92 Burn Injury Assessment Study: How Good Are We at Assessing Burn Injury?

Abstract

BackgroundOver 1 million burn injuries occur each year in the U.S. seeking evaluation in emergent or urgent care settings. A major component of the initial evaluation includes burn wound assessment (BWA). Unfortunately, historical studies have demonstrated that providers incorrectly assess the depth of burn wounds (BW) in 25- 30% of injuries, in part because BW evaluation has no adjunctive labs or imaging tools to aid in the determination of healing versus non-healing BW.Study ObjectivesThe goal of our investigation was to study a diverse group of emergency medicine (EM) providers’ assessments of BW to better discern baseline accuracy and practice patterns.MethodsIRB-approved, prospective study designed to collect data from emergency department providers. Using a tablet- based data entry device, EM participants enrolling in the study completed a brief questionnaire followed by a series of five scenarios of thermal BW each with images from ABA-verified burn centers. 21-day follow-up photos of the BW were used to determine healing and non-healing regions by a consensus panel of fellowship-trained burn surgeons. EM participants were asked to mark the non-healing portions of the wound using a stylus. Results from the EM participants were compared to the 21-day healed/non-healed consensus panel to determine accuracy on a pixel-for-pixel basis. Statistical analysis of the non-healing portions of the wounds was performed by Chi square. Non-healing BW were defined as thermal burns of severity 3rd degree and deep 2nd degree. Healing burn were defined as thermal burns of severity 1st degree and superficial 2nd degree. Sensitivity was the percentage of “non-healing burn” areas that were marked. Specificity was the percentage of the “healing burn + uninjured skin” areas that were not marked. Accuracy was the percentage of area in a participant’s image correctly marked by the participant.Results77 participants enrolled in the study. 5 participants did not complete BWA drawings and were excluded. 360 images were reviewed in total for healing/non-healing BW. In the final cohort, 80% of the participants identified as emergency physicians in-practice (20% residents) with a median of 4 years in practice primarily at facilities not adjoined to a burn center (73%). For non-healing wounds, EM participants selected no surgery needed or local wound care only in 82% of the images. For healing wounds, participants correctly identified the wound as needing local wound care only in 34% of the images. Pixel-based mapping demonstrated a mean accuracy of 74%, sensitivity of 38%, and specificity of 77%.ConclusionsThis is the largest study to date examining emergency medicine providers ability to assess BW. This is important because in the event of a disaster, large numbers of burn casualties could overwhelm an already strained health care system and EM providers would be called upon to perform large volumes of BWA. Improvements in determining BW healing are essential to aid emergency department providers in the appropriate treatment of burn-injured patients.FundingFunding and technical support is provided by the Biomedical Advanced Research and Development Authority (BARDA), under the Assistant Secretary for Preparedness and Response (ASPR), within the U.S. Department of Health and Human Services (HHS), under ongoing USG Contract No. 75A50119C00033.Yes, authors have interests to discloseDisclosureAvita Medica, PolyNovo, Spectral MD, Access Pro Medical- in lieu of compensation all proceeds are donated to charities supporting burn outreach, education, survivor programs, and research.Consultant/AdvisorAvita Medica, PolyNovo, Spectral MD, Access Pro Medical- in lieu of compensation all proceeds are donated to charities supporting burn outreach, education, survivor programs, and research.DisclosureServes as the medical monitor for the Spectral MD DeepView Training StudyConsultant/AdvisorServes as the medical monitor for the Spectral MD DeepView Training Study DisclosureSpectral MD IncEmployee Spectral MD Inc BackgroundOver 1 million burn injuries occur each year in the U.S. seeking evaluation in emergent or urgent care settings. A major component of the initial evaluation includes burn wound assessment (BWA). Unfortunately, historical studies have demonstrated that providers incorrectly assess the depth of burn wounds (BW) in 25- 30% of injuries, in part because BW evaluation has no adjunctive labs or imaging tools to aid in the determination of healing versus non-healing BW. Over 1 million burn injuries occur each year in the U.S. seeking evaluation in emergent or urgent care settings. A major component of the initial evaluation includes burn wound assessment (BWA). Unfortunately, historical studies have demonstrated that providers incorrectly assess the depth of burn wounds (BW) in 25- 30% of injuries, in part because BW evaluation has no adjunctive labs or imaging tools to aid in the determination of healing versus non-healing BW. Study ObjectivesThe goal of our investigation was to study a diverse group of emergency medicine (EM) providers’ assessments of BW to better discern baseline accuracy and practice patterns. The goal of our investigation was to study a diverse group of emergency medicine (EM) providers’ assessments of BW to better discern baseline accuracy and practice patterns. MethodsIRB-approved, prospective study designed to collect data from emergency department providers. Using a tablet- based data entry device, EM participants enrolling in the study completed a brief questionnaire followed by a series of five scenarios of thermal BW each with images from ABA-verified burn centers. 