Purpose: This study sought to evaluate the influence of methodologic aspects on variations in the findings of in vitro microleakage studies of the implant-abutment interface. Materials and Methods: The MEDLINE, EMBASE, and Cochrane Library databases were consulted for in vitro studies published between 1990 and August 2011. Date from the studies that met the inclusion and exclusion criteria were arranged in tables and subjected to descriptive analysis. Results: Twenty-one studies were found to be eligible for the analysis after application of the inclusion/exclusion criteria. Sixteen studies used bacteria (76.2%), one used a bacterial toxin (4.76%), one used saliva (4.76%), two employed dyes (9.52%), and one used a combination of dyes and bacteria (4.76%). Eight studies evaluated microleakage from the inner portion of the implant to the external portion (38.1%) and nine examined the reverse (42.85%), while four studies investigated the relationship between them (19.05%). The volume inoculated inside the implants ranged from 0.1 to 5.0 mL. The bacterial concentrations used in the tests ranged from 2.41 × 106 to 8 × 108 colony-forming units/mL. Oral bacterial flora; mixtures of bacteria, toluidine blue, and gentian violet; and lipopolysaccharide of Salmonella enterica bacterial toxins were used. The monitoring period of test results ranged from 24 hours to 11 weeks for bacteria, 5 minutes to 7 days for dye, and 7 days for bacterial toxins. In four studies, microleakage was correlated with the size of the implant-abutment microgap. The external-hexagon implant configuration showed the greatest microleakage, followed by internal-trilobe, internal-hexagon, and internal-taper configurations. Conclusion: The lack of standardization hinderd comparisons of the studies and could explain the divergent results. It is suggested for future studies that special emphasis be placed upon inoculation and analysis of the specific volume for each system, lower concentrations of inoculated bacterial suspensions, and shorter follow-up time when using bacteria. Schlagwörter: implant-abutment interface, literature review, microleakage