Although in the past years a reduced mortality in peri-operative care has been observed, cardiovascular mortality and morbidity still is a major burden in patients undergoing noncardiac surgery and its evaluation is still a difficult task. An accurate risk stratification can improve quality of peri-operative care and may improve survival, while reducing healthcare costs. In clinical practice, we make our assessment of a patient's cardiac status based on history, examination and investigations, together with risks related to the surgical procedure, to generate an 'individualized cardiac risk assessment'. At the present, risk stratification with clinical risk score and cardiac testing have been shown to be suboptimal in identifying high-risk patients. Surgery, like exercise, increases oxygen consumption. Indeed, one of the key elements in determining risk assessment is exercise intolerance, but future research in this field is needed to clarify this statement. Cardiopulmonary exercise testing (CPET) provides a global assessment of functional capacity involving and integrating the physiological measurement during incremental exercise. The pattern of CPET's variables identifies the abnormal exercise capacity, often providing an objective evaluation of cause and, moreover, predicting outcomes in both apparently healthy and chronic disease populations. An anaerobic threshold VO2 above 11ml/kg per min seems to identify individuals with a very low surgical risk even if undergoing major surgery. This review is focused on tools of risk assessment in patients undergoing noncardiac surgery and on the physiological basis for CPET in detecting patients 'at risk'. © 2013 Italian Federation of Cardiology.