- Email: [email protected]
Liver Transplantation for Autoimmune Hepatitis
Liver Transplantation for Autoimmune Hepatitis DAVID J. REICH,1 ISABEL FIEL,2 JAMES V. GUARRERA,3 SUKRU EMRE,4 STEPHEN R. GUY,5 MYRON E. SCHWARTZ,4 CHARLES M. MILLER,4 AND PATRICIA A. SHEINER4
SEE EDITORIAL ON PAGE 868
studies could elucidate the best posttransplantation immunosuppressive regimens for AIH patients. (HEPATOLOGY 2000;32:693-700.)
Publications about liver transplantation (LTX) for autoimmune hepatitis (AIH) have started to emerge, but many issues remain unresolved. We reviewed data on 32 patients transplanted for AIH to determine how pretransplantation and posttransplantation characteristics correlate with recipient outcome, including disease recurrence. Recipients were 37ⴞ 14 years old; 30 of 32 were women. Most had chronic disease (8 ⴞ 6 years); 25% had fulminant failure. The majority had ascites (91%), jaundice (88%), elevated prothrombin time (18 ⴞ 3 seconds), and hypoalbuminemia (2.7 ⴞ 0.6 g/dL). All had hypergammaglobulinemia (3.0 ⴞ 1.0 g/dL) and autoantibodies (72% antinuclear, 74% smooth muscle). Only one was HLA A1-B8-DR3 positive. Other autoimmune disorders affected 25% of patients; half improved after transplantation. Actuarial survival was 81% at 1 and 2 years posttransplantation. There was a high frequency of rejection (75% of recipients had 1.7 ⴞ 0.8 episodes), and 39% of rejections required OKT3. Among 24 recipients with long-term follow-up (27 ⴞ 14 months), histologically proven recurrent AIH occurred in 25%, 15 ⴞ 2 months posttransplantation; half (3 patients) required retransplantation 11 ⴞ 3 months after diagnosis. After retransplantation 2 of 3 patients had re-recurrence within 3 months; 1 received a third LTx. Recurrence occurred in 6 of 18 patients transplanted for chronic disease vs. 0 of 6 transplanted as fulminants (P ⴝ not significant [NS]). Patients with and without recurrence had similar rejection profiles. In summary, results of LTx for AIH are excellent. However, AIH patients have a high frequency of rejection and often require OKT3. Furthermore, severe recurrent AIH sometimes develops, particularly in chronic versus fulminant AIH patients and in those already retransplanted for recurrence. Multicenter
Liver transplantation (LTX) is perhaps the most dramatic advance in the management of autoimmune hepatitis (AIH) since the earliest description of this disease by Waldenstrom in 1950.1 Although immunosuppression remains the mainstay of therapy for patients with AIH, 13% to 20% deteriorate on traditional regimens, and of those who experience remission and discontinue therapy, 74% to 85% relapse.2,3 Medical therapy does not cure AIH. Therefore, many patients eventually develop end-stage liver disease and become candidates for LTX. Until a short time ago the literature on LTX for AIH was sparse, including one publication that addressed the role of LTX for AIH,2 and some reports about recurrence of AIH after LTX.4-7 Recently, several retrospective reviews on LTX for AIH have emerged from various transplantation programs8-12 as well as a few publications about rejection and immunosuppression after LTX for AIH.13-15 In general there is consensus about the demographics of AIH patients that come to LTX. However, there are unresolved issues regarding the posttransplantation course of these patients. For example, there is controversy about the prevalence, risk factors, and outcome of recurrent AIH, and about appropriate posttransplantation immunosuppression strategies that take into account the increased immunoreactivity of these patients. To address some of these issues, we reviewed our data on 32 patients who underwent LTX for a primary indication of AIH and we sought to determine whether specific pretransplantation characteristics or various aspects of the postoperative course could be used to predict posttransplantation outcome, including disease recurrence. PATIENTS AND METHODS
Abbreviations: LTX, liver transplantation; AIH, autoimmune hepatitis; FHF, fulminant hepatic failure; LFT, liver function test; ANA, antinuclear antibody; SMA, smooth muscle antibody; LKM, liver kidney microsomal antibody; AMA, antimitochondrial antibody; UNOS, United Network for Organ Sharing; HLA, human leukocyte antigen; re-LTx, retransplantation; NS, not significant. From the 1Department of Surgery, Albert Einstein Medical Center, Philadelphia, PA; 2Department of Pathology, Elmhurst Medical Center, Queens, NY; 3Department of Surgery, Columbia-Presbyterian Medical Center, New York, NY; 4Recanati/Miller Transplantation Institute, the Mount Sinai Hospital of Mount Sinai—NYU Health, New York, NY; 5Department of Surgery, Temple University Hospital, Philadelphia, PA. Received October 27, 1999; accepted July 5, 2000. Address reprint requests to: Patricia A. Sheiner, M.D., Mount Sinai Hospital, Box 1104, One Gustave L. Levy Place, New York, NY 10029. E-mail: patricia.sheiner@ mountsinai.org; fax: 212-996-9688. Copyright © 2000 by the American Association for the Study of Liver Diseases. 0270-9139/00/3204-0004$3.00/0 doi:10.1053/jhep.2000.16666
We retrospectively reviewed our data on patients in whom a clinical and pathologic diagnosis of AIH had been made on the basis of the diagnostic criteria of the International Autoimmune Hepatitis Group16 and who underwent transplantation at The Mount Sinai Medical Center (New York, NY) between September 1988 and January 1995. Patients with coexisting causes of hepatitis, such as viral or alcohol-induced hepatitis, were excluded from the study. Indications for transplantation were either complications of endstage liver disease or fulminant/subfulminant liver failure. Patients with fulminant hepatic failure (FHF) met the diagnostic criteria of Pappas.17 We recorded pretransplantation data (age, gender, pretransplantation disease duration, concomitant autoimmune disorders, pretransplantation corticosteroid and azathioprine use, indication for LTX, liver function tests [LFTs: transaminases, bilirubin, alkaline phosphatase, albumin, and prothrombin time], autoimmune markers [antinuclear antibody (ANA), smooth muscle anti-
693
694 REICH ET AL.
HEPATOLOGY October 2000
body (SMA), liver kidney microsomal antibody (LKM), antimitochondrial antibody (AMA), and gamma globulin levels], viral serologies, United Network for Organ Sharing [UNOS] status, and autoimmune-associated human leukocyte antigens [HLAs] A1, B8, Dr3) and posttransplantation data (donor HLA antigens, postoperative complications, immunosuppressive regimen, rejection episodes, histopathologic findings, and LFTs before liver biopsies and during the final month of follow-up for this study). In addition, during the final month of follow-up each patient’s autoimmune markers were measured. Posttransplantation immunosuppression varied among patients, and was based on cyclosporine or tacrolimus. Cyclosporine doses were adjusted to maintain blood trough levels of 250 to 400 ng/mL initially and tapered to yield 100 to 200 ng/mL by 6 months (TDx monoclonal FPIA; Abbott, Abbott Park, IL). Tacrolimus doses were adjusted to maintain trough plasma levels of 15 to 20 ng/mL initially and tapered to yield approximately 5 ng/mL by 6 months (IMx polyclonal FPIA; Abbott). Prednisone was begun at 20 mg/d and tapered to 5 mg/d by 6 months. Cyclosporine-treated patients also received azathioprine, 1 mg/kg/d. Single rejection episodes were usually treated with corticosteroids. OKT3 was used to treat steroid-resistant rejection and as initial therapy for some cases of severe rejection. After recurrent rejection patients were maintained on prednisone 10 mg/d and those on cyclosporine were switched to tacrolimus. None of the patients received OKT3 induction immunosuppression. Posttransplantation biopsies were performed to evaluate unexplained LFT elevations. Representative sections taken from hepatectomy specimens and posttransplantation liver biopsy slides were stained with hematoxylin and eosin and reviewed for this study by one pathologist blinded to patient scenarios except for the presumptive diagnosis of AIH. Findings of chronic hepatitis with plasma cell infiltrates were considered to be histologic evidence of AIH. Chronic hepatitis was graded and staged according to the scoring system of Hytiroglou et al.18 Staging was based on degree of portal fibrosis, and necroinflammatory activity was graded according to the extent of piecemeal necrosis and lobular activity. Histologic diagnosis of recurrent AIH was based on the same criteria, in the absence of rejection, infection, or a biliary tract problem. Autoantibodies (ANA, SMA, LKM, AMA) were detected using indirect immunofluorescence assays as follows: ANA (Quantafluor Hep-2 Cell-line Substrate Kit; Sanofi Diagnostics Pasteur, Chaska, MN); SMA and AMA (CT-2 Fluorokit; INCSTAR, Stillwater, MN); LKM (Liver Disease Western Blot; SCIMEDX, Denville, NJ). Titers ⱖ1:40 were considered elevated. Gamma globulin levels were determined by serum protein electrophoresis (Paragon Electrophoresis System; Beckman Coulter, Brea, CA) followed by densitometry (Appraise Densitometer; Beckman Coulter). Levels ⱖ1.6 g/dL were considered elevated. Anti-hepatitis C was sought with a second-generation enzyme-linked immunosorbent assay (HCV EIA 2.0; Abbott). Data are presented as mean ⫾ SD. Groups were compared using Fisher’s exact 2-tailed test or the Mantel-Haenszel test for categorical data, and 2-tailed t tests or the Cox proportional hazards model for continuous data. The Mantel-Haenszel test or the Cox proportional hazards model was used when length of follow-up differed between groups. P ⬍ .05 was considered significant. Actuarial patient survival was computed using the Kaplan-Meier estimate. RESULTS
Thirty-four patients with AIH, among a total of 671 patients (5%), received 41 of 793 liver transplantations during the study period. In 2 of the 34, hepatitis C was confirmed clinically, pathologically, and serologically both before and after LTX; these 2 patients were excluded, leaving for study 32 patients who underwent LTX for AIH. Demographics. Demographic characteristics of the patients transplanted for AIH are presented in Table 1. Thirty of 32 patients were women. The mean age at primary LTX was 37 ⫾14 years (range, 15-63). Twenty-four patients (75%) had
TABLE 1. Clinical Features of Chronic and Fulminant AIH Patients at the Time of Transplantation Chronic Patients (n ⴝ 24)
Age (yr)* 38 ⫾ 15 (15-63) Female gender† 23/24 (96%) Prior manifestations of liver disease† Ascites 22/24 (92%) Jaundice 20/24 (83%) Weakness 18/24 (75%) Spontaneous bacterial peritonitis 13/24 (54%) Variceal hemorrhage 13/24 (54%) Encephalopathy 11/24 (46%) Hepatorenal syndrome 3/24 (13%) Hepatopulmonary syndrome 1/24 (4%) Serum laboratory values at transplantation* Prothrombin time (sec) 17 ⫾ 4 (12-26) Albumin (g/dL) 2.7 ⫾ 0.7 (1.7-4.5) Total bilirubin (mg/dL) 12 ⫾ 11 (1-36) Alkaline phosphatase (U/L) 228 ⫾ 181 (5-804) ALT (U/L) 91 ⫾ 69 (21-361) AST (U/L) 111 ⫾ 65 (34-296) UNOS status† 1 7/24 (29%) 2 9/24 (38%) 3 8/24 (33%)
Fulminant/ Subfulminant Patients (n ⴝ 8)
33 ⫾ 14 (17-50) 7/8 (88%)
7/8 (88%) 8/8 (100%) 8/8 (100%) 2/8 (25%) 1/8 (13%) 8/8 (100%) — —
20 ⫾ 3 (16-23) 2.6 ⫾ 0.2 (2.3-2.8) 19 ⫾ 9 (6-33) 257 ⫾ 139 (129-560) 241 ⫾ 171 (76-614) 308 ⫾ 187 (99-676) 3/8 (38%) 5/8 (62%) —
*Mean ⫾ SD (range). †n (%).
