DOI: 10.1148/radiol.2411050597
(Radiology 2006;241:267-274.)
© RSNA, 2006
Prospective Evaluation of Vascular Complications after Liver Transplantation: Comparison of Conventional and Microbubble Contrast-enhanced US1
Benjamin K. Hom, BS,
Ruchi Shrestha, MD,
Suzanne L. Palmer, MD,
Michael D. Katz, MD,
R. Rick Selby, MD,
Zhanna Asatryan, BA,
Jabali K. Wells, BS and
Edward G. Grant, MD
1 From the Departments of Radiology (B.K.H., R.S., S.L.P., M.D.K., Z.A., J.K.W., E.G.G.) and Surgery (R.R.S.), University of Southern California, Keck School of Medicine, USC University Hospital, 1500 San Pablo St, Los Angeles, CA 90033. From the 2004 RSNA Annual Meeting. Received April 14, 2005; revision requested June 13; revision received July 25; accepted September 1; final version accepted February 1, 2006. Supported by a grant from GE Healthcare.
Address correspondence to E.G.G. (e-mail: edgrant{at}usc.edu).
 |
ABSTRACT
|
|---|
Purpose: To prospectively compare diagnostic performance of conventional Doppler ultrasonography (US) and microbubble contrast material–enhanced US for assessment of vascular complications after liver transplantation, with clinical follow-up or angiography as reference standard.
Materials and Methods: This study was approved by institutional review board and was HIPAA compliant. Written informed consent was obtained. Seventy-two patients (49 men, 23 women; average age, 52.3 years) were included in this study. Patients who had undergone liver transplantation underwent conventional color Doppler and contrast-enhanced US of the liver. Quality of hepatic artery (HA) and portal vein (PV) visualization, contrast material arrival time, and time for complete evaluation of vasculature were compared for both techniques. McNemar test was used to compare vascular flow visualization scores; Student t test was used to compare mean study times with both techniques. Patients without HA flow at Doppler US underwent angiography; those with flow were followed up clinically. McNemar test was used to compare sensitivity of both techniques.
Results: Contrast-enhanced US helped significantly improve flow visualization in hepatic vessels (P < .001). Mean contrast material arrival time was 13.7 seconds ± 3.8 (standard deviation) in proper HA and 20.7 seconds ± 6.3 in PV. Mean study time decreased from 27.4 minutes ± 13.9 to 9.3 minutes ± 4.5 (P < .01). Doppler US failed to depict HA flow in eight patients; contrast-enhanced US showed flow in six and no flow in two of these patients. Follow-up results confirmed contrast-enhanced US findings. Sensitivity, specificity, and accuracy for Doppler US were 91.3%, 100%, and 91.5%, respectively. Sensitivity, specificity, and accuracy of contrast-enhanced US were all 100%. Sensitivity and accuracy values of the two techniques were significantly different (P < .014); there was no significant difference in specificity (P > .99)
Conclusion: Contrast-enhanced US helped improve flow visualization in the HA and PV, decrease scanning time, and correctly differentiate between thrombosis and a patent artery in patients without HA flow at conventional Doppler US.
© RSNA, 2006
|
INTRODUCTION
|
|---|
In the United States, 5671 patients received orthotopic liver transplants in 2003, and more than 17 000 people were on the waiting list as of January 1, 2004 (1). Given the seriousness of the surgery and the lack of available livers for transplantation, graft loss is a serious problem. Hepatic artery (HA) thrombosis occurs in 3%–8% of transplants in adults, and it is the most common vascular complication after orthotopic liver transplantation (2,3). Acute HA thrombosis in this population almost invariably leads to graft loss from infarction and eventual abscess formation. Although portal vein (PV) thrombosis is less frequently encountered than HA thrombosis, it also represents a serious complication that often leads to graft loss as well (4). Ultrasonography (US) is usually the initial imaging technique used for identification of vascular complications in the early postoperative period and also is used for long-term follow-up (4–6).
Although the performance of Doppler US is reasonably acceptable in this role, with a reported sensitivity of 92% (6), its inability to depict flow in a patent HA remains a substantial problem despite improvements in Doppler technology (7). In patients in whom there is no flow identified in the HA, angiography, with its attendant risks, usually is required to obtain a definitive answer. In addition, Doppler study of the hepatic vasculature can be technically challenging and time consuming. We hypothesized that the use of a US contrast agent to assist in visualization of the hepatic vasculature would improve the quality of the scans, decrease scanning time, and potentially decrease the number of false-negative examinations for visualization of vascular flow and, therefore, the number of angiograms.
