underwent CT examination for regurgitation (1/9 for suspected vascular ring anomaly). Among asymptomatic patients, it was considered not responsible of esophageal constriction in 4/6 dogs and questionably constricting the esophagus in 2/6 dogs; upon further interrogation two owners reported clinical signs consistent with occasional regurgitation (one in each group). In 1/3 symptomatic dogs, regurgitation was considered secondary to generalized megaesophagus; in another dog it was presumed to be secondary to the aberrant RSA. In all the aforementioned cases, the aberrant RSA was an isolated malformation while in one severely symptomatic dog a right ligamentum arteriosum was suspected because of the position of the pulmonary artery bifurcation located ventral to the trachea. Surgery confirmed the severe esophageal stenosis due to the RSA but no definitive information is available about the anatomy of the ligamentum arteriosum.
Conclusion:
Boxer dogs might be predisposed to aberrant subclavian arteries. Both aberrant LSA and RSA are often an incidental finding in dogs and do not cause esophageal constriction, but in some cases they can be clinically significant. The most valuable information achievable from CT studies is the diagnosis of a clinically significant aberrant RSA that needs to be approached surgically via a right lateral thoracotomy (instead of a standard left lateral). Although the ligamentum arteriosum is not identifiable, a pulmonary artery bifurcation located ventral to the trachea might be suggestive of its right‐sided nature.
COMPARISON OF ABDOMINAL ULTRASOUND AND COMPUTED TOMOGRAPHY ENTEROGRAPHY IN 10 HEALTHY DOGS AND SIX DOGS WITH GASTROINTESTINAL DISEASE
S. M. Stieger‐vanegas, E. Ryan, C. Ruaux, G. Bobe From the Department of Clinical Sciences, College of Veterinary Medicine, Oregon State University, Corvallis, OR (Stieger‐Vanegas, Ruaux), Veterinary Medical Center, College of Veterinary Medicine, University of Minnesota, St. Paul, MN (Ryan), and Department of Animal and Rangeland Sciences, College of Agricultural Sciences, Linus Pauling Institute, Oregon State University, Corvallis, OR (Bobe).
Background:
The objective of this study was to compare the imaging findings of gastrointestinal ultrasonography (US) to computed tomography (CT) enterography using an oral iodinated contrast agent in healthy dogs and in dogs with gastrointestinal disease.
Purpose:
The initial purpose of this study was to establish a CT enterography protocol in healthy dogs. A further aim of the study was to compare the diagnostic accuracy of CT enterography to US in effectively identifying diseased segments of intestine in canine patients with gastrointestinal disease.
Methods:
CT enterography was performed in 10 healthy dogs and six dogs with gastrointestinal disease. After fasting for 12 h, all dogs underwent an abdominal ultrasound and CT enterography exam. After the US exam, nine of the 10 healthy dogs and three of the six diseased dogs got via orogastric intubation 15 ml/kg of a high concentration oral iodinated contrast agent containing diatrizoatemeglumine and diatrizoatesodium solution (Gastrografin, Brancco, Princeton, NJ) diluted 1:4 to 1:16 with water to a maximum volume of 500 mls. One of the healthy dogs and three of the dogs with gastrointestinal disease drank the contrast agent containing fluid voluntarily. Postanesthesia, a CT exam of the entire abdomen from the cranial aspect of the diaphragm to the anus before and 60 s after IV injection of 2 mls/kg iopamidol (Bracco, Princeton, NJ) using a power injector at a flow rate of 3 mls/s was performed. After the US and CT enterography exam gastrointestinal wall thicknesses and intestinal wall diameter were measured in both modalities. In addition, pre‐ and post‐IV iodinated contrast agent injection gastrointestinal wall attenuations were measured in the CT enterography images.
Findings:
No complications post oral and IV iodinated contrast agent injections were noted in any of the dogs. CT enterography allowed excellent visualization of the entire thickness of the gastrointestinal wall in healthy dogs and dogs with gastrointestinal disease. The gastrointestinal wall thickness measurements between ultrasound and CT enterography were similar. The diameter of the gastrointestinal tract was higher using CT enterography than ultrasound, which is likely secondary to the increased fluid influx into the gastrointestinal tract post oral iodinated contrast application.
Conclusion:
CT enterography provided increased confidence in the diagnosis and altered management plans in one patient with near penetrating gastric ulceration. Both gastrointestinal ultrasound and CT enterography proved effective in visualizing the gastrointestinal tract and diagnosing gastrointestinal disease.
FUNCTIONAL MAGNETIC RESONANCE IMAGING OF THE CANINE LIVER IN HEALTHY DOGS
F. Del Chicca, A. Schwarz, P. Grest, P. R. Kircher From the Division of Diagnostic Imaging (Del Chicca, Kircher), Section of Anesthesiology (Schwarz), and Institute of Veterinary Pathology, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland (Grest).
Background:
Magnetic resonance (MR) assessment of the liver largely relies on visual assessment of T1‐ and T2‐weighted images. However, the diagnostic value of structural alterations in conventional images is limited. Functional MR imaging techniques are designed to yield additional information about the pathologic process and attempt to quantify it, becoming the modality of choice in human medicine.
Purpose:
The purpose of the present study is to investigate the characteristics of the healthy canine liver through functional imaging techniques. These are dynamic contrast‐enhanced MR imaging, diffusion‐weighted MR imaging, in‐phase and opposed‐phase MR imaging (modified DIXON sequence).
Methods:
Eleven dogs with no clinically signs of abdominal disease were scanned with a 3T MR system (Philips Ingenia). Beside morphological sequences (T1 and T2 weighted), contrast angiography, diffusion‐weighted imaging, modified DIXON sequence, and blood flow quantification through phase contrast angiography were performed. Fine needle aspiration of all the dogs and biopsy of seven were interpreted as normal liver samples.
Findings:
During the contrast angiography, regions of interest were placed on the aorta, caudal vena cava vein, portal vein, hepatic, and splenic parenchyma. Mean maximal relative enhancement of the aorta is 1120% ± 287.4, of the caudal vena cava 328.8% ± 107.2, of the portal vein 478.9% ± 101.5. Mean time to peak of the three vessels is, respectively, 27 ± 10.4 s, 73.1 ± 55.1s, 75.2 ± 53.1 s. Mean wash‐in and wash‐out rates are, respectively, 172 ± 49 and 24.2 ± 16.9/s in the aorta, 116.7 ± 48.0 and 10.5 ± 6.2/s in the caudal vena cava, and 102.8 ± 45.4 and 6.0 ± 1.9/s in the portal vein. Mean relative enhancement of the liver parenchyma is 112.2% ± 40.7, of the splenic parenchyma 64.7% ± 33.3, time to peak, respectively, 76.9 ± 41.8 s and 166.2 ± 110.2 s, wash‐in and wash‐out rates 34.1 ± 20.2 and 3.0 ± 1.7/s in the liver, 9.5 ± 7.0 and 1.5 ± 1.5/s in the spleen. Mean ADC diffusion coefficient of the liver parenchyma is 0.84 ± 0.17 × 10−3, mean of the ratio between ADC liver to spleen parenchyma is 1.8 ±0.4, and mean of the ratio between ADC liver to gallbladder 0.38 ± 0.07. The fat quantification calculated through in‐phase and opposed‐phase technique ranges from 0.9% to 8.5%, when calculated through chemical shift images from 2.8% to 6.1%. Characteristics of the blood flow (as flow over time curve) on single slices at defined location of the aorta, caudal vena cava, portal vein, and celiac artery were studied and mean stroke volume, mean velocity, regurgitant fraction, and stroke distance are reported.
Conclusion:
The present data are suggested as baseline values in homogenous normal population and intended as reference in case of hepathopathies. It is reasonable to assume that in veterinary medicine, similar to human medicine, the reported values would be subject to change depending on the underlying pathologic process. Further investigations in this field are needed.
DISTINGUISHING MALIGNANT VS. NONMALIGNANT HEPATIC AND SPLENIC MASS LESIONS USING DUAL‐PHASE COMPUTED TOMOGRAPHY
I. D. Jones, C. R. Lamb, R. Drees, P. Mantis From the Department of Clinical Sciences and Services, The Royal Veterinary College, North Mymms, Hertfordshire, UK (Jones, Lamb, Drees, Mantis).
Background:
The current gold standard technique for the identification of hepatic and splenic masses is histopathology. Noninvasive diagnosis of malignant vs. nonmalignant hepatic and splenic mass lesions would aid clinical decision making. The CT (computed tomographic) appearance of hepatic (triple‐phase postcontrast) and splenic (single‐phase postcontrast) masses has been described. Malignant hepatic masses tend to be hypoattenuating with heterogeneous enhancement following contrast administration, whereas nonmalignant hepatic masses tend to be more iso/hyperattenuating with uniform contrast enhancement. Malignant splenic masses also tend to be more hypoattenuating relative to nonmalignant masses pre‐ and postcontrast administration.
Purpose:
The aim of this study was to identify features in dual‐phase postcontrast CT studies that could be used to distinguish malignant and nonmalignant hepatic and splenic mass lesions.
Methods:
Medical records in the period 2008–2012 were searched for dogs that had histopathological diagnosis of a hepatic or splenic mass lesion after dual‐phase postcontrast CT. Dogs were excluded if more than one distinct type of lesion was identified. CT images were acquired prior to and <30 s (early phase) and between approximately 90 and 480 s (late phase) following contrast administration. CT studies were reviewed by two observers (I.J., P.M.) without knowledge of the final diagnosis. The following features were recorded: maximum transverse dimension (mm), attenuation (HU), relative attenuation (HU), early enhancement (HU), late enhancement (HU), contrast retention (late minus early enhancement), enhancement pattern (diffuse/patchy/peripheral), margination (distinct/indistinct), capsular distortion (yes/no), mineralization (yes/no), and ascites (yes/no). Differences in imaging variables were tested using a t‐test and Mann–Whitney U‐test for parametric and nonparametric data, respectively. Likelihood ratios were calculated for categorical data.
Findings:
Thirty‐one dogs were studied, 19 hepatic, nine splenic, and three with both hepatic and splenic mass lesions. Sixteen mass lesions were nonmalignant and 18 were malignant. Malignant hepatic mass lesions had lower attenuation than nonmalignant mass lesions (45 ± 12 HU vs. 61 ± 14 HU, P = 0.02). No statistically significant differences in CT features were identified between malignant and nonmalignant splenic mass lesions. When hepatic and splenic data were combined, attenuation was significantly lower (45 ± 14 HU vs. 57 ± 15 HU, P = 0.02) and contrast retention significantly higher (53(−75 to 56) HU vs. −68(−108 to 27) HU, P = 0.02) for malignant vs. nonmalignant mass lesions.
Conclusion:
There is substantial overlap in the CT features of malignant and nonmalignant hepatic and splenic mass lesions. Malignant masses have lower average attenuation and higher contrast retention than nonmalignant masses, but overlap in these ranges limits diagnostic accuracy. Biopsy remains essential for diagnosis of splenic and hepatic mass lesions in dogs.
COMPUTED TOMOGRAPHIC FINDINGS IN 44 DOGS AND 10 CATS WITH GRASS SEED FOREIGN BODIES
D. Vansteenkiste, K. C. L. Lee, C. R. Lamb From the Department of Clinical Sciences and Services, The Royal Veterinary College, North Mymms, Hertfordshire, UK (Vansteenkiste, Lee, Lamb).
Background:
Migrating grass seeds or fragments (e.g., awns) are a well‐recognized problem, particularly in active dogs. Clinical signs depend on the affected site and frequently include pyrexia, soft tissue swelling, and/or cough. Radiographic signs in affected animals usually represent secondary inflammatory lesions rather than the foreign body. Computed tomographic (CT) findings reported in 13 dogs and one cat with intrathoracic grass seed fragments included focal pulmonary infiltrates or consolidation, pleural fluid, pleural thickening, and pneumothorax.
