Background: Patients with diabetic nephropathy and unprotected left main (LM) coronary artery disease suffer from high cardiovascular morbidity and mortality. Although surgical revascularization is currently recommended in this special patient population, the optimal revascularization method for this distinct patient group has remained unclear. Methods:We collected 99 consecutive patients with unprotected LMdisease and diabetic nephropathy, including 46 patients who had undergone percutaneous coronary intervention (PCI), and 53 who had coronary artery bypass grafting (CABG), with a mean age of 72  10; with 80.8% male. Diabetic nephropathywas defined as overt proteinuria (proteinuria > 500 mg/day) and estimated glomerular filtration rate (eGFR) by the modified Modification of Diet in Renal Disease (MDRD) equation of less than 60 mL/min/1.73m2. The baseline characteristics, angiographic results and long-term clinical outcomes were retrospectively analyzed. Results: The baseline characteristic of all patients were similar except for smokers, low density lipoprotein (LDL) level and extension of coronary artery disease involvement. The median follow-up period was 3.8 years. There were 73 patients (74%) considered as high risk with additive European System for Cardiac Operative Risk Evaluation (EuroSCORE)  6. During follow-up period, the long term rate of all-cause death (PCI vs. CABG: 45.7% vs. 58.5%, p = 0.20) and all-cause death/myocardial infarction (MI)/stroke (PCI vs. CABG: 52.2% vs. 60.4%, p = 0.41) were comparable between the PCI and CABG group, whereas the repeat revascularization rate was significantly higher in the PCI group (PCI vs. CABG: 32.6% vs. 9.4%, p < 0.01). eGFR remained an independent predictor for all-cause death [hazard ratio: 0.97, 95% confidence interval: 0.96 to 0.99; p = 0.002] in multivariate logistic regression. Conclusions: In the real-world practice of high-risk patients with unprotected LM disease and diabetic nephropathy, we found that PCI was a comparable alternative to CABG in terms of long-term risks of all-cause death/MI/stroke, with significantly higher repeat revascularization rate. Given the small patient number and retrospective nature, our findings should be validated by larger-scale randomized studies.