Thus, the purpose of our study was to prospectively compare the diagnostic performance of conventional and microbubble contrast material–enhanced US for the assessment of vascular complications after liver transplantation, with clinical follow-up or angiography as a reference standard.
|
MATERIALS AND METHODS
|
|---|
Study Population
From June 2003 to December 2004, any liver transplant recipient older than 18 years at our institution for whom a Doppler US examination was ordered was eligible for our study. Patients enrolled in other research protocols were excluded. This investigation was approved by our institutional review board and was compliant with the Health Insurance Portability and Accountability Act. Written informed consent was obtained from all patients. The authors had control of the data and the information submitted for publication.
Seventy-two patients (49 men, 23 women) were enrolled in the study. The average patient age was 52.3 years (age range, 30.5–68.2 years).
Conventional US
Patients who had undergone cadaveric or living related liver transplantation and were referred for Doppler examination underwent routine gray-scale, color, and spectral Doppler examination of the liver with a commercially available unit (Acuson Sequoia 512; Siemens Medical Solutions, Mountain View, Calif). The mean time between transplantation and conventional US was 17.0 days (range, 1–2326 days). Studies were performed in the US suite and at bedside in the intensive care unit. Studies were performed according to a previously delineated scanning protocol in one accredited US facility by two experienced technologists, including one with 15 years of experience and the other (Z.A.) with 17 years of experience in performance of US of the liver. The vascular evaluation included imaging of the proper, left, and right HAs; the hepatic veins; and the main, right, and left PVs. For this study, only the HAs and PVs were considered. The total time to image all three HA segments was recorded.
Contrast-enhanced US
Contrast-enhanced US was performed within 24 hours of conventional US. The contrast agent (Optison; GE Healthcare, Princeton, NJ) is a second-generation perfluorocarbon-based contrast material with a median bubble diameter of approximately 3.5 µm and a concentration of 5–8 x 108 microbubbles per milliliter. The agent was injected as a 0.5-mL bolus and was immediately followed by a 10-mL saline flush into either a peripheral vein or central venous catheter. Patients were imaged after contrast material injection by using a technique with a low mechanical index (<0.5) and the same scanner that was used for conventional US. A phase-inversion technique was used during the first half of the study, and a newer technique known as contrast pulse sequencing was used for the second half. With this technique, both phase inversion and amplitude modulation were used to differentiate tissue echoes from those arising from the contrast agent. The operator can vary the image to display only those echoes that arise from contrast material, only those echoes that return from soft tissues, or a combination of both. In the latter situation, the echoes that arise from contrast material are displayed in a shade of gold and are overlaid on the background of the gray-scale image. This was the method typically used in our patients.
Our contrast imaging protocol included three bolus injections followed by specific imaging of the proper, right, and left HAs and their adjacent PV segments. Patients received a maximum of 3 mL of contrast agent in six possible injections. Between injections, continuous imaging was performed until the contrast material disappeared from the circulation. This portion of the study was performed by one radiologist (E.G.G.) with 13 years of experience in the performance of contrast-enhanced US evaluations of the liver. The total time to image all three HA segments was recorded.
Image Evaluation and Comparisons
The quality of flow visualization in the proper, right, and left HAs and the main PV was subjectively assigned a grade on a scale of 0–3, as follows: grade 0, nonvisualization of the vessel segment; grade 1, patchy visualization with less than 50% of the expected length of the vessel depicted with color and/or contrast material; grade 2, patchy visualization with depiction of more than 50% of the expected length of the vessel; and grade 3, clearly defined visualization of all parts of the vessel segment in question. For the proper HA, flow was sought throughout the region of the porta hepatis. For the right and left HA segments, complete visualization was considered present when the entire segment of the vessel was visible on one section obtained at the region of the bifurcation. This process was repeated after the injection of contrast material.