Purpose:
The purpose of the present study was to describe CT findings in a larger series of patients with grass seeds in other anatomical sites.
Methods:
Medical records covering the period of 2004–2013 were searched for dogs and cats that had CT and subsequent endoscopic or surgical retrieval of an intact grass seed or grass seed fragment during the same period of hospitalization from an anatomical site included in the CT study. CT images were reviewed with knowledge of the diagnosis and sites from which grass seeds were retrieved.
Findings:
Records of 44 dogs and 10 cats were found that satisfied these criteria. Median (range) age was 4 years (0.3–13 years). There were 34 males (23 neutered) and 20 females (13 neutered). Most prevalent breeds were 11 spaniels (25% dogs) and six Domestic Shorthaired (60% cats). Forty‐three animals (80%) were admitted in July–December. The predominant clinical signs were coughing in 16 (30%) animals, sneezing in 15 (28%), swelling in 15 (28%), and discharge in eight (15%). Grass seeds were retrieved from the thorax in 19 (35%) animals, nasal cavity in 17 (31%), other sites in the head and neck in 16 (30%), sublumbar muscles in one (2%), and pelvic limb in one (2%). The grass seed was visible in CT images of the respiratory tract in nine cases and the external ear canal in one case. Secondary lesions were visible in CT images in 52 (96%) animals, including collection of exudate in 20 (37%), soft tissue swelling in 13 (24%), abscess in 12 (22%), enlarged regional lymph nodes in 12 (22%), gas pocket in four (7%), and pneumothorax in four (7%). In two dogs, CT images appeared normal.
Conclusion:
Grass seeds in gas‐containing structures are frequently visible in CT images, but in general CT appears to be more useful as a guide to the possible location of a grass seed than as a method of definite diagnosis. This series, in which grass seeds were retrieved predominantly from the respiratory tract, likely underrepresents anatomical sites from which retrieval is more difficult, such as the sublumbar muscles.
MAGNETIC RESONANCE IMAGING OF MASTICATORY MUSCLES IN BASSET HOUNDS
A. Lorek, R. Dennis From the Centre for Small Animal Studies, Animal Health Trust, Newmarket, Suffolk, UK (Lorek, Dennis).
Background:
Magnetic resonance imaging (MRI) can be a useful adjunctive test in the diagnosis of inflammatory myopathies such as masticatory muscle myositis in dogs. Regions of altered signal intensity are found in MR images of affected muscles.
Purpose:
At the AHT, it has been observed that normal Basset hounds show apparent atrophy and MR signal changes of masticatory muscles in the absence of clinical signs. The purpose of this study was to describe these typical MR characteristics in order to help differentiate inflammatory myopathies from a breed variant.
Methods:
Twenty‐six Basset Hounds (11 females, 15 males, age range 2–12 years) were identified that had undergone an MRI scan of the head for various reasons. The scan protocol routinely included transverse T2‐W, T1‐W, and postcontrast T1‐W sequences and also included some or all of short tau inversion recovery, fluid attenuation inversion recovery, GRE, and postcontrast T1‐W with fat suppression. The MR images were examined for changes in signal intensity and muscle bulk and the final diagnosis was recorded. The imaging findings were compared to those in dogs of several other breeds with a final diagnosis of masticatory muscle myositis or atrophy secondary to trigeminal nerve disease.
Findings:
In the vast majority of the Bassets, the superficial part of the temporalis muscles and the dorsal aspect of the masseter muscles showed atrophy with bilaterally symmetrical T2‐W and T1‐W hyperintense signal and reduced enhancement after contrast medium administration while pterygoids were normal. The combination of pulse sequences suggested that the affected muscle areas contained adipose tissue with no evidence of inflammation. The abnormal area in the temporalis muscles often extended over the whole length of the muscle but was usually centered at the level of the attachment of the base of the pinnae, especially in younger animals. This distribution is different from cases of acute masticatory muscle myositis or other inflammatory myopathies in which changes tend to affect the whole masticatory muscle bulk including the pterygoid muscles, with marked contrast enhancement. In addition, the degree of muscle atrophy was milder and the fat infiltration greater in the Basset Hounds than in cases of chronic masticatory muscle myositis or trigeminal disease.
Conclusion:
Atrophy and MR signal changes in the superficial parts of the temporalis and masseter muscles in Basset Hounds appear to be normal findings in this breed and should not be misinterpreted as evidence of an inflammatory myopathy. We speculate that the changes are possibly due to chronic mechanical irritation of these head muscles from the gravitational pull of the large and heavy ears. Awareness of these MR imaging characteristics will help to prevent overdiagnosis of inflammatory myopathies in Basset Hounds.
POSTERS
RADIOGRAPHIC AND COMPUTER TOMOGRAHIC FINDINGS IN A GOLDEN LION TAMARIN (LEONTOPITHE CUS ROSALIA) WITH OSTEOPENIA
T. KÖnning, M. Van Zijll Langhout, M. Passon‐vastenburg From the Diagnostic Imaging Department (Könning, Passon‐Vastenburg) and Emergency Care Department, Medisch Centrum voor Dieren, Amsterdam, The Netherlands (van Zijll Langhout).
Background:
Tamarins (Saguinus) are susceptible to secondary hyperparathyroidism related to nutrition mismanagement or limited exposure to ultraviolet light. Although chronic renal failure is a common cause of secondary hyperparathyroidism in humans, it is less likely in other primates. Imaging findings of Golden Lion Tamarins (Leontopithecus rosalia) with osteopenia, likely caused by renal secondary hyperparathyroidism, has not been documented yet.
Purpose:
The purpose of this case report is to document radiographic and computer tomographic (CT) findings of a Golden Lion Tamarin with osteopenia and to compare the radiographs of this Golden Lion Tamarin with the radiographs of a Golden Lion Tamarin with normal bone structure.
Methods:
A 21‐year‐old male Golden Lion Tamarin from a private collection was presented at the Medisch Centrum voor Dieren with a spiral fracture of the diaphysis of the left femur. The animal did not show any clinical symptoms before this episode. On radiographic examination generalized reduction in bone opacity was visible, with thinning of cortices and a heterogeneous, mottled bony structure. Another 18–year‐old Golden Lion Tamarin, kept in the same husbandry situation, showed normal bone structure on radiographic examination. Both animals received a balanced diet containing sufficient calcium. To further investigate the bony structure of the affected Golden Lion Tamarin, a total body CT scan was performed with 0.8 mm slice thickness and a bone algorithm. The images were reviewed in Osirix® in bone setting (WL 300, WW 1500) and a soft tissue setting (WL 40, WW 350). Because no IV catheter was placed, no contrast medium was administered.
Findings:
On CT images a generalized mottled bony structure was visible, most apparent in the vertebrae. The ribs showed focal thickening in multiple ribs, most likely due to healed fractures. In the region of the thyroid gland and parathyroid gland, a local bilateral enlargement of soft tissue opacity was visible. At the right side a small mineralized opacity was visible within the enlarged structure. This was best visible in soft tissue setting. Repair of the fractured femur had an unacceptable high risk of complications due to the abnormal bony structure. Because of the poor prognosis, it was decided to euthanize the Golden Lion Tamarin. Blood work, necropsy, and histology revealed renal secondary hyperparathyroidism with metabolic bone disease caused by a chronic, active nephritis. The rib changes seen in CT images were identified as healed, possibly pathologic rib fractures related to the osteodystrophy. The parathyroid glands within the thyroid gland showed hypertrophic enlargement.
Conclusion:
CT images allowed, in this case, visualization of the bony structure and showed abnormalities of the ribs and parathyroid glands. Comparison of the normal and affected Golden Lion Tamarin radiographs showed a clear difference in bony structure. This case report is the first documentation of radiographs and CT images of a Golden Lion Tamarin with osteopenia caused by a chronic, active nephritis.
THE RADIOLOGICAL RULER©: A NOVEL DEVICE TO MEASURE STRUCTURES IN DIAGNOSTIC IMAGES
C. R. Lamb From the The Royal Veterinary College, University of London, North Mymms, UK (Lamb).
Background:
Making radiographic measurements of the size of an organ or structure does not increase diagnostic accuracy. Nevertheless, anecdotal evidence suggests there are more than a few picky clinicians out there who insist that radiology reports of their patients include measurements and academic radiologists who continue to publish studies based on the premise that there is diagnostic value in making measurements of organs.
Purpose:
The aim of the present study was to put a stop to that.
Methods:
A prototype ruler was constructed easily by cutting up the plastic sheet supplied with film for a thermal printer (Drystar DT1; Agfa Healthcare). A scale of measurement employing vague severity terms (normal, slight, moderate, marked) was applied using self‐adhesive labels (LabelPoint b/w label printer; Dymo). Clinical trials of the prototype ruler were conducted away from the prying eyes of any surgical residents unable to take a joke. Experimental technique involved holding the ruler up in the vicinity of any radiograph on a view box or monitor; the investigator then estimated the size of a lesion or relevant structure according to the scale. A series of duplicate radiology reports of clinical patients seen during a particularly slow afternoon were prepared by two investigators working independently: one using a conventional clear plastic ruler graduated in millimeters; the other using the Radiological Ruler©. Reports were assessed for clarity, readability, and literary style by a third investigator blinded to the identity of the investigators.
Findings:
The Radiological Ruler© was easy to use. The method initially attempted proved to be robust, hence no modification of the experimental technique was considered necessary. Accuracy of diagnosis of the two investigators was the same (about 75%), but radiology reports prepared using the Radiological Ruler© were significantly shorter and easier to read and understand <0.05) than the reports produced using the conventional ruler.
Conclusion:
Commercialization of this device, though tempting, is not considered necessary because of the ease of construction. Even a klutz should be able to make one and it is more ecologically friendly to use scrap materials. Furthermore, accurate calibration during manufacture is unnecessary because the size of the ruler does not matter! Particularly when measurements of structures have no diagnostic utility (i.e., virtually all of the time), the Radiological Ruler© should be considered as an alternative to conventional measurement methods. Ben Felson said, “A radiologist with a ruler is a radiologist in trouble.” Not with this ruler.
B‐MODE AND A‐MODE ECHOGRAPHIC EVALUATION OF PRIMARY AND METASTATIC INTRAOCULAR TUMORS IN THE DOG AND CAT
C. S. Baptista, J. Santos, M. H. Villagrasa From the Department of Veterinary Clinics, Institute of Biomedical Sciences Abel Salazar, University of Porto (ICBAS‐UP), Porto, Portugal (Baptista, Santos) and Ophthalmologic Veterinary Centre, Madrid, Spain (Villagrasa).
Background:
Intraocular tumors are primary or metastatic and can cause significant sequels for vision and the comfort of the animal. Accordingly, an early diagnosis is an important aspect of successful management, which can be provided by ocular ultrasonography.
Purpose:
Our aim was to evaluate the topographic and acoustic characteristics, in B‐mode and A‐mode, of 25 primary and two metastatic intraocular tumors in the dog and cat. We hypothesized that ocular ultrasound may establish objective criteria for the evaluation and diagnosis of intraocular tumors and permits the reproducibility of results.
Methods:
Twenty‐three dogs and three cats with histologic and/or cytologic confirmation of an intraocular tumor and that had B‐mode and A‐mode sonograms performed were retrospectively selected. We used a 10 MHz or a 20 MHz sector transducer for both scans. Local anesthetic was instilled in the affected eye and the probe was applied directly to the gel‐coated cornea. The ultrasound beam was oriented along the optic axis in order to obtain sagittal and dorsal image planes. Several topographic/acoustic characteristics of the masses were evaluated, namely location, size, shape, extension, echotexture, internal reflectivity, sound attenuation, consistency, mobility, and vascularity.
