Quantification of bone tissue regeneration employing β-tricalcium phosphate by three-dimensional non-invasive synchrotron micro-tomography - A comparative examination with histomorphometry

Purpose: This methodical Study presents a novel approach to evaluate the validity of two-dimensional histomorphometric measurements of a bone biopsy specimen after sinus floor elevation by means of high contrast, high resolution, three-dimensional and non destructive synchrotron micro-tomography (SCT). The aim of this methodical description is to demonstrate the potential of this new approach for the evaluation of bone biopsy samples. Materials and methods: Unilateral sinus grafting was carried out exemplarily in two patients using a combination of beta-tricalcium phosphate (beta-TCP) and autogenous bone chips. For the first patient a beta-TCP with 35% porosity and in the second with 60% porosity was used. At implant placement, 6 months after sinus grafting, a cylindrical specimen was biopsied from the augmented area. Subsequent to the histological embedding in resin the specimens were imaged using a SCT facility resulting in three-dimensional (3-D) images with approximately 4 mu m spatial resolution (1.5 mu m pixel size) for each patient's specimen. Subsequent to the SCT acquisition, tissue sections were prepared for histomorphometric analysis. Results: Bone area fractions determined by two-dimensional (2-D)) quantitative histomorphometry and by analysis of the corresponding 2-D slice from the SCT volume data were similar. For the first biopsy specimen (beta-TCP with 35% porosity), the bone area fractions were 53.3% and 54.9% as derived by histomorphometry and by analyzing a SCT slice, respectively. For the second biopsy, specimen (beta-TCP with 60% porosity) the bone area fractions were 38.8% and 39% respectively. Although tilt agreement between the 2 D methods was excellent, the area fractions were somewhat higher than the volume fractions computed by 3-D image analysis On the entire SCT Volume data set. The Volume fractions were 48.8% (first biopsy specimen) and 36.3% (second biopsy specimen). Conclusion: Although the agreement between the 2-D methods is excellent in terms of computing the area fractions, the structural 3-D insight which can be derived from classical 2-D methods, including histomorphometric analysis is considerably limited. This fact is emphasized by the discrepancy between the measured areas and Volume fractions. (C) 2008 Elsevier Inc. All rights reserved.