Early-Onset Atrial Fibrillation

Definition

Early-onset atrial fibrillation (EOAF) is defined operationally as AF diagnosed before the age of 66 years. It is distinguished from AF arising in older patients because it is substantially more likely to have a monogenic (single-gene) cause. Early-onset AF can represent the first clinical manifestation of an underlying inherited cardiomyopathy or, less commonly, a channelopathy — rather than a primary electrical disease.

Epidemiology and Genetic Yield

• Prevalence of disease-associated variants: In a prospective cohort of 1293 patients with EOAF who underwent whole-genome sequencing analyzed against 145 cardiomyopathy and arrhythmia genes, 10.1% carried a pathogenic/likely pathogenic (P/LP) variant by ACMG criteria. This rate is comparable to the genetic yield of testing in dilated cardiomyopathy (~25%) and far exceeds the ~2% rate in the general population. (sources/eoaf-jama-2021)
• The yield is selection-dependent — 20% in a referred clinic against 10% unselected. In a dedicated AF precision medicine clinic where patients were referred for suspicion of genetic disease or prompted by an electronic alert, 48 of 242 (20%) tested positive — double the unselected rate, with younger age at diagnosis (median 39 years) and clinician pre-selection offered as the explanations. The same paper is explicit that this is an ascertained population, so neither figure is the "true" yield; they bracket it. (sources/genetic-eoaf-ehj-2026, rating: high)
• Age gradient: The yield of disease-associated variants is inversely related to age at AF diagnosis: 16.8% for AF diagnosed before age 30, declining to 7.1% for those diagnosed between age 60–65. A continuous cubic spline analysis confirmed a nonlinear relationship (P=.02). (sources/eoaf-jama-2021)
• Predominantly cardiomyopathy genes: The large majority of P/LP variants are in genes associated with inherited cardiomyopathy (DCM 7.2%, HCM 3.3%, AC/ARVC 2.9%) rather than channelopathies (LQTS 0.9%, Brugada 0.2%, CPVT 0.1%). This suggests EOAF often represents an atrial manifestation of a cardiomyopathy-susceptibility genotype. (sources/eoaf-jama-2021)
• VUS challenge: 62.8% of patients in the EOAF cohort carried a VUS as their highest-priority finding. Managing VUS results requires structured genetic counselling, particularly for genes such as RYR2 and FLNC that carry potential VT/SCD risk implications. (sources/eoaf-jama-2021)

Genetic Architecture

Population-Level Rare Variant Analysis (UK Biobank WES)

• WES gene-burden test in 403,990 individuals — 6 pLOF genes associated with AF: TTN, RPL3L, PKP2, CTNNA3, KDM5B, C10orf71; 5/6 externally replicated. Carrier prevalence: 1.27% of the incident-AF-free population. TTN (PSI90) had the strongest association: OR 3.85. This is the largest population-scale rare variant study in AF to date. (sources/Biobank-AF-JAMA-2024, rating: high)
• Combined PRS + pLOF multiplicative model: OR 7.08: Top-quintile AF PRS alone (OR ~3.5×) + pLOF variant → OR 7.08 (95% CI 6.03–8.28) for AF. 10-year absolute risk: 24% in males >60 years with top PRS + pLOF variant. This models the joint contribution of common and rare genetic variation to AF risk across the general population. (sources/Biobank-AF-JAMA-2024, rating: high)
• pLOF variants predict cardiomyopathy — AF is often first manifestation: In cause-specific Cox regression, pLOF carriers had CM HR 3.13 before AF and HR 2.98 after AF — both largely TTN-driven; attenuated to non-significant when TTN excluded. The AF PRS did not predict cardiomyopathy (HR 0.99). Population-level confirmation that rare pLOF variant-positive AF represents a cardiomyopathy-susceptibility phenotype requiring longitudinal cardiac surveillance. (sources/Biobank-AF-JAMA-2024, rating: high)
• Novel genes implicate epigenetic and ribosomal mechanisms: KDM5B (histone H3K4 demethylase, cardiac fibrosis role) and RPL3L (ribosomal function, splice-donor variant) suggest arrhythmia substrate mechanisms beyond ion channel dysfunction and structural protein disruption. (sources/Biobank-AF-JAMA-2024, rating: high)