known chronic liver disease; in these patients, the mean duration of disease pretransplantation was 8 ⫾ 6 years (range, 1-20). Eight patients (25%) had fulminant (n ⫽ 4) or subfulminant (n ⫽ 4) hepatic failure and underwent LTX 4 ⫾ 2 months (range, 1-6) after onset of illness. Autoantibodies. Autoimmune characteristics of the patients transplanted for AIH are presented in Table 2. LKM antibodies were measured in 9 patients and were undetectable in all 9. AMA titers were measured in 29 patients and were elevated in 2, both of whom were also ANA positive. There were no differences in any of these autoimmune markers between patients with chronic disease and those with FHF. HLA Typing. HLA typing was done in 27 patients, 11 of whom (41%) had at least one of the A1, B8, or Dr3 antigens. Six (22%) were A1 positive, 5 (19%) were B8 positive, and 7 (26%) were DR3 positive. Six (22%) had more than 1 of these antigens. All B8-positive patients were also A1 positive and/or DR3 positive. Only 1 patient had the A1-B8-DR3 phenotype. The distribution of these antigens was not significantly different between patients with chronic liver disease and those with FHF. Pretransplant Steroid Use. At the time of LTX, 17 of the 24 patients (71%) with known chronic liver disease were taking prednisone, 26 ⫾ 15 mg/d (range, 10-60), for 4.0 ⫾ 4.0 years (range, 0.30-14). Five of these 17 (29%) were also taking azathioprine, 50 mg/d, for 6.4 ⫾ 4.0 years (range, 2-14). Most of the other patients had been treated with prednisone, with or without azathioprine, earlier in the course of their disease. Of the 8 patients with FHF, 4 received a trial of prednisone; 3 of these 4 also received azathioprine.
HEPATOLOGY Vol. 32, No. 4, 2000
REICH ET AL.
TABLE 2. Autoimmune Characteristics of Patients Transplanted for AIH (n ⴝ 32) Autoantibodies* Either ANA or SMA ⱖ1:40 Both ANA and SMA ⱖ1:40 ANA ⱖ1:40 SMA ⱖ1:40 ANA⫹ titer (median, range) SMA⫹ titer (median, range) Gamma globulin* Gamma globulin elevation ⱖ1.5 g/dL Gamma globulin levels (mean ⫾ SD, range) HLA† A1⫹ B8⫹ Dr3⫹ A1-B8-Dr3 Associated autoimmune disorders At least one associated autoimmune disorder Rheumatoid arthritis Ulcerative colitis Hemolytic anemia Pericarditis Systemic lupus erythmatosus Glomerulonephritis Thyroiditis Psoriasis Raynaud’s disease
32/32 (100%) 15/31 (48%) 23/32 (72%) 24/31 (74%) 320 (40-1,280) 320 (40-2,560) 31/31 (100%) 3.0 ⫾ 1.0 (1.8-5.6) 6/27 (22%) 5/27 (19%) 7/27 (26%) 1/27 (4%) 8/32 (25%) 4/32 (13%) 3/32 (9%) 2/32 (6%) 2/32 (6%) 1/32 (3%) 1/32 (3%) 1/32 (3%) 1/32 (3%) 1/32 (3%)
NOTE. Data are presented as n (%) unless otherwise indicated. *ANA titers were measured in 32 patients, and SMA titers and gamma globulin levels in 31 patients. †HLA typing was performed in 27 of 32 patients.
Associated Autoimmune Disorders. Eight of the 24 patients (25%) with known chronic liver disease had other associated autoimmune disorders (Table 2), in some cases predating known liver disease by many years. These included 6 of 24 patients (25%) with chronic liver disease and 2 of 8 (25%) with FHF. All 8 patients had previously been unsuccessfully treated with steroids for these disorders. Posttransplantation, the associated autoimmune disorders resolved completely in 2 patients and significantly in another 2; in 4 patients, the associated disorders remained stable. One additional patient developed ulcerative colitis posttransplantation. Histologic Findings. Of the available native hepatectomy specimens from patients with chronic disease, most (16 of 20 explants) revealed cirrhosis with varying degrees of necroinflammatory activity and plasma cell infiltrates (Table 3). Most of the available FHF specimens (5 of 7 explants) revealed massive or submassive hepatic necrosis, broad areas of parenchymal collapse, bridging necrosis, frequent-to-dense plasma cell infiltrates, cholestasis, and bile duct proliferation. Cirrhosis was present in 2 of 7 FHF explants. Morbidity and Mortality. There were 8 deaths among these 32 AIH patients with 23 ⫾ 16 months of follow-up (range, 2 days to 52 months) (Table 4). Actuarial survival of the 32 patients was 81% at 1 and 2 years following transplantation (Fig. 1). In addition to the 6 of 32 patients who died of perioperative complications, another 2 patients were excluded from posttransplantation analysis because of other diseases that they developed: 1 contracted hepatitis B during LTX and required retransplantation (re-LTX) 5 months later, and the other developed adenocarcinoma of the pancreas several weeks after LTX and died of metastatic disease the following year.
695
TABLE 3. Native Hepatectomy Histology From Chronic AIH Patients (n ⴝ 20) Stage fibrosis/cirrhosis 0 (none) 1 (portal fibrosis) 2 (limited fibrous septa) 3 (bridging fibrosis) 4 (cirrhosis) Grade necroinflammatory activity—piecemeal necrosis 0 (none) 1 (mild) 2 (moderate) 3 (severe) Grade necroinflammatory activity—lobular activity 0 (none) 1 (mild) 2 (moderate) 3 (severe) Plasma cell infiltrates None Rare Frequent Dense
— — 1 (5%) 3 (15%) 16 (80%) — 8 (40%) 8 (40%) 4 (20%) — 6 (30%) 7 (35%) 7 (35%) — 10 (50%) 5 (25%) 5 (25%)
NOTE. Explants from 20 of 24 chronic AIH patients were available for review. Data are presented as n (%).
Twenty-four AIH recipients remained for posttransplantation analysis, with long-term follow-up of 27 ⫾ 14 months (range 6-52). Postoperative complications occurred in 17 of 24 patients (71%), not including acute rejection or recurrent AIH (Table 5). In patients receiving pretransplantation immunosuppression, there was a trend toward an increased incidence of wound infection (4 of 13 vs. 1 of 19) but no obvious increase in incidence of other bacterial or viral infections. Rejection Episodes. Eighteen of 24 patients (75%) had 1.7 ⫾ 0.8 episodes of acute rejection (range, 1-3). Four of these 18 patients (22%) required OKT3 for steroid-resistant rejection. An additional 3 patients with severe rejection were treated with OKT3 as initial therapy, so that a total of 39% of rejections (7 of 18) were treated with OKT3. One patient developed chronic rejection, diagnosed 10 months after LTX, and was eventually listed for re-LTX. The incidence of rejection did not correlate with recipient or graft HLA-A1, B8, or Dr3 status or with matching of these antigens. TABLE 4. Causes of Mortality (8 of 32 AIH Recipients) Early postoperative deaths (within 2 months) Sepsis following re-LTx for primary nonfunction Sepsis following re-LTx for severe acute cellular rejection Sepsis complicating original transplantation Myocardial infarction Cerebral hemorrhage Long-term deaths (after 1 year) Pancreatic adenocarcinoma* Chronic rejection complicating noncompliance†
2 1 1 1 1 1 1
*This developed several weeks after LTx, and the patient expired with metastatic disease a year later. She is not included among the 24 of 32 patients with long-term follow-up. †This patient became noncompliant after re-LTx for recurrent AIH performed 20 months after her first LTx, and died of chronic rejection a year later. She is included among the 24 of 32 patients with long-term follow-up only until re-LTx.
696 REICH ET AL.
HEPATOLOGY October 2000
FIG. 2. The impact of recurrent autoimmune hepatitis among 24 patients with long-term follow-up after liver transplantation.
FIG. 1. Kaplan-Meier survival of patients transplanted for autoimmune hepatitis (n ⫽ 32) was similar to that of patients transplanted for other liver diseases (n ⫽ 639) during the same time period (P ⫽ .39, log rank test).
Recurrent AIH. Among the 24 patients with long-term follow-up (27 ⫾ 14 months), 25% (6 patients) developed histologically documented recurrent AIH, 15 ⫾ 2 months after LTX (range, 12-18); 5 of the 6 had persistently abnormal LFTs and 4 of the 6 had elevated autoantibody titers. The impact of recurrent AIH is summarized in Fig. 2. Three of the 6 patients (50%) with recurrent AIH required re-LTX 11 ⫾ 3 months after the diagnosis of recurrence (range, 7-14). After re-LTX, 1 patient became noncompliant and died of chronic rejection and 2 had histologically proven re-recurrence within 3 months; 1 of these received a third LTX. All 4 recurrent AIH explants showed cirrhosis, necroinflammatory activity, plasma cell infiltrates, and severe cholestasis. Moderate to severe lobular activity, frequent to dense
TABLE 5. Complications After Transplantation (26 of 32 AIH Recipients) Biliary complications Bile leak from T-tube exit site, treated by endobiliary stent Biliary stricture, treated with choledochojejunostomy Infections Severe wound infection requiring debridement Herpes hepatitis De novo hepatitis B* Cytomegalovirus hepatitis Epstein-Barr virus hepatitis Immunosuppression toxicity Severe but reversible neurologic side effects Severe but reversible renal failure Miscellaneous Portal vein thrombosis, requiring thrombectomy Small bowel obstruction, requiring adhesion lysis Pulmonary embolism Retinal vein thrombosis Chronic rejection
6 2 5 1 1 1 1 1 3 1 1 1 1 1 1
NOTE. This table excludes acute rejection episodes, cases of recurrent AIH, and fatal complications occurring in 8 of 32 patients (described in Table 4). *This developed 3 months after LTX. He is not included among the 24 of 32 patients with long-term follow-up.