Findings:
We performed the ultrasound evaluation of two ciliary body adenocarcinomas, two anterior uveal adenomas, one metastatic mammary adenocarcinoma, one metastatic lymphosarcoma, and 21 anterior uveal melanomas (17 in dogs and four in cats; one cat presented, in the same eye, two melanomas histologically different). In the B‐mode examination, 11 tumors (40.74%) were isoechoic, nine (33.33%) hyperechoic, and seven (25.93%) hypoechoic. Medium size was 9.68 mm (range from 3 to 1.2 mm). All tumors were irregular (100%) and acoustic shadowing was not observed (100%). The A‐mode studies showed all masses were solid (100%). Thirteen tumors (48.14%) showed medium internal reflectivity, seven (25.93%) revealed low internal reflectivity, and seven (25.93%) presented with high internal reflectivity. We observed that the higher the content in melanin, the higher was the internal reflectivity of anterior uveal melanomas. Sound attenuation was low in 16 cases (59.26%) and medium in 11 cases (40.74%).
Conclusion:
The acoustic hallmark of anterior uveal melanomas is their low to medium reflectivity, internal vascularity, and hard consistency. Metastatic carcinomas and adenomas presented high internal reflectivity and low attenuation and they can be distinguished from high reflectivity melanomas because these tumors are heavily pigmented. Conversely, the differentiation between ciliary body adenocarcinomas and low reflectivity melanomas may be very challenging due to their low content in melanin. Interpreted in accordance with the clinical history and the ophthalmic examination, our data may improve the clinical applicability, reliability, and accuracy of echography in both the detection and differentiation of intraocular tumors.
ULTRASONOGRAPHIC CAUDOCRANIAL APPROACH OF THE CANINE SHOULDER JOINT
E. Daga, F. Miduri, A. Volta, M. Fabbi, S. Manfredi, G. Gnudi From the Dipartimento di Scienze Medico Veterinarie, University of Parma, Parma, PR, Italy (Daga, Miduri, Volta, Fabbi, Manfredi, Gnudi).
Background:
Ultrasonographic examination of the canine shoulder joint was described by Vandevelde. The caudal aspect of the humeral head was studied positioning the transducer distal to the acromion with a caudocranial direction with the joint adducted and endorotated.
Purpose:
The aim of this study is to evaluate the feasibility and usefulness of a caudocranial ultrasonographic approach of shoulder joint in normal dogs and in dogs affected by osteochondrosis (OC) of the caudal third of the humeral head.
Methods:
Eighteen shoulder joints of nine dogs were considered and examined with radiography, ultrasound, and arthroscopy. Ultrasound was performed by a well‐trained operator, unaware of the radiographic results. Ultrasound was performed according to Vandevelde et al. and with a caudocranial approach of the humeral head, with the joint in maximum extension. An 8 MHz microconvex transducer was employed. For the caudocranial approach, the probe was positioned against long head of the triceps muscle, moving it until the caudal surface of the humeral head appeared.
Findings:
Seven shoulder joints were normal and 11 were affected by OC. All the normal humeral heads were seen as a hyperechoic curvilinear interface with an overlying thin anechoic line representing the joint cartilage both with the ultrasonographic approach described by Vandevelde and the caudocranial approach. The caudal profile of all the humeral heads affected by OC appeared irregular and flattened both with Vandevelde and with caudocranial projections. In one case of OC, an anechoic distended caudal pouch of the joint was detected. The caudocranial ultrasonographic approach allowed a complete study of the caudal aspect of the humeral head, when the transducer was tilted laterally and medially. The results obtained by ultrasound were always confirmed by radiography and arthroscopy.
Conclusion:
The caudocranial ultrasonographic approach of the canine shoulder joint was feasible for the diagnosis of caudal humeral head anomalies such as OC and could be useful in addition to the approach described by Vandevelde. The difficulties encountered by the thickness of the muscles interposed by the skin and the humeral head interface might be minimized using a microconvex probe and a frequency ranging from 6 to 8 MHz. A limitation of this approach is the inability to scan the entire joint.
HEPATIC PARENCHYMAL GAS FOLLOWING BLUNT TRAUMA IN A CAT
M. Fabbi, F. Miduri, S. Manfredi, E. Daga, G. Gnudi, A. Volta From the Dipartimento di Scienze Medico Veterinarie, University of Parma, Parma, PR, Italy (Fabbi, Miduri, Manfredi, Daga, Gnudi, Volta).
Background:
Hepatic parenchymal emphysema is a rare occurrence in cats as in dogs, usually associated with infection with gas‐producing organisms. Clinical features are mostly severe, and prognosis is guarded to poor.
Purpose:
The purpose of this paper is to describe the first case of intrahepatic gas after blunt trauma in a cat.
Methods and findings:
A 1.5‐year‐old neutered male domestic shorthair cat was presented within 12 h after sustaining blunt trauma. On presentation, the cat was bright and responsive, but unable to stand, normothermic, with some facial lacerations. Blood workup showed increased levels of serum alanine aminotransferase. A focused assessment with sonography for trauma exam showed a moderate amount of pleural and peritoneal fluid compatible with hemorrhage and multiple hyperechoic foci associated with reverberation artifact in the right medial liver lobe, suggestive of hepatic injury with gas formation. Chest, abdominal, and hind limbs radiographs were obtained and revealed a mild pleural effusion and a Salter‐Harris type I fracture of the distal right femur. A small amount of gas was faintly visible in the right cranial abdomen, below the costal arch. The cat was initially treated with rest in cage, fluid therapy, and amoxicillin‐clavulanic acid. A strict monitoring of the patient was made. Twelve hours after, hepatic parenchymal gas was still visible ultrasonographically and resolved within 3 days. At that time, an area of mixed echogenicity in the site of the hepatic injury, representing an organizing hematoma, was detected. The cat underwent orthopedic surgery and was discharged 2 days later. No signs of infection developed. Follow‐up ultrasound at 20 and 50 days showed progressive resolution of the hepatic lesion.
Conclusion:
Parenchymal or portal vein gas is rare in canine and feline liver. In humans, few reports on hepatic parenchymal gas after abdominal trauma exist. It has been postulated that gas may arise from hepatic necrosis or from increased intraperitoneal pressure, after blunt trauma, which may allow gas to enter mesenteric veins by means of microscopic mucosal injury. In this report, infection with gas‐producing organisms on a liver injury or acute torsion of a hepatic lobe was not likely, due to the ultrasonographic features and the benign course of the clinical case. In conclusion, the presence of hepatic gas after abdominal trauma, in cats as in humans, may not necessarily indicate infection or a surgical emergency.
USE OF COMPUTED TOMOGRAPHY TO IDENTIFY PROXIMAL SUSPENSORY CHRONIC DESMITIS ASSOCIATED WITH EXOSTOSES ON THIRD METACARPAL BONE: CASE REPORT
G. F. Viana, C. A. Sarkis, A. C. S. Aguiar, V. M. D. V. Machado From the Department of Animal Reproduction and Veterinary Radiology, Sao Paulo State University, Rio Claro, SP, Brasil (Viana, Sarkis, Aguiar, Machado).
Background:
An extensive bone formation (large exostoses) present at the palmar metacarpal region has been incriminated as a cause of persistent lameness in horses. They can cause mechanical irritation of the suspensory ligament (SL). Previous studies report acute SL desmitis associated with exostoses diagnosed using MRI and ultrasound scans.
Purpose:
This case report aims to demonstrate the applicability of computed tomography (CT) to identify chronic desmitis of the proximal suspensory ligament associated with third metacarpal bone exostoses, as desmitis is not usually diagnosed with this imaging technique.
Methods:
An equine, Brazilian Jumper Horse, 12 years old, a 1.40 m show jumper horse was referred to the Department of Animal Reproduction and Veterinary Radiology of Sao Paulo State University and presented with history of left forelimb progressive lameness associated with local repetitive strain injuries. The animal was forwarded for CT examination. The equipment used was a CT helical scanner Shimadzu, model SCT‐7800TC. Horse was positioned in lateral recumbency under general anesthesia, maintained with isoflurane and oxygen. Contiguous 2 mm transverse slices were acquired from the proximal palmar region of the left third metacarpal bone (MC3) through the metacarpophalangeal joint and no application of IV or intra‐arterial contrast medium. The images were stored and then evaluated on axial plane in soft tissue and bone windows and in three‐dimensional reconstruction (3D) for the bony contours.
Findings:
In the CT images, in soft tissue window it was possible to observe a punctate hyperattenuating area with poorly defined contours (Hounsfield unit [HU]: 103) in the proximal suspensory ligament (HU: 32) and a discrete sharp exostosis on the palmar aspect of MC3. Just below this region, an extensive exostosis in MC2 was observed in bone window. On lateral aspect, at left, MC2 showed irregular bone margin and increased diameter on distal third, not correlated with desmitis area identified on CT. Three‐dimensional images provided detailed information regarding extension, features, and location of exostoses along MC2 and MC3.
Conclusion:
The CT examination enabled to verify bone and soft tissues lesions adjacent to each other. The identification of hyperattenuating area on the proximal region of the suspensory ligament was possible due to the lesion chronicity, the selection of the soft tissues window, and HU values at level of exostosis in MC3. Although CT associated with angiography is a diagnostic method widely used to highlight ligament injuries, this case showed that CT not associated with angiography is effective on detecting chronic lesions of the proximal suspensory ligament.
IMAGING DOGS WITH PRESUMED INTRASPINAL CYSTS USING COMPUTED TOMOGRAPHY OR MAGNETIC RESONANCE: CHANCES AND RISKS
B. Lorincz, S. Breit, A. Bohler, N. Katic, A. Fuchs‐baumgartinger, S. Kneissl From the Clinic for Companion Animals and Horses, Diagnostic Imaging (Lorincz, Bohler, Kneissl), Department of Pathobiology (Breit, Fuchs‐Baumgartinger), and Clinic for Companion Animals and Horses, Small Animal Surgery, University of Veterinary Medicine, Vienna, Austria (Katic).
Background:
Cysts in the canine spinal canal are rare and associated with the facet joint, dorsal longitudinal ligament, intervertebral disc, or subarachnoid cavity. No previous studies have compared the ability of CT and MR imaging to identify cysts in dogs.
Purpose:
The purpose of the study was to compare CT and MR imaging with regard to definitive identification of canine spinal cyst (rim‐enhanced soft tissue mass) and determination of the site, extent, and clinical relevance.
Methods:
Medical records from 2008 to 2013 were searched for dogs that had suspected spinal cysts in CT or MR images or both. Images were reviewed by two observers (B.L. and S.K.), who reached a consensus regarding the following features: location; extent; connection to facet joint, ligaments, disc, or subarachnoid cavity; signal intensity/density; contrast enhancement after IV application; and involvement of neural tissue (spinal cord or segmental nerves) or bones.
Findings:
Seven dogs fulfilled the inclusion criteria: two German Shepherd Dogs, a German Boxer, an American Staffordshire Terrier, a Rottweiler, a Beagle, and a French Bulldog (CT + MR, n = 3; CT, n = 3; MR, n = 1). IV contrast was applied in five cases. Age varied from 5 to 12 years. Trauma was suspected in three dogs. Six of the seven dogs had neurologic deficits, and one had only pain. Five presumed intraspinal cysts were associated with thoracic, thoracolumbar, or lumbar facet joints; one with a thoracolumbar intervertebral disc; and one with the subarachnoid cavity. Two of the facet joint cysts were associated with spondylarthrosis. All cysts showed rim enhancement. Four displaced neural structures. All dogs had at least one other imaging finding, including spondylosis, spondylarthrosis, transitional vertebra, or disc degeneration. MR was superior to CT in determining the lesion location in one case. Three dogs were treated surgically, four conservatively. The cyst was verified by pathohistologic examination in one case.
