Abstract Excessive pain after surgery causes many kinds of endocrine, metabolic and inflammatory responses, which may increase postoperative morbidity and mortality - especially among elderly patients. This study evaluated the effect of peripheral and central pain relief techniques after joint surgery. Intravenously administered doses of 100 mg, 200 mg and 300 mg of ketoprofen decreased the requirement for opioid (fentanyl) in a dose-dependent manner by 38%, 45% and 53%, respectively, compared with a placebo, without any noticeable ceiling-effect, when administered after hip and knee arthroplasty. Patients receiving a 300 mg dose of ketoprofen had significantly lower postoperative pain scores than those receiving a placebo. There were no significant differences in incidences of nausea and vomiting, or in the amount of bleeding between the ketoprofen and placebo groups. Intravenous doses of 200 mg of ketoprofen, 150 mg of diclofenac, and 120 mg of ketorolac produced similar postoperative pain scores and requirement for opioid (fentanyl) with no intergroup differences in the incidence of nausea and vomiting and in the amount of bleeding, when administered after hip arthroplasty. The addition of ropivacaine, 1 mg·ml-1, did not decrease the requirement for epidural fentanyl administered via a patient-controlled analgesia device for postoperative pain relief after hip arthroplasty. Both drug infusions provided effective pain relief. The most common adverse effect was pruritus, which occurred in a similar number of patients in both groups. An interscalene brachial plexus block with ropivacaine decreased the dose of PCA-delivered oxycodone by 78% after arthroscopic shoulder surgery while subacromial bursa blockade with ropivacaine decreased it by only 11 % compared to a placebo during the 20 hour study period. Postoperative pain scores were significantly lowest with a interscalene brachial plexus block.