Focal liver lesions are commonly identified during diagnostic evaluation of dogs with hepatic disease, during tumor staging in dogs with extrahepatic neoplasia, or as an incidental finding in dogs with problems not related to the liver. In all instances, characterization of these lesions presents a diagnostic challenge, given that the primary goal is to differentiate those that are benign from those that are malignant.1–3 Both identification and characterization of these lesions is vital to direct further diagnostic and therapeutic strategies. The ability to characterize focal hepatic lesions reduces the need for dogs to undergo biopsy procedures.4–6 For humans with extrahepatic primary tumors, the identification of metastatic disease facilitates staging and stratification for minimally invasive intervention, tumor resection, or medical management. If liver-specific MRI protocols could be developed that were equally effective in dogs, the result would be a major improvement in veterinary patient care.
Although commonly used, ultrasonography often fails to characterize or even detect many focal liver lesions.1,7 In humans, the evaluation of focal liver lesions by ultrasonography has been largely abandoned and replaced by MRI evaluation. Administration of standard extracellular gadolinium-based contrast agents followed by MRI has been shown to increase tumor detection, compared with results of MRI without prior contrast agent administration.7,8 However, extracellular contrast enhancement patterns are not specific to the type of hepatic lesion.3
Recently, MRI protocols involving liver-specific contrast agents have been developed to provide reliable noninvasive characterization of focal hepatic lesions in humans. Gadoxetic acid provides superior detection and classification of hepatic lesions in humans, compared with other hepatocyte-specific contrast agents (eg, gadobenate dimeglumine).9 Gadoxetic acid has the unique property of extracellular distribution in the initial phase of contrast enhancement, followed by hepatocyte-specific uptake in the delayed phase. Unlike other hepatocellular MRI contrast agents, gadoxetic acid can be administered as an IV bolus, allowing for angiography, and the hepatic phase occurs within a short time after injection. Thus, gadoxetic acid facilitates both identification and classification of focal lesions in the liver in a single MRI examination with use of a single contrast agent.10 In humans, specific patterns of enhancement are used to differentiate malignant from benign lesions.2,3,8,9
The purpose of the study reported here was to determine, by means of MRI, the time to maximal contrast enhancement in T1-weighted images following IV administration of gadoxetic acid in healthy dogs and assess the impact of gadoxetic acid on the signal intensity in T2-weighted images. The intent was to evaluate the contrast agent-related effects on T2-weighted images to develop a clinically applicable MRI protocol involving gadoxetic acid for use in dogs. If T2-weighted images could be obtained after contrast agent injection (during the interval between injection and maximal enhancement), then the overall duration of the MRI examination could be minimized. Because all of the veterinary patients are anesthetized for MRI examinations at our facility (and presumably so at other hospitals), we foresee that a reduction of the duration of the MRI procedure would decrease both the cost and the risk of morbidity associated with prolonged anesthesia.11 One of our hypotheses was that following the injection of gadoxetic acid in healthy dogs, the signal intensity of the liver parenchyma in T1-weighted images would be greater than that in precontrast images and that the pattern of signal intensity increase would be similar to that observed in humans, with maximum contrast enhancement occurring within 20 to 60 minutes after injection. Another hypothesis was that T2-weighted images would not be affected by gadoxetic acid injection and thus, signal intensity of T2-weighted images after contrast agent administration would not be significantly different from that in precontrast images.
Materials and Methods
DOGS
Seven healthy research Beagles were obtained from the University of Guelph Central Animal Facility for use in the study. The dogs’ weights ranged from 7.6 to 12.2 kg and their ages ranged from 1.5 to 4 years; 3 dogs were neutered males and 4 dogs were spayed females. The University Animal Care Committee approved this study according to Canadian Council on Animal Care guidelines. The dogs were determined to be healthy on the basis of results of a physical examination, CBC, serum biochemical analyses, urinalysis, and hepatic ultrasonography. Ultrasonography was performed with each dog in dorsal recumbency. The hair on the ventral aspect of the abdomen was clipped and acoustic coupling gel applied to the skin. The liver was examined with an 8-MHz curvilinear transducer. Dogs were excluded from study participation if any hepatic abnormalities were detected.
