Hepatic Artery Angioplasty after Liver Transplantation: Experience in 21 Allografts

Hepatic Artery Angioplasty after Liver Transplantation: Experience in 21 Allografts1 Philip D. Orons, DO Albert B. Zajko, MD Klaus M. Bron, MD Gregory T. Trecha, MD Robert R. Selby, MD John J. Fung, MD Index terms: Arteries, transluminal angi0plasty, 952.1282 Hepatic arteries, stenosis or obstruction, 952.458 Hepatic arteries, thrombosis, 952,458. Liver, transplantation, 761.458

.

JVIR 1995;6:523-529 Abbreviations: ALT = alanine aminotransferase, AST = aspartate aminotransaminase, HAS = hepatic artery steOLT = orthotopic liver transplantstion, PTA = percutaneous transluminal angioplasty

PURPOSE: To assess whether percutaneous transluminal angioplasty (PTA)can help prolong allograft survival and improve allograft function in patients with hepatic artery stenosis after liver transplantation. PATIENTS AND METHODS: Hepatic artery PTA was attempted in 19 patients with 21 allografts over 12 years. The postangioplasty clinical course was retrospectively analyzed. Liver enzyme levels were measured before and after PTA to determine if changes in liver function occurred after successful PTA. RESULTS: Technical success was achieved in 17 allografts (81%). Retransplantation was required for four of 17 allografts (24%)in which PTA was successful and four of four allografts in which PTA was unsuccessful; this difference was significant (P= .03). Two major procedure-relatedcomplications occurred: an arterial leak that required surgical repair and an extensive dissection that necessitated retransplantation 14 months after PTA. Hepatic failure necessitated repeat transplantation in seven cases from 2 weeks to 27 months (mean, 8.4 months) after PTA. Six patients died during follow-up, three of whom had undergone repeat transplantation. Markedly elevated liver enzyme levels at presentation were associated with an increased risk of retransplantation or death regardless of the outcome of PTA. CONCLUSION: PTA of hepatic artery stenosis after liver transplantation is relatively safe and may help decrease allograft loss due to thrombosis. Marked allograft dysfunction at presentation is a poor prognostic sign; thus, timely intervention is important.

VMCULAR mmplications remain a

'Fromthe Department ofRadiolog~,Divi-

sionofvascularflnterventionalRadiolo~~ (P.D.O.,A.B.Z.,K.M.B.,G.T.T.),andthe Department of Surgery, Division of Organ Transplantation (R.R.s., J.J.F.), University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213. From the 1994 SCVIR annual meeting. Received July 15, 1994; revision requested September 15; revision received January l , 1995; accepted January 9. Address reprint requests to P.D.O. SCVIR, 1995

major cause of morbidity and mortality after orthotopic liver transplantation (OLT). The most common vascular complication reported is hepatic artery thrombosis. Commonly seen in children, especially those younger than 5 years of age, hepatic artery thrombosis is a devastating complication that occurs in 4%-25% of transplant recipients (1--4). A less commonly described but potentially important cause of morbidity is hepatic artery stenosis ( U S ) ,which reportedly occurs in 6%-13% of transplant recipients (3,5). Recent experience with, and advancements in, duplex sonogra-

phy indicate the actual prevalence of HAS may be considerably higher (6). Patients with HAS who are symptomatic have traditionally been treated with anticoagulation therapy, revascularization, or retransplantation, if indicated (4,7). However, authors of several case reports and one small series have described promising clinical responses with percutaneous transluminal angioplasty (PTA) (8-12). We report our experience since 1982 with PTA of hepatic artery stenoses in 21 allografts. The purposes of our study were to determine whether successful hepatic artery PTA can help prolong allograft survival and to deter-

523

Journal of Vascular and Interventional Radiology

524

July-August 1995

Table 1 Summary of Results and Clinical Follow-up of Hepatic Artery PTA in 21 Allografts Allograft No.

