Perfusion Defects Detected After Liver Transplantation on Multidetector Computerized Tomography: Short- and Long-Term Follow-up Y.C. Shin, C.S. Choi, and J.S. Kim ABSTRACT Background. Perfusion defects are sometimes found during the follow-up computerized tomography (CT) after liver transplantation (OLT). This study sought to determine the short- and long-term outcomes of perfusion defects observed after OLT with the use of multidetector CT. Methods. From February 4, 2004, to December 8, 2011, a total 46 LTs were performed in our hospital owing to end-stage liver cirrhosis (n ¼ 43), liver cirrhosis with hepatocellular carcinoma (HCC; n ¼ 1), combined HCC with cholangiocellular carcinoma (CCC; n ¼ 1), or hepatic failure from acute hepatitis A (n ¼ 1). The transplanted livers were procured from cadaveric (n ¼ 24) or living related donors (n ¼ 22). The average age of the recipient was 53.3 10.4 years. The male-female ratio was 30:16. Postoperative multidetector CT was performed with a dynamic sequence in 203 examinations and with a portal phase in 46 examinations. The contrast media was Radisense. The rate of injection of 120 mL was 3 mL/s with a power injector; the iodine concentration was 300 or 370 mg/dL. Follow-up ranged from 3 months 3 days to 7 years 363 days. We classified perfusion defects as chronic segmental or subsegmental benign ischemia, transient focal perfusion defects, benign subcapsular ischemia, or fatal whole liver perfusion defects. Results. There were 3 cases of chronic segmental or subsegmental benign ischemia, 8 focal transient perfusion defects, 1 benign subcapsular ischemia, and 4 fatal whole liver perfusion defects. Except the fatal cases, all other perfusion defects occurred in the courses of benign conditions without resection or reoperation. Conclusions. Most perfusion defects were benign and uneventful, requiring no treatment, with the exception of fatal whole liver perfusion defects, which resulted in death after detection.
T
HE FINAL TREATMENT option for end-stage liver disease mainly associated with cirrhosis is liver transplantation using organs from living related or cadaveric donors. Using to the shortage of organs for transplantation from brain dead patients, livers from living related donors are the major source in Korea.1,2 The liver is a unique organ with a dual blood supply from the hepatic arterial and portal venous systems. In addition, there are numerous intrahepatic shunts, such as arteriovenous, arterioportal, and portovenous blood flows. Hepatic sinuses can communicate with each other when normal intrahepatic blood flow is disturbed,3 eg, by portal hypertension, inflammation, tumors, or
Budd-Chiari syndrome. These increase intravascular pressure in the involved segment to form shunts to preserve the liver.3
From the Department of Cardiovascular and Thoracic Surgery (Y.C.S.), Department of Radiology (C.S.C.), and Department of Surgery (J.S.K.), Kangdong Sacred Heart Hospital, College of Medicine, Hallym University, Seoul, Korea. Address reprint requests to Chul Soon Choi, Department of Radiology, Kangdong Sacred Heart Hospital, Kildong 445, Kangdongku, Seoul, Korea, zip code: 134-701. E-mail: [email protected]
ª 2013 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/13/$esee front matter http://dx.doi.org/10.1016/j.transproceed.2013.08.024
Transplantation Proceedings, 45, 3183e3186 (2013)
3183
3184
SHIN, CHOI, AND KIM
Fig 1. (A) Computerized tomography image in the portal phase 2 years 9 months 25 days after liver transplantation. There is a segmental perfusion defect with sharp anterior and posterior margins in segment 8, and it extends from the medial capsule to the lateral capsule of the liver (arrow). (B) Computerized tomography image in the portal phase 8 years 4 days after liver transplantation. Segment 8 is contracted, and capsular retraction is noted with bile duct dilatation (arrow).
Liver transplantation, regardless of the type of procured liver, requires anastomoses of the hepatic arteries, portal and hepatic veins and bile ducts between the transplanted liver and the host. The anastomotic sites pose risks for complications. Transplanted livers are vulnerable to ischemia or infarction after transplantation, especially regarding hepatic arterial or portal venous out flows.4 After contrast media enhancement,3 ischemic or infarcted areas of the liver show perfusion defects on multidetector computerized tomography (MDCT). These defects can be observed immediately after transplantation or at short- and/or long-term follow-up. In this study, we classified the perfusion defects as well as determined their frequencies and impact on follow-up outcomes.