plasma cell infiltrates and collapse, were all present in 3 of 4 explants. Piecemeal necrosis was usually mild. Biopsy-proven recurrence was present in all 3 patients whose posttransplantation autoantibody titers were greater than or equal to their pretransplantation titers, but in only 1 of 4 whose titers were less than pretransplantation titers. Five patients without recurrent AIH also had elevated autoantibody titers, 38 ⫾ 13 months posttransplantation (range, 1652), but these patients had essentially normal LFTs and did not require biopsy. Histologic recurrence was documented in 6 of 18 patients transplanted for chronic autoimmune disease versus none of 6 transplanted for fulminant disease (P ⫽ not significant [NS]). Follow-up was 29 ⫾ 15 months (range, 6-52) for patients with chronic disease and 20 ⫾ 12 months (range, 6-41) for patients with FHF. No other difference was found between patients with and without recurrent AIH. They had similar pretransplantation factors, recipient and graft autoimmune-associated HLA antigens and antigen matching (Table 6), immunosuppressive regimens, and number and severity of rejection episodes (Table 7). The 6 patients with recurrent AIH had not undergone immunosuppression reduction for at least several months before the diagnosis of recurrence. Their regimens consisted of prednisone 5 mg (n ⫽ 4) or 10 mg (n ⫽ 2, raised from 5 mg many months earlier because of repeated rejections) and tacrolimus 3 to 8 mg twice a day, trough 5 to 10 ng/mL (n ⫽ 6, 3 patients had been converted from cyclosporine many months earlier because of repeated rejections). Cyclophosphamide, 50 to 75 mg/d started early posttransplantation, was used for prophylaxis against recurrent AIH in 3 patients, 2 of whom had already been retransplanted once for recurrent AIH; all 3 of these patients developed recurrent disease despite the cyclophosphamide. Long-Term Laboratory Values. Mean laboratory values 27 ⫾ 14 months after LTX were essentially normal in 17 of 24 patients. The remaining 7 patients had abnormal laboratory values caused by resolving acute cellular rejection (n ⫽ 1), chronic rejection (n ⫽ 2, one case associated with noncompliance), mild recurrent AIH (n ⫽ 2), and re-recurrent AIH after re-LTX (n ⫽ 2). DISCUSSION
The clinical course of AIH is variable. This illness causes chronic liver disease or FHF. There is no cure for AIH, but most patients can be successfully managed with medication. Some AIH patients deteriorate to the extent that they require
HEPATOLOGY Vol. 32, No. 4, 2000
REICH ET AL.
697
TABLE 6. HLA Data From Patients Who Developed Recurrent AIH
Recipient HLA status Donor HLA status
A1ⴙ
A1ⴚ
P
B8ⴙ
B8ⴚ
P
Dr3ⴙ
Dr3ⴚ
P
1/5 1/5
4/14 4/14
NS NS
2/5 1/4
3/14 4/15
NS NS
1/4 1/3
4/15 4/16
NS NS
NOTE. HLA data available for 19 of 24 recipients and donors.
LTX, and approximately 6% of all liver transplantations in the United States are performed for patients with this disease.19 However, many questions regarding transplantation for AIH remain unanswered. Which Patients With AIH Are Likely to Need Transplantation?
Our patients were 37 ⫾ 14 years old and were predominantly women (30 of 32). Most often, they had suffered chronic disease for 8 ⫾ 6 years before LTX. Their most common disease manifestations were ascites (91%), jaundice (88%), elevated prothrombin time (18 ⫾ 3), and hypoalbuminemia (2.7 ⫾ 0.6). All had elevated ANA and/or SMA titers and all but 1 had hypergammaglobulinemia. Our patients’ pretransplantation mean duration of chronic illness, most frequent physical and biochemical abnormalities, and serologic profile are similar to those that other investigators have published regarding their transplanted AIH patients.2,9,10 Our series of patients is distinct because 25% presented with FHF, and underwent LTX 4 ⫾ 2 months after onset of illness. There are several reports that AIH patients with the HLA A1-B8-Dr3 phenotype or the B8 and/or Dr3 antigens suffer more aggressive disease,2,20,21 and that those who require liver transplantation are more likely to have the A1-B8-Dr3 phenotype than those who do not.20 Only 19% and 26% of our patients were B8 positive and Dr3 positive, respectively, and only one patient was A1-B8-Dr3 positive. Although we did not collect data about patient ethnicity, the low incidence of A1, B8, and Dr3 antigens among our patients might be explained by racial differences, which have been shown to affect the distribution of these genes even among closely related populations.20 AIH is a heterogeneous disease of multifactorial etiology, and severe illness is known to occur in the absence of these particular antigens.2,20,21 Does Pretransplantation Immunosuppressive Therapy for AIH Increase the Risk of Posttransplantation Infectious Complications?
There has been concern that patients with AIH who are taking prednisone at the time of LTX might suffer more complications or higher mortality. There was a trend toward increased risk of wound infection in our patients taking predTABLE 7. Rejection Data From Recipients Who Developed Recurrent AIH and Those Who Did Not Recurrent AIH Present (n ⴝ 6)
Rejection prevalence* Number of rejections† Mean ⫾ SD Range OKT3 required‡
Recurrent AIH Absent (n ⴝ 18)
P
6/6 (100%)
12/18 (67%)
NS
1.6 ⫾ 0.8 1-3 1/6 (17%)
1.6 ⫾ 0.6 1-3 3/18 (17%)
NS
*Patients with at least 1 rejection episode. Data presented as n (%). †Number of rejections per patient with at least 1 rejection episode. ‡For steroid-resistant rejection. Data presented as n (%).
NS
nisone before LTX (4 of 13 vs. 1 of 19). We could not demonstrate an increased risk of other bacterial or viral infections because of pretransplantation immunosuppression in this relatively small cohort. It is noteworthy that 4 of the 6 early postoperative deaths were related to sepsis; 2 of these 4 patients received pretransplantation immunosuppression, which may have contributed to their mortality. It is our current practice to taper or withdraw immunosuppression in chronic AIH patients nearing transplantation and to discontinue it in FHF patients who fail to show an expeditious significant response. What Are the Results of LTX for AIH? Survival. It is apparent that LTX for AIH can be accomplished with acceptable morbidity and mortality. The posttransplantation mortality of 25% of these 32 AIH patients, with 23 ⫾ 16 months of follow-up (range, 2 days to 52 months), approximates the mortality of 26% of the non-AIH patients (167 of 639) that we transplanted during the same time period. There is no difference in the Kaplan-Meier survival curves of the AIH and non-AIH groups (Fig. 1). Patients transplanted for AIH have been reported to have 74% to 92% 5-year survival.2,9,12,19 In our opinion, the mortality among these 32 patients is not representative of that among most patients transplanted for AIH because of the unusually high incidence in this group of early re-LTX (2 primary nonfunction, 1 severe rejection) and of cardiovascular deaths (1 myocardial infarction, 1 cerebral hemorrhage), neither of which is associated with AIH. Early sepsis accounted for half the deaths (4 of 8) and followed early re-LTX in 3 of 4 cases. We could not identify unusual clinical or biochemical risk factors for mortality among transplanted AIH patients. Rejection. These recipients suffered a high frequency of rejection (75% of patients, 1-3 episodes), and of steroid-resistant rejection; not only were 22% of the rejections (4 of 18) steroid resistant, but also 3 severe rejections were treated with OKT3 as initial therapy, so that a total of 39% of rejections (7 of 18) were treated with OKT3. Others have noted similarly high incidences of rejection and steroid-resistant rejection in AIH patients, ranging from 56% to 83% and 23% to 59%, respectively.7,9,11-13 At Stanford University (Stanford, CA), patients transplanted for AIH compared with a control group transplanted for alcohol-induced cirrhosis had a higher incidence of rejection (79% vs. 47%) and steroid-resistant rejection (33% vs. 13%) and a higher risk of multiple rejection episodes per patient, similar to our experience.13 The risk of rejection in our patients was not affected by the recipient or graft HLA A1, B8, or Dr3 status, or by the matching of these antigens, as has also been shown by the University of Pittsburgh (Pittsburgh, PA).7 Disease Recurrence. This review corroborates that patients transplanted for AIH have a significant incidence of recurrence in the allografts (Fig. 2). Among the 24 patients with long-term follow-up (27 ⫾ 14 months), biopsy-proven recurrent AIH occurred in 25% within 12 to 18 months. Of note,
698 REICH ET AL.
HEPATOLOGY October 2000
half of the patients with biopsy-proven recurrent AIH required re-LTX and then suffered a particularly high incidence of re-recurrence. Three of 6 patients with biopsy-proven recurrence required re-LTX 24 ⫾ 3 months after the first LTX. There is long-term follow-up for 2 of these 3 patients that required re-LTX and they both developed biopsy-proven rerecurrence of AIH within 3 months after re-LTX. One of these 2 received a third LTX 19 months after re-LTX. Table 8 highlights the experience with disease recurrence in 7 other series of AIH transplantation patients.7-12,22 Five groups, with 15 to 47 patients each, reported recurrence in 20% to 33% of the recipients, similar to 25% of our patients, also occurring predominantly during the second and third posttransplantation years.7-10,12 Another group reported that as many as 82% of their recipients (18 of 22) were found on biopsy to have mild to moderate hepatitis suggestive of recurrent AIH.11 However, none of the Mayo Clinic patients (n ⫽ 24) was diagnosed with recurrence, at 39 ⫾ 4 months followup. Most reported episodes of recurrent AIH were mild,7-8,12 but the need for re-LTX because of severe recurrence has been documented.9,10 Differences in the incidence and severity of recurrent AIH might, at least in part, be explained by differences in the immunosuppressive regimens of various programs. For example, the Mayo Clinic baseline immunosuppressive regimen, including prednisone 10 mg/d and azathioprine 2 mg/kg/d,2 tended to be more potent than those of the other centers that provided details about their regimens.8-10,12 When the Mayo group revisited their experience with longer follow-up, 11% of their patients (4 of 35) recurred, the diagnosis being made 72 ⫾ 20 months posttransplantation.22 They attributed recurrence to attempted reduction of maintenance immunosuppression. Recently, another group that reported an increased risk of recurrence over time (8% at 1 year, 68% at 5 years) also attributed the risk to concomitantly decreased immunosuppressive treatment.8 It is known that recurrent AIH can follow tapering of baseline immunosuppression, such as prednisone or azathioprine decrease or withdrawal, and remission can been attained in some patients with recurrence by increasing immunosuppression.4-6,8,10-12 We could not temporally relate recurrent AIH in our patients to reduction of immunosuppression, similar to the experience at Hopital Paul Brousse,
Villejuif, France.9 The immunosuppressive regimen of our 6 patients with recurrent AIH had not been reduced for at least several months before the diagnosis of recurrence. When recurrent AIH was diagnosed, we tended to increase baseline steroids and/or add cyclophosphamide. Cyclophosphamide did not appear to be helpful in either treating recurrent disease or preventing re-recurrence after re-LTX. More recently, we have used 6-mercaptopurine with better results. There is currently debate regarding the prudence of steroid withdrawal in AIH recipients.15 The University of Colorado group (Boulder, CO) was able to withdraw steroids from 17 of 25 AIH patients (68%), reducing hypercholesterolemia, hypertension, and diabetes.14 There was no recurrent AIH at a mean follow-up of 22 months. Rejection during steroid withdrawal was more common in AIH patients than in others (33% vs. 14%) but the episodes were mostly steroid responsive and there was no graft loss. Skeptics of the Colorado approach express concern that the follow-up period after steroid withdrawal was as short as 1 month in some patients, there were very few liver biopsies, the magnitude of nonsteroidal immunosuppression was not evaluated in detail, and the feasibility of steroid independence for posttransplantation AIH patients has not been corroborated.15 Indeed, attempts at steroid withdrawal by other groups have failed in 50% to 64% of AIH patients.10,12 It would be useful to identify other risk factors for the development of recurrent AIH in addition to length of follow-up and tapering of immunosuppression. We found that AIH was less likely to recur in those patients who originally presented with FHF as opposed to those who had known chronic disease (0 of 6 vs. 6 of 18), although the patient subgroups were too small to show statistical significance. Perhaps the chronic AIH patients who require LTX continue to have a propensity for immunosuppression resistance, just as they did pretransplantation, whereas fulminants, naive to chronic immunosuppression, are likely to respond to it. Another risk factor that we identified for development of recurrent AIH was re-LTX for recurrent AIH. Both of our patients with long-term follow-up after re-LTX for recurrent AIH subsequently re-recurred. This is similar to the experience at the University of Birmingham, Birmingham, UK, where all 3 patients who required re-LTX for recurrence re-recurred in the
TABLE 8. Recurrent AIH at Various Transplantation Programs Program 7
U. of Pittsburgh, 1992 Mayo Clinic,22 1995‡ Spanish Coop. Grp.,8 1998 Hopital Paul Brousse,9 1999 U. Birmingham, UK,10 1999 Humboldt U, Berlin,11 1999 U. CA San Fran.,12 1999 Mt. Sinai Hosp., current
Length of Follow-Up (mo Postop)*
NA 91 ⫾ 13 (NA) 44 ⫾ 28 (8-108) 58 ⫾ NA (24-102) 50 (median f/u) 47 ⫾ NA (20-101) 42 (median f/u) 27 ⫾ 14 (6-52)
Incidence of Recurrence n (%)
Time of Recurrence (mo Postop)*
Graft Loss (n)†
11/43 (26%) 4/35 (11%) 9/27 (33%) 3/15 (20%) 13/47 (28%) 18/22 (82%)§ 8/40 (20%)㛳 6/24 (25%)
18 ⫾ 4 (NA) 72 ⫾ 20 (NA) 31 ⫾ 18 (8-53) 24 ⫾ 12 (12-36) 29 ⫾ NA (6-63) NA 18 ⫾ 13 (NA) 15 ⫾ 2 (12-18)
— — — 2 3 — — 4¶
Abbreviation: NA, not available. *Data presented as mean ⫾ SD (range) unless otherwise indicated. †Allografts lost because of recurrent AIH. ‡This abstract updates their 1991 publication,2 when with less follow-up there were no cases of recurrent AIH. All of the other references cited in this table are complete publications. §Eighteen patients had mild-moderate graft hepatitis but only 5 of these had allograft dysfunction. 㛳Eight patients had histologic evidence of recurrence; 5 additional patients had presumed recurrence but did not undergo biopsy. ¶Four allografts were lost from 3 patients; 1 patient lost 2 grafts.
