Preoperative nutritional status was assessed by: the percentage weight loss (% WL), body weight in relation to reference weight (WI), arm muscle circumference (AMC), and S-albumin (S-Alb) in a prospective study of 215 noncancer patients classified into three groups according to type of surgery: major vascular, minor vascular, and abdominal. The clinical significance of the nutritional markers was assessed by correlations to postoperative outcome and the time spent in the hospital after surgery. The influence of age on nutritional markers and clinical variables was evident but was ruled out in all correlations. If malnutrition was classified as two or more abnormal values in the nutritional markers (% WL, WI, AMC, S-Alb), the overall frequency was 12%, highest in the major vascular surgery group (18%) and lowest in the minor vascular group (4%). Patients with low nutritional status stayed an average of 29 days in the hospital compared to 14 days if the nutritional status was normal (p less than 0.01). The overall complication frequency was higher in patients with low nutritional status compared to normal status (48% and 23%, respectively, p less than 0.01). The frequency of serious complications was 31% in undernourished and 9% in well-nourished patients (p less than 0.05). Various nonnutritional variables such as age, diagnosis, and duration of surgery were shown to increase the predictive ability of nutritional status. The results of this study confirm that nutritional state per se is predictive for postoperative outcome even when variables were stabilized for different backgrounds with covariation to nutritional status.