Conclusion:
MR imaging is the gold standard for diagnosing intraspinal cysts. More neurologic patients receive CT than MR imaging at the University of Veterinary Medicine, Vienna. However, nearly all patients with inconclusive CT undergo further MR imaging to explain the clinical signs. CT allowed the assessment of site, extent, and involvement of other structures in three of the seven patients. We found that MR outlined the cysts with better contrast, but CT offered better resolution. Thus, for small dogs CT would be the modality of choice and for large ones MR. Moreover, CT better delineated associated degenerative changes along the spine.
SONOGRAPHIC PARAMETERS OF THE COMMON CAROTID ARTERY IN AMERICAN QUARTER HORSES
A. C. S. Aguiar, N. R. Sousa, G. F. Viana, V. M. D. V. Machado From the Department of Animal Reproduction and Veterinary Radiology, FMVZ, UNESP, Botucatu, SP, Brasil (Aguiar, Sousa, Viana, Machado).
Background:
Common carotid artery (CCA) is the main source of arterial supply for the brain and reduced supply caused, for example by stenosis can lead to brain lesions, affecting white matter as already described in humans and rats. Physiologic data regarding sonographic evaluation of the carotid in horses are lacking.
Purpose:
The aim of this study was to obtain normal values for CCA intima‐media thickness (IMT), resistivity index (RI), and diameter in American Quarter Horses and check how age influenced these parameters.
Methods:
Sixteen adult American Quarter Horses aged between 3 and 10 years and 10 foals from the same breed, aged between 13 months and 1.6 years of age were submitted to bilateral ultrasonographic scans of CCA on the lower neck, around C5 and C6. The equipment used was My LAB30‐ESAOTE and a linear transducer of 18 MHz. Three to five measurements of IMT and lumen diameter (D) were made using at least three different images of each artery. Mean RI was obtained from at least three measurements of the same image with Doppler angle between 75° and 60°.
Findings:
Mean value for right IMT was 1.03 ± 0.12 mm for adults and 0.96 ± 0.16 mm for foals. Mean left IMT was 1.04 ± 0.14 mm for adults and 0.94 ± 0.14 mm for foals. Mean right CCA luminal diameter was 10.83 ± 1.16 mm for adults and 9.43 ± 0.71 mm for foals. Mean left CCA D was 10.24 ± 0.80 mm for adults and 9.20 ± 0.58 mm for foals. Mean right RI was 0.80 ± 0.06 for adults and 0.75 ± 0.08 for foals. Mean left IR was 0.81 ± 0.07 for adults and 0.73 ± 0.07 for foals. Adults presented higher values for RI, D, and IMT than foals, this suggest that these values increase with age. Right CCA diameter presented higher values than the right side on both age groups.
Conclusion:
There are very few studies on equine carotid artery, especially regarding normal values that could be related to neurologic findings and also to access blood supply during anesthesia. This study emphasizes the importance of further studies in specific breeds, once values of RI differ from those previously obtained by Schumucker et al. (2000) and D from those published by Cipone et al. (1997). Normal values for CCA intima‐media thickness, resistivity and diameter can be used to differentiate between diameter on each side or normal values for foals and specific breeds that have not been previously reported as far as the authors are concerned.
RENAL B‐MODE AND SPECTRAL DOPPLER VALUES OF CAPTIVE SAPAJUS APELLA (CAPUCHIN MONKEY)
L. S. Carantina, R. V. Santos, R. D. N. Dias Neto, V. M. D. V. Machado From the Universidade Estadual Paulista Julio de Mesquita Filho—UNESP, Botucatu, SP, Brasil (Carantina, Santos, Dias Neto, Machado).
Background:
Sapajus apella (capuchin monkey) is widely distributed in Brazil, however, there are few studies on normal ultrasonographic parameters of the kidneys in this species.
Purpose:
In this study renal B‐mode and pulsed‐wave Doppler values of healthy S. apella, including renal length, thickness, width, volume, and cortical thickness as well as peak systolic velocity (PSV), end diastolic velocity (EDV), and resistivity index (RI) of the renal artery were measured to standardize a normal range for this species.
Methods:
Ten adults S. apella (three males and seven females) were evaluated using a multifrequency (6–10 MHz) linear probe. All animals examined under general anesthesia obtained with ketamine and midazolam as premedication and isoflurane for maintenance. Longitudinal, transverse, and coronal sections of both kidneys were performed. The volume of the kidney was calculated automatically by the ultrasound software, and this value was compared with the value obtained using the formula renal volume = length × width × thickness × 0.523. Color Doppler was used to identify renal arteries and pulsed‐wave Doppler was used to measure PSV, EDV, and RI of renal arteries.
Findings:
The mean values ± SD obtained were as following. Length of the left and right kidney was 3.17 ± 0.53 and 3.14 ± 0.47 cm, respectively. Thickness of the left and right kidney was 1.13 ± 0.24 and 1.24 ± 0.20 cm, respectively. Width of the left and right kidney was 1.18 ± 0.21 and 1.25 ± 0.18 cm, respectively. Cortex thickness of the left and right kidney was 0.38 ± 0.08 and 0.37 ± 0.05 cm, respectively. Automatic volume of left kidney was 2.19 ± 0.73 ml and calculated volume of left kidney 2.42 ± 0.97 ml. Automatic volume right kidney was 2.33 ± 1.02 ml and calculated volume right kidney 2.58 ± 0.9 ml. Applying the Student's t‐test to the values obtained, there were no statistical differences between right and left kidney. Resistivity index of the left and right renal artery was 0.75 ± 0.07 and 0.71 ± 0.08, respectively. Peak systolic velocity of the left and right renal artery was 0.30 ± 0.09 and 0.27 ± 0.05 m/s, respectively. End diastolic velocity of the left and right renal artery was 0.08 ± 0.02 m/s. Applying the Student's t‐test to the values obtained showed statistical difference between the values of the RI of the left and right renal artery.
Conclusion:
B‐mode and spectral Doppler values of S. apella obtained in this work are important since they could help in the diagnosis of the diseases, contributing to the better maintenance and preservation of the species.
EVALUATION OF C3‐C4 CERVICAL STENOTIC MYELOPATHY IN A FOAL USING MAGNETIC RESONANCE IMAGING
L. S. Silva, M. C. R. Castiglioni, B. R. De Souza, V. M. V. Machado, L. C. Vulcano From the Department of Animal Reproduction and Veterinary Radiology, Sao Paulo State University, Rio Claro, SP, Brasil (Silva, Castiglioni, De Souza, Machado, Vulcano).
Background:
One of the most common causes of spinal ataxia in foals is cervical stenotic myelopathy and it can be differentiated between other vertebral injuries by cervical radiographic examination. Intravertebral minimum sagittal diameter (MSD) is measured from the dorsal aspect of the vertebral body to the ventral border of the dorsal laminae. The sagittal ratio is obtained by dividing the MSD of the vertebral canal by the height of the corresponding vertebral body.
Purpose:
The aim of this study is to describe the magnetic resonance imaging (MRI) findings in a foal with subluxation and the applicability of measurements of the MSD on MRI imaging.
Methods:
A 4‐month‐old male American Quarter Horse, with no history of trauma, was presented with signs of ataxia and absence of proprioception in all four limbs and reluctance to flex the neck. Radiography with myelography and MRI examinations of the cervical spine were performed with the animal in right lateral recumbency and under general anesthesia. Immediately, cerebrospinal fluid (CSF) was obtained from the atlanto‐occipital space. Neutral, dorsoflexion, and ventroflexion radiographic projections were obtained to evaluate the static and dynamic instability. Subsequently, MRI scan was performed using a low field (0.25 Tesla) open MR system (EsaoteR Vet MR Grande) with a flexible coil. The scanning protocol included spin echo T1 and fast spin echo T2 in dorsal, sagittal, and transverse plains, a fluid attenuation inversion recovery (FLAIR) and 3D HYCE (T2‐weighted) in sagittal plans. The images were evaluated in the current workstation software (Synapse; Fujifilm Medical Systems, Stamford, CT). Minimum sagittal diameter and sagittal ratio were measured in sagittal plains on T2 images.
Findings:
Ventroflexion radiographic projection showed slight dorsal displacement of the cranial physis of C4 when compared with the caudal physis of C3. MRI findings were malalignment of the C3‐C4 articulation and flaring of the caudal physis of C3 in the dorsal aspect. In addition, spinal cord diameter was 40% reduced compared to the total diameter in C3‐C4 on transverse plain, increased intramedullary signal intensity on T2 images and accumulation of CSF in the dorsal dural space of C2‐C3 and in the ventral dural space of vertebral body of C3. Intervertebral discs of C2–C5 were considered hypointense on T2 images when compared with small animals and presented elongate shape with normal intervertebral spaces. Minimum sagittal diameter and sagittal ratio values at the level of C4 were 12.34 mm and 42.8%, respectively. Some authors consider sagittal ratio values lower than 48% at C4 through C6 as moderate to severe bony malformation. Analysis of CSF presented within normal limits.
Conclusion:
In association with clinical signs, radiographic and MRI findings and measurement of the MSD and sagittal ratio, this foal was diagnosed with cervical stenotic myelopathy. However, the applicability of measurement of the MSD through MRI is still in progress and this abstract is only a case report.
SUBCLINICAL COMPUTED TOMOGRAPHY ABNORMALITIES IN THE LUMBOSACRAL SPINE IN YOUNG ADULT DOGS WITH LUMBOSACRAL TRANSITIONAL VERTEBRAE
P. Proks, L. Stehlik, M. Paninarova, K. Irova, A. Necas From the Small Animal Clinic, University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic (Proks, Stehlik, Paninarova, Irova, Necas).
Background:
The association between the occurrence of a lumbosacral transitional vertebra (LTV) and the cauda equine syndrome (CES) in dogs is assumed. Lumbosacral transitional vertebra is associated with lumbosacral disc degeneration at an early age. No information is available regarding CT detectable abnormalities in lumbosacral area in young dogs with LTV.
Purpose:
Purpose of this prospective study was to determine whether some abnormalities in lumbosacral region could be detected with CT examination in young adult dogs with LTV and without clinical signs of CES.
Methods:
In 19 young nonchondrodystrophic pure breed dogs, LTV was detected on radiographs during preventive hip dysplasia examination for breeding purpose. All dogs were neurologically normal. These dogs underwent CT examination of lumbosacral region in sternal recumbency with pelvic limbs extended. Contiguous 1.25‐mm‐thick slices were obtained between L5‐Cd1. Volume rendering and multiplanar reconstruction (MPR) was performed. Type of LTV was determined. Disappearance of epidural fat between disc and thecal sac, calcification in vertebral canal or in disc, and vacuum phenomenon in disc between last normal lumbar vertebrae and LTV was determined as pathology. Presence of bone spur (spondylosis) in this localization was detected too.
Findings:
There were 12 females and seven males with average age 1.5 years (range 1–3.5 years) and average weight 30 kg (range 20–40 kg). From 19 dogs, 11 had LTV type II (symmetrical), eight dogs had LTV type III (asymmetrical). Pathology was found in five cases (26.32%), three males and two females. Disappearance of epidural fat between last normal lumbar vertebrae and LTV was detected in three cases (one case LTV type III, two cases LTV type II). In one case disappearance of epidural fat, disc vacuum phenomenon, and bone spurs (spondylosis) were detected (LTV type III). In one case disappearance of epidural fat and nucleus pulposus calcification were detected (LTV type III). No significant association was detected (P = 0.34) between lumbosacral pathology and type of LTV (Fisher's exact test).
