The efficacy of thymopentin in reducing postoperative infections (PI) was prospectively evaluated in 138 patients with abdominal cancer, who underwent major surgery. Comparable subsets of patients were obtained according to age (cutpoint 65 years) and nutritional status (patients with serum albumin less than 30 g/l or weight loss greater than or equal to 10% with respect to their usual body weight were considered as malnourished). Patients were then at random attributed to a control group and to a thymopentin receiving (Thy) group, in the latter thymopentin (50 mg) was given three times before surgery and three times after operation. All patients received perioperative short term antibiotic prophylaxis and postoperative parenteral nutrition. The severity of Pl was expressed by the sepsis score which was calculated on all those patients developing Pl. Overall complication rate was 26.3% in the control group and 20.3% in the Thy group (p ns). Surgical-related infections occurred in 14 (20.3%) control group patients and in 10 (14.5%) Thy group patients (p ns). The average sepsis score was 10.11 +/- 6.69 in control group (2 patients died) and 6.85 +/- 3.80 in the Thy group (all patients survived) (p less than 0.05). By considering the elderly patients, a reduction in both Pl rate and the average sepsis score was observed in Thy group. Otherwise, no difference was observed in the young patients. Our data suggest that Thymopentin is useful in reducing both the incidence and the severity of Pl in elderly patients.