A comparison also was made between the time it took the technologist to identify each of the three segments of the HA by using color and/or spectral Doppler US versus the time it took the radiologist to complete this task after contrast material injection. The end point for the former was documentation of color and spectral images. For contrast-enhanced images, anatomic depiction of the HA and PV segments was considered sufficient on the basis of a presumed temporal differential appearance of contrast material in the HA and PV and excellent anatomic detail provided by the contrast agent. Because the arrival time for contrast material in the central hepatic vessels was unknown at the start of this study and could provide a secondary method for differentiating between the HA and PV, we recorded the time between the appearance of contrast material in the HA segment and in the adjacent PV after saline flush for each of the three injections. We calculated the mean time of contrast material arrival and standard deviations. The route of injection through a peripheral intravenous versus a central venous catheter was taken into consideration for this calculation. Because the determination of arrival time in patients with central venous catheters was inconsistent, only arrival times for patients with peripheral intravenous catheters were considered. Final results of the contrast-enhanced study were recorded at the conclusion of the examination.
Follow-up
Patients in whom flow was demonstrated at both conventional Doppler and contrast-enhanced US were followed up clinically because the majority of patients with HA or PV thrombosis develop severe complications ascribable to ischemia. Clinically significant hepatic ischemia is typically associated with graft loss and the need for repeat transplantation or signs of intrahepatic biliary stricture. The charts from all patients enrolled in the study were reviewed (B.K.H., J.K.W.) for clinical or imaging evidence of these complications.
Patients in whom flow could not be identified in the HA at conventional Doppler US underwent selective angiography. Those with absent flow in the PV or those suspected of having partial thrombosis were evaluated with contrast-enhanced computed tomography (CT). These studies represent the accepted follow-up studies for HA and PV thrombosis at our institution. Both the referring transplantation surgeons and the angiographer were blinded to the results of contrast-enhanced US. Selective catheterization of the HA conduit was performed, and contrast material (Omnipaque 300; GE Healthcare) was injected at a rate of 5 mL/sec (total volume, 20 mL). The examinations were performed with a digital system (Multistar; Siemens Medical Solutions, Erlangen, Germany) and recorded on film. The final diagnosis was determined at the conclusion of angiography.
Statistical Analysis
The results of conventional Doppler and contrast-enhanced US were compared with those of clinical follow-up or angiography, if the latter was performed. The sensitivity, specificity, and accuracy for determination of the patency of the HA and PV were calculated for each blood vessel before and after the injection of contrast material. Corresponding confidence intervals (CIs) were calculated by using the approach of Clopper and Pearson (8). The McNemar test was used to compare the performance of conventional Doppler US and contrast-enhanced US for the identification of vascular flow. Angiograms of all patients in whom the HA was not visualized at conventional US were evaluated individually to determine whether a cause for nonvisualization could be determined. The McNemar test also was used to compare the flow visualization scores of the HAs with and without contrast material. The mean time to visualization of the arterial segments and the mean arrival time of the contrast material in the HA and PV and standard deviations were calculated. The mean study times for both methods were compared by using the Student t test. A difference with P value of less than .05 was considered significant. Software (SAS, version 9.1 for Windows; SAS Institute, Cary, NC) was used for the statistical analysis.
|
RESULTS
|
|---|
Twenty-nine of 72 patients were examined at bedside in the intensive care unit. Figure 1 summarizes the underlying types of liver disease in all patients. Fifty-one patients received cadaveric liver transplants and 21 received living related grafts. During the study, 444 US examinations were performed at our institution in 172 patients who received liver transplants.
Flow
Among the 72 patients included in this investigation, contrast-enhanced scans were considered successful in 71. Contrast-enhanced scans were considered successful when the contrast agent was identified in the hepatic vasculature and parenchyma. Early in the study, adequate visualization was not achieved in one patient because of technical problems with the patient's central venous catheter. Conventional Doppler US demonstrated flow in all three HA segments in 63 of these 71 patients. Among those 63 patients, there were no instances in which findings at clinical follow-up were suggestive of HA thrombosis. There were no patients in whom the PV was not seen at conventional Doppler US. In one patient, findings at both Doppler and contrast-enhanced US were suggestive of a partial thrombosis. This potential determination was corroborated with CT findings. The flow visualization scores of all three HA segments and the PV improved significantly (Table 1) with contrast-enhanced US (Fig 2).
