Since the first report in 1984 [3], recurrent AIH is being increasingly recognised with nearly 20 publications [4–24] reporting recurrence rates of 20–30 % (Table 6.2). The other post-transplant condition resembling autoimmune hepatitis in patients grafted for other indication has been termed de novo AIH (dnAIH ) [25]. The median reported time to recurrence varies between 15 and 52 months [11, 14, 16–18, 26]. The length of follow-up also appears to be important in assessing the probability of rAIH. The rate of recurrence has been quoted at 8 % at 1 year and as much as 68 % after 5 years [16]. The median time to reported recurrence varies between series but is around 2–5 years. Studies following patients over a longer period have revealed that the risk of recurrent disease persists even over 10 years post-transplantation with the two patients developing severe recurrence at 10 and 14 years post-transplantation [10]. As protocol biopsies were not performed in all studies, there is the possibility that recurrent AIH is under-reported. This may be related to the tendency to reduce immunosuppression over time. Whether the use of current immunosuppression , particularly tacrolimus has an impact on long-term risk is uncertain, although the evidence so far is that the choice of calcineurin inhibitor (CNI) is not a risk factor for recurrence [23].

Many authors have used the criteria developed by the International Autoimmune Hepatitis Group (IAHG ) [27] to define rAIH: this is inappropriate, the IAHG criteria were developed to ensure that there was a consistent approach taken in clinical studies and extrapolation of this definition to the allograft is inappropriate and potentially misleading. The differential diagnosis of the histological findings in the allograft is much wider than in the native liver, and other causes of graft damage must be excluded. In contrast to the patient with a native liver, the liver allograft recipient will be taking immunosuppressive and other drugs, there is usually a different HLA (human leukocyte antigen ) and other antigenic environment, rejection itself is associated with the presence of autoantibodies, and there are many other causes of potential graft damage which, of course, may coexist with rAIH.

The role of autoantibodies in the diagnosis: Autoantibodies may be present in low titre post-transplant and do not necessarily correlate with histological features of graft inflammation [28]. Furthermore, histological abnormalities can precede changes in biochemical and immunological tests [10]. Changes in serum aminotransferases do not correlate with chronic hepatitis in children following transplantation [29], and biochemical improvement does not always correlate with histological remission [16]. Assessing response to the treatment of recurrent disease is probably best served by liver biopsy, since liver tests do not correlate with liver histology and significant histological inflammation can be present with normal biochemistry.

The role of protocol liver biopsy is controversial, and the risks of biopsy have to be balanced against the potential benefits associated with earlier and more effective treatment to prevent graft damage. Furthermore, interpretation of graft histology can be challenging and distinction between rAIH and allograft can be difficult, although there are some typical features in both as detailed in Table 6.3 [28]. One of the earliest findings is that of lobular lymphoplasmacytic hepatitis with acidophil bodies [6].

Rigamonti [30] suggested that transient elastography (TEG) may be helpful in diagnosing recurrent disease but the numbers in their series were very small, and there was no comparison with other causes of allograft damage. However, routine use of TEG may help guide the need for liver biopsy.