Article Text

Download PDFPDF

Corticosteroid treatment for autoimmune pancreatitis
  1. Rahul Pannala,
  2. Suresh T Chari
  1. Fiterman Center for Digestive Diseases, Mayo Clinic, Rochester, Minnesota, USA
  1. Correspondence to Dr Suresh T Chari, 200 First ST SW, Rochester, MN 55905, USA; chari.suresh{at}

Statistics from

Autoimmune pancreatitis (AIP) is the only pancreatic disorder responsive to steroid treatment. AIP has recently been subclassified into type 1 and type 2 AIP.1 Type 1 AIP is the pancreatic manifestation of immunoglobulin G4 (IgG4)-associated systemic disease (ISD), a fibroinflammatory autoimmune disorder that involves multiple organ systems and is characteristically associated with elevation of serum levels of IgG4.2 Commonly used diagnostic criteria for AIP include the Asian Consensus diagnostic criteria3 and the Mayo Clinic HISORt criteria4 (histology, imaging, serology, other organ involvement and response to corticosteroid treatment). Treatment protocols for AIP are still evolving, but corticosteroids are generally considered to be very effective in the initial inflammatory phase of the disease. Several small case series have reported that corticosteroids are effective in inducing remission in AIP, but larger series and controlled trials are lacking.

In this issue of Gut (see page 1504), Kamisawa and colleagues5 report results from a large, multicentre, retrospective survey of 563 patients with AIP treated at 17 referral centres in Japan. Their study highlights the fact that though a substantial proportion of patients with AIP may achieve spontaneous remission (77/104 patients, 74%), corticosteroids are also highly effective in inducing remission (451/459 patients, 98%). This study, despite the limitations of a retrospective and cross-sectional study design, provides the best evidence to date of the efficacy of corticosteroid treatment in the initial presentation of AIP. However, our knowledge of the natural history of AIP, predictors of relapse, need for maintenance treatment and the optimal type of maintenance treatment is still limited. The objective of this commentary is to provide a brief overview of the issues associated with corticosteroid treatment in AIP.

Rationale for steroid treatment of initial presentation of AIP

The goals of treatment in the initial inflammatory phase of AIP are symptom alleviation (eg, jaundice and abdominal discomfort), improvement of radiological and biochemical parameters and confirmation of diagnosis in select patients receiving a diagnostic steroid trial. Corticosteroids are very effective in alleviating symptoms and induce remission more quickly than no treatment.6 7 Patients with obstructive jaundice often undergo diagnostic biliary evaluation and biliary stenting to relieve jaundice. Steroids allow rapid removal of the stent by causing resolution of the biliary stricture. A clinical and radiological response to steroids is usually seen in 2–4 weeks.7 A rapid response to corticosteroid treatment is not only reassuring, but also helps substantiate the diagnosis of AIP in selected patients.4 For all the above reasons corticosteroids should be considered in all patients with AIP with active disease.

Steroid regimen for induction of remission in AIP

The optimal dose of corticosteroids, the duration of initial treatment and the assessment of clinical, serological and radiological response have not been evaluated in a rigorous fashion. In a consensus statement in the study by Kamisawa et al,5 the authors suggest initiating treatment with prednisolone at 0.6 mg/kg/day, tapering gradually to a maintenance dose of 5 mg/day over 3–6 months, and then continuing maintenance steroids for at least 6 months and possibly up to 3 years. In our practice, we most commonly give prednisone 40 mg/day for 4 weeks, repeat laboratory tests and imaging in 4–6 weeks after initiating treatment, and, if there is clinical and radiological response, we taper by 5 mg every week to complete a treatment course of 11 weeks.6 By then most patients have had their biliary stents removed and are clinically asymptomatic. We do not routinely put patients on maintenance treatment.

The two different approaches to treatment of AIP raise several questions. Is there a benefit to treating patients with a prolonged corticosteroid taper as compared with treating for a shorter duration, especially as many patients have symptomatic relief within a matter of a few weeks? In the study by Kamisawa et al, the relapse rate even among patients who were not treated with steroids was <40%. In other words, if 60% of patients do not have a relapse even without any treatment, do we need to treat all patients with corticosteroids for an extended duration? Future studies should help understand the risks of relapse versus risk of steroid-induced complications.

