Clinical Characteristics of Patients with Autoimmune Pancreatitis with or without Mikulicz's Disease and Mikulicz's Disease Alone.

BACKGROUND/AIMS: The objective of this study was to compare the clinical characteristics of patients with autoimmune pancreatitis (AIP) with or without Mikulicz's disease (MD) and with MD alone.
METHODS: We investigated the clinical findings in 15 AIP patients with MD (group A+M), 49 AIP only patients (group A), and 14 MD only patients (group M).
RESULTS: The male-female ratio was significantly higher in group A+M (73%, p<0.05) and group A (78%, p<0.01) than group M (21%). Serum immunoglobulin G (IgG) levels were significantly higher in group A+M than in group A (p<0.01) and group M (p<0.05). Serum IgG4 levels were significantly higher in group A+M than in group A (p<0.01). Other organ involvement was observed in 73% (11/15) of patients in group A+M. The number of patients with diabetes mellitus was significantly higher in group A+M (66%, p<0.01) and group A (51%, p<0.05) than in group M (7%). All of the patients responded well to steroid therapy, but the relapse rate in group A+M (33%) was significantly higher than that in group A (3%, p<0.01). Salivary gland function was impaired in all groups compared with the control group, but the degree of dysfunction was less in group A compared with group A+M and group M.
CONCLUSIONS: The relapse rate of AIP in MD patients was significantly higher than that of AIP in patients without MD.

INTRODUCTION

Mikulicz's disease (MD) refers to idiopathic, bilateral, painless, and symmetrical swelling of the lacrimal, parotid, and submandibular glands.1 Due to histological similarity, MD has been considered a subtype of Sjögren's syndrome.2 However, in contrast to Sjögren's syndrome, MD responds well to steroid therapy and lacks anti-SS-A and anti-SS-B antibodies.3,4 Recently it was revealed that MD patients showed high immunoglobulin G4 (lgG4) concentrations and abundant infiltration of IgG4-positive cells as well as lymphocytes and fibrosis was detected in the lacrimal and salivary glands of MD patients,3,4 leading to the suggestion that MD is an IgG4-related disease.5

Autoimmune pancreatitis (AIP) is frequently associated with sclerosing extrapancreatic lesions such as stenosis of the bile ducts (sclerosing cholangitis), gallbladder wall thickening (sclerosing cholecystitis), swelling of the salivary and lacrimal glands (sclerosing sialadenitis and dacryoadenitis), retroperitoneal fibrosis, and pseudotumors. Histological findings for these extrapancreatic lesions are similar to those of the pancreas in AIP patients, and included dense fibrosis with abundant infiltration of IgG4-positive plasma cells and lymphocytes.6 We proposed the existence of a novel clinicopathological entity, an "IgG4-related sclerosing disease." IgG4-related sclerosing disease represents a systemic disease characterized by extensive IgG4-positive plasma cells and T lymphocyte infiltration of various organs. Clinical manifestations are apparent in organs such as the pancreas, bile duct, gallbladder, salivary and lacrimal glands, retroperitoneum, lung, and prostate gland, where tissue fibrosis is pathologically induced. AIP is not simply a pancreatitis, but rather a pancreatic lesion reflecting an IgG4-related sclerosing disease. In some cases, only one or two organs are clinically involved, while in other cases three or four organs are affected.7,8

Some patients have AIP complicated by MD, but the difference between AIP cases complicated MD, AIP alone patients, and MD alone patients is not clear. We compared clinical characteristics of AIP patients with and without MD and MD patients alone.

