Almeida, Erika O. [UNESP]Freitas Júnior, Amilcar C.Bonfante, Estevam A.Rocha, Eduardo Passos [UNESP]Silva, Nelson R. F. A.Coelho, Paulo G.2014-05-272014-05-272013-02-01Clinical Oral Implants Research, v. 24, n. 2, p. 191-196, 2013.0905-71611600-0501http://hdl.handle.net/11449/74507Objectives: This study evaluated the reliability and failure modes of implants with a microthreaded or smooth design at the crestal region, restored with screwed or cemented crowns. The postulated null hypothesis was that the presence of microthreads in the implant cervical region would not result in different reliability and strength to failure than smooth design, regardless of fixation method, when subjected to step-stress accelerated life-testing (SSALT) in water. Materials and methods: Eighty four dental implants (3.3 × 10 mm) were divided into four groups (n = 21) according to implant macrogeometric design at the crestal region and crown fixation method: Microthreads Screwed (MS); Smooth Screwed (SS); Microthreads Cemented (MC), and Smooth Cemented (SC). The abutments were torqued to the implants and standardized maxillary central incisor metallic crowns were cemented (MC, SC) or screwed (MS, SS) and subjected to SSALT in water. The probability of failure versus cycles (90% two-sided confidence intervals) was calculated and plotted using a power law relationship for damage accumulation. Reliability for a mission of 50,000 cycles at 150 N (90% 2-sided confidence intervals) was calculated. Differences between final failure loads during fatigue for each group were assessed by Kruskal-Wallis along with Benferroni's post hoc tests. Polarized-light and scanning electron microscopes were used for failure analyses. Results: The Beta (β) value (confidence interval range) derived from use level probability Weibull calculation of 1.30 (0.76-2.22), 1.17 (0.70-1.96), 1.12 (0.71-1.76), and 0.52 (0.30-0.89) for groups MC, SC, MS, and SS respectively, indicated that fatigue was an accelerating factor for all groups, except for SS. The calculated reliability was higher for SC (99%) compared to MC (87%). No difference was observed between screwed restorations (MS - 29%, SS - 43%). Failure involved abutment screw fracture for all groups. The cemented groups (MC, SC) presented more abutment and implant fractures. Significantly higher load to fracture values were observed for SC and MC relative to MS and SS (P < 0.001). Conclusion: Since reliability and strength to failure was higher for SC than for MC, our postulated null hypothesis was rejected. © 2012 John Wiley & Sons A/S.191-196engDental implantFractographyReliabilityStep-stress accelerated life-testingWeibullArtigo10.1111/j.1600-0501.2012.02437.xWOS:000313834500010Acesso restrito2-s2.0-848726677473383392287039820