Each dog was anesthetized according to a protocol for a concurrent research project that included premedication with hydromorphone hydrochloride (0.05 mg/mL, IM; used only in 2 dogs) and induction of anesthesia with propofol (2 to 4 mg/kg, IV). Each dog was intubated, and anesthesia was maintained by inhalation of isoflurane and oxygen with an appropriate size ventilator (respiratory rate was maintained at 10 to 12 breaths/min and tidal volume was maintained at 10 to 20 mL/kg). For each dog, monitoring included assessment of heart rate, arterial blood pressure measurements, and capnography. Arterial blood pressure was monitored via a digital artery in a pelvic limb. Following the experimental procedure, each dog was allowed to recover from anesthesia and was returned to the housing facility. A blood sample (3 mL) was collected immediately and 24 hours following the experimental procedure for a CBC and serum biochemical analyses. Once-daily physical examinations were performed for 3 days following the experimental procedure. All dogs were adopted into family homes within 6 weeks after the study.
MRI PROTOCOL
Images were acquired with a 1.5T MRI unita with an 8-channel cardiac coil. Each dog was positioned in dorsal recumbency with pediatric respiratory bellows fitted around the caudal aspect of the thorax and adjusted to maintain good respiration triggering. Transverse T1-weighted (liver acquisition with volume acquisition) 3-D spoiled gradient echo, transverse T2-weighted fast spin echo, and a dorsal T2-weighted fast spin echo images (Appendix) of the liver were obtained 5 to 10 minutes prior to contrast agent administration (baseline data). A parallel imaging technique (sensitivity encoding) was used to reduce scan times and increase spatial resolution. Calibration sequences (array spatial sensitivity encoding technique) were acquired prior to imaging to remove aliasing caused by the reduced field of view that is required with the parallel imaging technique. For each dog, a total dose of 0.1 mL of gadoxetic acid/kg was administered IV (at 0 minutes) as a bolus at a rate of 1 mL/s followed by IV injection of 15 mL of saline (0.9%NaCl) solution at the same rate. Transverse T1-weighted (liver acquisition with volume acquisition) images were obtained before (baseline) and at 2, 5, 10, 15, 20, 25, 30, 35, 40, and 45 minutes after contrast agent injection. For 2 dogs, additional images were obtained at 50, 55, and 60 minutes after injection. Transverse T2-weighted (single shot fast spin echo) images were obtained before (baseline) and at 7, 12, 17, 22, 27, 32, 37, and 42 minutes after contrast agent injection. For 2 dogs, additional images were obtained at 47 and 62 minutes after injection. The T1-weighted images were acquired during a breath-hold to prevent motion artifacts from respiration. For T2-weighted sequences, pediatric bellows triggered the respiratory gating.
IMAGE ANALYSIS
Signal intensity of the liver tissue was measured before (baseline) and at each time point after contrast agent administration on the transverse T1-weighted and T2-weighted images. Investigators were instructed to select the representative slice that illustrated the branching of the hepatic veins. Regions of interest were elliptical (area, approx 1 cm2) and located to avoid major ducts and vessels. On each slice, 3 ROIs were placed in the liver parenchyma (in the dorsal aspect, peripherally in the left ventral portion, and laterally in the left portion of the liver; Figure 1). Two investigators (RCA and SGN) selected a single representative slice, and one investigator (AKB) selected 2 additional slices (1 cranial and 1 caudal to the representative slice) for placement of the ROIs. Mean ± SD signal intensity was recorded for each ROI. Background noise was measured with an ROI located external to the dog in the readout direction. To standardize signal measurements between dogs signal-to-noise ratio (SNR) was calculated on the basis of the signal intensity (SI) of the ROI and the SD of the background noise (SDN) for each dog as follows:
Representative transverse T1-weighted image of the liver of an anesthetized healthy dog that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. The dog was used in a study to investigate the time to maximal gadoxetic acid enhancement in T1-weighted images in 7 healthy dogs and assess the impact of gadoxetic acid on the signal intensity of T2-weighted images. The transverse T1-weighted image selected for analysis was obtained at the level of the branching of the hepatic veins. Three ROIs are placed in the liver parenchyma (in the dorsal aspect, peripherally in the left ventral portion, and laterally in the left portion of the liver [circles within tissue]) for measurement of hepatic signal intensity. One ROI is placed external to the dog in the readout direction (circle at top edge).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
