In vivo confocal Raman spectroscopy for intrinsic aging and photoaging assessment

Abstract

Background In vivo confocal Raman spectroscopy is a non-invasive method to assess either the epidermis or the dermis composition. Few studies have focused on dermis collagen alterations through intrinsic aging and photoaging. Objective This study evaluated the in vivo Raman spectra from the dermis of a photoexposed site versus a non-photoexposed region in different age groups, and evaluated the correlation between peak intensities and age, photoaging score and the amount of collagen assessed with histology and high frequency ultrasound (HFUS). Methods Fifteen volunteers aged 28–82 years were divided into three groups according to forearm photoaging degree. In vivo Raman spectra from the dermis were collected on the dorsal forearm (chronically photoexposed skin) and on the proximal medial arm (non-photoexposed skin). Cross-sectional images of the skin were obtained using a 20 MHz ultrasound unit exactly on the same sites, which were further submitted to punch biopsies for histologic study (collagen I immunohistochemistry, picrosirius red staining and Verhoeff). Principal Component Analysis (PCA) and Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) were taken in the spectral region of 796 cm−1–996 cm−1 to determine its potential to discriminate between different groups. The Spearman rank correlation coefficient of individual peak intensities and ratios with age, clinical score and the amount of collagen assessed by ultrasound and histology were calculated. Results PCA of pairs of groups and OPLS-DA could discriminate the intrinsically from the photoaged skin and the young group from the elderly one, with important contribution of the 938 cm−1 and 855 cm−1 peaks intensities. The intensity of the peaks in 855 cm−1 and/or 938 cm−1 presented moderate correlation with age (rho = 0.579, p = 0.049) and moderate to high inverse correlation with HFUS echogenicity (rho = −0.710, p = 0.010) and collagen I immunohistochemistry (rho = −0.833, p = 0.005) in the non-photoexposed region. The I1275/I1450 intensities ratio presented moderate to high correlation coefficients with age (rho = −0.730, p = 0.007), photoaging score (rho = −0.594, p = 0.042), HFUS echogenicity (rho = 0.760, p < 0.001) and histology (collagen I immunohistochemistry (rho = 0.643, p = 0.024), picrosirius (rho = 0.773, p = 0.005) and Verhoeff (rho = −0.727, p = 0.011)) in the photoexposed site. Conclusion The wavenumber region between 798 and 994 cm−1 is useful for the analysis of dermal collagen alterations through the intrinsic aging process, while photoaging is better assessed by the I1275/I1450 intensities ratio. This is the first skin aging study to show a correlation between Raman peaks and the amount of collagen assessed by HFUS and histology.