A Classification Algorithm for Types of Diabetes in Chronic Pancreatitis Using Epidemiological Characteristics

We developed an epidemiological algorithm to classify types of diabetes mellitus (DM) in chronic pancreatitis (CP), and applied it to a nationwide prospective longitudinal cohort of CP patients.Patients with definite CP (M-ANNHEIM criteria) were classified as having DM types 1, 2, or 3c, or no DM using an algorithm based on epidemiological characteristics: DM onset in relation to age, CP onset, exocrine insufficiency. Variables associated with development of DM were identified.Of 1130 included patients with CP between 2011 and 2018, 368 patients (33%) had DM at inclusion. Among patients with DM, 11 were classified as having type 1 (3%), 159 as type 2 (43%), and 191 as type 3c (52%). Patients with DM type 3c had longer duration of CP, more severe pain and lower physical quality of life. During longitudinal follow-up of median 47 months, 120 (20%) patients developed DM, of which 99 patients were classified as type 3c. This was independently associated with pancreatic endoscopy and surgery.The described algorithm based on epidemiological characteristics can help to classify types of DM in patients with CP. Diabetes mellitus type 3c is associated with longer duration of CP and more severe CP sequelae.