Conclusion:
CES is usually a problem of older dogs. Findings of this study indicate that CT examination is useful noninvasive technique for early detection of lumbosacral pathology in young adult dogs. Presence of LTV is a predisposition for early disc degeneration in some young dogs. Findings indicate that some lumbosacral CT detectable abnormalities may be clinically insignificant in young adult dogs. Decision making the surgery should be done according to the clinical findings and CT anatomical localization of compressive tissues.
USE OF COMPUTED TOMOGRAPHY FOR DIAGNOSIS OF CONGENITAL PERIFERAL ARTERIOVENOUS MALFORMATION IN A PUPPY
P. Proks, L. Stehlik, M. Paninarova, K. Irova, J. Lorenzova From the Small Animal Clinic, University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic (Proks, Stehlik, Paninarova, Irova, Lorenzova).
Background:
Arteriovenous (AV) malformation is aberrant communication between the arterial and venous system that does not involve transit through a capillary bad and may be acquired or congenital.
Purpose:
The diagnostic procedure of unusual case of peripheral AV malformation manifested as subcutaneous lesion is discussed.
Methods:
A 4‐month‐old, 20 kg, male Cane Corso dog was presented for slowly enlarging subcutaneous swelling on the left side of the flank. The dog had no history of previous trauma. On clinical examination there was a soft, painless pulsating swelling connected with pulsating peduncle going to the flank region. A B‐mode ultrasonographic examination of the left flank was performed including Color flow mapping (14 MHz linear probe). Ultrasonography of the left caudal abdominal wall revealed a large number of torturous vessels. Two large vessels of 4 mm diameter were detected continued subcutaneously to sublumbar region. Color flow mapping showed region with multicolored mosaic pattern represented high velocity turbulent flow. Finding was consistent with AV malformation. For more precise evaluation of suspected AV malformation, nonselective CT angiography was performed on multidetector 16‐slice CT unit. One native and two contrast‐enhanced (arterial and venous phase) abdominal scans were performed. Picture acquisition was performed in helical mode (slice thickness 1.25 mm). For contrast examination 45 ml of Iomeprol (300 mg/ml) was injected by power injector via 14 G IV catheter into right v. cephalica antebrachii. Arterial and venous phase was performed in ninth and 30th second after start of the contrast medium application. Raw data were examined and multiplanar (MPR) and 3D reconstruction (volume rendering) was performed.
Findings:
On the left caudal flank region plexus (approximate size of 2.7 × 4.9 × 1.8 cm) of aberrant torturous small vessels were revealed in arterial phase of study. Plexus was located under the skin and were directly connected with deep circumflex artery and deep circumflex vein. The left deep circumflex vein was contrast enhancing in the arterial phase. CT finding was consistent with congenital AV malformation. Arteriovenous malformation was surgically resected. Six month after operation the dog is clinically normal.
Conclusion:
Congenital peripheral AV malformation is a rare disease in dogs. Contrast‐enhanced CT has an important role in surgical planning because it can reveal exact location of the lesion, its extension, and relation to adjacent structures. CT is excellent noninvasive diagnostic modality to evaluate the exact morphology of the anomalous vessels and provide definitive diagnosis.
INTRAOPERATOR REPEATABILITY OF MITRAL ANNULAR VELOCITY ESTIMATION WITH PULSED TISSUE DOPPLER IN DOGS
Y. Vali, M. Molazem From the Department of Surgery and Radiology, Faculty of Veterinary Medicine, University of Tehran, Tehran, Iran (Vali, Molazem).
Background:
Limitations of using conventional echocardiography for estimating functional alterations of the myocardium during the early stages of cardiac diseases have led to the use of tissue Doppler imaging (TDI) as a sensitive and specific technique.
Purpose:
The objective of this work was the assessment of intraoperator repeatability of mitral annular velocity with pulsed‐wave tissue Doppler in large healthy dogs.
Methods:
A total of four clinically healthy mixed‐breed female dogs with mean 16.34 kg body weight (ranged between 15 and 17.68 kg) were studied. The dogs did not have any clinical abnormalities on physical examination, electrocardiography, radiography, and conventional echocardiography. To evaluate the longitudinal myocardial motion, a 2‐mm sample volume gate was placed on the annulus of the mitral valve using the left apical four‐chamber view. Systolic (Sa), early diastolic (Ea), and late diastolic (Aa) motion of the mitral annulus were recorded. Each dogs underwent nine Doppler echocardiographic examination within 3 days (n = 36) by the same operator and results were analyzed with interclass correlation (ICC) test by SPSS software.
Findings:
All dogs were scanned on nine occasions over a 3‐day period (n = 36). Intraclass correlation of results showed perfect intraoperator repeatability between the measured systolic mitral annulus velocity (ICC = 0.919, with 95% CI (0.717, 0.998) and moderate agreement of the Ea, Aa velocities (ICC = 0.47 and 0.41, respectively).
Conclusion:
Perfect intraoperator repeatability that is shown in the present study for estimating systolic mitral annulus velocity makes it more reliable. Although moderate correlation among the Ea and Aa assessment can be acceptable, they should be interpreted guarded. Intra‐ and interobserver repeatability is suggested for definitive evaluation of mitral annular velocity validity for further studies.
ANATOMO‐RADIOGRAPHIC DESCRIPTION OF THE HEART IN BROWN BROCKET DEER (MAZAMA GOUAZOUBIRA)
L. S. Silva, V. R. Babicsak, V. M. V. Machado From the Department of Animal Reproduction and Veterinary Radiology, Sao Paulo State University, Rio Claro, SP, Brasil (Silva, Babicsak, Machado).
Background:
The brown brocket deer is a true ruminant belonging to the Artiodactyla order and Cervidae family with a wide geographical distribution in Brazil, Argentina, Bolivia, Paraguay, and Uruguay.
Purpose:
Until now, the thoracic radiographic features in this species have not been reported in the literature. In order to provide reference images and values for radiological interpretation, the objective of this study was the radiological description of the heart of brown brocket deer.
Methods:
A retrospective study was performed using five traumatized wild animals, four adults and one young (with visible physeal lines), which were treated in the CEMPAS, FMVZ‐UNESP Botucatu. Radiographic examination was performed under general anesthesia (xylazine and ketamine) to rule out any thoracic disorder, using digital radiographic equipment (GER DR‐F) with a Potter–Bucky grid and an exposure of 68–75 kVp and 200 mA. The deer were positioned in right lateral and dorsal recumbency. Images were evaluated qualitatively and quantitatively with a current workstation software (Synapse; Fujifilm Medical Systems, Stamford, CT).
Findings:
In the ventrodorsal view, the heart showed an elongated and conical shape with a convex left aspect and right atrium and a concave right ventricle. In addition, the cardiac apex was rounded and showed a slight inclination to the right side. The greatest cardiac area was visible in the left hemithorax. The heart was placed between the third and the sixth intercostal spaces on lateral view and the fourth and the ninth intercostal spaces on the ventrodorsal view in all animals except in one adult deer. In this subject, the heart was placed between the fifth and the ninth intercostal spaces in the ventrodorsal projection. The cardiac height and width were measured in the lateral view with mean values of 10.98 ± 0.85 cm and 7.39 ± 1.12 cm, respectively. The vertebral heart score (VHS) was also calculated with an average value of 8.26 ± 0.61 vertebrae.
Conclusion:
According to recent studies using dissected brown brocket deer, the heart is placed between the third and the sixth ribs as found in the present research. Additionally it was reported that the heart position and orientation is similar to that of domestic ruminants, differentiated by a bicarotid trunk. However, the cardiac apex of domestic ruminants is located in the left hemithorax, whereas this study showed a right‐sided location of the cardiac apex in brown brocket deer. The general cardiac shape and appearance of the left ventricular border resembles the heart of sheep and goats, and the VHS values were similar to sheep (8.6 vertebrae). In conclusion, although the radiographic appearance of the heart of Mazama gouazoubira is generally similar to that of domestic ruminants, some features are peculiar to this species.
CORRELATION OF RADIOLOGIC AND NECROPSY FINDINGS OF VERTEBRAL THORACOLUMBAR NEW BONE FORMATION IN TWO AFFECTED DOGS AND A CONTROL
C. Strohmayer, S. Breit, I. Walter, S. Kummer, S. Lang, S. Kneissl From the Department for Companion Animals and Horses, Diagnostic Imaging (Strohmayer, Kneissl), Department for Pathobiology, Institute of Anatomy, Histology and Embryology (Breit, Walter), VetCore Facility for Research (Walter, Walter), and Clinical Institute of Pathology, Medical University of Vienna, Vienna, Austria (Lang).
Background:
In humans, inflammatory spondylarthropathies (ISA) and diffuse idiopathic skeletal hyperostosis (DISH) predominantly account for new bone formation (NBF). In dogs, little is known about neither the etiology nor the disease progress in ISA, however spondylarthropathies are thought to be degenerative mainly.
Purpose:
The aim of the present study is to analyze and understand radiographic features by means of corresponding pathological macro‐ and microscopic sections.
Methods:
Postmortem specimens (Th13/L1‐L6/L7) were collected from two dogs with severe spinal NBF (German Shepherd crossbred, 14 years; German Boxer, 10 years) and from a control (Labrador, 12 years), all having been euthanized for reasons unrelated to the study. After conventional radiography and computer tomography (CT) unilateral para‐median slices from the fresh cadaver were cut using a band saw and photographed. On the basis of imaging findings (C.S. and S.K.), tissue samples of approximately 2 × 2 cm at the ventral contour of the intervertebral disc including the adjacent vertebral endplates were cut, stained with hematoxylin and eosin, and analyzed.
Findings:
Radiographically in the Boxer, adjoining thoracolumbar vertebrae were fused by osteophytes bridging the intervertebral space (IS) mainly ventrolaterally, but also dorsally, including the facet joints. Throughout, the NBF at the level of the IS continued in equal thickness into NBF at the ventral aspect of the center of the vertebral body. In the crossbred, NBF was predominantly located at the level of the IS, but deficient ventrally to the center of the vertebral body and at the facet joints. Hence, the ventral contour of the spine resembled the undulating contour of a bamboo. Intervertebral discs were partly affected in both pathologic specimens on computer tomograms. Histologically, intervertebral discs were unorganized, myxomatous, and partly necrotic. The cortex of the vertebral endplates was heterogeneously thickened; the amount of osteoblasts was increased. The tip of the spurs usually showed areas of lymphocytes and plasma cells. The structure in between the spurs consisted of fibrocartilage. None of these findings were evident in the control.
Conclusion:
Morphologically, the grade and the distribution of NBF along the thoracolumbar spine varied between the Boxer and the crossbred: The Boxer shared findings seen in DISH, the crossbred findings seen in ISA. Intervertebral disc alterations were seen in both pathologic specimens, however radiographically in a lesser extent. Moreover, highly active zones of remodeling were seen on microscopic sections in both dogs. According to histology additional collagen typing of the fibrocartilaginous tissue within the subdiscal space could be of interest to further analyze whether the intervertebral disc or the enthesis are involved in this process.
COMPUTED TOMOGRAPHY FINDINGS OF ENZOOTIC NASAL TUMOR IN SMALL RUMINANTS: A REPORT OF THREE CASES
M. N. Patsikas, N. Giadinis, I. Vlemmas, P. Papadopoulou, C. Soultani, E. Ioannidou From the Faculty of Veterinary Medicine, School of Health Sciences, Aristotle University of Thessaloniki, Thessaloniki, Greece (Patsikas, Giadinis, Vlemmas, Papadopoulou, Soultani, Ioannidou).
Background:
Enzootic nasal tumor (ENT) of small ruminants is an important viral disease, which presents many difficulties in antemortem diagnosis.
