Direct comparison of reproducibility and reliability in quantitative assessments of burn scar properties

Abstract

IntroductionDetermining the efficacy of anti-scar technologies can be difficult as qualitative, subjective assessments are often utilized instead of systematic, objective measures. Perceptions regarding the reliability of instruments for quantitative measurements along with their high cost and increased data collection time may discourage their use, leading to use of scar scales which are relatively quick and low-cost. To directly evaluate the reliability of instruments for quantitative measurements of scar properties, instruments and two qualitative scales were compared by assessing a variety of cutaneous scars. MethodsScar height and surface texture were evaluated using a 3D scanner and a mold/cast technique. Scar color was evaluated by using a spectroscopy-based tool, the Mexameter®, and digital photography with image analysis. Scar biomechanics were evaluated using the BTC-2000™, Dermal Torque Meter (DTM®), and ballistometer®. The Vancouver Scar Scale (VSS) and Patient and Observer Scar Assessment Scale (POSAS) were used to qualitatively evaluate the same scar properties. Intraclass correlation coefficients (ICC) were used to determine inter- and intra-user reliability (poor, moderate, good, excellent) with all instruments and the kappa reliability statistic was used to asses inter-user reliability (poor, fair, moderate, good, very good) for VSS and POSAS. Time for measurement collection and after collection analysis was also recorded. ResultsThe Mexameter® was the most reliable method for evaluating erythema and pigmentation compared to digital photography and image processing, POSAS and VSS. Digital photography and analysis was more reliable than POSAS and VSS. Assessment of scar height was significantly more reliable when using a 3D scanner versus VSS and POSAS. The 3D scanner and mold-cast techniques also offered an additional benefit of providing an absolute value of scar height relative to the surrounding tissue. Intra-user reliability for all mechanical tests was moderate to good. Inter-user reliability was greater when using the BTC-2000™ and ballistometer® versus the DTM®. All quantitative measurements took less than 90 s for collection, with the exception of the mold/cast technique. ConclusionNon-invasive instruments allow scar properties to be quantitatively assessed with high sensitivity and as a function of time and/or treatment without the need for biopsy collection. Overall, the reliability of scar assessments was significantly improved when quantitative instruments were utilized versus scar scales. Quantitative assessment of color and biomechanics were swift, requiring less than 90 s per measurement while assessments of texture and height required additional analysis time after collection. With proper training of clinical staff and well-defined protocols for measurement collection, reliable, quantitative assessments of scar properties can be collected with little disruption to the clinical workflow.