Materials and Methods

Fourteen healthy Beagles, 1 to 4 years of age and weighing 5.7 to 11.5 kg, were scanned. All dogs were determined to be clinically normal on the basis of results of physical examination, serum biochemical profiles, and urinalysis. All results were within reference ranges. The procedures were approved by the Institute of Laboratory Animal Resources at the Seoul National University. Anesthesia was induced with atropine sulfate^a (0.5 mg/kg, SC), diazepam^b (0.2 mg/kg, IV), and thiopental sodium^c (11 mg/kg, IV) and maintained with isoflurane.^d Respiratory motion artifacts were minimized by inducing hyperventilation and breath holding. Images were acquired with a single-channel helical scanner.^e There were 2 imaging protocols, a sequential CT scan and a triphasic CT scan. The image acquisition parameters included a helical and single-slice dynamic acquisition, a matrix of 512 × 512, a 25-cm display field of view, and a small scan field of view.

The sequential CT scan was divided into an unenhanced helical scan and a cine scan. The unenhanced helical scan was performed with a thickness of 3 mm, interval of 3 mm, pitch of 2, voltage of 120 kV (peak), and amperage of 80 to 100 mA. Images were acquired from the cranial aspect of T13 to the caudal aspect of the L4 so as to include both kidneys. The cine scan was performed with a thickness of 2 mm, voltage of 120 kV (peak), amperage of 40 to 60 mA, and scan speed of 2.0 s/rotation at the right renal pelvic region. Seventy serial images of the pelvic region of the right kidney were acquired over 148 seconds. Iodinated contrast medium^f (300 mg of I/mL) was injected at a rate of 3 mL/s, at a dose of 3 mL/kg by use of a power injector.^g The scan was begun at the onset of contrast medium injection. Time-attenuation curves were obtained from the cine scan images with an imaging workstation and used to quantify enhancement time of the CMP and NP and evaluate renal parenchymal enhancement. Time zero was considered the time of onset of contrast medium injection.

An ROI was hand drawn over the aorta and the cortex, parenchyma, and pelvis of the right kidney on each image of the sequence. Within each ROI, attenuation was quantified on the basis of the HU value, which was expressed as a function of time. The corticoaortic time was the time when the cortical and aortic time-attenuation curves crossed, the corticomedullary time was the time when the cortical and medullary time-attenuation curves crossed, and the corticopelvic time was the time when the cortical and pelvic time-attenuation curves crossed following bolus injection of contrast medium.

One week after the sequential CT scan was performed, a triphasic CT scan was performed with the same technical conditions as the sequential scan. Unenhanced scanning was performed to confirm the scanning range of enhanced scans (CMP and NP) and obtain baseline HU values for the cortex and parenchyma of both kidneys in clinically normal dogs. Scanning was performed with a 3-mm thickness, a 3-mm interval, a pitch of 2, 20 kV (peak), and 80 to 100 mA. The scan range was from the cranial part of T11 to the caudal part of L4. For maximizing opacification of each phase, the delay time of each phase was determined by use of time-attenuation data from the sequential scan. Nephrographic images were acquired immediately after the corticomedullary images. The contrast medium dosage was 3 mL/kg at a rate of 3 mL/s. Postenhancement helical scan images were reconstructed with a 1-mm reconstruction interval. Baseline HU values and enhanced HU values were acquired at the cranial part, pelvic region, and caudal part of the kidney on helical images. Each ROI was hand drawn as close as possible to the margin by use of an abdomen window, and the mean HU value per voxel was recorded. A net HU value was calculated by subtracting the baseline value from the enhanced value.

Statistical analysis was performed by use of a statistical program.^h Two-way ANOVA was performed to analyze nonenhanced HU values, enhanced HU values, and net enhancement according to measurement site. Significant differences between mean values obtained after contrast medium injection and enhancement were determined by use of the Kruskal-Wallis test. A value of P < 0.05 was considered significant.

Results

In the cine scan, the cortex was enhanced immediately after aortic enhancement and the cortex had a lower HU value than the aorta. The medulla was gradually enhanced after the cortical enhancement. The differentiation between the cortex and the medulla continuously diminished over time. Functional indices were measured from the cine scan (Figure 1; Table 1). To obtain a peak enhancement image of the CMP, CMP scanning was started as close as possible to the time of peak cortical enhancement. Therefore, the delay time of the CMP was determined as an initial attenuation time (Figure 2). With time-attenuation curves, the CMP and NP delay times were measured, and applicable CMP and NP delay times were 10 and 44 seconds, respectively. The CMP scan was started at 10 seconds after contrast medium injection, and the NP scan was started immediately after the CMP scan was completed.

View Full Size

Time attenuation curves of the aorta, renal cortex, renal parenchyma, and renal pelvis in a clinically normal dog obtained by use of sequential CT. A—Time-attenuation curves of the aorta (1) and renal cortex (2). B—Time-attenuation curves of the renal cortex (2), medulla (1), and pelvis (3). CAj = Corticoaortic junction. CMj = Corticomedullary junction. CPj = Corticopelvic junction.

Citation: American Journal of Veterinary Research 72, 3; 10.2460/ajvr.72.3.345

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Sequential CT image of the renal cortex of a dog illustrating the ROI of the CMP (A) and time-attenuation curve corresponding to the ROI (B). Delay time of the CMP was the time at which the cortex was enhanced after contrast medium injection (arrow). Values on the x-axis indicate time (milliseconds). Values on the y-axis indicate CT values (HUs). The vertical line indicates time of CT initiation, and the horizontal line indicates CT value at that time.