Representative transverse T1-weighted image of the liver of an anesthetized healthy dog that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. The dog was used in a study to investigate the time to maximal gadoxetic acid enhancement in T1-weighted images in 7 healthy dogs and assess the impact of gadoxetic acid on the signal intensity of T2-weighted images. The transverse T1-weighted image selected for analysis was obtained at the level of the branching of the hepatic veins. Three ROIs are placed in the liver parenchyma (in the dorsal aspect, peripherally in the left ventral portion, and laterally in the left portion of the liver [circles within tissue]) for measurement of hepatic signal intensity. One ROI is placed external to the dog in the readout direction (circle at top edge).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
Representative transverse T1-weighted image of the liver of an anesthetized healthy dog that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. The dog was used in a study to investigate the time to maximal gadoxetic acid enhancement in T1-weighted images in 7 healthy dogs and assess the impact of gadoxetic acid on the signal intensity of T2-weighted images. The transverse T1-weighted image selected for analysis was obtained at the level of the branching of the hepatic veins. Three ROIs are placed in the liver parenchyma (in the dorsal aspect, peripherally in the left ventral portion, and laterally in the left portion of the liver [circles within tissue]) for measurement of hepatic signal intensity. One ROI is placed external to the dog in the readout direction (circle at top edge).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
To determine differences in signal intensity among observers, ROI locations, and slices locations, change in signal-to-noise ratio (ΔSNR) was calculated as follows:
where SNRpost is signal-to-noise ratio of the postcontrast images at each time and SNRpre is signal-to-noise ratio of the precontrast (baseline) images. The change in signal intensity was then used for statistical analysis and time-contrast enhancement curves.
STATISTICAL ANALYSIS
For each data set (dog and ROI), a time-contrast enhancement curve was generated and a best-fit model and adjusted correlation coefficientb were calculated. Data from the first 2 dogs acquired after 45 minutes were not used in the statistical analysis. Owing to individual variation in the shape and duration of the curve plateau, the time to maximal contrast enhancement was determined as the time at which the first derivative (slope of the curve) reached 1. For T1-weighted images, differences in change in signal intensity and time to peak enhancement were determined with an ANCOVA by use of combined data from all observers and all slices. For T2-weighted images, mean change in signal intensity was plotted and a slope estimated for each data set (by dog, slice, observer, and ROI). The slope was compared against a slope of 0 (ie, no change in signal intensity) with an ANOVA.c Significance was set at a value of P ≤ 0.05.
Results
DOGS
Results of physical examination, CBC, and serum biochemical analyses were within reference limits for all 7 dogs both before and 24 hours following completion of gadoxetic acid-enhanced MRI. No dog was excluded from study participation on the basis of detected hepatic abnormalities. No contrast agent– or injection-related adverse reactions were noted in any dog.
T1-WEIGHTED IMAGES
The T1-weighted image sequences were completed for all dogs (Figure 2). The T1-weighted contrast enhancement curve was similar for all dogs and ROIs. Rapid initial contrast enhancement was followed by a plateau and more gradual persistent increase in enhancement (Figure 3). The gradual persistent contrast enhancement resulted in no peak enhancement in any curve. Mean ± SD time to maximal enhancement after contrast agent injection was 10.5 ± 3.99 minutes, and was not significantly different among slices, locations, dogs, or observers. Absolute signal intensity varied among dogs but there was no significant difference in the change in signal-to-noise ratio from baseline (precontrast) among dogs, locations, observers, or slices. The change in signal-to-noise ratio was significantly (P < 0.02) different among locations, with the highest and lowest change in signal-to-noise ratio detected in the ventral and dorsal regions of the liver, respectively (Table 1).
