The relation between weight loss and the risks of major surgery have been investigated for more than 50 years. It can now be said, with some confidence, that the underweight patient has an increased risk of complications following major surgery. This understanding, however, is based on methods of nutritional assessment that are of limited relevance to hospitalized patients whose malnutrition might be due to sepsis, neoplasia, trauma, or starvation. A consequence is the widespread belief that protein-energy malnutrition (PEM) has been overemphasized as a surgical risk factor, and that the many nonnutritional risk factors ought to be implicated more often. An argument is made for a fresh approach to nutritional assessment in order to better identify the individual patient who, by virtue of PEM, stands an increased risk of a complicated postoperative course. It is suggested that an evaluation of the impact PEM has on vital physiologic function provides a clinically relevant defect to identify and treat and a means of monitoring response to nutritional intervention.