Biological aging and dietary fatty acid balance may influence the bidirectional progression between atrial fibrillation (AF) and heart failure (HF); however, most studies focus on single endpoints, overlooking intermediate states.

To evaluate the independent and joint associations of phenotypic age acceleration (PhenoAgeAccel) and the plasma omega-6/omega-3 (ω-6/ω-3) polyunsaturated fatty acid (PFUA) ratio with AF-HF transitions, and to examine mediation by lipids and C-reactive protein.

In a retrospective cohort of 191,091 UK Biobank participants free of baseline cardiovascular disease, PhenoAgeAccel was calculated as the residual from regressing phenotypic age on chronological age. The ω-6/ω-3 ratio was quantified by nuclear magnetic resonance. Incident AF and HF were modeled using clock-forward multistate Markov models for four transitions: baseline to AF, baseline to HF, AF to HF, and HF to AF. These transitions represent sequential disease progression, where either AF or HF may occur first and later progress to AF–HF comorbidity. Hazard ratios (HRs) were estimated per 1-SD increment. Joint exposure and mediation analyses were performed.

Over a median 15.4 years, 10,084 developed AF and 3,117 H F; 1,335 transitioned from AF to HF and 426 from HF to AF. Per 1-SD higher PhenoAgeAccel, risks increased for baseline-to-AF (HR 1.12 [95% CI 1.10–1.15]), baseline-to-HF (1.24 [1.21–1.26]), AF-to-HF (1.12 [1.09–1.15]), and HF-to-AF (1.06 [1.01–1.12]). Per 1-SD higher ω-6/ω-3 ratio, risks rose for baseline-to-AF (1.04 [1.02–1.06]), baseline-to-HF (1.07 [1.05–1.10]), AF-to-HF (1.12 [1.07–1.18]), and HF-to-AF (1.10 [1.01–1.20]). Mediation occurred via triglycerides (up to 38.5% of ω-6/ω-3–AF association) and CRP (up to 10.7% of PhenoAgeAccel–HF association).

Higher PhenoAgeAccel and ω-6/ω-3 PFUA ratios were independently associated with higher risks of AF–HF transitions, with these associations partly explained by lipid and inflammatory pathways.