Acute myocardial infarction continues to be a major cause of mortality and morbidity, and infarct size is the major determinant of patients’ prognosis. The only way to reduce infarct size is to restore coronary blood flow; however, reperfusion not only salvages myocardium from infarction, but also brings about additional ‘reperfusion injury’. Ischaemic post-conditioning, ie the repeated mechanical brief interruption of coronary blood flow during early reperfusion, attenuates reperfusion injury and reduces the ultimate infarct size. 1, 2 Ischaemic post-conditioning was first characterized in dogs 3 and confirmed in all species tested so far, provided they were young and healthy. 4 Staat et al. in a landmark proof-of-concept study translated the experimental finding to a small cohort of select patients with acute myocardial infarction; the cumulative creatine kinase (CK) release during reperfusion was reduced by a protocol of four cycles of 1min angioplasty balloon inflation/1min deflation before full reperfusion. The cumulative CK release vs. the area at risk, as determined using ventriculography before reperfusion from the abnormally contracting myocardium, was also reduced. 5 Reduction of infarct size by ischaemic post-conditioning in patients with acute myocardial infarction was subsequently confirmed in a number of small studies, using as endpoints CK 6–9 or troponin release, 9 ECG ST-segment resolution, 8 and anatomic infarct size, as determined by single photon emission computed tomography 7, 9 or delayed enhancement magnetic resonance imaging (MRI) 10. A more long-term benefit from ischaemic post-conditioning was seen in echocardiographic wall motion after 8 weeks 6 or left ventricular ejection fraction (LVEF) after 1 year. 9 One recent study 11 using the same ischaemic post-conditioning algorithm of four cycles of 1 min re-occlusion/reperfusion found no protection in terms of either CK and troponin release or infarct size by MRI. Now, a study by Freixa et al. 12 also does not confirm protection by ischaemic post-conditioning, again using the established algorithm of four cycles of 1 min re-occlusion/reperfusion: CK release was not reduced, and troponin I release was even enhanced. Infarct size, as determined from delayed enhancement MRI, was not reduced, nor was the relationship of infarct size to area at risk, as determined from coronary angiography using the BARI score. 13 A calculated myocardial salvage ratio was even less with ischaemic post-conditioning. Whereas control patients had a slight improvement of LVEF over 6 months by MRI, patients undergoing an ischaemic post-conditioning protocol did not. These two recent negative studies on ischaemic post-conditioning are disturbing, but also welcome because they challenge and make us critically re-examine the prevailing concept of cardioprotection before its broader translation to clinical practice. Technically, both negative studies used MRI, although both avoided the Achilles heel of MRI-based infarct size assessment, ie the quantification of area at risk from T2 weighted oedema, 14 but used either ventriculography 11 or angiography 12 to this end. Both also found no protection in terms of CK or troponin release, although it is unfortunate that in the study of Freixa et al. the area under the curve for CK was not reported, which would be less dependent than peak CK on release kinetics in the presence of microvascular obstruction. Also, no long-term benefit for LVEF was seen with either echocardiography or MRI in the study of Freixa et al., and therefore the apparent lack of protection in this study 12 and the previous 11 negative study cannot just be ascribed to the use of MRI. One …