Liver transplantation and autoimmune liver diseases


  • Rodrigo Liberal,

    1. Institute of Liver Studies, King's College London School of Medicine at King's College Hospital, London, United Kingdom
    2. Faculty of Medicine, University of Porto, Porto, Portugal
    Search for more papers by this author
  • Yoh Zen,

    1. Institute of Liver Studies, King's College London School of Medicine at King's College Hospital, London, United Kingdom
    Search for more papers by this author
  • Giorgina Mieli-Vergani,

    1. Institute of Liver Studies, King's College London School of Medicine at King's College Hospital, London, United Kingdom
    Search for more papers by this author
  • Diego Vergani

    Corresponding author
    • Institute of Liver Studies, King's College London School of Medicine at King's College Hospital, London, United Kingdom
    Search for more papers by this author

Address reprint requests to Diego Vergani, M.D., Ph.D., F.R.C.P., Institute of Liver Studies, King's College London School of Medicine at King's College Hospital, Denmark Hill, London SE5 9RS, United Kingdom. Telephone: +44 203 2993357; FAX: +44 203 2994224; E-mail:


Liver transplantation (LT) is an effective treatment for patients with end-stage autoimmune liver diseases such as primary biliary cirrhosis, primary sclerosing cholangitis, and autoimmune hepatitis. Indications for LT for these diseases do not differ substantially from those used for other acute or chronic liver diseases. Despite the good outcomes reported, the recurrence of autoimmune liver disease is relatively common in the allograft. In addition, it has become apparent that autoimmunity and autoimmune liver disease can arise de novo after transplantation for nonautoimmune liver disorders. An awareness of the existence of recurrent autoimmune liver diseases and de novo autoimmune hepatitis after LT has important clinical implications because their management differs from the standard antirejection treatment and is similar to the management of classic autoimmune liver diseases in the native liver. Liver Transpl 19:1065-1077, 2013. © 2013 AASLD.


acute cellular rejection


autoimmune hepatitis


anti-mitochondrial antibody


anti-nuclear antibody


anti–liver-kidney microsomal type 1 antibody


autoimmune sclerosing cholangitis




calcineurin inhibitor


de novo autoimmune hepatitis


glutathione-S-transferase T1


hepatocellular carcinoma


human leukocyte antigen


inflammatory bowel disease




liver transplantation


Model for End-Stage Liver Disease


mycophenolate mofetil


primary biliary cirrhosis


primary sclerosing cholangitis


anti–smooth muscle antibody


regulatory T cell


ulcerative colitis


ursodeoxycholic acid

Liver transplantation (LT) is widely accepted as a standard therapeutic approach for the treatment of end-stage acute and chronic liver diseases.[1] Liver diseases with an autoimmune pathogenesis that may require LT are primary biliary cirrhosis (PBC),[2, 3] primary sclerosing cholangitis (PSC),[4] and autoimmune hepatitis (AIH).[3, 5, 6] LT is indicated for PBC, PSC, and AIH patients with end-stage chronic liver disease, hepatic malignancies, or intractable symptoms[7] and for AIH patients presenting with acute liver failure unresponsive to steroid treatment.[8] Overall, autoimmune liver diseases account for approximately 24% of LT procedures performed in Europe and United States,[9] with a 5-year post-LT survival rate around 85%.[10] Transplant recipients, however, can experience various complications, including acute and chronic rejection, primary disease recurrence, chronic hepatitis, cancer, and intrahepatic and/or extrahepatic autoimmunity.[11] It is well established that, despite the use of immunosuppressive drugs after LT, autoimmune liver diseases can recur after LT.[12-14] More recently, it has been reported that AIH and autoimmunity can also arise de novo after LT for nonautoimmune liver diseases.[10, 15] Similarly, patients with PSC and AIH have an increased risk of reactivation or de novo development of ulcerative colitis (UC) after LT.[16]

The aims of this review are to discuss the indications and outcomes of LT for patients with autoimmune liver diseases and to review the latest developments in the understanding of the diagnosis, clinical course, and treatment of recurrent and de novo autoimmune liver diseases after LT.


