Aims/methods: Treatment of patients with multivessel coronary artery disease (CAD) has been an ongoing focus of recent clinical studies, questioning the ideal treatment. Randomized trials comparing coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) have so far only included a minority of screened patients. Therefore, we analyzed data from 679 consecutive "all-comer" patients, who underwent PCI in at least two main vessels. Expected in-hospital mortality for CABG was calculated using the EuroSCORE and compared to the observed mortality rate during in-hospital as well as long-term follow-up.
Results: The patients were suffering from 2.5 +/- 0.6 diseased vessels, and 2.8 +/- 1.0 lesions were stented (32% of patients received at least one drug-eluting stent [DES]; 20% of lesions were treated with DES). Forty-seven percent of patients were treated for acute coronary syndrome (ACS) (N = 176 ST-elevation myocardial infarction [STEMI]; N = 140 non-ST-elevation myocardial infarction [NSTEMI]). The EuroSCORE was significantly higher in ACS patients compared to stable patients (logistic: STEMI 16.3 +/- 17.2; NSTEMI 13.6 +/- 13.0; stable CAD 3.9 +/- 4.2). The observed in-hospital mortality (STEMI 13.0%; NSTEMI 2.9%; stable CAD 1.7%, P < 0.001) was far lower than the estimated 30-day mortality. Cox regression analysis identified an elevated logistic EuroSCORE (HR per quartile 2.7, P = 0.003), severely reduced left ventricular ejection fraction (HR 2.7, P < 0.001), elevated C-reactive protein (HR 1.8, P = 0.012), and chronic renal failure (HR 2.8, P = 0.001) as independent predictors of long-term mortality.
Conclusions: The EuroSCORE, which is routinely used to estimate the perioperative risk of patients undergoing CABG, also predicts short- and long-term prognosis of patients undergoing MV-PCI. The observed mortality of patients undergoing MV-PCI seems to be much lower than the estimated mortality of CABG.