- Email: [email protected]
The role of hepatic vein Doppler in diagnosing acute rejection following paediatric liver transplantation
Clinical Radiology (1992) 45, 228-232
The Role of Hepatic Vein Doppler in Diagnosing Acute Rejection Following Paediatric Liver Transplantation P. D. BRITTON, D. J. LOMAS, R. A. COULDEN, P. F A R M A N and S. REVELL* Departments of Radiology and *Paediatrics, Addenbrooke's Hospital and University of Cambridge, Cambridge Serial Doppler ultrasound examinations of the hepatic veins were performed on 50 consecutive paediatric liver transplants. Damping of the normally pulsatile signal was observed in 23 of the 32 biopsy-proven episodes of rejection. In 10 episodes, the reduction in hepatic vein pulsatility preceded clinical and biochemical evidence of rejection by up to 36 h. Seven cases had damped signals throughout the post-operative period which precluded assessment by this method. In two patients the hepatic vein signals remained pulsatile despite rejection, one patient having unsuspected tricuspid regurgitation, and the other a stenotic IVC anastomosis. In the 35 liver transplants with normal pulsatility, hepatic vein Doppler proved to be a valuable indicator of acute rejection during the first 2 weeks following transplantation (sensitivity 92%, specificity 100%, positive predictive value 100% and negative predictive value 83%). Britton, P.D., Lomas, D.J., Coulden, R.A., Farman, P. & Revell, S. (1992). ClinicalRadiology 45, 228-232. The Role of Hepatic Vein Doppler in Diagnosing Acute Rejection Following Paediatric Liver Transplantation
Acute graft rejection following liver transplantation is a common occurrence. Early diagnosis and treatment are important as they improve overall graft function and survival (Snover, 1986). The diagnosis of rejection is suggested by derangement of the liver function tests and, in many instances, non-specific clinical features - principally pyrexia. Histological confirmation is usually required before augmenting potentially hazardous immunosuppressive therapy_ Doppler ultrasound has been used widely in liver transplantation for the detection of early vascular comphcations (Flint et al., 1988). Doppler examination of the normal hepatic veins produces a pulsatile flow signal related to changes in right atrial pressure during the cardiac cycle. Damping of the hepatic vein signal has been observed in acute and chronic liver disease, and has been attributed to reduced hepatic compliance (Rees et al., 1989; Bolondi et al., 1991). We have investigated the use of serial hepatic vein Doppler assessment in the diagnosis of early graft rejection in 41 paediatric patients following orthotopic liver transplantation. P A T I E N T S AND M E T H O D S Serial Doppler ultrasound examinations were performed on 50 consecutive liver transplants in 41 patients managed on the Paediatric Unit (20 male, 21 female). Nine patients required retransplantation following graft failure. The patients ranged in age from 8 months to 20 years (mean 5 years 7 months). In 40 operations a complete liver was grafted, in nine a left lobe only, and in one a right lobe only_ An Aloka SSD-650 Ultrasound machine with 3.5 and 5 MHz curvilinear array probes was used throughout_ Initially, examinations were performed at 2 3 day intervals following transplantation (first 18 cases). This later proved inadequate for the early detection of rejection and Correspondence to: Dr P. D. Britton, Dept. of Radiology, Addenbrooke's Hospital, Hills Road, Cambridge, CB2 2QQ.
therefore daily examinations were performed for the first week or until rejection occurred. Thereafter, twice weekly examinations were undertaken until the patient was discharged from hospital. The patients were studied in the supine position during gentle respiration or during mechanical ventilation. The standard examination involved assessment of liver texture, bile duct calibre and the perihepatic areas for fluid collections. Pulsed Doppler signals were obtained from the hepatic artery, portal vein, and the three (when present) main hepatic veins. Examination of the hepatic veins and calculation of the venous pulsatility index (VPI) were performed according to the method described by Coulden et al. (1992). The left hepatic vein signals were not analysed owing to artefacts induced by the physical motion of the heart. The most pulsatile value was taken from the measurements of the right and middle hepatic veins. The normal range of pulsatility has been established. Using the mean _+3 s.d. (99.75% of a normal distribution) the lower limit of a normal VPI is 0%. All VPIs with a positive value were considered damped. Following surgery, routine liver function tests (bilirubin, alanine transaminase, alkaline phosphatase, prothrombin and partial thromboplastin times) were checked daily. If clinical or biochemical changes suggested graft rejection then a liver biopsy was taken. If rejection was confirmed then further immunosuppressive treatment was administered. As experience accrued, liver biopsy was performed on the basis of an abnormal hepatic vein Doppler examination alone.