Primary Biliary Cirrhosis

PBC is the most common liver disorder with a possible autoimmune pathogenesis. Considered by some a typical organ-specific autoimmune disease because of its female preponderance, its specific bile duct pathology, and its hallmark serological signature [anti-mitochondrial antibodies (AMAs)], PBC unexpectedly has a poor response to immunosuppressive treatment.[17-19] PBC is a slowly progressive cholestatic liver disease that is characterized histologically by chronic, destructive, nonsuppurative granulomatous/lymphocytic cholangitis affecting small and medium bile ducts[18] and serologically by the presence of highly disease-specific AMAs. AMAs are found in 90% to 95% of PBC patients and in less than 1% of healthy subjects.[20]

The natural course of PBC differs from patient to patient: although minimal or no progression is observed in some patients over several decades, others progress to liver failure within 10 years of their diagnosis.[21] The natural history of the disease, however, appears to have changed over the last few decades since the introduction and widespread use of ursodeoxycholic acid (UDCA) as a first-line therapy in patients with PBC.[19, 21-23] In support of this, data obtained from the United Network for Organ Sharing show that the number of patients with PBC requiring LT decreased by 20% from 1996 and 2006,[22] with PBC dropping from the most common indication for LT in the United States in the 1980s to the sixth most common.[21]

The indications for LT in PBC patients (Table 1) are mostly similar to the indications in patients with other end-stage chronic liver diseases,[7, 19] with the majority of cases undergoing transplantation because of progressive liver disease leading to decompensated cirrhosis with a Model for End-Stage Liver Disease (MELD) score higher than 16 points.[3] Other indications are hepatocellular carcinoma (HCC) within the Milan/San Francisco criteria, hepatopulmonary syndrome, and portopulmonary hypertension.[30] A small subset of PBC patients with severe noncirrhotic portal hypertension associated with nodular regenerative hyperplasia and/or obliterative portal venopathy benefit from LT.[31] Lastly, intractable symptoms such as refractory pruritus and chronic fatigue may constitute indications for LT in PBC patients,[30, 32] even in the presence of well-preserved liver function,[33] although the latter is a rarer and more controversial indication for LT than the former because fatigue may persist after LT.[59]

Table 1. Autoimmune Liver Disease: Indications for LT, Survival, and Recurrence After Grafting
DiseaseIndications for LT5-Year Patient Survival (%)Recurrence (%)
PBCMELD score > 16382-86[24, 25]0-50[26-29]
Hepatopulmonary syndrome[30]
Portopulmonary hypertension[30]
Obliterative portal venopathy[31]
Refractory pruritus[32, 33]
Chronic fatigue[32, 33]
PSCMELD score > 16485-86[34, 35]15-30[13, 35-37]
HCC[29, 38]
Intractable pruritus[4, 30]
Recurrent bacterial cholangitis[4, 30]
Limited stage cholangiocarcinoma[4]
AIHMELD score > 16[3, 39]80-90[40]12-46[41-58]
Hepatopulmonary syndrome[30]
Portopulmonary hypertension[30]
Fulminant liver failure[39]

A combination of corticosteroids (withdrawn over the course of 3 months), a calcineurin inhibitor (CNI), and mycophenolate mofetil (MMF) or azathioprine is the most common immunosuppressive regimen used after LT and leads to very successful outcomes.[3] In a retrospective series of 301 transplant recipients with PBC from the United Kingdom, the 1-, 3-, and 5-year actuarial patient survival rates were 94%, 91%, and 82%, respectively, and the graft survival rates were 85%, 83%, and 75%, respectively.[24] Similarly, a retrospective analysis of the United Network for Organ Sharing database for patients undergoing transplantation between 2002 and 2006 showed that among PBC transplant recipients, the estimated 1-, 3-, and 5-year patient survival rates were 93%, 90%, and 86%, respectively, for living donor LT and 90%, 87%, and 85%, respectively, for deceased donor LT; the estimated graft survival rates at 1, 3, and 5 years were 86%, 81%, and 77%, respectively, for living donor LT and 85%, 83%, and 81%, respectively, for deceased donor LT.[25]

Primary Sclerosing Cholangitis

PSC is a chronic cholestatic liver disease of unknown cause characterized by progressive inflammation and fibrosis of intrahepatic and/or extrahepatic bile ducts that leads to bile duct obliteration with the formation of multifocal bile duct strictures.[60, 61] PSC affects mainly males. It is not considered a classic autoimmune disease, although it shares many features characteristic of immune-mediated diseases.[62] PSC is a progressive disorder that often leads to cirrhosis, portal hypertension, and liver failure.[63] An important aspect of PSC is its frequent association with UC and, more generally, inflammatory bowel disease (IBD).[64, 65] In fact, PSC develops in 2.4% to 7.5% of patients with IBD, and up to 85% of patients with PSC ultimately develop IBD.[66] Prospective studies conducted in patients with IBD suggest that the liver disease progresses regardless of the severity of intestinal manifestations. PSC is associated with an increased risk for the development of hepatobiliary (cholangiocarcinoma or gallbladder carcinoma) and colorectal malignancies.[67] To date, no medical treatment has been proven to positively affect survival, and LT remains the only curative option for PSC patients with end-stage liver disease.[64]

