The influence of smoking on 3-year clinical success of osseointegrated dental implants.

Abstract

Health risks associated with smoking have been exhaustively documented and include increased incidence of periodontal disease, greater risk of osteitis following oral surgery, and compromised wound healing due to hypoxia. Information related directly to dental implants, although limited, points to higher rates of implant failures among smokers than non-smokers. This paper reports on long-term clinical outcomes of osseointegrated dental implants placed in smokers and non-smokers in a longitudinal clinical study of endosseous dental implants. In 1990, the Dental Implant Clinical Research Group (DICRG) of the Department of Veterans Affairs (DVA) launched an 8-year, randomized, prospective clinical study of more than 2,900 endosseous dental implants in more than 800 patients at 32 study centers. Confounding variables, including smoking patterns, were recorded. For this report, new follow-up data were analyzed for two groups: 1) current smokers and 2) those who never smoked combined with those who quit. Most of the variables recorded for each implant were screened on a univariate basis as possible predictors associated with implant survival/failure. Those with P values less than 0.15 and those likely to be a factor of clinical importance were placed in a logistic regression equation and analyzed for a simultaneous effect on survival. A step-wise procedure was used to eliminate those variables that showed the least significance, until only those variables with a Wald chi-square of significance in the presence of others remained. The effects of clustering within patients and of unbalanced distribution within hospitals were standardized to facilitate analysis of influence of demographic variables. The GEE analysis was performed with the patient as the primary cluster. Current data do not support earlier findings that smoking contributes to early implant failure (placement to uncovering). A trend of greater failures in smokers appeared between the time after uncovering and before insertion of the prosthesis. Hydroxyapatite (HA)-coated implants had significantly lower failure rates. For the entire 3-year period, overall failures were significantly higher for smokers than non-smokers. Results suggest that increased implant failures in smokers are not the result of poor healing or osseointegration, but of exposure of peri-implant tissues to tobacco smoke. Data also suggest that detrimental effects may be reduced by: 1) cessation of smoking; 2) the use of preoperative antibiotics; and 3) the use of HA-coated implants.