116 Short Wave Infrared Light Imaging Distinguishes Superficial from Deep Burns

Introduction Burns represent a significant public health burden that rely on accurate diagnosis of burn depth. The gold standard for burn injury depth remains visual clinician appraisal. Candidate modalities for objective and standardized diagnostics have failed to reach wide adoption due to limitations in practicality, cost, and reproducibility. Here we utilize short wave infrared spectroscopy (SWIR) which involves analysis of a subset of the infrared spectrum, between 1000–2500 nm, and has previously been described in the assessment of water composition in human skin. Given literature highlighting moisture as a potential biomarker for tissue viability, we hypothesized a similar application in a novel, healthcare context for assessing tissue damage following contact burn injury. Methods A custom imaging system was constructed with tungsten-filament light source, filters centered at 1200, 1650, 1940, and 2250 nm, and a SWIR camera. Phantom models of agar slabs were constructed at 1, 2, 3, 5, and 10 mm thickness and spectrally interrogated in epi-illumination and trans-illumination configurations (Fig A, top). For verification of moisture detection, filter paper was wetted with 50 µL H2O and imaged. For burns study, Yorkshire pigs (N=5) were injured with both short (20 s) and long (40 s) exposure burns to assess varying degrees of burn injury for SWIR and biopsy histology (H&E, picrosirius). In vivo imaging occurred 72 hours after burn injury. For quantifications, a 50x50 pixel region was selected from the center of each burn site to determine average reflectance. Results Absorbance of SWIR light was found to depend on agar thickness, increasing with depth and moisture in substrates. 1940 nm absorbance was sensitive with full absorbance even in agar slabs as thin as 1 mm (Fig A, bottom). Water directly applied to paper demonstrated sharp demarcation of dry vs. wet regions consistent with absorptive patterns seen in agar (Fig B). Short (S) and long (L) regions of porcine burns were imaged (Fig C-D). Two-way ANOVA analysis across 5 animals revealed that reflected fraction was different for each wavelength (p< 0.001) and for both burn exposure times (p=0.011). Imaged wavelengths of 1200, 1650 and 2250, but not 1940 nm, showed a significant 10–20% higher reflectance in the long duration burn in comparison to the short or unburned regions (Fig E). Analysis of biopsies in each treatment region confirmed deeper damage with longer injury times (long: 922±45 vs short: 590±57 mm, n=10/group, p< 0.001, Fig F). Conclusions There exists a need in burn care for a reliable, objective measure of tissue viability following burn injury. SWIR spectroscopy can delineate moisture content in phantom models and distinguish burn depth in porcine models. Applicability of Research to Practice This work to be particularly timely given the need and demand for more objective diagnostic modalities in contemporary burn depth assessment.