HEPATOLOGY Vol. 32, No. 4, 2000
second grafts,10 and at Hopital Paul Brousse, where the patient with recurrent AIH who was retransplanted went on to rerecur.9 We did not identify other risk factors for recurrent AIH. Patients with recurrent AIH had a similar incidence and severity of rejection as those patients who did not recur, and an increased frequency or severity of rejections in a patient was not a risk factor for recurrence. Some groups have found that Dr3-positive recipients are more susceptible to recurrence,7-8,12 whereas other groups similar to us have not found this to be true.9-11 Extrahepatic Autoimmune Disorders. Although these patients with pretransplantation autoimmunity continue to be “hyper” immune after LTX, exhibiting disease recurrence and an increased risk of rejection, their extrahepatic associated autoimmune disorders significantly improve posttransplantation. One quarter of the patients had associated autoimmune disorders before LTX and these improved significantly or resolved completely in half these patients; none of them became worse. Only 1 patient developed an autoimmune disorder after LTX (mild ulcerative colitis). Patients whose associated autoimmune disorders improved posttransplantation had already failed treatment for these disorders with prednisone before LTX; it is likely that the addition of cyclosporine or tacrolimus contributed to the posttransplantation improvement. Do Fulminant AIH Patients Have a Unique Posttransplantation Course?
A significant proportion of the AIH recipients had presented with FHF (25%). However, this high-risk group did not suffer greater mortality than did the chronic patients. AIH recurred more often in patients who presented with known chronic disease than in those who had FHF, although the groups were too small to show statistical significance. Does the Presence of Specific HLA Antigens or Autoantibodies Result in an Increased Incidence of Rejection or Recurrence of Disease After Transplantation?
The risk of rejection was not affected by the recipient or graft HLA A1, B8, or Dr3 status, or by the matching of these antigens, as has also been shown by the University of Pittsburgh.7 However, several groups have related these antigens, especially in recipients that are Dr3 positive, to disease recurrence.7,8,12 We did not demonstrate this association. Posttransplantation autoantibody titer elevation might occasionally be a marker for disease recurrence. In the Mayo Clinic series, all 24 patients became seronegative for ANA and SMA within 9 ⫾ 8 months after LTX,2 and thereafter titer elevation was sometimes present in patients with biopsyproven recurrence (2 of 4).22 However, others have found that autoantibodies frequently persist posttransplantation and are not per se a useful marker of recurrent AIH.7-9,11 Autoantibodies were present in 9 of 24 of our recipients (38%). In the Paul Brousse series, autoantibodies persisted in 12 of 14 recipients (86%), decreasing during the first 4 months posttransplantation, increasing over the following 2 years to a lower titer than pretransplantation, and then remaining stable.9 Perhaps the degree of titer elevation can predict biopsy-provable recurrence. Biopsy-proven recurrence was present in all 3 of our patients whose posttransplantation titers were greater than or equal to their pretransplantation titers, but in only 1 of 4
REICH ET AL.
699
whose titers were less than those pretransplantation. Neither pretransplantation nor posttransplantation autoantibody titer elevation was a risk factor for rejection. In conclusion, LTX for patients with AIH is highly successful, although it is complicated by a high risk of rejection and of steroid-resistant rejection, and severe recurrent AIH sometimes develops. Patients already retransplanted for recurrence are at particularly high risk of recurring again. Patients transplanted for fulminant AIH appear to be protected from recurrence. Recurrence can follow reduction of baseline immunosuppression, but this was not the case with our patients. Neither was recurrence more likely in our HLA Dr3-positive recipients. Multi-institutional collaborative studies would allow accrual of enough patients to determine the best posttransplantation immunosuppressive regimens for AIH patients. Gene research may elucidate the etiology of AIH and of posttransplantation recurrence, which would have therapeutic implications. Acknowledgment: The authors are grateful to Nancy Ehrlich for her invaluable editorial assistance. REFERENCES 1. Waldenstrom J. Leber, blutproteine und nahrungseiweiss. Dtsch Gesellsch Verdau Stoffwechselkr 1950;15:113-121. 2. Sanchez-Urdazpal L, Czaja AJ, Van Hoek B, Krom RAF, Wiesner RH. Prognostic features and role of liver transplantation in severe corticosteroid-treated autoimmune chronic active hepatitis. HEPATOLOGY 1991;15: 215-221. 3. Roberts SK, Therneau TM, Czaja AJ. Prognosis of histological cirrhosis in type 1 autoimmune hepatitis. Gastroenterology 1996;110:848-857. 4. Neuberger J, Portmann B, Calne R, Williams R. Recurrence of autoimmune chronic active hepatitis following orthotopic liver grafting. Transplantation 1984;37:363-365. 5. Devlin J, Donaldson P, Portmann B, Heaton N, Tan KC, Williams R. Recurrence of autoimmune hepatitis following liver transplantation. Liver Transpl Surg 1995;1:162-165. 6. Sempooux C, Horsmans Y, Lerut J, Rahier J, Geubel A. Acute lobular hepatitis as the first manifestation of recurrent autoimmune hepatitis after orthotopic liver transplantation. Liver 1997;17:311-315. 7. Wright HL, Bou-Abboud CF, Hassanein T, Block GD, Demetris AJ, Starzl TE, Van Thiel DH. Disease recurrence and rejection following liver transplantation for autoimmune chronic active liver disease. Transplantation 1992;53:136-139. 8. Prados E, Cuervas-Mons V, de la Mata M, Fraga E, Rimola A, Prieto M, Clemente G, et al. Outcome of autoimmune hepatitis after liver transplantation. Transplantation 1998;66:1645-1650. 9. Ratziu V, Samuel D, Sebagh M, Farges O, Saliba F, Ichai P, Farahmand H, et al. Long-term follow-up after liver transplantation for autoimmune hepatitis: evidence of recurrence of primary disease. J Hepatol 1999;20: 131-141. 10. Milkiewicz P, Hubscher SG, Skiba G, Hathaway M, Ellias E. Recurrence of autoimmune hepatitis after liver transplantation. Transplantation 1999;68:253-256. 11. Gotz G, Neuhaus R, Bechstein WO, Lobeck H, Berg T, Hopf U, Neuhaus P. Recurrence of autoimmune hepatitis after liver transplantation. Transplant Proc 1999;31:430-431. 12. Narumi S, Hakamada K, Sasaki M, Freise CE, Stock PG, Roberts JP, Ascher NL. Liver transplantation for autoimmune hepatitis: rejection and recurrence. Transplant Proc 1999;31:1955-1956. 13. Hayashi M, Keeffe EB, Krams SM, Martinez OM, Ojogho ON, So SKS, Garcia G, et al. Allograft rejection after liver transplantation for autoimmune liver diseases. Liver Transpl Surg 1998;4:208-214. 14. Trouillot TE, Shrestha R, Kam I, Wachs M, Everson GT. Successful withdrawal of prednisone after adult liver transplantation for autoimmune hepatitis. Liver Transpl Surg 1999;5:375-380. 15. Czaja AJ. The immunoreactive propensity of autoimmune hepatitis: is it corticosteroid-dependent after liver transplantation? Liver Transpl Surg 1999;5:460-463.
700 REICH ET AL. 16. Johnson PJ, McFarlane IG. Meeting report: International autoimmune hepatitis group. HEPATOLOGY 1993;18:998-1005. 17. Pappas SC. Fulminant hepatic failure. In: Gitnick G, La Brecque DR, Moody FG, eds. Diseases of the Liver and Biliary Tract. St. Louis: Mosby, 1992;615. 18. Hytiroglou P, Thung SN, Gerber MA. Classification and quantitation of the severity of chronic hepatitis: keep it simple! Semin Liver Dis 1995; 15:414-421. 19. Belle SH, Beringer KC, Detre KM. An update on liver transplantation in the United States: recipient characteristics and outcome. In: Cecka JM, Terasaki PI, eds. Clinical Transplants 1995. Los Angeles: UCLA Tissue Typing Laboratory, 1995;19-33.
HEPATOLOGY October 2000 20. Donaldson PT, Doherty DG, Hayllar KM, McFarlane IG, Johnson PJ, Williams R. Susceptibility to autoimmune chronic active hepatitis: human leukocyte antigens DR4 and A1-B8-DR3 are independent risk factors. HEPATOLOGY 1991;13:701-706. 21. Czaja AJ, Strettell MDJ, Thomson LJ, Santrach PJ, Moore SB, Donaldson PT, Williams R. Associations between alleles of the major histocompatibility complex and type 1 autoimmune hepatitis. HEPATOLOGY 1997;25: 317-323. 22. Roberts S, Czaja AJ, Charlton MR, Hay JE, Krom RAF, Poravko MK, Wiesner RH. Recurrent autoimmune hepatitis following orthotopic liver transplantation [Abstract]. HEPATOLOGY 1995;22(Suppl):215A.