Time Interval Arterial after OLT (mo) Anastomosis

Result of PTA

Followup (mo)

1 2

1 9

A-A A-A

S U

43 26

3

3

IRG

S

6

4 5 6 7 8

4 1 3 0.75 3

A-A IRG A-A A-A IRG

S S S S S

28 3 36 22 3

9 10 11

0.75 6 2

A-A A-A A-A

S S S

21 31 7

12 13 14

9 6 13

A-A A-A A-A

S U U

8 6 20

15

2

IRG

S

14

16 17

5 0.5

A-A A-A

S S

22

18 19 20 21

44 2 1.5 5

IRG A-A A-A IRG

S U S S

17 0.5 0.75 15

4

Clinical Outcome Stable Retransplantation because of ischemia, portal vein thrombosis Leak after PTA that required surgical repair*; patient died of recurrent hepatitis, sepsis, gastrointestinal hemorrhage Stable Died of sepsis, renal failure Stable Stable Retransplantation because of severe ischemic injury, abscess, infarct; patient died Stable Stable Died of multisystem organ failure, sepsis, pulmonary hemorrhage Stable Retransplantation because of rejection; patient died Retransplantation because of hepatic necrosis; patient died Retransplantation because of hepatic artery pseudoaneurysm related to extensive dissection after PTA, liver function stable Stable Retransplantation because of rejection (hepatic artery patent) Stable Retransplantation because of hepatic necrosis Retransplantation because of rejection Stable

Note.-A-A = direct hepatic artery-to-hepatic artery anastomosis, IRG = infrarenal graft; S = successful, U = unsuccessful. *Underwent concomitant arterial revascularization, biliary reconstruction, and drainage of abdominal abscess.

mine whether PTA affects allograft function as determined from liver enzyme levels. PATIENTS AND METHODS

Hepatic artery PTA was attempted in 19 patients (10 female, nine male; age range, 15 months to 75 years; mean age, 45 years) with 21 allografts a t our institution from 1982 to 1994. Seventeen of these procedures were attempted between 1990 and 1994. We retrospectively reviewed all angiographic examinations and the clinical findings after PTA in each patient, and noted whether repeat liver transplantation

was eventually necessary. The indication for retransplantation was determined from clinical and pathologic records. Patients were referred for initial angiographic examination on the basis of abnormal liver enzyme levels, biopsy results, or the results of Doppler ultrasound (US) of the hepatic artery. In most cases since 1990, HAS was suspected due to abnormal findings at Doppler US (6). Long-term follow-up information was obtained by means of review of patient records and telephone interviews with patients when records were believed to be incomplete. Liver enzyme levels were evaluated before and after PTA to determine whether changes or trends in liver

function occurred in response to successful PTA. Direct hepatic artery-to-hepatic artery anastomoses were present in 15 allografts, and the remaining six allografts were revascularized with infrarenal iliac artery homografts. One of the patients with a direct arterial anastomosis also had a replaced right hepatic artery that was anastomosed to the splenic artery stump. At angiography, stenosis was considered to be clinically relevant if the lumen diameter was decreased by more than 50%. Angiography was performed with standard catheter techniques from a transfemoral approach with conventional or digital

Orons et a1

525

Volume 6 Number 4

b. Figure 1. (a) Selective arteriogram of the iliac artery homograft obtained in a 56-year-old man 5 months after OLT demona.

strates an anastomotic stenosis (arrow). The patient had elevated y-glutamyl transferase and bilirubin levels and evidence of HAS a t US. Systolic and mean transstenotic arterial pressure gradients were 30 and 16 mm Hg, respectively. (b) After PTA with a 7-mm balloon, a normal lumen was restored and the systolic gradient was reduced to 4 mm Hg with complete ablation of the mean gradient. This allograft was later lost to rejection.

