Nutritional assessment and postoperative morbidity. A prospective study in 286 consecutive surgical patients

286 consecutive surgical patients went through a nutritional assessment following admission to the hospital. Serum albumin (S-alb), serum transferrin (S-transf), triceps skinfold (TSF) and arm muscle circumference (AMC) were recorded. The frequency of pathological values of TSF and AMC varied greatly depending on the reference population used to set the standard. There was no correlation between TSF or AMC and postoperative morbidity. S-alb less than 36 g/l and S-transf less than 2.0 g/l were associated with a marked increase in postoperative morbidity (29.4% and 20.0% respectively as compared to 7.6% in all operated patients). Thus S-alb and S-transf seemed to be able to predict postoperative morbidity. However, this ability varied depending on the diagnosis. Age was another factor with a marked influence on postoperative morbidity. It is suggested that attempts to make nutritional interpretations from S-alb and S-transf take into account patient age and the diagnosis. TSF and AMC seem to be of less value.