RESULTS A total of 447 Doppler examinations were performed (average 11 examinations per patient) for an average of 36 days following transplantation (maximum follow up 118 days). There were 32 biopsy-proven episodes of acute rejection in the 50 transplants studied_ The average time of diagnosis was the fifth post-operative day (range 2-18 days)_ Of the 32 rejection episodes, 23 (78%) were
HEPATIC VEIN DOPPLER AFTER PAEDIATRICLIVER TRANSPLANTATION
229
correctly diagnosed by detecting d a m p i n g o f the hepatic vein signal (Fig. 1). The mean value o f the V P I for d a m p e d signals was + 74% (range + 35 to + 100%). In 11 (44%) episodes, d a m p i n g o f the hepatic vein signals occurred 24 36 h before the development o f clinical or biochemical evidence o f rejection. Some patients had a d a m p e d signal on the first post-operative examination but normal pulsatility returned within 72 h. In such cases a subsequent diagnosis o f rejection could still be made. D a m p i n g o f the hepatic vein signal was also observed during episodes o f cholangitis (five patients) (Fig. 2) and cytomegalovirus (CMV) hepatitis (one patient). These infective changes were observed after the first 2 weeks_ Before this the only cause for signal d a m p i n g was rejection.
(c)
(a)
Fig. 1 - Hepatic vein signals in an 8-year-old female transplanted for Alpha l-antitrypsin deficiency. (a) Day l post-surgery the signal is normal (VPI = -56°/,,). (b) On day 5 the signal is almost completely flat (VPI = + 74%), biopsy at this time showed acute rejection. (c) On day 8, following a course of methylprednisolone, the signal returned to normal.
O f the nine cases of rejection not detected by ultrasound, seven occurred in patients--who had d a m p e d hepatic vein signals t h r o u g h o u t the post-operative period. This precluded any meaningful assessment in these cases. In the remaining two, the hepatic vein signals remained pulsatile during episodes of acute rejection_ In
(b)
Fig. 2 (a)
230
CLINICAL RADIOLOGY
(b)
(d)
(c)
(")
Fig. 2 Hepatic vein signals in an 8-year-old boy transplanted for fulminant hepatic failure. (a) On day 2 the signal is normal (VP! = - 34%). (h) On day 4 dt~ring a biopsy-proven episode of rejection the signal is completely damped (VP1 - 100%). (c) On day 12, following further anti-rejection therapy the signal has returned to normal (VPI - - 82%). (d) On day 22 during an episode of severe cholangitis following balloon dilatation ofa biliary stricture, the signal again is completely damped (VPI = 100%). (e) By day 28 the signal has returned to normal following antibiotic therapy.
HEPATIC VEIN DOPPLER AFTER PAEDIATRIC LIVER TRANSPLANTATION
Table 1 - The results using hepatic vein Doppler to diagnose rejection in the first 2 weeks following surgery in the 35 grafts with normal pulsatility in the early post-operative period
Episode of rejection
No episode of rejection
23 2
0 lO
231
vein thrombosis (n = 3), focal hepatic infarction (n = 5), and intra-abdominal fluid collections (n = 33)_ Of these, 12 were perihepatic.
