Comparison of ground sections, paraffin sections and micro-CT imaging of bone from the epiphysis of the porcine femur for morphometric evaluation

Abstract

The aim of this study was to compare data on the volume fraction of bone and the thickness of the cortical compact bone acquired during microcomputed tomography (micro-CT) analysis with data acquired from identical samples using stereological analysis of either decalcified paraffin sections or ground sections. Additionally, we aimed to compare adjacent tissue samples taken from the major trochanter of the porcine femur to map the basic biological variability of trabecular bone.

Fifteen pairs of adjacent tissue blocks were removed from the major trochanter of the proximal epiphyses of porcine femurs (female pigs aged 24–39 months, weight = 59.16 ± 8.15 kg). In each sample, the volume of the cortical compact bone, the volume of the trabecular bone, and the thickness of the cortical compact bone was assessed using micro-CT. Afterwards, half of the samples were decalcified and processed using paraffin histological sections. Another half was processed into ground sections. The volume and thickness of bone was assessed in histological sections using stereological techniques.

There were no significant differences in the bone volumes and thicknesses measured by micro-CT and the corresponding values quantified in decalcified sections. Similarly, there were no differences between the results from micro-CT and the analysis of the corresponding ground sections.

Histomorphometric studies based on relatively low numbers of undecalcified ground sections or demineralized paraffin sections of bone yield data on bone volume and the thickness of cortical compact bone that is comparable with three-dimensional micro-CT examination. The pilot data on the variability of cortical compact bone and trabecular bone volumes in the porcine major trochanter provided in this study aim for planning experiments in the field of bone healing and implantology.

Introduction

Histomorphometry is one of the most commonly used methods to quantify bone, and histomorphometry also enables a qualitative evaluation of tissue responses and remodeling. In general, there are three approaches available for processing and morphometry of bone samples: (i) histological sectioning of demineralized bone samples, (ii) grinding sections processed from bone samples without prior demineralization, and (iii) microcomputed tomography (X-ray microtomography, micro-CT) analysis of bone samples without sectioning (cf. Bancroft and Gamble, 2008, Mulisch and Welsch, 2015, Liu et al., 2016). Each of these techniques has its pros and cons. For example, decalcified sections provide an easy and inexpensive approach unless the evaluation of the integration of titan implants is needed (Gredes et al., 2015). Cells, non-mineralized osteoid and soft tissue can be differentiated by means of staining or immunohistochemistry. Undecalcified hard tissue ground sections embedded in resin (Mohammadi et al., 2000) or methacrylate (Botzenhart et al., 2015, Kunert-Keil et al., 2015) preserve the mineral bone and are methods of choice whenever metallic (e.g., titan) implants are part of the sample (Bissinger et al., 2017). The major disadvantage of both demineralized and undemineralized histological processing is that the methods rely on two-dimensional sections, while the third dimension has to be evaluated from a number of sections.

Compared to histology, microcomputed tomography provides an intrinsically three-dimensional approach without destroying the tissue sample. Therefore, it is an important method for the quantification of bone tissue in the study of bone functional adaptation (Hoechel et al., 2015) and bone metabolic disease (Wen et al., 2015), in experiments researching the possible use of tissue scaffolds (Hadzik et al., 2016) and stem cells to affect bone defects and in the implantology field (Friedmann et al., 2014). Micro-CT is a technique that enables visualization and quantification of three-dimensional structures, such as cortical compact bone and trabecular bone. Other advantages of micro-CT are its non-destructiveness, reduced need for sample preparation and, therefore, a shorter time for evaluation. The major disadvantages of micro-CT are a lower resolution than optical microscopy and image artifacts occurring from the presence of computed tomography artifacts (Boas and Fleischmann, 2012).

In the present implantology research, all three methods (demineralized sections, ground sections, micro-CT) are often combined when evaluating osseointegration and the healing of bone implants, both in human patients or in experimental animal models. The most common bones for implantation are the maxilla (Aparicio et al., 2011, Bissinger et al., 2017, Botzenhart et al., 2015, Saulacic et al., 2014, Stadlinger et al., 2012) and femur in pigs (Buser et al., 1991, Endres et al., 2005, Howashi et al., 2016, Isoda et al., 2012, Kulkova et al., 2014, Schwarz et al., 2009). Choosing a suitable animal model that mimics human bone is vital for the validity of experiments that study osseointegration of implants, because of different local ratios between the compact and trabecular bone (Babuska et al., 2016). Different proximal vs. distal segments of long bones, such as the femur, tibia or humerus may vary in the proportions between the compact and trabecular type of bone tissue. In contrast with a more compact mandible (Tonar et al., 2011), the maxilla has a substantially greater proportion of very thick trabecular bone. Localization of implantation during in vivo testing of implants should correspond with the intended place of use in human patients. Therefore, a reliable way to assess the ratio between the compact and spongy bone is necessary.

In our study, we focused on the epiphysis of the porcine femur, which is an important animal model used in implantology and bone healing research (Pithon et al., 2013, Kulkova et al., 2014). Recently, several papers found good agreement between morphometric data based on undecalcified ground histological sections and micro-CT, namely Gabler et al. (2015) in rat tibia, Liu et al. (2016) in rat mandible and tibia, and Bissinger et al. (2017) in maxilla of minipigs. However, to our best knowledge, we did not find a direct comparison of histomorphometric measurements of bone tissue using decalcified sections, ground sections, and micro-CT at the same time in large animal models. Therefore, the aim of the study was to compare data on the volume fraction of bone and thickness of the cortical compact bone acquired during micro-CT analysis with data acquired from identical samples using stereological analysis of either decalcified paraffin sections or ground sections. Additionally, we aimed to compare adjacent tissue samples from the major trochanter of porcine femur to map the basic biological variability of trabecular bone.

The following null hypothesis was formulated and tested:

H₀

There is no difference when comparing data on the volume and cortical bone thickness using micro-CT vs. stereological analysis of identical bone samples processed as decalcified paraffin sections or ground sections.