Hemodynamic kinetic in severe burns

Position du problème et objectif(s) de l’étude

Large burns are associated with a massive inflammatory response and catecholamine release leading to complex cardiovascular disorders (1). Our objective was to describe the kinetic of hemodynamic changes immediately after large burn injuries.

Matériel et méthodes

We retrospectively collected data from patients admitted to the burn center for severe burn injuries between January 2017 and January 2025. Inclusion criteria were: admission within 24 hours of burn and initiation of PiCCO monitoring within 24 hours of admission. Exclusion criteria were: secondary transfer to the burn center and PiCCO monitoring time of less than ten hours. Approval for conducting the study was obtained from the ethics committee of the French Armed Forces Health Academy (IRB00014579). Age, mortality rate, total burns surface area (TBSA), third degree burn area and smoke inhalation injury were assessed for all included patients. Cardiac index (CI) was measured using transpulmonary thermodilution, then monitored by pulse wave contour analysis. Systemic Vascular Resistance Index (SVRI) was calculated based on CI, central veinous pressure (CVP) and arterial mean pressure (MAP). Heart rate was also collected hourly. Fluid resuscitation followed a standardized local protocol (2). The same protocol was used for all patients during this period. Data were extracted from Centricity Critical Care® database and analyzed using the R software. Data were collected hourly, during the first 96 hours of care. Time curves for CI, SVRI and heart rate were plotted for each patient. For each variable, mean and standard deviation were hourly determined for all the population. Mann-Kendall tests were performed to assess trends over time. Each variable curve was smoothed by a generalized additive model. Then, hemodynamic profiles including CI, SVRI and heart rate were compared using dynamic time warping (DTW). Clusters were identified through hierarchical clustering. The optimal number of clusters was determined by the elbow method. Quantitative demographic data were compared between clusters using Kruskal-Wallis test followed by Dunn-Bonferroni post-hoc test, while categorical data were compared using chi-squared test with Bonferroni correction.

Résultats & Discussion

One hundred and two patients were included with a median age of 52 years (IQR: 26-78) and a median total burn surface area of 40% (IQR: 15-65). Twenty-one patients died during their hospital stay. Average CI significantly increased, while average RVSI significantly decreased over the first 96 hours. Three hemodynamic profiles were highlighted following the cluster analysis: one with continuously increasing CI, one with CI stabilization after initial rise, and one with CI decline after initial rise. SVRI decreased over time in all clusters. Age was significantly lower in the cluster with a continuous increase in CI compared to the other two clusters. Mortality was significantly different between clusters and was lower in the in the cluster with continuously increasing CI.

Conclusion

These results confirm a two-phases evolution after a severe burn: first a hypokinetic - vasoconstriction state followed by a hyperdynamic circulation with vasoplegia. These results are consistent with an old small sized study(3) and a more recent paediatric study(4). Additionally, it suggests a worse prognosis among patients unable to sustain CI increasing.

Auteurs

Thibaut BAUDIC, Matthieu LAURENT, Jean-Vivien SCHAAL, Julie RENNER, Thomas LECLERC, Nicolas DONAT - (1)Hôpital National D'instruction Des Armées Percy, Clamart, France

Orateur(s)

Thibaut BAUDIC (Clamart)