MATERIALS AND METHODS This retrospective study used data that had already been recorded in the Picture Archiving and Communication System (PACS; Infinit, Seoul, Korea). The examinations were part of follow-up for appropriate treatment. The ethical review performed by our hospital board granted permission owing to the study design. From February 4, 2004, to December 8, 2011, a total of 46 liver transplantations were performed in our hospital because of due to end-stage liver cirrhosis (Child-Pugh C; n ¼ 43), liver cirrhosis with hepatocellular carcinoma (HCC; n ¼ 1), mixed HCC and cholangiocellular carcinoma (CCC; n ¼ 1), or hepatic failure from acute hepatitis A (n ¼ 1). The CCC patient was considered to have only HCC before surgery, but was later shown to have mixed-type HCC and CCC according to pathologic examination of the explanted liver. Livers were procured from cadaveric (n ¼ 24) or living related donors (n ¼ 22). The average recipient age was 53.3 10.4 years, and the male-female ratio was 30:16. Postoperative MDCT used was either a 16-detector CT (MX 8000; Philips, Haifa, Israel) or a 256 detector CT (iCT256; Philips). The scans were performed
with dynamic sequences after the injection of the contrast media (CM): arterial phase at 30e40 seconds; portal phase 55e60 seconds: delayed phase; 2 minutes in 203 examinations (portal phase only in 46 examinations). The contrast medium was Radisense (Tae Joon Pharm Co, Seoul, Korea) with an iodine concentration of 300 mg/dL or 370 mg/dL. The CM was injected (3 mL/s) into the antecubital vein regardless of the sidedness with a power injector (Nemoto, Tokyo, Japan) using a 21-G or 18-G needle. The final follow-up ranged from 3 months 3 days to 7 years 363 days. The MX 8000’s scan parameters were pitch 1.2; collimation 1.5 mm, 16 detectors, rotation time 0.75 per rotation, and B medium filter. The reconstruction parameters were 5 mm thickness and interval algorithm. The powers of the tube were 120 kVp with automatic modulation of the mAs. The iCT used pitch 0.915, rotation time 0.4 ms, Y-sharp filter, and reconstruction interval and thickness 5 mm. The perfusion defects were defined as a lower attenuation zone or region compared with the nearby liver parenchyma or surrounding parenchyma in the portal phase. These defects were never detected in the arterial phase. Depending on the final results and their clinical impacts, we classified the perfusion defects into 4 types: chronic segmental or subsegmental benign ischemia, transient focal perfusion defects, benign subcapsular ischemia, and fatal whole liver perfusion defects, which indicated no enhancement of the entire liver parenchyma.
RESULTS
The 3 cases of chronic segmental or subsegmental benign ischemia were located in segment S8 (n ¼ 1), both S4 and S2 (n ¼ 1), or both S1 and S5 (n ¼ 1). They appeared to be perfusion defects initially, showing progressive volume loss with persistent perfusion defects upon follow-up. One patient with an S8 defect showed bile duct dilation after 2 years 9 months 25 days; this bile duct dilation persisted with whole segmental atrophy at follow-up 8 years 4 days (Fig 1). Although this patient experienced this lesion, there was
Fig 2. (A) Computerized tomography image in the portal phase 7 days after liver transplantation. A perfusion defect is noted in the anterior portion of the resection margin (arrow). (B) Computerized tomography image in the portal phase 10 months 17 days after liver transplantation. The perfusion defect noted in the resection margin has been resolved.
PERFUSION DEFECT
3185
Fig 3. (A) Computerized tomography image in the portal phase 10 days after liver transplantation. There is a subcapsular curvilinear perfusion defect with fluffy margins (arrow). Bilateral pleural effusion and ascites are noted. (B) Computerized tomography image in the portal phase 9 months 4 days after liver transplantation. The subcapsular perfusion defect has disappeared completely with clearance of the pleural effusion and ascites.
a transient liver enzyme elevation without jaundice. The remaining 2 patients showed no bile duct dilation, only a parenchyma enhancement defect. The defects persisted during follow-up which ranged from 111 months 5 days to 4 years 2 months 9 days. There were 8 focal transient defects. Although detected immediately after transplantation (Fig 2A), all of the perfusion defects disappeared at the next follow-up CT scan, which was performed from 3 days to 5 months after surgery (Fig 2B). All defects were located along the resection margins of the living related donor livers. There was 1 case of a subcapsular (right subphrenic area) curvilinear defect along the capsule below the right diaphragm (Fig 3A), which had disappeared by CT follow-up at 22 days (Fig 3B). There were 4 cases of whole liver perfusion defects detected at 1 day (n ¼ 2) or 17 days (n ¼ 1) after liver transplantation. One patient experienced a fatal outcome at 2 years 2 months after transplantation (Fig 4). All of these cases died after detection of whole liver perfusion defects.