subtraction imaging. PTA was most often performed with 5-F balloon catheters (Medi-tech/Boston Scientific, Watertown, Mass). Balloon size was determined by means of direct measurement of the patent portion of the hepatic artery on the conventional arteriogram. In all patients, heparin was systemically administered during the procedure. In several cases, low-profile balloons (Sub4; Medi-techBoston Scientific) were used because a 5-F balloon could not be advanced through the stenosis. In most cases since 1990, a 0.035-inch angled hydrophilic guide wire, either the standard or the long-taper type (Medi-techBoston Scientific) has been used to traverse the stenosis. Stiffer wires, such as heavy-duty exchange or TAD wires (Torque-Attenuating Diameter; Mallinckrodt Medical, St Louis, Mo) were often used for passage of the angioplasty balloon. PTA was considered successful if a normal or near-normal lumen diameter was restored (<20%residual stenosis). Nitroglycerine was admin-

istered intraarterially in patients in whom evidence of arterial spasm had been seen after PTA. Coaxial balloon systems were not used in patients in whom stenoses could not be crossed with conventional balloon catheters. Anticoagulation was not routinely maintained after angioplasty. Several variables were subjected to statistical analysis. Foremost, allograft survival time and time between PTA attempt and retransplantation were compared for patients who underwent successful versus unsuccessful PTA; the Fisher exact test was used. Pathologic findings in patients who underwent retransplantation were compared by means of the Fisher exact test to determine whether patients who underwent unsuccessful PTA had a higher rate of ischemic hepatic necrosis than patients who underwent successful PTA in whom retransplantation was required. In addition, liver function at the time of PTA was compared in those patients whose condition remained stable

during follow-up and those who died or required repeated liver transplantation; an analysis of variance for difference in means was used.

RESULTS Angioplasty was attempted a mean of 3.8 months after OLT (range, 12 days to 44 months). HAS was seen in 17 of 21 allografts (81%) within 6 months of OLT. Stenoses were anastomotic in 18 allografts (86%),intrahepatic in two (9.5%), and both anastomotic and intrahepatic in one (5%)(Table 1). Initial technical success was achieved in 17 allografts (81%)(Fig 1). In four cases (19%),all of which were direct arterial anastomoses, PTA could not be performed because the hepatic artery was tortuous or redundant, which prevented selective transstenotic catheter placement. Two major procedure-related complications occurred. One patient who underwent PTA approximately 2 months after OLT developed a dis-

Journal of Vascular and Interventional Radiology

526

July-August 1995

Table 2 Diagnoses in Patients with Failed Allografts in Whom Repeated OLT Was Required Allograft No.

Indication Unsuccessful PTA

2 13 14 19

Ischemic necrosis Rejection Ischemic necrosis Ischemic necrosis Successful PTA

8 15 17 20

Ischemic necrosis Hepatic artery pseudoaneurysm Rejection Rejection

Note.-For unsuccessful versus successful PTA, ischemic necrosis was present in 75% versus 25% of cases, respectively ( P = .5).

section that extended from the anastomosis into the hepatic artery bifurcation. At angiography, arterial flow appeared to remain satisfactory, and follow-up Doppler US showed no flow abnormalities. No immediate treatment was rendered. This patient underwent follow-up US examination 3 and 6 months after PTA, and both examinations were normal. However, during a routine US examination 14 months after PTA, a large pseudoaneurysm was identified. The pseudoaneurysm could not be repaired percutaneously or surgically. Although the patient was asymptomatic and had excellent allograft function, the risk of aneurysm rupture was believed to be high, and the patient underwent elective retransplantation. Another patient who underwent PTA approximately 3 months after OLT develo~edan arterial leak and hernoperitoneum, which necessitated surgical repair (Fig 2). Length of clinical follow-up ranged from 2 weeks to 43 months (mean, 16 months). Allograft survival was significantly better in the successful PTA group than in the

