• K Harris, D Edwards, and J Mant.
• General Practice & Primary Care Research Unit, Strangeways Research Laboratory, University of Cambridge, Wort's Causeway, Cambridge CB1 8RN, UK.
• J R Coll Physicians Edinb. 2012 Jan 1; 42 Suppl 18: 5-22.

AbstractAtrial fibrillation (AF) is an arrhythmia of increasing prevalence associated with a reducible risk of stroke. We conducted a systematic review to address five questions relating to how we can best detect AF: 1. Are there useful screening tests to determine who should have a 12-lead electrocardiogram (ECG)? Potential screening tests, all with acceptable sensitivity, include pulse palpation, single-lead ECG and newer technologies such as modified sphygmomanometers or a finger probe device. Pulse palpation has a high number of false positives, but is the cheapest method. 2. Is it more effective to offer 12-lead ECGs to the whole population (or specific sub-groups) or only to those who screen positive for AF? The cost-effectiveness of new devices, such as a modified blood pressure monitor, needs to be assessed. It is more cost-effective to opportunistically screen people rather than to offer a 12-lead ECG to everybody. 3. How accurate are different healthcare professionals and interpretative software at diagnosing AF on ECG? Definitive diagnosis of AF should be by 12-lead ECG, interpreted by someone with appropriate expertise. Computer software is not currently sensitive enough to be used alone to diagnose AF on ECG. Primary care practitioners may not accurately detect AF on ECG, but consistently high accuracy can be achieved by healthcare professionals with adequate training. 4. How best can we diagnose paroxysmal atrial fibrillation (PAF)? In patients in whom PAF is suspected, longer periods of monitoring will detect more cases of PAF. 5. What is the impact of the use of different ECG monitoring strategies (e.g. Holter monitoring, serial ECGs, continuous ECG) on AF detection rates post-stroke? In patients post-stroke, a single ECG will miss cases of PAF which can be detected by longer duration monitoring such as Holter monitoring, cardiac event recorders and serial ECGs. Further research into the cost-effectiveness of these methods, the duration of monitoring required and the clinical significance of the PAF detected is needed.

19 keywords...

hide…