Glycaemic variation and atrial fibrillation burden and progression in individuals without diabetes undergoing continuous rhythm monitoring
A Krebs Hessellund, K J Haugan, L Y Xing, E K Kongebro, O B Vad, C Graff, A Andersen, S Haedersdal, T Truelsen, S Hoejberg, A Brandes, L Koeber, R Frikke-Schmidt, J H Svendsen, S Z DiederichsenAbstract
Background
Glycaemic dysregulation is associated with clinically diagnosed atrial fibrillation (AF), but prior studies largely relied on intermittent or short-term monitoring of heart rhythm and glucose levels to investigate their relationship. Additionally, the relationship between glycaemic variation and AF progression over time remains undetermined.
Purpose
We investigated whether higher glucose exposure, reflected by increased haemoglobin A1c (HbA1c) levels, was associated with AF burden and progression in individuals recruited from a general population without known diabetes or AF undergoing long-term continuous monitoring with implantable loop recorder (ILR).
Methods
This is a post-hoc analysis of 884 LOOP Study participants, without diabetes at baseline, undergoing ILR screening for AF combined with serial HbA1c measurements. Individuals with diabetes at baseline were excluded to eliminate confounding by antidiabetic medication. AF burden was defined as percentage of time in AF and was calculated in a 60-day window prior to each blood sampling and analysed using a two-part mixed-effects model. Model adjustments included age, sex, blood sample count, and comorbidities (hypertension, thyrotoxicosis, valvular disease, heart failure, and previous acute myocardial infarction) and a random intercept for each participant. Additionally, AF progression was analysed in groups by highest recorded HbA1c level and assessed by cumulative AF duration and burden over time.
Results
The dataset combined >1 million days of heart rhythm monitoring (including 34,852 with AF) with 3079 HbA1c measurements. AF lasting at least 6 minutes was detected in 276 (31%) of the 884 participants. The AF burden was 0.52% [0.09–3.45%], 0.78% [0.25–4.07%], and 4.42% [0.98–13.1%] for HbA1c in the normal (≤38 mmol/mol), prediabetes (39–47 mmol/mol), and diabetes (≥48 mmol/mol) range, respectively (Figure 1). Each 1 mmol/mol increase in HbA1c was associated with a 5% relative increase in AF burden (adjusted estimate 1.05 [1.00–1.09], p=0.036). Individuals with at least one measurement in the diabetes and prediabetes range accumulated more AF over time compared to participants in the normal range (Figure 2).
Conclusion
Higher HbA1c levels were associated with higher AF burden and more progression of AF over time. Individuals with consistently normal HbA1c had low AF burden and were least likely to progress. These findings are exploratory but hold promise for future analysis of HbA1c to identify high-risk individuals and glycaemic control in AF management.Figure 1AF burden by HbA1c categoryFigure 2Cumulative AF by HbA1c group