Objectives: Intermittent rhythm monitoring (IRM) is the most frequently used modality employed
for the detection atrial fibrillation (AF) recurrence after therapeutic interventions.
The data obtained from IRM are then used to draw inferences on the success or failure
of surgical or interventional therapies for atrial fibrillation. We sought to investigate
the sensitivity of various IRM strategies using a novel probabilistic approach on
a continuous monitored patient population with proven AF recurrence.
Methods: Rhythm histories of 473 patients (mean AF burden, 0.16 ± 0.24, 491 patient*years
follow-up, mean follow-up: 1.03 ± 0.33 years) with implantable CM devices (Reveal
XT 9529 or AT500 pacemaker; Medtronic, Inc, Minneapolis, MN) and proven AF recurrence
were reconstructed and analyzed. With the use of computationally intensive simulation,
virtual IRMs were performed and the success rate and sensitivity of IRM of various
frequencies and durations on the identification of AF recurrence was evaluated.
Results: Prolonged-duration IRM (7-, 14-, 30- days monitoring) was superior to shorter IRM
(24h monitoring) (P< 0.0001). However, even with aggressive IRM strategies, AF recurrence was not detected
in a great proportion of patients. The most often used IRM strategy (4 24h Holter
per year) failed to identify AF recurrence in almost 50% of patients with proven AF
recurrence thus offering no better insight for AF recurrence detection in these patients
than the tossing of a fair coin.
Conclusions: IRM follow-up is unreliable and significantly inferior to CM. IRM strategies will
not identify AF recurrence in a great proportion of patients at risk. When relying
on IRM for rhythm monitoring to identify AF recurrence, chance has an immerse and
immeasurable effect on the outcomes. For the scientific, evidence-based evaluation
of AF treatments, CM should be strongly recommended.