As with cirrhosis due to other etiologies, LT is indicated for PSC patients progressing to end-stage liver disease (ie, those with a MELD score higher than 16).[4, 68] HCC can occur in patients with cirrhosis and PSC, and LT prioritization for these patients follows the same rule used for other patients with cirrhosis and HCC.[30, 38] Indications for LT in PSC patients that exempt them from the MELD score include intractable pruritus, recurrent bacterial cholangitis in the presence of dominant bile duct strictures untreatable endoscopically, and the presence of limited-stage cholangiocarcinoma[4, 60, 63] (Table 1). A select group of patients with solitary hilar cholangiocarcinoma who have undergone chemotherapy, radiotherapy, and staging laparotomy have a particularly high success rate after LT.[69, 70] Those PSC patients with a solitary tumor less than 1 cm in diameter within the explant have LT outcomes similar to those of patients without cholangiocarcinoma.[34]

The outcomes of LT for PSC are usually good.[33] Because the disease frequently affects the common bile duct and may recur after LT, the preferred biliary anastomosis in these patients is a Roux-en-Y choledochojejunostomy.[71] Previous biliary surgery or prior colectomy does not appear to affect LT outcomes,[34, 72] although colectomy may prevent the recurrence of PSC after LT (discussed later). A study of the outcomes of LT for PSC performed in 127 patients between 1984 and 1996 reported 1-, 2-, and 5-year actuarial patient survival rates of 90%, 86%, and 85%, respectively, and graft survival rates of 82%, 77%, and 72%, respectively.[34] A Mayo Clinic study of 150 transplant patients with PSC reported patient survival rates of 94%, 92%, 86%, and 70% and graft survival rates of 83%, 83%, 79%, and 61% at 1, 2, 5, and 10 years, respectively.[35]

In pediatrics, sclerosing cholangitis is often associated with florid autoimmune features, including elevated titers of anti-nuclear antibodies (ANAs) and/or anti-smooth muscle antibodies (SMAs), elevated immunoglobulin G levels, and interface hepatitis on histology.[73] This condition, often associated with IBD, is called autoimmune sclerosing cholangitis (ASC) and is as prevalent as AIH type 1 in childhood, but in contrast to AIH, it affects boys and girls equally. ASC responds satisfactorily to immunosuppression with steroids and azathioprine with respect to the parenchymal inflammation, but the bile duct disease progresses in 50% of cases, and LT is ultimately required.[73]

Autoimmune Hepatitis

AIH accounts for 2% to 3% of pediatric LT procedures and 4% to 6% of adult LT procedures performed in Europe and the United States.[74] Like PBC, AIH is characterized by a female preponderance and the presence of circulating autoantibodies, although the autoantibody profile differs between patients and allows a distinction between 2 types.[75] AIH type 1 is defined by the presence of ANAs and/or SMAs, whereas AIH type 2 is characterized by positivity for anti–liver-kidney microsomal type 1 (anti-LKM1) or anti–liver cytosol type 1 autoantibodies.[76] Histologically, AIH is characterized by a dense portal and periportal infiltrate rich in plasma cells, interface hepatitis, and central perivenulitis.[77]

Information on the natural course of AIH mostly stems from studies conducted more than 30 years ago when the last placebo-controlled treatment trial for AIH was published.[78-81] These studies showed that without treatment, as many as 40% of patients would die within the first 6 months after their diagnosis. In the immunosuppressive treatment era, the 10-year overall survival rate for patients with AIH has been calculated to range from 80% to 93%.[82-86]

LT is indicated for AIH patients presenting with acute liver failure not responsive to immunosuppressive treatment or end-stage chronic liver disease and for those with HCC who meet the transplant criteria.[3, 39] Although patients with a chronic presentation of AIH typically respond well to immunosuppressive treatment, some 10% will eventually require LT. There is no single predictor for the necessity of LT, but those patients who fail to achieve remission after 4 years of therapy are the most common candidates.[87] The indications for LT for end-stage chronic AIH are similar to those for PBC and other end-stage liver diseases[41] (Table 1).