Figure 2. (a) Selective hepatic arteriogram obtained in a 46-year-old man 3 months after OLT demonstrates a severe anastomotic stenosis (arrow). Also shown is segmental occlusion of the right hepatic artery (arrowheads). There was marked improvement of the lumen after dilation with a 6-mm balloon and no immediate complications. However, the patient subsequently developed hemobilia and gastrointestinal bleeding. (b) Repeated arteriogram obtained several hours after PTA shows extravasation (arrows) from the hepatic artery adjacent to the site of previous stenosis. The patient was treated with surgical repair. Drainage of a n abdominal abscess adjacent to the hepatic artery in this patient also may have contributed to vessel disruption with PTA.

Table 3 Causes of Death after Attempted Hepatic Artery PTA Allograft No.

Result of PTA

Repeated OLT

3

S

No

5 8 11

S S S

No Yes No

13 14

U U

Yes Yes

Note.-S

Cause of Death Recurrent hepatitis, sepsis, gastrointestinal hemorrhage 6 months after PTA Sepsis, renal failure 3 months after PTA Fungal sepsis 3 months after PTA Pulmonary hemorrhage and pulmonary edema superimposed on multisystem organ failure and sepsis; hepatic artery patent a t autopsy Sepsis, multisystem organ failure, pneumonia Hepatic necrosis, gastric hemorrhage after third liver transplantation

= successful, U = unsuccessful.

unsuccessful PTA group (P = .03). In addition to the retransplantation for hepatic artery pseudoaneurysm, retransplantation was performed for seven allografts that failed; thus, retransplantation was required in a total of eight allografts (38%)(Table 2). These eight allografts included

the four allografts in which PTA could not be performed, as well as four of the 17 allografts (24%) in which PTA was successful. Prior PTA did not appear to make retransplantation technically more difficult. The successfully treated patients who had to undergo retrans-

Orons et a1

527

Volume 6 Number 4

Table 5 Mean Liver Enzyme Levels at Presentation

AST

ALT

ALP

GGTP

BILI

Group

(<40 IUIL)

(<40 IUL)

(40-125 IUIL)

(<65 IUIL)

(0.3-1.5 mg1dL)

Patients who died or required transplantation Patients who remained stable P value

188 42 .001

161 79 .06

758 326 .05

1,131 349 .003

4.3 1.1 .02

Note.-The normal range of values is given in parentheses. ALP = alkaline phosphatase, ALT = alanine aminotransferase, AST = aspartate aminotransferase, BILI = serum bilirubin, GGTP = y-glutamyl transferase.

Table 4 Changes in Liver Enzyme Levels after Successful PTA

Finding

AST

ALT

ALP

GGTP

BILI

2 2 3 5 4 No. of allografts with increase >25% No. of allografts with decrease >25% 4 7 1 3 2 No. of allografts with change <25% 12 8 8 8 10 Note.-ALP = alkaline phosphatase, ALT = alanine aminotransferase, AST = aspartate aminotransferase, BILI = total serum bilirubin, GGTP = y-glutamyl transferase. All values not available for all allografts.

plantation (excluding the patient who underwent retransplantation for hepatic artery pseudoaneurysm) underwent the procedure a mean of 1.9 months after PTA, compared with a mean of 13.4 months among the unsuccessfully treated patients ( P = .16). There has been no clinical evidence of late hepatic artery thrombosis in patients who have undergone successful PTA. Although the majority of patients underwent Doppler US examination soon after PTA, most patients have not undergone sonographic or angiographic examination during extended followup, and the presumption of maintained hepatic artery patency is based on clinical evaluation. Six patients died (four patients with successful PTA, two patients with unsuccessful PTA), three of whom had undergone retransplantation (Table 3). No patient death was directly related to PTA. Liver function was evaluated both a t the time of presentation and 1-3 weeks after PTA to detect changes of greater than 25% in liver enzyme levels in response to successful PTA.