21-day follow-up photos of the BW were used to determine healing and non-healing regions by a consensus panel of fellowship-trained burn surgeons. EM participants were asked to mark the non-healing portions of the wound using a stylus. Results from the EM participants were compared to the 21-day healed/non-healed consensus panel to determine accuracy on a pixel-for-pixel basis. Statistical analysis of the non-healing portions of the wounds was performed by Chi square. Non-healing BW were defined as thermal burns of severity 3rd degree and deep 2nd degree. Healing burn were defined as thermal burns of severity 1st degree and superficial 2nd degree. Sensitivity was the percentage of “non-healing burn” areas that were marked. Specificity was the percentage of the “healing burn + uninjured skin” areas that were not marked. Accuracy was the percentage of area in a participant’s image correctly marked by the participant. IRB-approved, prospective study designed to collect data from emergency department providers. Using a tablet- based data entry device, EM participants enrolling in the study completed a brief questionnaire followed by a series of five scenarios of thermal BW each with images from ABA-verified burn centers. 21-day follow-up photos of the BW were used to determine healing and non-healing regions by a consensus panel of fellowship-trained burn surgeons. EM participants were asked to mark the non-healing portions of the wound using a stylus. Results from the EM participants were compared to the 21-day healed/non-healed consensus panel to determine accuracy on a pixel-for-pixel basis. Statistical analysis of the non-healing portions of the wounds was performed by Chi square. Non-healing BW were defined as thermal burns of severity 3rd degree and deep 2nd degree. Healing burn were defined as thermal burns of severity 1st degree and superficial 2nd degree. Sensitivity was the percentage of “non-healing burn” areas that were marked. Specificity was the percentage of the “healing burn + uninjured skin” areas that were not marked. Accuracy was the percentage of area in a participant’s image correctly marked by the participant. Results77 participants enrolled in the study. 5 participants did not complete BWA drawings and were excluded. 360 images were reviewed in total for healing/non-healing BW. In the final cohort, 80% of the participants identified as emergency physicians in-practice (20% residents) with a median of 4 years in practice primarily at facilities not adjoined to a burn center (73%). For non-healing wounds, EM participants selected no surgery needed or local wound care only in 82% of the images. For healing wounds, participants correctly identified the wound as needing local wound care only in 34% of the images. Pixel-based mapping demonstrated a mean accuracy of 74%, sensitivity of 38%, and specificity of 77%. 77 participants enrolled in the study. 5 participants did not complete BWA drawings and were excluded. 360 images were reviewed in total for healing/non-healing BW. In the final cohort, 80% of the participants identified as emergency physicians in-practice (20% residents) with a median of 4 years in practice primarily at facilities not adjoined to a burn center (73%). For non-healing wounds, EM participants selected no surgery needed or local wound care only in 82% of the images. For healing wounds, participants correctly identified the wound as needing local wound care only in 34% of the images. Pixel-based mapping demonstrated a mean accuracy of 74%, sensitivity of 38%, and specificity of 77%. ConclusionsThis is the largest study to date examining emergency medicine providers ability to assess BW. This is important because in the event of a disaster, large numbers of burn casualties could overwhelm an already strained health care system and EM providers would be called upon to perform large volumes of BWA. Improvements in determining BW healing are essential to aid emergency department providers in the appropriate treatment of burn-injured patients. This is the largest study to date examining emergency medicine providers ability to assess BW. This is important because in the event of a disaster, large numbers of burn casualties could overwhelm an already strained health care system and EM providers would be called upon to perform large volumes of BWA. Improvements in determining BW healing are essential to aid emergency department providers in the appropriate treatment of burn-injured patients.