OBJECTIVE To gain insight into the outcomes of postoperative pain treatment using patient-controlled intravenous analgesia (PCIA) with morphine and oral analgesics after total hip or knee arthroplasty (THA or TKA). METHODS Prospective observational study. METHODS In 2006-2007, 154 THA and 110 TKA patients who underwent arthroplasty at the Medical Centre Alkmaar were included in this study. Postoperative pain treatment consisted of a PCIA pump with morphine and oral analgesics. Primary endpoints were effectiveness of pain treatment and the incidence of side effects. RESULTS After THA, 65% of the patients had an acceptable pain score at rest on the day of surgery (day 0), increasing to >95% from 2 days after the operation onwards. On day 1, 43% had an acceptable pain score during movement, increasing to >90% from day 2 onwards. On day 0, nausea occurred in 24% and vomiting in 25% of patients; on day 1, these figures were 21% and 29% respectively. After TKA, 45% of the patients had acceptable pain at rest on day 0, increasing to >or=95% from day 3 onwards. On day 1, pain during movement was acceptable in 24% of the patients, increasing to 80% on day 4. On day 0 following TKA nausea occurred in 37% and vomiting in 39% of patients; on day 1, these figures were 26% and 23%, respectively. CONCLUSIONS In view of the somewhat disappointing effectiveness and the side effects of a PCIA pump with morphine and oral analgesics, adjustments in the treatment of pain following THA and TKA are required. The greatest improvement can be realised on the day of surgery and the first postoperative day.