Temporal association between elevated c-reactive protein and atrial fibrillation in ICD and CRT patients: a self-controlled case series of the HeartLogic sensor data

Background Low-grade systemic inflammation is increasingly recognized as a contributor to atrial fibrillation (AF), but whether short-term fluctuations in inflammatory activity trigger AF episodes remains uncertain. In patients with implantable cardiac devices enabling continuous rhythm monitoring through the HeartLogic algorithm, we investigated whether transient elevations in C-reactive protein (CRP) are associated with an increased risk of AF episodes using a self-controlled case series design. Methods We conducted a self-controlled case series study from 2018 to 2024 among patients with an implantable cardioverter-defibrillator or cardiac resynchronization therapy defibrillator equipped with the HeartLogic algorithm. The continuous rhythm monitoring enabled ascertainment of AF episodes. CRP measurements were obtained as part of routine clinical care, most commonly during evaluation of suspected infections. Periods of elevated inflammation were defined by CRP levels >10 mg/L. The primary outcome was device-detected AF episodes lasting >6 minutes with recurrent events considered independent if separated by ≥24 hours. Incidence rate ratios (IRRs) were estimated using conditional Poisson regression, comparing predefined post-exposure risk periods (1–14, 15–30, 31–100, and 101–180 days) following CRP >10 mg/L with baseline periods within the same individual. Baseline periods were defined as all observation time outside predefined post-CRP risk intervals. By using a within-person design, time-invariant confounders were inherently controlled. Prespecified sensitivity analyses included alternative CRP thresholds and restriction to patients with prevalent heart failure. Results A total of 216 patients with device-detected AF >6 minutes and at least one CRP measurement >10 mg/L were included. Mean age was 63.5 and 78% of the population were men. In this cohort elevated CRP levels were associated with a significantly increased incidence of AF episodes. The incidence rate ratio (IRR) for AF within 1–14 days following CRP >10 mg/L compared to the baseline period was 1.83 (95% CI 1.49–2.26; p<0.001). The association remained significant during days 15–30 (IRR 1.78, 95% CI 1.49–2.11; p<0.001), days 31–100 (IRR 1.45, 95% CI 1.26–1.66; p<0.001), and days 101–180 (IRR 1.36, 95% CI 1.19–1.54; p<0.001), with a gradual attenuation over time. Results were consistent across sensitivity analyses. Conclusion In this self-controlled case series of patients with implantable cardiac devices enabling continuous rhythm monitoring, elevated CRP was associated with a significantly increased incidence of atrial fibrillation episodes with the strongest effect observed in the weeks following CRP elevation but persisting through 180 days. These findings support a potential triggering role of systemic inflammation in atrial fibrillation and suggest that inflammatory activity may represent a modifiable risk marker in vulnerable patients.Forest plot