Polygenic × Monogenic Risk Interaction

• PRS modifies TTNtv penetrance for AF: In ~44,000 UK Biobank participants, TTNtv carriers in the highest polygenic risk score (PRS) tertile had 21.5% AF prevalence vs. 6.7% in the lowest tertile — a >3-fold range in penetrance explained by common variant background. This is a clinically important finding for pre-symptomatic counselling of TTNtv carriers. (sources/genetic-eoaf-ehj-2024, rating: high)
• Similar effect demonstrated in HCM: P/LP HCM variant carriers in the highest PRS centile had 14-fold increased risk vs. the median PRS; penetrance in P/LP carriers is significantly higher in the highest quintile vs. the lowest (OR 9.56 vs. 3.69). (sources/genetic-eoaf-ehj-2024)
• Clinical translation pending: How PRS should be integrated with rare variant findings for EOAF counselling remains undefined — an important research gap. (sources/genetic-eoaf-ehj-2024)

Pathogenesis

Atrial Myopathy Hypothesis

• EOAF attributable to cardiomyopathy-associated variants may reflect an "atrial myopathy" — where genetic defects in sarcomeric, cytoskeletal, or nuclear envelope proteins produce early atrial structural dysfunction before overt ventricular cardiomyopathy develops. Genes expressed in the atrium (especially MYH6) may selectively manifest as AF without ventricular involvement. (sources/eoaf-jama-2021)

Temporality of AF and Ventricular Cardiomyopathy

• AF often precedes ventricular cardiomyopathy: In TTNtv carriers who eventually develop both AF and ventricular cardiomyopathy, AF preceded the cardiomyopathy diagnosis in ~50% of cases (UK Biobank TTNtv analysis). This directly challenges the assumption that AF is purely a downstream manifestation of established structural disease. (sources/genetic-eoaf-ehj-2024, rating: high)
• LMNA: AF and atrial myopathy are established early features of LMNA disease, often preceding ventricular manifestations by years to decades. Isolated AF may be the sole initial presentation before cardiomyopathy becomes structurally detectable. (sources/genetic-eoaf-ehj-2024)
• Latvian EOAF cohort: 5 of 13 EOAF patients carrying P/LP cardiomyopathy variants had normal echocardiography at diagnosis but later showed ventricular enlargement on CMR — underscoring the need for longitudinal MRI-based surveillance, not just echocardiography, in EOAF variant carriers. (sources/genetic-eoaf-ehj-2024)
• Clinical implication: Identifying a cardiomyopathy variant in an EOAF patient should trigger ongoing MRI surveillance even when initial echocardiography is normal, as structural cardiomyopathy may develop years later. (sources/genetic-eoaf-ehj-2024)

Clinical Phenotype: AF Before Age 40 Without SHD

• Lifestyle risk factor burden is high: In the largest prospective cohort of AF before age 40 without SHD (n=122; mean age 31.2 years; 83% male; median follow-up 52 months), 89% had ≥1 lifestyle risk factor and 74% had at least one modifiable risk factor. The most common were overweight/obesity (30%), intensive sport activity >6 h/week (27%), and excessive alcohol (22%). Recreational drug use was present in 11%. (sources/eoaf-riskfactor-ehj-2026, rating: medium)
• Hereditary predisposition in 1 in 4: 25% had hereditary predisposition — familial AF (16%) or family history of cardiovascular disease (10%). Self-reported AF triggers were present in 48%: psychological stress (20%), exercise (11%), alcohol (11%). (sources/eoaf-riskfactor-ehj-2026)
• Genetic yield 11.6% P/LP in this phenotype: Among 69/122 patients (57%) accepting genetic panel testing, 8 carried P/LP variants (11.6%) and 27 carried VUS (39.1%). All 8 familial AF patients who were tested harbored at least one variant (3 P/LP, 5 VUS). P/LP variants were predominantly cardiomyopathy genes (74%): TTN ×4, LMNA ×2, PKP2 ×2, TNNI3 ×1. (sources/eoaf-riskfactor-ehj-2026)
• Structural heart disease absent at presentation does not rule out underlying cardiomyopathy genotype: CMR performed in 49% showed normal ventricular volumes without scar or fibrosis; LA enlargement was present in only 15%. Yet cardiomyopathy gene P/LP variants (LMNA, TTN, PKP2) were found — consistent with AF as the initial manifestation before structural cardiomyopathy emerges. (sources/eoaf-riskfactor-ehj-2026)
• Stroke in 4.1% despite CHA₂DS₂-VA = 0: Five patients (4.1%) had a stroke — 4 at first AF presentation and 1 during follow-up. All had CHA₂DS₂-VA = 0 and were not anticoagulated. This raises questions about whether conventional stroke risk models underestimate thromboembolic risk in very young AF without comorbidities. (sources/eoaf-riskfactor-ehj-2026)
• Excellent catheter ablation outcomes: Among the 63% who underwent catheter ablation, 1-year success rate was 86%, rising to 93% after two procedures. Progression to persistent AF occurred in only 6% and none developed permanent AF during follow-up. (sources/eoaf-riskfactor-ehj-2026)