View larger version (73K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a: Longitudinal oblique contrast-enhanced US scans obtained with contrast pulse sequencing in the mixed mode. With this technique, echoes arising from contrast material are displayed in a shade of gold overlaid on a background of gray-scale tissue echoes. (a) Image obtained 11.6 seconds after contrast material injection demonstrates a normal proper HA (arrow) and central branches. The PV (P) is devoid of echoes and displayed in black. (b) Image obtained through the same region 19.5 seconds after contrast material injection demonstrates intense enhancement of the HA (arrow); contrast material filled the PV (P). (c) Image obtained 45 seconds after contrast material injection demonstrates homogeneous parenchymal enhancement, with common bile duct (arrow).
|
|
View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b: Longitudinal oblique contrast-enhanced US scans obtained with contrast pulse sequencing in the mixed mode. With this technique, echoes arising from contrast material are displayed in a shade of gold overlaid on a background of gray-scale tissue echoes. (a) Image obtained 11.6 seconds after contrast material injection demonstrates a normal proper HA (arrow) and central branches. The PV (P) is devoid of echoes and displayed in black. (b) Image obtained through the same region 19.5 seconds after contrast material injection demonstrates intense enhancement of the HA (arrow); contrast material filled the PV (P). (c) Image obtained 45 seconds after contrast material injection demonstrates homogeneous parenchymal enhancement, with common bile duct (arrow).
|
|
View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c: Longitudinal oblique contrast-enhanced US scans obtained with contrast pulse sequencing in the mixed mode. With this technique, echoes arising from contrast material are displayed in a shade of gold overlaid on a background of gray-scale tissue echoes. (a) Image obtained 11.6 seconds after contrast material injection demonstrates a normal proper HA (arrow) and central branches. The PV (P) is devoid of echoes and displayed in black. (b) Image obtained through the same region 19.5 seconds after contrast material injection demonstrates intense enhancement of the HA (arrow); contrast material filled the PV (P). (c) Image obtained 45 seconds after contrast material injection demonstrates homogeneous parenchymal enhancement, with common bile duct (arrow).
|
|
Time
The mean total time needed for the technologist to identify all three HA segments was 27.4 minutes ± 13.9 (standard deviation) for Doppler US and 9.3 minutes ± 4.5 for contrast-enhanced US (P < .01). Differences between Doppler US and contrast-enhanced US in regard to the time to identify the HA segments were more pronounced when we excluded images from the patients in whom arterial flow was not depicted at conventional US. In these patients, the mean scanning time was 42.7 minutes ± 6.8 for Doppler US and 12.6 minutes ± 3.3 for contrast-enhanced US, despite that these patients often received six bolus injections of contrast material (P < .01).
After completion of the saline flush, mean time for visualization of the proper HA after the injection of contrast material was 13.7 seconds ± 3.8, that for visualization of the right HA was 14.5 seconds ± 4.7, and that for visualization of the left HA was 14.5 seconds ± 4.0. The main PV was seen a mean of 20.7 seconds ± 6.3 after contrast material injection. This represented an average difference of 7.1 seconds between the arrival time of contrast material in the HA segments and the arrival time of contrast material in the PV.
No Flow Visualization
In eight patients, flow was not identified in all three HA segments at Doppler US. In one of these patients, who underwent imaging late in the study, flow was not seen at Doppler US in the right HA but was present in the proper and left segments. Flow was clearly visualized in all three segments after the injection of contrast material. Given the patient's stable clinical condition, lack of liver function test abnormalities, and obvious flow in all three HA segments after contrast material injection, we did not believe that angiography was justified. Flow in the right HA was clearly demonstrated at follow-up conventional US. The patient continued to do well 6 months after the contrast-enhanced US. In two patients, neither technique provided visualization of any segment of the HA, and the artery was considered to be thrombosed. Findings at angiography confirmed thrombosis at the anastomosis in both of those patients (Fig 3), and repeat transplantation was necessary. In the remaining five patients, Doppler US did not depict flow in any of the three HA segments. In four of these patients, flow in all three HA segments was clearly depicted after contrast material injection, and patency was confirmed with selective angiography. In the fifth patient, flow to the HA bifurcation and into the proximal right HA was seen, but flow into the peripheral right HA or into the left HA was not seen clearly after contrast material injection. Given this pattern, we believed that the patient had attenuated flow rather than thrombosis. At angiography, the vessel was not seen after aortic injection; however, patent spindly arteries were confirmed at selective angiography.
