Clinical prediction of wound re-epithelisation outcomes in non-severe burn injury using the plasma lipidome

Abstract

Impaired wound healing in burn injuries can lead to complications such as skin graft loss, infection, and increased risk of scarring, which impacts long-term patient outcomes and quality of life. While wound repair in severe burns has received substantial research attention, poor wound outcomes in cases of non-severe burns (classified as <20% of the total body surface area (TBSA)) remain relatively understudied despite also having considerable physiological impact and constituting the majority of hospital admissions for burns. Predicting outcomes in the early stages of healing would decrease financial and patient burden, and aid in preventing long-term complications from poor wound healing. Lipids have been implicated in inflammation and tissue repair processes and may play essential roles in burn wound healing. Longitudinal plasma samples were collected from patients (n=20) with non-severe (<15% TBSA) flame or scald burns over a 6-week period including timepoints pre- and post-surgical intervention. Samples were analysed using liquid chromatography-tandem mass spectrometry and nuclear magnetic resonance spectroscopy to detect 850 lipid species and 112 lipoproteins. Statistical analyses, including orthogonal projection to latent structures-discriminant analysis was performed to identify changes associated with either re-epithelialisation or delayed wound re-epithelisation. The results demonstrated that the plasma lipid and lipoprotein profiles at admission could predict wound re-epithelisation outcomes at two weeks post-surgery, and that these discriminatory profiles were maintained over a 6-week period. Triacylglycerides, diacylglycerides (DAG) and low density lipoprotein (LDL) subfractions were associated with delayed wound closure, with DAG(18:2_18:3) and LDL/High density lipoprotein (HDL) ratio having the most influence (p-value < 0.02, Cliff's delta > 0.7), while HDL subfractions, phosphatidylinositols, phosphatidylcholines (PC), and phosphatidylserines were associated with re-epithelisation at two weeks post-surgery, with PC(16:0_18:1) and HDL-2 apolipoprotein-A1 showing the greatest influence on the model (p-value < 0.01 , Cliff's delta < -0.7). We demonstrate clinical prediction of wound re-epithelisation in non-severe burn patients using lipid and lipoprotein profiling. Further validation of the models will potentially lead to personalised intervention strategies to enhance injury outcomes, reducing the risk of chronic complications post-burn injury.