Clinical Genetic Evaluation — What a Dedicated Clinic Finds

The first prospective account of an AF precision medicine clinic (Vanderbilt; 264 referred, 242 completing a cardiomyopathy/arrhythmia panel; October 2020 – September 2024) reports a 20% P/LP yield — twice the ~10% from unselected retrospective cohorts, and comparable to yields at which genetic testing is already routine in DCM and Brugada syndrome. (sources/genetic-eoaf-ehj-2026, rating: high)

• The clinic made the diagnosis rather than confirming it. A cardiomyopathy or ventricular arrhythmia syndrome had been diagnosed before referral in only 6 of the 48 genotype-positive patients.
• Median age at AF diagnosis was 39 years; median time from diagnosis to evaluation was 3.8 years, but ranged from one week to 43 years — a reminder that a negative evaluation at one time point is not a lifetime result.
• TTN accounted for 23 of the 48 positives, LMNA and SCN5A three each, and 71% of positives carried a cardiomyopathy variant. Ten fell outside the four AF genetic subgroups, including three HFE and three CPT2 — treatable metabolic diagnoses returned incidentally.
• A gene-elusive AF overlap syndrome — diagnostic criteria met with negative genetic testing — was found in 22 patients (9%). Half of them had neither a suspicious VUS nor a suggestive family history, which is the argument for phenotyping every patient rather than only the genotype-positive.

Pre-Test Probability, Measured

• T1 mapping outperformed LGE, consistent with T1 being superior for diffuse fibrosis — and it is a more actionable finding than scar, because it suggests earlier adverse remodelling.
• Family history of AF did not predict a positive result, nor did AF type or cardioversion count.
• The aggregate risk score did not discriminate: HARMS2-AF showed no difference in clinical risk burden between genotype-positive and genotype-negative patients (P=.76).
• No single factor is strong enough to gate referral, and the authors say so — but the combination of cardiomyopathy history, conduction disease and abnormal CMR is what identifies the population that should be tested.

Penetrance of the Overlap Syndrome

Of 52 genotype-positive participants (48 probands plus 4 relatives found by cascade testing), 56% met criteria for the corresponding ventricular syndrome: ACM genes 89% (8/9), TTN-DCM 54% (13/24), channelopathy 50% (2/4), HCM 40% (2/5). The ACM subgroup was both the most penetrant and the most consequential.

Management Impact

Genetic and phenotypic evaluation changed management in 52% of genotype-positive patients — the figure that turns genetic testing in AF from a prognostic test into a therapeutic one. (sources/genetic-eoaf-ehj-2026, rating: high)

• Seven new ICDs and two pacemaker-to-ICD upgrades, across four distinct indications: genetic cardiomyopathy with LVEF <45% plus risk factors; ACM with syncope and ventricular arrhythmia; new HCM diagnoses with individualised primary prevention; and a pacing indication in a high-arrhythmia-risk genotype.
• Seven patients with a CHA₂DS₂-VA of ≤1 were started on anticoagulation after being newly diagnosed with HCM or ACM — the clearest example of a genetic result overriding a conventional thrombotic risk score.
• Guideline-directed medical therapy started or optimised in genotype-positive patients with any history of LV systolic dysfunction, including TTN carriers whose dysfunction had recovered with rhythm control.
• Drug avoidance in channelopathies: three SCN5A carriers told to avoid Class I antiarrhythmics and take fever precautions; one KCNQ1 carrier told to avoid QT-prolonging and Class III agents.
• Exercise restriction to ≤6 METs for four ACM participants with desmosomal variants.
• Management changed in 33% of DCM-phenotype patients against 92% of ACM-phenotype and 100% of HCM-phenotype patients — the yield of a management change tracks the penetrance of the syndrome, not the fact of a variant.