View larger version (47K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a: HA thrombosis in 34-year-old woman who had undergone living related liver donor transplantation. (a) Longitudinal oblique conventional Doppler US scan obtained 2 days after transplantation fails to show HA flow beyond the level of the porta hepatis. Only a small stump of the patent artery (PHA) is shown on this image. Visualization of the PV and its intrahepatic branches is normal. (b) Longitudinal oblique contrast-enhanced US scan obtained 4 hours after conventional Doppler US. The HA was not visualized. Normal filling of the PV (P) was demonstrated. This parenchymal phase image shows almost a complete lack of perfusion in the anterior right lobe. (c) Posteroanterior selective angiogram obtained after celiac artery injection confirms complete thrombosis of right HA at the level of the anastomosis (white arrow). The arterial anatomy is unusual. The left HA (black arrow) was ligated at surgery and arises from the common HA, which continues into the gastroduodenal artery. The patient underwent repeat transplantation 2 days later.
|
|
View larger version (60K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b: HA thrombosis in 34-year-old woman who had undergone living related liver donor transplantation. (a) Longitudinal oblique conventional Doppler US scan obtained 2 days after transplantation fails to show HA flow beyond the level of the porta hepatis. Only a small stump of the patent artery (PHA) is shown on this image. Visualization of the PV and its intrahepatic branches is normal. (b) Longitudinal oblique contrast-enhanced US scan obtained 4 hours after conventional Doppler US. The HA was not visualized. Normal filling of the PV (P) was demonstrated. This parenchymal phase image shows almost a complete lack of perfusion in the anterior right lobe. (c) Posteroanterior selective angiogram obtained after celiac artery injection confirms complete thrombosis of right HA at the level of the anastomosis (white arrow). The arterial anatomy is unusual. The left HA (black arrow) was ligated at surgery and arises from the common HA, which continues into the gastroduodenal artery. The patient underwent repeat transplantation 2 days later.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c: HA thrombosis in 34-year-old woman who had undergone living related liver donor transplantation. (a) Longitudinal oblique conventional Doppler US scan obtained 2 days after transplantation fails to show HA flow beyond the level of the porta hepatis. Only a small stump of the patent artery (PHA) is shown on this image. Visualization of the PV and its intrahepatic branches is normal. (b) Longitudinal oblique contrast-enhanced US scan obtained 4 hours after conventional Doppler US. The HA was not visualized. Normal filling of the PV (P) was demonstrated. This parenchymal phase image shows almost a complete lack of perfusion in the anterior right lobe. (c) Posteroanterior selective angiogram obtained after celiac artery injection confirms complete thrombosis of right HA at the level of the anastomosis (white arrow). The arterial anatomy is unusual. The left HA (black arrow) was ligated at surgery and arises from the common HA, which continues into the gastroduodenal artery. The patient underwent repeat transplantation 2 days later.
|
|
Diagnostic Performance
The sensitivity, specificity, and accuracy of Doppler US were 91.3% (95% CI: 82.0%, 96.7%), 100% (95% CI: 15.8%, 100%), and 91.5% (95% CI: 82.5%, 96.8%), respectively. The sensitivity, specificity, and accuracy of contrast-enhanced US were 100% (95% CI: 94.8%, 100%), 100% (95% CI: 15.8%, 100%), and 100% (95% CI: 94.9%, 100%), respectively. The differences between the sensitivity and accuracy values of both techniques were significant (P < .014); there was no significant difference in the specificity values of both techniques (P > .99).