Liver enzyme levels a t presentation were available in 20 allografts, 19 (25%) of which displayed abnormalities in liver function as determined from either serum aspartate aminotransaminase (AST), alanine aminotransferase (ALT), alkaline phosphatase, y-glutamyl transferase, or bilirubin levels. No consistent changes in liver function (either improvement or decline) occurred within 1-3 weeks after PTA (Table 4). However, baseline liver function in those patients who eventually required retransplantation or died was substantially worse than in those who remained stable after PTA (Table 5).

DISCUSSION The importance of hepatic artery patency after OLT has been well described (1-5,7). HAS may cause graft ischemia and dysfunction. If arterial thrombosis develops from HAS, then sepsis, abscess formation, bile duct strictures, bile leak, hepatic infarction, and allograft failure may ensue (3,7). Such complications

usually require retransplantation. Case reports in which PTA has been used to treat posttransplantation HAS have been encouraging in regard to the effect of PTA on allograft function and survival (8-10,12). Abad et a1 (8) described two cases of successful PTA of HAS; in both of these cases, the patients presented with elevated liver enzyme levels, and one had histologic evidence of hepatic necrosis. Normal liver enzyme levels after PTA were reported in both patients, and no further evidence of necrosis was seen a t later biopsy in the patient with necrosis. Castaneda et a1 (9) reported a similar recovery of hepatic function in a child whose liver biopsy before PTA was remarkable for ischemic changes. Raby et a1 (10) reported successful hepatic artery PTA that resulted in improvement in liver function in three patients. In two of these patients, however, repeat dilation was required to treat recurrent arterial stenoses (10). In a recently reported study by Mondragon et a1 (121, six of seven patients presented with clinical abnormalities attributed to HAS (three patients had bile leak and three patients had allograft dysfunction). Five of these patients, including the three with bile leak, exhibited marked improvement in AST levels within 1week after PTA. However, all patients with bile leaks developed biliary strictures, and two patients eventually required surgical biliary revision. Presumably, the reestablishment of a normal arterial lumen would restore normal blood flow to previously ischemic hepatic tissue and

528

Journal of Vascular and Interventional Radiology

July-August 1995

would result in improved allograft function. It seems intuitive, based on our knowledge of the natural history of stenoses in the peripheral vascular system, that many hepatic artery thromboses begin as stenoses. The treatment of HAS may therefore help prevent hepatic artery thrombosis. However, no randomized study has been performed to confirm or deny this assertion, and the performance of such a prospective study may raise serious ethical considerations. PTA was successful in restoring a normal or near-normal lumen in 17 of 21 allografts (81%).Had coaxial catheter systems been used, it might have been possible to accomplish successful PTA in patients with excessively tortuous arteries. Three of the four unsuccessful procedures were performed before 1991, when coaxial balloon systems were not in use at our institution. Transstenotic pressure measurements were not routinely obtained, and these measurements might have been helpful in the objective evaluation of hemodynamic significance. In more than 75% of stenoses, lumen narrowing of greater than 80% was demonstrated; these stenoses were believed to be hemodynamically significant based on their angiographic appearance. Comparison of the clinical outcome in patients with successful PTA versus that in patients with unsuccessful PTA revealed that allograft survival was significantly better in the successful PTA group (P = .03). Three of the four patients in the unsuccessful PTA group had to undergo retransplantation for ischemic necrosis, but only one patient in the successful PTA group experienced ischemic allograft loss (a patient who had hepatic ischemia and multiple hepatic abscesses for several weeks before PTA); however, this disparity was not statistically significant (P > .5) (Table 2). The mean time interval between the PTA attempt and retransplantation seemed to vary between the successful and unsuccessful PTA

