Background. One of the criteria for determining indications for myocardial revascularization is the data of invasive coronary angiography (CAG) and the task of clinicians is to select patients for this study and then determine a particular treatment strategy. For this purpose, the corresponding diagnostic algorithms have been developed, as reflected in the international recommendations. It would seem that the problem should not exist, however, real clinical practice shows that it is not possible to organize an effective selection, there is a noticeable difference in the frequency of detection of intact coronary arteries (CA) in different clinics and regions. The aim of the study was to analyze cases of detection of intact CA in clinics of different levels of subordination (federal and municipal). Material and methods. A retrospective analysis was made of the history of patients who were hospitalized with suspicion or with a previously diagnosed coronary artery disease (CAD) for conducting CAG in the clinics of the Kemerovo Cardiology Center and who did not have occlusive stenotic CA lesions. Depending on the hospital base, 2 groups of patients were formed: 1 group — patients of the federal clinic (NII KPSSZ, n=100), 2 group — patients of the municipal institution (CCD, n=84). Results. When the planned CAG was performed, the frequency of detection of intact CAs did not differ in clinics of various levels of subordination (federal and municipal) and amounted to about 25% in both cases. Among patients with intact CA, there were no differences in two institutions in the initial pre-test probability of CAD (53.5 and 51.5%, p=0.664) and the frequency of exercise tests (4.0 and 10.7%, p=0.076). About 1/3 of patients in group 1 and 1/2 — 2nd group were hospitalized on the initiative of a cardiologist (p=0.013), 39 and 50.1% of cases were sent by a therapist (p=0.115), 19 and 3.6% — by nurses (p=0.001), in 15 and 2.4% of cases — it was not possible to identify a specialist (p=0.003). In the federal center, more often the indication was the examination of patients before the operation for valvular heart disease (16.0 and 1.2%, p<0.001), less often — suspected coronary artery disease (44.0 and 61.9%, p=0.015) and changes in the electrocardiogram (6.0 and 20.2%, p=0.003). Conclusion. The results of this study should be taken into account when considering the issues of increasing the effectiveness of detection of obstructive lesions.