and after 2 years in 1 case. The whole liver perfusion defect can be caused by peripheral diffuse microthrombi in the portal vein and/or hepatic arteries.4 All 4 cases of whole liver perfusion defects were detected in the portal venous phase, but not in the arterial phase. We could not determine whether there was an hepatic artery insult, because they were not detected in the arterial phase. We should have performed urgent retransplantation immediately after detection of this type of defect; however, we did not. In the case detected after a delayed period, the early postoperative follow-up CT showed no perfusion defects in the arterial or portal venous phase, but the 2-year CT showed the perfusion defect in the portal venous phase. These results suggest that periodic CT follow-up should be required in liver transplantation. In conclusion, some liver transplant cases show perfusion defects detected by CT. We suggest that benign perfusion defects need routine follow-up CT without further
DISCUSSION
CT, ultrasonography, and magnetic resonance images (MRI) show postoperative complications such as hepatic arterial, hepatic and portal venous, and bile duct complications. On the basis of our findings, we can perform necessary interventions. MRI can also provide information regarding liver regeneration and liver volume changes.5 We reviewed the CT images of liver transplant patients to observe perfusion defects. We classified 4 types of perfusion defects. No serious outcome was associated with a chronic segmental or subsegmental benign ischemic change, a focal transient perfusion defect, or a benign subcapsular ischemic change. In contrast, all subjects with a fatal whole liver perfusion defect died immediately after detection of this abnormality. Perfusion defects can be caused by ischemia or an infarct. The chronic segmental or subsegmental benign ischemia persisted, resulting in chronic hypotrophic parenchyma in 1 case that showed peripheral bile duct dilation in the involved segment. All focal transient perfusion defects disappeared after several days or months. We think that their resolution was due to collateral blood flow which required time to form in these areas.3 Benign perfusion defects need only follow-up CT examination, but not further examinations or a liver biopsy. Whole liver perfusion defects were detected in the immediate postoperative period in 3 cases,
Fig 4. Coronal reformatted computerized tomography image in the portal phase 2 years 2 months after liver transplantation. The living related donor liver shows heterogeneous whole liver perfusion defects. The recipient died 1 day after the detection of this finding.
3186
intervention, but whole liver perfusion defect demand an urgent precise diagnosis and optimal treatment of retransplantation. REFERENCES 1. Adam R, McMaster P, O’Grady JG, et al. Evaluation of liver transplantation in Europe: report of the European liver transplant registry. Liver Transpl. 2003;9:1231e1243.
SHIN, CHOI, AND KIM 2. Paik SW. Optimal timing and evaluation of liver transplantation. CMH. 2004;10:177e184. 3. Quiroga S, Sebastià C, Pallisa E, et al. Improved diagnosis of hepatic perfusion disorders: value of hepatic arterial phase imaging during helical CT. Radiographics. 2001;21:65e81. 4. Quiroga S, Sebastià C, Margarit C, et al. Complications of orthotopic liver transplantation:spectrum of findings with helical CT. Radiographics. 2001;21:1085e1102. 5. Amesur NB, Zajko AB. Interventional radiology in liver transplantation. Liver Transpl. 2006;12:330e351.
T
HE FINAL TREATMENT option for end-stage liver disease mainly associated with cirrhosis is liver transplantation using organs from living related or cadaveric donors. Using to the shortage of organs for transplantation from brain dead patients, livers from living related donors are the major source in Korea.1,2 The liver is a unique organ with a dual blood supply from the hepatic arterial and portal venous systems. In addition, there are numerous intrahepatic shunts, such as arteriovenous, arterioportal, and portovenous blood flows. Hepatic sinuses can communicate with each other when normal intrahepatic blood flow is disturbed,3 eg, by portal hypertension, inflammation, tumors, or
Budd-Chiari syndrome. These increase intravascular pressure in the involved segment to form shunts to preserve the liver.3
From the Department of Cardiovascular and Thoracic Surgery (Y.C.S.), Department of Radiology (C.S.C.), and Department of Surgery (J.S.K.), Kangdong Sacred Heart Hospital, College of Medicine, Hallym University, Seoul, Korea. Address reprint requests to Chul Soon Choi, Department of Radiology, Kangdong Sacred Heart Hospital, Kildong 445, Kangdongku, Seoul, Korea, zip code: 134-701. E-mail: [email protected]
ª 2013 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/13/$esee front matter http://dx.doi.org/10.1016/j.transproceed.2013.08.024
Transplantation Proceedings, 45, 3183e3186 (2013)
3183
3184
SHIN, CHOI, AND KIM
Fig 1. (A) Computerized tomography image in the portal phase 2 years 9 months 25 days after liver transplantation. There is a segmental perfusion defect with sharp anterior and posterior margins in segment 8, and it extends from the medial capsule to the lateral capsule of the liver (arrow). (B) Computerized tomography image in the portal phase 8 years 4 days after liver transplantation. Segment 8 is contracted, and capsular retraction is noted with bile duct dilatation (arrow).
