The β-cell stress hypothesis suggests that increased insulin demand contributes to the development of type 1 diabetes. In the TRIGR trial we set out to assess the profile of plasma glucose and HbA1c before the diagnosis of clinical diabetes compared to nondiabetic children. A cohort of children (N = 2159) with an affected first-degree relative and increased HLA risk were recruited 2002–2007 and followed until 2017. To study the relationship between plasma glucose/HbA1c and the development of autoantibodies or clinical disease Kaplan-Meir curves were developed. Mixed models were constructed for plasma glucose and HbA1c separately. A family history of type 2 diabetes was related to an increase in plasma glucose (P < 0.001). An increase in glucose from the previous sample predicted clinical diabetes (P < 0.001) but not autoantibodies. An increase of HbA1c of 20% or 30% from the previous sample predicted the development of any autoantibody (P < 0.003 resp <0.001) and the development of diabetes (P < 0.002 resp <0.001. Participants without autoantibodies had lower HbA1c (mean 5.18%, STD 0.24; mean 33.08 mmol/mol, STD 2.85) than those who progressed to clinical disease (5.31%, 0.42; 34.46 mmol/mol, 4.68; P < 0.001) but higher than those who developed any autoantibody (5.10%, 0.30; 32.21 mmol/mol, 3.49; P < 0.001), or multiple autoantibodies (5.11%, 0.35; 32.26 mmol/mol, 3.92; P < 0.003). A pronounced increase in plasma glucose and HbA1c precedes development of clinical diabetes, while the association between plasma glucose or HbA1c and development of autoantibodies is complex. Increased insulin demand may contribute to development of type 1 diabetes.

• Objective. —To differentiate the insulin-dependent glucose intolerance associated with cystic fibrosis from type I diabetes mellitus in patients with cystic fibrosis. Design. —Patient report. Setting. —Tertiary care referral center. Participant. —An 11-year-old boy with cystic fibrosis who developed diabetic ketoacidosis. Measurement/Main Result. —Biochemical, immunologic, and molecular techniques were used to support the sporadic association of type I diabetes mellitus in a patient with cystic fibrosis. Cystic fibrosis was confirmed by sweat test and further supported by the demonstration of a heterozygous deletion of the F508 locus. Evidence for the diagnosis of type I diabetes mellitus was developed from the clinical presentation of diabetic ketoacidosis with hyperglycemia, ketonemia, and ketonuria. Immunologic evidence included the demonstration of anti-insulin antibodies. The demonstration of homozygous absence of aspartic acid at position 57 of the HLA DQ-β chain placed this child at high risk of type I diabetes mellitus. Conclusion. —The clinical presentation and the presence of immunologic and genetic markers characteristic of type I diabetes mellitus supports the concordance of cystic fibrosis and type I diabetes mellitus in this patient. (AJDC. 1992;146:1457-1458)