BACKGROUND: Renal artery stenosis (RAS) remains underdiagnosed because of nonspecific clinical manifestations, including in patients with coronary artery disease (CAD). AIMS: To estimate the prevalence and identify predictors of RAS in patients with CAD undergoing coronary angiography. SETTING: University-based medical centre. METHODS: We enrolled 650 consecutive patients (mean age=67+/-10 years, 80% men) with confirmed CAD. All patients underwent selective renal arteriography in the same procedure. We estimated the prevalence of RAS, defined as a >50% lesion. Multiple variable analysis of factors associated with presence of RAS was carried out using a logistic regression model. Variables that emerged as predictors by single-variable analysis were included in the model, along with variables that were tentatively associated with RAS, based on a literature review. RESULTS: RAS was detected in 94 patients (14.5%, 95% CI: 11.8-17.2%), including 20 (3.1%) with bilateral lesions. By single-variable analysis and presence and number of coronary artery stenoses (P<.001), hypertension (P=.001), and creatinine clearance <90 ml/min (P<.001) were associated with an increased risk of RAS. By multiple variable analysis, male sex (P<.05), presence and number of coronary artery lesions (P<.01), hypertension (P=.001), and renal insufficiency (P<.001) predicted the presence of RAS. CONCLUSIONS: The main clinical predictors of RAS in patients with CAD were hypertension, renal insufficiency, and multivessel CAD. These observations might help defining a high-risk subgroup of patients in need of meticulous investigations of both CAD and RAS.