Representative transverse T1-weighted 3-D gradient echo (liver acquisition with volume acquisition) images of the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Images at the level of the branching of the hepatic veins were obtained prior to (A) and at 10 (B), 20 (C), and 30 (D) minutes following contrast agent administration (at 0 minutes).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
Representative transverse T1-weighted 3-D gradient echo (liver acquisition with volume acquisition) images of the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Images at the level of the branching of the hepatic veins were obtained prior to (A) and at 10 (B), 20 (C), and 30 (D) minutes following contrast agent administration (at 0 minutes).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
Representative transverse T1-weighted 3-D gradient echo (liver acquisition with volume acquisition) images of the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Images at the level of the branching of the hepatic veins were obtained prior to (A) and at 10 (B), 20 (C), and 30 (D) minutes following contrast agent administration (at 0 minutes).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
A T1-weighted time enhancement curve for the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Transverse T1-weighted (liver acquisition with volume acquisition) images were obtained before (baseline) and at 2, 5, 10, 15, 20, 25, 30, 35, 40, and 45 minutes after contrast agent injection (at 0 minutes). Individual data points and best-fit curve are illustrated.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
A T1-weighted time enhancement curve for the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Transverse T1-weighted (liver acquisition with volume acquisition) images were obtained before (baseline) and at 2, 5, 10, 15, 20, 25, 30, 35, 40, and 45 minutes after contrast agent injection (at 0 minutes). Individual data points and best-fit curve are illustrated.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
A T1-weighted time enhancement curve for the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Transverse T1-weighted (liver acquisition with volume acquisition) images were obtained before (baseline) and at 2, 5, 10, 15, 20, 25, 30, 35, 40, and 45 minutes after contrast agent injection (at 0 minutes). Individual data points and best-fit curve are illustrated.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
Mean ± SD time to maximum hepatic signal-to-noise ratio and change in hepatic signal-to-noise ratio determined during MRI in 7 anesthetized healthy dogs following bolus IV injection of gadoxetic acid (0.1 mL/kg).
| ROI | Time to maximum change in signal-to-noise ratio (min) | Change in signal-to-noise ratio at time of maximum signal intensity |
|---|---|---|
| Lateral | 10.5 ± 3.99 | 265.09 ± 139.69 |
| Ventral | 15.55 ± 5.91 | 414.07 ± 218.20 |
| Dorsal | 11.14 ± 4.26 | 220.29 ± 111.08 |
Signal intensity of the liver tissue was measured at 3 ROIs before (baseline) and at 2, 5, 10, 15, 20, 25, 30, 35, 40, and 45 minutes after contrast agent administration on T1-weighted images. Means were calculated on the basis of values from all observers (3) and all dogs (7). The ROIs were placed in the liver parenchyma (in the dorsal aspect, peripherally in the left ventral portion, and laterally in the left portion of the liver). Signal-to-noise ratio (SNR) was calculated as SNR = SI/√SDN, where SI is the signal intensity and SDN is the SD of the background noise. Change in signal-to-noise ratio (ΔSNR) was calculated for each dog and each time (baseline and following contrast agent administration) as follows: ΔSNR = (SNRpost – SNRpre)/SNRpre, where SNRpost is signal-to-noise ratio of the postcontrast images at each time and SNRpre is signal-to-noise ratio of the precontrast (baseline) images.
T2-WEIGHTED IMAGES
The T2-weighted image sequences were completed for all dogs (Figure 4). The slope of the T2-weighted contrast enhancement curve was not significantly different from zero for any variable of interest (slice, region, observer, location, or dog), although all estimates of slope were slightly negative. Similar to the difference noted for the T1-weighted images following contrast agent injection, the absolute T2-weighted signal intensity was significantly (P = 0.02) different by location, with the highest and lowest change signal intensity detected in the ventral and dorsal regions of the liver of the dogs, respectively.
