PURPOSE OF REVIEW:To review the evidence for and against imaging as a means of selecting patients for cardiac resynchronization therapy (CRT). RECENT FINDINGS:There is no evidence that either the QRS interval on the surface ECG or dyssynchrony measured by imaging is of any practical value in predicting the clinical response to CRT in patients with a dilated and dysfunctional left ventricle. Careful assessment of the patient, so that therapy can be logically aligned with treatment goals, such as improving symptoms or prognosis, is the only useful method for selecting patients. Simple clinical evaluation may be as effective as, or more effective than, more complex assessments in predicting treatment benefits. Patients with a low blood pressure and moderate functional mitral regurgitation might benefit more, in absolute terms, from CRT. The benefits of adding a defibrillator to CRT are modest and, for many patients, uncertain. SUMMARY:Echocardiography, which was supposed to facilitate the introduction of CRT, may have become the greatest barrier to its appropriate implementation. Cardiac dyssynchrony, measured by echocardiography prior to implantation, may not be the substrate for the effects of CRT.