TY - JOUR T1 - Coronary angiography in heart failure: when and why? Uncertainty reigns JF - Heart JO - Heart SP - 548 LP - 549 DO - 10.1136/heartjnl-2017-312200 VL - 104 IS - 7 AU - Mark C Petrie AU - Paul Rocchiccioli Y1 - 2018/04/01 UR - http://heart.bmj.com/content/104/7/548.abstract N2 - Remarkably in 2017 cardiologists do not know when and why patients with heart failure should undergo coronary angiography. This uncertainty is strikingly illustrated in the Heart paper1, in a subanalysis of the BIOSTAT study (69 centres over 11 countries), which reports that only 12% of patients with worsening heart failure underwent coronary angiography. This low rate is similar to that reported in a study of 67 000 patients in the USA, in which 16% underwent coronary angiography within 90 days of a new diagnosis of acute decompensated heart failure.2 Why would a coronary angiogram be recommended for patients with heart failure and when should it be performed? One clear indication for coronary angiography is to establish which patients are potential candidates for coronary artery bypass grafting (CABG). The Surgical Treatment for Ischemic Heart Failure (STICH) trial reported benefit of CABG compared with medical therapy in terms of all-cause mortality, cardiovascular (CV) mortality and CV hospitalisations after 10 years of follow-up.3 If a patient is a candidate for CABG (from a perspective of functional status and comorbidity), only coronary angiography can inform the clinician whether or not the patient has coronary artery disease amenable to surgical revascularisation. Some cardiologists are advocates of performing non-invasive testing for ischaemia or viability prior to performing coronary angiography. These modalities have a very weak evidence base when it comes to deciding whether or not to recommend CABG for heart failure. In the STICH trial, the presence or absence of ischaemia or viability did not identify patients with more or less to gain from CABG.4 The viability literature is especially poor, relying on data such as a meta-analysis of historical, observational small … ER -