Nonsevere Burn Induces a Prolonged Systemic Metabolic Phenotype Indicative of a Persistent Inflammatory Response Postinjury

Abstract

Globally, burns are a significant cause of injury that can cause substantial acute trauma as well as lead to increased incidence of chronic comorbidity and disease. To date, research has primarily focused on the systemic response to severe injury, with little in the literature reported on the impact of nonsevere injuries (<15% total burn surface area; TBSA). To elucidate the metabolic consequences of a nonsevere burn injury, longitudinal plasma was collected from adults (n = 35) who presented at hospital with a nonsevere burn injury at admission, and at 6 week follow up. A cross-sectional baseline sample was also collected from nonburn control participants (n = 14). Samples underwent multiplatform metabolic phenotyping using ¹H nuclear magnetic resonance spectroscopy and liquid chromatography-mass spectrometry to quantify 112 lipoprotein and glycoprotein signatures and 852 lipid species from across 20 subclasses. Multivariate data modeling (orthogonal projections to latent structures-discriminate analysis; OPLS-DA) revealed alterations in lipoprotein and lipid metabolism when comparing the baseline control to hospital admission samples, with the phenotypic signature found to be sustained at follow up. Univariate (Mann-Whitney U) testing and OPLS-DA indicated specific increases in GlycB (p-value < 1.0e^-4), low density lipoprotein-2 subfractions (variable importance in projection score; VIP > 6.83e^-1) and monoacyglyceride (20:4) (p-value < 1.0e^-4) and decreases in circulating anti-inflammatory high-density lipoprotein-4 subfractions (VIP > 7.75e^-1), phosphatidylcholines, phosphatidylglycerols, phosphatidylinositols, and phosphatidylserines. The results indicate a persistent systemic metabolic phenotype that occurs even in cases of a nonsevere burn injury. The phenotype is indicative of an acute inflammatory profile that continues to be sustained postinjury, suggesting an impact on systems health beyond the site of injury. The phenotypes contained metabolic signatures consistent with chronic inflammatory states reported to have an elevated incidence postburn injury. Such phenotypic signatures may provide patient stratification opportunities, to identify individual responses to injury, personalize intervention strategies, and improve acute care, reducing the risk of chronic comorbidity.

Keywords: acute burn injury; inflammation; lipids; lipoproteins; liquid chromatography-tandem mass spectrometry; metabolic phenotyping; nonsevere burn; nuclear magnetic resonance; supramolecular phospholipid composite; thermal injury.

Figure 1

Overview of nonsevere burn patient and nonburn participant recruitment and experimental workflow. A) Main selection criteria for nonsevere burns and nonburn participants with final recruitment numbers used in the study. B) Sample collection time points for both groups with single plasma collection time point for nonburn controls and two time points from hospital admission and 6 weeks postburn debridement surgery for the nonsevere burn group. C) Basic workflow of nonsevere burn and nonburn plasma samples analyzed by ¹H nuclear magnetic resonance (NMR) spectroscopy and using diffusion and relaxation editing. Data analyzed included 112 lipoprotein subfractions, GlycB, and the supramolecular phospholipid composite (SPC). D) Basic lipidomic workflow of nonsevere burn and nonburn plasma analyzed by liquid chromatography tandem mass spectrometry to quantify 1163 lipid species across 20 subclasses. Image created with BioRender.com.

Figure 2

OPLS-DA analysis of lipoprotein profiles from paired burn plasma samples at admission to a hospital and 6 weeks postburn surgery compared to nonburn controls. A) OPLS-DA score plots of admission (n = 35; colored blue) (R²X = 0.25, cross validated-area under the receiver operating characteristic (CV-AUROC) = 0.95 compared to nonburn controls (n = 14; colored white). B) OPLS-DA score plots of 6 weeks postburn surgery (n = 35; colored orange) (R²X = 0.12, CV-AUROC = 0.96) compared to nonburn controls (n = 14; colored white). C) Bar plot of the top 12 variable importance in projection (VIP) scores, which summarizes and ranks the influence of the individual lipoproteins on the OPLS-DA model, representing burns at admission (blue) and nonburn control (white). D) Bar plot of the top 12 VIP scores representing the 6 weeks postburn surgery (orange) and nonburn controls (white).

Figure 3

Univariate analysis of SPC₁, SPC₂, SPC₃, GlycB, and SPC₁ to GlycB ratio values of nonburn controls compared to burns at admission and 6 weeks postsurgery. Box and whisker plots of nonburn controls (n = 14) (white) compared to burns at admission (n = 35) (blue) and 6 weeks postsurgery (n = 35) (orange) for SPC₁, SPC₂, SPC₃, GlycB, and SPC₁ to GlycB ratio values. Significance level from Mann–Whitney U tests between nonburn controls to burns at admission and nonburn controls to 6 weeks postsurgery are shown above in the corresponding plots and represented with “*”. Significance: ns = not significant; * = p-value < 0.05; ** = p-value < 0.01; *** = p-value < 0.001; **** = p-value < 0.0001.

Figure 4

OPLS-DA analysis of lipids from plasma samples collected from hospitalised patients postnonsevere burn injury both at admission to a hospital and 6 weeks postsurgery vs nonburn controls. A) OPLS-DA score plots of admissions (n = 35; colored blue) vs nonburn controls (n = 14; colored white) (R²X = 0.21, CV-AUROC = 0.95). B) OPLS-DA score plots of 6 weeks post burn surgery (n = 35; colored orange) vs nonburn controls (n = 14; colored white) (R²X = 0.11, CV-AUROC = 0.80). C) Bar plots corresponding to the top 12 loading variables of each OPLS-DA model (ranked by variable importance in projection (VIP) score), colors represent burn injury group at admission (blue) vs nonburn controls (white). D) Bar plots corresponding to the top 12 loading variables of each OPLS-DA model (ranked by VIP score), colors represent burn injury group at 6 weeks (orange) vs nonburn controls (white).

Figure 5

Top 20 lipids from univariate analysis of lipid profiles from nonburn controls compared to burns at admission and 6 weeks postsurgery. Box and whisker plots of nonburn controls (n = 14) (white) compared to burn admissions (n = 35) (blue) and 6 weeks postburn surgery (n = 35) (orange) for the top 20 lipids from univariate analysis of nonburn controls vs the burn injury group at each time point. Significance level from Mann–Whitney U tests between nonburn controls to burn admissions and nonburn controls to 6 weeks postburn surgery are shown above in the corresponding plots and represented with “*”. Significance: ns = not significant; * = p-value < 0.05; ** = p-value < 0.01; *** = p-value < 0.001; **** = p-value < 0.0001.