Association of Diabetic Ketoacidosis at Onset, Diabetes Technology Uptake, and Clinical Outcomes After 1 and 2 Years of Follow-up: A Collaborative Analysis of Pediatric Registries Involving 9,269 Children With Type 1 Diabetes From Nine Countries

Abstract

Objective: This study examined the association between diabetic ketoacidosis (DKA) at type 1 diabetes diagnosis and long-term glycemic outcomes, insulin requirements, BMI SD score (SDS), and diabetes technology uptake in youth.

Research design and methods: Data were from nine countries (Austria, Czechia, Germany, Italy, Luxembourg, New Zealand, Slovenia, Switzerland, and U.S. [Colorado]), including youth (0.5-15.9 years) diagnosed with type 1 diabetes in 2019-2020 and followed for 2 years thereafter. Participants were divided into three groups: no DKA, nonsevere, and severe DKA at diagnosis. HbA1c, insulin requirements, BMI SDS, and use of technology, including automated insulin delivery (AID), were assessed.

Results: The analysis included 9,269 individuals (54.8% males, mean age 9.0 years). DKA at diagnosis was observed in 34.2% of participants and severe DKA in 12.8%. After 1 year, adjusted mean HbA1c was higher in the severe DKA group (7.41%) compared with nonsevere DKA (7.23%, P = 0.001) and no DKA groups (7.14, P < 0.001), and this difference persisted after 2 years (7.58% vs. 7.38% [P < 0.001] and vs. 7.32% [P < 0.001]). Higher BMI SDS was observed in both DKA groups compared with no DKA. The use of AID was associated with lower HbA1c levels compared with other treatment modalities and moderated differences between DKA groups after 2 years of follow-up (P = 0.072).

Conclusions: Severe and nonsevere DKA at type 1 diabetes diagnosis were both associated with persistently higher HbA1c and higher BMI SDS. AID use diminishes the association of DKA at diagnosis and higher HbA1c over time.

Graphical abstract

Figure 1

Differences in HbA_1c (A), insulin requirement (B), and BMI SDS (C) after 1 and 2 years since onset between the severe DKA, nonsevere DKA, and no DKA groups. Data are shown as adjusted means (SEM). The P values are adjusted for age at type 1 diabetes diagnosis, sex, treatment modality, and included country as random intercept.

Figure 2

Differences in HbA_1c for DKA status and treatment modality. Data are shown as adjusted means (SEM). The P values are adjusted for age at type 1 diabetes diagnosis sex, stratified by treatment modality, and included country as random intercept.