Introduction Stress echocardiography (SE), single photon emission computed tomography (SPECT) and cardiac magnetic resonance (CMR) imaging are widely used for assessment of coronary artery disease (CAD). Patients with atrial fibrillation (AF) can provide technical challenges for imaging (e.g. irregular heart rates and excessive tachycardia). In addition, the usual practice of continuing beta-adrenoceptor antagonists (betablockers) may lead to sub-optimal stress. We undertook this study to determine the evidence supporting the diagnostic and prognostic value of these techniques in patients with AF and policies on betablocker usage in the identified studies.

Methods The PubMed database was searched for articles published between 1980–2017. Studies were included if they assessed SE, SPECT or stress perfusion CMR as a diagnostic test for CAD. We used relevant search terms to identify studies examining prognostic value of these tests in AF. Original research studies were only included if patients also underwent coronary angiography.

Results A total of 277 studies (SE-97, SPECT-135, CMR-93) met inclusion criteria, comprising 44 180 patients. The majority of studies [231/276 (83%)] did not specify whether patients with AF were enrolled. Only 4% (11/276) of studies specifically included patients with known AF but 13% (35/276) of studies specifically excluded patients with AF.

Regarding study protocols on betablocker medication, 50% of studies did not specify whether this was discontinued prior to imaging. Furthermore, whilst 23% (63/276) of studies specifically discontinued betablockers, 27% (74/276) studies continued them prior to the test.

We identified five studies (2 CMR, 2 SPECT and 1 SE), comprising 365 patients, that examined diagnostic accuracy specifically in AF patients. All patients underwent angiography in 3 studies and selected patients only in 2 studies. The range of diagnostic accuracy was 70%–88%. We identified only three studies examining the prognostic value of stress testing in AF patients (table 1).

Conclusions There is limited published data regarding both the diagnostic and prognostic value of three widely-used functional imaging techniques (SE, SPECT and CMR) in patients with suspected CAD and co-existent AF. There was also marked heterogeneity within these studies regarding continuation or withdrawal of betablockers, which may affect detection of ischaemia. Given the challenges AF can pose to these techniques, there is an urgent need for prospective studies to identify the optimal strategy in these patients.