Monitoring hypertrophic burn scars with a spectroscopic polarimetric optical system

Up to 67% of individuals surviving burn injury will develop hypertrophic scars,¹ which can cause serious body deformities, social reintegration difficulties, reduced physical function, and loss of ability in daily living (see Figure 1).^2,3 Unfortunately, scar formation is not fully understood, and the efficacy of interventional procedures, such as compression, laser treatment, and injected or topical pharmacotherapies (such as steroids) have shown inconclusive scientific and clinical results.^4,5

Figure 1. Examples of patients with hypertrophic scars (HTSs) from burn injuries. (A) Raised scar affecting ear structure. (B) Disfigurement of the palm of the hand by scar bands. (C) Limitation of movement at the anticubital fossa by a thick, dyschromic raised scar. (D) Contracted, dyschromic, webbed neck scarring limiting range of motion of the neck and mandible. (E) Compression/splinting apparatus used to treated lesions such as those in (D).

Assessment of hypertrophic scars is based on subjective clinician rankings using a four-parameter scale called the Vancouver scar scale (VSS) or the patient observer scar assessment scale (POSAS), but no objective, standardized tool for quantifying scar severity is available.^6–10 Non-invasiveness of the measurements is critical in this environment, for reasons of patient pain management and infection risks. Furthermore, payment for revisional surgeries and treatment of scars is subject to scrutiny due to lack of objective measurements documenting the degree of scar pathology and subsequent improvement after therapy.

We recently developed a spectroscopic polarimetric optical system (SPOS) that can monitor scar formation non-invasively.^11–14 We used a combined spatial frequency domain imager (SFDI) that is widely applied in medical settings, including burns,¹⁵ and this enabled assessment of tissue molecular components. We used SFDI to obtain images of scar oxygen saturation, water content, and blood volume fraction, and integrated the system with an out-of-plane Stokes polarimeter for the assessment of collagen orientation¹⁴ and rough surface.¹¹ Our group was the first to use this methodology macroscopically to study skin disease,¹⁶ and we have documented examples of rough surface models of normal skin for healthy, diseased,¹⁷ and scar tissues. The recovery of injured skin to its original roughness is desirable, as scars tend to be smooth and shiny, creating a strong contrast to uninjured skin. We assumed the polarimetric out-of-plane approach used in the characterization of scar directional growth by measuring global directional anisotropy.

We also developed a reproducible scar model and validated its similarity to a human burn scar, both grossly and histologically (see Figure 2).^11,18 We identified wound conditions to create hypertrophic scars (HTSs) in the red Duroc pig,¹³ and confirmed the similarity of key molecular and structural changes in skin during wound healing and scar maturation in humans.^12,13 We also developed a wearable automated pressure delivery system (APDS) (see Figure 2)¹⁸ to assess the effect of pressure on HTS development. Results showed that the system was capable of delivering a constant pressure of 30mmHg on scars for a period of two weeks without any adverse effects.

Figure 2. The automated pressure delivery system on the animal. (A) Surgical mounting of the device base on the animal flank. (B) Fixing the pressure delivery box in the designated place on the plate. (C) Covering the system with a custom-fitted neoprene protective vest. (D) The animal dressed during pressure therapy.

We conducted a six-animal study at the Firefighters' Burn and Surgical Research Laboratory, MedStar Health Research Institute, in Washington DC, to characterize our system's ability to monitor scar formation in vivo. We applied the APDS on the animal scars for a period of two weeks, using the SPOS in combination with a laser doppler imaging (LDI) system to monitor animals for 140 days. Experienced clinicians scored the scars using the VSS, and compared the results of the optical techniques with histologic and molecular examination of scar punch biopsies (those taken with a circular blade).

Figure 3 shows typical images obtained with our system. We collected regions of interest within the images, averaged the corresponding pixel values, and used these for longitudinal observation of each metric. We then compared the optical measurements to immunohistochemical data, showing an increase in perfusion, hemoglobin volume fraction, and number of vessels per unit area during the study of scars undergoing compression therapy. LDI showed increased levels of perfusion, during and after pressure application, which agreed with the optical measurement of blood volume fraction. Assessment of vascular endothelial growth factor transcript levels using reverse transcription PCR (polymer chain reaction) showed concordant increases.

Figure 3. (A) Scar appearance on day 105 after wounding. (B) The scar's blood perfusion map. (C) Map of hemoglobin volume fraction. (D) Map of oxygen saturation. The black string visible in (A) shows the reference suture.

Critical to scar pathology are the composition, cross-linking, and arrangement of collagen. Tightly arranged collagen fibrils will result in less pliable skin. We used the out-of-plane methodology included in the SPOS to measure^{14, 19} bulk collagen preferential orientation.^14,20 Scars that were treated with pressure showed a higher preferential angle orientation, and these results were in agreement with measurement of collagen orientation (orientation index) conducted by image processing of hematoxylin and eosin-stained biopsies.¹⁸ Furthermore, treated scars showed altered total collagen quantity and collagen type balance demonstrated using hydroxyproline assays, immunohistochemical staining, and Masson's trichrome staining.

To summarize, using both invasive and optical techniques, we detected and identified quantitative and qualitative collagen changes in scars undergoing pressure therapy. The system was also capable of quantifying the increase in melanin concentration in certain scars, the changes in water content, and 3D structure. In future work, we plan to further standardize our optical techniques with the ultimate goal of providing physicians with a tool not only for patient diagnosis, but also to compare treatment efficacy in scar formation.