Summary:Early non-invasive imaging of atherosclerosis and in particular the detection of lesions at risk with high specifi city could significantly affect cardiovascular morbidity and mortality. Conventional nuclear medicine approaches, in particular using autologous radiolabeled lipoproteins, can be related to histopathological fi ndings; however, they fail to identify lesions at risk. Positron emission tomography (PET) tracers with much better physical properties have been examined, the most detailed information being available for F-18-deoxyglucose (FDG) and F-18-sodium fl uoride (NaF). These two approaches are sensitive to different biochemical mechanisms, i.e. infl ammation and microcalcifi cation. Initial enthusiasm, in particular for F-18-FDG, has disappeared, although for F-18-NaF there is some hope, but this is not a breakthrough. No tracer is available so far that is able to identify a specifi c characteristic of a lesion prone to rupture. Other PET tracers in the pipeline have been examined, mainly in experimental models and only a few in patients, but they failed to contribute signifi cantly to early lesion discovery and do not support great expectations. The key question is: Do we understand what we see? Moreover, methodological problems, a lack of standardization of imaging protocols and aspects of quantifi cation provide a wide range for potential future improvements. While monitoring a therapeutic intervention seems to be possible for both F-18-FDG and F-18-NaF, highly specifi c early identifi cation of lesions at risk by PET imaging is still far away. As of today, PET is not ready for routine clinical judgment of atherosclerotic lesions at risk to rupture. Even if all these problems can be solved, radiation exposure will still remain a concern, in particular for repeated studies.