Representative transverse T2-weighted spin echo (single shot fast spin echo) images of the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Images at the level of the branching of the hepatic veins were obtained prior to (A) and 7 (B), 17 (C), and 32 (D) minutes following contrast agent administration (at 0 minutes).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
Representative transverse T2-weighted spin echo (single shot fast spin echo) images of the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Images at the level of the branching of the hepatic veins were obtained prior to (A) and 7 (B), 17 (C), and 32 (D) minutes following contrast agent administration (at 0 minutes).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
Representative transverse T2-weighted spin echo (single shot fast spin echo) images of the liver of 1 of 7 anesthetized healthy dogs that had been administered an IV injection of 0.1 mL of gadoxetic acid/kg. Images at the level of the branching of the hepatic veins were obtained prior to (A) and 7 (B), 17 (C), and 32 (D) minutes following contrast agent administration (at 0 minutes).
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.224
Discussion
Gadoxetic acid is a hepatocellular-specific contrast agent designed to differentiate normally functioning liver tissue from extrahepatic metastases. Hepatocyte-specific contrast agents are now routinely used in humans to identify and characterize small (< 1 cm in diameter) or indistinct hepatic lesions. Initial applications of gadoxetic acid in human studies12 focused on focal lesions to differentiate extrahepatic tumor metastases from focal nodular hyperplasia. Recently, differences in uptake patterns of gadoxetic acid have been used for identification and staging of hepatocellular carcinomas.9,13
Prior to hepatocellular uptake of gadoxetic acid, there is an extracellular phase similar to that associated with standard MRI contrast agents. Following the extracellular phase, gadoxetic acid also shows a delayed hepatocellular uptake phase. The uptake mechanism of gadoxetic acid, which has recently been elucidated in humans, involves anion transport proteins.14 Initial extracellular distribution delivers the gadoxetic acid to the sinusoidal space. In humans, the contrast agent is transported into the cell via organic anion transport polypeptides (OATP1B1 and OATBP1B3). Gadoxetic acid is then excreted unchanged into the bile canaliculi by multidrug-resistance protein 2 (MDRP2).15 Recent work has correlated expression of MDRP2 with histologic grade of hepatocellular carcinomas in humans.16
Through this anion transport protein mechanism, approximately 50% of gadoxetic acid is excreted via the biliary system,17 allowing for both a dynamic extracellular and delayed hepatic phase of imaging. Renal filtration is responsible for the remaining 50% of gadoxetic acid excretion. In humans, the ability to perform vascular and delayed hepatocyte-specific imaging with gadoxetic acid results in more efficient imaging protocols than those involving other liver-specific contrast agents that require a longer delay and cannot be injected as a bolus.9,18,19
In preclinical and early clinical phase trials in humans,20,21 peak contrast enhancement in T1-weighted images was reported to occur approximately 20 minutes after gadoxetic acid injection. Yonetomi et al22 described a hepatic imaging protocol involving gadoxetic acid for use in dogs, with which the hepatobiliary phase was acquired at 20 minutes following injection and allowed reliable lesion detection. Recently, protocols for hepatic imaging in humans have been revised because of the comparable lesion conspicuity at 10 to 15 minutes following gadoxetic acid injection.18,19 In the present study, the mean time to maximal liver contrast enhancement in T1-weighted sequences after gadoxetic acid injection in dogs was 10.5 ± 3.99 minutes. The contrast enhancement curves revealed an enhancement plateau rather than a distinct peak. This was in agreement with findings of another study23 in dogs, which indicated enhancement peaks between 1 and 10 minutes after contrast agent injection followed by a prolonged plateau period. These curve patterns are similar in appearance to the signal intensity and contrast enhancement curves detected in human studies.20,24 The signal intensity plateau was expected given that the mechanism of action of the hepatic uptake of the contrast agent requires transport into the hepatocyte and excretion into the bile canaliculi, allowing accumulation of contrast agent within the hepatic parenchyma. For development of an efficient clinical protocol, we chose to characterize the initial acceleration of contrast enhancement to the point at which the plateau occurred because increases in enhancement were minimal beyond this initial effect. This pattern was consistent throughout the duration of image acquisition for all dogs in the present study. This initial increase in enhancement was likely responsible for the shorter time to maximal enhancement detected in the present study, compared with initial peak enhancement reported for humans and for dogs in earlier preclinical trials.25
Variations in signal enhancement among ROIs identified in the present study were likely a function of the ROI location relative to the receiver coil. In all dogs, the signal-to-noise ratio in the ventral aspect of the liver was higher than that in the dorsal portion of the liver. Dogs were positioned in dorsal recumbency for image acquisition with the receiver coil placed over the ventrum. The signal intensity of each ROI reflected its position relative to the receiver coil, with the dorsal ROI being farthest from the coil, thereby producing the lowest signal intensity. Importantly, the time to maximal contrast enhancement was not different among ROIs, neither by slice location nor by location. This suggests that there was a homogeneous distribution of contrast uptake, although the small sample size (3 ROIs) and the location of all ROIs in the left portion of the liver may have biased this finding. Finally, we found no significant difference in time to maximal contrast enhancement among the observers, illustrating a measure of agreement among observers and the fact that that the time to enhancement was independent of observer.