Reasons for Nonvisualization
Angiograms and contrast-enhanced US scans were reviewed to determine whether there were underlying abnormalities that could account for nonvisualization of the HA and PV on Doppler US scans among those patients who had patent arteries. Among the four patients whose HA segments were patent, one, who had undergone surgery 36 hours earlier, had an anastomotic flap at angiography, and the flap was surgically revised. In a second patient, contrast-enhanced US and angiography depicted numerous vessels, which were presumed to represent collateral vessels, in the porta hepatis but normally patent right and left HA segments (Fig 4). This development of collateral vessels was likely the result of an earlier undetected HA thrombosis. This patient remains clinically asymptomatic and continues to be followed up; no intervention other than angiography was undertaken. In a third patient, weak arterial signals were identified after contrast material injection in all three arterial segments, and thin attenuated arteries were confirmed at angiography. This patient, who had undergone living related liver transplantation, required repeat transplantation but died 3 weeks later of other complications. The fourth patient had a normal angiogram, and no obvious cause for nonvisualization on the conventional Doppler US scan was found. He continues to do well clinically for 14 months. The fifth patient, in whom attenuated flow was present in the common and right HA segments after contrast material injection, was described earlier. The difficulty in identification of flow in this patient was attributed to a combination of poor cardiac output and spasm of the HAs. Table 2 provides a summary of the preceding information.
View larger version (44K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a: HA thrombosis with subsequent collateral vessel formation. (a) Longitudinal oblique contrast-enhanced US scan obtained in the contrast-only mode (with absence of gray-scale background) demonstrates multiple small vessels in the region of the porta hepatis (arrow) anterior to the PV (P). The image was obtained in the arterial phase, and the vein had not yet filled with contrast material. Scans obtained in the cephalad direction (not shown) demonstrated patency of the right and left HA branches. (b) Right anterior oblique selective angiogram failed to demonstrate a normally patent proper HA. The proper HA is replaced by a series of tiny collateral vessels. Note the reconstitution of the intrahepatic branches of the HA from collateral vessels.
|
|
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b: HA thrombosis with subsequent collateral vessel formation. (a) Longitudinal oblique contrast-enhanced US scan obtained in the contrast-only mode (with absence of gray-scale background) demonstrates multiple small vessels in the region of the porta hepatis (arrow) anterior to the PV (P). The image was obtained in the arterial phase, and the vein had not yet filled with contrast material. Scans obtained in the cephalad direction (not shown) demonstrated patency of the right and left HA branches. (b) Right anterior oblique selective angiogram failed to demonstrate a normally patent proper HA. The proper HA is replaced by a series of tiny collateral vessels. Note the reconstitution of the intrahepatic branches of the HA from collateral vessels.
|
|
|
DISCUSSION
|
|---|
Historically, the diagnostic performance for detection of HA thrombosis with conventional Doppler US is reported to be excellent (5,7). In our series of patients, the sensitivity to flow of Doppler US was 91.3%. False-positive examinations are unusual in adults (7) but have been reported in children as a result of silent HA thrombosis with collateral vessel formation (9). Nonvisualization of flow is usually simply ascribed to the weakness of signal strength in a relatively small artery and is considered an acceptable limitation of Doppler US. Color Doppler technology has improved considerably during the past 10 years, and it is likely that the number of studies in which flow was not visualized has decreased during that time. This finding is supported by a comparison of our lower incidence of HA nonvisualization with that of a study by Sidhu et al (7), who used older technology for scanning the majority of patients in their series. In their study, the HA was not visualized with conventional Doppler US in 4.4% of 794 patients. In our study, flow was not demonstrated in 1.8% of scans (eight of 444 scans). Regardless, false-negative diagnoses for flow visualization remain a limitation of conventional US. The addition of contrast material, however, clearly overcame this problem and increased the sensitivity for depiction of flow to 100%.
When we compared conventional Doppler US with contrast-enhanced US, the latter provided a statistically significant improvement in the visualization of all three main HA segments and the PV (P < .01). Although the PV may be difficult to visualize in patients before transplantation (10), it is well visualized in almost all patients after transplantation. Practically speaking, contrast material added little in this evaluation. This finding has been corroborated in other studies (11). In addition, the incidence of PV thrombosis is low in most centers at this time and does not represent as great a threat to graft survival as does HA thrombosis.
The use of contrast material provided a significant decrease, from 27.4 minutes to 9.3 minutes, in the mean time required to identify the HA. Although this time did not include the added time for setting up the injection, almost all patients had existing intravenous access in place, and the contrast material itself was easily withdrawn from premixed vials and rapidly administered. Furthermore, when we excluded those patients in whom flow was not seen at conventional Doppler US, the mean time difference became even more pronounced (42.7 vs 12.3 minutes) despite the fact that the maximum number of injections was almost invariably given.
