Abstract Objectives Using two groups of mini-implants (successful and failed) the objectives of this in vivo study were: to evaluate the microbial contamination by the checkerboard DNA-DNA hybridization technique and to quantify the bacterial endotoxin by the limulus amebocyte lysate assay. Material and Methods The 15 successful and 10 failed mini-implants (1.6 mm diameter × 7.0 or 9.0 mm long), placed in the maxilla and/or mandible, were obtained from 15 patients undergoing orthodontic treatment. Data were analyzed statistically by the Wilcoxon rank-sum test using the SAS software (a=0.05). Results All 40 microbial species were detected in both groups of mini-implants, with different frequencies. No differences were observed between the groups with respect to microbial complexes (blue, purple, yellow, green, orange, red and other species) and endotoxin quantification (p>0.05). Conclusion Neither microbial contamination nor endotoxin quantification was determinant for the early loss of stability of the mini-implants.