Radiographic changes in hepatic size, shape, location, and opacity are used to assess the liver for possible abnormalities. A normal liver size does not preclude the presence of liver disease, but liver size is useful for screening and making a differential diagnosis for liver diseases. Hepatomegaly may represent hepatic congestion, steroid hepatopathy, inflammatory and infiltrative disease, and primary and metastatic neoplasia. Microhepatica may represent congenital portosystemic shunts and chronic inflammation such as hepatic cirrhosis.1–3
Liver size assessed radiographically, particularly for the right lateral view, has the highest correlation with actual liver weight in dogs.4 Radiographs of the liver are often obtained at the end of expiration to maximize abdominal organ separation and decrease motion artifact. The radiographic view is obtained from just cranial to the diaphragm to a few centimeters caudal to the hip joints.1,4 Liver size can be determined on the basis of the gastric axis, which ranges from perpendicular to the vertebral column to parallel to the corresponding rib in most breeds of dog.5 Liver size can be assessed by determining the ratio of the distance between the ventral border of the caudal vena cava where the caudal vena cava meets the diaphragm and caudal margin of the liver to the length of the vertebral body of T11, with a mean ± SD reference interval of 5.5 ± 0.8.6–8 In addition, there should be an acute angle near the costal arch of the left lateral lobe of the liver.2 In generalized hepatomegaly, the caudoventral liver margin becomes rounded or blunted and extends beyond the costal arch, and the gastric axis can be displaced caudally.9,10 Marked microhepatica results in cranial displacement of the stomach and a decrease in the distance between the caudal vena cava and hepatic border.1
Patient or technique-related nonpathological conditions can affect radiographic liver size. Studies1,4,11 on age, body size and body weight, breed, and body conformation have revealed that older, obese, shallow-chested, brachycephalic, and chondrodystrophoid dogs have caudal extension of the liver. Neonatal and young dogs have a larger liver relative to body size than do older dogs, which creates the appearance of hepatomegaly without a true hepatic abnormality. The respiratory phase when radiographs are obtained also can influence the location of the caudoventral margin of the liver.1,2,4,8,12 Overexpansion of the thorax or deep inspiration causes caudal displacement of the diaphragm and liver margins. It has been proposed that the position of an animal during radiography can influence liver size calculations, but there is controversy about the lateral view that makes the liver appear larger.1,2 Radiographic liver size can also be affected by gastric distention. A distended stomach places pressure on the liver, which causes the liver's ventral aspect to move caudally beyond the costal arch, or it may mask liver volume to make the liver appear small.4,5 These nonpathological factors can affect the radiographic assessment of liver size and should be considered; however, there is controversy about the extent of the impact that each factor has on liver measurements.
Therefore, the study reported here was conducted to investigate the influence of technical and patient factors (eg, beam center location, body position, respiratory phase, and gastric distention) on assessment of liver size in dogs. In addition, CT and fluoroscopy were also performed to evaluate the effect of body position on the liver and adjacent organs, respiratory effects on the liver, and thoracic width and height.
Materials and Methods
Animals
Twelve purpose-bred Beagles (11 males and 1 female) were used in the study. Dogs were 3 to 5 years old, and body weight ranged from 8 to 14 kg. All dogs were healthy as determined on the basis of results of a physical examination, CBC, serum biochemical analysis, electrolyte analysis, and thoracic and abdominal radiography. Ultrasonography was used to determine that the liver parenchyma had a normal homogeneous hypoechoic echotexture in all dogs. The study protocol was approved by the Institutional Animal Care and Use Committee of Chonnam National University (CNU IACUC-YB-2017-36).
Radiographic assessment of liver size
The influence of beam center location, body position, respiratory phase, and gastric distention on radiographic assessment of liver size was evaluated. Food was withheld from all dogs for at least 12 hours, and radiographic examination was performed with a digital radiographic system.a Dogs were not sedated or anesthetized for radiography. Radiographs obtained included the right lateral view of the thorax, right and left lateral views of the abdomen, and ventrodorsal view of the abdomen; all views were acquired at maximal inspiration and maximal expiration (total of 8 radiographs). A feeding tube was orally inserted, and gastric distention was then induced by injecting 200 to 250 mL of air. The same radiographic protocol was repeated, and 8 additional radiographs were obtained.
