Background Dental implants are available in different materials, shapes and with different surface characteristics. In particular, numerous implant designs and surface modifications have been developed for improving clinical outcome. This is an update of a Cochrane review first published in 2002, and previously updated in 2003, 2005 and 2007. Objectives Primary: to compare the clinical effects of different root-formed osseointegrated dental implant types for replacing missing teeth for the following specific comparisons: implants with different surface preparations, but having similar shape and material; implants with different shapes, but having similar surface preparation and material; implants made of different materials, but having similar surface preparation and shape; different implant types differing in surface preparation, shape, material or a combination of these. Secondary: to compare turned and roughened dental implants for occurrence of early implant failure (before prosthetic loading) and occurrence of peri-implantitis. Search methods We searched the following electronic databases: the Cochrane Oral Health Group's Trials Register (to 17 January 2014), the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2013, Issue 12), MEDLINE via OVID (1946 to 17 January 2014) and EMBASE via OVID (1980 to 17 January 2014). We placed no restrictions on the language or date of publication when searching the electronic databases. Selection criteria We included any randomised controlled trial (RCT) comparing osseointegrated dental implants of different materials, shapes and surface properties having a follow-up in function of at least one year. Outcome measures were success of the implants, radiographic peri-implant marginal bone levels changes and incidence of peri-implantitis. Data collection and analysis At least two review authors independently conducted screening, risk of bias assessment and data extraction of eligible trials in duplicate. We expressed results using fixed-effect models (if up to three studies were present in a meta-analysis) or random-effects models (when there were more than three studies) using mean differences (MD) for continuous outcomes and risk ratios (RR) for dichotomous outcomes with 95% confidence intervals (CI). We reported the following endpoints: one, three, five and 10 years after functional loading. Main results We identified 81 different RCTs. We included 27 of these RCTs, reporting results from 1512 participants and 3230 implants in the review. We compared 38 different implant types with a follow-up ranging from one to 10 years. All implants weremade of commercially pure titanium or its alloys, and had different shapes and surface preparations. We judged two trials to be at low risk of bias, 10 to be at unclear risk of bias and 15 to be at high risk of bias. On a 'per participant' rather than 'per implant' basis, we found no significant differences between various implant types for implant failures. The only observed statistically significant difference for the primary objective regarded more peri-implant bone loss at Nobel Speedy Groovy implants when compared with NobelActive implants (MD -0.59 mm; 95% CI -0.74 to -0.44, different implant shapes). The only observed statistically significant difference for the secondary objective was that implants with turned (smoother) surfaces had a 20% reduction in risk to be affected by peri-implantitis than implants with rough surfaces three years after loading (RR 0.80; 95% CI 0.67 to 0.96). There was a tendency for implants with turned surfaces to fail early more often than implants with roughened surfaces. Authors' conclusions Based on the results of the included RCTs, we found no evidence showing that any particular type of dental implant had superior long-term success. There was limited evidence showing that implants with relatively smooth (turned) surfaces were less prone to lose bone due to chronic infection (peri-implantitis) than implants with much rougher surfaces (titanium-plasma-sprayed). These findings were based on several RCTs, often at high risk of bias, with few participants and relatively short follow-up periods.