There is little information available in the literature about the arrival time of contrast material in the HA versus that in the PV. The fact that the HA can be seen earlier than the PV is an accepted tenet of diagnostic imaging and an important part of many routine contrast-enhanced studies with CT and magnetic resonance (MR) imaging. We believed that it would be of benefit to evaluate this difference in arrival time with contrast-enhanced US because arrival time, if different for the PV and HA, could be of value for differentiation between the two. Our study showed that there is a significant temporal difference in the mean arrival time of contrast material (7.1 seconds) in the HA versus the PV, and, in practical terms, this may be used to facilitate the differentiation of these blood vessels. The arrival times of contrast material in the HA and PV in our study are quite different from those typically observed with other imaging studies such as CT. This is likely attributable to the fact that we were evaluating the larger central arteries and PVs, whereas the typical multiphase CT scan demonstrates parenchymal enhancement patterns. As such, contrast material arrives in the vessels more rapidly than it does with the usual parenchymal imaging times used in CT or MR imaging, and the temporal difference between the arrival time in the HA and PV is shorter.
Patients in whom conventional Doppler US fails to depict flow in the HA typically undergo selective angiography. The results of our study suggest that a diagnosis of HA thrombosis can be confidently confirmed with a contrast-enhanced examination and that angiography may be avoided in this group. Our findings are supported by those of the study by Sidhu et al (7); in that study, HA thrombosis was confirmed in two patients without flow at contrast-enhanced US.
In a large study of patients who had undergone liver transplantation and were evaluated with contrast-enhanced US, Sidhu et al (7) did not evaluate the potential causes of nonvisualization of the HA at conventional Doppler US. As described earlier, the number of patients in whom HA flow was not visualized at conventional Doppler US in our study was quite low, which was likely because we used more state-of-the-art US technology than did Sidhu et al (7). As such, our population was likely strongly selected for having possible complications involving the HA. Many of our patients with no flow at Doppler US had underlying unusual abnormalities; however, the use of contrast material helped identify the cause in most cases. Although the formation of collateral vessels around a silent HA thrombosis is rare in adults, it does occur (7) and can be correctly identified with contrast-enhanced US. Attenuated flow, whether secondary to spasm or low cardiac output, can also cause nonvisualization of flow at conventional Doppler US and can be differentiated from true thrombosis by using contrast-enhanced US. In one of our patients, arterial flow (although extremely slow) was clearly visualized in all segments, and minimal flow was confirmed at angiography. In another patient, because segmental thrombosis has, to our knowledge, never been reported, the visualization of most HA segments after contrast material injection was sufficient to guide us away from a diagnosis of HA thrombosis. Contrast-enhanced US further suggested the correct diagnosis because it depicted slow markedly reduced arterial flow in visualized segments at real-time evaluation.
There were several limitations to this study. First, flow visualization scores for conventional and contrast-enhanced US were determined by the person who performed US at the conclusion of the examination. Assignment of grades was, therefore, performed by two different people. To minimize possible bias, the grading system was established and standardized at the beginning of the project. Neither person knew the scores of the other. Second, the results of conventional Doppler US were known to the radiologist performing the contrast-enhanced study; this potentially could have introduced a bias. Finally, only those patients in whom flow was absent at conventional Doppler US underwent angiography. The performance of such an invasive procedure in patients with flow at conventional Doppler US was obviously not justifiable from an ethical and cost standpoint, and the complications of HA thrombosis are almost invariably so severe that we believed clinical follow-up represented an adequate reference standard in these patients. Only two patients with absent flow at both Doppler and contrast-enhanced US had HA thrombosis at angiography. This small number of cases resulted in a wide CI for specificity for the visualization of flow. Additional prospective studies will be necessary to evaluate the specificity of contrast-enhanced US.
In conclusion, results of our comparison of conventional Doppler and contrast-enhanced US indicate that contrast material provides improved visualization of the HA and PV. The use of contrast material also significantly helps to decrease scanning time; this decrease was particularly dramatic in patients in whom conventional Doppler US failed to show flow. Most important, the use of contrast material helped to improve the sensitivity of US for the visualization of flow and can aid in decreasing the need for angiography in most patients.
|
ADVANCES IN KNOWLEDGE
|
|---|
- Contrast-enhanced US, compared with routine color Doppler US, helps to improve the visualization of the hepatic artery and portal vein in patients who have received liver transplants.
