Objective: To compare the analgesic effects between the blockade and bipolar thermal radio frequency in the treatment of sacroiliac joint pain. Method: Prospective, randomised and experimental study conducted on 60 patients selected in the two hospitals over a period of nine months, who had intense sacroiliac joint pain (Visual Analogue Scale [VAS] > 6) that Lasted more than 3 month. Patients were randomised into three groups (n = 20): Group A (two intra-articular sacroiliac injections of local anaesthetic/corticosteroid guided by ultrasound in 7 days). Group B: conventional bipolar radiofrequency "palisade". Target points were the lateral branch nerves of S1, S2. and S3, distance needles 1 cm. Group C: modified bipolar radiofrequency "palisade" (needle distance > 1cm). Patients were evaluated at one month, three months, and one year. Demographic data, VAS reduction, and side effects of the techniques were assessed. Results: One month after the treatment, pain reduction was > 50% in the three groups P < .001. Three and 12 months after the technique, the patients of the group A did not have a significant reduction in pain. At 3 months, almost 50% patients of the group B referred to improvement of the pain (P=.03), and < 25% at 12 months, and those results were statistically significant (P=.01) compared to the baseline. Group C showed an improvement of 50% at 3 and 12 months (P < .001). All patients completed the study. Conclusions: Bipolar radiofrequency "palisade", especially when the distance between the needles was increased, was more effective and lasted longer, compared to join block and steroids, in relieving pain sacroiliac joint. (C) 2015 Sociedad Espanola de Anestesiologia, Reanimacion y Terapeutica del Dolor. Published by Elsevier Espana, S.L.U. All rights reserved.