Methods: Seventy-three patients scheduled S1 TFESI randomly allocated into anteroposterior (group AP, n=36) or oblique (group O, n=37) approach. In group AP, cephalad-caudad tilt was used to maximize the opening of S1 foramen. In group O, the L5-S1 endplates were lined up by caudal tilting and the C-arm was rotated ipsilateral oblique to view the S1 as “Scotty dog”. We recorded foramen passage time, total procedure time, presence of intravascular injection, pain score and complications.

Results: Foramen passage time and total procedure time were shorter in group O than in group AP. (19.2 ± 5.7 s vs. 59.4 ± 65.2 s, P=0.007; 94.4 ± 49.1 s vs. 168.1 ± 99.6 s, P=0.004, respectively, Table 1). And the incidence of overall intravascular injection was higher in group AP (27.8% vs. 5.4%, P=0.014). However, there were no significant differences in pain score (Table 2), paresthesia during procedure, and complication rates between the two groups.

Conclusions: The oblique “Scotty dog” approach improves the visualization of S1 foramen and reduces the incidence of intravascular injection during TFESI. Therefore, this approach allows both patients and practitioners to reduce radiation exposure.