Background/Purpose: Microvascular abnormalities, leading to hypoxia, drive the SSc disease process. Accurate measurement of skin oxygenation, both baseline and during ‘stress’/reperfusion (which can be mimicked in the laboratory by occlusion), would not only facilitate studies of pathogenesis but also provide a surrogate outcome measure for clinical trials of vasoactive therapies. But measuring skin oxygen is challenging, especially in darker skin tones. Our aim was to (a) explore the effect of skin tone on skin oxygenation measurements using multispectral imaging (MSI) (which measures oxygen by imaging the skin under white light) and (b) compare skin oxygenation between patients with SSc and healthy controls (HC).

Methods: Fifty patients with SSc (49 fulfilled ACR criteria or were VEDOSS; median age 55.0 (IQR 45.50-63.75) years; F:M 45:5) and 50 healthy controls (55.5, (44.75-64.00) years; F:M 45:5) were recruited. Objective measurements of participants’ ‘individual typology angle’ (ITA) were taken using a colorimeter, to record skin tone. ITA data from recruited participants are given in Table 1. Participants underwent an occlusion of the middle finger on their non-dominant hand to reduce oxygenation and their hands were imaged with MSI before, during, and after occlusion release. From the MSI data an index of oxygenation (Oxy) was calculated. The Oxy index was averaged over a region of interest on the occluded finger. A multiple linear regression model was used to estimate the impact of skin tone and cohort on the change in Oxy (ΔOxy) upon occlusion release.

Results: Median and IQR of Oxy over time is shown in Figure 1. The fitted model of ΔOxy, shown in Figure 2, was significant, F(2,97) = 45.66, p < 0.001, and showed that both cohort, t(97) = -5.596, p < 0.001, and ITA, t(97) = 7.041, p < 0.001, had a significant effect on ΔOxy. ΔOxy was reduced for lower ITA, i.e. darker skin tone, and for patients compared with HCs.

Conclusion: (a) Skin tone had a significant effect on Oxy measurements, with lower ΔOxy in darker skin tones. We believe this is due to a lack of sensitivity of the Oxy index to oxygenation in dark skin caused by strong absorption of light by eumelanin (which sits above the vascular network in the skin). This is a problem that is inherent to visible light imaging of the skin and needs to be addressed in MSI and similar optical techniques. This work represents a first step in understanding the effect melanin has on oxygenation measurements; future work is required to further understand and account for this effect and thus ensure equitable techniques for all patients.(b) ΔOxy was lower in patients, suggesting that patients display a muted response to the occlusion, with less pronounced reactive hyperaemia compared to HCs of a similar age and sex. This clear difference demonstrates the vascular dysfunction in patients, and Oxy could be a promising marker for disease measurement.

Table 1: The number of participants (healthy controls, patients, and total) recruited in each skin tone group by individual typology angle (ITA).

Figure 1: Oxy over time (before, during, and for 5 minutes after occlusion) colour-coded by skin tone group as given in Table 1, showing the median and IQR of each group for (a) healthy controls (HC) and (b) patients. ΔOxy is calculated as the difference in Oxy between release and occlusion.

Figure 2: The fitted multiple linear regression model showing the linear effect of individual typology angle (ITA), aka skin tone, on ΔOxy and the difference between cohorts. The red dashed line is the healthy control (HC) fit, while the blue solid line is the patient fit. The model accounts for 47% of variance in the data, with an R² of 0.47.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/assessing-the-impact-of-cutaneous-melanin-on-microvascular-oxygenation-measurement-in-systemic-sclerosis-using-multispectral-imaging/