groups (1.9 and 13.4 months, respectively); but this difference also was not statistically significant (P= .16). The exact cause of this disparity was unclear. Patients in the unsuccessful group appeared to have undergone retransplantation in a more elective fashion, but liver function within this group was generally poor during that period. However, when baseline liver function was evaluated. it became evident that poor liver function at presentation was an ominous prognostic sign, regardless of the success or failure of PTA. Also, in contradistinction to previous studies (8-10, 12), as a group, our patients demonstrated no consistent improvement in liver enzyme or bilirubin levels during follow-up after PTA. Mondragon et a1 (12) reported improved liver function in their patients within 1week after PTA. but no consistent liver enzyme response was demonstrated in our patients (either improvement or decline) (Table 4). Because the cause of allograft dysfunction is frequently multifactorial, a causal relationship between individual factors (ie, HAS, allograft rejection, sepsis) and distinct patterns of liver enzyme level elevation are often difficult or impossible to establish with certainty. When hepatic artery thrombosis occurs, rapid clinical deterioration is usually seen at presentation accompanied by a precipitous rise in the levels of transaminases (AST, ALT) and bilirubin (13). Biliary tract abnormalities (ie, stricture, obstruction, necrosis) and resultant biliary stasis are frequently associated with elevation of alkaline phosphatase and y-glutamyl transferase levels (13). However, no pattern of liver enzyme elevation has been ascribed specifically to liver dysfunction secondary to HAS because HAS may be associated with parenchymal ischemia as well as resultant biliary tract abnormalities. The small sample size in this study made statistical analysis of liver enzyme levels difficult. The study was carried out with an analy-

sis of variance for difference in t ~seudoanmeans. The ~ a t i e nwith eurysm formation was excluded from this analvsis because her liver function was not a factor in the need for retrans~lantation.When the liver function in patients who died or eventually required retransplantation was compared with that in the patients who remained stable, a significant disparity in baseline liver function was found. With the exception of ALT (borderline significance, P = .06). all liver enzvme levels were significantly higher fn the group of patients who died or required retransplantation (Table 5); as a group, those with poor long-term outcomes had markedly higher liver enzyme levels at time of presentation. Although allograft rejection may have contributed to this effect, these findings may indicate that biliary tract abnormalities or irreversible ischemic injury had already occurred in these patients by the time PTA was performed. Previous studies have shown that up to 84% of patients with hepatic artery thrombosis may present with cholangiographic abnormalities (51, but no data have been published regarding the prevalence of biliary tract disease associated with HAS. We have recently reviewed our experience with HAS and associated cholangiographic abnormalities in 33 patients. Each patient had angiographically proved HAS and a temporally associated diagnostic cholangiogram. Cholangiographic abnormalities were demonstrated in more than 60% of these patients. Unfortunately, in the present retrospective PTA study, a sufficient number of cholangiographic studies were not available to help confirm or deny the possibility of associated cholangiographic abnormalities in the study population. Also, liver biopsy specimens were not available in enough patients to allow a comprehensive analysis for the presence of coexistent rejection at the time of PTA. PTA of the hepatic artery appears to be associated with the same com-

Orons et a1

529

Volume 6 Number 4

plications inherent in balloon dilation of other visceral arteries. Previous studies on complications of allograft renal artery PTA have shown acceptable perioperative complication rates of 5%-9% (14-16). In our group, major complications occurred in two of 21 PTA attempts (9.5%), both of which were infrarenal grafts. In one case surgical repair was required, and in the other case retransplantation eventually was required a s a consequence of PTA complication. The patient who developed arterial leak may have been predisposed to arterial disruption because of the presence of a large abscess that surrounded the arterial graft and may have compromised the integrity of the vessel; this abscess was drained during surgical repair of the hepatic artery. This patient had to undergo concomitant biliary revision a s well to treat biliary stricture and eventually died. The cause of the extensive arterial dissection in the other patient was unclear. I t is possible that the dissection occurred when a guide wire was passed through the stenosis, but no dissection was recognized a t that time. Although we believe the safety of PTA is acceptable, the potential for major complication exists, and use of the procedure should be carefully considered. We recommend that arterial dilation not be attempted until several weeks after transplantation to help avoid potential rupture of the suture line, although we have performed PTA a s early a s 12 days after transplantation without complication. One recognized shortcoming of this study is the absence of objective long-term follow-up information (ie, hepatic angiographic or Doppler US findings) to help confirm t h a t patency was maintained in the hepatic artery in most patients in the successful PTA group. Although we could not definitively evaluate the presence of recurrent HAS, we be-