Among the autoimmune liver diseases discussed in this review, only AIH may present as acute liver failure and, therefore, may require urgent LT.[88] The management of patients with fulminant AIH is challenging because the available data are scarce and are based on small retrospective studies. On the basis of these data, it has been calculated that up to 19.8% of AIH patients may present with fulminant acute liver failure.[8, 88-91] Even though one-third of these patients will respond to corticosteroids, for the vast majority, LT is the only hope for survival. Among all the factors analyzed, the following were associated with an increased need for LT: a MELD score higher than 28 on admission, the presence of massive necrosis on histology, no improvement in bilirubin and INR levels within the first 4 days of treatment, and no change in the MELD-sodium score after day 7 of corticosteroids.[92, 93] In such cases, continuing corticosteroid therapy may be a futile exercise because it is most unlikely to alter the course of the disease and, on the contrary, could result in serious adverse events (particularly sepsis).[88, 94] Thus, it is of the utmost importance to closely follow the changes in the clinical and analytical parameters of disease activity during corticosteroid therapy because a failure to respond or a worsening of such indices within the first week of treatment justifies consideration for LT.[92, 93] If therapy with corticosteroids is maintained in these patients, it is imperative to actively and concomitantly pursue the assessment for LT.

A combination of prednisolone and a CNI is the most common immunosuppressive regimen used after LT[74] and leads to very successful outcomes, with reported 5- and 10-year patient survival rates of 80% to 90% and 75%, respectively,[95] and 1- and 5-year graft survival rates of 84% and 75%, respectively.[40, 95]


Recurrent Primary Biliary Cirrhosis

In 1982, Neuberger et al.[12] reported for the first time the recurrence of PBC after LT, but this remained a controversial issue until more recent studies confirmed that PBC does recur in patients undergoing deceased or living donor LT.[26-28, 96-98] The reported rate of disease recurrence ranges from 0% to 50%, and this discrepancy has been attributed to a decline in the number of per-protocol biopsies performed in patients undergoing transplantation for PBC.[26, 29] The reported median time from LT to the diagnosis of recurrent PBC is 3 to 5.5 years,[27] although the disease may recur even in the first year after LT.[99]

The diagnosis of PBC in the allograft is more challenging than the diagnosis in the native liver. After LT, high levels of AMAs and immunoglobulin M may persist, and other PBC-associated disorders may recur or arise de novo; this suggests that the underlying immune defect is not corrected by the removal of the diseased liver or by immunosuppression.[10] The best way of diagnosing PBC recurrence in the allograft is liver histology, and granulomatous bile duct injury is an infrequent but highly specific finding (Fig. 1).[26] Without a granulomatous reaction, cholangitis per se in recurrent PBC is sometimes reminiscent of acute cellular rejection (ACR),[100] but an association with periportal copper-associated protein deposition is more in keeping with recurrent PBC, whereas an additional finding of endotheliitis favors cellular rejection. The criteria for the diagnosis of recurrent PBC are summarized in Table 2.

Figure 1.

Histological features of recurrent PBC. The bile duct (arrow) is damaged and surrounded by a granulomatous reaction (granulomatous cholangitis; arrowheads).

Table 2. Criteria for the Diagnosis of Recurrent PBC
  1. NOTE: A definite diagnosis requires the fulfillment of all clinical and laboratory criteria plus at least 3 of the 4 histological features. A probable diagnosis requires the fulfillment of all clinical and laboratory criteria plus 2 histological features. This table has been adapted from Journal of Hepatology.[96] Copyright 1993, European Association for the Study of the Liver.

LT for PBC
Characteristic histological features of PBC
Lymphoplasmacytic infiltrate
Lymphoid aggregates
Epithelioid granulomas
Evidence of bile duct injury
Persistence of AMAs
Elevated immunoglobulin M level
Exclusion of other causes of graft dysfunction

Post-LT immunosuppressive regimens have been suggested to affect the frequency and timing of disease recurrence. Transplant patients on tacrolimus are more likely to experience earlier disease recurrence than patients on cyclosporine.[97, 101] Rapid weaning from antirejection drugs has been reported to favor recurrence.[102] Other reported risk factors for PBC recurrence include a young donor age, an old recipient age, and a long cold ischemia time[103, 104] (Table 3). It should be noted, however, that the reproducibility and validity of these variables remain to be validated.[99, 118]