Radiographs were separately reviewed in arbitrary order by 2 authors (AC and SP). Radiographs for which the caudoventral liver margin could not be clearly identified were excluded from the study. All radiographs were sent to a workstation, and measurements were made on a DICOM fileb by use of electronic calipers. Liver length and length of T11 were measured on lateral radiographic views. Liver length was measured as the length of the axis from the ventral border of the caudal vena cava at the point where the caudal vena cava met the diaphragm to the caudal border of the liver. Length of T11 was measured at the level of the midpoint parallel to the long axis of the vertebral body. Ratio of liver length to length of T11 was calculated for each dog.
CT assessment of liver size
A CT examination was performed to evaluate the effect of body position on the liver and adjacent organs in 6 dogs. Anesthesia was induced by IM injection of a combination of zolazepam-tiletaminec and medetomidined and maintained with isoflurane.e Anesthetized dogs were positioned in sternal recumbency, and CT images were obtained from the diaphragm to the pelvis. Dogs were then positioned in right and left lateral recumbency (arbitrary order), and iodine contrast mediumf was administered (880 mg of I/kg, IV). The CT scans were conducted by use of a 16-channel multidetector CT scannerg with the following settings: kV (peak), 130; mA, 100; slice thickness, 1 mm; and pitch, 0.8. Multiplanar reconstruction and maximum-intensity projection of the contrast-enhanced CT images were performed. Quantitative analysis of liver size involved measurement of liver length and diameter of the aorta on a sagittal image (Figure 1). Liver length was measured with the electronic calipers from the ventral border of the caudal vena cava where the caudal vena cava met the diaphragm to the apex of the caudal border of the liver; all sagittal CT images were examined to find each point. Diameter of the aorta was measured at the level of the celiac aortic branch. In each dog, ratio of liver length to aorta diameter was calculated. In addition, the caudal border of the liver on the sagittal CT images was compared to the level of the adjacent vertebrae to evaluate liver location, with the vertebral body divided into 3 parts (cranial, middle, and caudal). Changes of the spleen and stomach on the basis of body position were evaluated by use of the dorsal plane contrast-enhanced CT images. The proximal points of the spleen, gastric fundus, and pylorus were compared to the level of the adjacent ribs, with the intercostal space divided into 3 parts (cranial, middle, and caudal). The level of each point was determined by examining all CT images. In addition, the spleen was recorded as a reverse-L shape or C shape in the reconstructed dorsal plane images.
Sagittal CT images of a dog in right lateral recumbency (A) and left lateral recumbency (B) and reconstructed dorsal plane CT images for right lateral recumbency (C) and left lateral recumbency (D). For the sagittal CT images, liver length was measured from the ventral border of the caudal vena cava to the apex of the caudal border of the liver after evaluating all sagittal CT images (black dashed line). Diameter of the aorta was measured at the level of the celiac aortic branch (black line). The caudal border of the liver was identified at the level of the cranial border of L2 (white dashed line). For the reconstructed dorsal plane CT images, the proximal point of the gastric fundus was identified at the level of the ninth rib (white line).
Citation: American Journal of Veterinary Research 79, 11; 10.2460/ajvr.79.11.1133
Fluoroscopic assessment of liver size
Fluoroscopy was performed to assess the influence of respiratory phase on liver size. Dynamic fluoroscopic evaluation of the liver was performed by use of a C-armh with settings of 58 to 60 kV (peak) and 10 mA. Fluoroscopic images were obtained by setting the beam center at the level of the diaphragm. Dogs were not sedated or anesthetized for fluoroscopy. Each dog was positioned in right lateral, left lateral, and dorsal recumbency, and fluoroscopic images of the cranial aspect of the abdomen were recorded (32 frames/s) by use of integrated recording softwarei during 3 respiratory cycles for each body position. Recordings were transferred to a medical image–management software program.b Maximal inspiratory and expiratory points were selected on the basis of the most cranial and caudal locations of the diaphragm. Liver length and length of T11 were measured at the maximal inspiratory and maximal expiratory points on the right lateral view; measurements were as described for the radiographic assessment. Thoracic height was measured as the length of the line extending perpendicularly from the xiphoid process to the line between the cranioventral border of T8 and caudoventral border of T13 on the right lateral and left lateral views. Thoracic width was measured by drawing a line between the costodiaphragmatic recesses on the ventrodorsal view (Figure 2).
