Forty-three patients were included and 40 analyzed. Patients were eligible if they had leg or arm traumatic pain of moderate intensity. Pain intensity was measured using a numerical scale (0 - 10) at regular times for 120 minutes and the main endpoint was at 30 minutes. The hypothesis of non-inferiority was formulated from previous works with healthy volunteers. After pain assessment, patients received at baseline 1 g-iv-APAP or saline and concomitantly, 125 mg APAP in 1 mL hydroalcoholic solution (HAS) or placebo (HAS only) was applied in the left mucogingival sulcus. Non-inferiority of the primary outcome was assessed by one-sided 2 group t-test of equivalence in means with equal variances with a non-inferiority limit difference of 1. Other tests were two-sided, with a type I error set at α = 0.05