Dokument

Summary:
Endobronchial biopsies collected by fiberoptic bronchoscopy have been increasingly used in biomedical research on disease mechanisms and clinical therapy studies of chronic inflammatory airway disorders. Although less invasive techniques are available for the investigation of the inflammatory infiltrate of the bronchial tree, a standardization of their results with respect to the extent or level of the sampled airway proved impracticable. Moreover in a clinical setting the structural alterations of the airway mucosa can only be assessed by histopathological biopsy analysis, which makes this approach indispensable to airway research. More and more quantitative approaches in biopsy studies have been reported. The high variability of their results points out the need for reliable and robust quantitative methods and sampling designs in order to allow for an easier interpretation and corroboration of the outcomes of different studies. It is unclear whether classical 2D approaches and unbiased stereological 3D designs for counting inflammatory cells, measuring area fraction or layer thickness on histological sections are equally well suited for these purposes. The aim of this study was to characterise the agreement between 2D and 3D approaches for inflammatory cell counting by simultaneously applying them on bioptic material. Furthermore, stereological designs were proposed for quantifying the extent of epithelial desquamation and the mean thickness of the reticular basement membrane, and the results were related to previously published data gained by 2D tissue analyses. The hypotheses that the epithelial integrity depends on biopsy size or mean basement membrane thickness were also verified. Biopsies from the segmental bronchi were collected by fiberoptic bronchoscopy in a group of smokers (n=7) and a group of healthy non-smokers (n=7), embedded in paraffin and exhaustively sectioned. Systematic uniform random samples of sections were stained histochemically (PAS) or immunohistochemically for macrophages (CD68) and T-lymphocytes (CD3), respectively. On the same systematic uniform random samples of fields of view, cell numbers per unit volume were assessed using the physical disector and cell and nuclear profiles were counted and related to the subepithelial layer area. To obtain a zero-dimensional index allowing for a direct comparison of the two methods, the CD68+/CD3+ ratio was calculated for each approach. The extent of epithelial desquamation was assessed as area fraction of the basement membrane by counting the intersections of a line grid with the basement membrane on PAS stained sections. On the same sections the arithmetic mean thickness of the reticular basement membrane was estimated using a coherent test system of points and line segments. Counting cell profiles per unit area severely overestimated the number of larger cells (macrophages) relative to smaller cells (T-lymphocytes). Counting of nuclear profiles delivered average values similar to the physical disector but a bias proportional to the magnitude of the CD68+/CD3+ ratios was identified. The extent of epithelial desquamation was similar between the two groups and in accordance with previous studies in healthy volunteers and asthmatics. The lack of a difference between the (non-asthmatic) subjects of this study and published data on asthma patients confirms earlier similar findings. This strengthens the doubt about the morphopathological significance of the epithelial disruption, suggesting an artefactual cause. The arithmetic mean thickness of the reticular basement membrane, an important marker of airway remodelling in biopsy studies of asthma, showed no significant difference between healthy non-smokers and smokers in the small studied groups. The average values were very similar to the results of another published stereological design and to those obtained by image analysis of perpendicular sections. At the same time they were conspicuously lower than the data reported by studies employing direct point-to-point measurements on sections. This underlines the overestimation of the mean thickness introduced by tangential cutting of the basement membrane when relying on 2D measurements of this three-dimensional structure.

Das Dokument ist im Internet frei zugänglich - Hinweise zu den Nutzungsrechten