DISCUSSION Damping of the hepatic vein signal No damping of the hepatic vein signal Sensitivity (92%) Specificity ( 100 % ) Positive predictive value (100%) Negative predictive value (83 %)
Fig. 3 The hepatic vein signal in a 10-month-old male during an episode of rejection. There are refluxing waves during both atrial (A) and ventricular (V) contraction. A subsequent echocardiogram revealed previously unsuspected tricuspid regurgitation.
one the hepatic vein signal was abnormally pulsatile and exhibited a large refluxing °v' wave (Fig. 3). Subsequent echocardiography revealed unsuspected right ventricular hypertrophy and tricuspid regurgitation. The second patient had abnormal venous anatomy with an absent IVC. At operation the donor IVC was anastomosed to a small recipient azygos vein_ Ultrasound showed this anastomosis to be narrowed and high velocity turbulent flow on Doppler confirmed significant stenosis. Of the 18 grafts which did not develop acute rejection, eight had persistently damped signals throughout the post-operative period. Therefore, 15 (30%) of the 50 grafts examined had consistently damped signals. In these cases, it would have been impossible to detect the change in hepatic vein waveform needed to diagnose rejection. Considering only patients with normal hepatic vein pulsatility in the first 2 post-operative weeks, sensitivity of the test is 92% and specificity 100% (Table 1). After.the first 2 weeks there were nine biopsy-proven episodes of acute rejection. Damping of the hepatic vein signal was demonstrated in only one patient on day 22 following surgery. Of the four cases of chronic rejection occurring in patients still being regularly assessed by ultrasound, none revealed a change in the hepatic vein signal. Other .clinically relevant findings observed during the study included: hepatic artery thrombosis (n = 5), portal
Acute rejection is the commonest cause of graft dysfunction in the early post-operative period (Demetris et al., 1984). The diagnosis is usually made on the basis of non-specific clinical and biochemical abnormalities followed by biopsy and definitive histological assessment. Although Doppler ultrasound is established in the diagnosis of vascular complications following liver transplantation (Wozney et al., 1986), most investigators have not found it helpful in rejection (Longley et al., 1988). However, our preliminary results suggested that serial hepatic vein Doppler may be useful (Coulden et al., 1990). This larger series supports our initial findings but also demonstrates some limitations and pitfalls. The pulsatility of the hepatic vein signal is thought to reflect the overall compliance of liver tissue. During episodes of rejection there is hepatocellular oedema and an inflammatory infiltration of the portal tracts (Snover, 1986). This expands the liver within its confining capsule, reduces compliance and results in damping of the hepatic vein signal. In 78% of rejection episodes, damping of the hepatic vein signal was demonstrated. In the first 14 days following surgery, in grafts with normal pulsatility, the only cause of signal damping was acute rejection. After this period, similar damping occurred during episodes of cholangitis and CMV hepatitis. Distinguishing these processes from rejection clinically is often difficult and liver biopsy may still be required. Even in this situation in which a change in hepatic vein signal is non-specific, it may be useful in indicating the need for early biopsy. After the first 2 weeks acute and chronic rejection failed to elicit the same changes in the hepatic vein signal demonstrated in the early post-operative period. It is unclear why this should be. It may reflect the fact that, after the initial post-operative period, multiple pathological processes may affect the transplanted liver whereas acute rejection alone is the major cause of early graft dysfunction. In our experience, a pulsatile vein signal in the presence of acute rejection is unusual, occurring in only 2/32 cases. In both cases there were underlying pathological reasons. In one, tricuspid regurgitation produced abnormally large 'v' waves resulting in hepatic vein reflux. This effect has been described as a diagnostic feature of tricuspid regurgitation (Pennestri et al_, 1984). In the second case, the IVC anastomotic stenosis may have altered the overall balance between the systemic venous circulation and right atrial pressures promoting hepatic vein reflux_ Clearly in the presence of a known right heart abnormality or deranged venous anatomy, this technique may not be reliable. In both cases it is likely that the change in hepatic compliance was obscured. Our most useful finding was that in 10 of the 23 rejection episodes diagnosed by Doppler, damping Qf t h e hepatic vein signal was the first indication of rejection, occurring before any clinical or biochemical evidence. As the response to therapy is related to the severity of rejection at the time of diagnosis (Snover, 1986) we hope that earlier detection and prompt treatment will result in improved graft function and survival.
232
CLINICAL RADIOLOGY
Aknowledgements. We should like to thank Dr N. D. Barnes (Department of Paediatrics) and Professor Sir R. Y. Calne (Department of Surgery) who were responsible for the patients in this study, for their advice and support.