Liver transplantation, regardless of the type of procured liver, requires anastomoses of the hepatic arteries, portal and hepatic veins and bile ducts between the transplanted liver and the host. The anastomotic sites pose risks for complications. Transplanted livers are vulnerable to ischemia or infarction after transplantation, especially regarding hepatic arterial or portal venous out flows.4 After contrast media enhancement,3 ischemic or infarcted areas of the liver show perfusion defects on multidetector computerized tomography (MDCT). These defects can be observed immediately after transplantation or at short- and/or long-term follow-up. In this study, we classified the perfusion defects as well as determined their frequencies and impact on follow-up outcomes.
MATERIALS AND METHODS This retrospective study used data that had already been recorded in the Picture Archiving and Communication System (PACS; Infinit, Seoul, Korea). The examinations were part of follow-up for appropriate treatment. The ethical review performed by our hospital board granted permission owing to the study design. From February 4, 2004, to December 8, 2011, a total of 46 liver transplantations were performed in our hospital because of due to end-stage liver cirrhosis (Child-Pugh C; n ¼ 43), liver cirrhosis with hepatocellular carcinoma (HCC; n ¼ 1), mixed HCC and cholangiocellular carcinoma (CCC; n ¼ 1), or hepatic failure from acute hepatitis A (n ¼ 1). The CCC patient was considered to have only HCC before surgery, but was later shown to have mixed-type HCC and CCC according to pathologic examination of the explanted liver. Livers were procured from cadaveric (n ¼ 24) or living related donors (n ¼ 22). The average recipient age was 53.3 10.4 years, and the male-female ratio was 30:16. Postoperative MDCT used was either a 16-detector CT (MX 8000; Philips, Haifa, Israel) or a 256 detector CT (iCT256; Philips). The scans were performed
with dynamic sequences after the injection of the contrast media (CM): arterial phase at 30e40 seconds; portal phase 55e60 seconds: delayed phase; 2 minutes in 203 examinations (portal phase only in 46 examinations). The contrast medium was Radisense (Tae Joon Pharm Co, Seoul, Korea) with an iodine concentration of 300 mg/dL or 370 mg/dL. The CM was injected (3 mL/s) into the antecubital vein regardless of the sidedness with a power injector (Nemoto, Tokyo, Japan) using a 21-G or 18-G needle. The final follow-up ranged from 3 months 3 days to 7 years 363 days. The MX 8000’s scan parameters were pitch 1.2; collimation 1.5 mm, 16 detectors, rotation time 0.75 per rotation, and B medium filter. The reconstruction parameters were 5 mm thickness and interval algorithm. The powers of the tube were 120 kVp with automatic modulation of the mAs. The iCT used pitch 0.915, rotation time 0.4 ms, Y-sharp filter, and reconstruction interval and thickness 5 mm. The perfusion defects were defined as a lower attenuation zone or region compared with the nearby liver parenchyma or surrounding parenchyma in the portal phase. These defects were never detected in the arterial phase. Depending on the final results and their clinical impacts, we classified the perfusion defects into 4 types: chronic segmental or subsegmental benign ischemia, transient focal perfusion defects, benign subcapsular ischemia, and fatal whole liver perfusion defects, which indicated no enhancement of the entire liver parenchyma.
RESULTS
The 3 cases of chronic segmental or subsegmental benign ischemia were located in segment S8 (n ¼ 1), both S4 and S2 (n ¼ 1), or both S1 and S5 (n ¼ 1). They appeared to be perfusion defects initially, showing progressive volume loss with persistent perfusion defects upon follow-up. One patient with an S8 defect showed bile duct dilation after 2 years 9 months 25 days; this bile duct dilation persisted with whole segmental atrophy at follow-up 8 years 4 days (Fig 1). Although this patient experienced this lesion, there was
Fig 2. (A) Computerized tomography image in the portal phase 7 days after liver transplantation. A perfusion defect is noted in the anterior portion of the resection margin (arrow). (B) Computerized tomography image in the portal phase 10 months 17 days after liver transplantation. The perfusion defect noted in the resection margin has been resolved.