SEE EDITORIAL ON PAGE 868
studies could elucidate the best posttransplantation immunosuppressive regimens for AIH patients. (HEPATOLOGY 2000;32:693-700.)
Publications about liver transplantation (LTX) for autoimmune hepatitis (AIH) have started to emerge, but many issues remain unresolved. We reviewed data on 32 patients transplanted for AIH to determine how pretransplantation and posttransplantation characteristics correlate with recipient outcome, including disease recurrence. Recipients were 37ⴞ 14 years old; 30 of 32 were women. Most had chronic disease (8 ⴞ 6 years); 25% had fulminant failure. The majority had ascites (91%), jaundice (88%), elevated prothrombin time (18 ⴞ 3 seconds), and hypoalbuminemia (2.7 ⴞ 0.6 g/dL). All had hypergammaglobulinemia (3.0 ⴞ 1.0 g/dL) and autoantibodies (72% antinuclear, 74% smooth muscle). Only one was HLA A1-B8-DR3 positive. Other autoimmune disorders affected 25% of patients; half improved after transplantation. Actuarial survival was 81% at 1 and 2 years posttransplantation. There was a high frequency of rejection (75% of recipients had 1.7 ⴞ 0.8 episodes), and 39% of rejections required OKT3. Among 24 recipients with long-term follow-up (27 ⴞ 14 months), histologically proven recurrent AIH occurred in 25%, 15 ⴞ 2 months posttransplantation; half (3 patients) required retransplantation 11 ⴞ 3 months after diagnosis. After retransplantation 2 of 3 patients had re-recurrence within 3 months; 1 received a third LTx. Recurrence occurred in 6 of 18 patients transplanted for chronic disease vs. 0 of 6 transplanted as fulminants (P ⴝ not significant [NS]). Patients with and without recurrence had similar rejection profiles. In summary, results of LTx for AIH are excellent. However, AIH patients have a high frequency of rejection and often require OKT3. Furthermore, severe recurrent AIH sometimes develops, particularly in chronic versus fulminant AIH patients and in those already retransplanted for recurrence. Multicenter
Liver transplantation (LTX) is perhaps the most dramatic advance in the management of autoimmune hepatitis (AIH) since the earliest description of this disease by Waldenstrom in 1950.1 Although immunosuppression remains the mainstay of therapy for patients with AIH, 13% to 20% deteriorate on traditional regimens, and of those who experience remission and discontinue therapy, 74% to 85% relapse.2,3 Medical therapy does not cure AIH. Therefore, many patients eventually develop end-stage liver disease and become candidates for LTX. Until a short time ago the literature on LTX for AIH was sparse, including one publication that addressed the role of LTX for AIH,2 and some reports about recurrence of AIH after LTX.4-7 Recently, several retrospective reviews on LTX for AIH have emerged from various transplantation programs8-12 as well as a few publications about rejection and immunosuppression after LTX for AIH.13-15 In general there is consensus about the demographics of AIH patients that come to LTX. However, there are unresolved issues regarding the posttransplantation course of these patients. For example, there is controversy about the prevalence, risk factors, and outcome of recurrent AIH, and about appropriate posttransplantation immunosuppression strategies that take into account the increased immunoreactivity of these patients. To address some of these issues, we reviewed our data on 32 patients who underwent LTX for a primary indication of AIH and we sought to determine whether specific pretransplantation characteristics or various aspects of the postoperative course could be used to predict posttransplantation outcome, including disease recurrence. PATIENTS AND METHODS
Abbreviations: LTX, liver transplantation; AIH, autoimmune hepatitis; FHF, fulminant hepatic failure; LFT, liver function test; ANA, antinuclear antibody; SMA, smooth muscle antibody; LKM, liver kidney microsomal antibody; AMA, antimitochondrial antibody; UNOS, United Network for Organ Sharing; HLA, human leukocyte antigen; re-LTx, retransplantation; NS, not significant. From the 1Department of Surgery, Albert Einstein Medical Center, Philadelphia, PA; 2Department of Pathology, Elmhurst Medical Center, Queens, NY; 3Department of Surgery, Columbia-Presbyterian Medical Center, New York, NY; 4Recanati/Miller Transplantation Institute, the Mount Sinai Hospital of Mount Sinai—NYU Health, New York, NY; 5Department of Surgery, Temple University Hospital, Philadelphia, PA. Received October 27, 1999; accepted July 5, 2000. Address reprint requests to: Patricia A. Sheiner, M.D., Mount Sinai Hospital, Box 1104, One Gustave L. Levy Place, New York, NY 10029. E-mail: patricia.sheiner@ mountsinai.org; fax: 212-996-9688. Copyright © 2000 by the American Association for the Study of Liver Diseases. 0270-9139/00/3204-0004$3.00/0 doi:10.1053/jhep.2000.16666
We retrospectively reviewed our data on patients in whom a clinical and pathologic diagnosis of AIH had been made on the basis of the diagnostic criteria of the International Autoimmune Hepatitis Group16 and who underwent transplantation at The Mount Sinai Medical Center (New York, NY) between September 1988 and January 1995. Patients with coexisting causes of hepatitis, such as viral or alcohol-induced hepatitis, were excluded from the study. Indications for transplantation were either complications of endstage liver disease or fulminant/subfulminant liver failure. Patients with fulminant hepatic failure (FHF) met the diagnostic criteria of Pappas.17 We recorded pretransplantation data (age, gender, pretransplantation disease duration, concomitant autoimmune disorders, pretransplantation corticosteroid and azathioprine use, indication for LTX, liver function tests [LFTs: transaminases, bilirubin, alkaline phosphatase, albumin, and prothrombin time], autoimmune markers [antinuclear antibody (ANA), smooth muscle anti-
693
694 REICH ET AL.
HEPATOLOGY October 2000
body (SMA), liver kidney microsomal antibody (LKM), antimitochondrial antibody (AMA), and gamma globulin levels], viral serologies, United Network for Organ Sharing [UNOS] status, and autoimmune-associated human leukocyte antigens [HLAs] A1, B8, Dr3) and posttransplantation data (donor HLA antigens, postoperative complications, immunosuppressive regimen, rejection episodes, histopathologic findings, and LFTs before liver biopsies and during the final month of follow-up for this study). In addition, during the final month of follow-up each patient’s autoimmune markers were measured. Posttransplantation immunosuppression varied among patients, and was based on cyclosporine or tacrolimus. Cyclosporine doses were adjusted to maintain blood trough levels of 250 to 400 ng/mL initially and tapered to yield 100 to 200 ng/mL by 6 months (TDx monoclonal FPIA; Abbott, Abbott Park, IL). Tacrolimus doses were adjusted to maintain trough plasma levels of 15 to 20 ng/mL initially and tapered to yield approximately 5 ng/mL by 6 months (IMx polyclonal FPIA; Abbott). Prednisone was begun at 20 mg/d and tapered to 5 mg/d by 6 months. Cyclosporine-treated patients also received azathioprine, 1 mg/kg/d. Single rejection episodes were usually treated with corticosteroids. OKT3 was used to treat steroid-resistant rejection and as initial therapy for some cases of severe rejection. After recurrent rejection patients were maintained on prednisone 10 mg/d and those on cyclosporine were switched to tacrolimus. None of the patients received OKT3 induction immunosuppression. Posttransplantation biopsies were performed to evaluate unexplained LFT elevations. Representative sections taken from hepatectomy specimens and posttransplantation liver biopsy slides were stained with hematoxylin and eosin and reviewed for this study by one pathologist blinded to patient scenarios except for the presumptive diagnosis of AIH. Findings of chronic hepatitis with plasma cell infiltrates were considered to be histologic evidence of AIH. Chronic hepatitis was graded and staged according to the scoring system of Hytiroglou et al.18 Staging was based on degree of portal fibrosis, and necroinflammatory activity was graded according to the extent of piecemeal necrosis and lobular activity. Histologic diagnosis of recurrent AIH was based on the same criteria, in the absence of rejection, infection, or a biliary tract problem. Autoantibodies (ANA, SMA, LKM, AMA) were detected using indirect immunofluorescence assays as follows: ANA (Quantafluor Hep-2 Cell-line Substrate Kit; Sanofi Diagnostics Pasteur, Chaska, MN); SMA and AMA (CT-2 Fluorokit; INCSTAR, Stillwater, MN); LKM (Liver Disease Western Blot; SCIMEDX, Denville, NJ). Titers ⱖ1:40 were considered elevated. Gamma globulin levels were determined by serum protein electrophoresis (Paragon Electrophoresis System; Beckman Coulter, Brea, CA) followed by densitometry (Appraise Densitometer; Beckman Coulter). Levels ⱖ1.6 g/dL were considered elevated. Anti-hepatitis C was sought with a second-generation enzyme-linked immunosorbent assay (HCV EIA 2.0; Abbott). Data are presented as mean ⫾ SD. Groups were compared using Fisher’s exact 2-tailed test or the Mantel-Haenszel test for categorical data, and 2-tailed t tests or the Cox proportional hazards model for continuous data. The Mantel-Haenszel test or the Cox proportional hazards model was used when length of follow-up differed between groups. P ⬍ .05 was considered significant. Actuarial patient survival was computed using the Kaplan-Meier estimate. RESULTS
Thirty-four patients with AIH, among a total of 671 patients (5%), received 41 of 793 liver transplantations during the study period. In 2 of the 34, hepatitis C was confirmed clinically, pathologically, and serologically both before and after LTX; these 2 patients were excluded, leaving for study 32 patients who underwent LTX for AIH. Demographics. Demographic characteristics of the patients transplanted for AIH are presented in Table 1. Thirty of 32 patients were women. The mean age at primary LTX was 37 ⫾14 years (range, 15-63). Twenty-four patients (75%) had
TABLE 1. Clinical Features of Chronic and Fulminant AIH Patients at the Time of Transplantation Chronic Patients (n ⴝ 24)
Age (yr)* 38 ⫾ 15 (15-63) Female gender† 23/24 (96%) Prior manifestations of liver disease† Ascites 22/24 (92%) Jaundice 20/24 (83%) Weakness 18/24 (75%) Spontaneous bacterial peritonitis 13/24 (54%) Variceal hemorrhage 13/24 (54%) Encephalopathy 11/24 (46%) Hepatorenal syndrome 3/24 (13%) Hepatopulmonary syndrome 1/24 (4%) Serum laboratory values at transplantation* Prothrombin time (sec) 17 ⫾ 4 (12-26) Albumin (g/dL) 2.7 ⫾ 0.7 (1.7-4.5) Total bilirubin (mg/dL) 12 ⫾ 11 (1-36) Alkaline phosphatase (U/L) 228 ⫾ 181 (5-804) ALT (U/L) 91 ⫾ 69 (21-361) AST (U/L) 111 ⫾ 65 (34-296) UNOS status† 1 7/24 (29%) 2 9/24 (38%) 3 8/24 (33%)
Fulminant/ Subfulminant Patients (n ⴝ 8)
33 ⫾ 14 (17-50) 7/8 (88%)
7/8 (88%) 8/8 (100%) 8/8 (100%) 2/8 (25%) 1/8 (13%) 8/8 (100%) — —
20 ⫾ 3 (16-23) 2.6 ⫾ 0.2 (2.3-2.8) 19 ⫾ 9 (6-33) 257 ⫾ 139 (129-560) 241 ⫾ 171 (76-614) 308 ⫾ 187 (99-676) 3/8 (38%) 5/8 (62%) —
*Mean ⫾ SD (range). †n (%).