REFERENCES Bolondi, L, Bassi, SL, Gaiani, S, Zironi, G, Benzi, G, Santi, V et al. (1991). Liver cirrhosis: changes of Doppler waveform of hepatic veins. Radiology, 178, 513 516. Coulden, RA, Britton, PD, Farman, P, Noble-Jamieson, G & Wight, DGD (1990). Preliminary report: hepatic vein Doppler in the early diagnosis of acute liver transplant rejection. Lancet, ii, 273 275. Coulden, RA, Lomas, DJ, Farman, P & Britton PD (1992). Doppler ultrasound of the hepatic veins: normal appearances. Clinical Radiology, 45, 223-227. Demetris, A J, Lasky, S, Van Thiel, DH, Starzl, TE & Dekker, A (1984). Pathology of hepatic transplantation. American Journal of Pathology, 118, 151 161.
Flint, EW, Su_mkin, JH, Zajko, AB & Bowen, AD (1988). Duplex sonography of hepatic artery thrombosis after liver transplantation. American Journal of Roentgenology, 151, 481--483. Longley, DG, Skolnick, ML & Sheahan, DG (1988). Acute allograft rejection in liver transplant recipients: lack of correlation with loss of hepatic artery diastolic flow. Radiology, 169, 417 420. Pennestri, F, Loperfido, F, SalvatorL MP, Mongiardo, R, Ferrazza, A, Guccione, P e t al. (1984). Assessment of tricuspid regurgitation by pulsed Doppler ultrasonography of the hepatic veins. American Journal o/' Cardiology, 54, 363-368. Rees, JIS, Shyte~ A & Cochlin, DL (1989). Doppler ultrasound in the investigation o~hepatic parenchymal disease. In: Proceedings of the 21st Meeting of the British Medical Ultrasound Society, Torquay (abstr.). British Journal of Radiology, (1990) 63, 384. Snover, DC (1986). The pathology of acute rejection. Transplant Proceedings, 18, 123 127. Wozney, P, Zajko, AB, Bron, KM, Point, S & Starzl, TE (1986). Vascular complications after liver transplantation: a 5 year experience. American Journal of Radiology, 147, 657-663.
The Role of Hepatic Vein Doppler in Diagnosing Acute Rejection Following Paediatric Liver Transplantation P. D. BRITTON, D. J. LOMAS, R. A. COULDEN, P. F A R M A N and S. REVELL* Departments of Radiology and *Paediatrics, Addenbrooke's Hospital and University of Cambridge, Cambridge Serial Doppler ultrasound examinations of the hepatic veins were performed on 50 consecutive paediatric liver transplants. Damping of the normally pulsatile signal was observed in 23 of the 32 biopsy-proven episodes of rejection. In 10 episodes, the reduction in hepatic vein pulsatility preceded clinical and biochemical evidence of rejection by up to 36 h. Seven cases had damped signals throughout the post-operative period which precluded assessment by this method. In two patients the hepatic vein signals remained pulsatile despite rejection, one patient having unsuspected tricuspid regurgitation, and the other a stenotic IVC anastomosis. In the 35 liver transplants with normal pulsatility, hepatic vein Doppler proved to be a valuable indicator of acute rejection during the first 2 weeks following transplantation (sensitivity 92%, specificity 100%, positive predictive value 100% and negative predictive value 83%). Britton, P.D., Lomas, D.J., Coulden, R.A., Farman, P. & Revell, S. (1992). ClinicalRadiology 45, 228-232. The Role of Hepatic Vein Doppler in Diagnosing Acute Rejection Following Paediatric Liver Transplantation
Acute graft rejection following liver transplantation is a common occurrence. Early diagnosis and treatment are important as they improve overall graft function and survival (Snover, 1986). The diagnosis of rejection is suggested by derangement of the liver function tests and, in many instances, non-specific clinical features - principally pyrexia. Histological confirmation is usually required before augmenting potentially hazardous immunosuppressive therapy_ Doppler ultrasound has been used widely in liver transplantation for the detection of early vascular comphcations (Flint et al., 1988). Doppler examination of the normal hepatic veins produces a pulsatile flow signal related to changes in right atrial pressure during the cardiac cycle. Damping of the hepatic vein signal has been observed in acute and chronic liver disease, and has been attributed to reduced hepatic compliance (Rees et al., 1989; Bolondi et al., 1991). We have investigated the use of serial hepatic vein Doppler assessment in the diagnosis of early graft rejection in 41 paediatric patients following orthotopic liver transplantation. P A T I E N T S AND M E T H O D S Serial Doppler ultrasound examinations were performed on 50 consecutive liver transplants in 41 patients managed on the Paediatric Unit (20 male, 21 female). Nine patients required retransplantation following graft failure. The patients ranged in age from 8 months to 20 years (mean 5 years 7 months). In 40 operations a complete liver was grafted, in nine a left lobe only, and in one a right lobe only_ An Aloka SSD-650 Ultrasound machine with 3.5 and 5 MHz curvilinear array probes was used throughout_ Initially, examinations were performed at 2 3 day intervals following transplantation (first 18 cases). This later proved inadequate for the early detection of rejection and Correspondence to: Dr P. D. Britton, Dept. of Radiology, Addenbrooke's Hospital, Hills Road, Cambridge, CB2 2QQ.