Although relative enhancement was not different among individual dogs in the present study, there were significant differences in absolute contrast enhancement. Several factors, including anatomic or biochemical variation, may account for these differences. For example, in smaller dogs, greater contact and shorter distances between the liver and the receiver coil may contribute to differences in absolute signal intensity among dogs. Variations in vascular anatomy among individual dogs may also have resulted in variable distribution of contrast agent throughout the liver, which may have affected absolute signal intensity. Importantly, differences in uptake efficiency of membrane-bound organic anion transport polypeptides may also have varied among dogs. Although well described for humans, organic anion transport polypeptides in dogs have only recently been sequenced and described.26 Furthermore, with regard to uptake rates and substrates, canine organic anion transport polypeptides correlate well with human organic anion transport polypeptides.27 Variations in the distribution and structure of organic anion transport polypeptides among individual dogs in the present study may also have accounted for the variation in hepatic uptake. Regardless, following administration of gadoxetic acid, it is the relative enhancement of normal hepatocytes that allows for identification of nonenhancing lesions.
Results of the present study with regard to the effect of gadoxetic acid on T2-weighted images were similar to those reported for human studies.9 Although there was a tendency toward a slightly negative change, the slopes of the enhancement curves in T2-weighted images were not significantly different from zero for all variables of interest (ie, slices, ROI locations, and observers). Because the signal intensity before and after contrast agent injection was similar, this will allow for a more efficient protocol for veterinary patients by minimizing the duration of anesthesia. These results were similar to T2-weighted sequence data for humans, for whom hepatic signal intensity and lesion conspicuity are similar before and after gadolinium enhancement achieved by use of extracellular gadolinium-based contrast agents.28 However, Kim et al10 reported a slight, but significant loss of signal intensity in T2-weighted images following gadoxetic acid administration in humans. In the present study, a negative slope to the T2-weighted time enhancement curve was evident, albeit not significant. As with all gadolinium chelates, gadoxetic acid is a paramagnetic contrast agent that causes shortening of T1, T2, and T2* relaxation. The strongest effect is in T1 relaxation shortening, which results in an increased T1 signal intensity characteristic of all gadolinium contrast agents. However, in T2-weighted images obtained after contrast administration, there is significant signal loss (up to 40% in the liver) because of magnetization transfer and T2* shortening.29 The slight loss of T2 signal intensity in the dogs of the present study likely reflected an accumulation of gadoxetic acid in hepatocytes, thereby increasing the T2 shortening and resulting in signal loss. Such signal loss of the hepatic parenchyma results in increased conspicuity (greater contrast-to-noise ratio) of several types of malignant lesions10 and facilitates identification of pancreatic duct lesions in humans.30 Further evaluation is required to validate this observation in dogs with malignant lesions.
The data obtained in the present study have illustrated the efficacy of gadoxetic acid for hepatic MRI in healthy dogs. Administration of gadoxetic acid in healthy dogs resulted in an enhancement pattern that was similar to that described for humans, namely a rapid increase in signal intensity in the initial phase followed by a plateau. Maximal enhancement occurred between 10 and 15 minutes after contrast agent injection in the study dogs. In addition, the present study revealed that gadoxetic acid does not significantly affect signal intensity on T2- weighted images. Further evaluation of the use of gadoxetic acid during MRI in dogs with known hepatic lesions is required to determine the efficacy of the agent for identification of small or indistinct nodules and masses and for differentiation of benign from malignant lesions.