lieve the catastrophic results that are virtually always associated with hepatic artery thrombosis would have been clinically recognizable in this population; therefore, we believe hepatic artery thrombosis was avoided in patients who underwent successful PTA. In conclusion, our data indicate that PTA for treatment of HAS after OLT is relatively safe; when PTA is successful, i t may prolong allograft survival. The risk of allograft loss secondary to subsequent hepatic artery thrombosis may be lessened with early intervention, but further research is indicated. Timely intervention is extremely important, a s poor liver function a t presentation is a n ominous prognostic sign, regardless of the outcome of the PTA attempt.

Acknowledgment: The authors thank Marcia Kurs-Lasky for assistance with statistical analysis. References 1. Langnas AN, Marujo W, Stratta RJ, Wood RP, Shaw BW Jr. Vascular complications after liver transplantation. Am J Surg 1991; 161:76-83. 2. Blumhardt G, Ringe B, Lauchart W, Burdelski M, Bechstein WO, Pichlmayr R. Vascular problems in liver transplantation. Transplant Proc 1987; 19:2412. 3. Wozney P, Zajko AB, Bron KB, Point S, Starzl TE. Vascular complications after liver transplantation: a five year experience. AJR 1986; 147:657463. 4. Merion RM, Burtch GD, Ham JM, Turcotte JG, Campbell DA Jr. The hepatic artery in liver transplantation. Transplantation 1989; 48:438443. 5. Zajko AB, Campbell WL, Logsdon GA, et al. Cholangiographic findings in hepatic artery occlusion after liver transplantation. AJR 1987; 149:485-489. 6. Dodd GD, 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. Hesselink EJ, Slooff MJH, Schuur KH, Bijleveld C, Gips C. Consequences of hepatic artery pathology after orthotopic liver transplantation. Transplant Proc 1987; 19:2476-2477. Abad J, Hidalgo EG, Cantarero JM, et al. Hepatic artery anastomotic stenosis after transplantation: treatment with percutaneous transluminal angioplasty. Radiology 1989; 171:661-662. Castaneda F, So SKS, Hunter DW, Castaneda-Zuniga WR. Reversible hepatic transplant ischemia: case report and review of literature. Cardiovasc Intervent Radio1 1990; 13188-90. Raby N, Karani J, Thomas S, O'Grady J, Williams R. Stenoses of vascular anastomoses after hepatic transplantation: treatment with balloon angioplasty. AJR 1991; 157:167-171. Zajko AB, Bron KM, Starzl TE, et al. Angiography of liver transplant patients. Radiology 1985; 157:305311. Mondragon RS, Karani JB, Heaton ND, et al. The use of percutaneous transluminal angioplasty in hepatic artery stenosis after transplantation. Transplantation 1994; 57:228231. Sher LS, Pan SH, Hoffman AL, et al. Liver transplantation. In: Makowka L, ed. The handbook of transplantation management. Austin, Texas: R.G. Landes, 1991; 192-253. McMullin ND, Reidy JF, Koffman CG, et al. The management of renal transplant artery stenosis in children by percutaneous transluminal angioplasty. Transplantation 1992; 53:559-563. Curry NS, Cochran S, Barbaric ZL, et al. Interventional radiologic procedures in the renal transplant. Radiology 1984; 152:647-653. Raynaud A, Bedrossian J, Remy P, Brisset JM, Angel CY, Gaux JC. Percutaneous transluminal angioplasty of renal transplant arterial stenoses. AJR 1986; 1462353-857.