Table 3. Risk Factors for the Development of Recurrent PBC, PSC, and AIH and d-AIH After LT
DiseaseRisk Factors
Recurrent PBCTacrolimus-based immunosuppression[97, 101]
Rapid weaning from antirejection drugs[102]
Young donor age[103, 104]
Old recipient age[103, 104]
Long cold ischemia time[103, 104]
Recurrent PSCOld recipient age[105]
Male recipient[106]
Recurrent ACR[107]
OKT3 therapy for steroid-resistant ACR[107]
CMV infection[106]
Presence of cholangiocarcinoma before transplantation[108]
Persistent UC requiring maintenance steroids[109]
Recurrent AIHHLA-DR3 or HLA-DR4 recipient[43, 51]
Weaning from corticosteroids[41, 110, 111]
Severe necroinflammatory activity in native liver[49, 112]
d-AIHHLA-DR3 or HLA-DR4 recipient or donor[113, 114]
Previous episodes of ACR[115]
Steroid dependence[116]
Female donor[117]
Older donor age[117]
Tacrolimus-based immunosuppression[117]

Although supportive data are scarce, treatment with UDCA is generally recommended after the diagnosis of recurrent PBC[99] (Table 4). A recent study reported that more than 50% of PBC patients receiving UDCA after LT experienced a complete normalization of serum alkaline phosphatase and alanine aminotransferase levels over a period of 36 months, whereas only 22% of untreated patients did. However, the same study showed that UDCA did not affect patient and graft survival.[119]

Table 4. Treatment of Recurrent PBC, PSC, and AIH and d-AIH After LT
Recurrent PBCUDCA[99]Complete normalization of alkaline phosphatase levels in >50% of patients[119]
No impact on graft or patient survival[119]
Recurrent PSCUDCA[120]Improvement in biochemical indices of liver disease[120]
Impact on graft and patient survival uncertain[120]
Recurrent AIHReintroduction (or increase in dose) of prednisolone and azathioprine[41]Graft and patient survival[51]
d-AIHPrednisolone alone or in combination with azathioprine or MMF[121]Graft and patient survival[121]

Most published series show that the impact of recurrent disease on long-term survival after LT is modest. In a series of 486 PBC patients undergoing transplantation, only 3 needed retransplantation as a result of recurrent disease.[122] Another study analyzing the survival of 100 patients undergoing transplantation for PBC with a median follow-up of 118 months reported 5-, 10-, and 15-year patient survival rates of 87%, 84%, and 82%, respectively; only 2 patients developed organ dysfunction due to PBC recurrence.[103]

Recurrent Primary Sclerosing Cholangitis

The recurrence of PSC after LT has also been reported. Its prevalence ranges from 15% to 30% among LT recipients, and the median time from LT to its development is approximately 3 to 5 years.[13, 35-37] The diagnosis of recurrent PSC is often challenging because it may be difficult to distinguish it from de novo secondary cholangitis.[123] Causes that may account for the development of nonanastomotic biliary strictures in the allograft include chronic ductopenic rejection, ischemic biliary insults (especially hepatic artery thrombosis), bacterial or fungal cholangitis, and ABO incompatibility between the donor and the recipient.[9, 124] Thus, the diagnosis of recurrent PSC relies not only on a radiological demonstration of diffuse biliary stricturing but also on the exclusion of all other possible causes of this complication (Table 5). Bile duct strictures may be demonstrated by magnetic resonance cholangiography or percutaneous transhepatic cholangiography. Because most patients undergoing transplantation for PSC have a Roux-en-Y loop rather than a duct-to-duct anastomosis, endoscopic diagnosis is difficult.[125] The diagnosis may also be supported by liver histology, although the pathognomonic histological finding of fibro-obliterative bile duct lesions or periductal concentric fibrosis is seen in fewer than 10% of patients with disease recurrence (Fig. 2A).[123, 126] Other chronic cholangiopathic features, including periportal copper-associated protein deposition and biliary-type interface activity, also support a diagnosis of recurrent PSC in an appropriate clinical context (Fig. 2B).

Figure 2.

Histological features of recurrent PSC. (A) The bile duct displays characteristic periductal concentric fibrosis. (B) The portal tract is enlarged by a bile ductular reaction and biliary-type interface activity, and these features are suggestive of recurrent PSC.

Table 5. Mayo Clinic Criteria for the Diagnosis of Recurrent PSC
Inclusion CriteriaExclusion Criteria
  1. NOTE: This table has been adapted from Hepatology.[35] Copyright 1999, American Association for the Study of Liver Diseases.