Acknowledgments
Supported by the Pet Trust Fund, Ontario Veterinary College, University of Guelph, ON, Canada.
Presented as a poster at the 2012 Annual Scientific Conference of the American College of Veterinary Radiology, Las Vegas, October 2012.
The authors thank Gabrielle Monteith for assistance with statistical analysis.
ABBREVIATION
| ROI | Region of interest |
Footnotes
1.5T GE Signa Excite II, version 11.1, General Electric, Milwaukee, Wis.
TableCurve 2D, 2012, Systat Software Inc, San Jose, Calif.
SAS/STAT, version 9.3, SAS Institute Inc, Cary, NC.
References
1. Liptak J, Stud F, Dernell W. Tumors in cats and dogs. Compend Contin Educ Pract Vet 2004; 26: 50–56.
2. Martin DR, Semelka RC. Magnetic resonance imaging of the liver: review of techniques and approach to common diseases. Semin Ultrasound CT MR 2005; 26: 116–131.
3. Irausquin RA, Scavelli TD, Corti L, et al. Comparative evaluation of the liver in dogs with a splenic mass by using ultrasonography and contrast-enhanced computed tomography. Can Vet J 2008; 49: 46–52.
4. Martí-Bonmatí L. MR imaging characteristics of hepatic tumors. Eur Radiol 1997; 7: 249–258.
5. Zech CJ, Herrmann KA, Reiser MF, et al. MR imaging in patients with suspected liver metastases: value of liver-specific contrast agent Gd-EOB-DTPA. Magn Reson Med Sci 2007; 6: 43–52.
6. Reimer P, Schneider GN, Schima W. Hepatobiliary contrast agents for contrast-enhanced MRI of the liver: properties, clinical development and applications. Eur Radiol 2004; 14: 559–578.
7. Clifford CA, Pretorius ES, Weisse C, et al. Magnetic resonance imaging of focal splenic and hepatic lesions in the dog. J Vet Intern Med 2004; 18: 330–338.
8. Semelka RC, Martin DR, Balci NC. Magnetic resonance imaging of the liver: how I do it. J Gastroenterol Hepatol 2006; 21: 632–637.
9. Frydrychowicz A, Lubner MG, Brown JJ, et al. Hepatobiliary MR imaging with gadolinium-based contrast agents. J Magn Reson Imaging 2012; 35: 492–511.
10. Kim YK, Kwak HS, Kim CS, et al. Detection and characterization of focal hepatic tumors: a comparison of T2-weighted MR images before and after the administration of gadoxectic acid. J Magn Reson Imaging 2009; 30: 437–443.
11. Clark-Price SC, Dossin O, Jones KR, et al. Comparison of three different methods to prevent heat loss in healthy dogs undergoing 90 minutes of general anesthesia. Vet Anaesth Analg 2013; 40: 280–284.
12. Marin D, Brancatelli G, Federle M, et al. Focal nodular hyperplasia: typical and atypical MRI findings with emphasis on the use of contrast media. Clin Radiol 2008; 63: 577–585.
13. Hammerstingl R, Zangos S, Schwarz W, et al. Contrast-enhanced MRI of focal liver tumors using a hepatobiliary MR contrast agent detection and differential diagnosis using Gd-EOB-DTPA–enhanced versus Gd-DTPA–enhanced MRI in the same patient. Acad Radiol 2002; 9 (suppl 1): S119–S120.
14. Leonhardt M, Keiser M, Oswald S, et al. Hepatic uptake of the magnetic resonance imaging contrast agent Gd-EOB-DTPA: role of human organic anion transporters. Drug Metab Dispos 2010; 38: 1024–1028.
15. Van Beers BE, Pastor CM, Hussain HK. Primovist, Eovist: what to expect? J Hepatol 2012; 57: 421–429.
16. Narita M, Hatano E, Arizono S, et al. Expression of OATP1B3 determines uptake of Gd-EOB-DTPA in hepatocellular carcinoma. J Gastroenterol 2009; 44: 793–798.
