BACKGROUND This study was performed to prospectively evaluate the safety, efficacy, and cost of injecting P-colloid into joints of children with hemophilia and synovitis to decrease the rate of joint bleeding. PATIENTS AND METHODS Eligibility included a diagnosis of hemophilia, history of more than six hemorrhages into a joint within a 6-month period, and evidence of synovitis by objective imaging. With written, informed consent, 0.25 to 1.0 mCi of P-colloid was injected into the problem joints. Safety was monitored by external beta-scanning and physical assessment. Efficacy was determined by analysis of the change in joint hemorrhage frequency from 6 months before and up to 96 months after the injection using a signed-rank test. Physical assessment and pain assessment were analyzed similarly using values obtained within 1 week before and 6 months after the radiosynoviorthesis. Cost was modeled using charges from the authors' institution in relation to existing alternative therapies. RESULTS One hundred injections were given into 91 joints in 59 children. Seven children had high-titer neutralizing antibodies to factor VIII or IX. Nine children were infected with HIV. Joints injected included 44 ankles, 19 knees, 27 elbows, and 1 shoulder. Nine joints required reinjection. All children showed a significant decrease in bleeding rate (P < 0.0001) and pain (P = 0.03), with improved physical function (P = 0.02). In one child acute lymphocytic leukemia developed, but it was judged unrelated to the two P injections that he had received 3 and 10 months before the leukemia diagnosis. There were no cases of bleeding, infection, or inflammation caused by the injection. Cost was substantially less than medical and surgical alternatives. CONCLUSIONS Radiosynoviorthesis is effective in limiting the frequency of joint hemorrhage, decreasing pain and improving function in children with hemophilia. However, long-term safety studies are needed.