PERFUSION DEFECT
3185
Fig 3. (A) Computerized tomography image in the portal phase 10 days after liver transplantation. There is a subcapsular curvilinear perfusion defect with fluffy margins (arrow). Bilateral pleural effusion and ascites are noted. (B) Computerized tomography image in the portal phase 9 months 4 days after liver transplantation. The subcapsular perfusion defect has disappeared completely with clearance of the pleural effusion and ascites.
a transient liver enzyme elevation without jaundice. The remaining 2 patients showed no bile duct dilation, only a parenchyma enhancement defect. The defects persisted during follow-up which ranged from 111 months 5 days to 4 years 2 months 9 days. There were 8 focal transient defects. Although detected immediately after transplantation (Fig 2A), all of the perfusion defects disappeared at the next follow-up CT scan, which was performed from 3 days to 5 months after surgery (Fig 2B). All defects were located along the resection margins of the living related donor livers. There was 1 case of a subcapsular (right subphrenic area) curvilinear defect along the capsule below the right diaphragm (Fig 3A), which had disappeared by CT follow-up at 22 days (Fig 3B). There were 4 cases of whole liver perfusion defects detected at 1 day (n ¼ 2) or 17 days (n ¼ 1) after liver transplantation. One patient experienced a fatal outcome at 2 years 2 months after transplantation (Fig 4). All of these cases died after detection of whole liver perfusion defects.
and after 2 years in 1 case. The whole liver perfusion defect can be caused by peripheral diffuse microthrombi in the portal vein and/or hepatic arteries.4 All 4 cases of whole liver perfusion defects were detected in the portal venous phase, but not in the arterial phase. We could not determine whether there was an hepatic artery insult, because they were not detected in the arterial phase. We should have performed urgent retransplantation immediately after detection of this type of defect; however, we did not. In the case detected after a delayed period, the early postoperative follow-up CT showed no perfusion defects in the arterial or portal venous phase, but the 2-year CT showed the perfusion defect in the portal venous phase. These results suggest that periodic CT follow-up should be required in liver transplantation. In conclusion, some liver transplant cases show perfusion defects detected by CT. We suggest that benign perfusion defects need routine follow-up CT without further
DISCUSSION
CT, ultrasonography, and magnetic resonance images (MRI) show postoperative complications such as hepatic arterial, hepatic and portal venous, and bile duct complications. On the basis of our findings, we can perform necessary interventions. MRI can also provide information regarding liver regeneration and liver volume changes.5 We reviewed the CT images of liver transplant patients to observe perfusion defects. We classified 4 types of perfusion defects. No serious outcome was associated with a chronic segmental or subsegmental benign ischemic change, a focal transient perfusion defect, or a benign subcapsular ischemic change. In contrast, all subjects with a fatal whole liver perfusion defect died immediately after detection of this abnormality. Perfusion defects can be caused by ischemia or an infarct. The chronic segmental or subsegmental benign ischemia persisted, resulting in chronic hypotrophic parenchyma in 1 case that showed peripheral bile duct dilation in the involved segment. All focal transient perfusion defects disappeared after several days or months. We think that their resolution was due to collateral blood flow which required time to form in these areas.3 Benign perfusion defects need only follow-up CT examination, but not further examinations or a liver biopsy. Whole liver perfusion defects were detected in the immediate postoperative period in 3 cases,
Fig 4. Coronal reformatted computerized tomography image in the portal phase 2 years 2 months after liver transplantation. The living related donor liver shows heterogeneous whole liver perfusion defects. The recipient died 1 day after the detection of this finding.
3186
intervention, but whole liver perfusion defect demand an urgent precise diagnosis and optimal treatment of retransplantation. REFERENCES 1. Adam R, McMaster P, O’Grady JG, et al. Evaluation of liver transplantation in Europe: report of the European liver transplant registry. Liver Transpl. 2003;9:1231e1243.
SHIN, CHOI, AND KIM 2. Paik SW. Optimal timing and evaluation of liver transplantation. CMH. 2004;10:177e184. 3. Quiroga S, Sebastià C, Pallisa E, et al. Improved diagnosis of hepatic perfusion disorders: value of hepatic arterial phase imaging during helical CT. Radiographics. 2001;21:65e81. 4. Quiroga S, Sebastià C, Margarit C, et al. Complications of orthotopic liver transplantation:spectrum of findings with helical CT. Radiographics. 2001;21:1085e1102. 5. Amesur NB, Zajko AB. Interventional radiology in liver transplantation. Liver Transpl. 2006;12:330e351.