Fluoroscopic images of a dog in left lateral recumbency (A) and dorsal recumbency (B) that were used to measure thoracic height and width. A—To measure thoracic height, a line was drawn from the cranioventral margin of T8 to the caudoventral margin of T13 (black line). Then a second line (black line with arrowheads) that was perpendicular to the first line and extended to the xiphoid process was measured. B—Thoracic width was measured by drawing a line between the costodiaphragmatic recesses (white line with arrowheads).
Citation: American Journal of Veterinary Research 79, 11; 10.2460/ajvr.79.11.1133
Statistical analysis
Data were reported as mean ± SD. Statistical analysis was performed with the values measured on the right lateral radiographic view, which were obtained by setting the beam center at the 13th rib at maximal expiration, as the criterion-referenced standard. The Mann-Whitney U test for paired data was used to determine significant differences between liver length and the ratio of liver length to length of T11 obtained for each view and for each respiratory phase, body position, beam center location, and degree of gastric distention. Reproducibility of assessments between the 2 investigators was analyzed by calculating the intraclass correlation coefficient. All statistical analyses were performed with standard software.j Values of P < 0.05 were considered significant.
Results
Liver length and the liver length-to-T11 length ratio on the basis of body position, respiratory phase, beam center location, and gastric distention status were summarized (Table 1). Liver size was evaluated by use of the radiograph obtained with a dog in right lateral recumbency at maximal expiration with the beam center at the 13th rib, which was considered the standard. In all dogs, liver length was significantly greater on radiographs obtained at maximal expiration than at maximal inspiration. However, the liver length-to-T11 length ratio did not differ significantly between expiration and inspiration.
Mean ± SD values for radiographic assessment of liver size in 12 healthy Beagles on the basis of body position, respiratory phase, beam center location, and gastric distention.
| Food withheld | Gastric distention | ||||
|---|---|---|---|---|---|
| Variable | RL Exp at 13th rib (n = 12) | RL Insp at 13th rib (n = 12) | RL Exp at scapula (n = 12) | LL Exp at 13th rib (n = 12) | RL Exp at 13th rib (n = 11) |
| Liver length (mm) | 103.77 ± 7.73a | 98.18 ± 8.83b | 103.38 ± 8.48 | 103.23 ± 10.30 | 105.92 ± 9.22 |
| Ratio of liver length to T11 length | 5.98 ± 0.40 | 5.66 ± 0.47 | 5.97 ± 0.48 | 5.95 ± 0.49 | 5.99 ± 0.47 |
Within a row, values with different superscript letters differ significantly (P < 0.05).
Exp = Maximal expiration. Insp = Maximal inspiration. LL = Left lateral recumbency. RL = Right lateral recumbency. 13th rib = Beam center located at the 13th rib. Scapula = Beam center located at the caudal border of the scapula.
The liver length and liver length-to-T11 length ratio measured with dogs in left lateral recumbency were not significantly different from those measured with dogs in right lateral recumbency. There was no significant effect of beam center location on liver length and the liver length-to-T11 length ratio between the caudal border of the scapula and the 13th rib locations. Liver length was compared after food was withheld and after gastric distention to determine the effect of gastric distention on liver size. The caudal border of the liver was not clearly delineated after gastric distention in 1 dog because it overlapped with the silhouette of the pylorus. Therefore, data for that dog were excluded from further analysis. For the remaining 11 dogs, liver size typically was greater, but not significantly different, after gastric distention than after food was withheld. For the radiographic assessment, only respiratory phase had a significant effect on liver size. The interclass correlation coefficient of liver length and liver length-to-T11 length ratio was 0.935 and 0.863, respectively.
