Development and internal evaluation of a composite electrophysiological criterion for the diagnosis of paroxysmal atrial fibrillation

Paroxysmal atrial fibrillation (PAF) is frequently missed by electrocardiography due to its unpredictable episodes. Electrophysiological study (EPS) can induce atrial fibrillation (AF), but standardized diagnostic criteria are lacking. We aimed to develop and internally evaluate a composite EPS-based criterion for PAF in a derivation cohort. This prospective study included 226 patients (117 PAF, 109 supraventricular tachycardia (SVT) controls) undergoing a standardized EPS protocol with programmed stimulation, decremental burst pacing, and low-dose isoproterenol. Spontaneous AF occurred in 32 PAF patients and no controls. In the remaining 194 patients, the area under the receiver operating characteristic (ROC) curve (AUC) for induced AF was 0.88 (95% CI 0.83–0.92), with an optimal cutoff value of ≥ 162 s. Thus, the composite criterion was spontaneous AF or, in its absence, an induced AF duration of ≥ 162 s. This composite criterion yielded a sensitivity of 88.9% (95% CI 83.2–93.9%), a specificity of 79.8% (95% CI 72.0–87.2%), a positive predictive value of 82.5% (95% CI 75.56–89.21%), and a negative predictive value of 87.0% (95% CI 80.58–93.20%). Decision curve analysis showed net benefit across threshold probabilities of 13%-82.5%, and performance remained consistent across age, sex, hypertension, smoking, and drinking subgroups in this derivation cohort. A composite EPS criterion-spontaneous AF or induced AF ≥ 162 s-shows high sensitivity and moderate specificity for PAF identification in this case-control setting. The standardized protocol offers promising clinical utility and safety. External validation in suspected PAF cohorts with negative noninvasive monitoring is needed before routine application.