- Contrast-enhanced US, compared with routine color Doppler US, helps to improve the sensitivity and accuracy for hepatic artery flow in patients who have received liver transplants.
- Contrast-enhanced US helps to decrease scanning time in patients who have received liver transplants.
- Temporal arrival times of US contrast material can help differentiate the hepatic artery and its segments from the portal vein in patients who have received liver transplants.
|
ACKNOWLEDGMENTS
|
|---|
The authors acknowledge Todd Alonzo, PhD, and Hossein Jadvar, MD, PhD, MPH, for their invaluable assistance with the statistical analysis in this article.
|
FOOTNOTES
|
|---|
Abbreviations: CI = confidence interval • HA = hepatic artery • PV = portal vein
See Materials and Methods for pertinent disclosures.
Author contributions: Guarantor of integrity of entire study, E.G.G.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, B.K.H.; clinical studies, B.K.H., R.R.S., E.G.G.; statistical analysis, B.K.H.; and manuscript editing, B.K.H., E.G.G.
|
References
|
|---|
- United Network for Organ Sharing and Scientific Registry Data. Data from the Organ Procurement and Transplantation Network and the U.S. Scientific Registry of Transplant Recipients. Available at: http://www.unos.org. Accessed August 18, 2004.
- Jain A, Reyes J, Kashyap R, et al. Long-term survival after liver transplantation in 4000 consecutive patients at a single center. Ann Surg 2000;232:490–500.[CrossRef][Medline]
- Shaw AS, Ryan SM, Beese RC, Sidhu PS. Ultrasound of non-vascular complications in the post liver transplant patient. Clin Radiol 2003;58:672–680.[CrossRef][Medline]
- Langnas AN, Marujo W, Stratta RJ, Wood RP, Shaw BW Jr. Vascular complications after orthotopic liver transplantation. Am J Surg 1991;161:76–83.[CrossRef][Medline]
- Flint EW, Sumkin JH, Zajko AB, Bowen A. Duplex sonography of liver transplantation. AJR Am J Roentgenol 1988;151:481–483.[Abstract/Free Full Text]
- Dodd GD III, Memel DS, Zajko AB, Baron RL, Santaguida LA. Hepatic artery stenosis and thrombosis in transplant recipients: Doppler diagnosis with resistive index and systolic acceleration time. Radiology 1994;192:657–661.[Abstract/Free Full Text]
- Sidhu PS, Shaw AS, Ellis SM, Karani JB, Ryan SM. Microbubble ultrasound contrast in the assessment of hepatic artery patency following liver transplantation: role in reducing frequency of hepatic artery arteriography. Eur Radiol 2004;14:21–30.[CrossRef][Medline]
- Clopper CJ, Pearson ES. The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika 1934;26:404–413.[Free Full Text]
- Hall TR, McDiarmid SV, Grant EG, Boechat MI, Busuttil RW. False-negative duplex Doppler studies in children with hepatic artery thrombosis after liver transplantation. AJR Am J Roentgenol 1990;154:573–575.[Abstract/Free Full Text]
- Marshall MM, Beese RC, Muiesan P, Sarma DI, O'Grady J, Sidhu PS. Assessment of portal venous system patency in the liver transplant candidate: a prospective study comparing ultrasound, microbubble-enhanced colour Doppler ultrasound, with arteriography and surgery. Clin Radiol 2002;57:377–383.[CrossRef][Medline]
- Herold C, Reck T, Ott R, et al. Contrast-enhanced ultrasound improves hepatic vessel visualization after orthotopic liver transplantation. Abdom Imaging 2001;26:597–600.[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
A. Garcia-Criado, R. Gilabert, A. Berzigotti, and C. Bru
Doppler Ultrasound Findings in the Hepatic Artery Shortly After Liver Transplantation
Am. J. Roentgenol.,
July 1, 2009;
193(1):
128 - 135.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. H. M. Caiado, R. Blasbalg, A. S. Z. Marcelino, M. da Cunha Pinho, M. C. Chammas, C. da Costa Leite, G. G. Cerri, A. C. de Oliveira, T. Bacchella, and M. C. C. Machado
Complications of Liver Transplantation: Multimodality Imaging Approach
RadioGraphics,
September 1, 2007;
27(5):
1401 - 1417.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