Effects of body position on liver size were investigated by use of CT to evaluate adjacent organs. Mean ± SD liver length and liver length-to-aorta diameter ratio were 113.82 ± 14.73 mm and 11.89 ± 0.83 respectively, for right lateral recumbency and 111.89 ± 13.65 mm and 11.69 ± 0.58, respectively, for left lateral recumbency. No significant differences were found between right and left lateral recumbency. In the reconstructed dorsal plane of the contrast-enhanced CT images obtained after placing dogs in right and left lateral recumbency, the spleen had a reverse-L shape in 3 dogs and a C shape in the other 3 dogs (Figure 3). For all dogs, shape of the spleen did not change after changing body position. The caudal border of the liver was compared with adjacent vertebrae, and locations of the proximal point of the spleen, gastric fundus, and pylorus were compared with adjacent ribs (Supplementary Table S1, available at http:/avmajournals.avma.org/doi/suppl/10.2460/ajvr.79.11.1133). Locations did not change markedly with changes in body position. For both left and right lateral recumbency, the liver was typically located adjacent to the cranial or middle third of L1 or L2, the proximal point of the spleen was typically located adjacent to the cranial or middle third of the 10th to 12th intercostal space, the proximal point of the gastric fundus was typically located adjacent to the 8th to 10th intercostal space, and the proximal point of the pylorus was typically located adjacent to the 9th to 11th intercostal space. Location and space occupation of the spleen and stomach also did not change on the basis of body position and did not affect liver size.
Schematic depiction of a spleen with a reverse-L shape (A) or C shape (B) on reconstructed dorsal plane CT images. Shape of the spleen was evaluated by examination of all dorsal plane images.
Citation: American Journal of Veterinary Research 79, 11; 10.2460/ajvr.79.11.1133
Effect of respiratory phase on liver size was investigated with fluoroscopy (Table 2). Liver length was greater at maximal expiration than at maximal inspiration. The diaphragm, thoracic width, and thoracic height changed on the basis of respiratory phase. At maximal inspiration, the diaphragm was flattened and displaced caudally, which caused the caudal portion of the liver to be displaced from the costal arch. In contrast, at maximal expiration, the diaphragm was displaced cranially, which caused the liver to lie more within the costal arch and made the liver appear longer as the thoracic width became smaller (Figure 4). Mean ± SD thoracic width at maximal expiration (195.05 ± 16.99 mm) was significantly less than at maximal inspiration (213.51 ± 19.08 mm); however, respiratory phase did not have a significant effect on thoracic height.
Fluoroscopic images obtained at maximal expiration (A and C) and maximal inspiration (B and D) for a dog positioned in dorsal recumbency (A and B) and left lateral recumbency (C and D). Notice that the liver appears to be longer as the thoracic width narrows. In the lateral views, length of the liver tip appears to be greater at maximal inspiration because of caudal displacement of the diaphragm; however, liver length measured from the caudal vena cava to the caudal border of the liver was greater at maximal expiration. Dotted lines represent the caudal border of the liver.
Citation: American Journal of Veterinary Research 79, 11; 10.2460/ajvr.79.11.1133
Mean ± SD values for fluoroscopic measurements of liver length and thoracic height at maximal inspiration and maximal expiration for the same dogs in Table 1.
| Maximal inspiration | Maximal expiration | |||
|---|---|---|---|---|
| Variable | RL | LL | RL | LL |
| Liver length (mm) | 130.41 ± 11.46 | 122.90 ± 4.18 | 135.40 ± 11.71 | 126.49 ± 5.44 |
| Ratio of liver length to T11 length | 5.73 ± 0.78 | 5.59 ± 0.46 | 5.94 ± 0.17 | 5.54 ± 0.39 |
| Thoracic height (mm) | 187.68 ± 5.15 | 187.57 ± 5.40 | 185.85 ± 5.01 | 186.22 ± 4.90 |
See Table 1 for key.
Discussion
In the study reported here, technical and patient factors (eg, respiratory phase, body position, beam center location, and gastric distention) that could affect radiographic assessment of liver size were comprehensively evaluated. The effect of body position on adjacent organs, such as the stomach and spleen, and on liver size were further assessed with CT. Changes in the thoracic cavity during the respiratory cycle were evaluated with fluoroscopy. Among the factors evaluated, only the respiratory phase significantly affected liver length, which was greater at maximal expiration than at maximal inspiration for both radiography and fluoroscopy. However, the liver length-to-T11 length ratio did not change significantly during the respiratory cycle. Thoracic width was significantly narrower at maximal expiration; this change in the thoracic cavity could have induced caudal elongation of the liver, which could have led to greater liver length than that measured at maximal inspiration.