LT for PSCHepatic artery thrombosis or stenosis
ANDEvidence of ductopenic rejection
Cholangiographic evidence of intrahepatic/extrahepatic biliary stricturing more than 90 days after LTNonanastomotic strictures less than 90 days after LT
ORABO incompatibility between donor and recipient
Histological evidence of 
Fibrous cholangitis 
Fibro-obliterative lesions with or without ductopenia, fibrosis, or cirrhosis 
Interface hepatitis 

Many variables, including recipient age and sex,[105, 106] recurrent ACR, OKT3 therapy for steroid-resistant ACR,[107] cytomegalovirus (CMV) infection,[106] the presence of cholangiocarcinoma before LT,[108] and persistent UC requiring maintenance steroids, have been associated with an increased risk for PSC recurrence (Table 3). On the other hand, it has been shown that in those patients with concomitant UC, colectomy before or during LT confers protection against the development of recurrent PSC.[109]

There is no established treatment for PSC recurrence after LT.[60] UDCA has been advocated because it improves biochemical indices of liver disease, but whether its use has an impact on outcomes remains uncertain[120] (Table 4). Dominant strictures should be dilated by interventional cholangiographic means whenever possible.[120] The impact of recurrent PSC on graft survival remains controversial, with some studies reporting no effect[127] and others reporting diminished graft survival.[42, 108]

Recurrence of disease after LT is very common (∼70%) in patients with ASC, the juvenile autoimmune form of sclerosing cholangitis, particularly if they have concurrent IBD, and it is associated with seriously compromised graft survival.[128]

Recurrent Autoimmune Hepatitis

Although LT is a highly successful mode of treatment for AIH, the primary disease recurs in approximately 30% of patients.[14, 129] The reported recurrence rate for AIH after LT ranges from 12% to 46% and depends on the diagnostic criteria, immunosuppressive regimen, length of follow-up, and performance of per-protocol biopsy.[41-58] Recurrence has been reported in adult and pediatric[45] populations, with a mean time to recurrence of 4.6 years,[74] but it may occur as early as 35 days after LT,[49] with the rate increasing with the postsurgical interval.

The diagnosis of recurrent AIH is based on the reappearance of clinical symptoms and signs, positivity for autoantibodies, and interface hepatitis on histology along with elevated transaminases and immunoglobulin G levels and responses to prednisolone and azathioprine[54, 74] (Table 6). Histological features of AIH may precede clinical and biochemical evidence of recurrence.[52] The aforementioned criteria are those published by the International Autoimmune Hepatitis Group[130-132] to diagnose AIH in the native liver, and they have not been tested systematically for the diagnosis of AIH recurrence in the graft. However, because of the similarity between spontaneous AIH and AIH recurrence after LT, they are a useful diagnostic tool.

Table 6. Diagnosis of Recurrent AIH
LT for AIH
Elevation of transaminases
Interface hepatitis
Elevation of immunoglobulin G
Presence of autoantibodies (ANA, SMA, and/or anti-LKM1)
Corticosteroid dependency
Exclusion of other causes of graft dysfunction

There are no consistent predictors of AIH recurrence, but some factors conferring a predisposition to recurrence have been reported. Although this has not been universally confirmed, possession by the recipient of human leukocyte antigen DR3 (HLA-DR3) or HLA-DR4 has been associated with a risk of recurrence.[43, 51] Caution should be exercised in weaning patients off immunosuppression because recurrence has been associated with the discontinuation of corticosteroids.[41, 110, 111] Ayata et al.[49] and others[112] found that the presence of severe necroinflammatory activity in the native liver at the time of LT is a predictor of AIH recurrence. AIH appears to be less likely to recur in patients presenting with fulminant hepatic failure versus patients with a chronic presentation,[39] and this suggests that chronic patients requiring LT may have propensity for immunosuppressive resistance, whereas patients with fulminant hepatitis, naive to immunosuppression, are more likely to respond. At variance with early studies showing an association between tacrolimus and AIH recurrence,[47, 49] a recent systematic review[129] reported that primary immunosuppression with either cyclosporine or tacrolimus did not influence the risk of recurrence (Table 3).

Most patients with recurrent AIH respond to the reintroduction (or to an increase in the dose) of corticosteroids and azathioprine, which should be implemented as soon as the diagnosis is made[10, 41, 51] (Table 4). In treatment failure, alternatives that have been tried with success include the addition of MMF to the standard therapeutic regimen,[74] the replacement of tacrolimus with cyclosporine,[133] and the replacement of a CNI with sirolimus.[134] The successful management of recurrent AIH relies greatly on its early diagnosis.[135]


Posttransplant de novo autoimmune hepatitis (d-AIH) was initially described in 1998, and in contrast to the recurrence of the disease in patients undergoing transplantation for AIH, this condition affects patients undergoing transplantation for disorders other than AIH.[135] In the first report of d-AIH,[113] 7 children (4% of 180 LT recipients) over a 5-year period developed a form of graft dysfunction with features identical to those of classic AIH: hypergammaglobulinemia, positivity for circulating autoantibodies (ANAs in 1 patient, ANAs and SMAs in 2 patients, gastric parietal cell antibodies in 1 patient, and atypical anti-LKM1 in 3 patients), and histological features of chronic hepatitis with portal and periportal inflammation (Fig. 3A). As in classic AIH, plasma cell–rich central perivenulitis is commonly seen in d-AIH (Fig. 3B).

