The authors report the results of two clinical studies on postoperative pain relief with PCA. In the first clinical study 44 patients, undergoing gynecologic surgery, were assigned at random to two groups. The first was treated by PCA (infusor Baxter) with morphine i.v. (basal bolus 0.05 mg/kg, loading doses 1 mg every 6-15'), the second with 10 mg morphine i.m. at the end of surgery and then on demand with a lock-out of 6h at least. In the 2nd clinical study, 40 elder patients submitted to orthopedic surgery, were assigned at random to two groups treated with PCA (morphine i.v., basal bolus 0.07 mg/kg, bolus PCA 0.007 mg/kg lock-out 15') and with continuous infusion (i.c.) (basal bolus 0.07 mg/kg, i.c. 0.02 mg/kg/h). Our data were analyzed with Student's "t" unpaired test and showed lower doses of the drug in the groups PCA (1st study PCA 25.98 mg/48h, i.m. 45.45 mg/48h, 2nd study PCA 0.155 mg/kg/12h, i.c. 0.311 mg/kg/12h) and lower rate of side effects in the same groups. Side effects were well controlled using symptomatic drugs. We proposed to our patients, at the end of observation, a questionnaire about general conditions, sleep, pain evaluation using a descriptive scale and retrospective evaluation. Patients and nurses agree PCA. Nursing staff expressed a positive opinion and patients said they benefitted from PCA. As reported, PCA appears from our results, valid and safe in postoperative pain relief.