Purpose: The purpose of this study was to assess tobacco use as a risk factor for early implant failure. Materials and Methods: To address the research aim, a retrospective study design was used, and a study sample was derived from among patients who had 1 or more implants inserted between July 1996 and July 2004. The main predictor variable was the frequency of tobacco use. The major outcome variable was early implant failure. Appropriate descriptive, bivariate, and multivariate statistics were computed. Results: The study sample was composed of 650 patients who had 1,628 implants inserted and were selected for analysis. The early implant loss rates found in the nonsmoking group and the smoking group were 3.32% and 2.81%, respectively. Univariate and bivariate analyses showed no statistical significance for early implant losses associated with the frequency of tobacco use. Conclusions: The results of this study suggest that tobacco use alone cannot be considered as a factor for risk related to early implant failures. Prospective studies are needed to assess the risk of early implant failure in conjunction with smoking. (C) 2008 American Association of Oral and Maxillofacial Surgeons.66510041009