O-16: Screening for insulin resistance in paediatric type 1 diabetes: a pilot insulin-anthropometric model
J Vinhais, I Fontes, C Gonçalves, M Oliveira, A Simões, R Pereira, L Lopes, C Limbert, J GalhardoAbstract
Introduction
Overweight and obesity are increasingly common in children and adolescents with type 1 diabetes (T1D), contributing to insulin resistance and rising insulin requirements. Escalating insulin doses alone may further promote weight gain, underlining the need for practical markers of reduced insulin sensitivity to identify patients who may benefit from review for adjunctive therapy.
Objectives
To develop a simple screening approach for insulin resistance in paediatric T1D using routine clinical data.
Methods
We conducted a single-center cross-sectional pilot study in children and adolescents with T1D. Insulin resistance was defined as eGDR-BMI < 8.0 mg/kg/min. Candidate markers included overnight insulin requirements from 03:00 to 06:00 during sleep (no bolus or carbohydrate intake; sensor glucose 80-120 mg/dL), daily basal insulin, and BMI z-score. ROC analysis and internal validation were performed.
Results
A total of 127 participants had complete data; 17 (13.4%) had eGDR-BMI < 8.0 mg/kg/min. Overnight insulin alone showed modest discrimination (AUC 0.760, 95% CI 0.670-0.830). Total daily insulin dose/kg also showed limited contrast (AUC 0.635, 95% CI 0.450-0.807). The best pragmatic model combined BMI z-score and daily basal insulin, with a leave-one-out cross-validated AUC of 0.830 (95% CI 0.720-0.850). A simple high-priority review rule (BMI percentile ≥85 with basal insulin ≥23.7 U/day or overnight insulin 03:00-06:00 ≥4.8 U) yielded sensitivity 74%, specificity 90.5%, PPV 54.0%, and NPV 95.2%.
Conclusions
In this pilot cohort, a simple insulin-anthropometric triage rule showed promising ability to identify children and adolescents with T1D at higher risk of insulin resistance. This pragmatic approach may help prioritise patients for structured assessment and consideration of adjunctive therapy.