Collagen bundle morphometry in skin and scar tissue: a novel distance mapping method provides superior measurements compared to Fourier analysis

• Published in: Journal of microscopy (Oxford) 2012-01, Vol.245 (1), p.82-89
• Main Authors: VERHAEGEN, PAULINE D.H.M., VAN MARLE, JAN, KUEHNE, ANDREAS, SCHOUTEN, HENNIE J., GAFFNEY, EAMONN A., MAINI, PHILIP K., MIDDELKOOP, ESTHER, VAN ZUIJLEN, PAUL P.M.
• Format: Article
• Language: English
• Subjects: Biopsy; Bundle spacing; bundle thickness; Cicatrix; collagen; Collagen - analysis; Collagen - ultrastructure; fibrosis; Fourier Analysis; Histocytochemistry - methods; Humans; Microscopy, Confocal; morphometry; Observer Variation; Pathology - methods; Reproducibility of Results; scar; skin; Skin - chemistry; Skin - ultrastructure
• ISSN: 0022-2720; 1365-2818
• DOI: 10.1111/j.1365-2818.2011.03547.x

Summary Histopathological evaluations of fibrotic processes require the characterization of collagen morphology in terms of geometrical features such as bundle orientation thickness and spacing. However, there are currently no reliable and valid techniques of measuring bundle thickness and spacing. Hence, two objective methods quantifying the collagen bundle thickness and spacing were tested for their reliability and validity: Fourier first‐order maximum analysis and Distance Mapping, with the latter constituting a newly developed morphometric technique. Histological slides were constructed and imaged from 50 scar and 50 healthy human skin biopsies and subsequently analyzed by two observers to determine the interobserver reliability via the intraclass correlation coefficient. An intraclass correlation coefficient larger than 0.7 is considered as representing good reliability. The interobserver reliability for the Fourier first‐order maximum and for the Distance Mapping algorithms, respectively, showed an intraclass correlation coefficient above 0.72 and 0.89. Additionally, we performed an assessment of validity in the form of responsiveness, in particular, demonstrating medium to excellent results via a calculation of the effect size, highlighting that both methods are sensitive enough to measure a treatment effect in clinical practice. In summary, two reliable and valid measurement methods were demonstrated for collagen bundle morphometry for the first time. Due to its superior reliability and more useful measures (bundle thickness and bundle spacing), Distance Mapping emerges as the preferred and more practical method. Nevertheless, in the future, both methods can be used for reliable and valid collagen morphometry of skin and scars, whereas further applications evaluating the quantitative microscopy of other fibrotic processes are anticipated.