therefore daily examinations were performed for the first week or until rejection occurred. Thereafter, twice weekly examinations were undertaken until the patient was discharged from hospital. The patients were studied in the supine position during gentle respiration or during mechanical ventilation. The standard examination involved assessment of liver texture, bile duct calibre and the perihepatic areas for fluid collections. Pulsed Doppler signals were obtained from the hepatic artery, portal vein, and the three (when present) main hepatic veins. Examination of the hepatic veins and calculation of the venous pulsatility index (VPI) were performed according to the method described by Coulden et al. (1992). The left hepatic vein signals were not analysed owing to artefacts induced by the physical motion of the heart. The most pulsatile value was taken from the measurements of the right and middle hepatic veins. The normal range of pulsatility has been established. Using the mean _+3 s.d. (99.75% of a normal distribution) the lower limit of a normal VPI is 0%. All VPIs with a positive value were considered damped. Following surgery, routine liver function tests (bilirubin, alanine transaminase, alkaline phosphatase, prothrombin and partial thromboplastin times) were checked daily. If clinical or biochemical changes suggested graft rejection then a liver biopsy was taken. If rejection was confirmed then further immunosuppressive treatment was administered. As experience accrued, liver biopsy was performed on the basis of an abnormal hepatic vein Doppler examination alone.
RESULTS A total of 447 Doppler examinations were performed (average 11 examinations per patient) for an average of 36 days following transplantation (maximum follow up 118 days). There were 32 biopsy-proven episodes of acute rejection in the 50 transplants studied_ The average time of diagnosis was the fifth post-operative day (range 2-18 days)_ Of the 32 rejection episodes, 23 (78%) were
HEPATIC VEIN DOPPLER AFTER PAEDIATRICLIVER TRANSPLANTATION
229
correctly diagnosed by detecting d a m p i n g o f the hepatic vein signal (Fig. 1). The mean value o f the V P I for d a m p e d signals was + 74% (range + 35 to + 100%). In 11 (44%) episodes, d a m p i n g o f the hepatic vein signals occurred 24 36 h before the development o f clinical or biochemical evidence o f rejection. Some patients had a d a m p e d signal on the first post-operative examination but normal pulsatility returned within 72 h. In such cases a subsequent diagnosis o f rejection could still be made. D a m p i n g o f the hepatic vein signal was also observed during episodes o f cholangitis (five patients) (Fig. 2) and cytomegalovirus (CMV) hepatitis (one patient). These infective changes were observed after the first 2 weeks_ Before this the only cause for signal d a m p i n g was rejection.
(c)
(a)
Fig. 1 - Hepatic vein signals in an 8-year-old female transplanted for Alpha l-antitrypsin deficiency. (a) Day l post-surgery the signal is normal (VPI = -56°/,,). (b) On day 5 the signal is almost completely flat (VPI = + 74%), biopsy at this time showed acute rejection. (c) On day 8, following a course of methylprednisolone, the signal returned to normal.