Citation: American Journal of Veterinary Research 72, 3; 10.2460/ajvr.72.3.345

• Download Figure
• Download figure as PowerPoint slide

In the CMP, the cortex was enhanced and demarcated from the medulla. Renal arteries and aorta were enhanced, and their anatomic structures were partially identified (Figure 3). In the NP, the cortex and medulla were not demarcated. The pelvic region was enhanced or not enhanced, according to the individual difference (Figure 4). Baseline and enhanced HU values of the cortex and parenchyma at the cranial part, caudal part, and pelvic region of both kidneys were acquired and analyzed. There were no significant differences among the measurement regions between right and left kidney or cortex and parenchyma (Table 2).

View Full Size

Reformatted images of the CMP in a dog obtained by use of triphasic helical CT. In the transverse plane (A) and sagittal plane (B), the cortex is hyperattenuated by contrast medium and demarcated from the medulla (arrows indicate the aorta [A] and renal artery [B]).

Citation: American Journal of Veterinary Research 72, 3; 10.2460/ajvr.72.3.345

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Reformatted images of the NP in a dog obtained by use of triphasic helical CT. In the transverse plane (A) and sagittal plane (B), the cortex and medulla are homogeneously hyperattenuated by contrast medium and not differentiated.

Citation: American Journal of Veterinary Research 72, 3; 10.2460/ajvr.72.3.345

• Download Figure
• Download figure as PowerPoint slide

Mean baseline, enhanced, and net HU (enhanced HU - baseline HU) values were calculated and analyzed. There were no significant differences between enhanced and net HU values (Table 3).

Discussion

In humans, CT imaging techniques coupled with dynamic contrast medium injection can clearly demonstrate the progression of the contrast medium from the CMP to the NP. In addition, a helical CT scan combined with an imaging workstation provides various types of information about the renal parenchyma. In a cine scan, it is possible to observe sequential changes in the renal parenchyma. In addition, physiologic functions can be evaluated because the scan reveals quantitative changes in the parenchymal distribution of contrast medium over time.¹⁰ In 1 study,⁹ the author suggested that several variables, including cortex maximum enhancement, medulla maximum enhancement, corticoaortic junction, corticomedullary junction, and corticopelvic junction, could be used to evaluate renal function by use of a sequential CT. These variables may be useful for diagnostic or prognostic evaluations of dogs with renal dysfunction.⁴

An unenhanced CT image is commonly used as a baseline image and can be important for characterization of renal lesions. In an enhanced scan, to acquire high-quality imaging of the CMP, the scan time should be as short as possible and the scan should be operating at the time of peak enhancement of the cortex. In the present study, therefore, the delay time of this phase was no more than 10 seconds, which was the initial attenuation time of the cortex after injection of contrast medium. The delay time of the CMP in humans ranges from 30 to 60 seconds.⁴ The CMP reflects renal perfusion of the contrast medium during its first circulatory pass. Renal abnormalities with potential vascular or perfusional aspects are likely to be visualized best in this phase.¹

In the NP scan, uniform contrast enhancement of the renal parenchyma is achieved. The NP mainly reflects advanced distribution of the contrast medium in the renal interstitial space and filtered contrast medium that enters Henle's loop and the collecting tubules.

Visualization of lesions in the renal medulla is expected to be clearer during the NP.^3,4,10 In humans, the NP starts approximately 70 seconds after administration of the contrast medium. It is maintained until the EP begins. As the EP begins, the contrast medium is concentrated continuously in the collecting system, causing streak artifacts.¹¹ In the present study, the NP began approximately 44 seconds after contrast medium injection and terminated before the onset of the EP, which began at 65 seconds.

The onset of the NP depends on the contrast medium injection protocol and differences among individuals.¹² When the injection rate is slow and the volume of contrast medium is large, the NP onset time is delayed. The NP onset time is more affected by the volume of the contrast medium than the injection rate. In humans, for triphasic helical CT, volume of the contrast medium is usually 120 to 150 mL administered at a flow rate of 2 to 3 mL/s. The duration of the injection is approximately 40 to 75 seconds. In the present study, on the other hand, the volume of the contrast medium was less and the injection duration was shorter. Therefore, the onset time was faster and duration was shorter, compared with values obtained in humans. In humans, several studies^3,4,6 have established the advantage of triphasic helical CT, compared with conventional CT or ultrasonography. Further investigations are necessary to make comparisons between triphasic helical CT and other modalities.

In the present study, HU values of the cortex in the CMP were often slightly higher in the caudal part than in the cranial part of the kidney and in the right kidney than in the left kidney. The HU values of the parenchyma in the NP were often slightly lower in the caudal part than in the cranial part of the kidney and in the right kidney than in the left kidney. This occurred because of the limitations of the CT equipment and the relatively short duration of the CMP and NP. However, there were no significant differences. Moreover, the net HU values obtained from the cortex and parenchyma were not significantly different from the enhancement HU values in these clinically normal dogs.

Results of the present study may provide reference range values for delay times and HU values in each phase of a triphasic renal CT scan in dogs. Triphasic CT scanning performed with optimal techniques can be useful in the diagnosis and evaluation of renal diseases in dogs. Further investigations are necessary to determine any differences associated with breeds and body weights and to apply the technique to dogs with renal diseases.