Background: Left main stenosis (LMS) occurs in 5–7% of patients with coronary artery disease. Half of patients with left main coronary artery (LMCA) disease die within few years after the diagnosis.

Aim: To evaluate survival of patients with LMCA disease treated with coronary artery bypass grafting (CABG), percutaneous coronary intervention (PCI), or managed medically due to lack of consent for CABG or being considered unsuitable candidatesfor CABG/PCI.

Methods: In 2006–2008, a significant LMS was found in 257 (5.14%) patients, and 98.44% of these patients were followed upfor on average 15.1 months. The patients were divided into 5 groups according to the treatment used. CABG was performedin 67% of patients, PCI of an unprotected LMS in 8% of patients, and 12% of patients were treated with PCI after a previous CABG (protected LMS). The remaining patients were managed medically: 4% were not considered suitable for CABG, and9% did not give their consent for CABG.

Results: Total mortality in the overall study group (n = 253) was 14.6%. Multivessel disease was more frequent in the CABG group (60.9% vs. 15.8%, p < 0.001). Mortality in CABG and PCI groups was comparable (11.4% vs. 15.8%). Patients in the PCI group were more frequently hospitalised due to recurrent angina (21.1% vs. 3.0%, p < 0.001) and the need for repeated revascularisation (15.8% vs. 1.2%, p < 0.001). Compared to the CABG group, patients considered not suitable for CABG hadlower left ventricular ejection fraction (LVEF) (36.55% vs. 51.04%, p < 0.001) and a higher mortality risk as estimated by the EuroScore. Mortality among patients deemed unsuitable for CABG was 54.6% (p < 0.001) and myocardial infarctions were observed more frequently in this group (18.2% vs. 2.4%, p < 0.01). In comparison to the CABG group, patients who did not consent to CABG were older (71.04 vs. 65.99 years, p = 0.027), had lower LVEF (44.05% vs. 51.04%, p = 0.004), were less frequently hospitalised due to acute coronary syndromes (17.4% vs. 40.8%, p = 0.03), and had a smaller degree of LMS (63%vs. 71%, p = 0.027). Mortality in this group was comparable to the CABG group (17.4% vs. 11.4%). The majority of patients who underwent previous CABG needed repeated revascularisation: PCI of a protected LMS was performed in 27% of patients,PCI of other native coronary arteries in 39% of patients, and PCI of a bypass graft in 7% of patients.

Conclusions: PCI of unprotected LMCA may be an equally effective revascularisation method as CABG. High mortality (55%) due to concomitant diseases was observed among patients with LMS who were deemed unsuitable candidates for CABG. Prognosis among patients who declined CABG was relatively good and might have been related to the small number of patients and different patient characteristics in this group.