known chronic liver disease; in these patients, the mean duration of disease pretransplantation was 8 ⫾ 6 years (range, 1-20). Eight patients (25%) had fulminant (n ⫽ 4) or subfulminant (n ⫽ 4) hepatic failure and underwent LTX 4 ⫾ 2 months (range, 1-6) after onset of illness. Autoantibodies. Autoimmune characteristics of the patients transplanted for AIH are presented in Table 2. LKM antibodies were measured in 9 patients and were undetectable in all 9. AMA titers were measured in 29 patients and were elevated in 2, both of whom were also ANA positive. There were no differences in any of these autoimmune markers between patients with chronic disease and those with FHF. HLA Typing. HLA typing was done in 27 patients, 11 of whom (41%) had at least one of the A1, B8, or Dr3 antigens. Six (22%) were A1 positive, 5 (19%) were B8 positive, and 7 (26%) were DR3 positive. Six (22%) had more than 1 of these antigens. All B8-positive patients were also A1 positive and/or DR3 positive. Only 1 patient had the A1-B8-DR3 phenotype. The distribution of these antigens was not significantly different between patients with chronic liver disease and those with FHF. Pretransplant Steroid Use. At the time of LTX, 17 of the 24 patients (71%) with known chronic liver disease were taking prednisone, 26 ⫾ 15 mg/d (range, 10-60), for 4.0 ⫾ 4.0 years (range, 0.30-14). Five of these 17 (29%) were also taking azathioprine, 50 mg/d, for 6.4 ⫾ 4.0 years (range, 2-14). Most of the other patients had been treated with prednisone, with or without azathioprine, earlier in the course of their disease. Of the 8 patients with FHF, 4 received a trial of prednisone; 3 of these 4 also received azathioprine.
HEPATOLOGY Vol. 32, No. 4, 2000
REICH ET AL.
TABLE 2. Autoimmune Characteristics of Patients Transplanted for AIH (n ⴝ 32) Autoantibodies* Either ANA or SMA ⱖ1:40 Both ANA and SMA ⱖ1:40 ANA ⱖ1:40 SMA ⱖ1:40 ANA⫹ titer (median, range) SMA⫹ titer (median, range) Gamma globulin* Gamma globulin elevation ⱖ1.5 g/dL Gamma globulin levels (mean ⫾ SD, range) HLA† A1⫹ B8⫹ Dr3⫹ A1-B8-Dr3 Associated autoimmune disorders At least one associated autoimmune disorder Rheumatoid arthritis Ulcerative colitis Hemolytic anemia Pericarditis Systemic lupus erythmatosus Glomerulonephritis Thyroiditis Psoriasis Raynaud’s disease
32/32 (100%) 15/31 (48%) 23/32 (72%) 24/31 (74%) 320 (40-1,280) 320 (40-2,560) 31/31 (100%) 3.0 ⫾ 1.0 (1.8-5.6) 6/27 (22%) 5/27 (19%) 7/27 (26%) 1/27 (4%) 8/32 (25%) 4/32 (13%) 3/32 (9%) 2/32 (6%) 2/32 (6%) 1/32 (3%) 1/32 (3%) 1/32 (3%) 1/32 (3%) 1/32 (3%)
NOTE. Data are presented as n (%) unless otherwise indicated. *ANA titers were measured in 32 patients, and SMA titers and gamma globulin levels in 31 patients. †HLA typing was performed in 27 of 32 patients.
Associated Autoimmune Disorders. Eight of the 24 patients (25%) with known chronic liver disease had other associated autoimmune disorders (Table 2), in some cases predating known liver disease by many years. These included 6 of 24 patients (25%) with chronic liver disease and 2 of 8 (25%) with FHF. All 8 patients had previously been unsuccessfully treated with steroids for these disorders. Posttransplantation, the associated autoimmune disorders resolved completely in 2 patients and significantly in another 2; in 4 patients, the associated disorders remained stable. One additional patient developed ulcerative colitis posttransplantation. Histologic Findings. Of the available native hepatectomy specimens from patients with chronic disease, most (16 of 20 explants) revealed cirrhosis with varying degrees of necroinflammatory activity and plasma cell infiltrates (Table 3). Most of the available FHF specimens (5 of 7 explants) revealed massive or submassive hepatic necrosis, broad areas of parenchymal collapse, bridging necrosis, frequent-to-dense plasma cell infiltrates, cholestasis, and bile duct proliferation. Cirrhosis was present in 2 of 7 FHF explants. Morbidity and Mortality. There were 8 deaths among these 32 AIH patients with 23 ⫾ 16 months of follow-up (range, 2 days to 52 months) (Table 4). Actuarial survival of the 32 patients was 81% at 1 and 2 years following transplantation (Fig. 1). In addition to the 6 of 32 patients who died of perioperative complications, another 2 patients were excluded from posttransplantation analysis because of other diseases that they developed: 1 contracted hepatitis B during LTX and required retransplantation (re-LTX) 5 months later, and the other developed adenocarcinoma of the pancreas several weeks after LTX and died of metastatic disease the following year.
695
TABLE 3. Native Hepatectomy Histology From Chronic AIH Patients (n ⴝ 20) Stage fibrosis/cirrhosis 0 (none) 1 (portal fibrosis) 2 (limited fibrous septa) 3 (bridging fibrosis) 4 (cirrhosis) Grade necroinflammatory activity—piecemeal necrosis 0 (none) 1 (mild) 2 (moderate) 3 (severe) Grade necroinflammatory activity—lobular activity 0 (none) 1 (mild) 2 (moderate) 3 (severe) Plasma cell infiltrates None Rare Frequent Dense
— — 1 (5%) 3 (15%) 16 (80%) — 8 (40%) 8 (40%) 4 (20%) — 6 (30%) 7 (35%) 7 (35%) — 10 (50%) 5 (25%) 5 (25%)
NOTE. Explants from 20 of 24 chronic AIH patients were available for review. Data are presented as n (%).
Twenty-four AIH recipients remained for posttransplantation analysis, with long-term follow-up of 27 ⫾ 14 months (range 6-52). Postoperative complications occurred in 17 of 24 patients (71%), not including acute rejection or recurrent AIH (Table 5). In patients receiving pretransplantation immunosuppression, there was a trend toward an increased incidence of wound infection (4 of 13 vs. 1 of 19) but no obvious increase in incidence of other bacterial or viral infections. Rejection Episodes. Eighteen of 24 patients (75%) had 1.7 ⫾ 0.8 episodes of acute rejection (range, 1-3). Four of these 18 patients (22%) required OKT3 for steroid-resistant rejection. An additional 3 patients with severe rejection were treated with OKT3 as initial therapy, so that a total of 39% of rejections (7 of 18) were treated with OKT3. One patient developed chronic rejection, diagnosed 10 months after LTX, and was eventually listed for re-LTX. The incidence of rejection did not correlate with recipient or graft HLA-A1, B8, or Dr3 status or with matching of these antigens. TABLE 4. Causes of Mortality (8 of 32 AIH Recipients) Early postoperative deaths (within 2 months) Sepsis following re-LTx for primary nonfunction Sepsis following re-LTx for severe acute cellular rejection Sepsis complicating original transplantation Myocardial infarction Cerebral hemorrhage Long-term deaths (after 1 year) Pancreatic adenocarcinoma* Chronic rejection complicating noncompliance†
2 1 1 1 1 1 1
*This developed several weeks after LTx, and the patient expired with metastatic disease a year later. She is not included among the 24 of 32 patients with long-term follow-up. †This patient became noncompliant after re-LTx for recurrent AIH performed 20 months after her first LTx, and died of chronic rejection a year later. She is included among the 24 of 32 patients with long-term follow-up only until re-LTx.
696 REICH ET AL.
HEPATOLOGY October 2000
FIG. 2. The impact of recurrent autoimmune hepatitis among 24 patients with long-term follow-up after liver transplantation.
FIG. 1. Kaplan-Meier survival of patients transplanted for autoimmune hepatitis (n ⫽ 32) was similar to that of patients transplanted for other liver diseases (n ⫽ 639) during the same time period (P ⫽ .39, log rank test).
Recurrent AIH. Among the 24 patients with long-term follow-up (27 ⫾ 14 months), 25% (6 patients) developed histologically documented recurrent AIH, 15 ⫾ 2 months after LTX (range, 12-18); 5 of the 6 had persistently abnormal LFTs and 4 of the 6 had elevated autoantibody titers. The impact of recurrent AIH is summarized in Fig. 2. Three of the 6 patients (50%) with recurrent AIH required re-LTX 11 ⫾ 3 months after the diagnosis of recurrence (range, 7-14). After re-LTX, 1 patient became noncompliant and died of chronic rejection and 2 had histologically proven re-recurrence within 3 months; 1 of these received a third LTX. All 4 recurrent AIH explants showed cirrhosis, necroinflammatory activity, plasma cell infiltrates, and severe cholestasis. Moderate to severe lobular activity, frequent to dense
TABLE 5. Complications After Transplantation (26 of 32 AIH Recipients) Biliary complications Bile leak from T-tube exit site, treated by endobiliary stent Biliary stricture, treated with choledochojejunostomy Infections Severe wound infection requiring debridement Herpes hepatitis De novo hepatitis B* Cytomegalovirus hepatitis Epstein-Barr virus hepatitis Immunosuppression toxicity Severe but reversible neurologic side effects Severe but reversible renal failure Miscellaneous Portal vein thrombosis, requiring thrombectomy Small bowel obstruction, requiring adhesion lysis Pulmonary embolism Retinal vein thrombosis Chronic rejection
6 2 5 1 1 1 1 1 3 1 1 1 1 1 1
NOTE. This table excludes acute rejection episodes, cases of recurrent AIH, and fatal complications occurring in 8 of 32 patients (described in Table 4). *This developed 3 months after LTX. He is not included among the 24 of 32 patients with long-term follow-up.
plasma cell infiltrates and collapse, were all present in 3 of 4 explants. Piecemeal necrosis was usually mild. Biopsy-proven recurrence was present in all 3 patients whose posttransplantation autoantibody titers were greater than or equal to their pretransplantation titers, but in only 1 of 4 whose titers were less than pretransplantation titers. Five patients without recurrent AIH also had elevated autoantibody titers, 38 ⫾ 13 months posttransplantation (range, 1652), but these patients had essentially normal LFTs and did not require biopsy. Histologic recurrence was documented in 6 of 18 patients transplanted for chronic autoimmune disease versus none of 6 transplanted for fulminant disease (P ⫽ not significant [NS]). Follow-up was 29 ⫾ 15 months (range, 6-52) for patients with chronic disease and 20 ⫾ 12 months (range, 6-41) for patients with FHF. No other difference was found between patients with and without recurrent AIH. They had similar pretransplantation factors, recipient and graft autoimmune-associated HLA antigens and antigen matching (Table 6), immunosuppressive regimens, and number and severity of rejection episodes (Table 7). The 6 patients with recurrent AIH had not undergone immunosuppression reduction for at least several months before the diagnosis of recurrence. Their regimens consisted of prednisone 5 mg (n ⫽ 4) or 10 mg (n ⫽ 2, raised from 5 mg many months earlier because of repeated rejections) and tacrolimus 3 to 8 mg twice a day, trough 5 to 10 ng/mL (n ⫽ 6, 3 patients had been converted from cyclosporine many months earlier because of repeated rejections). Cyclophosphamide, 50 to 75 mg/d started early posttransplantation, was used for prophylaxis against recurrent AIH in 3 patients, 2 of whom had already been retransplanted once for recurrent AIH; all 3 of these patients developed recurrent disease despite the cyclophosphamide. Long-Term Laboratory Values. Mean laboratory values 27 ⫾ 14 months after LTX were essentially normal in 17 of 24 patients. The remaining 7 patients had abnormal laboratory values caused by resolving acute cellular rejection (n ⫽ 1), chronic rejection (n ⫽ 2, one case associated with noncompliance), mild recurrent AIH (n ⫽ 2), and re-recurrent AIH after re-LTX (n ⫽ 2). DISCUSSION
The clinical course of AIH is variable. This illness causes chronic liver disease or FHF. There is no cure for AIH, but most patients can be successfully managed with medication. Some AIH patients deteriorate to the extent that they require
HEPATOLOGY Vol. 32, No. 4, 2000
REICH ET AL.