17. Purysko AS, Remer EM, Veniero JC. Focal liver lesion detection and characterization with Gd-EOB-DTPA. Clin Radiol 2011; 66: 673–684.
18. Fidler J, Hough D. Hepatocyte-specific magnetic resonance imaging contrast agents. Hepatology 2011; 53: 678–682.
19. van Kessel CS, Veldhuis WB, van den Bosch MA, et al. MR liver imaging with Gd-EOB-DTPA: a delay time of 10 minutes is sufficient for lesion characterisation. Eur Radiol 2012; 22: 2153–2160.
20. Reimer P, Rummeny EJ, Shamsi K, et al. Phase II clinical evaluation of Gd-EOB-DTPA: dose, safety aspects, and pulse sequence. Radiology 1996; 199: 177–183.
21. Hamm B, Staks T, Mühler A, et al. Phase I clinical evaluation of Gd-EOB-DTPA as a hepatobiliary MR contrast agent: safety, pharmacokinetics, and MR imaging. Radiology 1995; 195: 785–792.
22. Yonetomi D, Kadosawa T, Miyoshi K, et al. Contrast agent Gd-EOB-DTPA (EOB•Primovist) for low-field magnetic resonance imaging of canine focal liver lesions. Vet Radiol Ultrasound 2012; 53: 371–380.
23. Marks A, Hecht S, Stokes J, et al. Effects of gadoxetic disodium (Eovist) contrast on magnetic resonance characteristics of the liver in clinically healthy dogs. Vet Radiol Ultrasound 2013; 55: 286–291.
24. Small WC, Nelson RC, Sherbourne GM, et al. Enhancement effects of a hepatocyte receptor-specific MR contrast agent in an animal model. J Magn Reson Imaging 1994; 4: 325–330.
25. Benness G, Khangure M, Morris I, et al. Hepatic kinetics and magnetic resonance imaging of gadolinium ethoxybenzyl diethylenetriaminepentacetic acid (Gd-EOB-DTPA) in dogs. Australas Radiol 1993; 37: 252–255.
26. Gui C, Hagenbuch B. Cloning/characterization of the canine organic anion transporting polypeptide 1b4 (Oatp1b4) and classification of the canine OATP/SLCO members. Comp Biochem Physiol C Toxicol Pharmacol 2010; 151: 393–399.
27. Wilby AJ, Maeda K, Courtney PF, et al. Hepatic uptake in the dog: comparison of uptake in hepatocytes and human embryonic kidney cells expressing dog organic anion-transporting polypeptide 1B4. Drug Metab Dispos 2011; 39: 2361–2369.
28. Chang SD, Thoeni RF. Effect of T1 shortening on T2-weighted MRI sequences: comparison of hepatic mass conspicuity on images acquired before and after gadolinium enhancement. AJR Am J Roentgenol 2008; 190: 1318–1323.
29. McRobbie DW, Moore EA, Groves MJ, et al. Measuring relaxation times in vivo. In: MRI: from picture to proton. 2nd ed. New York: Cambridge University Press, 2007; 208–214.
30. Kanematsu M, Aatsuo M, Shiratori Y, et al. Thick-section half-Fourier rapid acquisition with relaxation enhancement MR cholangiopancreatography: effects of I.V. administration of gadolinium chelate. AJR Am J Roentgenol 2002; 178: 755–761.
Appendix
Settings used for MRI examinations of 7 anesthetized healthy dogs that were administered an IV injection of 0.1 mL of gadoxetic acid/kg to investigate the time to maximal gadoxetic acid enhancement in T1-weighted images and assess the impact of gadoxetic acid on the signal intensity of T2-weighted images.
| MRI sequence | ||
|---|---|---|
| Variable | T1 | T2 |
| Matrix | 225 × 192 | 225 × 192 |
| Slice (mm) | 3.4 | 5 |
| Gap (mm) | 1.7 | 0 |
| No. of excitations | 0.75 | 2 |
| Echo time (ms) | 3.356 | 91 |
| Repetition time (ms) | 1.564 | 5,000 |
| Flip angle (°) | 12 | — |
| Echo train length | — | 11 |
— = Not applicable.