Purpose:
In this study, computed tomography (CT) findings of ENT in three small ruminants (two goats and one sheep) are described, as to the author's knowledge they have not been reported before.
Methods:
Plain radiographs of the thorax and head were performed in three small ruminants with weight loss, dyspnea, and stertorous breathing. The animals were subsequently euthanized and their heads underwent CT examination using a spiral CT scanner with slice thickness of 3 mm with no overlap. Necropsy as well as histopathological and virological examinations was performed. Correlation between CT findings and histopathological lesions were done.
Findings:
Plain radiographs of the head showed diffuse soft tissue opacity in the nasal cavities in all animals. Thoracic radiographs detected a well‐defined soft tissue opacity in the caudodorsal mediastinum of the sheep, which was identified as abscess due to Corynebacterium pseudotuberculosis (caseous lymphadenitis) in postmortem examination. CT of the heads revealed obliteration of both nasal cavities by soft tissue density mass (24–60 HU) with areas of mineralization. Destruction of the nasal turbinates and lysis of the nasal septum, frontal and sphenoid bones were also evident. Frontal sinuses were obscured. All CT findings were confirmed at necropsy. Histopathological and virological diagnosis was ENT.
Conclusion:
CT findings of ENT resemble those of malignant neoplasms invading nasal cavities. CT is useful for the diagnosis and extension of the lesions of ENT in small ruminants.
THE USEFULNESS OF DY NAMIC TISSUE PERFUSION MEASUREMENT IN SMALL ANIMAL VETERINARY PRACTICE
M. DziĘcioŁ, A. Wojciech, T. Scholbach, K. Wojciech, N. Wojciech From the Department of Reproduction (Dzięcioł, Wojciech), Department of Surgery, Wrocław University of Environmental and Life Sciences, Wrocław, Poland (Wojciech, Wojciech), and Chemnitz Clinics, Hospital for Children and Adolescents, 09116, Flemmingstr., Chemnitz, Germany (Scholbach).
Background:
The evaluation of tissue perfusion in small animals with the use of traditional methods could be difficult because of small size of organs and vessels.
Purpose:
The aim of this study was to evaluate the feasibility of assessment of tissue perfusion in various organs in dogs with the use of color Doppler techniques and with objective, digital evaluation of the images.
Methods:
The study was conducted in the Department of Reproduction and Department of Surgery of the local university. In six dogs (two 3‐year‐old male and four 3‐year‐old female beagles), the perfusion of thyroid gland and ovary was analyzed. For ultrasonographic examinations the ultrasound system MyLab25 Gold (Esaote, Italy) equipped with 12 MHz microconvex probe was used. The resulting video files were analyzed with the use of the modern software Pixel FluxR (Chameleon Software, Germany).
Findings:
In the thyroid gland as well as the ovary the ROI (region of interest) covered the whole organ, which allowed for analysis of total organ perfusion. While in some cases problems with visualization of the thyroid arteries hampers the proper performing of traditional pulsed Doppler analysis, the color Doppler technique combined with the use of PXFX software in all performed trials allowed for the digital, objective evaluation of perfusion of this organ. In the bitches, examined during different parts of the ovarian cycle, analysis of ovarian perfusion showed an increased number of pixels in this area after ovulation, which confirmed increased perfusion associated with corpus luteum formation and its activity. This was in accordance with findings of other authors who in previously published studies concluded that assessment of perfusion of the corpus luteum is more important than analysis of the size of this structure. Compatibility of our observations with above‐mentioned conclusions seems to confirm the usefulness of the presented method for the objective, digital evaluation of ovarian perfusion in dogs.
Conclusion:
The analysis of tissue and organ perfusion can be extremely important for the evaluation of its proper work and/or activity. In veterinary practice the small size of organs often limits the possibilities of the use of traditional pulsed Doppler technique and lack of routine methods for objective evaluation of the results obtained with the use of color Doppler technique limits its usefulness, especially in research. The application of the Pixel Flux software eliminates this inconvenience and allows for precise, objective analysis of the perfusion of all or part of an organ visualized with color Doppler technique.
COMPUTED TOMOGRAPHY FEATURES OF A LACRIMAL GLAND TUMOR IN A DOG
A.‐L. Etienne, M. Grauwels, G. Storms, A. Liotta, G. Bolen From the Diagnostic Imaging Section (Etienne, Liotta, Bolen) and Ophtalmology Section, Department of Small Animals and Equidae Clinic, Faculty of Veterinary Medicine, University of Liege, Liege, Belgium (Grauwels, Storms).
Background:
In the dog, neoplasms of the lacrimal gland are rare. Lacrimal cysts were also reported. Computed tomography (CT) scan findings of these lacrimal gland abnormalities have not been described in dogs.
Purpose:
The present report describes clinical and CT features of a lacrimal gland tumor in a dog.
Methods:
A 3.5‐year‐old female intact Leonberger with a good general health status was presented with a progressive swelling at the level of the superotemporal area of the left upper eyelid and of the bulbar conjunctiva since 3 weeks. Ocular examination revealed a slight enophthalmia and a ventronasal strabismus, the presence of a firm mass in the upper eyelid that was anteriorly well circumscribed and extending posteriorly between the globe and the orbital ligament.
Findings:
A CT scan pre‐ and postcontrast studies of the head were performed. A 3 cm hypoattenuating (±10 HU) ovoid mass was observed dorsolateral to the indented left globe. This mass was well circumscribed by an unevenly thick irregular soft tissue attenuating wall (50 HU), which was strongly contrast‐enhanced (150 HU) and measured about 2–10 mm in thickness. The left globe was displaced caudoventromedially. No adjacent bone lesion or foreign body was observed. No normal left lacrimal gland was observed. These findings were suggestive of an abscess or an orbital tumor with a necrotic/cystic center; a lacrimal cyst was considered less likely because of the unevenness and the thickness of the wall. Complete surgical ablation was done by a modified lateral orbitotomy. Histopathology revealed a mixed benign tumor of the lacrimal gland and a lymphocytic necrotic adenitis. Six months postoperatively no recurrence is noted and the tear production is 15 mm/min.
Conclusion:
The lacrimal gland is located dorsolateral to the globe and produces tears. In human medicine, a wide range of lacrimal gland pathologies were described and assessed by computed imaging techniques. Pleomorphic adenomas (mixed benign tumor) could reveal irregular bone erosion and could undergo malignant transformation. In dogs, lacrimal cysts were described as thin‐walled structures. Only few articles reported lacrimal gland tumor. To the authors’ knowledge, CT findings of a lacrimal gland tumor have not been described previously. In conclusion, lacrimal gland tumor should be included in the differential diagnosis of a firm ovoid cystic/necrotic mass dorsolateral to the orbit.
CONTRAST‐ENHANCED MAGNETIC RESONANCE IMAGING IN HORSES USING INTRAVENOUS OR INTRA‐ARTERIAL INJECTION OF GADOLINIUM
D. De Zani, M. Di Giancamillo, M. Longo, D. D. Zani From the Reparto di Radiologia, Azienda Polo Veterinario di Lodi, Universita degli Studi di Milano, Milano, Italy (De Zani, di Giancamillo, Longo, Zani).
Background:
The use of contrast‐enhanced MRI for the detection of the equine orthopedic pathologies is poorly described. Nevertheless, in these few studies, enhanced MRI seems to allow to differentiate active lesions from chronic pathologies and highlight ambiguous lesions.
Purpose:
The aim of this study is describe and compare the effectiveness of enhanced MRI in equine patient with lameness localized to the foot region using a single IV bolus dose of gadolinium contrast or regional intra‐arterial delivery of contrast medium.
Methods:
Six horses with lameness localized to the foot that underwent contrast‐enhanced MRI were included in the study. Precontrast sequences included Turbo 3D, T1‐weighted gradient echo (T3DT1), short tau inversion recovery (STIR), and dual‐echo sequences. Postcontrast T3DT1 were acquired after IV (0.1 ml/kg in the jugular vein) or intra‐arterial (0.02 ml/kg in the radial artery) gadolinium (Omniscan, GE Healthcare, Milan, Italy) administration. Via a catheter inserted before entering the MRI room, a bolus of the contrast agent was injected within 20 s. Regions of interest (ROI) were collected from both pre‐ and postcontrast images. The ROIs included prearranged regions and areas with abnormal findings. In order to evaluate the value of contrast enhancing, ratios between pre‐ and postcontrast ROIs were calculated.
Results:
Gadolinium was injected IV in two patients and in four horses contrast medium was administered using intra‐arterial regional delivery. The most consistent contrast enhancing was noted at the level of the medial palmar digital artery, of the dorsal aspect of the cortex of the middle phalanx, and of the distal interphalangeal joint. Injured structures that revealed greater increase in signal in postcontrast images were the deep digital flexor tendon (DDFT), the navicular spongiosa, and the peritendinous tissues. No significant enhanced was noted in the DDFT in absence of lesions. In two horses distal impar sesamoidean ligament (DSIL) injuries were observed only on postcontrast images. No involvement of the peritendinous tissues was noted on precontrast sequences while in postcontrast images it was observed in three cases. Enhanced MRI highlighted the presence of reactive scar tissue in four horses. Regional intra‐arterial administration of gadolinium provided high ratio of contrast enhancement.
Conclusion:
Enhanced MRI allowed to differentiate active lesions from chronic ones and to identify lesions otherwise non detectable. The high signal on the dorsal aspect of the cortex of the middle phalanx was considered a partial volume averaging artifact. Regional intra‐arterial administration of gadolinium was considered the best choice due to the higher signal and lower volumes required. Nevertheless, enhanced MRI using both IV or intra‐arterial injection of gadolinium increased the diagnostic capability of MRI in horses with foot lesions. Further studies are compulsory to statistically evaluate the sensitivity and specificity of MRI using regional intra‐arterial administration of gadolinium.
COMPARISON OF COMPUTED TOMOGRAPHY AND RADIOGRAPHY FOR THE DIAGNOSIS OF EQUINE SINONASAL DISEASE
G. Manso‐diaz, M. Garcia‐lopez, L. Maranda, O. Taeymans From the Department of Animal Medicine and Surgery, School of Veterinary Medicine, Universidad Complutense de Madrid, Madrid, Spain (Manso‐Diaz), Department of Clinical Sciences, Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA (Garcia‐Lopez), Department of Quantitative Health Sciences, University of Massachusetts Medical School, Worcester, MA (Maranda), and Department of Clinical Sciences, Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA (Taeymans).
Background:
Radiographic assessment of the equine paranasal sinuses is limited by their complex anatomic conformation and their superimposition with other structures, such as teeth and nasal passages. Computed tomography (CT) has become the gold standard for studying this region as it avoids superimposition of surrounding structures, and has higher inherent contrast resolution.
Purpose:
The purpose of this study was to compare radiographic and CT findings in cases of equine sinonasal disorders.
Methods:
A retrospective review of radiographic and CT images in 42 horses with sinonasal disorders was performed. Affecting conditions included dental disease (n = 32), sinus cyst (n = 6), ethmoid hematoma (n = 3), and nasal tumor (n = 1). Two observers simultaneously reviewed the images (first radiographs, then CT for each horse). Observers were blinded to final outcome, and interpretation was reached by consensus. Paranasal sinuses were individually assessed for presence of pathology, mucosal thickening, fluid lines, and/or increased opacity. Contingency tables were constructed to classify horses according to their radiography and CT results. Estimates of sensitivity (Se), specificity (Sp), and positive and negative predictive values (PPV and NPV) were calculated for radiographic findings, using CT results as the gold standard.
