Objective: To establish if DKA at onset or HbA1c at onset is closer associated with metabolic control three years after diagnosis in a large, heterogeneous, international T1D cohort.

Methods: Data were collected from the international diabetes registry SWEET. 1420 T1D children (54% boys; 994 from Northern Europe, 341 from Southern Europe, 85 from America/Canada), diagnosed before 18 years, with documented clinical presentation and HbA1c at onset and follow-up data available were included. Patients were grouped according to presentation at onset: DKA with coma, DKA without coma and no DKA; and HbA1c at onset: low (HbA1c <10%), medium (10-<12%) and high (≥12%). To adjust for age at onset, gender and region, linear regression models were applied (adjusted means with 95% CI). Vuong test was employed to compare the association between third year HbA1c and both HbA1c at onset and presentation at onset.

Results: Median age at T1D onset was 9.1 [Q1; Q3: 5.8; 12.2] years. DKA with coma was present in 6% of children, DKA without coma in 37% and no DKA in 57%. No significant differences were found in third year HbA1c between DKA with coma 7.7 (7.4;8.0)%, DKA without coma 7.5 (7.4;7.6)% and no DKA group 7.6 (7.5;7.7)% [adjusted means (95% CI)]. Third year HbA1c was lower (7.4 (7.3;7.5)%) in low HbA1c compared to high HbA1c at onset group (7.7 (7.6;7.8)%, p<0.01). In the medium HbA1c group, third year HbA1c was 7.5 (7.4;7.6)% (no differences). Comparing the model with HbA1c at onset groups vs. the model with presentation at onset groups, the model with HbA1c at onset was closer associated with the third year HbA1c (p=0.02).

Conclusions The high percentage of DKA at T1D onset is still a concern worldwide. The level of HbA1c at onset was more closely associated with long term T1D control than the DKA at onset. Early T1D diagnosis, preventing high HbA1c at onset as well as DKA at onset, is important for long term metabolic control.


B. Piccini: None. A. Schwandt: None. I. Arslanoglu: None. R. Cardona-Hernandez: None. R. Coutant: None. C.A. Jefferies: None. J. Kim: None. O. Kordonouri: None. C. Limbert: None. R. Preiksa: None. A. Pundziute Lyckå: None. B. Rami-Merhar: Speaker's Bureau; Self; Eli Lilly and Company, Medtronic, Menarini Group, Roche Diabetes Care. E. Richmond: None. R. Savova: None. S. Todorovic: None. H.J. Veeze: Consultant; Self; Medtronic MiniMed, Inc. S. Toni: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.