Outcomes and their measurement in clinical trials of endosseous oral implants.

Abstract

The clinical suitability of an oral implant system should be substantiated by well-planned clinical trials that meet the expectations of both the patient and the clinician; i.e., to offer anchorage to a dental prosthesis for one or more decades. The literature often reports on the survival rather than on the success rate, and for the latter a large variety of more or less stringent criteria have been proposed. Although the use of cumulative percentages to assess survival or success on consecutive patients, the so-called Kaplan-Meier statistics, is essential, many publications only report on absolute percentages, thus giving much too optimistic results. Indeed, the large number of recently inserted implants tend to dilute the small fraction that has gone through the maximum observation period. Multi-center studies on consecutive patients who are not selected on the basis of bone quality and volume should be recommended. Otherwise the labeling of the implant should clearly mention which groups; e.g., smokers or patients with limited bone volume are not part of the validated target group. The final outcome measurement would be that an endosseous implant is no longer able to carry out its purpose: to anchor or support successfully a functional dental prosthesis. Like many other clinical evaluations, surrogate parameters have been sought to assess interim success rates. The stability of the marginal bone level is one parameter proposed by many, although bone loss as such is not synonymous with failure. Several implant systems with a roughened surface (plasma sprayed or coated) suffer from progressive marginal bone loss and consequently ongoing loss of implants even after 5 or more years. Those implant systems should be evaluated with cumulative success rates over a period of 10 years. Only those that demonstrated a clustering of both bone and implant loss during the first 1 or 2 years after insertion and subsequently a levelling off for those two parameters-a so-called plateau pattern, should be allowed to draw conclusions after 5 years. Since radiographs reveal only the approximal areas, measurement of the attachment level by means of probing towards a reference point on the abutment should also be performed annually. Intra-oral radiographs can only reveal absence of radiolucency at the bone-implant interface, but do not prove osseointegration. The use of an electronic device offers objective results and reveals even subclinical mobilities of endosseous implants. One can conclude that there is an urgent need for a consensus between health authorities, third parties, and the scientific community to define a set of universally-acceptable success criteria, to impose the use of cumulative success rates, and to ask for observation periods of 10 years for implants that do not demonstrate stable results after one or two years.