Neutrophil gelatinase-associated lipocalin (NGAL) has been proposed as an early marker for estimating the risk of contrast-induced acute kidney injury (CI-AKI). However, the predictive value of baseline serum NGAL levels for CI-AKI remains unclear. Serum NGAL was measured before percutaneous coronary intervention in 633 patients with coronary artery disease. The primary clinical endpoints were a composite of major adverse cardiac and cerebrovascular events (MACCEs; cardiac death, myocardial infarction, stroke, and any revascularization). The mean follow-up duration was 29.4 months. Ninety-eight (15.5%) patients developed CI-AKI. Compared with patients without CI-AKI, baseline serum NGAL was higher in patients with CI-AKI (149.6 ± 88.8 ng/mL vs. 138.0 ± 98.6 ng/mL, p = 0.0279), although serum creatinine and estimated glomerular filtration rate were not different between groups. Patients in the highest tertile of baseline serum NGAL showed a significantly higher rate of MACCEs (10.5% vs. 3.8%, p = 0.02). Using the first tertile as a reference, the adjusted hazard ratios for MACCEs in patients in the second and third tertiles of NGAL were 2.151 (confidence interval (CI) 0.82 to 5.59, p = 0.116) and 2.725 (CI 1.05 to 7.05, p = 0.039), respectively. Baseline serum NGAL is a reliable marker for predicting CI-AKI, and high serum NGAL levels are associated with a higher incidence rate of long term MACCEs.