Figure 3.

Histological features of d-AIH. (A) The portal tract is enlarged with a plasma cell–rich inflammatory infiltrate and interface activity (arrows). (B) Perivenular confluent hepatocyte loss is noted. The inflammatory infiltrate contains several plasma cells (central perivenulitis; arrow).

The index case did not respond to a short course of high-dose steroids for rejection but instead responded only to the classic treatment for AIH. None of the children underwent transplantation for autoimmune liver disease, none were hepatitis C virus–positive, and all had serum concentrations of cyclosporine A or tacrolimus within therapeutic antirejection levels at the time of the diagnosis of d-AIH.[113] Since that report, several other groups have reported the occurrence of d-AIH after LT in children, with the prevalence ranging from 2.35% to 6.2%; the indications for LT so far reported include extrahepatic biliary atresia, Alagille syndrome, acute liver failure, alpha-1-antitrypsin deficiency, primary familial intrahepatic cholestasis, PSC, and Budd-Chiari syndrome.[115, 116, 136-139]

d-AIH has also been described in the adult population, although at a lower prevalence than that seen in children. The original indications for LT have included PSC, PBC, alcoholic cirrhosis, hepatitis C cirrhosis, Wilson's disease, and acute liver failure.[114, 121, 140, 141]

Several reports have investigated whether the development of d-AIH is associated with the possession of specific major histocompatibility complex antigens by either the recipient or the donor. In the original report, 5 of the 7 children received livers from donors who were HLA-DR3–positive or HLA-DR4–positive.[113] In adults, Heneghan et al.[114] found HLA-DR3 or HLA-DR4 in either donors or recipients in all cases, and Salcedo et al.[121] noted an overrepresentation of DR3 in recipients. In an attempt to identify possible risk factors leading to d-AIH, Miyagawa-Hayashino et al.[115] showed that in 69% of patients, at least 1 episode of ACR had been identified before the development of d-AIH; however, other series reported that d-AIH was preceded by ACR in only 20% to 50% of patients.[114, 136, 139] Venick et al.[116] in a matched case-control study found that previous episodes of ACR and steroid dependence constituted risk factors for the development of pediatric d-AIH. More recently, a study investigating possible risk factors for the development of d-AIH in adults suggested that grafts from female and older donors and tacrolimus treatment of the recipient were factors associated with a higher risk for d-AIH[117] (Table 3).

An awareness that treatment with prednis(ol)one alone or in combination with azathioprine or MMF is successful in d-AIH has led to excellent graft and patient survival[121] (Table 4). It is interesting that these patients do not respond satisfactorily to short courses of high-dose steroids for cellular rejection but respond only to the standard treatment for classic AIH; this makes it essential to make an early diagnosis to prevent graft loss. Children should be given a starting dose of 1 to 2 mg/kg prednis(ol)one (without exceeding a daily dose of 60 mg) in combination with azathioprine (1-2 mg/kg); the steroids should then be tapered over 4 to 8 weeks to reach a maintenance dose of 5 to 10 mg/day. Adults are generally given an initial dose of 30 mg prednis(ol)one in combination with 1 to 2 mg/kg azathioprine per day. The dose of the steroid is reduced according to the response to a maintenance dose of 5 to 10 mg/day. In the absence of a response, azathioprine should be replaced by MMF.[135] The importance of maintenance therapy with steroids was shown in a study comparing treatment with steroids and treatment without steroids: although all steroid-untreated patients developed cirrhosis and either died or required retransplantation, none of the steroid-treated patients had progressive disease.[121]


