Principal component analysis: a novel analysis to evaluate the characteristics of osseointegration of different implant surfaces.

ABSTRACT

OBJECTIVE: To apply a new statistical method (principle component analysis; PCA) to evaluate osseointegration. MATERIALS AND METHODS: Two different commercially available implants were selected for the study. Twenty implants, 10 of each type, were placed in the rabbit tibiae (n = 10). The fluorochromes (FLCs) alizarin complexone and calcein green were administered after 20 days and 4 days before sacrifice for labeling. On the day of implantation and retrieval (6 weeks), implant stability was measured with a resonance frequency analyzer (RFA). The retrieved samples were ground sectioned for histomorphometric and FLC quantification. The collected data were analyzed by a PCA software program (Qlucore Omics Explorer, Lund, Sweden) to explore and determine the correlation between different study variables and to analyze the differences between different implants. RESULTS: The RFA presented no significant differences at either time point. The bone-to-implant contact was significantly higher for the TiUnite (NobelBiocare, Gothenburg, Sweden); however, the bone area and FLC quantification showed higher values for the Osseotite (3i Implant Innovation, FL). Consistent with these results, the PCA indicated a strong correlation between TiUnite and high bone-to-implant contact values and between Osseotite and high bone area and FLC values. No correlation between RFA and the biological responses were found. CONCLUSION: The application of the PCA analysis may help interpret and correlate results obtained from numerous evaluations.