PURPOSE: A number of reports have shown a direct analgesic effect of opioids by way of the peripheral receptors. Nevertheless, only a very few studies have reported using opioids in the temporomandibular joint (TMJ), and nearly all of them were connected to surgical patient groups. The present study was designed to evaluate the analgesic efficacy and safety of repeated intra-articular morphine applications compared with a local anesthetic and saline solution in the management of TMJ pain. PATIENTS AND METHODS: A total of 48 patients with articular pain related to the TMJ were entered in a clinical, prospective, randomized, double-blind, single-center study. The analgesic effect of repeated intra-articular infiltration with morphine (5 or 10 mg morphine sulfite), bupivacaine 0.5% (Carbostesin; AstraZeneca, London, UK), and isotonic saline solution as a placebo in the TMJ was examined. The efficiency after 3 injections of the same substance with an interval of 48 hours between each application was measured using a pain relief scale, visual analog scale, pain intensity scale, and the potential need for accessory peripheral analgesics (paracetamol). RESULTS: All patients showed, independent of the treatment group, pain relief within 60 minutes after the first injection. Patients with saline and Carbostesin reported almost complete pain recurrence before the second injection. At 1 week after the last and third injection, the 10-mg morphine group still showed a distinct effect, with 16.7% reporting complete (no pain) and 41.7% distinct pain relief. In addition, 33.3% had a poor response and 8.3% had no improvement. None of the other groups reported complete improvement; however, 25% of the patients who received 5 mg morphine had distinct pain relief, and 50% had at least poor pain relief. In the Carbostesin group, distinct improvement was reported by 8.3%, with a poor response in 41.7%, and no effect in the remaining 50%. Patients treated with saline had a poor response in 16.7%, but most (83.3%) reported no improvement 1 week after treatment. CONCLUSIONS: Independent of the applied substances, initial pain relief can be registered in the TMJ: either from the arthrocentesis effect or at least the placebo effect. Morphine at a dosage of 10 mg showed the best and most long-lasting analgesic efficiency. Morphine, in general (5 and 10 mg), and, with limitations, Carbostesin were more or less efficient for postoperative pain control but without distinct effects in the long term. With regard to our results, we can recommend intra-articular morphine application at a dose of 10 mg for pain management. Carbostesin showed no promising long-term effects.