IMPROVED SPECIFICITY OF MATRIX ASSESSMENT OF NORMAL AND REPAIR CARTILAGE USING INFRARED IMAGING ANALYSIS AND MULTIVARIATE METHODS

Abstract

Articular cartilage is a connective tissue that lines the long bones and provides a near frictionless load-bearing surface. Cartilage degeneration, which is associated with chemical and structural changes in its extracellular components including water, type II collagen and proteoglycans, can progress to severe forms of osteoarthritis. There are several approaches for repair of cartilage, and quality of the tissue formed correlates to the success rate of the clinical procedure. Conventionally, histology and immunohistochemistry have been used to evaluate tissue characteristics, but they are expensive, time consuming, and require tissue harvesting. Therefore, development of a method that can assess cartilage tissue molecular structure, and could ultimately be used for in vivo analyses, is of great interest. Fourier transform infrared (FT-IR) spectroscopy is a vibrational spectroscopic technique that is sensitive to tissue chemical and structural properties. The studies described in this thesis aimed to improve the FT-IR spectroscopy applications in cartilage tissue assessment in the following areas: 1. Validation of FT-IR derived parameters developed in previous studies for tissues sectioned on the newly developed low emissivity infrared slides. 2. Evaluation of the clinical outcome of the autologous chondrocyte implantation (ACI) cartilage repair process using histology, immunohistochemistry and FT-IR parameters. 3. Development of an in situ infrared data collection method using infrared fiber optic probe (IFOP) spectroscopy to assess quality of degenerative cartilage. 4. Development of multivariate data analysis models to predict chemical changes in repair and normal tissue and to map the distribution of biochemical components in cartilage. 5. Assessment of the sensitivity of FT-IR parameters for evaluation of collagen crosslinks compared with biochemical methods, and 6. Assessment of cartilage water content using the mid infrared region of the spectra. Infrared spectra were collected from normal, degenerative and repair cartilage using different modalities including FT-IR analysis of bulk tissue samples, infrared fiber optic probe spectroscopy, and FT-IR imaging spectroscopy (FT-IRIS). Histology and immunohistochemistry were used as standards for comparison. Chemical and structural properties of matrix components including water, collagen and proteoglycan content, collagen fiber orientation, and collagen maturity were measured using univariate and multivariate analysis of infrared spectra. Results from the 6 studies showed: 1. FT-IR data collected in transflectance mode are significantly different from transmittance mode and these differences should be taken into account during analysis. 2. FT-IR derived molecular parameters, and histology and immunohistochemistry parameters, were correlated to clinical outcome of the cartilage repair process. 3. Quantitative data from multivariate analysis of infrared fiber optic probe spectroscopy evaluation of degenerated cartilage were correlated to histology grading. 4. A partial least squares model based on infrared spectra of pure components (collagen and proteoglycan) was able to predict tissue matrix components. 5. The FT-IR derived 1660/1690 cm-1 parameter previously correlated to collagen crosslinks was only sensitive to collagen maturity in cartilage and bone of different ages, but not to specific biochemically-determined crosslinks. 6. The mid-infrared absorbance centered at 2125cm-1 was sensitive to water distribution in cartilage and likely arises from bound water, based on correlation with near-infrared data and collagen content. In conclusion, the potential of FT-IR spectroscopy in assessment of chemical and structural properties of articular cartilage and other connective tissues was established. The results of these studies lay the groundwork for application of FT-IR spectroscopy in clinical and in situ applications.