The liver moves into the thoracic portion of images during expiration and moves caudally during inspiration because it is anchored to the diaphragm by ligaments, which causes it to move in unison with the diaphragm. Liver size is reportedly larger at maximal inspiration than at maximal expiration.1,2,4,8,12 In a study8 that involved use of a liver tip, (the liver lobe protruding at the level of the 12th rib) to determine liver length, the liver tip was longer during inspiration than during expiration because of the caudal movement of the diaphragm. Investigators in that study8 assessed the change in length of the part of the liver that protruded beyond the rib cage; therefore, displacement of the liver related to the caudal motion of the diaphragm during respiration mainly affected liver length (as measured by use of the liver tip).
The caudal border of the liver of dogs may extend caudal to the ventral portion of the costal arch, or it may be completely contained within the rib cage.13 The liver of deep-chested dogs lies more completely within the costal arch, whereas greater caudal extension of the liver is visible in dogs with a shallow, wide thoracic conformation.3 In the study reported here, the effect of respiration on liver size was investigated with fluoroscopy because liver length is not influenced by thoracic conformation.6,11 Moreover, because liver length was measured from the diaphragmatic level to the caudal border, cranial and caudal movements of the liver (as influenced by diaphragmatic motion during the respiratory cycle) were considered to be negligible. However, changes in the thoracic width were considered to be a major factor for greater liver length at maximal expiration because the formula liver length × thoracic depth × thoracic width is significantly correlated with liver volume.6 In other words, liver volume was constant in the dogs of the present study, and there was not a significant change in thoracic depth; therefore, liver length appeared to be larger as the thoracic width became narrower during expiration. Thus, changes in the size of the thoracic cavity can be related to the perceived greater liver length in the expiratory phase, compared with the size during the inspiratory phase, in the same dog.
Body position of the dogs was a nonpathological factor that could have affected radiographic liver size. Radiographic liver length measured for dogs in right lateral recumbency is reportedly larger than liver length measured for dogs in left lateral recumbency.14,15 Relative movements of the diaphragmatic crura and shifting of the left lateral lobe of the liver were proposed as underlying causes for this result.14 In contrast, caudal movement of the left lobe of the liver was considered to be the cause of the greater liver size measurements for dogs in right lateral recumbency, compared with results for dogs in left lateral recumbency.15 However, liver length measured by use of radiography, CT, and fluoroscopy in the present study did not differ significantly between right and left lateral recumbency, which is compatible with results of other studies.8,12 Previous theories about a greater liver length for dogs in right lateral recumbency were evaluated by use of CT. Location of the liver, spleen, and stomach did not change on the basis of body position of the dogs when a side-by-side comparison of the CT images of the reformatted dorsal plane for right and left lateral recumbency was performed.
Beam center is one of the more important technical factors for radiography because the target object can be distorted if it is out of the beam center. However, beam center location did not cause significant changes to liver length, regardless of whether the beam was centered at the caudal border of the scapula or at the 13th rib. Similarly, investigators of another study8 also did not find a significant difference in length of a liver lobe when the beam center location was at the middle of the 13th rib and iliac crest or at the middle of the 10th rib and iliac crest. Considering that thoracic radiographs are acquired during maximal inspiration and abdominal radiographs are acquired at maximal expiration, liver size on abdominal radiographs cannot be compared with liver size on thoracic radiographs, even under ideal imaging conditions.
The effect of gastric distention on liver length was investigated in the study reported here. It has been reported4,5 that the liver may appear larger because a distended stomach places pressure on the liver; conversely, the liver may appear smaller because the stomach may mask liver volume. In the present study, gastric contents obscured the caudal border of the liver in 1 dog, which made it difficult to interpret liver size. In addition, liver length after gastric distention typically was larger, although not significantly different, from liver length after food was withheld.