O f the nine cases of rejection not detected by ultrasound, seven occurred in patients--who had d a m p e d hepatic vein signals t h r o u g h o u t the post-operative period. This precluded any meaningful assessment in these cases. In the remaining two, the hepatic vein signals remained pulsatile during episodes of acute rejection_ In
(b)
Fig. 2 (a)
230
CLINICAL RADIOLOGY
(b)
(d)
(c)
(")
Fig. 2 Hepatic vein signals in an 8-year-old boy transplanted for fulminant hepatic failure. (a) On day 2 the signal is normal (VP! = - 34%). (h) On day 4 dt~ring a biopsy-proven episode of rejection the signal is completely damped (VP1 - 100%). (c) On day 12, following further anti-rejection therapy the signal has returned to normal (VPI - - 82%). (d) On day 22 during an episode of severe cholangitis following balloon dilatation ofa biliary stricture, the signal again is completely damped (VPI = 100%). (e) By day 28 the signal has returned to normal following antibiotic therapy.
HEPATIC VEIN DOPPLER AFTER PAEDIATRIC LIVER TRANSPLANTATION
Table 1 - The results using hepatic vein Doppler to diagnose rejection in the first 2 weeks following surgery in the 35 grafts with normal pulsatility in the early post-operative period
Episode of rejection
No episode of rejection
23 2
0 lO
231
vein thrombosis (n = 3), focal hepatic infarction (n = 5), and intra-abdominal fluid collections (n = 33)_ Of these, 12 were perihepatic.
DISCUSSION Damping of the hepatic vein signal No damping of the hepatic vein signal Sensitivity (92%) Specificity ( 100 % ) Positive predictive value (100%) Negative predictive value (83 %)
Fig. 3 The hepatic vein signal in a 10-month-old male during an episode of rejection. There are refluxing waves during both atrial (A) and ventricular (V) contraction. A subsequent echocardiogram revealed previously unsuspected tricuspid regurgitation.
one the hepatic vein signal was abnormally pulsatile and exhibited a large refluxing °v' wave (Fig. 3). Subsequent echocardiography revealed unsuspected right ventricular hypertrophy and tricuspid regurgitation. The second patient had abnormal venous anatomy with an absent IVC. At operation the donor IVC was anastomosed to a small recipient azygos vein_ Ultrasound showed this anastomosis to be narrowed and high velocity turbulent flow on Doppler confirmed significant stenosis. Of the 18 grafts which did not develop acute rejection, eight had persistently damped signals throughout the post-operative period. Therefore, 15 (30%) of the 50 grafts examined had consistently damped signals. In these cases, it would have been impossible to detect the change in hepatic vein waveform needed to diagnose rejection. Considering only patients with normal hepatic vein pulsatility in the first 2 post-operative weeks, sensitivity of the test is 92% and specificity 100% (Table 1). After.the first 2 weeks there were nine biopsy-proven episodes of acute rejection. Damping of the hepatic vein signal was demonstrated in only one patient on day 22 following surgery. Of the four cases of chronic rejection occurring in patients still being regularly assessed by ultrasound, none revealed a change in the hepatic vein signal. Other .clinically relevant findings observed during the study included: hepatic artery thrombosis (n = 5), portal
Acute rejection is the commonest cause of graft dysfunction in the early post-operative period (Demetris et al., 1984). The diagnosis is usually made on the basis of non-specific clinical and biochemical abnormalities followed by biopsy and definitive histological assessment. Although Doppler ultrasound is established in the diagnosis of vascular complications following liver transplantation (Wozney et al., 1986), most investigators have not found it helpful in rejection (Longley et al., 1988). However, our preliminary results suggested that serial hepatic vein Doppler may be useful (Coulden et al., 1990). This larger series supports our initial findings but also demonstrates some limitations and pitfalls. The pulsatility of the hepatic vein signal is thought to reflect the overall compliance of liver tissue. During episodes of rejection there is hepatocellular oedema and an inflammatory infiltration of the portal tracts (Snover, 1986). This expands the liver within its confining capsule, reduces