patients, mean age 81 years, with AF, paroxysmal SVT, or sinus rhythm detected by 24-hour ambulatory electrocardiograms. New thromboembolic stroke developed at 31-month follow-up in 87 of 201 patients (43%) with AF in 84 of 493 patients (17%) with paroxysmal SVT, and in 143 of 782 patients (18%) with sinus rhythm. These data confirmed that AF was an independent predictor of thromboembolic stroke in elderly patients, and that paroxysmal SVT was not associated with thromboembolic stroke. Compared to our study or other studies on stroke, it appeared that the incidence of stroke was very high in this aged population. In our study mean age was 49 ± 19 years and studies were not comparable. Limitations Fundamental limitations must be underlined. The study was a retrospective analysis. There was no comparison group (case-controlled patients without SVT). History of stroke/TIA at the time of EPSwasnotedbuttherewasnofurtherevaluation for stroke/TIA. History of AF at the time of EPS was used without any screening for asymptomatic AF in follow-up monitoring done only for symptoms of palpitations and in patients treated with ablation. Onthispoint,letusrecallthenotionof“silentAF” that is likely in patients with SVT. In the group of patients with paroxysmal SVT, asymptomatic arrhythmia events were significantly less frequent than symptomatic arrhythmia events.25 In a group of patients with paroxysmal AF, sustained asymptomatic AF occured far more frequently than symptomatic AF.25
Age and prior AF were predictors of AF in follow-up but difficult to interpret in context without a comparison group in whom it is likely that these same factors would predict AF. Another limitation is the inequality in the period of follow-up of patients. The patients included in 1980 will have a larger time of followup than those included in 2010.