Forty-four disease-free patients were evaluated a mean of three years after total gastrectomy (TG) or subtotal gastrectomy (SG) for cancer. The investigation encompassed nutritional assessment by standard anthropometric and biochemical indices; evaluation of the nutritional intake based on 24 h recall and of appetite status on a visual analogue scale; and anamnestic analysis of postcibal symptoms and syndromes. Body weight had declined progressively until the 15th postoperative month after both TG and SG. Weight loss, as well as the general nutritional status index (actual body weight/usual body weight +/- actual body weight/desirable body weight + measured arm muscle circumference/reference arm muscle circumference x 33), had dropped more significantly in patients undergoing TG than those having SG (p less than 0.05). The principal body compartment change was observed in the fat content which was severely depleted, whereas the somatic proteins were relatively spared and the visceral proteins and remaining biochemical variables were in the normal range. Protein intake was not significantly different in the two groups, but caloric intake was significantly lower and the number of meals significantly higher after TG (p less than 0.05). These data suggest that malnutrition after TG is relatively mild and that this operation causes only a limited impairment of the nutritional state, and spares most of the nutritional variables of clinical interest in comparison with SG. These findings argue in favor of TG when clinically indicated without excessive concern about postoperative nutrition.