Disease relapse

Relapses occur in 30–40% of AIP. The relapse may involve different organ systems compared with the initial presentation. For example, a patient who initially presents with a pancreatic tail mass may present during relapse with intrahepatic biliary strictures and no active pancreatic disease. Relapses occur in patients experiencing spontaneous resolution, following surgical resection for suspicion of cancer or after stopping steroid treatment. Does corticosteroid treatment of initial presentation reduce the likelihood of future relapse? In the study by Kamisawa et al, steroid-treated patients were less likely to relapse compared with patients with AIP not treated with corticosteroids (110/451 patients, 24% vs 32/77, 42%, p = 0.003).5 However, the vast majority (82%) of the steroid-treated patients continued to receive maintenance corticosteroids. Therefore, it is unclear if an initial course of corticosteroids truly decreases the relapse rate. Most relapses occurred in the first 3 years following initial diagnosis and were corticosteroid responsive.

What predicts relapse in AIP? In the study by Kamisawa et al,5 the authors suggest that serial IgG4 measurements are helpful in identifying early relapse. However, in this study 115/182 (63%) of patients with AIP failed to normalise IgG4 levels with treatment and 37/54 (69%) developed serological relapse during maintenance treatment.5 Only 30% of patients with persistent IgG4 elevation relapsed, whereas 10% of those who normalised IgG4 levels relapsed. These data suggest that lack of a serological relapse is helpful in predicting continued clinical remission. However, the majority of patients with AIP will have elevated IgG4 levels on withdrawal of steroids, of whom only a minority will develop a clinical relapse.

In our experience the only clinical predictor of relapse has been the presence of extrapancreatic biliary involvement.8 In our study, the relapse rate in patients with intrahepatic or proximal extrahepatic biliary disease was 65% compared with 25% in those without proximal biliary disease. Therefore, in patients with IgG4-associated cholangitis (IAC), we typically monitor liver tests every 12 weeks for the first 1–2 years.8 In others, we generally evaluate for relapse based on symptoms.

Treatment of relapse and maintenance of remission

Corticosteroids are just as effective in treating relapses as they are for the initial presentation. Unlike our practice of using maintenance treatment only in those who relapse after an initial 11 week course of steroids, the Japanese practice has been to use maintenance steroids in most patients. In the study by Kamisawa et al,5 maintenance treatment with low-dose corticosteroids, usually 2.5–10 mg, was given to 82% of patients initially treated with steroids. Despite this, nearly a quarter of patients relapsed. At our centre, nearly a third of patients with AIP have required maintenance immunosuppressive drugs for relapse after withdrawal of steroids.8 Drugs used to maintain disease remission include azathioprine, 6-mercaptopurine, mycophenolate mofetil and rituximab. However, published data on the use of these medications are limited. Church et al9 have reported on the efficacy of azathioprine in a limited number of patients.

Complications of maintenance treatment of AIP

The use of maintenance corticosteroids and immunomodulatory medications poses multiple challenges, especially as AIP mostly affects elderly patients. Kamisawa et al5 report that 13 patients had complications related to steroid treatment, but the true incidence of steroid-related complications during the study period is difficult to assess from review of the reported data. In a previous study, Hirano et al reported that among 19 patients treated with corticosteroids, 4 (21%) developed serious complications including avascular necrosis of the femoral head (n = 1), lumbar vertebral fracture (n = 1) and diabetes requiring discontinuation of steroids (n = 2).10 A particular problem with corticosteroid treatment in AIP is worsening glycaemic control among patients with diabetes. While diabetes in the initial acute phase may improve with steroid treatment or even spontaneously, in the late phase glycaemic control often worsens with steroid treatment. The authors state that glycaemic control was achieved in patients with diabetes using insulin and oral hypoglycaemic agents.5 However, it has been our experience that many patients have difficulty maintaining euglycaemia while on steroids.

As with corticosteroids, immunomodulatory medications also have various adverse effects, which have not been evaluated in AIP. In our series of 58 patients, azathioprine had to be discontinued in one patient due to septicaemia (unpublished data). Therefore, the ideal maintenance strategy for AIP is unclear, emphasising the need for prospective, controlled trials of these treatments.

In summary, corticosteroids, usually in the dose range of 30–40 mg/day, are effective in the acute inflammatory phase of AIP and should be considered in all patients. The optimal treatment duration of initial treatment remains to be determined. Relapses occur in 30–40% of patients on taper or withdrawal of steroids and are corticosteroid responsive. Data on maintenance treatment, either with low-dose corticosteroids or with immunomodulatory drugs such as azathioprine or mycophenolate mofetil, are limited but nearly a third of the patients with AIP require maintenance treatment to prevent frequent relapses.


View Abstract


  • Competing interests None.

  • Provenance and Peer review Commissioned; not externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Pancreas
    T Kamisawa T Shimosegawa K Okazaki T Nishino H Watanabe A Kanno F Okumura T Nishikawa K Kobayashi T Ichiya H Takatori K Yamakita K Kubota H Hamano K Okamura K Hirano T Ito S B H Ko M Omata