Abstract 145
Advanced insulin pump therapy does not trigger early worsening of diabetic retinopathy:
a 6-month longitudinal analysis in a real-world cohort a 6-month longitudinal analysis in a real-world cohort
E. FortinE. Fortin1, N. Widén2, A. Elksnis1, S. Karayiannides1, S.B. Catrina1, N. Rajamand-Ekberg1;
1Department for Molecular Medicine and Surgery, Karolinska Institute, Stockholm, Sweden, 2Centre for Diabetes, Academic Specialist
Centre, Region Stockholm, Stockholm, Sweden.
Background and aims:
The rapid correction of chronic hyperglycaemia is historically associated with the risk of the early worsening of
diabetic retinopathy.
While automated Insulin Delivery (AID) systems facilitate substantial and rapid glycaemic gains, the short-term
microvascular safety of this ”glycaemic rescue” remains under-explored in real-world settings. We evaluated whether intensive
improvements in Time in Range (TIR) influence retinopathy progression.
Materials and methods:
This longitudinal study included adults with type 1 diabetes initiating advanced pump therapy (2021-2025).
Standardized 14-day glycaemic metrics (baseline, 3 and 6 months) were retrieved via cloud-based platforms or electronic records.
Retinopathy progression was evaluated as an ordinal outcome (grades 0-4) using mixed-effects ordinal logistic regression to model the
probability of grade shifting over 6 months.
The model was adjusted for age, sex, diabetes duration, and technology type (AID vs.
Sensor-Augmented Pump [SAP]). Glycaemic trajectories (TIR, HbA1c, and Time Below Range [TBR]) were analysed using linear mixed-
effects models (REML) with time-by-technology interaction term and age, sex, diabetes duration as adjusting factors.
Results:Results: Among 92 participants with available baseline fundus photography (age 41.0 ± 13.4 years, diabetes duration 22.0 ± 13.4 years),
the AID group (n=71) presented with a significantly higher baseline glycaemic burden compared to SAP (
n=21; TIR 46.4% vs. 59.5%, p=0.01). Despite this higher-risk profile, no significant evidence of early retinopathy progression was observed at 3 or 6 months (OR
1.04 [95% CI 0.39-2.75] and 1.34 [0.55-3.25], respectively).
While longer diabetes duration (OR 1.37 [95% CI 1.16-1.61], p<0.001) and
lower baseline TIR (OR 0.87 [95% CI 0.80-0.94], p=0.001) were associated with higher retinopathy grades at study entry, the technology-
induced ”rescue” did not trigger further progression (pinteraction=0.34). Metabolic optimization was robust across the cohort: mean
HbA1c decreased by -10.2 mmol/mol (-1.0%; p<0.001), and AID users achieved a superior absolute TIR improvement compared to SAP
(+17.5% vs. -1.0%; pinteraction=0.002) while maintaining a lower risk of Level 1 hypoglycaemia (mean difference -2.15%; p=0.001).
Clinically significant (Level 2) hypoglycaemia remained exceptionally low and stable over time in the whole population (from 0.42% to
0.29% at 6 months).
Conclusion:
Initiating advanced insulin pump therapy, particularly AID, achieves rapid metabolic optimization without triggering early
worsening of diabetic retinopathy over 6 months, even in patients with high-risk baseline metabolic profiles.
These real-word data provide important clinical reassurance that modern automated technology can safely bridge the gap to glycaemic goals.
Clinical Trial Registration Number:
N/A
Disclosure:
E. Fortin: None.
Nyhetsinfo
www red DiabetologNytt