Sensitivity of resonance frequency analysis for detecting early implant failure: a case-control study

The aim of this study was to test the sensitivity of the resonance frequency analysis for detecting early implant failure.

In all, 3,786 implants placed from June 2007 to January 2013 were retrospectively evaluated. A total of 20 implants (in 20 patients) placed in pristine bone were found to have failed before loading. The implant stability quotient (ISQ) values were extracted from these 20 implants at baseline (immediate) and 4 months after placement (delayed). Simple linear regression, logistic regression, and two-way contingency tables were used to test for the relationships between ISQ values and early implant failure.

Immediate ISQ values were significantly related to failure (odds ratio [OR] = 4.27). Furthermore, the results of the second regression showed a significant relationship between ISQ at delayed measurement and implant failure (OR = 9.20). For immediate ISQ, it seems that the 73.7% correct classifications were obtained at the cost of an incorrect classification of 55% of the implant failures. However, for the delayed ISQ, 86.2% correct classifications were obtained at the cost of assuming that all implants will survive.

The present study showed that ISQ values are not reliable in predicting early implant failure. In addition, the real cutoff ISQ value to differentiate between success and early implant failure remains to be determined.