This study aimed to evaluate the effectiveness of low intensity laser therapy (LILT) in 30 patients presenting temporomandibular joint (TMJ) pain and mandibular dysfunction in a random and double-blind research design. The sample, divided into experimental group (1) and placebo group (2), was submitted to the treatment with infrared laser (780 nm, 30 mW, 10 s, 6.3 J/cm2) at three TMJ points. The treatment was evaluated throughout six sessions and 15, 30 and 60 days after the end of the therapy, through visual analogue scale (VAS), range of mandibular movements and TMJ pressure pain threshold. The results showed a reduction in VAS (p < 0.001) and through the ANOVA with repeated measures it was observed that the groups did not present statistically significant differences (P = 0.2060), as the averages of the evaluation times (P = 0.3955) and the interaction groups evaluation times (P = 0.3024), considering the MVO. The same occurred for RLE (P = 0.2988, P = 0.1762 and P = 0.7970), LLE (P = 0.3265, P = 0.4143 and P = 0.0696), PPTD (P = 0.1558, P = 0.4695 and P = 0.0737) and PPTE (P = 0.2376, P = 0.3203 and P = 0.0624). For PE, there were not statistically significant differences for groups (P = 0.7017) and the interaction groups evaluation times (P = 0.6678), even so in both groups the PE varied with time (P = 0.0069). © 2005 Blackwell Publishing Ltd.