UC occurs in up to 85% of patients with PSC and in up to 20% of patients with AIH.[66, 142] The exacerbation of preexistent UC has been described after LT for PSC or AIH, and it appears to have a more aggressive course than UC before LT.[16, 143-145] IBD can also develop de novo after LT.[16] In one study,[146] the use of tacrolimus conferred an increased risk for UC, whereas azathioprine exerted a protective effect, but a confounding factor was that patients using cyclosporine were also treated with azathioprine and patients using tacrolimus were not. Another study[147] reported that CMV donor-recipient mismatch confers an increased risk for IBD after LT; however, CMV infection itself can cause a colitis that closely resembles UC and may thus also constitute a confounding factor. The management of IBD and particularly UC after LT is not well established, and the risks and benefits of treatment with biological agents in this setting remain to be determined.[60] The rate of proctocolectomy for intractable IBD is higher after LT than before transplantation.[143] Because the cumulative risk for the development of recurrent or de novo UC after LT increases over the first 10 years after LT and because there is a high risk for the development of colonic neoplasia, careful observation and annual surveillance with colonoscopy are recommended for transplant recipients with concomitant UC.[16, 35, 148, 149]


The pathogenic mechanisms underlying post-LT autoimmunity remain to be conclusively identified, although there are a number of hypotheses for the recurrence of autoimmunity after LT. It has been suggested that a pool of memory cells, sensitized to species-specific antigens, are restimulated upon exposure to autoantigens presented to the recipient's T cells by either recipient or donor antigen-presenting cells sharing histocompatibility antigens with the recipient.[10]

There are several nonmutually exclusive mechanisms by which autoimmunity and AIH could arise de novo in patients undergoing transplantation for nonautoimmune conditions.[135] First, autoantigens could be released from tissue damaged during transplantation. Second, molecular mimicry, the process by which immune responses to external pathogens become directed toward structurally similar self-components, could occur.[150] In support of this hypothesis, in one study,[121] all cases of d-AIH were associated with CMV, Epstein-Barr virus, or Parvovirus infection. Viral infections could also trigger autoimmunity via other mechanisms, including polyclonal stimulation, the enhancement and induction of membrane expression of major histocompatibility complex class I and II antigens, and/or interference with immunoregulatory cells.[10, 135]

An alternative hypothesis has been proposed in light of animal studies investigating the impact of CNIs on T cell maturation and/or regulatory T cell (Treg) development. CNI-associated autoimmunity has been described in animals treated neonatally and in animals immunocompromised by irradiation;[151] therefore, patients may be predisposed to developing autoimmunity through the influence of CNI use after LT. Cyclosporine blocks the activation-induced cell death of effector T cells and interferes with tolerance induction by costimulation blockade,[152] and it reduces the production of interleukin-2 (IL-2).[153] IL-2 is essential for Treg homeostasis and survival, and the absence of IL-2 has been associated with impaired Treg function.[154] A low number of Tregs has also been reported in LT recipients after the development of ACR.[155] In keeping with the observation that Treg impairment is inversely correlated with the production of autoantibodies such as anti-soluble liver antigen and anti-LKM1 in classic AIH,[156] it is possible that Treg impairment contributes to the onset and perpetuation of an autoreactive immune response after transplantation and, therefore, to the development of d-AIH.

Finally, glutathione-S-transferase T1 (GSTT1) antibodies have been reported in d-AIH patients.[157] GSTT1 is a drug-metabolizing enzyme expressed at high levels within the liver and kidneys. It is encoded by a gene that is absent in 20% of Caucasians. It has been demonstrated that the transplantation of a graft from a GSTT1-positive donor into a GSTT1-null recipient triggers the production of anti-GSTT1 antibodies.[158] This suggests that GSTT1 donor/genotype mismatch and the presence of anti-GSTT1 antibodies constitute risk factors for the development of d-AIH.[159] Other studies, however, show that a lack of the GSTT1 gene and anti-GSTT1 antibodies do not account for the development of d-AIH in most cases; therefore, alternative mechanisms, including those mentioned previously, must be involved in the development of d-AIH.[160]


The criteria for LT in patients with autoimmune liver diseases are similar to those for patients with other chronic liver disorders: LT is indicated for PBC, PSC, and AIH patients with end-stage chronic liver damage and for AIH patients presenting with fulminant liver failure. The results of LT for these conditions are excellent despite the relatively high disease recurrence rate. The early diagnosis of disease recurrence is essential for appropriate life- and graft-saving treatment. Autoimmunity and autoimmune liver disease can arise de novo after LT for nonautoimmune liver disorders. d-AIH should be suspected and investigated in all patients with graft dysfunction not attributable to rejection or surgical complications and, once diagnosed, should be treated promptly. d-AIH does not respond satisfactorily to classic antirejection regimens but responds only to the standard treatment for AIH. IBD may occur after LT for whatever indication, but it particularly affects patients undergoing transplantation for PSC and AIH, in whom it may represent a flare-up of a disease process present before LT. Because the risk of developing IBD increases over time, close surveillance is warranted.