OBJECTIVES: To investigate the effect of a percutaneous radiofrequency (RF) heat lesion compared with a sham procedure, applied to the lateral branches of L5, S1, S2, S3, and S4 nerve roots. MATERIALS AND METHODS: Sixty patients aged 18 years and above with a medical history and physical examination suggestive for sacroiliac joint pain and a reduction of 2 or more on a numerical rating scale (NRS, 0 to 10) after a sacroiliac joint test block were included in this study. Treatment group: percutaneous RF heat lesion at the lateral branches of S1, S2, S3, and S4 nerve roots and the posterior ramus dorsalis of L5; sham group: same procedure as the treatment group except for the RF heat lesion. PRIMARY OUTCOME MEASURE: pain reduction (NRS). Secondary outcome measure: Global Perceived Effect. RESULTS: No statistically significant differences in pain level over time between the groups (Group×Period) (F1,58=0.353; P=0.56) nor within the treatment Group (F1,58=0.212; P=0.65) were found. The Period factor, however, yielded a significant difference (F1,58=61.67; P<0.001), that is, when pooled together the mean pain level of the patients was significantly reduced at T1 compared with T0. In the crossover group, 42.1% experienced a reduction in NRS of 2 or more at 1 month (P=0.65). No statistically significant difference in satisfaction over time between the groups was found (F1,50=2.1; P=0.15). The independent factors Group (F1,50=2.02; P=0.16) and Period (F1,50=0.95; P=0.33) also showed no statistically significant difference. The same applies to recovery: no statistically significant Group×Period effect (F1,51=0.09; P=0.77) was found, neither an effect of Group (F1,51=0.004; P=0.95) nor of Period (F1,51=0.27; P=0.60). DISCUSSION: The hypothesis of no difference in pain reduction or in Global Perceived Effect between the treatment and sham group cannot be rejected. LEVEL OF EVIDENCE: Level 1A.