697
TABLE 6. HLA Data From Patients Who Developed Recurrent AIH
Recipient HLA status Donor HLA status
A1ⴙ
A1ⴚ
P
B8ⴙ
B8ⴚ
P
Dr3ⴙ
Dr3ⴚ
P
1/5 1/5
4/14 4/14
NS NS
2/5 1/4
3/14 4/15
NS NS
1/4 1/3
4/15 4/16
NS NS
NOTE. HLA data available for 19 of 24 recipients and donors.
LTX, and approximately 6% of all liver transplantations in the United States are performed for patients with this disease.19 However, many questions regarding transplantation for AIH remain unanswered. Which Patients With AIH Are Likely to Need Transplantation?
Our patients were 37 ⫾ 14 years old and were predominantly women (30 of 32). Most often, they had suffered chronic disease for 8 ⫾ 6 years before LTX. Their most common disease manifestations were ascites (91%), jaundice (88%), elevated prothrombin time (18 ⫾ 3), and hypoalbuminemia (2.7 ⫾ 0.6). All had elevated ANA and/or SMA titers and all but 1 had hypergammaglobulinemia. Our patients’ pretransplantation mean duration of chronic illness, most frequent physical and biochemical abnormalities, and serologic profile are similar to those that other investigators have published regarding their transplanted AIH patients.2,9,10 Our series of patients is distinct because 25% presented with FHF, and underwent LTX 4 ⫾ 2 months after onset of illness. There are several reports that AIH patients with the HLA A1-B8-Dr3 phenotype or the B8 and/or Dr3 antigens suffer more aggressive disease,2,20,21 and that those who require liver transplantation are more likely to have the A1-B8-Dr3 phenotype than those who do not.20 Only 19% and 26% of our patients were B8 positive and Dr3 positive, respectively, and only one patient was A1-B8-Dr3 positive. Although we did not collect data about patient ethnicity, the low incidence of A1, B8, and Dr3 antigens among our patients might be explained by racial differences, which have been shown to affect the distribution of these genes even among closely related populations.20 AIH is a heterogeneous disease of multifactorial etiology, and severe illness is known to occur in the absence of these particular antigens.2,20,21 Does Pretransplantation Immunosuppressive Therapy for AIH Increase the Risk of Posttransplantation Infectious Complications?
There has been concern that patients with AIH who are taking prednisone at the time of LTX might suffer more complications or higher mortality. There was a trend toward increased risk of wound infection in our patients taking predTABLE 7. Rejection Data From Recipients Who Developed Recurrent AIH and Those Who Did Not Recurrent AIH Present (n ⴝ 6)
Rejection prevalence* Number of rejections† Mean ⫾ SD Range OKT3 required‡
Recurrent AIH Absent (n ⴝ 18)
P
6/6 (100%)
12/18 (67%)
NS
1.6 ⫾ 0.8 1-3 1/6 (17%)
1.6 ⫾ 0.6 1-3 3/18 (17%)
NS
*Patients with at least 1 rejection episode. Data presented as n (%). †Number of rejections per patient with at least 1 rejection episode. ‡For steroid-resistant rejection. Data presented as n (%).
NS
nisone before LTX (4 of 13 vs. 1 of 19). We could not demonstrate an increased risk of other bacterial or viral infections because of pretransplantation immunosuppression in this relatively small cohort. It is noteworthy that 4 of the 6 early postoperative deaths were related to sepsis; 2 of these 4 patients received pretransplantation immunosuppression, which may have contributed to their mortality. It is our current practice to taper or withdraw immunosuppression in chronic AIH patients nearing transplantation and to discontinue it in FHF patients who fail to show an expeditious significant response. What Are the Results of LTX for AIH? Survival. It is apparent that LTX for AIH can be accomplished with acceptable morbidity and mortality. The posttransplantation mortality of 25% of these 32 AIH patients, with 23 ⫾ 16 months of follow-up (range, 2 days to 52 months), approximates the mortality of 26% of the non-AIH patients (167 of 639) that we transplanted during the same time period. There is no difference in the Kaplan-Meier survival curves of the AIH and non-AIH groups (Fig. 1). Patients transplanted for AIH have been reported to have 74% to 92% 5-year survival.2,9,12,19 In our opinion, the mortality among these 32 patients is not representative of that among most patients transplanted for AIH because of the unusually high incidence in this group of early re-LTX (2 primary nonfunction, 1 severe rejection) and of cardiovascular deaths (1 myocardial infarction, 1 cerebral hemorrhage), neither of which is associated with AIH. Early sepsis accounted for half the deaths (4 of 8) and followed early re-LTX in 3 of 4 cases. We could not identify unusual clinical or biochemical risk factors for mortality among transplanted AIH patients. Rejection. These recipients suffered a high frequency of rejection (75% of patients, 1-3 episodes), and of steroid-resistant rejection; not only were 22% of the rejections (4 of 18) steroid resistant, but also 3 severe rejections were treated with OKT3 as initial therapy, so that a total of 39% of rejections (7 of 18) were treated with OKT3. Others have noted similarly high incidences of rejection and steroid-resistant rejection in AIH patients, ranging from 56% to 83% and 23% to 59%, respectively.7,9,11-13 At Stanford University (Stanford, CA), patients transplanted for AIH compared with a control group transplanted for alcohol-induced cirrhosis had a higher incidence of rejection (79% vs. 47%) and steroid-resistant rejection (33% vs. 13%) and a higher risk of multiple rejection episodes per patient, similar to our experience.13 The risk of rejection in our patients was not affected by the recipient or graft HLA A1, B8, or Dr3 status, or by the matching of these antigens, as has also been shown by the University of Pittsburgh (Pittsburgh, PA).7 Disease Recurrence. This review corroborates that patients transplanted for AIH have a significant incidence of recurrence in the allografts (Fig. 2). Among the 24 patients with long-term follow-up (27 ⫾ 14 months), biopsy-proven recurrent AIH occurred in 25% within 12 to 18 months. Of note,
698 REICH ET AL.
HEPATOLOGY October 2000
half of the patients with biopsy-proven recurrent AIH required re-LTX and then suffered a particularly high incidence of re-recurrence. Three of 6 patients with biopsy-proven recurrence required re-LTX 24 ⫾ 3 months after the first LTX. There is long-term follow-up for 2 of these 3 patients that required re-LTX and they both developed biopsy-proven rerecurrence of AIH within 3 months after re-LTX. One of these 2 received a third LTX 19 months after re-LTX. Table 8 highlights the experience with disease recurrence in 7 other series of AIH transplantation patients.7-12,22 Five groups, with 15 to 47 patients each, reported recurrence in 20% to 33% of the recipients, similar to 25% of our patients, also occurring predominantly during the second and third posttransplantation years.7-10,12 Another group reported that as many as 82% of their recipients (18 of 22) were found on biopsy to have mild to moderate hepatitis suggestive of recurrent AIH.11 However, none of the Mayo Clinic patients (n ⫽ 24) was diagnosed with recurrence, at 39 ⫾ 4 months followup. Most reported episodes of recurrent AIH were mild,7-8,12 but the need for re-LTX because of severe recurrence has been documented.9,10 Differences in the incidence and severity of recurrent AIH might, at least in part, be explained by differences in the immunosuppressive regimens of various programs. For example, the Mayo Clinic baseline immunosuppressive regimen, including prednisone 10 mg/d and azathioprine 2 mg/kg/d,2 tended to be more potent than those of the other centers that provided details about their regimens.8-10,12 When the Mayo group revisited their experience with longer follow-up, 11% of their patients (4 of 35) recurred, the diagnosis being made 72 ⫾ 20 months posttransplantation.22 They attributed recurrence to attempted reduction of maintenance immunosuppression. Recently, another group that reported an increased risk of recurrence over time (8% at 1 year, 68% at 5 years) also attributed the risk to concomitantly decreased immunosuppressive treatment.8 It is known that recurrent AIH can follow tapering of baseline immunosuppression, such as prednisone or azathioprine decrease or withdrawal, and remission can been attained in some patients with recurrence by increasing immunosuppression.4-6,8,10-12 We could not temporally relate recurrent AIH in our patients to reduction of immunosuppression, similar to the experience at Hopital Paul Brousse,
Villejuif, France.9 The immunosuppressive regimen of our 6 patients with recurrent AIH had not been reduced for at least several months before the diagnosis of recurrence. When recurrent AIH was diagnosed, we tended to increase baseline steroids and/or add cyclophosphamide. Cyclophosphamide did not appear to be helpful in either treating recurrent disease or preventing re-recurrence after re-LTX. More recently, we have used 6-mercaptopurine with better results. There is currently debate regarding the prudence of steroid withdrawal in AIH recipients.15 The University of Colorado group (Boulder, CO) was able to withdraw steroids from 17 of 25 AIH patients (68%), reducing hypercholesterolemia, hypertension, and diabetes.14 There was no recurrent AIH at a mean follow-up of 22 months. Rejection during steroid withdrawal was more common in AIH patients than in others (33% vs. 14%) but the episodes were mostly steroid responsive and there was no graft loss. Skeptics of the Colorado approach express concern that the follow-up period after steroid withdrawal was as short as 1 month in some patients, there were very few liver biopsies, the magnitude of nonsteroidal immunosuppression was not evaluated in detail, and the feasibility of steroid independence for posttransplantation AIH patients has not been corroborated.15 Indeed, attempts at steroid withdrawal by other groups have failed in 50% to 64% of AIH patients.10,12 It would be useful to identify other risk factors for the development of recurrent AIH in addition to length of follow-up and tapering of immunosuppression. We found that AIH was less likely to recur in those patients who originally presented with FHF as opposed to those who had known chronic disease (0 of 6 vs. 6 of 18), although the patient subgroups were too small to show statistical significance. Perhaps the chronic AIH patients who require LTX continue to have a propensity for immunosuppression resistance, just as they did pretransplantation, whereas fulminants, naive to chronic immunosuppression, are likely to respond to it. Another risk factor that we identified for development of recurrent AIH was re-LTX for recurrent AIH. Both of our patients with long-term follow-up after re-LTX for recurrent AIH subsequently re-recurred. This is similar to the experience at the University of Birmingham, Birmingham, UK, where all 3 patients who required re-LTX for recurrence re-recurred in the
TABLE 8. Recurrent AIH at Various Transplantation Programs Program 7
U. of Pittsburgh, 1992 Mayo Clinic,22 1995‡ Spanish Coop. Grp.,8 1998 Hopital Paul Brousse,9 1999 U. Birmingham, UK,10 1999 Humboldt U, Berlin,11 1999 U. CA San Fran.,12 1999 Mt. Sinai Hosp., current
Length of Follow-Up (mo Postop)*
NA 91 ⫾ 13 (NA) 44 ⫾ 28 (8-108) 58 ⫾ NA (24-102) 50 (median f/u) 47 ⫾ NA (20-101) 42 (median f/u) 27 ⫾ 14 (6-52)
Incidence of Recurrence n (%)
Time of Recurrence (mo Postop)*
Graft Loss (n)†
11/43 (26%) 4/35 (11%) 9/27 (33%) 3/15 (20%) 13/47 (28%) 18/22 (82%)§ 8/40 (20%)㛳 6/24 (25%)
18 ⫾ 4 (NA) 72 ⫾ 20 (NA) 31 ⫾ 18 (8-53) 24 ⫾ 12 (12-36) 29 ⫾ NA (6-63) NA 18 ⫾ 13 (NA) 15 ⫾ 2 (12-18)
— — — 2 3 — — 4¶
Abbreviation: NA, not available. *Data presented as mean ⫾ SD (range) unless otherwise indicated. †Allografts lost because of recurrent AIH. ‡This abstract updates their 1991 publication,2 when with less follow-up there were no cases of recurrent AIH. All of the other references cited in this table are complete publications. §Eighteen patients had mild-moderate graft hepatitis but only 5 of these had allograft dysfunction. 㛳Eight patients had histologic evidence of recurrence; 5 additional patients had presumed recurrence but did not undergo biopsy. ¶Four allografts were lost from 3 patients; 1 patient lost 2 grafts.
