Aims Cardiovascular diseases remain the most common cause of death in older adults. Guidelines state that advanced age alone should not limit the use of invasive therapy. However, coronary angiograms and subsequent revascularization are often not carried out in octogenarians. The benefit/risk balance of an invasive strategy and the decision‐making process are not clearly defined. The aim of the present study was to assess the decision‐making process, and the in‐hospital and long‐term mortality based on the clinical presentation, the diagnostic approach (coronary angiogram or conservative) and the therapeutic management (revascularization or not). Methods The present study was a single‐center retrospective analysis. Results A total of 522 patients aged ≥80 years, with a diagnosis of coronary disease were included from 2003 to 2009. The mean age was 82 ± 2.6 years. A total of 195 of 522 (37%) presented with a ST segment elevation myocardial infarction (STEMI). A coronary angiogram was carried out in 316 patients (60.5%) and 71% were treated by percutaneous coronary revascularization. A total of 39.5% were considered ineligible for a coronary angiogram due to cardiological reasons or comorbidities. Excluding cardiogenic shock, overall in‐hospital mortality was 4.9%. Clinical presentation strongly influenced both in‐hospital and 6‐month mortality rates (cardiogenic shock 20% and 28.7%, stable angina 1% and 4.1%, respectively, P < 0.001). Long‐term mortality was reduced in the coronary angiography arm compared with the conservative group (14.3% vs 20.9%, P = 0.04) whether or not revascularization was carried out. Conclusion In the present study, in octogenarians, long‐term mortality was lower in the group of patients who underwent a coronary angiogram, regardless of revascularization. The selection process for coronary angiography and angioplasty was mostly influenced by the existence of age‐associated comorbidities. Risk prediction models are required to reduce age‐dependent biases. Geriatr Gerontol Int 2014; ●●: ●●–●●.