Type 1 diabetes is often associated with other autoimmune diseases (1), including primary biliary cirrhosis (2). Furthermore, type 1 diabetes and primary biliary cirrhosis may share similar pathogenetic pathways (3). In type 1 diabetic patients, the identification of markers for associated autoimmune diseases may permit earlier diagnosis and more effective treatment.

A 46-year-old man with type 1 diabetes (age of onset 26 years) was admitted into our hospital due to poor glycemic control (HbA1c 11.3%) with severe daily hypoglycemia and significant hyperglycemic spikes. At admission, routine blood tests showed mild eosinophilia (6.7%, 482.4/mmc versus normal 1–4%, 72–282/mmc) and markedly elevated values for γ-glutamyl transpeptidase (γGT) (203 units/l versus normal, 8–61) and alkaline phosphatase (571 units/l versus normal, 91–258). Aspartate, alanine aminotransferase, and bilirubin values were normal. Alkaline phosphatase gradually increased during hospitalization (from 571 to 683 units/l), whereas γGT did not change significantly. Mild eosinophilia (5.9%, 403.9/mmc) occurred ∼18 months before hospitalization, but all common causes of eosinophilia were excluded. Twelve months before hospitalization, γGT and alkaline phosphatase values were normal. The patient did not show any history of jaundice, pruritus, or dyspepsia. During hospitalization, any causes of hepatobiliary disease, including viral infections, were accurately excluded. Moreover, common causes of eosinophilia were also excluded. Screening for autoimmunity showed normal values for the common panel of autoantibodies (anti-nuclear, anti-thyroid peroxidase, anti-thyroglobulin, and anti-cardiolipin) except for anti-mitochondrial antibodies (titer 1:40).

Abdominal ultrasonography did not reveal any abnormal findings. Extrahepatic biliary tracts were not dilated. Ultrasound-guided liver biopsy was then performed. Histological findings showed flogistic infiltration of the portal tract and hepatic lobules. Moreover, there was portal tract fibrosis with focal infiltration of lobules, including a picture of intrahepatic biliary duct disease. This picture was consistent with stage 2 primary biliary cirrhosis according to Scheuer classification (4). Ursodesoxicholic acid treatment was begun, and since then cholestasis values have decreased and glycemic control has improved.

The present case shows an association between type 1 diabetes and asymptomatic primary biliary cirrhosis. One year before hospitalization, the patient did not show abnormal markers for cholestasis, but 18 months beforehand, he did show mild eosinophilia. In the last decade, evidence for an association between mild eosinophilia and primary biliary cirrhosis has constantly increased. Moreover, according to most recent studies, mild eosinophilia seems to be an indicator of early disease stages and is considered a strong predictor of good response to ursodesoxicholic acid treatment and of better prognostic outcomes (5).

To the best of our knowledge, this is the first case of mild eosinophilia associated with primary biliary cirrhosis in type 1 diabetic patients. This case suggests that in type 1 diabetic patients, isolated mild eosinophilia should be carefully regarded when common causes of eosinophilia have been excluded. Indeed, when considering the possible association between type 1 diabetes and primary biliary cirrhosis (13) in type 1 diabetic patients with unexplained eosinophilia, γGT, alkaline phosphatase, and anti-mitochondrial antibodies should be evaluated to discern which subjects are at risk for primary biliary cirrhosis. In patients with positive anti-mitochondrial antibodies but normal γGT and alkaline phosphatase values, the latter should be strictly monitored. Patients with anti-mitochondrial antibodies and elevated γGT and alkaline phosphatase values should undergo a liver biopsy. In this way, mild eosinophilia may be considered a marker of asymptomatic primary biliary cirrhosis at earlier stages, when biochemical and clinical responses to ursodesoxicholic acid treatment can lead to better results. In addition, an early and effective treatment of primary biliary cirrhosis may permit better diabetes control.

1.
Hanukoglu A, Mizrachi A, Dalal I, Admoni O, Rakover Y, Bistritzer Z, Levine A, Somech E, Lehmann D, Tuval M, Boaz M, Golander A: Extrapancreatic autoimmune manifestations in type 1 diabetes patients and their first-degree relatives: a multicenter study.
Diabetes Care
26
:
1235
–1240,
2003
2.
Prince MA, Vialettes B, Zevaco-Mattei C, Vague P: Clinical characteristics and etiological markers in insulin-dependent diabetes associated with an organ-specific autoimmune disease.
Acta Diabetol Lat
20
:
221
–229,
1983
3.
Mason AL, Xu L, Guo L, Munoz S, Jaspan JB, Bryer-Ash M, Cao Y, Sander DM, Shoenfeld Y, Ahmed A, Van de Water J, Gershwin ME, Garry RF: Detection of retroviral antibodies in primary biliary cirrhosis and other idiopathic biliary disorders.
Lancet
351
:
1620
–1624,
1998
4.
Scheuer PJ: Ludwig Symposium on biliary disorders, part II: pathologic features and evolution of primary biliary cirrhosis and primary sclerosing cholangitis.
Mayo Clin Proc
73
:
179
–183,
1998
5.
Yamazaki K, Nakadate I, Suzuki K, Sato S, Masuda T: Eosinophilia in primary biliary cirrhosis.
Am J Gastroenterol
91
:
516
–522,
1996