Comparison of three-dimensional digital analyses and two-dimensional histomorphometric analyses of the bone-implant interface

Abstract

Histological analysis is considered to be the gold standard method of evaluating osseointegration around a bone-implant. However, this method requires invasive specimen preparation and is capable of representing only one plane. By comparison, micro-computed tomography (μCT) is a fast and convenient method that offers three-dimensional information but is hampered by problems related to resolution and artifacts, making it a supplementary method for osseointegration analysis. To verify the reliability of μCT for osseointegration evaluation, this animal model study compared bone-to-implant contact (BIC) ratios obtained by the gold standard histomorphometric method with those obtained by the μCT method, using a rabbit tibia implant model. A sandblasted, large-grit, acid-etched (SLA) implant and a machined surface implant were inserted into each tibia of two rabbits (giving eight implants in total). Bone-implant specimens were analyzed using μCT with a spiral scan technique (SkyScan 1275) and histological sections were prepared thereafter. Three-dimensional (3D) reconstructed μCT data and four two-dimensional (2D) μCT sections, including one section corresponding to the histologic section and three additional sections rotated 45°, 90°, and 135°, were used to calculate the BIC ratio. The Pearson's test was used for correlation analysis at a significance level of 0.05. The histomorphometric BIC and the 2D-μCT BIC showed strong correlation (r = 0.762, P = 0.046), whereas the histomorphometric BIC and 3D-μCT BIC did not (r = -0.375, P = 0.385). However, the mean BIC value of three or four 2D-μCT sections showed a strong correlation with the 3D-μCT BIC (three sections: r = 0.781, P = 0.038; four sections: r = 0.804, P = 0.029). The results of this animal model study indicate that μCT can be used to complement the histomorphometric method in bone-implant interface analyses. With the limitations of this study, 3D-μCT analysis may even have a superior aspect in that it eliminates random variables that arise as a consequence of the selected cutting direction.