Improving Thromboprophylaxis Using AFib Diagnostic Capabilities in ICDs: The Multicentre Italian ANGELS of AFib Project

Editor's Note: This article is based on Boriani G, Santini M, Lunati M, et al. Improving Thromboprophylaxis Using Atrial Fibrillation Diagnostic Capabilities in Implantable Cardioverter-Defibrillators: The Multicenter Italian Angels of AF Project. Circ Cardiovasc Qual Outcomes 2012; 5:182-188.


Atrial fibrillation (AF) is a known risk factor for stroke. The increased use of cardiac implantable devices allows continuous rhythm monitoring and accurate detection of atrial fibrillation. Prior studies have shown increased thromboembolic rates in patients with asymptomatic atrial fibrillation detected by implantable cardiac devices.1-3 Implementation of pathways for anticoagulation determination based on device diagnostics has not been uniform. The Angels of AF Project4 (NCT01007474) looked at an ICD population and evaluated the presence of AF and appropriate anticoagulation, stroke and thromboembolic rates, and implementation of a strategy to enhance appropriate oral anticoagulation.


This study included 50 centers that participate in the Italian Clinical Service Project, which is a national cardiovascular data repository. All patients received either dual chamber or triple chamber ICDs for standard guideline indications and data were prospectively acquired through routine visits or follow-up phone calls. Device AF detection was programmed to atrial rates greater than 171 beats per minute for at least 32 ventricular events during which there was a >1:1 A:V pattern consistent with an atrial arrhythmia. Device diagnostic algorithms were utilized for measurement of AF occurrence and duration.

A total of 3,438 consecutive patients with ICDs were followed as part of a national registry of 50 centers involved. The active intervention arm of this study involved 15 centers that received reports regarding AF occurrence in duration, stroke risk factors, and oral anticoagulation use. The remaining 35 centers did not receive specialized reports and followed standard practice.

In the intervention arm, AF duration of 6 hours and was considered clinically significant based on the TRENDS study.2 The CHADS2 score and American College of Chest Physicians stroke risks were calculated by automated algorithms. Medications were cross referenced in those found to be at moderate or high risk for stroke, and this information was presented to the treating physician if the patient was not taking oral anticoagulation. The treating physician then had to document the reasoning for withholding such therapy.  


A total of 3,438 patients were prospectively followed for a median follow-up duration of 25 months. 57% were in the active intervention arm and 43% in the usual care control arm. 683 patients in the intervention arm were found to be at moderate to high risk for stroke and 69% of these were on oral anticoagulation. For the remaining 209 patients, the reports were made available to the patients’ attending physician. These reports resulted in initiation of anticoagulation in 22 patients (10.5%). In 158 patients (75.6%), antiplatelet therapy was deemed to be the best choice and in 27 (12.9%) patients, no anticoagulation was prescribed. The rate of appropriate oral anticoagulation increased from 46% at baseline to 73% at the end of the observation period in the intervention arm. In the control arm, this rate increased from 47% to 57%.

The annual rate of embolic events in patients with atrial fibrillation was 0.7 per 100 patient years versus 0.2 per 100 patient years in those who never experienced atrial fibrillation (p=0.001). 64% of patients with a stroke experienced at least a single AF episode lasting 5 minutes. During the 30 days prior to a thromboembolic event, 33% experienced an episode of atrial fibrillation and 15% experienced embolic events during atrial fibrillation. There were trends for decreased rates of thromboembolism in AF patients treated with anticoagulation with a CHADS2 score of 1-2 as well as a trend towards less thromboembolism in AF patients in the active intervention arm; however, neither of these reached statistical significance.

A post hoc analysis was performed to exclude the potential bias that the centers involved in the active intervention arm may have had a higher compliance with oral anticoagulation therapy. Oral anticoagulation was prescribed in 59% of moderate to high risk patients in the 48 months prior to intervention at those sites, which was close to 57% rates seen in the usual care arm.


Real time feedback to physicians from continuous device monitoring about atrial fibrillation burden and thromboembolic risk factors improves oral anticoagulation prescription in accordance with the available guidelines. 


Atrial fibrillation has a prevalence of up to 1% in the general population. It is a disorder of aging making the burden on health care delivery systems only larger with time. Although guidelines for appropriate anticoagulation to mitigate the morbidity of stroke and thromboembolism exist5, these were largely developed prior to the advent of continuous arrhythmia monitoring from implantable pacemakers and defibrillators. Prior studies have clearly shown an increase in thromboembolic events with AF as detected from an implantable device.1-3 How the frequency and duration of episodes modulates this risk is not completely clear. For example, in the TRENDS2 study, 5.5 hour episodes of atrial fibrillation showed a trend toward a doubling of stroke risk in the following 30 days. An increase in thromboembolic events was also seen in the ASSERT3 trial; however, events lasting only 6 minutes were shown to have a significant impact.

In the Angels of AF Project,4 a clear improvement was seen in guideline adherence of anticoagulation prescription. This did not result in a significant improvement in overall stroke rates but favorable trends were certainly seen. Even though active intervention was performed, more than 25% of those with a moderate to high stroke risk score were not receiving anticoagulation. The reasons for this are not completely delineated in this study; however, these findings highlight the difficulty of anticoagulation use in clinical practice. More data will be forthcoming from the IMPACT study6 (NCT00559988) looking at intervention with anticoagulation based upon atrial fibrillation discovered from ICD interrogation. Additionally, data regarding the CHA2DS2-VASc score in relation to device diagnostics is not available.7

Given that device clinics are not often directly affiliated with the physicians who provide long term continuity of care and often the best judges of candidacy for oral anticoagulation, this study highlights the importance and potential impact of formal communication regarding findings on device interrogation that confer an increased risk of thromboembolism.


  1. Glotzer TV, Hellkamp AS, Zimmerman J, et al. Atrial high rate episodes detected by pacemaker diagnostics predict death and stroke: report of the Atrial Diagnostics Ancillary Study of the Mode Selection Trial (MOST). Circulation 2003; 107:1614-9.
  2. Glotzer TV, Dauod EG, Wyse DG, et al. The relationship between daily atrial tachyarrhythmia burden from implantable device diagnostics and stroke risk: the TRENDS study. Circ Arrhythm Electrophysiol 2009; 2:474-80.
  3. Healey JS, Connolly SJ, Gold MR, et al. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med 2012; 366:120-9.
  4. Boriani G, Santini M, Lunati M, et al. Improving Thromboprophylaxis Using Atrial Fibrillation Diagnostic Capabilities in Implantable Cardioverter-Defibrillators: The Multicenter Italian Angels of AF Project. Circ Cardiovasc Qual Outcomes 2012; 5:182-188.
  5. Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001; 285:2864-70.
  6. The IMPACT of BIOTRONIK Home Monitoring Guided Anticoagulation on Stroke Risk in Patients With ICD and CRT-D Devices. Clinical Last Accessed July 18, 2012.
  7. Lip GY, Nieuwlaat R, Pisters R,et al. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest 2010; 137:263-72.

Keywords: Atrial Fibrillation, Defibrillators, Stroke, Thromboembolism

< Back to Listings