Findings:
Compared to CT, radiographs correctly identified sinus abnormalities in rostral and caudal maxillary sinuses in19/28 and 18/24 horses, respectively. Involvement of the conchofrontal sinus was identified radiographically in 15/23 horses, while evidence of ventral conchal and sphenopalatine involvement was seen in 10/23 horses and 3/18 horses, respectively. Radiography demonstrated higher sensitivity for caudal (75%) and cranial (68%) maxillary sinus pathology, while the lowest sensitivity was observed for sphenopalatine sinus pathology (17%). Specificity was high for radiography in all cases, and varied between 93% and 100% for the cranial maxillary and the sphenopalatine sinuses, respectively. Sensitivity of radiographs was optimal when respiratory mucosa was thicker than 10 mm in combination with fluid levels, or when total obstruction of the sinus occurred. Radiographic changes were infrequently seen when respiratory mucosa thickening was the only finding on CT.
Conclusion:
Using CT as a gold standard, the sensitivity of radiology was poor to fair when evaluating rostral and caudal maxillary or conchofrontal sinuses. Radiology appeared to be less accurate than CT at revealing changes within individual sinus compartments, especially in cases of ventral conchal and sphenopalatine sinus involvement. Radiographic changes are not usually present in horses that only have thickening of the respiratory epithelium.
COMPUTED TOMOGRAPHIC DIAGNOSIS OF A MIDBRAIN HEMATOMA IN A ROYAL EAGLE (AQUILA CHRYSAETOS HOMEYERI)
J. C. Santos, C. Monteiro, H. Fernandes, A. Canadas, L. Guardao, C. S. Baptista From the Department of Veterinary Clinics, UPVET, Institute of Biomedical Sciences Abel Salazar—University of Porto (ICBAS‐UP), Porto, Portugal (Santos, Guardao, Baptista), Santo Inacio Zoo, Vila Nova de Gaia, Portugal (Monteiro, Fernandes), and Department of Molecular Pathology and Immunology, Institute of Biomedical Sciences Abel Salazar—University of Porto (ICBAS‐UP), Porto, Portugal (Canadas).
Background:
In birds, head trauma, usually due to crashes into trees, windows, or cars can be a cause of central neurologic signs. To the authors knowledge there are no reports describing the computed tomographic (CT) characteristics of hematomas in Aquila chrysaetos homeyeri.
Purpose:
The aim of this case report is to characterize the clinical signs and the CT features of a histologically confirmed midbrain hematoma in a royal eagle (A. chrysaetos homeyeri).
Methods:
A 30‐year‐old royal eagle (A. chrysaetos homeyeri) presented with acute opisthotonus, suspected seizures, left head tilt, and inability to fly and stand. There was no known history of trauma and the main differentials were hypovitaminosis B or E and intracranial disease due to infection or neoplasia. CT of the head, before and after IV injection of iodinated contrast medium (250 mg I/Kg) administered into the basilica vein, was performed with a 16‐slices scanner. The animal was anesthetized with isofluorane and positioned in sternal recumbency. Axial images were acquired using a helical scanning mode with 0.6 mm slice thickness and soft tissue and bone reconstruction algorithms.
Findings:
CT scans showed a hyperdense nonenhancing homogeneous well‐delineated round area, ventral to the third ventricle, with a density of 65 HU, and 6 mm in its highest diameter (6 × 4.4 × 5.4 mm). This lesion did not produce deviation of the falx cerebri and was confined to the midbrain (the ventricular system was normal). Such lesion was suggestive of acute intraparenchymal hemorrhage but additional differentials such as granuloma or neoplasia could also be considered. The animal was humanely euthanized and a necropsy was performed. Brain examination revealed the presence of an acute cerebral hematoma which was histologically confirmed.
Conclusion:
CT has been used successfully to locate intracranial lesions in birds, although it was only 80% sensitive. CT is an excellent method for the detection of acute intraparenchymal brain hemorrhage due to the high content of globin and fibrin resulting in a hyperattenuated area. The attenuation values of this lesion (65 HU) were within the limits recognized for acute hemorrhage. Additionally, the well‐defined limits and its nonenhancing nature, due to the poor blood perfusion of the acute phase of the hemorrhagic event, strongly suggest the diagnosis of a cerebral hematoma.
SONOGRAPHIC EVALUATION OF ECHOGENICITY AND ECHOTEXTURE OF THE MAMMARYGLAND OF PREPUBERTAL EWE LAMBS BY HISTOGRAM WITH GRAY LEVELS
A. Dantas, E. R. Siqueira, S. Fernandes, V. M. D. V. Machado, R. V. Santos, P. T. R. Santos From the FMVZ‐UNESP, Sao Paolo, Brasil (Dantas, Siqueira, Fernandes, Machado, Santos, Santos).
Background:
The mammary gland is formed of variable amounts of adipose, connective, and epithelial tissue of which the amount varies during the growth, especially in the period before puberty. The use of ultrasonography and, later, digital processing of the image obtained with the preparation of a gray‐level histogram, is a potential tool for the study of mammary gland in ruminants.
Purpose:
The present study aims at processing images of breast ultrasound to verify the applicability of the gray‐level histogram in lambs during growth, providing quantitative values for the echotexture and echogenicity of the breast tissue.
Methods:
We used 20 healthy lambs, with an initial age of 90 days. Four ultrasound examinations (with an interval of 60 days each) were performed by the same operator, using the same apparatus (Esaote R brand Mylab 30 model) and an 18 MHZ transducer. The ultrasound images from the last two tests were used, because it was when the best visibility of breast tissue occurred. The images of breast structure were divided in areas of 0.5 cm by 0.5 cm and analyzed using specific computer software (Photoshop CS4 R). The intensity of each pixel is expressed by shades of gray scales with a minimum (absolute black) and maximum (absolute white) set at 0 and 255. In order to analyze, the echotexture and echogenicity of the breast tissue n most, n all, standard deviation, and mean were measured. Quantitative measurements were evaluated by n most and variables n all, being n most referring to the number of identical pixels and n all being the total number of pixels in the selected area, thereby, the ratio between n most/n all represents the echotexture value of the selected study area. The standard deviation was considered as the echo amplitude. The echotexture was considered homogeneous when the value of the n most/ n all ratio was higher than the value measured by the standard deviation and heterogeneous when the ratio was lower. The echogenicity of the tissue was determined by l mean variable, which indicates the average brightness of the selected region, being the lower its value, the lower the echogenicity. ANOVA was held in order to determine differences between the experimental moments, with a significance level of 5%.
Findings:
The mean values for the ratio n most/ n all, standard deviation and l mean in both periods studied were, respectively, 11.14 and 11.50; 23.44 and 22.37; and 93.93 and 92.37. Although the presented parameters have not yet been standardized in the mammary gland of ruminants, the results of this study demonstrate that the mammary gland of lambs during growth showed high echogenicity and heterogeneous echotexture. Since there was no difference for the values on the gray‐level histogram during this study time, one cannot infer that there is a higher or lower deposition of adipose tissue.
Conclusion:
This study demonstrates that the gray‐level histogram breast ultrasound pictures of healthy lambs during growth proved to be an efficient and easy to perform method, indicating its applicability in the breast tissue analysis in different ages.
COMPUTED TOMOGRAPHY FINDINGS OF ABDOMINAL AORTIC THROMBOSIS ASSOCIATED WITH PROTEIN LOSING NEPHROPATHY IN A DOG
M. N. Patsikas, N. Tsimitris, P. L. Papadopoulou, C. S. N. Soultani, G. Mprellou From the Faculty of Veterinary Medicine, School of Health Sciences, Aristotle University of Thessaloniki, Thessaloniki, Greece (Patsikas, Tsimitris, Papadopoulou, Soultani, Mprellou).
Background:
Aortic thrombosis is a rare condition in the dog and ultrasonography (US) and computed tomography (CT) are used to determine the absence of blood flow in the aortic lumen.
Purpose:
The purpose of this work is to describe the CT findings of abdominal aortic thrombosis in a dog with concurrent subclinical protein‐losing nephropathy. To the authors knowledge this condition has not been described before.
Methods:
An 8‐year‐old mixed breed dog was referred to the Companion Animal Teaching Clinic with a 5‐day history of pelvic limb paraplegia, partial anorexia, and polydipsia. Physical examination revealed cold extremities, absent femoral pulses bilaterally, and sinus bradycardia. Laboratory abnormalities included a stress leukogram, severe hypoalbuminemia and proteinuria, antithrombin III deficiency along with elevated d‐dimer concentration. Plain abdominal radiography, abdominal ultrasonography, and CT angiography of the caudal thoracic and abdominal aorta using a helical CT scanner were performed. Pre‐ and postcontrast CT scans of 3 mm slice thickness with no overlap were made with manual injection of Iohexol 350 mgI/ml at a dose of 2 ml/kg BW.
Findings:
Plain abdominal radiographs were within normal limits. Abdominal ultrasonography showed mildly echogenic masses in the lumen of the abdominal aorta, caudal of the origin of nephric arteries. Color and pulse Doppler technique demonstrated no blood flow in the caudal aorta. Echocardiographic examination was found within normal limits. CT angiography of the aorta revealed severe attenuation of contrast medium in the aorta caudal to the L1‐L2 intrervertebral space, compatible with nearly complete occlusion of the aortic lumen. The enhancement of the kidneys was found normal. In precontrast sections the HU values in the center of the aortic lumen were ranged 19–49 HU and 18–36 HU in not occluded and in occluded regions, respectively. In postcontrast sections the HU values in not occluded and in occluded regions were ranged 206–165 HU and 16–37 HU, respectively. Euthanasia was elected due to poor prognosis of the case. Postmortem examination showed aortic thrombosis caudal to nephric arteries extending to the left and right femoral arteries, with nearly complete occlusion of the vessels’ lumen. Histopathologic examination of the kidneys revealed membranous glomerulonephritis, glomerulosclerosis, and chronic interstitial nephritis.
Conclusion:
CT angiography is considered an important advancement in the diagnosis of aortic thrombosis in dogs, determining the exact thrombus location, as well the degree of lumen obstruction.
DETERMINATION OF REFERENCE VALUES AND FREQUENCY OF OCCURRENCE OF THE PATELLA ALTA IN THE GERMAN SHEPHERD DOGS
A. Lojszczyk‐szczepaniak, P. Silmanowicz From the Laboratory of Radiology and Ultrasonography, Department and Clinic of Animal Surgery (Lojszczyk‐Szczepaniak) and Department and Clinic of Animal Surgery, Faculty of Veterinary Medicine, University of Life Sciences in Lublin, Lublin, Poland (Silmanowicz).
Background:
Orthopedic diseases are a very common reason for medical consultation. In many cases it is caused by patellofemoral disorders, which constitute up to 10–40% of all problems associated with the stifle joint. One of the most common diseases is medial patellar luxation, which is increasingly found in large breed dogs. It has been documented that one of the predisposing factors is an abnormal high vertical position of the patella (patella alta). Recognition of patellofemoral malalignment and maltracking is based on history, clinical examination, and on radiological evaluation. Radiological measurements allow for precise determination of anatomical relationships between particular structures of the normal stifle joint as well as for detection of pathological disorders.
Purpose:
The research in the work aims at:
clinical and radiographic evaluation (with measurements of selected parameters) of the stifle joints in clinically healthy German Shepherd Dogs with no evidence of lameness,
determination of the reference values for the vertical position of patella in the stifle joint and comparison of these values with the results of previous studies.
Methods:
The research included 65 working German Shepherd Dogs. Radiological evaluation considered mediolateral and craniocaudal radiographs of the stifle joints as well as skyline views of patellofemoral joints. Measurements of particular parameters were made: the degree of knee flexion, the patella (P) and the patellar ligament (L) length (mm), the vertical position of the patella (Insall Salvati ratio, L:P). The vertical position of the patella was measured on the basis of two methods. The first method was based on the calculation of the patellar ligament length from the distal part of the patella to the small indentation on the tibial tubercle (Lj). In the second method, the length was measured from the distal part of the patella to the proximal extent of the tibial tuberosity (Lm). In both methods patellar length (P) was measured in the same manner. All above‐mentioned parameters were measured on mediolateral radiographs. The craniocaudal and skyline views were used to determine the position of the patella in the stifle joint.
Findings:
The outcome of the work allowed to determine reference values for the vertical position of the patella in German Shepherd Dogs. They ranged from 1.46 to 1.79 using the first method of measurement (Lj/P) and 1.72–2.1 using the second one (Lm/P). Simultaneously it was observed that the high‐riding patella (patella alta) occurs in German Shepherd Dogs ranging from 14.78% to 26.95% depending on the used range of reference values: own or from the previous studies.
Conclusion:
The examined dogs with patella alta were clinically healthy dogs, without signs of musculoskeletal disorders and lameness or pain in the stifle joint, which is commonly associated with a high‐riding patella. There were also no observations of trochlear dysplasia symptoms and the characteristics of patellar displacement. Therefore the significance of the high‐riding patella in dogs should be further observed.
UNDERSTANDING EQUINE DISTAL LIMB ANATOMY AND RADIOLOGY: AN INTEGRATED APPROACH TO LEARNING USING AN INTERACTIVE MODEL
S. Dickey, R. Weller From the Department of Veterinary Clinical Sciences, The Royal Veterinary College, North Mymms, Hertfordshire, UK (Dickey, Weller).
Background:
Amalgamation of anatomy and radiology is an essential skill for any veterinarian, but the learning process by which this is best achieved remains unclear. While there is no “one size fits all” teaching process, there is still a need to assess the usefulness of teaching aids and learning styles in veterinary education.
Purpose:
The aim of this study was to create an interactive model of the equine distal limb with removable bones and an accompanying booklet of relevant clinical anatomy and radiology. This model was compared to traditional resources including textbooks and websites.
Methods:
Data were collected from 49 fourth‐year veterinary students that had completed lectures in anatomy and radiology and begun clinical rotations. The students were randomly assigned to either the model or control group after performing an initial baseline knowledge test. The groups were then given 20 min with their allocated resources for independent study. Postsession knowledge was assessed with a second test. Confidence before and after the session and enjoyment was evaluated with a questionnaire.
Findings:
The study showed a difference in postsession knowledge scores, with the model group scoring significantly higher than the control (P = 0.014). When anatomy and radiology components were assessed individually and divided by gender, males using the model performed significantly better than males in the control group (P = 0.017), but not females. Questionnaire responses indicated that more than one‐third of students preferred interactive models as their primary learning resource.
Conclusion:
The results indicate that the model‐aided anatomical understanding and a significant number of students found it to be a useful resource. An interactive model may prove useful to a subset of students as a supplement to clinical rotation teaching as it allows students to interact with the model and each other, as well as work at their own pace.
IMAGING FEATURES OF AN ECTOPIC THYROID CARCINOMA IN A HORSE
G. Manso‐diaz, R. A. Garcia‐fernandez, M. De Los Angeles Jimenez Martinez, R. H. Vilella, I. Santiago From the Department of Animal Medicine and Surgery, School of Veterinary Medicine, Universidad Complutense de Madrid, Avda. Puerta de Hierro s/n, Madrid, Spain (Manso‐Diaz, Garcia‐Fernandez, de los Angeles Jimenez Martinez, Vilella, Santiago).
Background:
Mediastinal tumors are uncommon in horses. Lymphoma is the most common thoracic neoplasm, but remains a rare cause of thoracic disease in this species.
Purpose:
The diagnosis of thoracic neoplasia can be difficult due to the nonspecific clinical signs and their overlap with other pulmonary diseases.
Methods:
A 30‐year‐old Standardbred gelding was presented with a 5 months history of pleural effusion. The horse was previously treated with a combination of antibiotics and nonsteroidal anti‐inflammatories, but no improvement was observed.
Findings:
On presentation, clinical examination revealed tachypnea, tachycardia, and the presence of ventral edema. Upon auscultation, there were bilaterally increased lung sounds in the dorsal lung fields, primarily on the right side. Hematology revealed anemia and hyperfibrinogenemia, and serum chemistry revealed mild hyperglycemia and azotemia. Thoracic ultrasonography showed the presence of a large amount of anechoic fluid within the pleural space bilaterally. Subsequently right hemithorax thoracocentesis was performed, obtaining 26 l of clear pleural effusion. Samples of pleural fluid and tracheal secretions were cultured, but no growth was observed. A second ultrasonography revealed a soft tissue structure located ventrally and extending from the fifth to 11th intercostal spaces on the right hemithorax. Radiographic examination showed a soft tissue opacity superimposed over the heart, obscuring the cranioventral lung fields and deviating the trachea dorsally. Diffuse increase of the caudal lung fields opacity was also observed, associated to several poorly marginated soft tissue nodular masses with a cavitated appearance. Fine needle aspiration of the mass was not diagnostic. A presumptive diagnosis of cranial mediastinal mass with pulmonary metastasis was made, and following consultation with the owner the horse was euthanized. At necropsy, a large, well demarcated, mass was located adjacent to the heart base. The mass was multilobular, densely cellular, encapsulated, and contained multifocal necrosis and hemorrhage. Neoplastic cells were arranged in compact aggregates or thick cords separated by thick fibrous stroma. Cells were polygonal to spindle shaped, with eosinophilic, finely granulated cytoplasm and central ovoid nuclei. Mitotic index was high, and there was marked anisocytosis and anisokaryosis. Within the lungs were multifocal to coalescing areas of fibrosis, which expanded and delineated septal walls. Alveoli were lined by type II pneumocytes and the majority of air spaces contained numerous macrophages and neutrophils. The neoplasm at the heart base was an undifferentiated carcinoma arising from ectopic thyroid tissue.
Conclusion:
To authors’ knowledge this is the first reference describing the imaging findings of this kind of tumor in horses, although uncommon it should be included as a differential diagnosis of mediastinal masses in this species. The lesions in lungs were consistent with equine multinodular pulmonary fibrosis; therefore it also should be considered as a possibility in horses with radiographic pulmonary nodular pattern associated with tumors.
DEVELOPMENT OF A RADIOGRAPHIC PROTOCOL FOR WHITE RHINOCEROS (CERATOTHERIUM SIMUM) FORE AND HIND FEET AND A DESCRIPTION OF THEIR NORMAL RADIOGRAPHIC ANATOMY
R. J. Dudley, S. P. Wood, R. Weller, J. R. Hutchinson From the Department of Veterinary Clinical Sciences, The Royal Veterinary College, North Mymms, Hertfordshire, UK (Dudley, Wood, Weller, Hutchinson).
Background:
Foot pathology is a common and significant health concern in captive rhinoceroses worldwide. Soft tissue diseases of the feet are frequently diagnosed but reports of osteopathologies in live rhinoceroses feet are relatively scarce in the literature. Recent research suggests that this may reflect underdiagnosis rather than true prevalence. The normal radiographic anatomy of rhinoceros feet has not been established and there are no radiographic protocols currently described.
Purpose:
Our aims were to develop a radiographic protocol for standing white rhinoceros (Ceratotherium simum) fore and hind feet and to describe their normal radiographic anatomy.
Methods:
Computed tomographic (CT) images were obtained of 11 cadaveric fore feet and nine cadaveric hind feet from nine different white rhinoceroses and assessed for pathology. A single manus and pes deemed free of pathology were radiographed using a range of different projections and exposures to determine the best protocol. Optimal projections were determined for each bone focusing on highlighting areas where pathology has been previously described. The normal radiographic anatomy of the white rhinoceros manus and pes were described using radiographs and 3D models produced from the CT images.
Findings:
The most useful projections for the fore foot were found to be D45Pr‐PaDiO (digit III), D40Pr60M‐PaDiLO (digit II), and D35Pr50L‐PaDiMO (digit IV). The projections deemed to be most useful for the hind foot were D60Pr‐PaDiO (digit III), D45Pr45M‐PaDiLO (digit II), and D40Pr35L‐PaDiLO (digit IV). The primary beam was centered 5–7 cm proximal to the cuticle on the digit of interest.
Conclusion:
The radiographic protocol we have developed along with the established normal radiographic anatomy we have described will allow for more accessible and effective diagnosis of white rhinoceros foot osteopathologies.
MAGNETIC RESONANCE IMAGING OF MULTIPLE LESIONS OF THE NUCHAL STRUCTURES IN A SHOW JUMPER HORSE—CASE REPORT
V. M. Vasconcelos Machado, A. C. Saldanha De Aguiar, L. M. Cintra Bueno, G. Fernandes Viana From the Department of Animal Reproduction and Veterinary Radiology, Sao Paulo State University, Rio Claro, SP, Brazil (Vasconcelos Machado, Saldanha de Aguiar, Cintra Bueno, Fernandes Viana).
Background:
Neck pain can lead to lameness and poor performance in horses. The anatomy of this region is very complex. Magnetic resonance imaging has been used to describe the occiptoatlantoaxial ligaments in people and recently in dogs. Additionally, there are few MRI anatomy studies as a reference for affected nuchal structures in horses.
Purpose:
This case report aims to highlight the importance of MRI to evaluate lesions of the nuchal structures in horses.
Methods:
A 7‐year‐old Westfalen mare with history of stiffness in flexing the neck, swollen area in the left cervical region, and no neurologic signs was referred to the Department of Veterinary Radiology for an MRI scan of the cervical region. Previously, the diagnosis of bursitis of nuchal bursa, nuchal ligament desmopathy, and an entesophyte in C2 was reported with an ultrasound study. MRI examination was performed using an Esaote® Vet‐MR Grande, 0.25 Tesla and a table for horses (Shanks MRI Veterinary Equipment). The horse was positioned in lateral recumbency, with the neck and head in neutral position, under general anesthesia, maintained with isoflurane and oxygen using Mallard Medical Inc‐2800‐RM equipment and Medrad® Vertis RM 8600 monitor. Dorsal, sagittal, and transverse images were acquired and the sequences included were STIR (short tau inversion recovery), FSE‐T2 (fast spin echo). Additionally, the gradient Echo sequences SST1‐3D, 3D HYCE, and TURBO‐3D were acquired.
Findings:
MRI revealed heterogeneous hyperintense signal on sagittal and transverse images of the dorsal funicular portion of the nuchal ligament (FSE‐T2; 3D HYCE) extending from its insertion on the occipital tuberosity to C2. Cranial bursa distension and hyperintense signal (FSE‐T2; 3D HYCE) on sagittal and transverse images along the ventral aspect of the funicular part of the nuchal ligament. Hyperintense signal (STIR; SST1‐3D) and irregularity of occipital tuberosity at the nuchal funicular ligament insertion. Heterogeneous signal of the semispinalis capitis tendon at its occipital insertion (FSE‐T2; 3D HYCE), more evident on the left side. Hyperintense signal in the semispinalis capitis muscle and heterogeneous signal of the nuchal laminar ligament insertion at C2 dorsal aspect (FSE‐T2; 3D HYCE). These features are compatible with nuchal funicular ligament desmopathy and enthesopathy, significant cranial bursitis, tendon semispinalis capitis tendinitis, semispinalis capitis myositis, occipital enthesopathy, and nuchal laminar ligament enthesopathy. Considering MRI diagnosis the horse was submitted to Kinesio taping therapy.
Conclusions:
Even though some lesions could be detected by other imaging modalities MRI scan provided much more detailed information of these lesions. Additionally, MRI revealed other lesions that were not detected previously. MRI can help diagnosing subtle changes in soft tissue and allow a better visualization of complex regions such as the nuchal region.