Genetic Testing Recommendations

• 2023 ACC/AHA AF guidelines (Class IIb): Genetic testing is reasonable in patients with AF diagnosed at ≤45 years with no obvious risk factors — the first formal guideline recommendation for genetic testing specifically in AF. For those <30 years with unexplained AF, an EP study to assess/treat SVT should also be considered (Class IIb). (sources/genetic-eoaf-ehj-2024, rating: high)
• ESC AF guidelines (current): No recommendation for genetic testing in young individuals without comorbidities — a notable gap vs. ACC/AHA. (sources/genetic-eoaf-ehj-2024)
• Multi-society consensus (European, US, Latin American, Asian-Pacific): Initial consideration for genetic testing in familial AF <60 years. (sources/genetic-eoaf-ehj-2024)
• Comparator — DCM: Genetic testing is a Class IB recommendation in ESC guidelines for DCM; diagnostic yield in non-familial DCM (~10%) is comparable to EOAF — making the lack of a formal EOAF recommendation a notable inconsistency. (sources/genetic-eoaf-ehj-2024)
• Clinical implication: Patients with EOAF may benefit from evaluation in an inherited heart disease clinic with genetic counselling and commercial cardiomyopathy/arrhythmia panel testing. If a disease-associated variant is identified, additional diagnostic evaluation for the corresponding inherited syndrome (e.g., echocardiogram, cardiac MRI, family screening) is indicated. (sources/eoaf-jama-2021)