MATERIALS AND METHODS

1. Patients

The subjects of this study were 64 patients with AIP and 14 patients with MD (group M) treated in Tokyo Metropolitan Komagome Hospital between 1990 and 2010. AIP was diagnosed according to the Asian Diagnostic Criteria for AIP:9 enlargement of the pancreas on computed tomography and/or ultrasonography (n=64); irregular narrowing of the main pancreatic duct on endoscopic retrograde pancreatography (n=64); elevation of serum IgG4 levels (n=48); elevation of serum IgG levels (n=22); presence of autoantibodies (n=30); histology of lymphoplasmacytic sclerosing pancreatitis (n=18); and steroid responsiveness (n=49). MD was diagnosed according to diagnostic criteria of MD proposed by Japanese Sjögren's Syndrome Society:10 1) symmetrical swelling of at least two pairs of the lacrimal, parotid, or submandibular glands continuing for more than 3 months (n=14), and 2) elevated serum IgG4 levels (>135 mg/dL) (n=12), or 3) histological features including lymphocyte and IgG4-positive plasma cell infiltration with typical tissue fibrosis or sclerosis (n=11). Patients in group M showed no abnormality in pancreatic imaging. The 64 AIP patients were divided into 15 AIP patients complicated by MD (group A+M) and 49 AIP patients without MD (group A). The 15 patients of group A+M had bilateral salivary gland swelling (n=15) and lacrimal gland swelling (n=2). Histological confirmation of MD was performed in 11 of the 15 AIP patients complicated by MD. AIP occurred in eight conservatively followed-up MD patients during 3 months to 4 years. MD and AIP were diagnosed simultaneously in the other seven patients. Steroid therapy was used in 12 patients in group A+M, 36 patients in group A, and 11 patients in group M according to the standard regimen.11

2. Clinical analysis

We investigated age, sex, serum IgG levels (11 patients in group A+M, 35 patients in group A, and 13 patients in group M), serum IgG4 levels (15 patients in group A+M, 44 patients in group A, and 14 patients in group M), anti-nuclear antibody (ANA) (10 patients in group A+M, 35 patients in group A, and 14 patients in group M), rheumatoid factor (RF) (11 patients in group A+M, 32 patients in group A, and 13 patients in group M), and association with diabetes mellitus (DM). Other organ involvement including intrahepatic or hilar sclerosing cholangitis, sclerosing cholecystitis, retroperitoneal fibrosis, and distant lymphadenopathy were examined.

Remission after steroid therapy was defined as disappearance of clinical symptoms, and resolution of the imaging manifestations of the pancreas, salivary and/or lacrimal glands, and/or other organ involvements.12 Relapse after steroid therapy was defined as reappearance of symptoms with development of imaging abnormalities.12

3. Salivary gland function

To examine salivary gland function, submandibular scintigraphy was performed in six patients in group A+M, 15 patients in group A, and seven patients in group M. After intravenous injection of 180 to 200 Mbq (99mTc) pertechnetate, anterior sequential imaging was performed every minute for 30 minutes using a gamma camera. At 15 minutes after injection of radionuclide, ascorbic acid (20 mg) was administered intraorally to stimulate salivary secretion. The region of interest was set manually on the submandibular glands and time-activity curves were generated. The ratio of cumulative peak count to injected radionuclide (PCR) and washout ratio (WR, peak count before ascorbic acid administration-lowest count after administration/peak count before administration) was examined as a functional parameter. The minimum levels of PCR and WR were analyzed.13,14 Data for salivary gland function examined before in 30 normal individuals were used as controls.14

4. Statistical analysis

Data for three groups were compared to each other using Fisher's exact probability test and the Mann-Whitney U test. The p-value was corrected by Bonferroni's method, and p-values of less than 0.05 were considered statistically significant.

RESULTS

Although there were no differences in age in diagnosis in each group, the male-female ratio was significantly higher in group A+M (73%, p<0.05) and group A (78%, p<0.01) than group M (21%). Serum IgG levels were significantly higher in group A+M (median, 2,389 mg/dL) than group A (1,640 mg/dL, p<0.01) and group M (1,708 mg/dL, p<0.05). Serum IgG4 levels were significantly higher in group A+M (median, 970 mg/dL) than group A (267 mg/dL, p<0.01). There were no significant differences in presence of ANA and RF among the three groups. In addition to salivary and/or lacrimal lesions, other organ involvement was seen in 11 (73%) of 15 patients in group A+M, but there were no significant differences among the three groups. Intrahepatic or hilar sclerosing cholangitis and sclerosing cholecystitis were detected in only group A+M and group A. The number of patients complicated by DM was significantly higher in group A+M (66%, p<0.01) and group A (51%, p<0.05) than group M (7%). All patients responded well to steroid therapy, but relapse rate in group A+M (33%) was significantly higher than in group A (3%, p<0.01) (Table 1). Periods from initial diagnosis of AIP to its relapse were 0.5, 1, 1, and 1.8 years in group A+M, and 1 year in group A.

