Why the study?
LV longitudinal strain is a sensitive marker of cardiovascular function in septic shock but is not universally available in standard clinical echocardiograms.
Can a machine learning model predict left ventricular longitudinal strain using standard echocardiographic and clinical variables in early sepsis?
Can a machine learning model predict left ventricular longitudinal strain using standard echocardiographic and clinical variables in early sepsis?
Standard echocardiographic and clinical variables can be used in a machine learning model to predict left ventricular longitudinal strain, a sensitive marker of cardiovascular function, in early sepsis.
May identify echocardiographic surrogates for LV strain in septic shock; leaves open whether these improve outcomes or warrant prospective validation.
Septic shock is a common deadly disease often associated with cardiovascular dysfunction. Left ventricular longitudinal strain (LV LS) has been proposed as a sensitive marker to measure cardiovascular function; however, it is not available universally in standard clinical echocardiograms. We sought to derive a predictive model for LV LS, using machine learning techniques with the hope that we may uncover surrogates for LV LS. We found that left ventricular ejection fraction, tricuspid annular plane systolic excursion, sepsis source, height, mitral valve Tei index, LV systolic dimension, aortic valve ejection time, and peak acceleration rate were all predictive of LV LS in this initial exploratory model. Future modeling work may uncover combinations of these variables which may be powerful surrogates for LV LS and cardiovascular function.
No takes yet. Share an insight, caveat, or question.
Knox et al. (2021) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: