Introduction
The use of elastography, which is widely used in human medicine to measure the elastic properties and stiffness of soft tissue, is gradually increasing in veterinary medicine. Ultrasonographic elastography techniques include strain imaging and shear wave imaging. Shear wave imaging techniques include 1-D transient elastography and acoustic radiation force elastography.1 Acoustic radiation force elastography can be further classified as PSWE and 2-D shear wave elastography.1
Strain imaging has been proposed as a method for examining the liver, spleen, and kidneys in cats.2 This technique involves analysis of tissue stiffness on the basis of compression force, similar to the principles of palpation. Therefore, the technique is operator dependent and influenced by the subject's body conformation (ie, body weight and depth of the liver).3 It is also prone to significant interoperator variability4 and cannot be used to quantitatively evaluate organ stiffness. In contrast, shear wave imaging involves use of an acoustic radiation force impulse, which allows for quantitative evaluation of stiffness.4 Furthermore, shear wave imaging yields highly reproducible results, with lower interoperator variation.3–6
Elastography has been valuable for the diagnosis of liver diseases such as steatosis, fibrosis, and neoplasia in humans.7 Moreover, research in dogs has demonstrated its usefulness for assessment of hepatic fibrosis,8 differentiation between benign and malignant mammary tumors,9 and diagnosis of testicular disorders.10 Several studies11–14 have investigated the use of shear wave imaging for assessment of the spleen, kidneys, mammary glands, and testes in cats. Findings of a recent study15 suggest that renal parenchymal stiffness in cats is correlated with renal functional impairment. However, to our knowledge, elastography has not been evaluated for assessment of the liver parenchyma in healthy or diseased cats. Such research would be important to determine whether elastography can be used in the evaluation of hepatic diseases such as hepatic lipidosis and fibrosis in cats and to distinguish between benign and malignant liver neoplasms. The aim of the study reported here was to evaluate the stiffness of the liver parenchyma in healthy adult cats by means of PSWE and generate summary data that might be useful when interpreting results for other cats.
Materials and Methods
Animals
Eighteen healthy client-owned adult cats were recruited for inclusion in the study from the Veterinary Medical Teaching Hospital of Seoul National University. This number of cats was based on results of an a priori, 2-tailed sample size calculationa (α = 0.05; effect size = 0.8; power = 0.85), which indicated that a total sample size of 17 would be needed to find differences in Vs between the right and left portions of the liver if they truly existed. Cats were included in the study if they had unremarkable results of physical examination, hematologic analysis (including a CBC), serum biochemical analysis (alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, and γ-glutamyltransferase activities and total bilirubin concentrations), and thoracic and abdominal radiography. Conventional B-mode ultrasonography of the abdominal organs was also performed to rule out abnormalities in echotexture or echogenicity of the liver parenchyma or in the size or margins of any abdominal organ. Cats < 1 year or > 6 years of age were excluded, as were cats with evidence of hepatic or other diseases in their medical records. The study protocol was approved by the Institutional Animal Care and Use Committee of Seoul National University in Korea (protocol No. SNU-190821-7). Informed consent was obtained from all owners of participating cats.
Ultrasonography and PSWE
To prepare the cats for abdominal ultrasonography and PSWE, food was withheld for at least 8 hours. Hair over the ventral abdominal region was clipped. Ultrasonic coupling gel was applied to ensure optimal transducer-skin contact. Cats were restrained on a table in dorsal recumbency throughout the ultrasonographic examination, with no sedation or anesthesia used. Conventional B-mode ultrasonography was performed by the same operator (SP), who used a small 4- to 8-MHz curved-array transducer and 2- to 12-MHz linear-array transducer.b The echotexture (homogeneous or heterogeneous) and echogenicity (hyperechoic or hypoechoic) of the liver parenchyma and size (increased, decreased, or normal) and margins (rounded or sharp) of the abdominal organs were assessed and subjectively evaluated.
The Vs of the right and left portions of the liver was ultrasonographically measured by means of PSWE with a 5- to 18-MHz linear array transducerb under visual control in B mode. The right portion of the liver was defined as the right ventral portion (including the right medial and lateral lobes), typically located to the right of the gall bladder and cranial to the right kidney. The left portion of the liver was defined as the left ventral portion (including the left medial, lateral, quadrate, and caudate lobes) located medial to the stomach (Figure 1). All measurements were acquired with the probe positioned parallel to the ribs and within the intercostal space, while avoiding rib shadowing as much as possible. Minimal scan pressure was applied to reduce variations in measurement and avoid liver compression. Most PSWE examinations of the liver were performed during the normal end-expiratory phase to minimize the effects of respiratory movement.
Representative ultrasonographic images of the liver in a healthy cat during PSWE. A—Right portion of the liver as observed through an intercostal window. Notice the gallbladder (GB) located near the medial aspect of the liver. B—Left portion of the liver as observed through an intercostal window. Acoustic shadowing (asterisks) by the ribs is evident. Notice the stomach (S) located dorsolateral to the liver parenchyma. All measurements were performed at depths of fixed regions of interest between 1.1 and 2.0 cm.
Citation: American Journal of Veterinary Research 82, 4; 10.2460/ajvr.82.4.286
Quantitative evaluation of Vs was performed by use of the shear velocity function of the ultrasound machine. The operator positioned the caliper over the liver parenchyma, taking care to avoid blood vessels. The area of the fixed region of interest was 10 × 5 mm. Measurements of each portion (right or left) of the liver were continued until 10 valid measurements were obtained at depths between 1.1 and 2.0 cm of fixed regions of interest. Sufficient liver parenchyma was visualized through each intercostal window to enable PSWE at the determined depth. The number of Vs measurements that failed because of cat movement (eg, breathing or body movement) was automatically determined by the ultrasound machine, and the overall success rate (number of successful measurements/total measurements × 100) was displayed, as were summary statistics for Vs (ie, median, IQR [calculated as the difference between the 75th and 25th percentiles], and IQR/median [calculated as IQR/median × 100]). The reliability of the PSWE data was confirmed by application of criteria commonly used to determine the reliability of liver stiffness evaluation in humans16,17: 10 valid measurements, success rate ≥ 60%, and IQR/median value ≤ 30%.
Statistical analysis
Data were analyzed by 1 author (SP), who used statistical software.c Normality of the Vs data was assessed with the Kolmogorov-Smirnov test. Mean values for Vs and associated 95% CIs were determined for the liver overall and for the right and left portions specifically. Differences between right and left portions of the liver were evaluated with the paired t test. Correlations between Vs and cat body weight and between Vs and depth of Vs measurement were evaluated by means of Spearman correlation analysis.
Results
Animals
The 18 cats included 12 castrated males, 4 spayed females, and 2 sexually intact males. Mean ± SD age and body weight were 3.2 ± 1.6 years (range, 1 to 6 years) and 5.0 ± 1.4 kg (range, 3.3 to 8.6 kg), respectively. Cats were classified as domestic shorthair (n = 12), Siamese (2), Abyssinian (1), Ragdoll (1), Russian Blue (1), and Persian (1). B-mode ultrasonography and PSWE measurements were performed while a few cats (4/18) were vocalizing because these cats generally vocalized with every exhalation. Nevertheless, the measurements showed that a stable liver window had been achieved despite this vocalization, resulting in adequate success rates (≥ 60%) similar to those for other cats.
PSWE
The mean success rate for the repeated measurements of Vs per cat was 78% for the right portion of the liver and 81% for the left portion. The mean of the IQR/median values was 33% and 29%, respectively. In 1 cat, the liver parenchyma was too deep to be sufficiently visualized owing to obesity (body weight, 8.6 kg). Thus, values for the left portion of the liver could not be obtained for this cat (1/36 sets) and these data were not included in the statistical analysis, resulting in a total of 35 sets (10 valid measurements for each set) of measurements (18 for the right portion of the liver and 17 for the left portion). A success rate of < 60% was found for 2 of 35 (6%) sets of measurements, and an IQR/median value of > 30% was found for 10 of 35 (29%) sets.
The mean Vs was 1.46 m/s (95% CI, 1.36 to 1.55 m/s) for the right portion of the liver, 1.36 m/s (95% CI, 1.26 to 1.47 m/s) for the left portion, and 1.43 m/s (95% CI, 1.35 to 1.51 m/s) overall. The mean Vs of the right portion was significantly higher than that of the left portion (P = 0.03; Figure 2). The mean ± SD depth of Vs measurement was 1.54 ± 0.25 cm. No significant correlation was found between Vs and cat body weight (P = 0.54) or between Vs and the depth of Vs measurement (P = 0.71; Figure 3).
Box-and-whisker plots comparing the stiffness (Vs) of the right (n = 18) and left (17) portions of the liver in healthy cats, as measured via PSWE. Mean Vs of the right portion (1.46 m/s) was significantly (P = 0.03) higher than that of the left portion (1.36 m/s). The central horizontal line within each box represents the median, the boundaries of each box represent the IQR, whiskers represent maximum and minimum values, and the circle represents an outlier (a value between 1.5 and 3 times the IQR).
Citation: American Journal of Veterinary Research 82, 4; 10.2460/ajvr.82.4.286
Scatterplot showing the distribution of values for Vs in the cats of Figure 2 at various depths of measurement. No significant (P = 0.71) correlation was found between Vs and measurement depth.
Citation: American Journal of Veterinary Research 82, 4; 10.2460/ajvr.82.4.286
Discussion
Diffuse liver diseases, such as inflammatory liver disease, vacuolar hepatopathy, round-cell neoplasia, prenodular (early) metastatic disease, lipidosis, and other non-nodular liver diseases, are encountered commonly in cats and dogs.18 Results of conventional B-mode ultrasonography are mainly based on subjective evaluation. Differentiation between the acoustic impedances of the liver parenchyma in diffusely diseased versus healthy liver is difficult with current ultrasound technology owing to the lack of prominent differences between these liver states.18
Recent research demonstrated a significant difference in Vs between dogs with clinically relevant hepatic fibrosis and those without clinically relevant hepatic fibrosis, similar to findings in human patients.8 Elastography has also been shown to be a feasible technique for distinguishing between benign and malignant mammary tumors in humans,19 female cats,13 and female dogs.9 Although biopsy and histologic examination is the gold standard technique for diagnosing diffuse liver diseases,20 elastography may be used as a noninvasive diagnostic tool in the future, possibly in the monitoring of the clinical stage of disease during treatment.
Hepatic lipidosis is one of the most common hepatobiliary disorders with a high mortality rate in cats.21,22 The prevalence of NAFLD has gradually been increasing in humans, and the applicability of elastography for diagnosing NAFLD has already been demonstrated. Specifically, a significant difference in Vs has been observed between humans with steatosis and healthy volunteers, whereby Vs decreases with each increase in steatosis grade.7 Because the lipid accumulation within hepatocytes in cats with hepatic lipidosis is similar to that found in humans with NAFLD (simple steatosis),21,23 elastography might be expected to be useful to identify cats with hepatic lipidosis.
Results of liver stiffness assessment for the healthy cats of the present study (mean Vs overall, 1.43 m/s) were within the range of those reported for healthy humans (mean ± SD, 1.197 ± 0.25 m/s).16 Moreover, these results were within the IQR reported for healthy adult dogs: 1.18 to 1.88 m/s at depths of 0 to 2 cm and 1.09 to 1.71 m/s at depths of 2 to 4 cm.24 These relationships in Vs among different species seem to be related to the unique histologic properties of the liver.25
To validate the feasibility of an elastography technique, it is recommended that sets of measurements include at least 10 valid measurements and have an IQR/median value of ≤ 30% and success rate of ≥ 60%.16 Unfortunately, several sets of measurements did not meet these criteria in the present study owing to movement or rapid breathing of the cat. Nevertheless, these reliability criteria for liver stiffness evaluation may not be relevant, given that a study17 showed that 10 valid measurements and a success rate of > 60% did not significantly influence the reliability of liver stiffness measurements in humans. Indeed, median and IQR/median values were independent predictors of fibrosis stage and were not significantly influenced by whether 10 valid measurements were obtained or by the success rate, with results considered very reliable (IQR/median value, ≤ 10%), reliable (> 10% but ≤ 30% or > 30% with a median liver stiffness of < 7.1 kPa), and poorly reliable (> 30% with a median liver stiffness ≥ 7.1 kPa).17 Per these criteria, 1 of the 35 (3%) sets of measurements of IQR/median values in the present study was considered very reliable because it was ≤ 10%, 24 (69%) sets were considered reliable, and the remaining 10 (29%) sets were considered reliable or poorly reliable. Because median liver stiffness was not measured in kilopascals, we were unable to distinguish between values deemed reliable and poorly reliable. These criteria for reliability should be used with caution because they are based on human data.
The mean Vs values for the 2 portions of the liver in the cats of the present study differed significantly. Several factors have been proposed to explain the influence of surrounding structures, including the diaphragm, stomach, and aorta, on measurements of the left liver lobe in humans, such as respiration, presence of food in the stomach, and pulsation of the aorta, respectively.26 Concern has also been expressed in the veterinary literature2 regarding the compression of the left liver lobe caused by a filled stomach. Moreover, findings in dogs indicate significantly higher Vs values for the right versus left side of the liver,25 and we observed a similar pattern in the cats of the present study. Although food was withheld from all cats in the present study, other factors may have influenced liver stiffness. Therefore, when elastography is used to evaluate or monitor patients, consecutive follow-up assessments should be conducted of the same liver portion for consistency.
Both the right ventral and left ventral portions of the liver were reliably imaged through intercostal windows during PSWE in the present study, contrary to the results of a previous study,2 which showed that only the left ventral and ventral midline portions of the feline liver could be reliably imaged by means of strain elastography. The liver is located within the rib cage in most cats, which makes the ventral midline portion difficult to image and requires that some degree of force be applied to the ultrasound probe at the subcostal region. On the other hand, in the authors’ experience, the right and left ventral portions of the liver are easier to image through the intercostal windows, with lesser variation in pushing pressure, because the ribs and thoracic wall exert limited pressure relative to that encountered when subxiphoidal or subcostal approaches are used. The present study was performed without anesthesia or sedation to mimic clinical situations in which cats are usually awake and to rule out the impact of any drug.24 Satisfactory results were nevertheless obtained through the end-expiratory period and during long expiration (during vocalization), resulting in inadequate measurements for only 6% (success rate of < 60%) and 29% (IQR/median value > 30%) of the 35 measurement sets.
The depth of PSWE measurements did not significantly affect the Vs in the cats of the present study, contrary to findings reported for dogs,24 wherein the mean Vs value decreases by 0.152 m/s for every 1-cm increase in depth. A depth of 1.1 to 2.0 cm was used for the cats of the present study, which is a narrower range in depth than used for dogs in the other study24 (0 to 4 cm). Therefore, this narrow depth range likely had no important effect on Vs measurements in our study.
The inclusion of a small number of cats of both sexes and various breeds could be considered a limitation of the study reported here. Although body weight was not significantly correlated with Vs (the cats were generally similar in body weight) and 12 of the 18 cats were castrated males, additional studies with larger sample sizes are needed to characterize the potential differences in Vs for specific body conformations (eg, body condition score), breeds, and sexes. Another limitation was that the included cats were presumed healthy on the basis of clinical signs, physical examination, clinicopathologic testing, radiography, and B-mode ultrasonography, and no fine-needle aspiration or biopsy was performed to cytologically or histologically confirm liver health owing to the guidelines of the institution's animal care and use committee. Therefore, studies that include cytologic and histologic assessment of healthy cats and cats with liver disease along with elastography should be considered for the future to determine the usefulness of PSWE for differentiating hepatic abnormalities in cats.
Overall, our findings suggested that quantitative PSWE of the liver was feasible in healthy adult cats. The data obtained could be helpful for interpretation of and comparison with PSWE findings in cats with hepatic disease. Additional research is needed to explore the potential usefulness of PSWE for diagnostic purposes.
Acknowledgments
Funded in part by Myoungin Meditech, which was not involved in the study design, data analysis and interpretation, or writing or publication of the manuscript.
The authors declare that there were no conflicts of interest.
Abbreviations
| IQR | Interquartile (25th to 75th percentile) range |
| NAFLD | Nonalcoholic fatty liver disease |
| PSWE | Point shear wave elastography |
| Vs | Shear wave velocity |
Footnotes
G-Power software, University of Trier, Trier, Germany.
ARIETTA 850 ultrasound machine, Hitachi, Tokyo, Japan.
SPSS Statistics for Windows, version 23.0, IBM Corp, Armonk, NY.
References
- 1. ↑
Sigrist RMS, Liau J, Kaffas AE, et al. Ultrasound elastography: review of techniques and clinical applications. Theranostics 2017;7:1303–1329.
- 2. ↑
White J, Gay J, Farnsworth R, et al. Ultrasound elastography of the liver, spleen, and kidneys in clinically normal cats. Vet Radiol Ultrasound 2014;55:428–434.
- 3. ↑
Carvalho CF, Chammas MC. Elastography—a new technology associated with ultrasonography. Clin Vet (Milano) 2013;18:62–70.
- 4. ↑
Chang JM, Won JK, Lee KB, et al. Comparison of shear-wave and strain ultrasound elastography in the differentiation of benign and malignant breast lesions. AJR Am J Roentgenol 2013;201:W347–W356.
- 5.
Ferraioli G, Tinelli C, Zicchetti M, et al. Reproducibility of real-time shear wave elastography in the evaluation of liver elasticity. Eur J Radiol 2012;81:3102–3106.
- 6.
Guzmán-Aroca F, Reus M, Berna-Serna JD, et al. Reproducibility of shear wave velocity measurements by acoustic radiation force impulse imaging of the liver. J Ultrasound Med 2011;30:975–979.
- 7. ↑
Yoneda M, Suzuki K, Kato S, et al. Nonalcoholic fatty liver disease: US-based acoustic radiation force impulse elastography. Radiology 2010;256:640–647.
- 8. ↑
Tamura M, Ohta H, Shimbo G, et al. Usefulness of noninvasive shear wave elastography for the assessment of hepatic fibrosis in dogs. J Vet Intern Med 2019;33:2067–2074.
- 9. ↑
Feliciano MAR, Maronezi MC, Pavan L, et al. ARFI elastography as a complementary diagnostic method for mammary neoplasia in female dogs—preliminary results. J Small Anim Pract 2014;55:504–508.
- 10. ↑
Feliciano MAR, Maronezi MC, Simoes APR, et al. Acoustic radiation force impulse (ARFI) elastography of testicular disorders in dogs: preliminary results. Arq Bras Med Vet Zootec 2016;68:283–291.
- 11.
Feliciano MAR, Maronezi MC, Crivellenti LZ, et al. Acoustic radiation force impulse (ARFI) elastography of the spleen in healthy adult cats – a preliminary study. J Small Anim Pract 2015;56:180–183.
- 12.
Garcia PHS, Feliciano MAR, Carvalho CF, et al. Acoustic radiation force impulse (ARFI) elastography of kidneys in healthy adult cats: preliminary results. J Small Anim Pract 2015;56:505–509.
- 13. ↑
Feliciano MAR, Maronezi MC, Brito MBS, et al. Doppler and elastography as complementary diagnostic methods for mammary neoplasm female cats. Arq Bras Med Vet Zootec 2015;67:935–939.
- 14.
Brito MBS, Feliciano MAR, Coutinho LN, et al. ARFI elastography of healthy adults felines testes. Acta Sci Vet 2015;43:1303.
- 15. ↑
Thanaboonnipat C, Sutayatram S, Buranakarl C, et al. Renal shear wave elastography and urinary procollagen type III amino-terminal propeptide (uPIIINP) in feline chronic kidney disease. BMC Vet Res 2019;15:54.
- 16. ↑
Madhok R, Tapasvi C, Prasad U, et al. Acoustic radiation force impulse imaging of the liver: measurement of the normal mean values of the shearing wave velocity in a healthy liver. J Clin Diagn Res 2013;7:39–42.
- 17. ↑
Boursier J, Zarski JP, de Ledinghen V, et al. Determination of reliability criteria for liver stiffness evaluation by transient elastography. Hepatology 2013;57:1182–1191.
- 18. ↑
Feeney DA, Anderson KL, Ziegler LE, et al. Statistical relevance of ultrasonographic criteria in the assessment of diffuse liver disease in dogs and cats. Am J Vet Res 2008;69:212–221.
- 19. ↑
Bai M, Du L, Gu J, et al. Virtual touch tissue quantification using acoustic radiation force impulse technology: initial clinical experience with solid breast masses. J Ultrasound Med 2012;31:289–294.
- 20. ↑
Mattoon JS, Nyland TG. Small animal diagnostic ultrasound. 3rd ed. St Louis: Saunders Elsevier, 2015;366.
- 21. ↑
Zawie DA, Garvey M. Feline hepatic disease. Vet Clin North Am Small Anim Pract 1984;14:1201–1230.
- 22. ↑
Dimski DS, Taboada J. Feline idiopathic hepatic lipidosis. Vet Clin North Am Small Anim Pract 1995;25:357–373.
- 23. ↑
Brunt EM, Tiniakos DG. Histopathology of nonalcoholic fatty liver disease. World J Gastroenterol 2010;16:5286–5296.
- 24. ↑
Holdsworth A, Bradley K, Birch S, et al. Elastography of the normal canine liver, spleen and kidneys. Vet Radiol Ultrasound 2014;55:620–627.
- 25. ↑
Tamura M, Ohta H, Nisa K, et al. Evaluation of liver and spleen stiffness of healthy dogs by use of two-dimensional shear wave elastography. Am J Vet Res 2019;80:378–384.
- 26. ↑
Toshima T, Shirabe K, Takeishi K, et al. New method for assessing liver fibrosis based on acoustic radiation force impulse: a special reference to the difference between right and left liver. J Gastroenterol 2011;46:705–711.
