Cardiac telemetry identifies AF in cryptogenic stroke

Caroline Helwick

February 02, 2012

New Orleans, LA - Outpatient cardiac telemetry over 21 days detects occult paroxysmal atrial fibrillation (AF) in almost 20% of patients with cryptogenic cerebral ischemia, a new study shows [1].

Investigators reported their results here at the International Stroke Conference 2012.

In one-third of patients discharged with stroke or transient ischemic attack (TIA), the etiology of the event remains undetermined, the researchers note. While paroxysmal AF is known to be a possible cause, factors predicting paroxysmal AF have not been defined.

"Potentially, one in three stroke patients could be monitored with this technology," said lead author Dr Daniel J Miller (Henry Ford Hospital, Detroit, MI) at a press briefing. "The reason it is important to identify paroxysmal AF is that it may change treatment."

He explained that standard antiplatelet treatment would not be sufficient for this population, which would benefit instead from anticoagulation.

Monitoring for 21 days detected paroxysmal AF in 17%

Investigators hypothesized that mobile cardiac outpatient telemetry (MCOT, CardioNet), if continued for 21 days, would detect a high rate of paroxysmal AF and that risk factors could be identified for predicting it.

The study was a retrospective analysis on 156 patients (mean age 68.5) evaluated by MCOT monitoring within six months of a cryptogenic stroke or TIA. Almost all (97%) were not on anticoagulation, and 51% were not on rate-control medication.

Half the patients were female and half were African American; 79% of the index events were stroke, and 35% of patients had a history of prior stroke or TIA, with strokes being fairly mild (NIH Stroke Scale 3.0). Hypertension was present in 87%, but heart failure was seen in only 8%, Miller noted.

A multivariate analysis with survival regression analysis was performed using demographic, clinical, neuroimaging, and cardiac imaging characteristics to determine predictive risk factors for the detection of paroxysmal AF on MCOT.

Twenty-seven of the patients (17.3%) were found to have paroxysmal AF on monitoring. Two-thirds of the episodes lasted <30 seconds; 26% were >30 seconds, and 7.5% of episodes were persistent, he reported.

The study also showed that as the duration of monitoring increased, so did the rate of detection of paroxysmal AF.

Paroxysmal-AF detection significantly increased from 3.8% in the initial 48 hours, to 9.2% at seven days, to 15.1% at 14 days, to 19.5% by 21 days (p<0.05).

"MCOT is a robust diagnostic tool for the identification of paroxysmal AF in patients with cryptogenic stroke and TIA," Miller said. "Length of monitoring is strongly associated with the likelihood of detection of the condition."

In a multivariate Cox regression analysis, the strongest predictors of paroxysmal AF were premature atrial complexes on ECG and female gender. Other significant factors were increase by 1 cm in left atrial diameter, decrease by 10% in left ventricular ejection fraction (LVEF), and increase by 1 point in NIH Stroke Scale.

Predictors of paroxysmal AF

Hazard r atio (95% C I )
All p atients

6.2 (1.9-19.5)

13.7 (2.7-68.6)

2.3 (1.1-5.0)

1.8 (1.2-2.7)
Stroke p atients

4.6 (1.3-16.4)

17.0 (3.2-91.4)

1.2 (1.1-1.4)

Study corroborates other findings

Dr Jose Biller (Loyola University Stritch School of Medicine, Chicago, IL) commented that the study corroborates findings from two previous studies that have observed that some 20% of patients labeled as "cryptogenic" may have paroxysmal AF.

He said that more information about duration of the paroxysmal AF would be helpful. "This is important in deciding whether to change therapy."

During the discussion, Miller said that for "very short episodes of atrial fibrillation" he would prescribe medication on a case-by-case basis. "Some patients have short episodes, but many of them, if they have other risk factors putting them at high risk for future stroke or have multiple recurrent short episodes under 30 seconds, they do end up being anticoagulated," he said.

"But as of now, we don't know if these episodes less than 30 seconds are associated with the same stroke risk as those greater than 30 seconds," he added. "But short episodes tend to become longer episodes and then persistent atrial fibrillation over time."

Biller said he uses MCOT and estimated that it alters treatment about 15% of the time. "The study's observations are important, because they can change the paradigm of management," he said.

Would even more prolonged monitoring be incrementally more beneficial? "My cardiology colleagues at our institution believe that beyond 30 days, there is no added value," he said.

The study received no outside funding. CardioNet provided monitoring resul ts and records at no charge. Miller and Biller reported no conflicts of interest.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.