For the study reported here, influences of various technical factors (which included respiratory phase, beam center location, body position, and gastric distention) on liver size were evaluated, and only the respiratory phase had a significant effect. Liver length measured at maximal expiration was greater than that measured at maximal inspiration as a result of narrowing of the thoracic width. However, the ratio of liver length to T11 length was not significantly affected by any technical factors, which confirmed that the ratio can be used as a reliable variable for the evaluation of liver size. Effect of the respiratory phase on liver size and thoracic width was confirmed with fluoroscopy. In addition, CT to determine the anatomic relationship between the liver and adjacent stomach and spleen based on body position revealed minimal changes in size and location. The study reported here provided a comprehensive elucidation of technical factors that could affect liver size, and the results can be useful for interpretation of liver size by practitioners.
Acknowledgments
Supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (NRF-2018R1A2B6006775).
Footnotes
EVA-HF525, Gemss-Medical, Seongnam-si, Republic of Korea.
Infinitt PACS, Infinitt Healthcare, Seoul, Republic of Korea.
Zoletil, Virbac, Carros Cedex, France.
Domitor, Orion Corp, Espoo, Finland.
Terrell, Piramal Critical Care, Bethlehem, Pa.
Omnipaque 300, GE Healthcare, Oslo, Norway.
Siemens Emotion 16, Siemens, Forchheim, Germany.
KMC-950, Gemss-Medical, Seongnam-si, Republic of Korea.
CXView, version 3910, Gemss-Medical, Seoul, Republic of Korea.
SPSS Statistics, version 21, IBM Corp, Armonk, NY.
References
1. Larson MM. The liver and spleen. In: Thrall DE, ed. Textbook of veterinary diagnostic radiology. 6th ed. St Louis: Elsevier, 2013;679–684.
2. Dennis R, Kirberger RM, Barr F, et al. Other abdominal structures: abdominal wall, peritoneal and retroperitoneal cavities, parenchymal organs. In: Dennis R, Kirberger RM, Barr F, eds. Handbook of small animal radiology and ultrasound. 2nd ed. St Louis: Elsevier, 2010;239–241.
3. Wrigley RH. Radiographic and ultrasonographic diagnosis of liver disease in dogs and cats. Vet Clin North Am Small Anim Pract 1985;15:21–38.
4. Partington BP, Biller DS. Hepatic imaging with radiology and ultrasound. Vet Clin North Am Small Anim Pract 1995;25:305–335.
5. Thrall DE, Robertson ID. The abdomen. In: Thrall DE, Robertson ID, eds. Atlas of normal radiographic anatomy and anatomic variants in the dog and cat. 2nd ed. St Louis: Elsevier, 2016;245–248.
6. van Bree H, Jacobs V, Vandekerckhove P. Radiographic assessment of liver volume in dogs. Am J Vet Res 1989;50:1613–1615.
7. Lee S-K, Park S, Cheon B, et al. Body weight, blood pressure, and systemic changes following low-dosage prednisolone administration in dogs. Am J Vet Res 2017;78:1091–1097.
8. van Bree H, Sackx A. Evaluation of radiographic liver size in twenty-seven normal deep-chested dogs. J Small Anim Pract 1987;28:693–703.
9. Evans SM. The radiographic appearance of primary liver neoplasia in dogs. Vet Radiol 1987;28:192–196.
10. Liptak JM, Dernell WS, Withrow SJ. Liver tumors in cats and dogs. Compend Contin Educ Pract Vet 2004;26:50–57.
11. Choi J, Keh S, Kim H, et al. Radiographic liver size in Pekingese dogs versus other dog breeds. Vet Radiol Ultrasound 2013;54:103–106.
12. Cockett PA. Radiographic anatomy of the canine liver: simple measurements determined from the lateral radiograph. J Small Anim Pract 1986;27:577–589.
13. Evans HE, de LaHunta A. The digestive apparatus and abdomen. In: Evans HE, de LaHunta A, eds. Miller's anatomy of the dog. 6th ed. St Louis: Elsevier, 2013;327–331.
14. Grandage J. The radiology of the dog's diaphragm. J Small Anim Pract 1974;15:1–18.
15. Kealy JK. The abdomen. In: Evans HE, de LaHunta A, eds. Diagnostic radiography of the dog and the cat. Philadelphia: WB Saunders, 1979;19–24.