HEPATOLOGY Vol. 32, No. 4, 2000
second grafts,10 and at Hopital Paul Brousse, where the patient with recurrent AIH who was retransplanted went on to rerecur.9 We did not identify other risk factors for recurrent AIH. Patients with recurrent AIH had a similar incidence and severity of rejection as those patients who did not recur, and an increased frequency or severity of rejections in a patient was not a risk factor for recurrence. Some groups have found that Dr3-positive recipients are more susceptible to recurrence,7-8,12 whereas other groups similar to us have not found this to be true.9-11 Extrahepatic Autoimmune Disorders. Although these patients with pretransplantation autoimmunity continue to be “hyper” immune after LTX, exhibiting disease recurrence and an increased risk of rejection, their extrahepatic associated autoimmune disorders significantly improve posttransplantation. One quarter of the patients had associated autoimmune disorders before LTX and these improved significantly or resolved completely in half these patients; none of them became worse. Only 1 patient developed an autoimmune disorder after LTX (mild ulcerative colitis). Patients whose associated autoimmune disorders improved posttransplantation had already failed treatment for these disorders with prednisone before LTX; it is likely that the addition of cyclosporine or tacrolimus contributed to the posttransplantation improvement. Do Fulminant AIH Patients Have a Unique Posttransplantation Course?
A significant proportion of the AIH recipients had presented with FHF (25%). However, this high-risk group did not suffer greater mortality than did the chronic patients. AIH recurred more often in patients who presented with known chronic disease than in those who had FHF, although the groups were too small to show statistical significance. Does the Presence of Specific HLA Antigens or Autoantibodies Result in an Increased Incidence of Rejection or Recurrence of Disease After Transplantation?
The risk of rejection was not affected by the recipient or graft HLA A1, B8, or Dr3 status, or by the matching of these antigens, as has also been shown by the University of Pittsburgh.7 However, several groups have related these antigens, especially in recipients that are Dr3 positive, to disease recurrence.7,8,12 We did not demonstrate this association. Posttransplantation autoantibody titer elevation might occasionally be a marker for disease recurrence. In the Mayo Clinic series, all 24 patients became seronegative for ANA and SMA within 9 ⫾ 8 months after LTX,2 and thereafter titer elevation was sometimes present in patients with biopsyproven recurrence (2 of 4).22 However, others have found that autoantibodies frequently persist posttransplantation and are not per se a useful marker of recurrent AIH.7-9,11 Autoantibodies were present in 9 of 24 of our recipients (38%). In the Paul Brousse series, autoantibodies persisted in 12 of 14 recipients (86%), decreasing during the first 4 months posttransplantation, increasing over the following 2 years to a lower titer than pretransplantation, and then remaining stable.9 Perhaps the degree of titer elevation can predict biopsy-provable recurrence. Biopsy-proven recurrence was present in all 3 of our patients whose posttransplantation titers were greater than or equal to their pretransplantation titers, but in only 1 of 4
REICH ET AL.
699
whose titers were less than those pretransplantation. Neither pretransplantation nor posttransplantation autoantibody titer elevation was a risk factor for rejection. In conclusion, LTX for patients with AIH is highly successful, although it is complicated by a high risk of rejection and of steroid-resistant rejection, and severe recurrent AIH sometimes develops. Patients already retransplanted for recurrence are at particularly high risk of recurring again. Patients transplanted for fulminant AIH appear to be protected from recurrence. Recurrence can follow reduction of baseline immunosuppression, but this was not the case with our patients. Neither was recurrence more likely in our HLA Dr3-positive recipients. Multi-institutional collaborative studies would allow accrual of enough patients to determine the best posttransplantation immunosuppressive regimens for AIH patients. Gene research may elucidate the etiology of AIH and of posttransplantation recurrence, which would have therapeutic implications. Acknowledgment: The authors are grateful to Nancy Ehrlich for her invaluable editorial assistance. REFERENCES 1. Waldenstrom J. Leber, blutproteine und nahrungseiweiss. Dtsch Gesellsch Verdau Stoffwechselkr 1950;15:113-121. 2. Sanchez-Urdazpal L, Czaja AJ, Van Hoek B, Krom RAF, Wiesner RH. Prognostic features and role of liver transplantation in severe corticosteroid-treated autoimmune chronic active hepatitis. HEPATOLOGY 1991;15: 215-221. 3. Roberts SK, Therneau TM, Czaja AJ. Prognosis of histological cirrhosis in type 1 autoimmune hepatitis. Gastroenterology 1996;110:848-857. 4. Neuberger J, Portmann B, Calne R, Williams R. Recurrence of autoimmune chronic active hepatitis following orthotopic liver grafting. Transplantation 1984;37:363-365. 5. Devlin J, Donaldson P, Portmann B, Heaton N, Tan KC, Williams R. Recurrence of autoimmune hepatitis following liver transplantation. Liver Transpl Surg 1995;1:162-165. 6. Sempooux C, Horsmans Y, Lerut J, Rahier J, Geubel A. Acute lobular hepatitis as the first manifestation of recurrent autoimmune hepatitis after orthotopic liver transplantation. Liver 1997;17:311-315. 7. Wright HL, Bou-Abboud CF, Hassanein T, Block GD, Demetris AJ, Starzl TE, Van Thiel DH. Disease recurrence and rejection following liver transplantation for autoimmune chronic active liver disease. Transplantation 1992;53:136-139. 8. Prados E, Cuervas-Mons V, de la Mata M, Fraga E, Rimola A, Prieto M, Clemente G, et al. Outcome of autoimmune hepatitis after liver transplantation. Transplantation 1998;66:1645-1650. 9. Ratziu V, Samuel D, Sebagh M, Farges O, Saliba F, Ichai P, Farahmand H, et al. Long-term follow-up after liver transplantation for autoimmune hepatitis: evidence of recurrence of primary disease. J Hepatol 1999;20: 131-141. 10. Milkiewicz P, Hubscher SG, Skiba G, Hathaway M, Ellias E. Recurrence of autoimmune hepatitis after liver transplantation. Transplantation 1999;68:253-256. 11. Gotz G, Neuhaus R, Bechstein WO, Lobeck H, Berg T, Hopf U, Neuhaus P. Recurrence of autoimmune hepatitis after liver transplantation. Transplant Proc 1999;31:430-431. 12. Narumi S, Hakamada K, Sasaki M, Freise CE, Stock PG, Roberts JP, Ascher NL. Liver transplantation for autoimmune hepatitis: rejection and recurrence. Transplant Proc 1999;31:1955-1956. 13. Hayashi M, Keeffe EB, Krams SM, Martinez OM, Ojogho ON, So SKS, Garcia G, et al. Allograft rejection after liver transplantation for autoimmune liver diseases. Liver Transpl Surg 1998;4:208-214. 14. Trouillot TE, Shrestha R, Kam I, Wachs M, Everson GT. Successful withdrawal of prednisone after adult liver transplantation for autoimmune hepatitis. Liver Transpl Surg 1999;5:375-380. 15. Czaja AJ. The immunoreactive propensity of autoimmune hepatitis: is it corticosteroid-dependent after liver transplantation? Liver Transpl Surg 1999;5:460-463.
700 REICH ET AL. 16. Johnson PJ, McFarlane IG. Meeting report: International autoimmune hepatitis group. HEPATOLOGY 1993;18:998-1005. 17. Pappas SC. Fulminant hepatic failure. In: Gitnick G, La Brecque DR, Moody FG, eds. Diseases of the Liver and Biliary Tract. St. Louis: Mosby, 1992;615. 18. Hytiroglou P, Thung SN, Gerber MA. Classification and quantitation of the severity of chronic hepatitis: keep it simple! Semin Liver Dis 1995; 15:414-421. 19. Belle SH, Beringer KC, Detre KM. An update on liver transplantation in the United States: recipient characteristics and outcome. In: Cecka JM, Terasaki PI, eds. Clinical Transplants 1995. Los Angeles: UCLA Tissue Typing Laboratory, 1995;19-33.
HEPATOLOGY October 2000 20. Donaldson PT, Doherty DG, Hayllar KM, McFarlane IG, Johnson PJ, Williams R. Susceptibility to autoimmune chronic active hepatitis: human leukocyte antigens DR4 and A1-B8-DR3 are independent risk factors. HEPATOLOGY 1991;13:701-706. 21. Czaja AJ, Strettell MDJ, Thomson LJ, Santrach PJ, Moore SB, Donaldson PT, Williams R. Associations between alleles of the major histocompatibility complex and type 1 autoimmune hepatitis. HEPATOLOGY 1997;25: 317-323. 22. Roberts S, Czaja AJ, Charlton MR, Hay JE, Krom RAF, Poravko MK, Wiesner RH. Recurrent autoimmune hepatitis following orthotopic liver transplantation [Abstract]. HEPATOLOGY 1995;22(Suppl):215A.