Contradictions / Open Questions

• Causal vs. coincidental association: The 10.1% P/LP variant yield in EOAF exceeds the general population (~2%) but falls substantially below the yield in DCM (~25–30%) or HCM (~40–60%). It is unclear in every individual case whether the cardiomyopathy gene variant is causative of AF or is a coincidental finding in a population with common underlying risk factors. Without mechanistic evidence linking atrial structural dysfunction specifically to the variant, classification remains probabilistic. (sources/eoaf-jama-2021)
• VUS management — no clinical pathway for 62.8% of patients: The majority of patients in the EOAF cohort received a VUS as their highest-priority result, not a P/LP variant. No standardized clinical pathway exists for managing VUS findings in EOAF — current guidance (don't act on VUS alone) leaves the majority of patients with genetic uncertainty and potential anxiety without actionable guidance. (sources/eoaf-jama-2021)
• Age cutoff for testing remains unvalidated: The 2021 study suggested testing up to age 60 may be reasonable (yield 7.1% at 60–65), but no prospective study has validated whether genetic testing at this threshold improves clinical outcomes compared to standard AF management without genetic evaluation. (sources/eoaf-jama-2021)
• Stroke in CHA₂DS₂-VA = 0 young AF — anticoagulation guidance absent: In the prospective <40 AF cohort, 4.1% had a stroke despite CHA₂DS₂-VA = 0 and none being anticoagulated. Current guidelines do not provide anticoagulation guidance for very young AF patients with score 0. Whether these events reflect occult atrial substrate, paradoxical embolism, or a true limitation of the score in this age group is unresolved. (sources/eoaf-riskfactor-ehj-2026)
• Voluntary genetic testing creates selection bias: The 57% acceptance rate in the prospective cohort limits the ability to define true population-level genetic yield for AF <40 without SHD. No significant differences between tested and non-tested patients were detected, but with n=122 the study is underpowered to exclude meaningful selection bias. (sources/eoaf-riskfactor-ehj-2026)
• UK Biobank pLOF associations: European ancestry only: All UK Biobank WES findings are from European ancestry individuals; generalisability to other populations is uncertain. The combined OR 7.08 (PRS + pLOF) cannot be assumed to apply outside this ancestral group. (sources/Biobank-AF-JAMA-2024, rating: high)
• Non-TTN pLOF genes may not require the same CM surveillance as TTNtv: When TTN variants are excluded from the UK Biobank analysis, the cardiomyopathy and HF associations become non-significant. Gene-specific surveillance (CM monitoring for TTNtv; ARVC surveillance for PKP2; epigenetics research for KDM5B) may be more appropriate than a uniform policy for all pLOF AF carriers. (sources/Biobank-AF-JAMA-2024, rating: high)
• No prospective outcome data for genetic testing in EOAF: No prospective RCT has demonstrated that genetic testing in EOAF improves clinical outcomes (mortality, HF prevention, SCD prevention) vs. standard AF management. The clinical benefit of identifying a cardiomyopathy variant in an EOAF patient — and the optimal surveillance protocol — remains to be validated prospectively. (sources/genetic-eoaf-ehj-2024, rating: high)
• No standardised definition of EOAF across studies: Age cut-offs of <30, <40, <45, <60, and <66 years are all used across studies — making yield estimates incomparable. An age ≤45 years is the ACC/AHA-endorsed threshold; no international consensus definition exists. (sources/genetic-eoaf-ehj-2024)
• Ablation outcomes for variant carriers unknown: No data exist on whether EOAF patients with cardiomyopathy P/LP variants have different ablation success rates, arrhythmia recurrence, or progression to persistent AF compared to genotype-negative EOAF patients. (sources/genetic-eoaf-ehj-2024)
• Selection criteria and real-world findings point in opposite directions. The AHA/ACC 2023 Class IIb recommendation restricts testing to AF ≤45 years without obvious risk factors, and the EHJ 2024 pre-test framework lists classical modifiable risk factors as reasons to lower suspicion. Yet in the prospective clinic cohort the aggregate AF risk score (HARMS2-AF) did not separate genotype-positive from genotype-negative patients at all (P=.76), and 14% of patients with obstructive sleep apnoea — the strongest single negative predictor — still carried a pathogenic variant. Risk factors may be second hits that precipitate AF in a genetically susceptible atrium rather than evidence against a genetic cause. (sources/genetic-eoaf-ehj-2026, rating: high; sources/AF-AHA-2023, rating: very high)
• Family history of AF is a poor gate for genetic testing. It is the intuitive criterion, it appears in most selection frameworks, and in the prospective cohort it did not predict a positive result — plausibly because familial clustering of AF is dominated by polygenic risk and shared environment rather than by rare large-effect variants. Family history of cardiomyopathy, sudden death or ICD is a different question and still matters. (sources/genetic-eoaf-ehj-2026)
• Penetrance estimates from AF cohorts are snapshots of a moving target. The 54% ventricular penetrance in TTN carriers was measured at a median of ~4 years after AF diagnosis but with a range of one week to 43 years, in a syndrome that commonly declares itself after age 60. The 89% penetrance in ACM-gene carriers may be correspondingly inflated by earlier onset and by the small denominator (nine patients). Neither number should be used to counsel an individual at diagnosis. (sources/genetic-eoaf-ehj-2026)
• Management changed for half of genotype-positive patients, but no outcome was measured. Seven ICDs, seven anticoagulation starts and a set of drug-avoidance recommendations are real clinical actions based on guideline-concordant reasoning — and none of them has been shown in this cohort to prevent a stroke, an arrhythmic death or a heart failure admission. The case for testing currently rests on the quality of the decisions it enables, not on demonstrated benefit. (sources/genetic-eoaf-ehj-2026)

Connections

• Related to entities/Atrial-Fibrillation
• Related to concepts/Genetic-Testing-in-AF
• Related to entities/TTN
• Related to entities/MYH7
• Related to entities/MYH6
• Related to entities/LMNA
• Related to entities/PKP2
• Related to entities/KCNQ1
• Related to concepts/Arrhythmogenic-Cardiomyopathy
• Related to concepts/Cascade-Family-Screening
• Related to concepts/CHA2DS2-VA
• Related to concepts/Catheter-Ablation-AF
• Related to entities/SCN5A · entities/FLNC · entities/DSP · entities/EMD — the other genes a prospective AF cohort returns
• Related to entities/ATTR-Amyloidosis — p.Val142Ile found twice in an AF cohort
• Related to entities/HCM · entities/DCM — the overlap phenotypes a positive result predicts

Sources

• sources/genetic-eoaf-ehj-2026
• sources/Biobank-AF-JAMA-2024
• sources/eoaf-jama-2021
• sources/eoaf-riskfactor-ehj-2026
• sources/genetic-eoaf-ehj-2024