Table 1

Clinical Differences between Group A+M, Group A, and Group M

ANA, anti-nuclear antigen; RF, rheumatoid factor; OOI, other organ involvement; DM, diabetes mellitus.

*Median (quartile range); †p<0.05 compared with Group M; ‡p<0.01 compared with Group M; §p<0.01 compared with Group A.

In terms of salivary gland function, PCR and WR by submandibular scintigraphy were significantly lower in group A+M (p<0.01), group A (p<0.05), and group M (p<0.01) than controls. PCR was significantly lower in group A+M and group M than group A (p<0.05). WR was significantly lower in group M than group A (p<0.05) (Table 2).

Table 2

Submandibular Gland Function Examined by Scintigraphy

Data are presented as median (quartile range).

PCR, ratio of the cumulative peak count to injected radionuclide; WR, washout ratio.

*p<0.01 compared with controls; †p<0.05 compared with controls; ‡p<0.05 compared with Group A.

DISCUSSION

Salivary gland lesions, usually in the bilateral submandibular glands, are one of the common locations of other organ involvement in patients with AIP.6,15 The histology of swollen salivary glands in AIP patients is fibrosis with abundant infiltration of lymphocytes and IgG4-positive plasma cells, similar to findings in the pancreas in AIP patients. Therefore, swollen salivary glands associated with AIP are considered a salivary gland lesion of IgG4-related disease.6 MD is characterized by bilateral, painless, and symmetrical swelling of the lacrimal, parotid, and submandibular glands. From the findings of high lgG4 concentrations and abundant infiltration of IgG4-positive cells in the lacrimal and salivary glands of MD patients, MD is also considered an IgG4-related disease.3-5 Therefore, salivary gland lesions are thought to be the same as MD. However, it is unknown whether there are clinical differences among patients with AIP alone (group A), AIP with MD (group A+M), and MD alone (group M).

AIP showed a male preponderance even in AIP with MD, but the sex ratio of the group M was 3:11 in favor of females. According to Yamamoto et al.,4 the sex ratio of M patients was approximately 3:1 in favor of females, which was similar to our results. The reason for the gender gap between AIP and MD is unknown.

Serum IgG and IgG4 levels were significantly higher in group A+M. In addition to salivary and/or lacrimal lesions, other organ involvement was seen in 73% of group A+M, although there were no significant differences among groups. In our previous study,16 AIP patients with serum IgG4 levels of more than 220 mg/dL frequently had other organ involvement. Ghazale et al.17 reported that AIP patients with serum IgG4 lesser than 140 mg/dL were less likely to present with other organ involvement compared with those with serum IgG4 greater than 140 mg/dL. It was also reported that AIP patients with organ involvement of three other organs had significantly higher IgG4 levels than those with no lesions.15,16 Serum IgG4 levels appear to show disease activity of an IgG4-related disease.

Salivary gland function was impaired in AIP patients without MD (group A) compared with controls, although the degree of salivary gland dysfunction was greater in group A+M and group M than group A. AIP occurred in eight conservatively followed-up MD patients during 3 months to 4 years. Compared with AIP, swelling of the salivary or lacrimal glands can be easily noticed clinically even without symptoms. Given these findings, AIP might exist subclinically when preceding salivary or lacrimal gland lesions are diagnosed. Furthermore, salivary gland involvement may occur subclinically in most AIP patients. DM occurred in about half of AIP patients and improved in half of them after steroid therapy.11,13 The mechanisms for this effect were suggested to be due to regeneration of damaged islets and suppression of cytokines released from infiltrated lymphocytes.18 However, glucose intolerance was rare in MD patients.

The relapse rate was significantly higher in group A+M than group A. It has been reported that predictors for relapse are the presence of proximal bile duct stenosis and elevated serum IgG4 levels.12,19,20 However, Kubota et al.21 reported that three of five AIP patients with MD relapsed. AIP patients with MD should be treated more carefully with maintenance steroid therapy. Although there are several potential methods such as higher dose of initial steroids, longer duration of initial steroids, longer duration of maintenance therapy, and other immunosuppressive agents to reduce the relapse rate of highly active disease, further studies may be warranted.

In conclusion, the relapse rate of AIP with MD patients was significantly higher than AIP alone patients. AIP with MD may be associated with high disease activity of an IgG4-related disease and should be treated carefully with maintenance steroid therapy.