compliance and results in damping of the hepatic vein signal. In 78% of rejection episodes, damping of the hepatic vein signal was demonstrated. In the first 14 days following surgery, in grafts with normal pulsatility, the only cause of signal damping was acute rejection. After this period, similar damping occurred during episodes of cholangitis and CMV hepatitis. Distinguishing these processes from rejection clinically is often difficult and liver biopsy may still be required. Even in this situation in which a change in hepatic vein signal is non-specific, it may be useful in indicating the need for early biopsy. After the first 2 weeks acute and chronic rejection failed to elicit the same changes in the hepatic vein signal demonstrated in the early post-operative period. It is unclear why this should be. It may reflect the fact that, after the initial post-operative period, multiple pathological processes may affect the transplanted liver whereas acute rejection alone is the major cause of early graft dysfunction. In our experience, a pulsatile vein signal in the presence of acute rejection is unusual, occurring in only 2/32 cases. In both cases there were underlying pathological reasons. In one, tricuspid regurgitation produced abnormally large 'v' waves resulting in hepatic vein reflux. This effect has been described as a diagnostic feature of tricuspid regurgitation (Pennestri et al_, 1984). In the second case, the IVC anastomotic stenosis may have altered the overall balance between the systemic venous circulation and right atrial pressures promoting hepatic vein reflux_ Clearly in the presence of a known right heart abnormality or deranged venous anatomy, this technique may not be reliable. In both cases it is likely that the change in hepatic compliance was obscured. Our most useful finding was that in 10 of the 23 rejection episodes diagnosed by Doppler, damping Qf t h e hepatic vein signal was the first indication of rejection, occurring before any clinical or biochemical evidence. As the response to therapy is related to the severity of rejection at the time of diagnosis (Snover, 1986) we hope that earlier detection and prompt treatment will result in improved graft function and survival.
232
CLINICAL RADIOLOGY
Aknowledgements. We should like to thank Dr N. D. Barnes (Department of Paediatrics) and Professor Sir R. Y. Calne (Department of Surgery) who were responsible for the patients in this study, for their advice and support.
REFERENCES Bolondi, L, Bassi, SL, Gaiani, S, Zironi, G, Benzi, G, Santi, V et al. (1991). Liver cirrhosis: changes of Doppler waveform of hepatic veins. Radiology, 178, 513 516. Coulden, RA, Britton, PD, Farman, P, Noble-Jamieson, G & Wight, DGD (1990). Preliminary report: hepatic vein Doppler in the early diagnosis of acute liver transplant rejection. Lancet, ii, 273 275. Coulden, RA, Lomas, DJ, Farman, P & Britton PD (1992). Doppler ultrasound of the hepatic veins: normal appearances. Clinical Radiology, 45, 223-227. Demetris, A J, Lasky, S, Van Thiel, DH, Starzl, TE & Dekker, A (1984). Pathology of hepatic transplantation. American Journal of Pathology, 118, 151 161.
Flint, EW, Su_mkin, JH, Zajko, AB & Bowen, AD (1988). Duplex sonography of hepatic artery thrombosis after liver transplantation. American Journal of Roentgenology, 151, 481--483. Longley, DG, Skolnick, ML & Sheahan, DG (1988). Acute allograft rejection in liver transplant recipients: lack of correlation with loss of hepatic artery diastolic flow. Radiology, 169, 417 420. Pennestri, F, Loperfido, F, SalvatorL MP, Mongiardo, R, Ferrazza, A, Guccione, P e t al. (1984). Assessment of tricuspid regurgitation by pulsed Doppler ultrasonography of the hepatic veins. American Journal o/' Cardiology, 54, 363-368. Rees, JIS, Shyte~ A & Cochlin, DL (1989). Doppler ultrasound in the investigation o~hepatic parenchymal disease. In: Proceedings of the 21st Meeting of the British Medical Ultrasound Society, Torquay (abstr.). British Journal of Radiology, (1990) 63, 384. Snover, DC (1986). The pathology of acute rejection. Transplant Proceedings, 18, 123 127. Wozney, P, Zajko, AB, Bron, KM, Point, S & Starzl, TE (1986). Vascular complications after liver transplantation: a 5 year experience. American Journal of Radiology, 147, 657-663.